Professional Documents
Culture Documents
Capitulo 4
Capitulo 4
Version 1.0
by C C Parkman, J Lowe, A B Sterling, J Rolt and H R Kerali
This is an unpublished report prepared for the Asian Development Bank and must not be referred to in any
publication without the permission of the Asian Development Bank. The views expressed are those of TRL and not
necessarily those of the Asian Development Bank.
Approvals
Project Manager
Quality Reviewed
The information contained herein is the property of the Transport Research Laboratory and does not reflect
the views or policies of the customer for whom the report was prepared. Whilst every effort has been made to
ensure that the matter presented in this report is relevant, accurate and up-to-date at the time of publication,
the Transport Research Laboratory cannot accept liability for any error or omission.
DRAFT
4. DISSEMINATION STRATEGY............................................................................. 5
4.1 Justification for dissemination .......................................................................... 5
4.2 Existing modes of dissemination....................................................................... 6
4.2.1 Newsletter.................................................................................................. 6
4.2.2 Brochure .................................................................................................... 6
4.2.3 Conference papers and presentations and published articles .................... 7
4.2.4 Website at University of Birmingham, PIARC and roadsource.com ....... 7
4.2.5 Training courses ........................................................................................ 8
4.2.6 Support from University of Birmingham and other institutions ............... 8
4.3 Options for dissemination ................................................................................. 8
4.3.1 Long term HDM-4 ‘hub’ overseeing dissemination ................................. 8
4.3.2 Regional road management working groups............................................. 9
4.3.3 Workshop sessions .................................................................................. 10
4.3.4 Targeted conference sessions .................................................................. 10
4.3.5 Published papers and articles .................................................................. 11
4.3.6 Website.................................................................................................... 11
4.3.7 Audio-visual and electronic media.......................................................... 12
4.3.8 Update information with new releases of program ................................. 12
4.3.9 Newsletter and brochure.......................................................................... 12
5. TRAINING STRATEGY....................................................................................... 13
5.1 Identified learning needs ................................................................................. 13
5.2 Access to training ............................................................................................ 14
5.2.1 Modes of training .................................................................................... 14
5.2.2 Methods of training ................................................................................. 15
5.2.3 Developing the strategy during the ADB project.................................... 15
5.3 Modular structure ............................................................................................ 16
5.4 Module content................................................................................................ 16
5.5 Assessment ...................................................................................................... 17
5.6 Training the trainers ........................................................................................ 17
5.7 Major risk / assumptions ................................................................................. 18
5.7.1 Sustainability........................................................................................... 18
5.7.2 Champions and Key Trainers.................................................................. 18
5.7.3 Demand for training ................................................................................ 19
5.7.4 Priority given to HDM-4 ......................................................................... 19
5.7.5 Modes of delivery.................................................................................... 19
5.7.6 Calibre of Trainers .................................................................................. 19
5.7.7 Road management practice versus HDM-4 capability............................ 19
ANNEX 1.................................................................................................................... 21
LOGICAL FRAMEWORK FOR TRAINING AND DISSEMINATION OF HDM-4 .
ANNEX 2.................................................................................................................... 22
LOGICAL FRAMEWORK FOR ADB FUNDED TRAINING AND
DISSEMINATION PROJECT
ANNEX 3.................................................................................................................... 23
TRAINING PROGRAMME DEVELOPMENT
ANNEX 4.................................................................................................................... 24
NEEDS ASSESSMENT AND PROPOSED MODULAR STRUCTURE FOR HDM-
4 TRAIN THE TRAINER PROGRAMME
HDM-4 TRAINING AND DISSEMINATION STRATEGY
1. INTRODUCTION
This draft strategy is to meet the training and dissemination needs that arise from the
introduction of HDM-4, a replacement for HDM-III. An overview and background to
HDM-4 and its technical characteristics are covered in the “Overview of the New
HDM-4 System” by H R Kerali, C J Hoban and W D O Paterson (1999) given at the
PIARC XXIst World Congress held in Kuala Lumpur.
HDM-III has been in use for over 10 years and the essential differences are that
HDM-4 extends the scope of the HDM-III program by incorporating:
2.1 Overview
The Terms of Reference for the project require the production of a strategy for
sustainable training and dissemination of HDM-4. It requires the strategy to be
produced in particular forms as follows:
All the above items go beyond the project timeframe and, as such, aim to create
sustainable training and dissemination processes and structures.
i) An overall training and dissemination strategy for this project in the Asia
Pacific region, which will take place over the next two years;
ii) Dissemination: a 5 year global dissemination programme; and
iii) Training: a 3 year global training programme.
This report summarises the conclusions from those discussions and presents revised
logframes based on the results of the workshop supplemented by further details
developed since that time.
The workshop confirmed most of the broad details of the three logframes, while
making the following changes:
i) The Goal was broadened to extend its scope and to remove the geographical
limitation of reference to the Asia-Pacific region. The draft final version
reads:
‘To improve the development and performance of the road sector and its use
of resources, within the context of national objectives, in order to minimise
total transport costs.’
ii) The Purpose was altered to emphasise not only increasing capability but also
actual application of HDM-4. The draft final version reads:
(b) emphasise the full range of media through which the training and
dissemination materials will be available.
iv) Verifiable Indicators were expanded to describe some details of the scope of
training and dissemination intended.
i) the existing logframes did not develop the strategy to the required extent
beyond the timeframe of the existing project; and
ii) the three existing logframes demonstrated considerable overlap, and were
repetitious in content.
The team have therefore decided that two logframes would meet the objectives of the
Terms of Reference and provide a clearer and more focused strategy.
i) Logframe for overall strategy (Annex 1). This key logframe represents the
overall strategy for the training and dissemination component of HDM-4, and
identifies the key outputs which we believe the HDM-4 stakeholders are
aiming to achieve in the long term.
ii) Logframe for ADB funded project (Annex 2). A more detailed logframe for
the two year project has also been developed from the original project
logframe presented at Kuala Lumpur.
The contents of the logframes are described in more detail in the following sections.
First, the overall strategy is discussed (Sections 4 and 5) and then aspects of the
current project which contribute to this overall strategy are described (Section 6).
The Goal and Purpose of each logframe is the same, and so these aspects are
discussed first in a separate section.
It should be noted that, at present, there are no specific provisions for project support
to extend beyond the first two years. This is an issue that will need to be reviewed
during the project as it needs to be commensurate with the overall strategy for
managing HDM-4 in the future. This, in turn, depends on the evolving Business Plan
prepared by PIARC and future institutional arrangements which are yet to be
finalised.
These aims can be encapsulated in the logframes at the Goal and Purpose level, as
agreed at Kuala Lumpur and described in Section 2.2. Whilst achieving the Goal and
Purpose is outside the control of the project, it is important to clarify those risks which
have been identified so that stakeholders can take steps to mitigate against such
possibilities.
Four risks, or assumptions, have been noted in the logframes. Three of these (suitable
data for and calibration of HDM-4, HDM-4 producing improved analyses, application
of results by policymakers without undue distortion) are self explanatory. The final
assumption that ‘HDM-4 gains credibility and acceptance by government agencies’
warrants further comment, based on discussions in Kuala Lumpur.
3.2 Sustainability
3.2.1 Acceptance of HDM-4
From discussions at the PIARC World Congress in Kuala Lumpur it is evident that
there are a number of key stakeholders who have concerns about the performance
parameters of HDM-4 and the strength of support/desire to adopt it. Difficulties with
calibration due to the complexity of the model, and the possible negative reactions of
some potential users to lower levels of projected benefits arising from the use of the
new version of HDM, have also been raised. These are some of the issues the project
team will have to resolve with stakeholders if HDM-4 development and adoption is to
be successfully prosecuted.
Failure to acknowledge and resolve these issues will seriously impair the willingness
of individuals, groups, institutions and governments to take ownership for the long-
term development and maintenance of HDM-4 in their region. Stakeholder
partnerships and participation at all levels will be fundamental to securing a dynamic
and broadly based post-project training and dissemination programme.
3.2.2 Motivation
In developing the project and post-project strategy it must be remembered that the
adoption of HDM 4 is not compulsory, so its benefits and advantages may need to be
made clear and, in some cases, incentives created.
The ADB intend that the project inputs should provide the training and dissemination
programmes with a direction and internal momentum to take the training programme
through to the end of year three and the dissemination programme to the end of year
five. How this post project phase will be supported and funded has yet to be resolved
but it is essential that the project identifies and pursues strategies that lead to early
local commitment and ownership of the HDM-4 programme so that post-project
activity can be considered and planned during the current project phase.
Motivation is a complex issue but two possible lines to follow in the project are:
In truth such a division is simplistic and it is likely that various combinations of these
will exist in the environment in which the project will have to “sell” HDM-4. The
project team must carefully assess and exploit such factors when promoting the
formation of local and regional HDM-4 steering/support institutions/groups.
4. DISSEMINATION STRATEGY
• Alert international donors and financiers to the role of HDM-4 so that they
design projects which have the potential to incorporate the use of HDM-4;
• Alert national policymakers to the role of HDM-4, so that they facilitate its
correct use within road and transport organisations;
• Alert users (including those who train others in the use of HDM-4) to the
scope and limitations of HDM-4; and
• Update the above groups progressively as HDM-4 is developed.
The need to generate genuine support for the use of HDM-4 and belief in its value,
before simply informing target groups of its capabilities, has been identified and
emphasised in Section 3. This includes all parties who might be involved with HDM-
4, from policy makers, whose support will be required, through to users, whose
genuine use of the system will be needed in order to provide meaningful outputs. A
sustainable strategy must address this issue.
In order to achieve the above and also to make best use of the resources available for
dissemination and training, it is recommended that the content of dissemination
materials is the prime initial focus. Addressing the many different and more
• target the most strategic audiences, in order to gain long term commitment
• target the greatest size of audience for given cost (benefit-cost)
• require the least amount of on-going input after initial set-up.
4.2.1 Newsletter
The ISOHDM Technical Secretariat at the University of Birmingham has been
publishing biannual Newsletters since 1996. This is a four-page document printed on
A3 size paper and folded in half to form the four pages of A4 size. The current print
format is black on white, although a multi-colour print format is under consideration.
The Newsletter is printed at the University of Birmingham at a cost of approximately
GBP 1,500 per print run thus totalling around GBP 3,000 per year.
The Newsletter is widely circulated in printed form to over 3,000 registered readers.
The vast majority of these are from the ISOHDM database which records details of all
individuals and organisations who have enquired about HDM-4 or are current users of
HDM-III/95. Electronic copies of the Newsletters are also posted on the ISOHDM
Web site hosted by the University of Birmingham.
Topics covered in the Newsletter have been wide ranging including the following:
• Overview of the HDM-4 system
• Descriptions of technical models
• Proposed plans for world-wide training programmes
• Pilot implementations of HDM-4
• Plans for release of HDM-4, etc.
Feedback from readers over the past few years indicates that the Newsletter has been
a successful and effective dissemination medium. Some readers have sent comments
on various articles in the Newsletter. However, only few readers have so far
contributed articles to the Newsletter.
4.2.2 Brochure
A full colour brochure was designed and published by the ISOHDM technical
secretariat at the University of Birmingham in 1998. Since then, over 5,000 copies of
the brochure have been distributed at conferences, by mail and to individuals visiting
PIARC and the University of Birmingham. The brochure is printed on hard-glossy A3
sheet and folded to form four-pages of A4. It includes several screen shots of HDM-4,
photographs of several roads with different types of traffic, pavement condition and
terrain.
There have been several papers published at international and at regional conferences
covering topics directly related to HDM-4. Recent conferences include:
In addition to conferences, there have been a few published articles in learned journals
and other periodicals.
The annual circulation of the conference proceedings and journals in which articles
and papers on HDM-4 have been published is not immediately available. However,
the effectiveness of these publications is assessed to be high given that they are
targeted at the more technical audience and potential HDM-4 users. These
publications are seen to be essential in encouraging proper technical use of HDM-4
covering topics such as; calibration, adaptation, implementation, data requirements,
integration with other pavement management systems, etc.
It is therefore proposed that the ISOHDM secretariats at PIARC and at the University
of Birmingham should continue to encourage HDM-4 users to publish the results of
selected applications of HDM-4 which are judged to be of interest to other users. Such
articles and papers must, of necessity, meet the editorial requirements of the
conference and journal publishers.
The longest established is the site under the University of Birmingham’s site. This
contains details of the earlier studies, copies of newsletters, product information,
contacts and sponsors. There is an opportunity to submit feedback, and a page on
seminars (under preparation). The newsletters and product information appear up to
date; at present, the remaining pages are less comprehensive.
The third site, the roadsource site, is more extensive than either of the above, and
focuses on the technical aspects of the HDM-4 system. It is not clear that it is official,
although there is a registration page which seems to be sponsored by PIARC. Other
pages include a discussion site (not clear that this is operational), a contacts page, and
downloads of various technical papers. The papers and documents included are not
comprehensive and suffer from the fact that updates to various technical issues are
continually being made, so that some of the documents are obsolete.
The success of the sites is unknown although the number of visitors to each site is
recorded. This is of the order of one to two thousand per annum.
Other courses also exist which include sessions on HDM within a broader context.
These have included the annual Management and Financing of Roads Course at
University of Birmingham, the annual Roads and Transport Course at TRL and other
similar events.
The intention would be that Country Champions (one per country), trainers and users
might all contribute to such a group, which might meet on an annual basis at, for
example, a regional road management conference, with the purpose of exchanging
experiences of using HDM-4 within the road management context.
The use of Champions has been discussed at Kuala Lumpur and it is felt that this is
the best way to proceed with the current ADB funded project. Country champions
will be high profile, high calibre personnel with considerable experience within their
country. As such they will have experience of operating at the policy level, and
understand the political context of road management. It is expected that one per
country will be appointed and their area of contribution would ideally include, but not
be limited to:
Country champions will be identified for the six countries during the ADB funded
project. On-going support for them, and future country champions elsewhere, is
beyond the scope of this project. The setting up of any working groups would also be
beyond the scope of the current ADB funded project.
Potential impact High. The target audience would combine users and stakeholders, and
generate a sense of local ownership.
Potential costs (1) Initial set up costs would be administrative time of a coordinating body
to promote the idea (PIARC).
(2) Some on-going support would be needed to fund members to attend
working groups, if they were unable to obtain funding elsewhere. The cost
of use of a suitable venue / facilities / publishing ‘proceedings’ etc. would
also need to be covered.
It is therefore recommended that the project sponsors are proactive in targeting this
group by arranging workshop sessions in which HDM-4 is promoted. These
workshop sessions might be held on a national basis, at an in-country road conference
or similar occasion, or a regional event. The former is preferred, in order to focus on
the needs of each country.
Potential impact High. The target group is the key to long-term sustainability.
Potential costs (1) Coordinating contacts etc to arrange suitable meetings
(2) Contributors will need some preparation time to provide well prepared
contributions.
(3) Costs of attendance by donors and financiers (national policymakers
attendance assumed). Costs of a workshop coordinator.
Major risks Low. Assuming HDM-4 is appropriate, the target group should be
persuaded.
Actions required Identification of suitable forums, and coordination of set-up.
Potential impact Medium to low. Such articles and papers only tend to be read by those
proactively interested.
Potential costs (1) A resource guide, identifying all published material, will be developed
under the ADB project in electronic and hardcopy format (subject to ADB
approval).
(2) On-going updates of the guide, and an annual sum for a prize.
Major risks None identified.
Actions required (1) Resource guide under ADB project. Cost estimate to be developed by
project team.
(2) Funding for on-going updates and prize put in place.
4.3.6 Website
As noted in 4.2.3, there are existing official websites under the University of
Birmingham and PIARC, and also a site sponsored by HTC Infrastructure
Management Ltd (www.roadsource.com). Whilst the user who is unaware of where
the websites are might enter the internet in a variety of locations (and so references on
key sites are recommended), it is recommended that a rational strategy is adopted for
the websites that hold information – this will avoid repetition, possible conflict, and
minimise costs (since if a website is to be maintained up to date, funding will be
required for its establishment and on-going management).
Potential impact Difficult to assess but will become more important with time
Potential costs Management fees for web maintenance
Major risks Without a working website, those wishing to attempt independent study or
upgrading of their current HDM-4 knowledge could be severely
compromised.
Actions required (1) Project sponsors rationalise ideas on how many websites are required
The use of electronic (CD-ROM) media, either as an addition to the HDM-4 program,
or stand-alone, would also be expensive. This aspect is discussed under training, as it
is seen more as a training activity.
The assumption above is that a database of contacts has been maintained up to date.
This supporting activity will enable targeted dissemination for this, and other
purposes such as the newsletter (see 4.3.9). The on-going maintenance of such a
database will require administration and management.
It is also proposed that the brochure (see Section 4.2.2) should now be updated in
order to incorporate the latest features in HDM-4 and the distribution arrangements.
5. TRAINING STRATEGY
All these groups have differing training and dissemination needs and the project will
have to define and prioritise these needs so that appropriate strategies can be
developed to meet them within the Project and subsequently in the post-project period
(Annex 3 – Training programme development).
The programme will be developed to meet the identified learning needs. The
identification process will be carried out through close collaboration with the project
technical advisory team, Country Champions and a range of other stakeholders. It is
recognised that the likely numbers of people who require training will not be large
and therefore it will not always be possible to assemble groups who will be
homogeneous in their requirements. The training material and methods of delivery
will also need to be tailored to cater for this diversity.
A first assessment of one of the target audiences - the Key Trainers who are to be
trained under the ADB project (see Section 5.6) – in terms of their required
characteristics and needs is provided as Annex 4. The training needs assessments for
future trainees in the countries where the trained Key Trainers will operate are
Table 5.1
Advantages and disadvantages of Trainer led courses and workshops
Advantages The trainer is able to tailor the training to meet the precise needs of
the trainees.
The trainees can get instant feedback and guidance from the trainer.
Trainees have a wider learning experience through interaction with
peers. (Group exercises and discussion).
Trainer can enhance the training content with topical and local
examples.
Training material can be produced in most simple print form and
subsequently enhanced by individual trainer to suit target groups.
Disadvantages Requires skilled trainers.
Requires trainees to come together at same time.
Needs appropriate training venue and equipment.
Trainees have to progress at pace of the group.
Not all trainers can accommodate a wide range of learning ability
and experience.
May be difficult to identify good trainers.
Table 5.2
Advantages and disadvantages of Supported Independent Individual/Group
Learner(s)
(For this method, a local tutor or coach is available to provide intermittent support,
which might be face-to-face by appointment, or telephone, or e-mail.)
Table 5.3
Advantages and disadvantages of Independent Individual/Group Learner(s)
The three tables above set out the main features but in practice it is likely that the
methods summarised in Tables 5.1 and 5.2 would be the most effective modes for
learners who are encountering HDM-4 for the first time. Table 5.3 could provide the
mode for training which is intended to enhance the learners ability to use HDM-4 in
specific circumstance or to add a specific process to their current experience i.e.
upgrading.
While the use of distance and independent study approaches will be developed it is
wise to recognise that the inputs and support needed to create such a sustainable
training system is well outside the current project budget. The project will carry out
research to establish what realistic provision might be feasible and viable within the
parameters of need and numbers. It will be part of the project strategy to establish a
network of local trainers/experts who can provide support and advice to those wishing
to adopt independent modes of study. The essence of considering a variety of
delivery modes and methods is so that various emphases and mixes can be achieved to
suit different needs.
If the training programme is to be sustainable it will be essential that the project team
continue to provide the Key Trainers with support and guidance throughout the
project. It is also clear that the Key Trainers have a major role to play in the long-
term development and sustainability of the programme in their home countries. The
project will endeavour to ensure that training programme ownership is established at
country, regional and international levels. This issue will be included in the wider
strategy development discussed in Sections 3 and 4 of this document.
Account will then need to be taken of different possible training methods and modes
of delivery (See 5.2 above and Annex 4 – Proposed training module structure).
Module development will begin as soon as possible and an early priority must be on
providing training to those concerned with the use of HDM-4 in policy and planning
roles e.g. senior civil servant, planners and sector leaders. A commitment to using
HDM 4 in planning and management at senior level will ensure a greater likelihood of
commitment to, and bottom up demand for, training. A general consensus amongst
key stakeholders is that this is a vital to ensuring the success of the programme.
It is uncertain how many of those nominated to become key trainers in the project will
have the experience or standing to deliver this “policy” level of training. The project
may have separately to identify and train suitable candidates to carry out this work.
5.5 Assessment
The practicalities and desirability of assessment and possible certification of HDM-4
training programmes is not yet clear. However it is desirable that the effectiveness of
training should be tested and feedback used to improve module content, structures and
delivery. To assist this process the development of the training modules will include
the specification of expected competencies and levels of performance to be achieved
at the end of each module. The political and practical issues of evaluating and
certifying individual learners will be explored and resolved locally as the project
develops.
The required standards for accreditation for the different levels of HDM-4 may be, for
example
The three-year plan must not restrict the assessment policy by becoming fixed once it
is documented. It will need to be responsive to the evolving HDM-4 application
environment.
The selection of key trainers will be critical to the success of the training programme.
Selection should take note of the nominees interpersonal and communication skills,
previous training experience and capacity to promote training and market HDM-4 as
well as their technical expertise and experience.
It is intended that the initial T3 programme will aim to achieve the strategy objectives
by ensuring that trainees will:
Given the developmental nature of the Project it will be essential to support “Key
Trainers” in launching new modules and modes of study as they become available.
Following the T3 course the project team will review the original train the trainer
thinking with the aim of strengthening the longer-term in-country support of trainers
and present ADB with an update. This may need some reflection and discussions
with “Key Trainers” and Champions before a formal proposal can be formulated.
5.7.1 Sustainability
Risk There is a concern that sustainability of the strategy is not assured within the
region, particularly after the timeframe of the Project.
Actions required This risk has been discussed in Section 3.2.
6. THIS PROJECT
The ADB funded HDM-4 Training and Dissemination Project will be implemented
over the next 18 months to two years. The detailed logframe for this project is
provided in Annex 2. The project will build a project training capacity by
The Training and Dissemination process and its subsequent development and
sustainability will have to be owned and operated by local, regional and international
stakeholders, as stated previously. The early and full involvement of stakeholders is a
critical factor in developing such ownership. Stakeholder commitment is also critical
to Project purpose and goal attainment. In this regard, the early identification and
engagement of country Champions will be crucial to Project progress.
The project team will develop and introduce a selection of training and dissemination
models so that options are available for stakeholders in different countries to utilise
according to their needs and resources. The on-going development of training and
learning materials and modes of delivery will be informed by the feed back from
stakeholders and field experience. However, the principle will be to start from a
simple basis, and develop different media etc from that point. In this regard, the first
task of development of materials will be to ensure there are a core set of documents
from which the training modules can be developed.
3. 24 T3 trainers trained. 3.1 TT Candidates qualifications 3.1 Candidates CVs and project
satisfy selection criteria (week 26). records.
3.2 At least 50% of TT Candidates 3.2 Project records and course
satisfy performance criteria (week 41) evaluation.
3.3 Action plans produced by country 3.3 Course evaluation.
groups (week 41).
4. 10 National end-users (and others) 4.1 One in-person training course run 4.1 Project records.
in each pilot country trained to at selected institutes in each of 6
appropriate level. selected countries in Asia-Pacific
region (Wk 68)
4.2 End of course evaluations and 4.2 Future courses incorporate
employer feedback. lessons learned.
4. Increased volume and scope of 4.1 HDM-4 represented at targeted 4.1 Reports
communication and dissemination conferences (4.3.4)
system (4.3.4 – 4.3.9) 4.2 Published papers and articles 4.2 Conference proceedings and
supporting HDM-4 use (4.3.5) journals
4.3 Website has increased range of 4.3 Site visits and user feedback
information and services (4.3.6)
4.4 Increased use of media systems in 4.4 Budget spend, availability of
promotion and training (4.3.7) materials, and training and
4.5 Program updates disseminated dissemination reports
(4.3.8) 4.5 HDM-4 updates with new releases
4.6 Newsletters published biannually 4.6 Mailing list and recipient feedback
and new brochure produced
(4.3.9)
4.7 Help network established and 4.7 System reports and user feedback
operating at local and international
levels by early 2001
5. Sustainable system of maintaining 5.1 Training materials review team 5.1 Reports and feedback from field
and upgrading training modes and formed
materials established (5.1-5.4) 5.2 Upgrades available and distributed 5.2 Reports from trainers and
champions
6. Sustainable training programme 6.1 Training provision meeting 6.1 Country surveys
extended throughout region (5.5) demand with low level of exclusion
Activities to Output
Activities 1. There is increasing demand for
1.1 ToR drawn up for empowering competent users of HDM-4
body to include training and 2. There are candidates who want to
dissemination coordinating function be trained.
1.2 Body (currently PIARC) appointed 3. There is funding available to pay for
for long term candidates to be trained.
4. Fees from courses (and other
2.1 Suitable forums identified and sources) are sufficient to make
sessions targeted and coordinated training financially sustainable.
5. Universities & other institutions
3.1 Concept explored and developed perceive HDM-4 training to be
with champions and trainers at first important.
T3 course. 6. Trainers / institutions continue to
3.2 Proposal for logistics and funding produce training materials for all
requirements in TDS Version 2 media.
3.3 HDM-4 sponsors implement 7. The standards established in the
funding and activities to ensure pilot project are at least maintained.
future developments 8. There is sufficient interest to lead to
WEB sites being used and
4.1 Current dissemination volume and maintained.
scope reported, and conceptual
ideas for increased levels
proposed, in TDS Version 1
4.2 Costed proposals included for
approved ideas, in TDS Version 2
4.3 HDM-4 sponsors implement
projects accordingly
2. To provide the trainers with set of training methods and skills which are
appropriate for the target groups they will be training and compatible with the
technical content of the course.
A vital aspect of the T3 programme must be that the training methods used in the
technical inputs should be models of good practice for the trainees to adopt. All the
evidence is that trainers will teach as they are taught. We must ensure that this is not
a case of “do as we say - not as we do”. This will need inputs to the course on
training methods and approaches.
e.g.
The learning process
Target group profiling and training need assessment
Using training objectives to define training methods and content
Planning training sessions
Participatory learning methods
Communication skills, learning aids
The reflective practitioner - Programme feedback, evaluation and improvement.
A4-1
Note that these include only the technical modules, and do not include details on the
modules for training the participants to train. Such training modules will be detailed
in due course; however they will clearly only be applicable to the T3 course. The
following modules will subsequently be developed, however, to be applicable for
future courses held by the trainers in their own countries, and for other HDM training
purposes.
In addition, note that whilst the participants at the T3 course will already have
experience in the use of HDM-III, the following modules will be prepared with the
assumption of little knowledge of HDM (although a general background in highway
management and engineering). This is because these materials will then be used by
the trainers and pared back for the less competent trainee. Hence at the T3 course the
training materials for each module, as well as being used to teach the participants
themselves, will also provide a means of introducing the participants to the materials
from which they will draw for their own courses in the future.
A4-2
Module Title Road investment appraisal
Key contents Three submodules:
(1) Economic analysis – definitions, components of total transport cost,
types of analyses, difference between financial and economic, different
indicators (NPV, IRR etc);
(2) Risk and sensitivity analysis – definitions and the need, examples and
experience of key variables;
(3) Advanced analyses including different optimisation techniques –
different techniques for ranking competing projects, their role in the
context of broader multi-criteria analyses
Expected existing As users of HDM, trainers will have an understanding of economic
competencies before T3 analyses. However, they might not be aware of the more advanced
course application of risk and sensitivity analyses, or of the various optimisation
techniques.
Proposed ability on Full understanding of the subject. Experience will be gained in application
completion of T3 course during the case studies and worked examples (project, programme and
strategy analyses modules below).
A4-3
Module Title Traffic
Key contents Two submodules:
(1) Types of traffic and their effects – different types of traffic, volume,
different vehicle types (complexities of different parameters related to each
vehicle for use in RUE models discussed in RUE module), axle loads,
fourth power law, traffic growth;
(2) Case studies – on the transformation of data from traffic studies into
HDM
Expected existing Participants should be well aware of these concepts from HDM-III
competencies before T3
course
Proposed ability on Full knowledge of all aspects, and ability to manipulate data for use in
completion of T3 course HDM-4.
A4-4
Module Title Programme analyses
Key contents A minimum of four submodules to cover the different sorts of programme
analysis.
Expected existing Participants will have some experience of particular type of analyses, but
competencies before T3 not necessarily a knowledge of the breadth of analyses of which HDM-4 is
course capable.
Proposed ability on Ability to design and develop analyses for the breadth of possibilities
completion of T3 course within their own countries.
A4-5
TITLE: Priorities in the collection of information for use in UNIT:
road management.
SHORT TITLE: Information strategies
Primary Module: Road and Transport Information
Duration: Presentation 1 hour
Target Group: Engineers/Planners
OBJECTIVES
RELATED UNITS/MODULES
i) Information types.
ii) Calibration of RDWE and RUE.
iii) Modelling RDWE and RUE.
iv) Configuration and implementation of HDM-4.
INTRODUCTION
This session is best taught by linking the aim to two main issues,
namely the difficulties of sustaining data hungry decision
support tools and the overall sensitivity of decisions to key
parameters.
DEVELOPMENT
SUMMARY
RESOURCE MATERIALS
Visual Aids:
1. Powerpoint presentation (15 – 20 slides).
Case studies/Practicals:
1. As required.
TRAINING FACILITIES
MISCELLANEOUS
OBJECTIVES:
The trainee will:
1. Identify decisions which have to be made in the road sector
2. Explain the principle of total transport costs, including
costs and benefits
3. Describe the concept of life cycle analysis, including
discounting
4. Explain the difference between economic and financial
costs
5. Identify primary and secondary benefits
6. Explain the concept of base case and do-minimum
comparisons
7. Describe the differences between estimation of benefits
from Normal, Diverted and Generated traffic
8. Describe the key economic indicators (NPV, IRR, NPV/C)
and their use in investment appraisal
RELATED UNITS/MODULES
i) Operation of HDM-4 / Modelling framework
ii) HDM-4 Analyses
iii) Overview of Highway management
INTRODUCTION
Appraisal of the funds being spent on the Highway network is
essential to ensure the most effective use of resources. As such,
an understanding of the appraisal methodology and economic
analyses carried out in HDM-4 is key to effective investment
appraisal. However, phrases such as “investment appraisal” and
economic analyses” are not naturally of interest to many
engineers. Hence the relevance to their work needs to be made
to obtain their interest.
It is important to draw out the experience of the trainees and it
may be useful to ascertain the background of the trainees
(engineering / economics/planning). This could assist is
selecting the right words to motivate the trainees.
DEVELOPMENT
The purpose and scope of the Unit should be described together
with the order and duration of various parts and practical
sessions. Indicative parts and timings are as follows:
SUMMARY
Separate summaries and reporting back should be invited for at
the end of each presentation / practical session.
Visual Aids:
5. PowerPoint presentation (15-20) slides including photos)
Case studies/Practicals:
1. HDM-4 Case Study
TRAINING FACILITIES
PowerPoint projector, white board/flip chart, overhead
projector/slide projector (optional) and computers.
MISCELLANEOUS
OBJECTIVES:
The trainee will
1. Explain the contribution of road user effects to investment
appraisal, including costs and benefits.
2. Describe the main modelling approach to RUE in HDM-4
and the use of unit costs
3. Define the key components of RUE and their relative
effects on total RUE
4. Recall the location of input parameters for RUE in the
vehicle fleet
5. Describe the speed input requirements for the speed model
6. Recall impact of traffic congestion / traffic interaction on
RUE (fuel, tyres)
7. Describe the impact on tyre consumption, parts and labour
costs
8. Explain the options for calculation of utilisation and service
life
9. Describe the way in which HDM-4 deals with vehicle
depreciation and interest
10. Describe the effect of travel time components and their
contribution to RUE
11. List the effects of non-motorised traffic.
RELATED UNITS/MODULES
i) Operation of HDM-4 / Modelling framework
ii) HDM-4 Analyses
iii) Overview of Highway management
iv) Traffic effects
v) Vehicle classification
Are costs likely to be the same for all vehicle types? This is
aimed to draw out ideas on different costs (and hence) benefits
to transporters or buses compared to car owners and the idea of
commercial use and value of time.
DEVELOPMENT
The purpose and scope of the Unit should be described together
with the order and duration of various parts and practical
sessions. Indicative parts and timings are for short or long
sessions are:
Time costs
The importance of travel time in total RUE should be stressed.
Effects of NMT
These can be described. In addition to having an impact on MT
in terms of congestion, RUE are estimated for NMT in a similar
manner to those for MT. However, many of the models are
conceptual and need further validation.
RESOURCE MATERIALS
Hardcopy (including Manuals):
1. HDM-4 documentation, Volume 4, Part E
2. HDM-4 documentation, Volume 5, Appendix E
3. HDM-4 documentation, Volume 7 (Draft).
4. PowerPoint presentations as additional handouts.
Visual Aids:
1. Short PowerPoint presentation (10 slides including photos
per section)
2. Long PowerPoint resource (As required)
Case studies/Practicals:
1. HDM-4 Case Study
TRAINING FACILITIES
PowerPoint projector, white board/flip chart, overhead
projector/slide projector (optional) and computers.
MISCELLANEOUS
OBJECTIVES
1. etc
RELATED UNITS/MODULES
i) etc
INTRODUCTION
DEVELOPMENT
SUMMARY
RESOURCE MATERIALS
Visual Aids:
1. etc.
Case studies/Practicals:
1. etc.
TRAINING FACILITIES
MISCELLANEOUS
RELATED UNITS/MODULES
1. Information Strategies
2. Modelling Framework
3. Modelling RDWE & RUE
4. Calibration of RDWE and RUE
5. Software Structure, Installation and Operation
6. Road Networks and Vehicle Fleets
7. Implementation of HDM-4
INTRODUCTION
DEVELOPMENT
RESOURCE MATERIALS
Visual Aids:
1. PowerPoint Slides (HDM-4 Configuration.PPT)
Case studies/Practicals:
1. HDM-4 Case Studies Workspace
TRAINING FACILITIES
MISCELLANEOUS
OBJECTIVES
The trainees will:
1. Describe the methods of data collection used in their
countries.
2. Describe the database management system used to store
road network inventory and condition attributes.
3. Describe the methods of data exchange between HDM-4
and road information databases.
4. Define the decision support systems required by road
managers.
5. Identify the role of HDM-4 as a decision support system for
road managers.
6. Define the management information that can be produced
using HDM-4.
RELATED UNITS/MODULES
INTRODUCTION
This unit puts into practical context the role that HDM-4 plays
within a road management system. HDM-4 is essentially a
decision support tool that can be used to process data collected
about the road network, the environment, the traffic, and other
road users in order to produce pertinent management
information. HDM-4 does not constitute a road management
system on its own. The above points need to be stressed during
the delivery of this unit.
DEVELOPMENT
• Data collection
• Database management system
• Decision support system
• Management information
RESOURCE MATERIALS
Visual Aids:
1. PowerPoint Slides (HDM-4 Implementation.PPT)
Case studies/Practicals:
1. HDM-4 Default Case Studies.
TRAINING FACILITIES
MISCELLANEOUS
OBJECTIVES
RELATED UNITS/MODULES
1. Overview of HDM-4
2. HDM-4 Software Structure and Installation
3. HDM-4 Configuration
4. Road Networks and Vehicle Fleets
5. Road Work Standards
6. HDM-4 Project Analysis Application
INTRODUCTION
DEVELOPMENT
SUMMARY
RESOURCE MATERIALS
Visual Aids:
1. PowerPoint Slides (T3 Modelling Framework.PPT)
Case studies/Practicals:
1. HDM-4 default Case Studies
TRAINING FACILITIES
MISCELLANEOUS
OBJECTIVES
RELATED UNITS/MODULES
1. Overview of HDM-4
2. Software Structure, Installation and Operation
3. HDM-4 Configuration
4. Road Networks and Vehicle Fleets
5. Road Work Standards
6. HDM-4 Programme Analysis Application
7. HDM-4 Strategy Analysis Application
INTRODUCTION
The first step in the procedure would be to calibrate the road sections
Those Trainers who complete the above exercise within a short time
could attempt to derive the optimum grading and regravelling
frequencies for the Gravel road that is included in the Road Network
database.
DEVELOPMENT
SUMMARY
RESOURCE MATERIALS
Visual Aids:
1. PowerPoint slides
Case studies/Practicals:
1. HDM-4 Default Case Studies.
TRAINING FACILITIES
MISCELLANEOUS
OBJECTIVES
RELATED UNITS/MODULES
1. Modelling Framework
2. Software Structure and Installation
3. HDM-4 Configuration
4. Road Networks and Vehicle Fleets
5. Road Work Standards
6. HDM-4 Project Analysis Application
INTRODUCTION
This is a key unit that introduces the HDM-4 life cycle analysis
concept and its applications within road management. Life cycle
analysis involves modelling the future performance of roads together
with the impact of road characteristics on user costs. Various titles are
often used to describe the same concept; whole life cycle, total
transport costs, road investment appraisal, etc. It is important that
trainers should stress that these all refer to a common modelling
framework.
With regard to the HDM-4 applications (i.e. use of HDM-4 for specific
purposes), the three applications are; Strategy, Programme and Project
analyses. The three applications differ in terms of scope and data
detail. Strategy analysis deals with an entire road network with input
data that can be at an aggregate level (e.g. good/fair/poor pavement
DEVELOPMENT
SUMMARY
RESOURCE MATERIALS
Visual Aids:
1. PowerPoint slides (T3 Role of HDM-4.PPT)
Case studies/Practicals:
1.
TRAINING FACILITIES
MISCELLANEOUS
OBJECTIVES
RELATED UNITS/MODULES
1. Overview of HDM-4
2. Software Structure, Installation and Operation
3. HDM-4 Configuration
4. Road Networks and Vehicle Fleets
5. Road Work Standards
6. Investment Appraisal
INTRODUCTION
DEVELOPMENT
SUMMARY
RESOURCE MATERIALS
Case studies/Practicals:
1. HDM-4 Default Case Studies.
TRAINING FACILITIES
MISCELLANEOUS
OBJECTIVES
RELATED UNITS/MODULES
1. Overview of HDM-4
2. Software Structure, Installation and Operation
3. HDM-4 Configuration
4. Road Networks and Vehicle Fleets
5. Road Work Standards
6. HDM-4 Programme Analysis Application
7. HDM-4 Strategy Analysis Application
INTRODUCTION
The first step in the procedure would be to calibrate the road sections
Those Trainers who complete the above exercise within a short time
could attempt to derive the optimum grading and regravelling
frequencies for the Gravel road that is included in the Road Network
database.
DEVELOPMENT
SUMMARY
RESOURCE MATERIALS
Visual Aids:
1. PowerPoint slides
Case studies/Practicals:
1. HDM-4 Default Case Studies.
TRAINING FACILITIES
MISCELLANEOUS
OBJECTIVES
RELATED UNITS/MODULES
1. Overview of HDM-4
2. Software Structure, Installation and Operation
3. HDM-4 Configuration
4. Investment Appraisal
INTRODUCTION
Road Networks and Vehicle fleets are the key data items
required in all HDM-4 applications. Road Networks can be
defined as a group of road sections with similar functional or
design characteristics, or are in the same geographic region.
Examples of road networks include; Trunk (or Main) roads,
Feeder roads, Paved roads, Unsealed roads, Expressways,
Northern Province roads, etc. In the context of a country’s road
network, these could be sub-networks. HDM-4 does not impose
any rules on the definition of a road network – it is up to the
user to decide which road sections should be included in the
network.
DEVELOPMENT
SUMMARY
RESOURCE MATERIALS
Visual Aids:
1. PowerPoint Slides (RoadNet & VehFleets.PPT)
Case studies/Practicals:
1. HDM-4 Default Case Studies.
TRAINING FACILITIES
MISCELLANEOUS
OBJECTIVES
RELATED UNITS/MODULES
1. Overview of HDM-4
2. HDM-4 Configuration
3. Road Networks and Vehicle Fleets
4. Road Work Standards
5. HDM-4 Programme Analysis Application
6. HDM-4 Strategy Analysis Application
INTRODUCTION
It is important that Trainees are able to properly install and operate the
HDM-4 software. They should also appreciate the licensing conditions
under which the HDM-4 software is supplied.
DEVELOPMENT
SUMMARY
RESOURCE MATERIALS
Visual Aids:
1. PowerPoint Slides (Software Installation and Operation.PPT)
Case studies/Practicals:
1. HDM-4 Case Studies Workspace
TRAINING FACILITIES
MISCELLANEOUS
OBJECTIVES
RELATED UNITS/MODULES
1. Overview of HDM-4
2. Software Structure, Installation and Operation
3. HDM-4 Configuration
4. Road Networks and Vehicle Fleets
5. Road Work Standards
6. HDM-4 Programme Analysis Application
INTRODUCTION
DEVELOPMENT
SUMMARY
RESOURCE MATERIALS
Visual Aids:
1. PowerPoint Slides (Strategy.PPT)
Case studies/Practicals:
1. HDM-4 Default Case Studies.
TRAINING FACILITIES
OBJECTIVES
RELATED UNITS/MODULES
1. Overview of HDM-4
2. Software Structure, Installation and Operation
3. HDM-4 Configuration
4. Creating Road Networks and Vehicle Fleets
5. Investment Appraisal
INTRODUCTION
DEVELOPMENT
SUMMARY
Visual Aids:
1. PowerPoint Slides (Work Standards.PPT)
Case studies/Practicals:
1. HDM-4 Default Case Studies.
TRAINING FACILITIES
MISCELLANEOUS
OBJECTIVES
1. etc
RELATED UNITS/MODULES
i) etc
INTRODUCTION
DEVELOPMENT
SUMMARY
RESOURCE MATERIALS
Visual Aids:
1. etc.
Case studies/Practicals:
1. etc.
TRAINING FACILITIES
MISCELLANEOUS
• In what way could traffic affect the road and hence road
management? Ask what would happen to the road of there
were no vehicles. This should aim to draw out ideas on the
aspects of vehicles which could affect road deterioration
such as weight and axle loading, as well as volume.
DEVELOPMENT
The session could start by reiterating the key uses of traffic data
in an investment appraisal tool such as HDM-4. This aims to put
The key parameters for traffic related to the road sections are
volume, composition and growth and these should be explained
in the context of their location in the road network files. It
should be explained that traffic volume is specified as AADT
and is specified separately for MT and NMT. The definition of
AADT should be clear, as it affects many aspects of the
analysis. The input of traffic growth is very flexible with ability
to specify values for different time periods and by different
calculation methods, such that all conventions should be
satisfied. The input of composition is represented simply as a
percentage of the volume.
Each road section must nominate a Speed Flow type and Traffic
flow pattern, both of which are specified in the Configuration
section. These can be user defined, although there are some
defaults available. The traffic flow pattern is derived from an
hourly flow distribution and the calculations of this can be
described.
Visual Aids:
1. Short PowerPoint presentation (15-20 slides including
photos)
2. Long PowerPoint resource (As required)
Case studies/Practicals:
1. HDM-4 Case Study
TRAINING FACILITIES
DEVELOPMENT
The overall scope of the Unit and the order and duration of the
various parts and practical sessions explained. Indicative parts
and timings are as follows:
SUMMARY
RESOURCE MATERIALS
Visual Aids:
Short Powerpoint presentation (10-15 slides including photos)
Long Powerpoint resource (As required)
Case studies/Practicals:
1. HDM-4 Project Case Study No 1 – Upgrading an unsealed
road.
2. HDM-4 T3 Extended Case Study No 1 – Optimum timing
of upgrading and effect of variation in grading frequency on
performance.
TRAINING FACILITIES
MISCELLANEOUS
RELATED UNITS/MODULES
i) Operation of HDM-4 / Modelling framework
ii) HDM-4 Analyses
iii) Traffic Effects
iv) Road User Costs
The session could start by reiterating the key uses of traffic and
vehicle data in an investment appraisal tool such as HDM-4.
This could be done by asking some questions (see above) and
aims to put the value of vehicle information into context.
Brief descriptions of the use of the data can be given but the
trainer should avoid detail as this will be covered in the Road
User costs session. Broad categories such as definition, physical
data, tyre data, utilisation, loading, and economic and financial
costs, including time values should be used.
RESOURCE MATERIALS
Visual Aids:
1. Short PowerPoint presentation (15-20 slides including
photos)
Case studies/Practicals:
1. HDM-4 Case Study for navigation exercise
2. Vehicle type input data sheets
TRAINING FACILITIES
PowerPoint projector, white board/flip chart, overhead
projector/slide projector (optional) and computers.
MISCELLANEOUS
Background
There are often situations when the budget available for projects (e.g. in the road sub-
sector) will not be sufficient to undertake all projects shown to have a positive return (i.e.
projects with positive NPV or IRR > MARR). In such situations, a formal method of
selecting projects to be included within the budget can be applied.
One such method is to apply the NPV decision rules in capital budgeting. The NPV rule
states that projects with the highest NPV should be given highest priority as they provide
the largest benefit to society.
Capital budgeting or rationing can be applied to a group of projects which meet either of
the following conditions:
(i) projects that are independent of each other (e.g. list of road projects from different
parts of the country),
(ii) mutually exclusive projects (i.e. projects that are alternatives to each other). Only
one alternative can be selected.
The NPV capital budgeting rules can be applied in both situations where sufficient funds
are available, and also when there is a budget constraint. The rules are summarised below:
(ii) Select mutually exclusive project alternatives with the highest NPV
(ii) Select mutually exclusive projects using the incremental NPV/Cost method
described below.
Example 1
A total budget of $1.5 million is given for the following projects. Use the Capital
Rationing methods to select the group of projects which will maximise the NPV.
Solution 1
The Budget limit of $1.5 million could be exhausted by selecting projects with the highest
NPV, i.e. G, F & E. The total NPV for these would be $ 225,000.
If the NPV/Cost ranking is used instead, projects B, A, C, G & F would be selected giving
a total of $ 275,000.
Incremental Analysis
The incremental analysis is used to test whether the ratio of the increase in NPV to the
increase in costs between alternative projects (mutually exclusive) is greater than a
specified marginal ratio. The formula is defined below:
If the above ratio is greater than a specified value, then the project alternative is included
among those to be funded.
Example 2
The table below summarises the capital costs (C) and Net Present Values (NPV) of 8
independent road projects, one of which has 2 mutually exclusive alternatives. If the total
budget available is $ 180,000, select the group of projects will maximise the total NPV.
In a capital rationing situation, the choice between independent projects is made on the
basis of the NPV/Cost ratio, and the choice between mutually exclusive alternatives is
made by comparing the incremental NPV/Cost ratio against the 'marginal' NPV/Cost ratio
of other competing independent project.
Working down the ranking order until the budget of $180,000 is exhausted, we would
first select projects C, A, H and E. For project H we need to decide between alternatives 1
or 2; Although H2 has the higher NPV, H1 has a higher NPV/Cost ratio. The incremental
NPV/Cost ratio for choosing H2 instead of H1 would be calculated as below:
= 0.0915
The above is then compared against the 'marginal' NPV/Cost ratio given by the next
projects in the NPV/Cost ranking order which can be carried out with the remaining
budget. We should therefore select project H2 because it has a higher incremental
NPV/Cost ratio when compared to that of projects D or G. Note that projects D and G are
the only ones which could utilise the $20,000 balance from the budget if projects C, A,
H1 and E are selected (totalling $160,000). Project G in this case provides the 'marginal'
NPV/Cost ratio of 0.0600 which is less than the incremental NPV/Cost ratio calculated
for selecting H2 instead of H1.
The selection of projects C, A, H2 and E gives the highest total NPV than any other
possible combination within the specified budget of $180,000.
ECONOMIC ANALYSIS
Questions & Decisions 1.
Is the p
project
j justified
j ?- Are benefits
greater than costs?
Which is the best investment if we have a
set of mutually exclusive alternatives?
If funds
f d are limited,
li it d how
h should
h ld different
diff t
schemes be ranked?
When should the road be built?
Questions & Decisions 2.
Management
g
Labour
Equipment
Materials
Land
Primary Effects 1
z overheads
Changes
g in road maintenance costs
Changes in accident rates
I
Increased
d travel
t l
Environmental effects
Change in value of goods moved
Secondary Effects
Due to changes
g in:
z Agricultural output
z Services
z Industrial output
z Consumer behaviour
z Land
a d values
a ues
Coverage & Double Counting
Cost +
Additional Benefits from
C1 Generated Traffic
C2
Demand Curve
T1 T2 Traffic
Estimating Benefits
VOC data relates to each road section and its condition at the time
Economic Decision Criteria
Fuel Uncongested
g Vehicle
T
Travel
l Ti
Time F t lit
Fatality
Consumption Travel Time Emissions
Parts
Consumption
Labour Hours
Depreciation
Interest
Overheads
Key Changes from HDM-III
z MT Vehicle
V hi l operating
ti costs
t (VOC)
z MT Travel time costs (TTC)
z NMT Time and operating costs (NMTOC)
z Accident costs (AC)
Ï Speed
p change
g cycle
y (Acceleration
( noise))
Ï Vehicle operating costs
Ï Accident costs
Mechanistic Models
Calculates:
z aerodynamic resistance
vel and
(Fa) D ire c tio n of Tra
Accele
ration
Driver Gradient
Vehicle Roughness
Vehicle Speed
Other Factors
Speed Model
z Desired speeds
Ï User specified for 2
2-lane
lane road
Ï Adjusted for Width, Friction, NMT, Speed Limit
z Free speeds
Ï Uphill (steady state)
Ï Downhill (steady state)
z Operating Speeds
Ï Calculated for traffic flow periods
Ï Adjusted for Speed
Speed-Flow
Flow effects
Desired Speeds
VDES = f (VDES2, CW)
VDESIR0 = VDES*XFRI*XNMT
VDESIR = min(VDESIR0,PLIMIT*ENFAC/3.6)
z VDES2 = Specified desired speed on a 2- lane road
z VDES = Calculated desired speed
z VDESIR0 = Adjusted for NMT & side-friction
z VDESIR = Adjusted for speed limit enforcement
z PLIMIT = Posted speed limit in km/h
z ENFAC = Enforcement factor (1.10)
z XFRI = Roadside friction factor
z XNMT = Non-motorised transport factor
Effect of Width on Speed
VDESIR' in m/s
a3
1m
a2
1m
a1
a0
Hoban et al
Hoban, al. (1984) HDM-Q Pan (1995) PhD Thesis
Constrained by:
y
Ï Desired Speed (VDESIR)
Ï Drive Power (VDRIVE)
Ï Braking Power (VBRAKE)
Ï Road Curvature (VCURVE)
Ï Riding Quality (VROUGH)
Speed Model Form
Model form:
⎛ σ2 ⎞
exp⎜ ⎟
⎝ 2⎠
VSS = β
⎡ 1 1 1 1 1
⎤
⎛
⎢⎜ 1 ⎞ β ⎛ 1 ⎞ β ⎛ 1 ⎞ β ⎛ 1 ⎞ β ⎛ 1 ⎞ β
⎟ +⎜ ⎟ +⎜ ⎟ +⎜ ⎟ +⎜ ⎟ ⎥
⎢⎝ VDRIVE ⎠ ⎝ VBRAKE ⎠ ⎝ VCURVE ⎠ ⎝ VROUGH⎠ ⎝ VDESIR ⎠ ⎥
⎣ ⎦
Examples of Speed Predictions
80
60
50
S pe e d in m /s
30
V DES IR - Des ired Speed
20
10
0
-10 -8 -6 -4 -2 0 2 4 6 8 10
Gra die nt in pe r ce nt
Traffic Interactions
Uncongested
Congested
0
Acceleration in m/s/s
Acceleration Noise
ARRB ARFCOM
model TOTAL POWER
power usage:
IFC = max(MinFuel, ξ Ptot)
Model Parameters
180
160 India-1
India-2
140 India-3
onsumption in L/1 000km
Caribbean
K eny a
120
100
80
Fuel C o
60
40
20
0
0 20 40 60 80 100 120
Speed in km/h
Implications of New Model
1.2
1.0
0.8
dFUEL 0.6
0.4
0.90
0.2
0 75
0.75
0.60
Acce le ra tion Noise
0.45 in m /s/s
0.0
0.30
10
20
0.15
30
40
50
60
70
0.00
80
0
7
90
S pe e d in km /h
100
Tyre Consumption
Tread wear
z amount of the tread worn due the
mechanism of the tyre coming into contact
with
ith the
th pavementt surface
f
Carcass wear
z combination
bi ti off fatigue
f ti and
d mechanical
h i l
damage to the tyre carcass - affects
number of retreads
Factors Influencing Tyre Consumption
Vehicle
Rubber
Tyre Load
Composition
Applied Force
Properties
Suspension Type
Tyre
Pavement Texture
Type/Construction
yp
Microtexture Tread Pattern
Macrotexture Aspect Ratio
Inflation Pressure
Tread Wear
Energy Per
Rate Per Unit
Unit Distance
of Energy
T
Temperature
t T
Temperature
t
Air Air
Road Road
Tyre Tyre
MECHANISTIC EMPIRICAL
Detailed models Usually aggregate
Relate tyre models
consumption to Based on fleet survey
fundamental data
equations of motion
Developed from
controlled
experiments
Retreads
9.0
8.0
7.0
6.0
5.0
dTYRE
4.0
3.0
2.0
0.90
0.75
1.0
0.60
0.45 Acce le ra tion Noise
0.0 in m /s/s
0.30
10
20
0.15
30
40
50
60
70
0.00
80
0
7
90
S pe e d in km /h
100
Parts and Labour Costs
1.6
m .
hicle Price/1000 km
1.4
1.2
PC and LDV
Parts Consumpttion as % New Veh
1.0
0.8
MT
0.6
HT
AT
0.4
0.2 HB
0.0
0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0
4.0
3.8
3.4
IRI0 = 3.5
3.2
3.0
IRI0 = 3.25
2.8
IRI0 = 3.1
2.6
24
2.4
2.2
2.0
2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 3.8 4.0
Actual Roughness in IRI m/km
Parts Model Parameters
HDM-III
Contained three utilisation methods:
z Constant Kilometreageg
z Constant Hours
z Adjusted
j Utilisation
Contained two service life methods:
z Constant Service Life
z de Weille’s Varying Service Life
Utilisation and Service Life
RUN(OL)
Costs per year
C
OL
Vehicle Age in years
Application of OL Method
100
90
80
Optimal Life as Percentage of Baseliine Utilisation .
70
60
50
40
30
20
10
0
0 2 4 6 8 10 12 14 16 18 20
Roughness in IRI m/km
Capital Costs
HDM-III
Used a simple linear model for depreciation
Affected
Aff t d by
b operating
ti conditions
diti through
th h
the effects of speed on utilisation and
speed on service life (de Weille’s method)
Depreciation in HDM-4
PC LT MT
HT AT LB
5
ht/km .
MB HB MC
Deprreciation Cost in Bah
0
0 5 10 15 20 25
Roughness in IRI m/km
Interest Costs
Linda Parsley
y
Overview of Presentation
pavement deterioration
p
vehicle operating costs
t
travel
l time
ti
quantities of vehicle exhaust emissions
calculation of energy use
economic analysis
Location of input traffic data
Specification of Alternatives:
Improvement
Impro ement standard
Configuration
Traffic Categories
Normal traffic
z Existing traffic and expected growth on the
road both with and without the investment
Diverted traffic
z Traffic diverted to the road as a result of
th investment
the i t t
Generated traffic
z N
New ttraffic
ffi iinduced
d d by
b the
th investment
i t t
Representative Vehicles
Purpose:
z to group vehicles with similar
characteristics to simplify input
Method of grouping:
z composite vehicle - average characteristics
of vehicles within the group
z actual
t l vehicle
hi l - actual
t l characteristics
h t i ti off
most common vehicle of the group
Vehicle Hierachy
Three levels:
Categories
z motorised and non-motorised traffic;
Classes e.g.
z passenger cars, trucks, buses, carts….
Types e.g.
z small car, medium car, large car, Toyota
Corolla, Mercedes Benz…...
Motorised traffic
Motorised
M di
Medium car Li h goods
Light d H
Heavy truck
k Li h bus
Light b
Category Coach
Class
Type
Non-Motorised traffic
Non-Motorised
Freight
(Commercial)
Freight
Category (Private)
Class
Type
Volume, composition and growth rates
z A t l AADT
Actual
PCU/PCE
z consider two factors:
z space occupied by vehicle
z speed effects
Length Gap
Space (m)
HDM-4 Speed-Flow Model
S3
S2
S1
Snom
Sult
Speed
Qo Flow in PCSE/h Qnom Qult
Km/h
Speed-flow model requirements
Requires speeds
z Free speed (calculated)
z Snom - Speed at nominal capacity (calculated)
z S lt
Sult - Speed
S d att ultimate
lti t capacity
it
Requires capacities
z Q0 - Volume where interactions affect speeds
z Qnom - Nominal capacity
z Qult - Ultimate capacity
specified in Configuration
Types of Capacities
Nominal Capacity
z the capacity where the traffic stream is
travelling at a constant speed
z sustainable over long time periods
Ultimate Capacity
z maximum possible capacity before flow
breaks down
Annual Distribution of Hourly Flows
Flow Periods
Peak
ow Next to Peak
r Medium flow
N t tto L
Next Low
Overnight
N mber of Ho
Number Hours
rs in the Year
Non-Motorised Traffic
Impact of NMT
NMT affect
z Motorised traffic flow and speed
z Motorised traffic operating costs
NMT also
l affect
ff t
z Safety related costs
z Road maintenance costs
these are not considered in HDM-4
Other effects on speeds
Specification of Alternatives:
Improvement standard
z Diverted traffic / generated traffic
Configuration
z speed flow types, traffic flow patterns
Summary
1 Background
This case study demonstrates the use of HDM-4 in budget forecasting. The objective of this
study is to determine the required funding levels for user defined network performance
standards, and to demonstrate the effect of budgetary constraints on the long term
performance trends. This involves defining the road network in terms of representative
sections and assigning alternative maintenance standards (investment alternatives) to each
section.
Strategy analysis is normally undertaken at infrequent intervals and is primarily a tool for
resource allocation by, or for, the top managers within a road agency. It is to be used for the
analysis of a chosen road network as a whole to prepare long-range planning estimates of
expenditure needs for road development and preservation under various budget scenarios.
Strategy analysis will be required to produce expenditure estimates for medium to long term
periods of between 5 to 40 years, under various budget constraints, maintenance standards,
and road use scenarios. It will predict key performance indicators of the road network such as
average roughness trends, surface distress, road user costs, safety levels and environmental
effects.
When conducting a strategy analysis, the physical road network may be categorised into a
matrix defined by user-defined parameters such as road class, surface type, pavement
condition or traffic flow. Strategy analysis utilises a model of the road network (or sub-
network) to simulate network-wide performance under specified road use characteristics,
maintenance standards, development standards and resource constraints.
The main difference between strategy and programme analysis is that programme analysis
deals with physical links and sections that can be identified from the road network. In strategy
analysis, the road system essentially loses its individual link and section characteristics as
these are grouped into representative sections to model the road network being analysed.
Thus, whereas in programme analysis the basic unit of the network is an identifiable road
section, the basic unit in strategy analysis is a representative section. Since sections are the
basic units of analysis in HDM-4 for both applications, similar types of economic analysis and
optimisation methods can be utilised for both modules.
Road Network
The road network has been modelled using eleven representative sections. The representative
sections for bituminous (Paved) roads have been based on traffic volume (High, Medium or
Low) and road condition (Good, Fair and Poor), giving a total of nine sections.
Unsealed (Gravel) roads have been classified by traffic volume alone, since condition may
change rapidly in any given year. For this case study, no gravel sections are subjected to high
traffic, so the gravel network is represented by two sections only, one for Medium traffic and
one for Low traffic.
Double-click on a Section Description (or selecting the section and clicking the View/Edit
Section button) gives access to the section's details. Some relevant characteristics of the
representative sections are summarised in Table C1.1and Table C1.1. Note that, for both
bituminous and gravel sections, the condition must be defined at the start of the analysis
period. As the analysis period commences in the year 2000, this means that the condition data
must be defined for the year 1999 (that is, end of 1999) or earlier.
Table C1.1
Details of representative sections for gravel roads
Condition (1998)
Gravel thickness (mm) 100 70
A tabulated data summary is useful in checking consistency of data across sections and also
within a section. This includes checking that the condition classification is compatible with
the condition parameters, and, with bituminous pavements, that the pavement type (defined
under Section/General) is compatible with the pavement details (surfacing material type,
surfacing thicknesses and dates of previous works) defined under Section/Pavement.
Remember that the dates of previous works refer to HDM-4 Works Types. Hence the date of
last resurfacing refers to Works Type resurfacing which includes works activities such as
surface dressing and slurry seal. Works activities such as overlaying, mill and replace, and
inlay are classed as Works Type rehabilitation and would be recorded under date of last
rehabilitation. The bituminous pavement types are reviewed below.
Table C1.1
Details of representative sections for bituminous roads
Section ID PHTGC PHTFC PHTPC PMTGC PMTFC PMTPC PLTGC PLTFC PLTPC
Length (km) 234 392 437 306 483 615 410 670 720
Traffic level High High High Medium Medium Medium Low Low Low
AADT 6200 5240 5180 2500 2300 2060 1400 1150 970
AADT year 1998 1998 1998 1998 1998 1998 1998 1998 1998
Condition (1999) Good Fair Poor Good Fair Poor Good Fair Poor
Roughness (IRI) 2.5 4.0 6.5 2.5 4.4 6.5 3 4.5 6.1
Skid resistance (SCRIM 50 kph) 0.7 0.5 0.4 0.7 0.5 0.4 0.7 0.5 0.3
Drainage Excellent Fair Poor Excellent Good Fair Good Fair Poor
Pavement type AMAP AMAP AMAP STGB STGB STGB STGB STGB STGB
Section ID PHTGC PHTFC PHTPC PMTGC PMTFC PMTPC PLTGC PLTFC PLTPC
Date of last works types:
- reconstruction 1991 1991 1991 1991 1991 1991 1991 1991 1991
- rehabilitation (overlay) 1996 1996 1996 1991 1991 1991 1991 1991 1991
- resurfacing (reseal) 1996 1996 1996 1996 1996 1991 1991 1991 1991
- preventative treatment 1996 1996 1996 1996 1996 1991 1991 1991 1991
SN after last treatment 2.70 2.70 2.70 2.07 2.07 2.26 1.67 1.67 1.67
Key:
Pavement type Traffic class Pavement Condition
P = Paved HT = High Traffic GC = Good Condition
G = Gravel MT = Medium Traffic FC = Fair Condition
LT = Low Traffic PC = Poor Condition
Vehicle Fleet
The vehicles included in the analysis are selected from the pre-defined Western Province
fleet. Individual vehicle attributes may be reviewed by double-clicking on the appropriate
vehicle type description (or by selecting the section and clicking the Edit Vehicle button).
Normal Traffic
The initial traffic composition and growth rates (by vehicle type) can be reviewed by double-
clicking the section description. The initial vehicle compositions are summarised in Table
C1.3. Note that the initial composition is defined for same year as the AADT. The annual
growth rates may be specified for several periods, but must at least be specified for a period
commencing from the analysis period start year (that is, 2000 in this case study). For example
the AADT for section PHTGC is specified for 1998, and the initial vehicle composition
applies to that year. The annual growth rates have been defined only for the period
commencing 2000, and in this case are also applied between 1998 and 2000.
Alternatives
For this case study, four investment alternatives are considered for each of the representative
sections, ranging from the provision of routine pavement maintenance only (Alternative 1) to
an ideal maintenance case (Alternative 4). For each investment alternative, Road Works
Standards have been pre-defined as shown in Table C1.2 and Table C1.4.
Table C1.2
Assigned Road Works Standards - short codes
Gravel (unpaved)
1. Grade & Spot regravel G1 G1
With bituminous road sections, maintenance standards have been assigned by alternative and
traffic class. For gravel road sections, a maintenance standard has been defined for each
investment alternative, and an improvement standard has been introduced to Alternative 4
(that is, upgrading gravel roads with medium traffic followed by paved road maintenance, and
widening gravel roads with low traffic). Note that Road Authorities may wish to assign
standards by Road Class, which may be represented in some countries by Surface Class alone.
With condition responsive maintenance intervention, the interval between maintenance
applications will clearly be influenced by the traffic volume.
The definition of investment alternatives for each of the representative sections is indicated in
Table C1.4. This shows the short code identifier associated with each of the pre-defined Road
Works Standards (that is, Maintenance Standards and Improvement Standards).
Maintenance Standards for bituminous sections
The Maintenance Standards for bituminous sections are reviewed first. These are based on
different combinations of the following works activities/operations.
If more than one of these activities is triggered in a given year, the works item that is highest
in the hierarchy of road works (that is, with lowest ranking value) will be applied. Note that
the routine pavement works (patching and crack sealing) have the same ranking, and both can
be applied in the same maintenance year. The hierarchy is defined by the ranking of road
works shown in Chapter D2 of the Analytical Framework and Model Descriptions.
Table C1.3
Details of initial vehicle composition and growth rates
Traffic level
Leyland bus 10 3 10 3 0 0
Mercedes heavy truck 10 3 25 3 10 3
Nissan pickup 15 3 20 3 20 3
Tata 7.5t lorry 15 3 0 0 10 3
Toyota Corolla car 40 3 30 3 40 3
Toyota Hiace minibus 10 3 15 3 20 3
Total (%) 100 100 100
Note: Traffic details (AADT & associated year, initial composition and associated growth rates) are assigned by section.
Table C1.4
Definition of investment alternatives for representative sections
2 PHT2 - -
PHTGC B H
3 PHT3 - -
4 PHT4 - -
1 PHT1 - -
2 PHT2 - -
PHTFC B H
3 PHT3 - -
4 PHT4 -
1 PHT1 - -
2 PHT2 - -
PHTPC B H
3 PHT3 - -
4 PHT4 - -
1 PMT1 - -
2 PMT2 - -
PMTGC B M
3 PMT3 - -
4 PMT4 - -
1 PMT1 - -
2 PMT2 - -
PMTFC B M
3 PMT3 - -
4 PMT4 - -
1 PMT1 - -
2 PMT2 - -
PMTPC B M
3 PMT3 - -
4 PMT4 - -
1 PLT1 - -
2 PLT2 - -
PLTGC B L
3 PLT3 - -
4 PLT4 - -
1 PLT1 -
2 PLT2 -
PLTFC B L
3 PLT3 - -
4 PLT4 - -
2 PLT2 - -
PLTPC B L
3 PLT3 - -
4 PLT4 - -
1 G1 - -
2 G2 - -
GMT U M
3 G3 - -
4 G1 GU4 PLT3
1 G1 - -
2 G2 - -
GLT U L
3 G3 - -
4 G1 GW4 G2
Note: Details of Road Works Standards identified by the above short codes are given in Table
C1.8 and Table C1.9 for bituminous and gravel sections respectively.
Table C1.5
Works activities considered for bituminous sections
Crack sealing 22
! Alternative 1
Comprises patching and crack sealing only. This represents the base case (or do minimum)
alternative. In this case study the intervention levels for patching vary from 10 potholes/km
for sections with high traffic, to 50 potholes/km for sections with low traffic (see Table
C1.8).
! Alternative 2
Includes patching, resealing and reconstruction. The objective of this alternative is to adopt
relatively inexpensive treatments which will maintain the existing road in a reasonable
condition for as long as possible until the eventual need for reconstruction.
! Alternative 3
Includes patching, resealing, overlaying and reconstruction. This alternative combines the
benefits of resealing and overlaying within one standard, which should reduce the required
frequency of the relatively expensive overlay works.
The Maintenance Standards for alternatives 2, 3 and 4 include reconstruction as a Works Item.
Details of the pavement layer thicknesses adopted for different road classes are given in Table
C1.8
The pavement details will normally be derived using appropriate design charts, based on a design
subgrade CBR and estimated cumulative traffic loading over the desired design life. The
subgrade CBR value is not requested within the reconstruction works item, as this is already
specified at section level (Section/Pavement). Details of the Maintenance Standards associated
with each alternative are summarised in Table C1.8.
Table C1.6
Pavement reconstruction details for bituminous sections
Note: M/Design refers to the Design Tab under the appropriate Maintenance Standard
Maintenance and Improvement Standards for gravel sections
The Maintenance and Improvement Standards for gravel sections are based on combinations of
the works activities listed in Table C1.7.
Table C1.7
Works activities considered for gravel sections
Spot regravelling 7
M Resurfacing Regravelling 6
Lane addition 4
The order of ranking of the Maintenance works items gives regravelling highest priority (that is,
with lowest ranking value). Spot regravelling and Grading have equal ranking, and both of these
can be performed in a given analysis year. Details of precedence rules for road works are given in
Chapter D2 of the Analytical Framework and Model Descriptions.
! Alternative 1
Comprises grading and spot regravelling, both condition responsive. This represents the base
case (do-minimum alternative).
! Alternative 2
Includes three works items, grading, spot regravelling and regravelling (resurfacing), all
condition responsive. Grading is triggered by roughness, while spot regravelling and
regravelling are both triggered by the thickness of gravel surfacing. This alternative aims to
maintain a reasonable thickness of gravel surfacing and so provide continuous protection to
the pavement subgrade.
! Alternative 3
Includes the same works items as Alternative 2, but the intervention levels are set to provide
a higher level of serviceability that should trigger the works more frequently.
! Alternative 4
Introduces upgrading to a paved standard for sections with medium traffic (representative
section GMT), and widening for sections with low traffic (section GLT). This represents the
ideal investment alternative. Suitable maintenance is provided before and after the respective
improvements. Note that the respective Improvement Standards are effective from the year
2003, with works scheduled to start in the year 2004. This means that no maintenance will
be applied during the year 2003, the year prior to improvement.
3 Funding requirements
Table C1.8
Details of investment alternatives for bituminous sections
Investment Maintenance Standard / Condition responsive Units Intervention levels by traffic class
alternative Works item criteria
High traffic Medium traffic Low
traffic
1 Patch & crack seal (PHT1) (PMT1) (PLT1)
Pothole patching Potholing no/km >=10 >=25 >=50
Notes:
1 Short code identifiers for Maintenance Standards shown in parentheses, for example (PHT1)
2 SBSD = Single Bitumen Surface Dressing
Table C1.9
Details of investment alternatives for gravel sections
Alt Sections M/I Road Works Standard Effective Responsive Criteria [M] Details (design)
from
Works Item year Start Date/Duration [I]
Spot regravelling GThk <= 100 mm Replace 10% annual material loss
Spot regravelling GThk <= 100 mm Replace 10% annual material loss
Spot regravelling GThk <= 125 mm Replace 10% annual material loss
I GU4: Pave section GMT in 2004 2003 Start year 2004 Upgrade to STGB with 2 m widening
Cracking >= 5%
I GW4: Widen section GLT in 2004 2003 Start year 2004 Widen by 3 m
Notes:
Column 1 Alt = Investment Alternative
Column 3 M/I = Maintenance or Improvement Standard
Column 6 GThk = Gravel thickness
1 Unit Aims
1. To introduce the HDM-4 Strategy Analysis module concepts so that Trainees are able to
forecast long term road budget requirements and road network performance trends.
2. To highlight the differences between Strategy and Programme Analysis Applications of
HDM-4.
3. To familiarise Trainers with the use of HDM-4 Strategy Analysis data input and output
modules.
2 Unit Objectives
1. To describe the concepts of the HDM-4 Strategy analysis application.
2. To use the HDM-4 model to calculate a long term budget requirement
3. To alter maintenance standards and unit costs and study the impact on budget
requirements.
4. To export the Strategy Analysis results and plot performance indicators using a
spreadsheet.
3 Background
This case study demonstrates the use of HDM-4 in budget forecasting. The objective of this
study is to determine the required funding levels for user defined network performance
standards, and to demonstrate the effect of budgetary constraints on the long term
performance trends. This involves defining the road network in terms of representative
sections and assigning alternative maintenance standards (investment alternatives) to each
section.
Strategy analysis is normally undertaken at infrequent intervals and is primarily a tool for
resource allocation by, or for, the top managers within a road agency. It is to be used for the
analysis of a chosen road network as a whole to prepare long-range planning estimates of
expenditure needs for road development and preservation under various budget scenarios.
Strategy analysis will be required to produce expenditure estimates for medium to long term
periods of between 5 to 40 years, under various budget constraints, maintenance standards,
and road use scenarios. It will predict key performance indicators of the road network such as
average roughness trends, surface distress, road user costs, safety levels and environmental
effects.
When conducting a strategy analysis, the physical road network may be categorised into a
matrix defined by user-defined parameters such as road class, surface type, pavement
condition or traffic flow. Strategy analysis utilises a model of the road network (or sub-
network) to simulate network-wide performance under specified road use characteristics,
maintenance standards, development standards and resource constraints.
The main difference between strategy and programme analysis is that programme analysis
deals with physical links and sections that can be identified from the road network. In strategy
analysis, the road system essentially loses its individual link and section characteristics as
these are grouped into representative sections to model the road network being analysed.
Thus, whereas in programme analysis the basic unit of the network is an identifiable road
section, the basic unit in strategy analysis is a representative section. Since sections are the
basic units of analysis in HDM-4 for both applications, similar types of economic analysis and
optimisation methods can be utilised for both modules.
Road Network
The road network has been modelled using eleven representative sections. The representative
sections for bituminous (Paved) roads have been based on traffic volume (High, Medium or
Low) and road condition (Good, Fair and Poor), giving a total of nine sections.
Unsealed (Gravel) roads have been classified by traffic volume alone, since condition may
change rapidly in any given year. For this case study, no gravel sections are subjected to high
traffic, so the gravel network is represented by two sections only, one for Medium traffic and
one for Low traffic.
Double-click on a Section Description (or selecting the section and clicking the View/Edit
Section button) gives access to the section's details. Some relevant characteristics of the
representative sections are summarised in Table C1.1and Table C1.2. Note that, for both
bituminous and gravel sections, the condition must be defined at the start of the analysis
period. As the analysis period commences in the year 2000, this means that the condition data
must be defined for the year 1999 (that is, end of 1999) or earlier.
A tabulated data summary is useful in checking consistency of data across sections and also
within a section. This includes checking that the condition classification is compatible with
the condition parameters, and, with bituminous pavements, that the pavement type (defined
under Section/General) is compatible with the pavement details (surfacing material type,
surfacing thicknesses and dates of previous works) defined under Section/Pavement.
Remember that the dates of previous works refer to HDM-4 Works Types. Hence the date of
last resurfacing refers to Works Type resurfacing which includes works activities such as
surface dressing and slurry seal. Works activities such as overlaying, mill and replace, and
inlay are classed as Works Type rehabilitation and would be recorded under date of last
rehabilitation. The bituminous pavement types are reviewed below.
Table C1.1
Details of representative sections for bituminous roads
Section ID PHTGC PHTFC PHTPC PMTGC PMTFC PMTPC PLTGC PLTFC PLTPC
Length (km) 234 392 437 306 483 615 410 670 720
Traffic level High High High Medium Medium Medium Low Low Low
AADT 6200 5240 5180 2500 2300 2060 1400 1150 970
AADT year 1998 1998 1998 1998 1998 1998 1998 1998 1998
Condition (1999) Good Fair Poor Good Fair Poor Good Fair Poor
Roughness (IRI) 2.5 4.0 6.5 2.5 4.4 6.5 3 4.5 6.1
Skid resistance (SCRIM 50 kph) 0.7 0.5 0.4 0.7 0.5 0.4 0.7 0.5 0.3
Drainage Excellent Fair Poor Excellent Good Fair Good Fair Poor
Pavement type AMAP AMAP AMAP STGB STGB STGB STGB STGB STGB
Section ID PHTGC PHTFC PHTPC PMTGC PMTFC PMTPC PLTGC PLTFC PLTPC
Date of last works types:
- reconstruction 1991 1991 1991 1991 1991 1991 1991 1991 1991
- rehabilitation (overlay) 1996 1996 1996 1991 1991 1991 1991 1991 1991
- resurfacing (reseal) 1996 1996 1996 1996 1996 1991 1991 1991 1991
- preventative treatment 1996 1996 1996 1996 1996 1991 1991 1991 1991
SN after last treatment 2.70 2.70 2.70 2.07 2.07 2.26 1.67 1.67 1.67
Key:
Pavement type Traffic class Pavement Condition
P = Paved HT = High Traffic GC = Good Condition
G = Gravel MT = Medium Traffic FC = Fair Condition
LT = Low Traffic PC = Poor Condition
Table C1.2
Details of representative sections for gravel roads
Condition (1998)
Gravel thickness (mm) 100 70
Vehicle Fleet
The vehicles included in the analysis are selected from the pre-defined Western Province
fleet. Individual vehicle attributes may be reviewed by double-clicking on the appropriate
vehicle type description (or by selecting the section and clicking the Edit Vehicle button).
Normal Traffic
The initial traffic composition and growth rates (by vehicle type) can be reviewed by double-
clicking the section description. The initial vehicle compositions are summarised in Table
C1.4. Note that the initial composition is defined for same year as the AADT. The annual
growth rates may be specified for several periods, but must at least be specified for a period
commencing from the analysis period start year (that is, 2000 in this case study). For example
the AADT for section PHTGC is specified for 1998, and the initial vehicle composition
applies to that year. The annual growth rates have been defined only for the period
commencing 2000, and in this case are also applied between 1998 and 2000.
Alternatives
For this case study, four investment alternatives are considered for each of the representative
sections, ranging from the provision of routine pavement maintenance only (Alternative 1) to
an ideal maintenance case (Alternative 4). For each investment alternative, Road Works
Standards have been pre-defined as shown in Table C1.3 and Table C1.5.
Table C1.3
Assigned Road Works Standards - short codes
Gravel (unpaved)
1. Grade & Spot regravel G1 G1
Table C1.4
Details of initial vehicle composition and growth rates
Traffic level
Leyland bus 10 3 10 3 0 0
Mercedes heavy truck 10 3 25 3 10 3
Nissan pickup 15 3 20 3 20 3
Tata 7.5t lorry 15 3 0 0 10 3
Toyota Corolla car 40 3 30 3 40 3
Toyota Hiace minibus 10 3 15 3 20 3
Total (%) 100 100 100
Note: Traffic details (AADT & associated year, initial composition and associated growth rates) are assigned by section.
With bituminous road sections, maintenance standards have been assigned by alternative and
traffic class. For gravel road sections, a maintenance standard has been defined for each
investment alternative, and an improvement standard has been introduced to Alternative 4 (that is,
upgrading gravel roads with medium traffic followed by paved road maintenance, and widening
gravel roads with low traffic). Note that Road Authorities may wish to assign standards by Road
Class, which may be represented in some countries by Surface Class alone. With condition
responsive maintenance intervention, the interval between maintenance applications will clearly
be influenced by the traffic volume.
The definition of investment alternatives for each of the representative sections is indicated in
Table C1.5. This shows the short code identifier associated with each of the pre-defined Road
Works Standards (that is, Maintenance Standards and Improvement Standards).
Table C1.5
Definition of investment alternatives for representative sections
2 PHT2 - -
PHTGC B H
3 PHT3 - -
4 PHT4 - -
1 PHT1 - -
2 PHT2 - -
PHTFC B H
3 PHT3 - -
4 PHT4 -
1 PHT1 - -
2 PHT2 - -
PHTPC B H
3 PHT3 - -
4 PHT4 - -
1 PMT1 - -
2 PMT2 - -
PMTGC B M
3 PMT3 - -
4 PMT4 - -
1 PMT1 - -
2 PMT2 - -
PMTFC B M
3 PMT3 - -
4 PMT4 - -
1 PMT1 - -
2 PMT2 - -
PMTPC B M
3 PMT3 - -
4 PMT4 - -
2 PLT2 - -
PLTGC B L
3 PLT3 - -
4 PLT4 - -
1 PLT1 -
2 PLT2 -
PLTFC B L
3 PLT3 - -
4 PLT4 - -
1 PLT1 - -
2 PLT2 - -
PLTPC B L
3 PLT3 - -
4 PLT4 - -
1 G1 - -
2 G2 - -
GMT U M
3 G3 - -
4 G1 GU4 PLT3
1 G1 - -
2 G2 - -
GLT U L
3 G3 - -
4 G1 GW4 G2
Note: Details of Road Works Standards identified by the above short codes are given in Table
C1.9 and Table C1.10 for bituminous and gravel sections respectively.
Maintenance Standards for bituminous sections
The Maintenance Standards for bituminous sections are reviewed first. These are based on
different combinations of the following works activities/operations.
If more than one of these activities is triggered in a given year, the works item that is highest in
the hierarchy of road works (that is, with lowest ranking value) will be applied. Note that the
routine pavement works (patching and crack sealing) have the same ranking, and both can be
applied in the same maintenance year. The hierarchy is defined by the ranking of road works
shown in Chapter D2 of the Analytical Framework and Model Descriptions.
Table C1.6
Works activities considered for bituminous sections
Crack sealing 22
! Alternative 1
Comprises patching and crack sealing only. This represents the base case (or do minimum)
alternative. In this case study the intervention levels for patching vary from 10 potholes/km
for sections with high traffic, to 50 potholes/km for sections with low traffic (see Table
C1.9).
! Alternative 2
Includes patching, resealing and reconstruction. The objective of this alternative is to adopt
relatively inexpensive treatments which will maintain the existing road in a reasonable
condition for as long as possible until the eventual need for reconstruction.
! Alternative 3
Includes patching, resealing, overlaying and reconstruction. This alternative combines the
benefits of resealing and overlaying within one standard, which should reduce the required
frequency of the relatively expensive overlay works.
The Maintenance Standards for alternatives 2, 3 and 4 include reconstruction as a Works Item.
Details of the pavement layer thicknesses adopted for different road classes are given in Table
C1.8
The pavement details will normally be derived using appropriate design charts, based on a design
subgrade CBR and estimated cumulative traffic loading over the desired design life. The
subgrade CBR value is not requested within the reconstruction works item, as this is already
specified at section level (Section/Pavement). Details of the Maintenance Standards associated
with each alternative are summarised in Table C1.9.
Table C1.7
Pavement reconstruction details for bituminous sections
Note: M/Design refers to the Design Tab under the appropriate Maintenance Standard
Maintenance and Improvement Standards for gravel sections
The Maintenance and Improvement Standards for gravel sections are based on combinations of
the works activities listed in Table C1.8.
Table C1.8
Works activities considered for gravel sections
Spot regravelling 7
M Resurfacing Regravelling 6
Lane addition 4
The order of ranking of the Maintenance works items gives regravelling highest priority (that is,
with lowest ranking value). Spot regravelling and Grading have equal ranking, and both of these
can be performed in a given analysis year. Details of precedence rules for road works are given in
Chapter D2 of the Analytical Framework and Model Descriptions.
! Alternative 1
Comprises grading and spot regravelling, both condition responsive. This represents the base
case (do-minimum alternative).
! Alternative 2
Includes three works items, grading, spot regravelling and regravelling (resurfacing), all
condition responsive. Grading is triggered by roughness, while spot regravelling and
regravelling are both triggered by the thickness of gravel surfacing. This alternative aims to
maintain a reasonable thickness of gravel surfacing and so provide continuous protection to
the pavement subgrade.
! Alternative 3
Includes the same works items as Alternative 2, but the intervention levels are set to provide
a higher level of serviceability that should trigger the works more frequently.
! Alternative 4
Introduces upgrading to a paved standard for sections with medium traffic (representative
section GMT), and widening for sections with low traffic (section GLT). This represents the
ideal investment alternative. Suitable maintenance is provided before and after the respective
improvements. Note that the respective Improvement Standards are effective from the year
2003, with works scheduled to start in the year 2004. This means that no maintenance will
be applied during the year 2003, the year prior to improvement.
5 Funding requirements
Table C1.9
Details of investment alternatives for bituminous sections
Investment Maintenance Standard / Condition responsive Units Intervention levels by traffic class
alternative Works item criteria
High traffic Medium traffic Low
traffic
1 Patch & crack seal (PHT1) (PMT1) (PLT1)
Pothole patching Potholing no/km >=10 >=25 >=50
Notes:
1 Short code identifiers for Maintenance Standards shown in parentheses, for example (PHT1)
2 SBSD = Single Bitumen Surface Dressing
Table C1.10
Details of investment alternatives for gravel sections
Alt Sections M/I Road Works Standard Effective Responsive Criteria [M] Details (design)
from
Works Item year Start Date/Duration [I]
Spot regravelling GThk <= 100 mm Replace 10% annual material loss
Spot regravelling GThk <= 100 mm Replace 10% annual material loss
Spot regravelling GThk <= 125 mm Replace 10% annual material loss
I GU4: Pave section GMT in 2004 2003 Start year 2004 Upgrade to STGB with 2 m widening
Cracking >= 5%
I GW4: Widen section GLT in 2004 2003 Start year 2004 Widen by 3 m
Notes:
Column 1 Alt = Investment Alternative
Column 3 M/I = Maintenance or Improvement Standard
Column 6 GThk = Gravel thickness
1 Introduction
This case study presents the economic analysis of a project to upgrade an existing gravel road to a
paved standard. The existing road is 50 km long and passes through varying topography. For
analysis purposes, three sections, based on geometry, pavement condition, and traffic volume can
represent the road. Traffic and condition data are available from surveys undertaken in 1998. The
gravel thickness in 1998 was 150 mm.
The objective of the case study is to demonstrate the structure and capabilities of HDM-4, and the
steps needed to undertake the project appraisal.
The purpose of the appraisal is to assess the economic benefits resulting from the proposed
investment. (This differs from a financial appraisal that is concerned with the means of financing
a project and the financial profitability of the project). The economic feasibility of the project is
assessed by comparison against a base-line project alternative (that is, a without project
alternative). The project alternatives are:
! Without Project
Table A1.1
Case study 1 - Summary of section attributes
Number of lanes 2 2 2
Number of shoulders 0 0 0
Geometry (before
upgrading)
Rise and Fall (m/km) 10 15 30
Condition 1998
Gravel Thickness 150 150 150
Notes:
3 Alternatives
The two project alternatives considered in this case study are defined below. The first alternative,
Without Project, represents a continuation of current maintenance practice. The second
alternative, With Project, represents the implementation of the project to upgrade the existing
gravel road to paved standard.
! Without Project
This project alternative comprises grading every six months; spot regravelling to replace
30% of material lost each year (if the gravel thickness falls below 100 mm, and gravel
resurfacing (whenever the gravel thickness falls below 50 mm).
Note that in a given year, if gravel resurfacing is triggered, this operation supersedes grading
and spot regravelling.
! With Project
This project alternative includes upgrading the road in the year 2004 (duration of works is
two years for road sections B001-01 and B001-03, and one year for section B001-02).
Before upgrading, the existing gravel road will be maintained, by grading, every six months
and spot regravelling to replace 30% of material lost each year (if the gravel thickness falls
below 100 mm). The latter works will maintain some protection of the subgrade prior to
upgrading.
After upgrading, the road will receive routine maintenance in the form of crack sealing (if
wide structural cracking reaches 5%) and patching (if the severely damaged area reaches
5%).
The analysis period is 20 years, starting in the year 2000. Note that Analysis by project option
has been selected meaning that the economic analysis results will be combined for all three
sections in the Project.
A schematic overview of the Project Alternatives and their constituent Road Works Standards is
shown in Table A1.2. Note that Road Works Standards are sub-divided into Maintenance
Standards denoted by M , and Improvement Standards denoted by I .
The Road Works Standards and associated works assigned to each project alternative are
summarised in Table A1.2. A summary of the intervention limits for maintenance works is given
in Table A1.3. These tables allow the user to check consistency of data across alternatives.
Each Road Works Standard has an associated date, assigned at project level, which represents the
year from which the standard takes effect. For the Without Project alternative, each of the three
road sections has been assigned a Maintenance Standard named Gravel Road Maintenance.
This assignment commences in the year 2000, which is the first year of the 20-year analysis
period (2000-2019).
For the With Project alternative, each section has been assigned a set of Road Works Standards
as follows:
! M Maintenance Standard for gravel road prior to upgrading
Note that (for a given feature) only one Maintenance Standard or Improvement Standard will be
effective in any analysis year. However a Maintenance Standard may include more than one
works item which could be implemented in a given year, for example, grading and spot
regravelling on unsealed roads, crack sealing and patching on bituminous roads.
Table A1.2
Details of road works standards for each project alternative
Table A1.3
Intervention limits for maintenance works
M Gravel Road 2000 Grading S 2019 30 n/a 180 days 750 days 0 100,000
Maintenance
Spot regravelling R 2019 30 100 m3/km/yr n/a n/a 0 100,000
M Maintenance before 2000 Grading S 2003 30 n/a 180 days 10000 days 0 100,000
upgrading
Spot regravelling R 2003 30 100 m3/km/yr n/a n/a 0 100,000
M Crack sealing and 2006 Crack sealing R 2019 12.5 1500 n/a n/a 0 100,000
patching paved road 2
m /km/yr
Notes:
n/a not applicable
* S = Scheduled intervention
R = Responsive intervention
4 Project alternatives
The use of the HDM-4 software to set up the Road Works Standards for this case study is
described in more detail below. Each Project Alternative includes one or more Maintenance
Standards and/or Improvement Standards that are assigned to be effective from a specified date.
Each Maintenance Standard may define several Works Items, and each Improvement Standard
defines a specific improvement type.
Project Alternative:
Without Project
M
Gravel Road Maintenance
Project Alternative:
With Project
M M
I
Maintenance before Crack sealing and
Upgrading works
upgrading patching paved roads
The Works Items and Operations associated with the Maintenance Standard are shown in
Figure A1.1. Full details of the logical rules associated with the implementation of works are
given in Chapter D1 of the Analytical Framework and Model Descriptions.
Regravelling/
Operation/Works Activity: Grading Spot regravelling
Resurfacing
Prior to upgrading, the existing gravel road will be given routine maintenance in the form of
grading every six months and spot regravelling (if the gravel thickness falls below 100 mm). The
works are specified on the Maintenance Standard window.
For this case study, the geometric characteristics are significantly different for each road section,
as are the pavement design details and duration of the upgrading works. The different
Improvement Standards for the three sections are reviewed below. A comparison of selected
section details, before and after upgrading, is given in Table A1.4.
Table A1.4
Section details before and after upgrading works
A 7 6 8
(excluding subgrade)
(US dollars/km)
Notes:
B = Before upgrading
A =After upgrading
The Improvement Standard for this section has a different pavement design to that specified for
Section B001-01. The construction period for Section B001-02 is one year.
The Improvement Standard section B001-03 has a different pavement design (detailed under
Pavement below) to the previous sections. The construction period for Section B001-03 is two
years.
For this road section, the proposed pavement design comprises a 150 mm stabilised sub-base
placed on existing formation level with a 200 mm stabilised road base and a double surface
dressing.
5 HDM-4 results
The costs and benefits of the Upgrade gravel road alternative will be compared with those for
the Maintain gravel road alternative. Note that the HDM-4 Workspace may hold data for any
number of projects with associated network, fleet and work standards data. The run that has been
set up will consider only those project alternatives defined under 'Alternatives', with the specified
selected sections and vehicles.
For this case study, accident costs, energy balance analysis, emissions calculations, and
acceleration effects are not included in the analysis.
The pre-defined report categories available are:
! Deterioration/Work Effects
! Road User Effects
! Environmental Effects
! Cost Streams
Note that reports generally refer to data presented in tabular form. Certain reports are also
available in graphical form (Charts). For this case study, the Timing of Works (by section) report
should be printed.
Considering the Without Project alternative, the report indicates that, at Section B001-01 (Town
A-Town B), gravel resurfacing would be implemented in the years 2003, 2007, 2010, 2013 and
2016 (with associated preparatory spot regravelling). Spot regravelling would be implemented
every year except the first year and each year immediately following gravel resurfacing. Grading
is implemented every six months as scheduled. These works are specified under the Maintenance
Standard Gravel Road Maintenance described previously.
The With Project alternative for the same road section (B001-01) confirms that the upgrade to
paved standard would be implemented during 2004 and 2005. Before upgrading, the existing
road is graded at six monthly intervals between 2000 and 2003 with spot regravelling. After
upgrading, the paved road has no works specified until 2016, when patching is listed each year
between 2016 and 2019 (end of analysis period). These works are specified under the three works
standards:
! M Maintenance before upgrading
! I Pave Section B001-01 in 2004
! M Crack sealing and patching paved road
The effect of these works on roughness is indicated in the Deterioration report and associated
chart showing roughness at the end of the year.
For this case study, the overall NPV is reported as US$ 0.624 millions. The breakdown by
section indicates that two sections give a positive NPV.
(millions of US Dollars)
B001-01 Town A to Town B 0.117
Note that the traffic levels are lowest for Section B001-02 (negative NPV), also that maximum
benefits are derived for Section B001-03 which has highest traffic flow. A revised project
comprising sections B001-01 and B001-03 only would give an overall NPV of US$ 0.940
millions.
UNIT AIMS
1. To practice the specification and assignment of HDM-4 maintenance standards so that Trainers are able to
apply a range of maintenance works.
UNIT OBJECTIVES
1. To prepare input data required for investigating optimum road maintenance standards for paved roads using
HDM-4.
2. To assign a range of maintenance standards to road sections and run HDM-4.
3. To interpret the results and recommend optimum maintenance standards for paved roads.
GUIDELINES
This case study presents the economic analysis of alternative maintenance standards for two paved road sections
that are 36.5 km and 10.6 km long, respectively. The existing roads have surface dressing and asphalt concrete
pavement surfacing in good condition, and carry approximately 6 000 and 16 000 vehicles per day (AADT),
respectively. The objective of this case study is to determine the optimum maintenance intervals for resealing
(surface dressing) and for overlays on the two roads, respectively.
The case study data is provided in the T3 Case studies workspace/database installed with the HDM-4 software.
The data for this case study are located in the Projects folder in the Case Studies Workspace. The name of
the case study is Case 2: Optimum maintenance intervals. To locate the data for this case study:
! Select the T3 Case Studies Workspace
The case study data includes the road network data set (CASE3 Project Sections) and the vehicle fleet (CASE2
Vehicle Fleet) required for the analyses.
The first step in the procedure would be to calibrate the road sections using the calibration factors derived from
the previous case study.
The second step in the procedure should be to create additional maintenance standards for the two roads. The
data provided includes one example each for:
! Surface dressing at 5 year intervals (scheduled), and
! Overlay 50mm at 6 IRI (responsive)
Create additional maintenance standards by copying the above to create separate standards for:
! Surface dressing intervals (scheduled) at 3, 7, 9 and 11 years
! Overlay intervals (responsive) at 4, 8, 10 and 12 IRI
After creating the additional maintenance standards, open the Projects Folder and go to the Specify
Alternatives group-tab. Select one of the sections and Edit Alternatives in order to assign the maintenance
standards. Note that the surface dressing standards should be applied to the section with ST surfacing, and the
overlays should be applied to the AC section.
When all of the alternatives have been assigned maintenance standards, go to the Analyse Projects group-tab
and specify the run data directory, and finally run HDM-4.
The results of the analyses can be viewed from the Generate Reports group-tab. The main report containing a
summary of the economic analysis results is the Benefit Cost Ratios within the Cost Streams report group.
Another useful plot would be of the NPV against the increase in agency cost. The resulting curve from this is
defined as the efficiency frontier. This will be described in detail in the Programme Analysis case study.
Those Trainers who complete the above exercise within a short time could attempt to derive the optimum
grading and regravelling frequencies for the Gravel road that is included in the Road Network database.
REFERENCES
1 Description
This case study presents the economic analysis of alternative maintenance standards for two
paved road sections, 36.5 km and 10.6 km long, respectively. The existing roads have surface
dressing and asphalt concrete surfacing, and carry approximately 6 000 and 16 000 vehicles
per day (AADT), respectively. The objective of this case study is to determine the optimum
maintenance intervals for resealing (surface dressing) and for overlays on the two roads,
respectively.
1.1 Locate the case study data
The case study data is provided in the T3 Case studies workspace/database installed with the
HDM-4 software. The data for this case study are located in the Projects folder in the Case
Studies Workspace. The name of the case study is Case 2: Optimum maintenance
intervals.
To locate the data for this case study:
! Go to the Case Studies Workspace
! Open the Projects folder
1.1.1 General
The General Tab page confirms the project description, analysis type, analysis period and the
pre-defined Road Network and Vehicle Fleet.
This case study is presented as a section analysis. The road under study is represented by one
section, and the different rehabilitation proposals represent section alternatives. The road
section will be selected from the Northern Province network, stored in the Road Network
folder. The vehicles using the road will be selected from the Northern Province vehicle fleet,
stored in the Vehicle Fleet folder.
It is also possible to conduct the analysis by Project although we have only one road section.
The analysis by Project option would combine the results of selected sections into one
Project. The analysis period is defined by a base year 2000, and an analysis period of 20
years, (that is, 2000 - 2019).
The Definition Tab gives details of basic section characteristics including road class,
speed-flow type and traffic flow pattern. The Pavement Type has been specified as
Details of the road section's geometry, including horizontal and vertical alignment, are
held under Section/Geometry and Section/Details/Alignment. The data indicates that
the road section under study is in rolling terrain (Rise + Fall = 20 m/km) and subject to a
speed limit of 100 kph.
Alignment data can be assessed from the existing mapping, if available, or measured by a
topographical survey.
! Pavement
After surface dressing (reseal) in 1995 (last resurfacing), Table A2.3 gives:
The dates Age2, Age3 and Age4 are used as a base line for the timing of scheduled
treatments (Age2 for surface treatments, Age3 for rehabilitation, Age4 for
reconstruction). For example, if surface dressing was requested every 4 years within a
Maintenance Standard effective from the year 2000, the first possible application would
be done immediately based on:
100 mm
Previous Surfacing
1988 (AMGB) 50mm AC Surfacing
Figure A2.1 Case Study 2: Details of existing pavements and pavement type
resets
Table A2.1
Generic HDM-4 bituminous pavement types
Note: AM and ST surfacings on concrete pavements, that is, AMCP and STCP, are
modelled in HDM-4 as concrete pavement types in the rigid pavement sub-model.
Table A2.2
Descriptions of surface and base materials
Table A2.3
Pavement type resets after maintenance works
Mill & replace **AP **AP **AP **AP N/A **SB **AB **AP
to intermediate
surface layer
Mill & replace **GB **SB **AB **AP **GB **SB **AB **AP
to base
Notes:
1 The pavement type depends on the critical thickness (Hmin) of the existing
bituminous surfacing that is user-definable in HDM Configuration.
** Indicates that these two characters are dependent on the specific works activity (or
operation) and the surface material.
The Pavement Tab page shows the current pavement strength, defined by the Adjusted
Structural Number of the pavement (Adjusted SNP). The Adjusted SNP incorporates a
weighting factor that reduces the contribution from the sub-base and subgrade, as
described in the Analytical Framework and Model Descriptions. (Note that the Adjusted
SNP is not the same as the Modified Structural Number used in HDM-III).
The following HDM-4 conventions should be noted:
1 SNP refers to the full structural number including contributions from the sub-base
and subgrade.
2 SN refers to the structural number comprising contributions from the surfacing,
roadbase and sub-base only.
For this case study, the Structural Number (SN) was entered directly together with the
subgrade CBR using Option (1) on the Section/Pavement screen. By clicking Option 1,
the calculated SNP value is displayed at the top of the screen.
Table A2.4
Ranking of road works applicable to the carriageway
Inlay 11 per m2
Edge-repair1 22 per m2
Routine
Patching1 22 per m2
Pavement
Crack sealing1 22 per m2
Note that Option (2) allows the user to calculate the SNP value with the help of the SNP
Calculation Wizard, which can calculate SNP based on deflections or layer thickness
and coefficients.
! Condition
The condition of the road in 1998 can be reviewed under the Condition Tab page.
For project analysis, this data should preferably be collected as part of a detailed
condition survey. However, users can set up aggregate data specifying detailed values
under HDM Configuration. Default values are shown in Chapter D2.
1.1.5 Alternatives
The Alternatives Tab is displayed after selecting the Specify Alternatives button. The
Alternatives screen is split into two boxes. The upper box shows the names of the six section
alternatives that have been set up for this case study.
The bottom box shows details of the Road Works Standards (assignments) associated with
each section alternative.
The six section alternatives considered for this case study are defined below. The analysis
period is 20 years (from year 2000 - 2019).
Alternative Description
Alternative Description
4 Mill 75 mm and replace 75 mm if Roughness >= 6 IRI AND Cracking >= 20%
carriageway area
5 Mill 75 mm and replace 100 mm if Roughness >= 6 IRI AND Cracking >= 20%
carriageway area
Note that each alternative includes routine pavement maintenance in the form of:
! Crack sealing if area of wide structural cracking > = 5%
! Patching if Severely Damaged Area >= 10%
The Road Works Standards and associated works assigned to each alternative are summarised
in Table A2.5. In this case study, each alternative is defined by one Maintenance Standard.
Note that an identical set of routine pavement works (crack sealing and patching) is common
to each Maintenance Standard. To be considered for implementation during any analysis
year, the routine works and rehabilitation works must be specified within the same
Maintenance Standard. (If routine works and rehabilitation works are specified as two
separate Maintenance Standards with the same effective from date, the software accepts only
one standard, that is, that which is assigned to the alternative first).
The intervention limits for the Works Items are summarised in Table A2.6.
Table A2.5
Case Study 2: Details of road works standards for each project alternative
Table A2.6
Case Study 2: Intervention limits for maintenance works
Notes:
n/a = not applicable; * S = Scheduled intervention, R = Responsive intervention
Note that for certain works items, the unit costs of preparatory works are entered separately on
the Works Item/Costs Tab page. In Case Study 2, this applies to the following works items.
Patching
50 mm overlay
Edge repair
3 Patching
Crack Sealing
Patching
Crack Sealing
Where HDM-4 does not offer a separate facility for entry of preparatory costs, such costs (if
any) should be included in the costs of the Works Item.
Issues relating to the Maintenance Works Standard for each alternative are discussed below:
! Alternative 1 - Maintenance Standard: Routine + Reconstruct
The works included in this Maintenance Standard are the routine pavement works (crack
sealing and patching) and reconstruction. The works are listed in the opening screen for
the Maintenance Standard:
The specification for each of these works can be reviewed by double-clicking the
appropriate works item name.
! Works Item: Crack Sealing and Patching
The specifications for these works, included in all alternatives for this case study, are
similar to those described for Case Study 1 within the Maintenance Standard Crack
Sealing and Patching Paved Road (see Appendix A1).
! Works Item: Reconstruct at IRI 12 and 40% damage
The specification for Reconstruction is held under the tabs:
General/Design/Intervention/Costs/Effects
Note that the Works Activity (Pavement Reconstruction) is specified under General.
The Design Tab shows the proposed new Pavement Type as Asphalt Mix on
Granular Base (AMGB) together with pavement details required by HDM-4.
Note that the details under Design refer to the new pavement construction. The
requested structural number refers to the layers above the subgrade.
The reconstruction Costs to be specified include all costs associated with the works.
The Tab pages specifying the Reconstruction Works Item are shown below:
The works included in the Maintenance Standard are the routine pavement works (crack
sealing and patching) and a 50 mm thick overlay (condition responsive, based on IRI).
The works are listed in the opening screen for the Maintenance Standard:
The specification for the overlay works can be reviewed by double-clicking the
appropriate works item name.
Note that two works items for overlaying are specified. Both are condition responsive,
one based on roughness, the other on structural cracking. This means that either
roughness OR structural cracking can trigger overlaying. The Works Items are identical
except for the Intervention Tab page. The Works Item based on roughness is reviewed
below:
! Maintenance Works: Overlay 50 mm at IRI 6
The specification for the Overlay works can be reviewed under the tabs:
General/Design/Intervention/Costs/Effects
The Works Activity (overlay dense-graded asphalt) is specified under General. The
overlay material, thickness and dry season strength coefficient are specified under
Design. The Construction defect indicators for the bituminous overlay (CDS) is also
specified.
Note that if overlay is implemented, the Pavement Type will be automatically reset
according to Table A2.3:
The Tab pages specifying the Overlay Works are reviewed below:
! Alternative 3 - Maintenance Standard: Routine + SD + 50 mm overlay
The works included in this Maintenance Standard are the same as for Alternative 2 plus
an extra works item, Surface Dressing (SD).
The specification for the Surface Dressing Works Item can be reviewed by double-
clicking the appropriate works item name:
! Maintenance Works: Surface Dressing at SFC <= 0.4
The specification for Surface Dressing can be reviewed under the tabs:
General/Design/Intervention/Costs/Effects
The Works Activity is specified as Surface Dressing Single (General tab), with
responsive intervention based on the skid resistance (Intervention tab). Note that a
double surface dressing is also available, and shape correction may be applied to
either single or double surface dressings. Shape correction adds to the cost, but has
a more beneficial effect in terms of roughness (details are given in Chapter D2 of the
Analytical Framework and Model Descriptions).
For this case study, intervention is based on the skid resistance. Additional criteria
that may be used with surface dressing are cracking, ravelling, texture depth, and
total damaged area.
Note that the unit costs of surface dressing works and preparatory works are
specified under Costs.
The pavement type (STAP) will be reset automatically according to Table A2.3 if
either the surface dressing or overlay works is implemented.
For example:
The Tab pages specifying the Surface Dressing Works are given below:
! Alternative 4 - Maintenance Standard: Routine + Mill 75 mm & Replace 75 mm
In addition to routine pavement works (crack sealing and patching) this Maintenance
Standard includes Mill and Replace (condition responsive based on IRI).
The specification for Mill and Replace can be reviewed under the tabs:
General/Design/Intervention/Costs/Effects
The Tab pages specifying the Mill and Replace Works for Alternative 4 are
reviewed below:
This Maintenance Standard is similar to that specified for Alternative 4, except in this
case a 100 mm thick surfacing is added after milling out 75 mm.
As with Alternative 4, the Pavement Type after implementation of the specified Mill &
Replace works would be given by:
The Tab pages specifying the Mill and Replace Works for Alternative 5 are reviewed
below:
! Alternative 6 - Maintenance Standard: Routine + Inlay
In addition to the routine pavement works (crack sealing and patching), this Maintenance
Standard includes Inlay, that is condition responsive, based on the mean rut depth.
The specification for the Inlay works can be reviewed under the tabs:
General/Design/Intervention/Costs/Effects
2000 Overlay at 15% cracking including preparatory patching and edge repair
2007 Overlay at 15% cracking
2015 Overlay at IRI 6
Observations The absence of routine pavement works between overlay applications
should be investigated by comparing the intervention criteria with the
progression of defects on the Deterioration Summary report.
The works proposed by these alternatives are summarised below. Note the different
thickness of new surfacing (after milling) for alternatives 4 and 5.
Observations Inlay is triggered five times during the analysis period, based on the
mean rut depth reaching 20 mm. Note that inlay is specified for 25%
carriageway area.
(US$ millions)
1 Routine and Reconstruct 18.8 (base)
2 Routine + 50 mm overlay 18.0 78.1
3 Routine + SD + 50 mm overlay 56.0 50.6
4 Routine + Mill 75 mm & Replace 75 mm 21.2 92.5
5 Routine + Mill 75 mm & Replace 100 24.8 86.5
mm
6 Routine + Inlay 10.2 86.9
Note that the significant condition parameters available as intervention criteria for Case Study
2 are:
! Roughness (IRI)
May be used as an intervention criteria for Reconstruction, Overlay, and Mill & Replace
May be used as an intervention criterion for Reconstruction and Surface Dressing, and
defined by:
where:
* Note :
where:
where:
Note also that transverse thermal cracks (no/km) may be adopted as an intervention
criterion for crack sealing
! Severely Damaged Area (ADAMS)
1 Unit Aims
1. To introduce the HDM-4 Programme Analysis module concepts so that Trainers and Trainees
are able to prepare annual or multi-year work programmes.
2. To introduce the concept of incremental benefit/cost ranking method so that Trainers
appreciate optimisation techniques.
3. To familiarise Trainers with the use of HDM-4 data input and output modules.
2 Unit Objectives
1. To describe the concepts of the HDM-4 Programme analysis application.
2. To apply the incremental benefit cost optimisation techniques through simple manual
calculation.
3. To use the HDM-4 model to generate a work programme.
4. To alter maintenance standards and unit costs and study the impact on budget requirements.
3 Introduction
This Case Study demonstrates the application of Programme Analysis to the production of a
rehabilitation programme for a road network in the Western Province of a country. The study
demonstrates the Life Cycle Analysis method described in Chapter B2.
The objective here is to prepare a prioritised list road projects from a selected group of road
sections that have been identified to be in poor condition. In practice, this would follow some
condition survey of the road network and the application of pavement condition trigger levels or
thresholds to identify those sections in need of some form of rehabilitation.
The Life Cycle analysis method used for this Case Study is the same as that used in Project
Analysis and described in Chapter B1.
To demonstrate this Programme analysis method, the Case Study is presented in the following
steps:
Step 1 : Overview of the case study.
Step 2 : Review of the case study input data.
Step 3 : Run programme analysis with the life-cycle option
Step 4: Examine the results of the work programme
T3 Case 3 Programme.doc 1
PROGRAMME ANALYSIS CASE STUDY
! Road network containing the candidate road sections for periodic maintenance and/or
improvement.
! Vehicle fleet defining the characteristics of typical vehicles commonly found on the road
network.
! The average annual daily traffic (AADT) on each road section.
! Maintenance and/or improvement standards to be assigned to each candidate road section.
The above data will usually be pre-defined in HDM-4 before a Programme Analysis study is
created. The following data managers are used for this:
! Road Network manager - to create and edit the road section data
To review the case study data, open the Programmes folder and select the Western Province
Rehabilitation (Life Cycle Analysis) study.
5 Input data
The data used for this Case Study has already been created and therefore the procedure described
below does not include the creation of a new study.
The procedure involves defining details of the programme analysis to be carried out, assignment
of work standards, generation of the work programme, optimisation under budget constraints, and
finally generation of reports. This procedure should normally be followed sequentially, although
iteration between the above tasks can be done in order to review previously specified data.
The data is arranged in task-groups and, within these, the data are presented in different Tabs.
6.1 General
The General Tab can be used to define the study description, analysis type, analysis period (start
year and duration), the currency conversion rates and the discount rate to be used in the economic
analysis. This Tab also shows the selected Road Network and Vehicle Fleet.
This Case Study involves Life Cycle analysis carried out over a fifteen year analysis period
starting from 2000 to 2014.
T3 Case 3 Programme.doc 2
PROGRAMME ANALYSIS CASE STUDY
It is recommended that work programmes should be prepared for short term budget periods, for
example five years or less.
Details of each road section can be edited from within this Tab (press the View/Edit Network
button).
7.1 Alternatives
The Life Cycle analysis method requires a comparison to be made between a base case (do
minimum) alternative against a project case (do something) alternative. Details of at least two
alternatives can be specified in this Tab.
For this Case Study, two alternatives have been defined for each road section; a base case
alternative and a rehabilitation alternative. The base case comprises routine pavement
maintenance only (crack sealing and patching potholes only). The rehabilitation alternative
includes routine maintenance, resealing, overlays and reconstruction for bituminous pavements.
Details of the maintenance standards are summarised in Table B1.1.
T3 Case 3 Programme.doc 3
PROGRAMME ANALYSIS CASE STUDY
Table B1.1
Details of Maintenance Standard (Brehab) assigned to bituminous road sections
Resealing Single Surface Dressing Total damaged area >= 30% of pavement 16
surface area
Note: * Ranking of road works from Volume 4: Analytical Framework and Model Descriptions,
Chapter D2
If more than one of these activities are triggered in a given year, the works item that is highest in
the hierarchy of road works (that is, with lowest ranking value) will be applied.
Note that routine pavement works (for example, patching and crack sealing) have the same
ranking, and therefore both can be applied in the same year.
8 Generate Programme
For life-cycle analysis, the Base alternative for economic analysis is specified in the Run Setup
screen. This can be displayed by selecting the Run Setup button within the Perform Run Tab.
On completing Run Setup, press the Start button to commence the economic analysis, which
produces an unconstrained work programme. This contains, for each road section, the alternative
with the highest NPV (refer to Chapter B2). This represents an unconstrained programme, with
the total financial cost given in the cumulative cost column.
The ideal budget required for periodic maintenance in year 2000 is approximately US$ 22
million.
T3 Case 3 Programme.doc 4
PROGRAMME ANALYSIS CASE STUDY
Table B1.2
Details of alternatives for life-cycle analysis
Alternative Surface Maintenance Standard Description Code Effective Maintenance Works Code
Class from year
Base Case Bituminous Patching and Crack Sealing Broutine 2000 Pothole Patching PATPOT
Rehabilitation Bituminous P&CS, Reseal, Overlay & Reconstruct Brehab 2000 Pothole Patching * PATPOT
Reseal RSL30
Overlay OV5IRI
Reconstruct REC12
Notes:
1 Details of the intervention criteria for road work activities are given in Table B1.1.
2 The routine pavement works in the Rehabilitation alternative (identified by *) are identical to those in the Base Case.
T3 Case 3 Programme.doc 5
PROGRAMME ANALYSIS CASE STUDY
Details of the budget optimisation options are specified by pressing the Optimisation Setup
button. For this case study, optimisation is by incremental benefit/cost ranking (see HDM-4
Volume 4). The parameters to be specified are given below:
Minimum incremental value = 0.1
T3 Case 3 Programme.doc 7
ERROR! NO TEXT OF SPECIFIED STYLE IN DOCUMENT. APPENDIX B1 - PROGRAMME ANALYSIS CASE STUDY 1
Year Section Length AADT Surface Work Description NPV/Cost Financial Cumulative
(km) Class Costs Costs
2000 MTS 549 km 0 - 1.0 1.0 2521 Bituminous Overlay 50mm 3.38 0.13 0.13
2000 MSC 112 km 35 - 80 21.0 1770 Bituminous Overlay 50mm 1.85 2.65 2.77
2000 MAN 203 km 254 - 262.8 8.8 1781 Bituminous Overlay 50mm 1.83 1.13 3.91
2000 MSC 141 km 28 - 49.5 21.5 2154 Bituminous Resealing SBSD 1.28 0.92 4.83
2000 MSC 131 km 16.5 - 26 9.5 2154 Bituminous Resealing SBSD 1.14 0.42 5.25
2000 MTS 754 km 0 - 1.8 1.8 1074 Bituminous Overlay 50mm 1.11 0.25 5.50
2000 MSW 959 km 0.1- 5, 10- 15.4 10.3 1265 Bituminous Overlay 50mm 1.10 1.26 6.76
2000 MSE 203 km 80 - 90 10.0 1191 Bituminous Overlay 50mm 0.76 1.26 8.02
2000 MAN 409 km 105 - 110 5.0 1076 Bituminous Overlay 50mm 0.60 0.63 8.65
2000 MTN 836 km 0 - 13 13.0 926 Bituminous Overlay 50mm 0.60 1.64 10.29
2000 MAN 243 km 0 - 21 21.0 1044 Bituminous Overlay 50mm 0.53 2.65 12.94
2000 MTN 805 km 145 - 152.6 7.6 873 Bituminous Overlay 50mm 0.41 1.01 13.94
2000 MSC 138 km 5 - 10 & 65 - 70 10.0 1315 Bituminous Resealing SBSD 0.21 0.42 14.36
This report was generated by version 1.0 of HDM-4. Page -1 of 1
1 Background
This case study presents the economic analysis of alternative rehabilitation standards for a 50
km long paved road. The existing road carries 8000 vehicles per day (AADT) and exhibits
significant levels of roughness and surface distress. This case study evaluates several possible
rehabilitation alternatives including overlay, mill and replace, and inlay.
The objective of the case study is to present the definition of section alternatives appropriate
to this type of analysis and demonstrate the HDM-4 deterioration model for paved roads. The
definition and timing of the Maintenance Road Works Standards are discussed, and the results
are examined.
To demonstrate this case study the following steps are followed:
! Locate the case study data (see Section 1.1)
! Specify Alternatives
1.2.1 General
The General Tab page confirms the project description, analysis type, analysis period and the
pre-defined Road Network and Vehicle Fleet.
This case study is presented as a section analysis. The road under study is represented by one
section, and the different rehabilitation proposals represent section alternatives. The road
section will be selected from the Northern Province network, stored in the Road Network
folder. The vehicles using the road will be selected from the Northern Province vehicle fleet,
stored in the Vehicle Fleet folder.
It is also possible to conduct the analysis by Project although we have only one road section.
The analysis by Project option would combine the results of selected sections into one
Project.
The analysis period is defined by a start year 2000, and a duration 20 years, (that is, 2000 -
2019).
The Definition Tab gives details of basic section characteristics including road class,
speed-flow type and traffic flow pattern. The Pavement Type has been specified as
Surface Treatment on Asphalt Pavement (STAP). It is important that the correct
Pavement Type is specified, as the deterioration relationships held in HDM-4 are defined
by Pavement Type (see Table A2.1).
Details of the existing pavement are discussed under Pavement below:
! Geometry
Details of the road section's geometry, including horizontal and vertical alignment, are
held under Section/Geometry and Section/Details/Alignment. The data indicates that
the road section under study is in rolling terrain (Rise + Fall = 20 m/km) and subject to a
speed limit of 100 kph.
Alignment data can be assessed from the existing mapping, if available, or measured by a
topographical survey.
! Pavement
thick granular sub-base. The in situ subgrade CBR is 8%. The pavement details required
by HDM-4 for this pavement type (Surface Treatment on Asphalt Pavement) are
indicated on the Pavement Tab page below:
Note that the specification of Pavement Type (on the Section/Definition page) refers to
the current pavement construction and should be compatible with the Previous Works
details specified on the Pavement Tab page.
The Pavement Type is automatically updated immediately after any maintenance works.
The definition of bituminous Pavement Types based on surface and base types is given in
Table A2.1. (Table A2.2 contains descriptions of surface and base materials.) A
summary of Pavement Type resets after maintenance works is given in Table A2.3.
For the road section under study, the changes in Pavement Type since the date of the last
construction (1988) are summarised in Figure A2.1.
The last reconstruction was 1988, given by pavement type AMGB. After the application
of overlay in 1992, Table A2.3 shows that:
After surface dressing (reseal) in 1995 (last resurfacing), Table A2.3 gives:
Figure A2.1 Case Study 2: Details of existing pavements and pavement type
resets
Table A2.1
Generic HDM-4 bituminous pavement types
Note: AM and ST surfacings on concrete pavements, that is, AMCP and STCP, are
modelled in HDM-4 as concrete pavement types in the rigid pavement sub-model.
The abbreviations in Table A2.1 are described in Table A2.2.
Table A2.2
Descriptions of surface and base materials
Table A2.3
Pavement type resets after maintenance works
Mill & replace **AP **AP **AP **AP N/A **SB **AB **AP
to intermediate
surface layer
Mill & replace **GB **SB **AB **AP **GB **SB **AB **AP
to base
Notes:
1 The pavement type depends on the critical thickness (Hmin) of the existing
bituminous surfacing that is user-definable in HDM Configuration.
** Indicates that these two characters are dependent on the specific works activity (or
operation) and the surface material.
N/A Not applicable.
The Pavement Tab page shows the current pavement strength, defined by the Adjusted
Structural Number of the pavement (Adjusted SNP). The Adjusted SNP incorporates a
weighting factor that reduces the contribution from the sub-base and subgrade, as
described in the Analytical Framework and Model Descriptions. (Note that the Adjusted
SNP is not the same as the Modified Structural Number used in HDM-III).
The following HDM-4 conventions should be noted:
1 SNP refers to the full structural number including contributions from the sub-base
and subgrade.
2 SN refers to the structural number comprising contributions from the surfacing,
roadbase and sub-base only.
For this case study, the Structural Number (SN) was entered directly together with the
subgrade CBR using Option (1) on the Section/Pavement screen. By clicking Option 1,
the calculated SNP value is displayed at the top of the screen.
Table A2.4
Ranking of road works applicable to the carriageway
Inlay 11 per m2
Edge-repair1 22 per m2
Routine
Patching1 22 per m2
Pavement
Crack sealing1 22 per m2
Note that Option (2) allows the user to calculate the SNP value with the help of the SNP
Calculation Wizard, which can calculate SNP based on deflections or layer thickness
and coefficients.
! Condition
The condition of the road in 1998 can be reviewed under the Condition Tab page.
For project analysis, this data should preferably be collected as part of a detailed
condition survey. However, users can set up aggregate data specifying detailed values
under HDM Configuration. Default values are shown in Chapter D2.
1.2.5 Alternatives
The Alternatives Tab is displayed after selecting the Specify Alternatives button. The
Alternatives screen is split into two boxes. The upper box shows the names of the six section
alternatives that have been set up for this case study.
The bottom box shows details of the Road Works Standards (assignments) associated with
each section alternative.
The six section alternatives considered for this case study are defined below. The analysis
period is 20 years (from year 2000 - 2019).
Alternative Description
Alternative Description
4 Mill 75 mm and replace 75 mm if Roughness >= 6 IRI AND Cracking >= 20%
carriageway area
5 Mill 75 mm and replace 100 mm if Roughness >= 6 IRI AND Cracking >= 20%
carriageway area
Note that each alternative includes routine pavement maintenance in the form of:
! Crack sealing if area of wide structural cracking > = 5%
! Patching if Severely Damaged Area >= 10%
The Road Works Standards and associated works assigned to each alternative are summarised
in Table A2.5. In this case study, each alternative is defined by one Maintenance Standard.
Note that an identical set of routine pavement works (crack sealing and patching) is common
to each Maintenance Standard. To be considered for implementation during any analysis
year, the routine works and rehabilitation works must be specified within the same
Maintenance Standard. (If routine works and rehabilitation works are specified as two
separate Maintenance Standards with the same effective from date, the software accepts only
one standard, that is, that which is assigned to the alternative first).
The intervention limits for the Works Items are summarised in Table A2.6.
Table A2.5
Case Study 2: Details of road works standards for each project alternative
Table A2.6
Case Study 2: Intervention limits for maintenance works
Notes:
n/a = not applicable; * S = Scheduled intervention, R = Responsive intervention
Note that for certain works items, the unit costs of preparatory works are entered separately on
the Works Item/Costs Tab page. In Case Study 2, this applies to the following works items.
Patching
50 mm overlay
Edge repair
3 Patching
Crack Sealing
Patching
Crack Sealing
Where HDM-4 does not offer a separate facility for entry of preparatory costs, such costs (if
any) should be included in the costs of the Works Item.
Issues relating to the Maintenance Works Standard for each alternative are discussed below:
! Alternative 1 - Maintenance Standard: Routine + Reconstruct
The works included in this Maintenance Standard are the routine pavement works (crack
sealing and patching) and reconstruction. The works are listed in the opening screen for
the Maintenance Standard:
The specification for each of these works can be reviewed by double-clicking the
appropriate works item name.
! Works Item: Crack Sealing and Patching
The specifications for these works, included in all alternatives for this case study, are
similar to those described for Case Study 1 within the Maintenance Standard Crack
Sealing and Patching Paved Road (see Appendix A1).
! Works Item: Reconstruct at IRI 12 and 40% damage
The specification for Reconstruction is held under the tabs:
General/Design/Intervention/Costs/Effects
Note that the Works Activity (Pavement Reconstruction) is specified under General.
The Design Tab shows the proposed new Pavement Type as Asphalt Mix on
Granular Base (AMGB) together with pavement details required by HDM-4.
Note that the details under Design refer to the new pavement construction. The
requested structural number refers to the layers above the subgrade.
The reconstruction Costs to be specified include all costs associated with the works.
The Tab pages specifying the Reconstruction Works Item are shown below:
The works included in the Maintenance Standard are the routine pavement works (crack
sealing and patching) and a 50 mm thick overlay (condition responsive, based on IRI).
The works are listed in the opening screen for the Maintenance Standard:
The specification for the overlay works can be reviewed by double-clicking the
appropriate works item name.
Note that two works items for overlaying are specified. Both are condition responsive,
one based on roughness, the other on structural cracking. This means that either
roughness OR structural cracking can trigger overlaying. The Works Items are identical
except for the Intervention Tab page. The Works Item based on roughness is reviewed
below:
! Maintenance Works: Overlay 50 mm at IRI 6
The specification for the Overlay works can be reviewed under the tabs:
General/Design/Intervention/Costs/Effects
The Works Activity (overlay dense-graded asphalt) is specified under General. The
overlay material, thickness and dry season strength coefficient are specified under
Design. The Construction defect indicators for the bituminous overlay (CDS) is also
specified.
Note that if overlay is implemented, the Pavement Type will be automatically reset
according to Table A2.3:
The Tab pages specifying the Overlay Works are reviewed below:
! Alternative 3 - Maintenance Standard: Routine + SD + 50 mm overlay
The works included in this Maintenance Standard are the same as for Alternative 2 plus
an extra works item, Surface Dressing (SD).
The specification for the Surface Dressing Works Item can be reviewed by double-
clicking the appropriate works item name:
! Maintenance Works: Surface Dressing at SFC <= 0.4
The specification for Surface Dressing can be reviewed under the tabs:
General/Design/Intervention/Costs/Effects
The Works Activity is specified as Surface Dressing Single (General tab), with
responsive intervention based on the skid resistance (Intervention tab). Note that a
double surface dressing is also available, and shape correction may be applied to
either single or double surface dressings. Shape correction adds to the cost, but has
a more beneficial effect in terms of roughness (details are given in Chapter D2 of the
Analytical Framework and Model Descriptions).
For this case study, intervention is based on the skid resistance. Additional criteria
that may be used with surface dressing are cracking, ravelling, texture depth, and
total damaged area.
Note that the unit costs of surface dressing works and preparatory works are
specified under Costs.
The pavement type (STAP) will be reset automatically according to Table A2.3 if
either the surface dressing or overlay works is implemented.
For example:
The Tab pages specifying the Surface Dressing Works are given below:
! Alternative 4 - Maintenance Standard: Routine + Mill 75 mm & Replace 75 mm
In addition to routine pavement works (crack sealing and patching) this Maintenance
Standard includes Mill and Replace (condition responsive based on IRI).
The specification for Mill and Replace can be reviewed under the tabs:
General/Design/Intervention/Costs/Effects
The Tab pages specifying the Mill and Replace Works for Alternative 4 are
reviewed below:
This Maintenance Standard is similar to that specified for Alternative 4, except in this
case a 100 mm thick surfacing is added after milling out 75 mm.
As with Alternative 4, the Pavement Type after implementation of the specified Mill &
Replace works would be given by:
The Tab pages specifying the Mill and Replace Works for Alternative 5 are reviewed
below:
! Alternative 6 - Maintenance Standard: Routine + Inlay
In addition to the routine pavement works (crack sealing and patching), this Maintenance
Standard includes Inlay, that is condition responsive, based on the mean rut depth.
The specification for the Inlay works can be reviewed under the tabs:
General/Design/Intervention/Costs/Effects
2000 Overlay at 15% cracking including preparatory patching and edge repair
2007 Overlay at 15% cracking
2015 Overlay at IRI 6
Observations The absence of routine pavement works between overlay applications
should be investigated by comparing the intervention criteria with the
progression of defects on the Deterioration Summary report.
The works proposed by these alternatives are summarised below. Note the different
thickness of new surfacing (after milling) for alternatives 4 and 5.
Observations Inlay is triggered five times during the analysis period, based on the
mean rut depth reaching 20 mm. Note that inlay is specified for 25%
carriageway area.
(US$ millions)
1 Routine and Reconstruct 18.8 (base)
2 Routine + 50 mm overlay 18.0 78.1
3 Routine + SD + 50 mm overlay 56.0 50.6
4 Routine + Mill 75 mm & Replace 75 mm 21.2 92.5
5 Routine + Mill 75 mm & Replace 100 24.8 86.5
mm
6 Routine + Inlay 10.2 86.9
Note that the significant condition parameters available as intervention criteria for Case Study
2 are:
! Roughness (IRI)
May be used as an intervention criteria for Reconstruction, Overlay, and Mill & Replace
May be used as an intervention criterion for Reconstruction and Surface Dressing, and
defined by:
where:
* Note :
where:
where:
Note also that transverse thermal cracks (no/km) may be adopted as an intervention
criterion for crack sealing
! Severely Damaged Area (ADAMS)
1 Introduction
This Case Study demonstrates the application of Programme Analysis to the production of a
rehabilitation programme for a road network in the Western Province of a country. The study
demonstrates the Life Cycle Analysis method described in Chapter B2.
The objective here is to prepare a prioritised list road projects from a selected group of road
sections that have been identified to be in poor condition. In practice, this would follow some
condition survey of the road network and the application of pavement condition trigger levels or
thresholds to identify those sections in need of some form of rehabilitation.
The Life Cycle analysis method used for this Case Study is the same as that used in Project
Analysis and described in Chapter B1.
To demonstrate this Programme analysis method, the Case Study is presented in the following
steps:
Step 1 : Overview of the case study.
Step 2 : Review of the case study input data.
Step 3 : Run programme analysis with the life-cycle option
Step 4: Examine the results of the work programme
To review the case study data, open the Programmes folder and select the Western Province
Rehabilitation (Life Cycle Analysis) study.
3 Input data
The data used for this Case Study has already been created and therefore the procedure described
below does not include the creation of a new study.
The procedure involves defining details of the programme analysis to be carried out, assignment
of work standards, generation of the work programme, optimisation under budget constraints, and
finally generation of reports. This procedure should normally be followed sequentially, although
iteration between the above tasks can be done in order to review previously specified data.
The data is arranged in task-groups and, within these, the data are presented in different Tabs.
4.1 General
The General Tab can be used to define the study description, analysis type, analysis period (start
year and duration), the currency conversion rates and the discount rate to be used in the economic
analysis. This Tab also shows the selected Road Network and Vehicle Fleet.
This Case Study involves Life Cycle analysis carried out over a fifteen year analysis period
starting from 2000 to 2014.
road section to others, thereby permitting each road section to have a different road set of traffic
growth characteristics, if required.
5.1 Alternatives
The Life Cycle analysis method requires a comparison to be made between a base case (do
minimum) alternative against a project case (do something) alternative. Details of at least two
alternatives can be specified in this Tab.
For this Case Study, two alternatives have been defined for each road section; a base case
alternative and a rehabilitation alternative. The base case comprises routine pavement
maintenance only (crack sealing and patching potholes only). The rehabilitation alternative
includes routine maintenance, resealing, overlays and reconstruction for bituminous pavements.
Details of the maintenance standards are summarised in Table B1.1.
Table B1.1
Details of Maintenance Standard (Brehab) assigned to bituminous road sections
Resealing Single Surface Dressing Total damaged area >= 30% of pavement 16
surface area
Note: * Ranking of road works from Volume 4: Analytical Framework and Model Descriptions,
Chapter D2
If more than one of these activities are triggered in a given year, the works item that is highest in
the hierarchy of road works (that is, with lowest ranking value) will be applied.
Note that routine pavement works (for example, patching and crack sealing) have the same
ranking, and therefore both can be applied in the same year.
6 Generate Programme
For life-cycle analysis, the Base alternative for economic analysis is specified in the Run Setup
screen. This can be displayed by selecting the Run Setup button within the Perform Run Tab.
On completing Run Setup, press the Start button to commence the economic analysis, which
produces an unconstrained work programme. This contains, for each road section, the alternative
with the highest NPV (refer to Chapter B2). This represents an unconstrained programme, with
the total financial cost given in the cumulative cost column.
The ideal budget required for periodic maintenance in year 2000 is approximately US$ 22
million.
Table B1.2
Details of alternatives for life-cycle analysis
Alternative Surface Maintenance Standard Description Code Effective Maintenance Works Code
Class from year
Base Case Bituminous Patching and Crack Sealing Broutine 2000 Pothole Patching PATPOT
Rehabilitation Bituminous P&CS, Reseal, Overlay & Reconstruct Brehab 2000 Pothole Patching * PATPOT
Reseal RSL30
Overlay OV5IRI
Reconstruct REC12
Notes:
1 Details of the intervention criteria for road work activities are given in Table B1.1.
2 The routine pavement works in the Rehabilitation alternative (identified by *) are identical to those in the Base Case.
Details of the budget optimisation options are specified by pressing the Optimisation Setup
button. For this case study, optimisation is by incremental benefit/cost ranking (see HDM-4
Volume 4). The parameters to be specified are given below:
Minimum incremental value = 0.1
Year Section Length AADT Surface Work Description NPV/Cost Financial Cumulative
(km) Class Costs Costs
2000 MTS 549 km 0 - 1.0 1.0 2521 Bituminous Overlay 50mm 3.38 0.13 0.13
2000 MSC 112 km 35 - 80 21.0 1770 Bituminous Overlay 50mm 1.85 2.65 2.77
2000 MAN 203 km 254 - 262.8 8.8 1781 Bituminous Overlay 50mm 1.83 1.13 3.91
2000 MSC 141 km 28 - 49.5 21.5 2154 Bituminous Resealing SBSD 1.28 0.92 4.83
2000 MSC 131 km 16.5 - 26 9.5 2154 Bituminous Resealing SBSD 1.14 0.42 5.25
2000 MTS 754 km 0 - 1.8 1.8 1074 Bituminous Overlay 50mm 1.11 0.25 5.50
2000 MSW 959 km 0.1- 5, 10- 15.4 10.3 1265 Bituminous Overlay 50mm 1.10 1.26 6.76
2000 MSE 203 km 80 - 90 10.0 1191 Bituminous Overlay 50mm 0.76 1.26 8.02
2000 MAN 409 km 105 - 110 5.0 1076 Bituminous Overlay 50mm 0.60 0.63 8.65
2000 MTN 836 km 0 - 13 13.0 926 Bituminous Overlay 50mm 0.60 1.64 10.29
2000 MAN 243 km 0 - 21 21.0 1044 Bituminous Overlay 50mm 0.53 2.65 12.94
2000 MTN 805 km 145 - 152.6 7.6 873 Bituminous Overlay 50mm 0.41 1.01 13.94
2000 MSC 138 km 5 - 10 & 65 - 70 10.0 1315 Bituminous Resealing SBSD 0.21 0.42 14.36
This report was generated by version 1.0 of HDM-4. Page -1 of 1
Derivation of Optimum
Maintenance Standards
Aims & Objectives
Aim:
z To understand
T d t d theth specification
ifi ti and
d
assignment of HDM-4 maintenance standards
Objectives:
z prepare input data required for investigating
optimum road maintenance standards
z assign a range of maintenance standards to
road sections
z interpret the results and recommend optimum
maintenance standards
Overview
2 Road Sections
z ST, 36
ST 36.5
5 km
km, AADT = 6 000
000.
z AC, 10.6 km, AADT = 16 000.
Maintenance Standards
z Base Case: Patching potholes, Crack sealing,
Drainage maint + Other cyclic works.
z P i di maint
Periodic i t 1 for
f ST road
d section:
ti Reseal
R l every 5
years (scheduled)
z Periodic maint 1 for AC road section: Overlay 50mm
at 6 IRI (responsive)
Procedure
NPV
Agency Cost
PROJECT ANALYSIS -
WORKED EXAMPLES
Overview of HDM-4 Cases Studies
UoB
OBJECTIVES
z Network Level
z Total energy use by transport mode
z Long term energy requirements
z Renewable energy sources
z Project Level
z Compare energy use for investment
options
z Specific energy use by transport mode
ENERGY COMPONENTS
z Motorised Transport
z Vehicle Operation
z Fuel Production
z Vehicle Manufacture
z Non-Motorised Transport
z Animal drawn carts,
carts cycling,
cycling head-loading
z Road Construction and Maintenance
z Road works
MOTORISED TRANSPORT
z Vehicle manufacture
z raw materials,
t i l processing,
i components,t
transportation, assembly, distribution
z Fuel production
z raw materials, stock transport, processing,
distribution
z Vehicle use
z fuel oil & tyre consumption
fuel,
z Vehicle maintenance
z component manufacture,
manufacture distribution
VEHICLE MANUFACTURE
z Motorcycle 20 GJ
z Medium car 100
z Light
g goods
g vehicle 160
z Four wheel drive 180
z Medium truck 600
z Articulated truck 1,500
z Mini-bus
Mini bus 300
z Large bus 1,000
FUEL PRODUCTION CYCLE
Petrol, Depot
p
Oil Di l
Diesel
Pipeline storage
refinery road tankers
Gas Petrol,
Separation Crude Oil Service Diesel
plant Tanker stations vehicl
es
FUEL PRODUCTION ENERGY
UoB
OBJECTIVES
z Road characteristics
z Traffic congestion
z V hi l technology
Vehicle t h l
APPLICATIONS
z Network level
z Total emissions by transport mode
z Regional long term impact of pollution
z Project level
z Compare levels of emissions for different
investment options
z Pollution impact on local areas near the
project road
TYPES OF POLLUTANTS
z Hydrocarbon (HC)
z Carbon monoxide (CO)
z Nitrous oxides (NOx)
z Sulphur dioxide (SO2)
z Carbon dioxide (CO2)
z P ti l t (Par)
Particulates (P )
z Lead (Pb)
THE MODELLING
z Quantities of emissions
z by component
z by vehicle type
z for each traffic flow period
z Annual quantities of emissions
z by component
z for each investment option
z Annual net quantities of emissions for
each pair of investment alternatives
CONCRETE PAVEMENTS
Pavement Classification
z J i
Jointed
d Plain
Pl i without
ih dowels
d l
z J i t d Plain
Jointed Pl i with
ith dowels
d l
z Jointed Reinforced
z Continuously Reinforced
Jointed Plain without dowels
Joint Spacing
3 - 6 m
Aggregate
Slab Interlock
Base
Jointed Plain with dowels
Joint Spacing
3 - 6 m
Dowels
Jointed Reinforced
10 - 20 m
Slab
Dowels
Base
Cracks separation
Slab
Base
Reinforcement Steel
0,6 - 0,8 % Area
Distress Modes Modelled in HDM-4
z S
Serviceability
i bilit di
dimensionless
i l JR CR
JR,
z Roughness
g inches p
per mile All
Primary Variables
z Structural characteristics
z Traffic
z E i
Environment
t
z Road geometry
z Pavement condition
z Age
z Previous maintenance
Structural Characteristics
Distress Distress
width width
A B C D
Longitudinal Joint
C
L
C D
Transv. Transv.
Joint Joint
A B Traffic
Sl b
Slab
Shoulder
Cracking - JP Pavements
z Temperature gradient
z Slab thickness
z Material properties
z Joint spacing
z Shoulder
Sh ld supportt
Cracking - JR Pavements
z Temperature gradient
z Slab thickness
z Material properties
z Joint spacing
z Shoulder
Sh ld supportt
Faulting
Faultingg represents
p the
elevation difference between
g of a transverse
the edges
joint or crack
Faulting
A
faulting B
Longitudinal Joint
C
L
Transv.
Transv. Joint
Joint
A B
Traffic
Slab
Faulting
• Cumulative ESALs
• Slab thickness
• Joint spacing
• Properties of material
• L d transfer
Load t f efficiency
ffi i
• Climate/environment
• Drainage factor
• Base type
• Width of lane
Spalling
A B C D
Crack Joint
Joint
Transv.
T Transv.
T Transv.
Transv
Joint Joint Joint
Low Sev.:
18m
1,8 Low Sev
Sev.:: C D
2m Moder. Sev.:
2,5 m
High Sev.:
15m
1,5 Traffic
A B
Shoulder
Spalling
z Pavement age
z Joint spacing
z Type of seal
z Dowel corrosion protection
z B
Base ttype
z Climate/environment
Failures in CR Pavements
The number
n mber of failures
fail res is a ffunction
nction of
of:
• Slab thickness
• Percentage of reinforcement steel
• Cumulative ESALs
• Base type
Present Serviceability Rating
For CR pavements,
F t th
the change
h iin PSR is
i a
function of slab thickness, cumulative ESALs
and pavement age
Roughness
Road Deterioration
AIMS AND OBJECTIVES
z SESSION AIMS
z To introduce the key relationships.
z To make participants aware of the key reference documents for
more detailed explanation.
z SESSION OBJECTIVES
z To outline the principles behind modelling of bituminous pavements.
z T identify
To id tif the
th various
i factors
f t which
hi h affect
ff t pavementt performance.
f
z To explain the basis of the various pavement defects and condition
indicators which are modelled in HDM-4, and to highlight the most
i
important
t t explanatory
l t variables.
i bl
z PRACTICALS
z REFERENCES
Overview of the presentation
z Moisture environment
z Rainfall
z Drainage from the pavement structure
z Temperature environment
RAINFALL / MOISTURE CLASSIFICATION
Temperature Classification
z Affected by:
z Pavement materials and thicknesses
z Subgrade
z Drainage regime and seasonality
z Adjusted Structural Number (SNP) is used
z Similar to modified structural number of
HDM-III / AASHO
z BUT contribution of pavement layers /
subgrade weighted by depth
z Relationship
p between SNP and DEF
EFFECT OF DRAINAGE
1.00
DF = 1
0.90
DF = 2
DF = 3
0.80
SNPw / SNPd
0.70
DF = 4
0.60
DF = 5
0.50
MMP = 200 mm/month
0.40
0 0
0 10 20 30 40 50 60 70 80 90 100
ACRA (%)
USE OF DEFLECTION RESULTS
3.5
3.0
n Beam Deflection (mm)
2.5
Granular base
2.0
1.0
0.5
0.0
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Adjusted Structural Number
INTERACTIONS BETWEEN PARAMETERS
Environment
Traffic
Loading
Pavement
Strength Pavement
(annual average) Deformation Roughness
Strength
(crack adjusted)
Structural
Cracking
Rutting
Potholing
ROUGHNESS
SNPd = 3
HS = 50 mm
YE4 = 0.1
5 Environmental component
m = 0.023
hness (IRI, m//km)
Rutting
g component
p
4
Cracking component
Rough
Structural component
2
0 5 10 15 20
Pavement Age (years)
ROUGHNESS (3)
8
SNPd = 6
7 HS = 100 mm Environmental component
YE4 = 1.0
m = 0.023
m/km)
6
Rutting component
ghness (IRI, m
5
Cracking component
Roug
4 Structural component
2
0 5 10 15 20
Pavement Age (years)
DISTRESS INITIATION & PROGRESSION
100
90
80
70
ent)
d Area (perce
Initiation Progression
60
Period Period
50
Affected
40
30
20
10
0
Age of Surfacing
CRACKING
CDS = 1.25
12
CDS = 1.00
eriod (years)
CDS = 0.75
10
All Crack Initiation Pe
6 SNP = 5
4
SNP = 2
0
0.0 0.2 0.4 0.6 0.8 1.0
Axle Loading (MESAL/year)
STRUCTURAL CRACKING PROGRESSION
100
SNPd = 6 SNPd = 3
80 YE4 = 1.0 YE4 = 0.1
uctural Crackking (%)
60
Area of Stru
40
20
All Cracking
Wide Cracking
0
0 2 4 6 8 10 12 14 16
Time Since Crack Initiation (years)
TRANSVERSE THERMAL CRACKING
120
m)
ermal Cracking (No/km
100
80 sub-tropical hot
arid
id
60
sverse The
40 temperate freeze
humid
Trans
20
0
0 5 10 15 20
Surface Age (Years)
RAVELLING INITIATION
18
16 CDS = 1.25
14
d (years)
12
Ravelling Iniitiation Period
CDS = 1.0
10
8
CDS = 0.75
6
0
0 1000 2000 3000 4000 5000
AADT
RAVELLING PROGRESSION
100
80
70
a of Ravelling (%)
60
50
40
Area
30
20
10
0
0 2 4 6 8 10 12 14 16
Time Since Ravelling Initiation (years)
POTHOLING INITIATION
9
Granular base
8
MMP = 100 mm/month
7
Time to Pothole Initiation (years)
HS = 150 mm
6
HS = 100 mm
5
HS = 50 mm
4
HS = 20 mm
3
0
0 1 2 3 4 5 6 7 8 9 10
Traffic (million axles/lane/year)
POTHOLING PROGRESSION
1000
HS = 50 mm
HS = 25 mm
800
HS = 100 mm
year)
oling (no/km/y
600
400
Potho
Granular base
200 ACW = 40%
MMP = 100 mm/month
TLF = 1
0
0 1 2 3 4 5 6 7 8 9 10
Traffic (million axles/lane/year)
EDGE BREAK
90
Edge Step = 50 mm
80
Average Speed = 50 km/h
R i f ll = 200 mm/month
Rainfall / th
70
Width = 3.5 m
m/year)
60
Break (m3/km
50
Width = 4.5 m
40
Edge B
30 Width = 5.5 m
20
10
0
0 500 1000 1500 2000
AADT
RUTTING - DENSIFICATION
Surface layer
y
Subgrade
RUTTING- PLASTIC FLOW
Depression within
Heaving alongside
wheelpath
wheelpath
Asphalt layer
Pavement layer 1
Pavement layer 2
S b d
Subgrade
Mix deformation
STRUCTURAL DEFORMATION
Structural Deformation
With Cracking
Rut Depth
R
Structural Deformation
No Cracking
Initial Densification
First Year
Age
PLASTIC DEFORMATION
120
YE4 = 1.0
Sh = 60 km/h
m)
ormation (mm
100 HS = 100 mm
CDS = 1.25
80
an Rut Depth - Plastic Defo
60
CDS = 1.0
40
CDS = 0.75
Mea
20
0
0 5 10 15 20
Surfacing Age (years)
SURFACE WEAR
40
35 Salted
Unsalted
30
Speed = 80 km/h
25
m)
Rut Depth (mm
20
Speed = 40 km/h
15
10
0
0 2 4 6 8 10
Vehicle Passes with Studded Tyres (million)
PAVEMENT TEXTURE
SCALE OF TEXTURE
SURFACE MACRO MICRO
COARSE HARSH
A TD~ 2mm SFC ~ 0.6
50
COARSE POLISHED
FINE HARSH
FINE POLISHED
z Routine works
z Cyclic and reactive works types
z Periodic works
z Preventive, resurfacing, overlay and
reconstruction works types
z Special works
z Emergency and winter maintenance works
types
z Development works
z Widening,
Wid i realignment
li t and
d new
construction works types
THE IMPACTS OF ROAD MANAGEMENT
z Planning
z Setting standards and policies
z Long term estimates of expenditure
z Programming
z Medium term work programmes
z Preparation
z Detailed project design and work
packaging
z Operations
z Implementation of works in field
DIFFERENCES BETWEEN THE FUNCTIONS
Planning Operations
z Sections/network considerations
z Time horizon
z Staff involved
z Data details
z Computer processing
z U off private
Use i t sector
t
THE MANAGEMENT CYCLE
Aims
A dit
Audit Needs
DATA
Implem. Action
s
Cost/Pr
DIFFERENT COMPUTER SYSTEMS HAVE
DIFFERENT AIMS
Processing
IMPROVEMENTS NEED TO BE SUSTAINABLE
z C lt
Cultural
l issues
i z Training
Sustainability
IMPROVEMENT PRIORITIES
Technical
Institutional
External
HOW DOES HDM-4 CONTRIBUTE?
Annual
7.0 Budget
50%
6.0
hness
80%
50
5.0
Rough
100%
4.0
Target
3.0
1998 1999 2000 2001 2002 2003 2004
PROGRAMME ANALYSIS - Multi-year works
3 Agree objectives
Aims
A dit
Audit Needs
DATA
Implem. Action
s
Cost/Pr
DECISION SUPPORT ACTIVITIES
z Assess Needs
z Road inventory
z Condition and traffic assessment
z Condition and traffic projection
z Determine Actions
z Apply standards
z Identify and select treatments
z Estimate Quantities and Costs
z Assign Priorities under Budget Constraints
z Analyse Past and Future Trends
Treatment selection methods
z NORMS
z Pre-set assumed treatments based on historical data
z SIMPLE RULES
z Responsive to level of distress, perhaps using
up to 2 defects
z Applicable to carriageway and non
carriageway
i defects
d f t
z COMPLEX RULES
z Responsive
R i tot multiple
lti l distresses
di t
z Can be based on optimum intervention levels
z Mainly applicable to carriageway defects
PAVED ROAD - SIMPLE RULES
Drains 10 1 1
RM Surface 507 151 167
Other RM 1 4 4
PM surface 609 780 269
Other PM 28 52 52
TOTAL 1155 988 493
COST
(MTsh)
Holding 22 29 5
Treats (n)
EXAMPLE - COMPLEX RULES
Works type
yp Planningg Programming
g g Design
g
Special Norms S/Norms S
Routine off Norms S/Norms S
c/w
R ti
Routine N
Norms/C
/C S/C S/C
surface
Periodic - S/C S/C S/C
reseal
Periodic - C C NA
rehab
PRIORITIES FOR RURAL ROADS
C1
Track
Costs
Improved road
C2
C3
T1 Traffic T2 T3
IMPORTANCE OF EFFECTIVE PLANNING
99% of y
year,, access established
Marrginal pro
1100
Poor performance - Too plastic
1000
Good in dry weather
900 Slippery with potholes when wet
us
Plasticity modulu
800
700
Too fine Too coarse
Good under both wet and dry conditions
600
500
P
70
DT
oss per yearr per 100 AD
40
(mm)
30
Material lo
20
10
0
0 200 400 600 800 1000 1200
Plasticity modulus
IMPORTANCE OF EFFECTIVE PLANNING
14
12
R o u g h n e sss a fte r g ra d i n g m / k m (IR I)
10
2
2 4 6 8 10 12 14 16
Roughness before grading m/km (IRI)
UNPAVED ROADS: GRADING
TRACTOR TOWED - GRADER BLADE DRAG
Long term average roughness
35
30 No grading
25
20
10
Grading 12 per year
0
0 20 40 60 80 100 120 140 160 180 200
AADT
Costs of Tractor-Towed and Motor Graders
z Light
g and Heavy y Grading
g
Deterioration
zOptimum grading cycles
10
9
8
7
6
5
1 2 3 4 5 6 7 8 9 10 11
Transport Costs & Light
Grading Frequency 40 Vpd (Zw $ per yr)
210000
200000
190000
180000
170000
160000
150000
1 3 5 7 9 11
Optimal Heavy Grading (reprocessing)
150 2 7 45
7.45 17 45
17.45 676
80 3 85
8.5 26 8
26.8 399
z GRAVEL
z BLOCK PAVEMENTS
KENYA - ENGINEERED EARTH ROAD
ETHIOPIA - CINDER GRAVEL ROAD
THAILAND - RURAL ROAD ON INCLINE
ETHIOPIA - WEAK & POORLY DRAINED SOIL CONDITIONS
MONGOLIA - STEPPE TRACK
DETERIORATION MECHANISMS - ??
z Dry weather
z Roughness
z Material Loss
z Rutting
z Looseness
z Impassability
Preservation Development
z Routine z Improvement
p s
z Off carriageway z Widening
z Grading z Realignment
z Spot regravelling
g g z Off-carriageway
g y works ??
z Other surface z Construction
z Periodic z Upgrading
z Resurfacing z New sections
z Special
z Emergencies
z Wi t maintenance
Winter i t
MAINTENANCE WORKS TYPES
z Roughness:
z Progression
z Effect of mechanical compaction
z Effect of grading
z Average annual roughness
z Steady
y state roughness
g
z Material Loss
z Passability
y
ROUGHNESS PROGRESSION - WITHOUT COMPACTION
30
25
20 10 veh/day
IRI (m/km)
15
10
0
0 1 2 3 4 5
Surface Age (years)
Roughness progression at different moisture levels
Effect of Surface Maintenance ??
14
12
R o u g h n e sss a fte r g ra d i n g m / k m (IR I)
10
2
2 4 6 8 10 12 14 16
Roughness before grading m/km (IRI)
REPROCESSING
UNPAVED ROADS: GRADING
ZIMBABWE - TOWED GRADER
TRACTOR TOWED - GRADER BLADE DRAG
Long term average roughness
35
30 No grading
25
20
10
Grading 12 per year
0
0 20 40 60 80 100 120 140 160 180 200
AADT
Material loss: Effect of rainfall and traffic
Material loss: Effect of rainfall and terrain
CONCLUSIONS ?
BITUMINOUS PAVEMENTS
Modelling
g of Works Effects
AIMS
z SESSION AIMS
z To introduce the modelling framework.
z To identify the treatments which are modelled
within HDM-4
z To explain the improvements which result
from implementing maintenance or
improvement activities
z To make participants aware of the key
reference documents for more detailed
explanation.
Road Work Classification
Preservation Development
z Routine z Improvement
p s
z Patching, Edge repair z Widening
z Drainage, Crack sealing z Realignment
z Periodic z Off-carriageway
g y works
z Preventive treatments z Construction
z Rehabilitation z Upgrading
z Pavement reconstruction z New sections
z Special
z Emergencies
z Wi t maintenance
Winter i t
Maintenance Works - Preservation
z Routine z Recurrent
z Periodic z Capital
z Special z Special
z I
Improvementst z C it l
Capital
z Construction z Capital
Works Standards
Section A
[[Operation
p 1]+[Intervention
] [ criteria]+[Limits]+[Effects]+[Cost]="Works
] [ ] [ ] [ ] item 1"
Scheduled
• Fixed intervals of time
• Fixed points in time
Responsive
• Pavement condition
• Pavement
P t strength
t th
• Surface age
• Vehicle speeds
• Traffic volumes / loadings
• Accident rates
Limits
Responsive
R i Cbo = Ca1 Cb1
intervention
Cao Cb4
Cb2 = Ca3
Ca2
0 1 2 3 4 Years
Scheduled intervention
Percent
total
costs
100
0 1 2 3 4 Years
Works Effects and Costs
Note:
Cay = Variable at the beginning of year y
Cby = Variable at the end of year y
Road
variable Cb1 = Ca2 Cb2 = Ca3
Responsive Cbo = Ca1 Cb3
intervention
Cao
Cb4 = Ca5
Ca4
0 1 2 3 4 Y
Years
40
30
0 1 2 3 4 Years
Pavement type resets after maintenance works
Works activity Existing pavement type
Routine works AMGB AMSB AMAB AMAP STGB STSB STAB STAP
I l
Inlay AMGB AMSB AMAB AMAP STGB STSB STAB STAP
Mill & replace to **AP **AP **AP **AP N/A **SB **AB **AP
intermediate
surface layer
Mill & replace to **GB **SB **AB **AP **GB **SB **AB **AP
base
Source: NDLI (1995
Routine Maintenance Operations
z DF affects
ff t ratio
ti between
b t wett and
d dry
d
season pavement strengths
z Better drainage means higher average
pavement strength
z Higher pavement strength means lower
roughness progression
Effect of Drainage on Roughness
12
AADT = 2,000
SNPd = 3
3.0
0
MMP = 100 mm/month no drainage
10 WET = 0.5 maintenance
hness (IRI, m/kkm)
rehabilitate
when DF = 5
8 annual
maintenance
DF = 2
Rough
2
0 5 10 15 20
Years
Sealing of Wide Structural Cracking
10
AADT = 2,000
SNPd = 3.0 maintain drainage
MMP = 100 mm/month
/ th no crack sealing
WET = 0.5
8
m/km)
no drainage maintenance
ughness (IRI, m
6
Rou
maintain drainage
4 seal all wide cracks
2
0 5 10 15 20
Years
Annual Areas of Crack Sealing
30 100
Seal 100% Wide Cracks
90
25
CSL = 3 years 70
a of Crack Sealing
CSL = 5 years
20
60
15 50
40
Annual Area
CSL = 10 years
10
30
20
5 Cumulative
Wide Cracking
10
0 0
0 2 4 6 8 10 12 14 16 18 20
Age (years)
Effect of Sealing Percentage
8
AADT = 2,000
SNPd = 3.0
MMP = 100 mm/month
WET = 0.5 seal 25% wide cracking
m/km)
6
ughness (IRI, m
2
0 5 10 15 20
Years
Sealing of Transverse Thermal Cracking
8
AADT = 2,000
SNPd = 3.0
MMP = 100 mm/month
/ th
WET = 0.5 seal 100% wide cracking
HS = 50 mm
m/km)
6
ughness (IRI, m
2
0 5 10 15 20
Years
Repair of Pavement Edge Break
z Preventive treatments
z Resurfacing
z Overlays
z Reconstruction
Preventive Treatments
z Either proactive
Eith ti as preventive
ti treatments
t t t
or reactive to low levels of surface
distress
z Includes surface treatments, slurry seals
or thin
thi overlays
l
z Resurfacings do not give a significant
increase in pavement strength
Overlays
150
150mm G
Granular
l S Sub-base
bb
0 – reset to zero
ero
N – no reset
Effects on Cracking and Ravelling
12
10
Surface treatment
IRI After Seal (m/km)
0
0 2 4 6 8 10 12
IRI Before Seal (m/km)
Roughness Effects - Seals with Shape Correction
12
10
Thickness including shape correction 10mm
m/km)
8 30mm
RI After Seal (m
50mm
6
IR
0
0 2 4 6 8 10 12
IRI Before Seal (m/km)
Roughness Effects - Overlays
8
Overlay thickness 20mm
7
40mm
IRI After Overlay (m/km)
6
60mm
5
80mm
4
0
0 2 4 6 8 10 12 14
IRI Before Overlay (m/km)
Roughness Effects - Reconstruction
z Lane addition
z Partial widening - no increase in number
of lanes
z Widening does not change alignment
Effects of Realignment
Work
Inventory Programs
Projects
Survey
y
Etc.
RDBMS HDM-4
Calibration/Adaptation Activities
z Vehicle Fleet
z Works Standards
z HDM 4 Configuration Data
HDM-4
z Model Calibration
z R
Reports
t
z Language
Calibration/Adaptation Activities
- Vehicle Fleet Calibration
z Works types
yp
z maintenance
z improvement
z Unit costs
z Construction quality
q y
z material properties
z construction quality
q y
z compaction, etc.
Calibration/Adaptation Activities
(e.g. Bangladesh Work Types)
z Works types:
z SBST
z DBST
z Thin Overlay
z Thick Overlay
z Rehabilitation
z Widening
z Intervention levels
Calibration/Adaptation Activities
- HDM-4 Configuration Data
Typical RD relationship:
d C = Kcpw CRP zw [(
dACW [(zw a4 a5 δtW YE4 (S
(SNPa6) + SCW
SC a5 )1/a5
CDS - SCW]
( h
(change in
i area off wide
id structural
t t l cracking)
ki )
Calibration/Adaptation Activities
- RD Model Calibration
Calibration Model
Factor Coefficients
Calibration/Adaptation Activities
- RD Model Calibration
RD calibration factors:
Calibration/Adaptation Activities
- RD Model Calibration
RD model coefficients:
Calibration/Adaptation Activities
- Other Activities
z Reports
z Special report requirements
z Using user-defined reports facility
z Language
z Is native language version required?
z User interface
z Reports
z Documentation, etc.
Calibration/Adaptation Activities
- Revised Data Flow Schematic
Work
Inventory Programs
Projects
Survey
y
Special
Calibration Reports
Factors RDBMS HDM 4
HDM-4 Etc.
Interfacing HDM-4 with External Systems
Work
Inventory Programs
Projects
Survey
Special
S i l
Calibration Reports
factors HDM 4
HDM-4 Etc
Etc.
RDBMS
S
Generating the Export File
- Outline Procedure
Selected Representative
RDBMS Section Look-up
Links Table
Generating the Export File
- Export Options
Generating the Export File
- Outline Procedure
P bl
Problem
Road networks in HDM-4 version 1.0 only
support TOTAL motorised and NMT traffic,
traffic and
not traffic by vehicle type.
Solution
Don t use a road network export file to transfer
Don’t
data - use a programme analysis export file.
Interfacing HDM-4 with External Systems
Any Questions?
Institutional Impacts of Implementing HDM-4
z Should we:
z adapt
d t HDM-4
HDM 4 tto suit
it local
l l methods*,
th d * or
z adapt local methods to suit HDM-4?
z Planning
z Setting aims and objectives
z Determining standards and policies
z Estimating long term expenditure
z Programming
z Medium term work programmes
z Preparation
z Detailed project design and work packaging
z Operations
z Implementation of works in field
THE MANAGEMENT CYCLE
Aims
A dit
Audit Needs
DATA
Implem. Action
s
Cost/Pr
DECISION SUPPORT ACTIVITIES
z Assess Needs
z Road inventory
z Condition and traffic assessment
z Condition and traffic projection
z Determine Actions
z Apply standards
z Identify and select treatments
z Estimate Quantities and Costs
z Assign Priorities under Budget Constraints
z Analyse Past and Future Trends
QUESTIONS
z Reliability of results
z Application of results
z Calibration focus
z RUE/RDWE calibration
z Calibration levels
z Reliability concepts
z What to focus on
z Information Quality Levels
Reliability of Results Depend On
z Input Data
z Must have a correct interpretation of the
input data requirements
z Have a quality of input data appropriate for
the desired reliability of results
z C lib ti
Calibration
z Adjust model parameters to enhance the
accuracy of its representation of local
conditions
Steps in Analysis
Predictive
Input Data Analysis Results
Models
This Presentation
Calibration Focus
180
2-Axle Truck 5-Axle Truck 7-Axle Truck
160
140
120
Crew
Cost in cents/km
100 Interest
Depreciation
80 Maintenance
Tires
40
20
0
Observed HDM Observed HDM Observed HDM
RDWE CALIBRATION IN ZIMBABWE
z Roughness Progression
z Crown height (and earthworks quantities)
z Drainage quality
z Design standard
z Unstable subgrades (Expansive and Dispersive)
1 0 MESA
1.0 No data No data (?) 05
0.5
INDONESIA - ROUGHNESS CALIBRATION
CRP
dACW = Kcpw [(zw a1 a2 δtW YE4 ((SNPa3) + SCWa4 )1/a5 -
zw [(
CDS SCW]
Calibration Model
Factor Coefficients
Calibration Levels
Time Required
Experimental
Years
Surveys and
Research
Months
Field Surveys
Weeks
Desk Studies
Resources
Limited Moderate Significant q
Required
z Comprised of
z Improved data collection
z Fundamental research
z Leads to more accurate data by observing
over long time period
z Often leads to alternative local
relationships
Reliability Concepts
Data
Predicted
Predicted
Data
Observed Observed
dicted
Pred
Pred
Data
Data
Observed Observed
Correction Factors
A Rotation B Translation
Translation
Rotation
Observed = Predicted Observed = Predicted
Predicted
Predicted
Data Data
Observed Observed
Translation
Rotation and
C Translation
Rotation
Observed = Predicted
Prediccted
Data
Observed
Bias and Precision in Input Data
frequency
Bias
Population Sampled
value
frequency Precision
Sampled
P
Population
l ti
value
Important Considerations
z Simulation of Past
z take sample of roads with historical data
(traffic, design, etc.)
z simulate deterioration from construction to
current age
z compare results
z Average predicted condition should be
similar to current condition
RDWE Calibration - 2
z Controlled Studies
z collects detailed data over time on traffic,
roughness, deflections, condition, rut
d th
depths
z sections must be continually monitored
z long term (5 yr) commitment to quality data
long-term
collection
Road User Effects
Regression to
6000 tariff data
5000
Tariff in Rs/trip
4000
2000
1000
0
0 100 200 300 400 500 600 700 800
Trip Length in km
RUE - Fleet Surveys
Impact
p Sensitivity
y Impact
p
Class Elasticity
High S-I > 0.50
Medium S-II 0.20 – 0.50
Low S-III 0.05 – 0.20
Negligible S IV
S-IV <0
0.05
05
Sensitivity Impact Variables or data important Variables or data important
class2 elasticity for total VOC3 for VOC savings4
Replacement vehicle price Replacement vehicle price
S-I > 0.50 Parts model exponent kp Parts model exponent kp
Assume
Required Priority 1 Priority 2
Defaults
Mass
Power
Capacity
Tyre Parameters
Level 1 Unit Costs Speed
Depreciation
All Others
Service Life
Parameters
Utilisation
Fuel
Speed
S d
Power
Level 2 Capacity
Service Life
All Others
PCSE
Utlilisation
Priorities
Depend on
Level 3 Resources and
Objectives
Bituminous Pavement Deterioration Priorities
Deterioration factor Impact class for Impact Typical Net Sensi-
given criteria 1 Elast- values Impact tivity
icity of class
Factor (%)
1 2 3 4 5 6
Roughness-age- D D B C B B 0.20 0.2 - 5.0 10
environment
Condition Confidence
Interval
Existing
Condition Criteria
for
1 Intervention
Condition Condition
2 1 2
Trigger Trigger
Interval Interval
Time
Information Quality Levels - IQL
HIGH LEVEL DATA
System Performance
IQL-5 Performance
Monitoring
Planning and
IQL-4 Structure Condition
Performance Evaluation
Programme Analysis or
IQL-3 Ride Distress Friction
Detailed Planning
Project Level or
IQL-2 Detailed Programme
Project Detail or
IQL-1
Research
Item 1 2 3 4
Roughness Profile elevations IRI Precise IRI IRI
W b d analysis
Waveband l i approx estimate
ti t
IRI
Cracking Type, Extent, severity, Type, extent Type and Extent
intensity & location and severity extent
z Yes, if calibrated
z HDM has proved suitable in a range of
countries
z As with any model, need to carefully
scrutinise output against judgement
z If unexpected predictions problem with (a)
data (b) calibration (c) the models, or (d)
your judgement
j d t
PRACTICAL: CREATING A ROAD
PRACTICAL
NETWORK & VEHICLE FLEET
Road Network - Key Skills / Points
z C eat g a Road
Creating oad Network
et o
z Creating a Road Section
z Concept of homogeneous section
z Use of aggregate data
z Categories of road section data
z ALL road data is mandatory
z Re use of existing sections
Re-use
z Use of representative section set
z NB importing of road data
NB.
Road Section Data Categories
Road Section
Data
Calibration
Road Network - Documentation
4 Road Networks
4.1 Keyy concepts
p 7
4.2 Software overview 7
4.3 Creating new Road Networks with no Sections 9
4.4 Creating Road Networks based on existing networks 10
45
4.5 Viewing / editing Road Networks 10
4.6 Deleting Road Networks 11
4.7 Renaming Road Networks 11
4.8 Exporting Road Networks 12
4.9 Importing Road Networks 12
4.10 Adding Sections based on aggregate data 13
4.11 Adding Sections based on existing Sections 14
4 12
4.12 Editing Section attributes 15
4.13 Specifying which Sections are displayed in the 17
Road Network window
4.14 Editing Section details directly from the Road 18
N
Networkk window
i d
4.15 Deleting Sections 18
Vehicle Fleet - Key Skills / Points
Vehicle
V hi l Type
T
Data
Calibration
Vehicle Fleet - Documentation
5 Vehicle Fleets
5.1 Key Concepts 20
5.2 Software overview 21
5.3 Creating Vehicle Fleets 22
5.4 Creating Vehicle Fleets based on existing fleets 23
5.5 Viewing / editing Vehicle Fleets 23
5.6 Deleting Vehicle Fleets 25
5.7 Renaming Vehicle Fleets 25
5.8 Importing Vehicle Fleets 26
5.9 Exporting Vehicle Fleets 26
5 10
5.10 Adding motorised Vehicle Types (MT) to a 27
Vehicle Fleet
5.11 Adding non-motorised Vehicle Types (NMT) 28
to a Vehicle Fleet
5.12
5 12 Editi Vehicle
Editing V hi l Type
T attributes
tt ib t 29
5.13 Specifying which Vehicle Types are displayed in 32
the Vehicle Fleet window
5.14 Editing Vehicle Type attributes directly from the
Vehicle Fleet window 33
5.15 Deleting Vehicle Types 34
RD Inputs and Outputs
Climate zones
Road network data
z Road class
z Speed-flow types
z Traffic flow pattern
z Aggregate variables
z Default data
Currency
CLIMATE ZONES
Road class
z functional hierarchy
y to facilitate management
g or
analysis of the road network
Speed-flow types
z models the effects of traffic volume on speed and the
economic consequences of road improvements
Traffic flow pattern
z models traffic congestion at different hours of the
day and on different days of the week and year
day,
Aggregate data (defaults)
z pavement condition
condition, strength,
strength road geometry,
geometry etc
etc.
ROAD CLASSES
Speed (km/hr)
S1
S2
S3
S lt
Sult Flow
Qo Qnom Qult PCSE/h
DEFINING A SPEED-FLOW TYPE
Name
Road type
Ultimate capacity (Qult) in PCSE/lane/h
Free flow capacity as a proportion of Qult
Nominal capacity as a proportion of Qult
Jam speed
p at capacity
p y in km/h
TRAFFIC FLOW PATTERN
Flow Periods
Peak
Flow Next to Peak
Medium flow
Next to Low
Low/Overnight
Name
R d use
Road
Number of flow-frequency periods
Number of hours in each flow-frequency
period (HRYRp)
Percentage of AADT in each flow-
frequency period (PCNADTp), or
Hourly flow in each period (HVp) as a
proportion
p p of AADT
AGGREGATE DATA
User defined
z Traffic levels: e
e.g.,
g low
low, medium
medium, high
high.
z Geometry class: in terms of parameters reflecting
horizontal and vertical alignment
z Pavement characteristics: structure and strength
parameters defined according
p g to pavement
p surface
class
z Road condition: ride quality, surface distress and
surface texture
z Pavement history: construction quality, pavement
age etc.
age, etc
DEFINING TRAFFIC BANDS
Defined by:
y
z Description e.g., high, medium, low
z Pavement surface class (e
(e.g.,
g bituminous
bituminous,
concrete or unsealed)
z Mean AADT
Unlimited number of traffic bands
DEFINING A ROAD GEOMETRY CLASS
Defined by :
z Description e.g., mostly straight and gently
undulating, bendy and winding, etc.
z A
Average rise
i plus
l fall
f ll (m/km)
( /k )
z Average number of rises and falls (no./km)
z Average horizontal curvature (deg/km)
z Average super-elevation (%)
z Other speed related data (e.g., speed limit)
Unlimited number of geometry classes
BITUMINOUS PAVEMENT STRENGTH
Method 1:
z Define several ranges
g of SNP values
z For each range and for each pavement type
define HSNEW, HSOLD, HBASE
z Specify default surface material for each
surface type (i.e., AM or ST), and resilient
modulus for stabilised road base type
Method 2 (HDM-4 Default):
z Layer thickness
L thi k derived
d i d usingi pre-defined
d fi d
range of SNP values, layer strength coefficients
for surfacing
g and base material,, and subgrade
g
CBR of 8%
CONCRETE PAVEMENTS
z Sponsors:
z World Bank
z Dept for International Development
z Asian Development Bank
z SNRA & Others
z I progress for
In f 5 years comprising
i i
z technical studies
z software development and testing
z PIARC leading the implementation phase
Objective:
Pavement standards
Alignments
Objective:
mproved framework for investigating road investments
Non-motorised
transport
Traffic congestion
Vehicle emissions
Travel times
Transport costs
Road accidents
HDM-4 Applications
Road pricing
p g
z fuel levies
z congestion charges
z weight distance charges
weight-distance
Vehicle regulations
z axle load limits
z energy consumption, vehicle emissions & noise
Engineering Standards
z sustainable road network size
z pavement design and maintenance standards
Analytical Framework
z Strategy
gy
z Programme
z Project
Network Condition Forecasts
Secondary
Roads
Roughn
5.0 $35m/yr
4.0
0 Trunk
Roads
$20m/yr
Good 3.0
1998 1999 2000 2001 2002 2003 2004
Effect of budget levels
Annual
7.0 Budget
50%
6.0
hness
80%
50
5.0
Rough
100%
4.0
Target
3.0
1998 1999 2000 2001 2002 2003 2004
Optimal budget requirements
250
200
Development
150
Improvement
Periodic
100
Routine
50
0
2000 2001 2002 2003
Multi-year programme
z Routine maintenance
z Drainage effects
z Periodic maintenance,, including g recycling.
y g
z Road improvements, including widening,
lane separation, diverted traffic.
z Intervention criteria, including road user
effect parameters
Integrating HDM-4 within
Road Management Systems
Long Term
Road
R d St t i
Strategic
Inventory Road Plan
DATABASE HDM-4
Road Multi -Year
Year
Condition Work
HDM4 Program
Traffic
Data
Detailed
Project Level
Bridges
Appraisal
Inventoryy
Bridges Policies
Condition Standards
Research
Summary
Long Term
Road Strategic
Inventory
DATABASE HDM--4
HDM Road Plan
Bridges Policies
P li i
Condition Standards
Research
INPUTS OUTPUTS
Data Collection
Road inventory
z Road
R d links,
li k class,
l pavementt types,
t
z Geometric characteristics (widths,
shoulders etc)
shoulders,
Pavement condition
z Roughness surveys (automated)
z Visual surveys (manual)
z Deflections (sampled once in 5 years)
Traffic data
Bridges
Calibration/Adaptation Activities
Vehicle Fleet
Unit Costs
Works Standards
HDM-4 Configuration Data
Model Calibration
Reports
L
Language
Linking HDM-4 to External Systems
Work
Inventory Programs
Projects
Surveys
DXF Special
Calibration R
Reports
t
factors Etc.
RDBMS HDM - 4
Vehicle Fleet Calibration
Representative
p Vehicle Types
yp
Basic Characteristics (Level 1)
z mass, loading, speed, vehicle life, utilisation
Unit Costs
Detailed Calibration (Level 2)
z Tyre wear, depreciation, fuel consumption
Works types
yp
z maintenance
z improvement
Unit costs
Construction quality
q y
z material properties
z construction quality
q y
z compaction, etc.
Bangladesh Work Types
Works types:
z SBST
z DBST
z Thin Overlay
z Thick Overlay
z Rehabilitation
z Widening
Intervention levels
HDM-4 Configuration Data 1
Reports
z Special report requirements
z Using user-defined reports facility
Language
z Is native language version required?
z User interface
z Reports
z Documentation, etc.
Interfacing HDM-4 with External Systems
Work
W k
Inventory Programs
P j t
Projects
Survey
Special
Calibration Reports
Factors RDBMS HDM-4 Etc.
Generating the Export File
Selected Representative
RDBMS Section Look-up
Links Table
The End
Questions?
Institutional Impacts of Implementing HDM-4
Should we:
z adapt
d t HDM-4
HDM 4 tto suit
it local
l l methods*,
th d * or
z adapt local methods to suit HDM-4?
z software basics
z installation
z user interface basics
z software structure
z i iti l configuration
initial fi ti
z getting help / support
z documentation
Software Basics - Software Description
"Educational
Educational Purposes
Purposes" means the use of ISOHDM
Controlled Products in formal vocational and
professional teaching courses (for example by
f
faculty
l and d students
d off universities
i i i and
d technical
h i l
training institutes), undergraduate and
postgraduate
p g project,
p j , thesis or research activities,,
and formal HDM-4 training activities. It does not
include consulting activities conducted by
professional researchers and educators
educators, which are
not primarily educational in nature.
Installation - Basics
z Release types:
z Version
z Update
z Upgrade
z Distribution types:
z CD
z Internet download
z Username / Software Licence Key (SLK)
Installation - Instructions
z Autorun screen
z User registration details
z HDM-4 program group / menu
z Readme file
z Uninstalling
z Documentation
User Interface Basics - Starting HDM-4
z IInputt data
d t
z Using multiple databases (§12.1)
z O t t / run data
Output d t
z Rundata ‘Clean-up’ facility
z Importing / exporting data
z Backing-up (§12.3)
z Reports (§10)
z Log-file
Initial Configuration
z On-line Help
z Documentation
z Distributor
z User Groups / Networks
z Web-sites / Knowledge base
z Technical Support
pp (inc.
( UoB))
User Support - Web Sites Key Facilities
z Knowledge Base
z Frequently Asked Questions (FAQs)
z Technical Articles
z Implementation projects
z Discussion / News groups
z Links
User Support / Getting Help
z On-line Help
z Distributor
z User Groups / Networks
z Web-sites / Knowledge base
z Technical Support (inc. UoB)
z Documentation
Documentation - Basics
z 1 - Overview
z 2 - Applications Guide
z 3 - Software User Guide
z 4 - Analytical Framework & Model Descriptions
z 5 - A Guide to Calibration & Adaptation
z 6 - Modelling Road Deterioration & Works Effects
z 7 - Modelling Road User & Environmental Effects
Documentation - Hands-on
z software basics
z installation
z user interface basics
z software structure
z i iti l configuration
initial fi ti
z getting help / support
z documentation
Any questions?
Improving Analysis Performance
Linda Parsley
HDM-4 SYSTEM ARCHITECTURE
z Strategy analysis
z For
o preparing
p epa g medium
ed u to long
o g term
te
planning estimates of funding needs for
road network development & maintenance
z Programme analysis
z For preparing single or multi-year work
programmes underd budget
b d t constraints
t i t
z Project analysis
z For evaluating
F l ti the
th economic i or
engineering viability of different road
investment projects
HDM-4 ANALYTICAL FRAMEWORK
$ Total
Optimum
Road
User
Construction/
Rehabilitation
Maintenance
R d St
Road Standard
d d
ANNUAL EFFECTS & COSTS
z Input data
z Start loop
z calculate traffic
z estimate road deterioration
z Return loop
z Repeat for each analysis year
TRAFFIC ESTIMATES
z Volume AADT
z Composition and growth rates
z Categories
g - Normal,, Diverted,, Generated
z Axle loading
z Road capacity
p y and speed-flow
p relationship
p
z Traffic flow pattern
z Degree of conflict between MT and NMT
z Roadside friction
VEHICLES
z Categories:
z Motorised transport (MT)
( )
z Non-motorised transport (NMT)
z Cl
Classes
z For example: motorcycles, cars, buses,
trucks
z Types
VEHICLE DATA
z General
z Name, category, class, type
z Utilisation
z Performance
z Loading g characteristics
z Unit costs
z RUE
U Ca Calibration
b at o pa
parameters
a ete s
z Sections
z a length of road that is homogeneous in
terms of its physical attributes (basic unit
off analysis)
l i )
z Network
z S b
Sub-networks
t k
z Nodes
z Links
ROAD SECTION DATA
z General
z ID, inventory, traffic, road use and class
z Geometry
z Pavement characteristics
z Road condition
z Calibration parameters
z Surface category
z Paved
z Unpaved
z Surface class
z bituminous
z concrete
z block
z unsealed
z Pavement type
ROAD WORKS
z MAINTENANCE / PRESERVATION
z Routine - undertaken each year
z Periodic - undertaken at intervals of years
z Special - emergencies, winter maintenance
z DEVELOPMENT
z improvement - widening, realignment
z new construction
WORKS STANDARDS
Section A Alternative 2
z End loop C
z Calculate
C l l t nett benefits
b fit over all
ll the
th road
d sections
ti
z End loop B
z Calc late economic indicators (NPV,
Calculate (NPV IRR,
IRR etc
etc.))
z End loop A
z P f
Perform optimisation
ti i ti (Programme
(P & Strategy)
St t )
PROJECT LEVEL OUTPUT
z Economic indicators
z NPV, IRR, FYRR, BCR,..
z Sensitivity analysis results
z Scenario analysis
z Road condition indicators
z Road user cost details
z Energy
e gy & e emissions
ss o s
PROGRAMME ANALYSIS
z Method
z Life Cycle Analysis
z Budget Period Analysis
z Objective functions
z maximise NPV
z maximise road network condition
improvement
z Constraints
z budget limits (capital, recurrent, etc.)
PRIORITISATION
Road
R d Section
S ti
Data
Calibration
Road Network - Documentation
4 Road Networks
4.1 Keyy concepts
p 7
4.2 Software overview 7
4.3 Creating new Road Networks with no Sections 9
4.4 Creating Road Networks based on existing networks 10
45
4.5 Viewing / editing Road Networks 10
4.6 Deleting Road Networks 11
4.7 Renaming Road Networks 11
4.8 Exporting Road Networks 12
4.9 Importing Road Networks 12
4.10 Adding Sections based on aggregate data 13
4.11 Adding Sections based on existing Sections 14
4 12
4.12 Editing Section attributes 15
4.13 Specifying which Sections are displayed in the 17
Road Network window
4.14 Editing Section details directly from the Road 18
N
Networkk window
i d
4.15 Deleting Sections 18
Vehicle Fleet - Key Skills / Points
Vehicle
V hi l Type
T
Data
Calibration
Vehicle Fleet - Documentation
5 Vehicle Fleets
5.1 Key Concepts 20
5.2 Software overview 21
5.3 Creating Vehicle Fleets 22
5.4 Creating Vehicle Fleets based on existing fleets 23
5.5 Viewing / editing Vehicle Fleets 23
5.6 Deleting Vehicle Fleets 25
5.7 Renaming Vehicle Fleets 25
5.8 Importing Vehicle Fleets 26
5.9 Exporting Vehicle Fleets 26
5 10
5.10 Adding motorised Vehicle Types (MT) to a 27
Vehicle Fleet
5.11 Adding non-motorised Vehicle Types (NMT) 28
to a Vehicle Fleet
5.12
5 12 Editi Vehicle
Editing V hi l Type
T attributes
tt ib t 29
5.13 Specifying which Vehicle Types are displayed in 32
the Vehicle Fleet window
5.14 Editing Vehicle Type attributes directly from the
Vehicle Fleet window 33
5.15 Deleting Vehicle Types 34
PRACTICAL
Procedure:
Create a Road Network
z Create several road sections and enter data
Create a Vehicle Fleet
z Create several vehicle types and enter data
Create a sample project
z Enter required data and run HDM-4
Review typical outputs
Programme Analysis
Programme Analysis
z Method
z Life Cycle Analysis
z Budget Period Analysis
z Objective functions
z maximise NPV
z maximise road network condition
improvement
z C
Constraints
t i t
z budget limits (capital, recurrent, etc.)
Analysis Procedure
Independent Projects:
z unrelated road projects e.g. from different
parts of the country,
p y,
Mutually Exclusive Projects:
z alternatives to one project where only one
can be selected.
Capital Budgeting Rules
G 500,000
, 78,000
, 0.156 4
Incremental Analysis
G 20 000
20,000 1 200
1,200 0 0600
0.0600 6
H2 100 000
100,000 16 829
16,829 0 1683
0.1683
Efficiency Frontier
∆ BENEFITS
Opt
p 4
Opt 3 O t5
Opt
Opt 2
Opt 1
Opt 0
∆COST
Efficiency Frontier
BENEFITS
A4 A5
A3
A2
B3 B4
Priority by
B1 slope:
B2 1. B1
2 A1
2.
3. A2
A1 4. A3
A, B - Independent Projects 5. B3
1,2,3,.. - Mutually Exclusive 6. A4
7. A5
PROJECT COST
Budget Period Analysis
Budget Period
TIME
Multi-Year Work Programme
Road
R dNNo. (km))
(k Work
W k Type
T Year
Y ($m))
($ Σ($m))
Σ($
A10031 12.5 Reconstruct 2001 5.4 5.4
A10004 23.4 Resurface 2001 2.9 8.3
B30106 19 8
19.8 Reconstruct 2001 86
8.6 16 9
16.9
A20122 5.5 Patch&Reseal 2001 1.4 18.3
: : : : : :
A11028 34.5 Reconstruct 2002 13.8 13.8
C34567 22.6 Reconstruct 2002 9.7 23.5
: : : : : :
B32109 15.0 Resurface 2003 4.2 4.2
C76543 36.9 Overlay 2003 24.1 28.3
: : : : : :
Strategy Analysis
Strategy Analysis
Objectives:
z Determine budget allocations for road
maintenance and improvement
z Prepare for work programmes
z Determine long term network performance
z Assess impact on road users
Procedure
z Categorise road network into matrix cells
z D fi representative
Define t ti traffic
t ffi volume
l and
d loading
l di
z Define maintenance and improvement standards
z Specify budget constraints
z Model pavement deterioration for each matrix
cell
ll
z Apply maintenance and improvement standards
z Calculate road user benefits
z Select maintenance standards which optimise
user benefits
b fit
Road Network Matrix
z Recurrent budget
z Routine maintenance
z Top sliced from specified budget
z Capital budget
z New construction, upgrading, rehabilitation, etc.
z Multi-year limit
z Optimisation methods
z Maximise NPV
z Maximise network condition improvement
z Minimise cost for target network condition
Pavement Deterioration and Maintenance
Condition
Reconstruct
Overlay
Traffic / Time
Maintenance and improvement costs
W t
West 27 2
27.2 20 4
20.4 19 5
19.5 12 3
12.3
Budget Allocation
UNIT AIMS
1. To introduce the modelling framework used in HDM-4 so that the HDM-4 applications can be described
accurately by trainers.
UNIT OBJECTIVES
1. State the main applications of HDM-4 and their relevant functions in road management.
2. Explain the analytical framework in terms of the modelling units, e.g. road deterioration, road work effects,
road user effects, etc.
3. Describe the concept of homogeneous sections, road links, sub-networks and networks.
4. Describe the pavement and vehicle classification system (i.e. pavement and vehicle types).
5. Describe the hierarchy embodied in the analysis loop.
6. Illustrate the key concept of section alternatives through a flow diagram.
GUIDELINES
It is important that the modelling framework is described to Trainees in such a way that they appreciate the
complexity of trying to model the future performance of roads and the impact on road users. HDM-4 essentially
comprises a set of mathematical models that have been put together in a coherent analytical framework in order
to provide useful road management information. The mathematical models are for predicting road deterioration
and the effects of road works, for calculating the effects of road characteristics on road user costs, and for
calculating other socio-economic effects (energy, emissions, noise, etc).
The HDM-4 models are applied to homogeneous road sections. This assumes that pavement deterioration will
be the same on the entire road section. Consequently, road user effects will be constant for that section.
Characteristics of the road section are defined in terms of the pavement type which is user defined and the type
of vehicles using the road. The modelling framework applies the concept of life cycle analysis to calculate over
a number of years (typically, 15 ~ 40 years) the road deterioration, works effects and road user effects for each
section-alternative.
The above concept is the key to understanding HDM-4. The applications (Project, Programme and Strategy) are
simply multiples of the analytical framework, i.e. a project analysis comprises one of more homogeneous
sections-alternatives, programme analysis comprises several projects and strategy analysis is simply a model of
the entire road network broken down into representative sections.
MATERIALS
1. PowerPoint Slides (T3 Modelling Framework.PPT)
2. Handout (T3 Modelling Framework Handout Notes.DOC)
REFERENCES
1. HDM-4 Volume 4: Analytical Framework and Model Description
UNIT TITLE: HDM-4 PROGRAMME ANALYSIS CASE STUDY
UNIT AIMS
1. To introduce the HDM-4 Programme Analysis module concepts so that Trainers and Trainees are able to
prepare annual or multi-year work programmes.
2. To introduce the concept of incremental benefit/cost ranking method so that Trainers appreciate optmisation
techniques.
3. To familiarise Trainers with the use of HDM-4 data input and output modules.
UNIT OBJECTIVES
1. To describe the concepts of the HDM-4 Programme analysis application.
2. To apply the incremental benefit cost optimisation techniques through simple manual calculation.
3. To use the HDM-4 model to generate a work programme.
4. To alter maintenance standards and unit costs and study the impact on budget requirements.
GUIDELINES
KEY POINTS
REFERENCES
UNIT TITLE: DERIVATION OF OPTIMUM MAINTENANCE STANDARDS
UNIT AIMS
1. To describe the specification and assignment of HDM-4 maintenance standards so that Trainers are able to
apply a range of maintenance works.
UNIT OBJECTIVES
1. To prepare input data required for investigating optimum road maintenance standards for paved roads using
HDM-4.
2. To assign a range of maintenance standards to road sections and run HDM-4.
3. To interpret the results and recommend optimum maintenance standards for paved roads.
GUIDELINES
This case study presents the economic analysis of alternative maintenance standards for two paved road sections
that are 36.5 km and 10.6 km long, respectively. The existing roads have surface dressing and asphalt concrete
pavement surfacing in good condition, and carry approximately 6 000 and 16 000 vehicles per day (AADT),
respectively. The objective of this case study is to determine the optimum maintenance intervals for resealing
(surface dressing) and for overlays on the two roads, respectively.
The case study data is provided in the T3 Case studies workspace/database installed with the HDM-4 software.
The data for this case study are located in the Projects folder in the Case Studies Workspace. The name of
the case study is Case 2: Optimum maintenance intervals. To locate the data for this case study:
! Select the T3 Case Studies Workspace
The case study data includes the road network data set (CASE2 Project Sections) and the vehicle fleet (CASE2
Vehicle Fleet) required for the analyses.
The first step in the procedure would be to calibrate the road sections using the calibration factors derived from
the previous case study. The second step in the procedure should be to create additional maintenance
standards for the two roads. The data provided includes one example each for:
! Surface dressing at 5 year intervals (scheduled), and
! Overlay 50mm at 6 IRI (responsive)
Create additional maintenance standards by copying the above to create separate standards for:
! Surface dressing intervals (scheduled) at 3, 7, 9 and 11 years
! Overlay intervals (responsive) at 4, 8, 10 and 12 IRI
KEY POINTS
The optimum economic maintenance standards are defined as the maintenance interval (scheduled or
responsive) with the maximum NPV. This can be illustrated by exporting the Benefit Cost Ratios report to an
Excel spreadsheet and plotting the NPV against the maintenance interval (years) or intervention criterion (IRI).
Another useful plot would be of the NPV against the increase in agency cost. The resulting curve from this is
defined as the efficiency frontier. This will be described in detail in the Programme Analysis case study.
Those Trainers who complete the above exercise within a short time could attempt to derive the optimum
grading and regravelling frequencies for the Gravel road that is included in the Road Network database.
MATERIALS
1. Project Case Study 2 Handout
2. Project Case Study 2 HDM-4 Workspace
REFERENCES
1. HDM-4 Volume 2: Applications Guide (Appendix A)
2. HDM-4 Volume 3: Software User Guide (Chapter 7)
3. HDM-4 Volume 5: A Guide to Calibration and Adaptation (Chapters 6 and 7)
4. T3 Case Study 1 Handouts
UNIT AIM
1. To introduce the concept of HDM-4 road investment appraisal and its applications in road management so
that the HDM-4 system can be properly applied within road management.
UNIT OBJECTIVES
1. Describe the concept of life cycle analysis that is used in road investment appraisal.
2. Describe the HDM-4 applications (Strategy, Programme and Project Analysis).
3. Explain the role of the HDM-4 applications within the road management cycle.
4. Illustrate the HDM-4 applications by using sample output.
GUIDELINES
This is a key unit that introduces the HDM-4 life cycle analysis concept and its applications within road
management. Life cycle analysis involves modelling the future performance of roads together with the impact of
road characteristics on user costs. Various titles are often used to describe the same concept; whole life cycle,
total transport costs, road investment appraisal, etc. It is important that trainers should stress that these all refer
to a common modelling framework.
The road management process is covered in another Unit. Trainers should refer to that unit in order to be able to
link HDM-4 applications with the management functions of planning, programming and preparation. The
operations management function is not explicitly covered by HDM-4.
With regard to the HDM-4 applications (i.e. use of HDM-4 for specific purposes), the three applications are;
Strategy, Programme and Project analyses. The three applications differ in terms of scope and data detail.
Strategy analysis deals with an entire road network with input data that can be at an aggregate level (e.g.
good/fair/poor pavement condition, etc.). The objective is to model the performance of the road network in
response to different budget scenarios or maintenance standards. Programme analysis deals with a selection of
road sections (i.e. a long list of candidate projects). The objective is to prioritise these and select short list of
projects that maximise benefits under specified budget constraints. Project analysis deals with individual road
sections and models mutually exclusive alternatives in order to estimate the economic benefits that would accrue
from each alternative. The three applications can also be used to conduct various research and policy studies, for
example to verify pavement design standards, to model impacts of changing axle load limit regulations, etc.
MATERIALS
1. PowerPoint slides (T3 Role of HDM-4.PPT)
2. Handout notes (T3 Role of HDM-4 Handout.DOC)
REFERENCES
UNIT AIMS
1. To introduce the HDM-4 software components and the installation so that Trainees are able to correctly
install and use HDM-4 for the first time.
UNIT OBJECTIVES
1. Describe the HDM-4 installation procedure
2. Describe the main components of the HDM-4 software (i.e. Road networks, Vehicle fleet, Road work
standards, Projects, Programmes, Strategies, and Configuration).
3. Demonstrate use of the HDM-4 software by running one of the pre-defined case studies.
GUIDELINES
It is important that Trainees are able to properly install and operate the HDM-4 software. They should also
appreciate the licensing conditions under which the HDM-4 software is supplied.
Software installation is achieved through a wizard that guides users through the installation process. There are
several choices that the user must select, for example the installation language, the default HDM-4 software
language, the HDM-4 components to be installed (e.g. Reports, Documentation, etc.). The wizard provides the
opportunity to step back through the process and ensure that the correct choices have been made.
After successful installation, the Trainees should be taken through a guided demonstration of the HDM-4
software. This should aim to illustrate the procedure for creating and running one of the Project Cast Studies
(e.g. Case Study 2a) by following the steps below:
• Review the configuration data (climate zones, speed-flow types, traffic flow patterns, default data)
• Select a Road Network and review the data for one or two selected road sections
• Select a vehicle fleet and review the data for one or two selected vehicle types
• Open the Road Works standards folder and review the maintenance standards used in one of the Project
Case Studies
• Open the Project folder and select one Case Study.
• Review the Case Study data and finally run HDM-4 and review the output Reports.
MATERIALS
1. HDM-4 Case Studies Workspace
2. PowerPoint Slides (Software Installation and Operation.PPT)
REFERENCES
1. HDM-4 Volume 3: Software User Guide
Country Action Plan
For
Implementation of HDM-4 in Highway Planning and Management in
Bangladesh
Objectives:
1. The road management authorities will use HDM-4 as a decision making tool.
2. Prepare optimized maintenance plan within budget constraint
3. Develop maintenance strategies and work standards.
Action Plan:
NB:- Tentative cost of training = 155 persons @ Tk 100,000 per person = Tk. 15,500,000
Objectives:
Group A: The participants will describe how HDM-4 could enhance their decision and
policy-making activities in the road infrastructure and development process.
Group B: The participants will explain and use HDM-4 efficiently in highway planning and
maintenance.
Group C: The participants will describe the role of HDM-4 in determine road user effects.
Course Contents:
See Table B1, B2 and B3.
TABLE C1: BANGLADESH TEAM
Group-A: Policy Makers
Program topics and timing for policy makers
1.1 Basics of Economic Describe the basic Description with 2 hrs. 1 hr.
Analysis and economic criteria and diagram / flowchart
interpretation of interpret HDM outputs support, handout for
HDM output participants, discussion
and demonstration of
HDM run.
1.2 Introduction to Describe the main Demonstration of 0.5 hrs. 0.5 hrs
HDM-4 Work space components in HDM-4 workspace folders and
workspace. describe the
functionality
3.1 Modeling Vehicle Explain and describe Description with 0.5 hrs.
Speeds Including the speed model and diagram / flowchart
Congestion. the effects of support, handout for
congestion. participants
3.2 Impact of Road Explain the effect of Description with 1.0 hrs.
Geometry, Road road geometry on diagram / flowchart
Side Use and traffic. support, handout for
Surface Condition. participants
4.1 Worked Example Explain and interpret Practice session 0.5 hrs.
without Economic project analysis
Analysis. reports.
6.1 Case Study Practice and describe Practice session with 0.5 hr. 1.0 hr.
Examples. the effects of road worked examples
deterioration in HDM
results
1.00 hr.
7. Economic Analysis Describe different Description with
and Road user economic criteria and diagram / flowchart
Costs. parameters of road support, handout for
user costs. participants, case
7.1 Concept of NPV, study, Demonstration,
IRR, B/C ratio. discussion.
7.3 Components of
Road User Costs.
Draft Activity Action Plan
Country Training - Pakistan
Time
No of
S.No. Activity Period Action By * Jul Aug Sep Oct Nov Dec Jan Remarks
Participants
(Days)
1 Submission of HDM T3 Course 15 - CG
2 Review meeting for Country Training Prog. 15 - CG-CC-TRL In consultation with TRL
3 Preparation of Final Country Plan 30 - CG-CC-TRL In consultation with TRL
4 Approval of Final Country Plan 15 - CC-LA
5 Coordination and Preparation of Training Material 45 - CG
6 Dove Tailing country pHs with HMM 4 15 - CG-CC-TRL In consultation with TRL
6.1 Calibration of HMM 4 to local Conditions 15 -
6.2 VOICE and Program Analysis of Network 60 - CG-TRL In consultation with TRL
7 Necessary arrangement of resources 45 - CG-CC-TRL Material and Financial
8 Implementation Consensus & Approval 45 - CG-CC-TRL In consultation with TRL
9 Implementation of Training Pro 15 -
9.1 One Day seminar on HMM 4 at Islamabad (for Policy and Decision Makers) 1 30 CG-CC-TRL In consultation with TRL
9.2 Three days training course for Professionals (Engineers & Economists) at Islamabad 3 12 CG-CC-TRL In consultation with TRL
9.3 Three days training course for Professionals (Engineers & Economists) at 4 Provinces 15 112
9.3 One day training course for Students in NED-Krc, UET-Lhr, UET Taxila, UET Pesh 10 15 CG
9.4 Five Days Training course for Technicians at Islamabad 7 12 CG
9.5 Three Days seminar on Unsealed Roads Management by HDM 4 5 15 TRL-CG In consultation with TRL
10 Subsequent Implementation and Training Prog Onwards - CG-CC-TRL In consultation with TRL
Note *
CG : Country Group ( 1.Waqar, 2.Azeem, 3.Saqlain, 4.Atiq)
CC : Country Champion ( Raja Nowsherwan Sultan)
LA : Local Authorities
Draft Activity Action Plan
for
HDM--4 Training by Trainers
HDM
Country Training - Pakistan
Activity Programme for Pakistan
Time
No of
S.No. Activity Period Action By * Jul Aug Sep Oct Nov Dec Jan Remarks
Participants
(Days)
1 Submission of HDM T3 Course 15 - CG
2 Review meeting for Country Training Prog. 15 - CG-CC-TRL In consultation with TRL
3 Preparation of Final Country Plan 30 - CG-CC-TRL In consultation with TRL
4 Approval of Final Country Plan 15 - CC-LA
5 Coordination and Preparation of Training Material 45 - CG
6 Dove Tailing country pHs with HMM 4 15 - CG-CC-TRL In consultation with TRL
6.1 Calibration of HMM 4 to local Conditions 15 -
6.2 VOICE and Program Analysis of Network 60 - CG-TRL In consultation with TRL
7 Necessary arrangement of resources 45 - CG-CC-TRL Material and Financial
8 Implementation Consensus & Approval 45 - CG-CC-TRL In consultation with TRL
9 Implementation of Training Pro 15 -
9.1 One Day seminar on HMM 4 at Islamabad (for Policy and Decision Makers) 1 30 CG-CC-TRL In consultation with TRL
9.2 Three days training course for Professionals (Engineers & Economists) at Islamabad 3 12 CG-CC-TRL In consultation with TRL
9.3 Three days training course for Professionals (Engineers & Economists) at 4 Provinc 15 112
9.3 One day training course for Students in NED-Krc, UET-Lhr, UET Taxila, UET Pesh 10 15 CG
9.4 Five Days Training course for Technicians at Islamabad 7 12 CG
9.5 Three Days seminar on Unsealed Roads Management by HDM 4 5 15 TRL-CG In consultation with TRL
10 Subsequent Implementation and Training Prog Onwards - CG-CC-TRL In consultation with TRL
Note *
CG : Country Group ( 1.Waqar, 2.Azeem, 3.Saqlain, 4.Atiq)
CC : Country Champion ( Raja Nowsherwan Sultan)
LA : Local Authorities
An Overview of Activity Plan
• Evaluation of
HDM4 Training & Dissemination
1.1 Preliminaries
The Trainers will, within a period of ten days after the arrival back to their
respective places of work, submit to the Country Champion a report
containing the essential details of the knowledge and experience gained
on HDM4 Capabilities through the “HDM4: Train the Trainers Course” held
in Kuala Lumpur from 5 June to 23 June 2000. The Trainers would make a
presentation on the major aspects of the model and their relevance in
regard to the strategic planning, program analysis, project analysis,
execution including monitoring and evaluation.
The report will cover the brief details on the adaptation requirements for
the default Road Deterioration and Works Effect (RDWE) and VOC
relationships. The following aspects would also be brought out in the
report:
The Trainers will have to discuss with the Country Champions in regard to
the "Trainers Training at Malayasia","Review Of The Existing Management
System In India" and complementing HDM4 software with the system.
The other relevant essential aspects concerning the “Country Action Plan”
are briefly described in the paragraphs that follow.
1) Politicians/Policy Makers
2) Ministry of Surface Transport
3) Public Works Departments
4) Academicians and Researchers
5) Consultants
6) BOT Operators
7) Financial Institutions
The HDM4 training will need resources in terms of funding, equipment and
personnel (trainers). The funding for the training is to come from the
Government of India and International Financial Institutions (World Bank,
Asian Development Bank). It is estimated that a total amount of INR 50
Million will be required for the expenses towards the training of about 550
persons in the first two years of the programme. This cost of INR 50 Million
includes institutional charges but not the cost of transportation and
accomodation and TA/DA of trainees from different parts of the country.
HDM4 Training & Dissemination
2000 2001
Month JL AU SE OC NO DE JA FE MA AP MY JU JL AU SE OC NO DE
Preparation
Group-Pol
Group-Eng
Group-Con
Group-Aca
Review
ACTION PLAN BY THE MALAYSIAN GROUP
ACTIVITIES PERSON IN-CHARGE TARGET DATE
2000 2001
A. CREATING AWARENESS JULY AUGUST SEPTEMBER OCTOBER NOVEMBER DECEMBER JANUARY FEBRUARY MARCH APRIL MAY JUN JULY AUGUST SEPTEMBER OCTOBER NOVEMBER DECEMBER
1 Report to Champion
- Training activities in KL
- Proposed Action Plan
3 Study
;-Institutional aspect
- Analyse the position of HDM-4 and existing IRMS
- Training needs analysis
- Calibration level- 1
- Calibration level- 2 and level 3
4 Training
5 Implementation HDM-4
1 Report to Champions
- Training activities in KL
- Proposed Follow-up Actions
2 Introductory Seminar
- awareness
- commitment
3 Study
- Institutional aspect
- Position of HDM-4 in the existing Systems
- Training needs analysis
- Calibration level- 1
- Calibration level- 2 and level 3
4 Training
5 Implementation HDM-4
Item No. DESCRIPTION OF ACTIVITY PRESENTED GROUP TYPE TARGET GROUP PERIOD
BY
1 Introduction to Senior Management National Government User - Dept. of Works
Senior Executive Management
Prepare and submitt report ba/pn Secretary and Deputy Secretaries Departmental Head & Champions 1 Week
Give an Overview on HDM-4 ba/pn Secretary and Deputy Secretaries 5 persons 1 hour
Prepare a session plan & other relevant authorities
Train the Trainers ba/pn A: One senior engineer from each Region Engineers 2 weeks
Introduction & Training B: Planning & Maintenance Branch Engineers & Planners 3 days
10 persons
Conduct Training (Highlands) Trainer - Highlands C: Provincial Works Managers, Provincial Engineers 2 weeks
Civil Engineers - Highlands Region 15 persons
Conduct Training (Southern) Trainer - Southern D: Provincial Works Managers, Provincial Engineers
Civil Engineers - Southern Region 15 persons 2 weeks
Conduct Training (Northern) Trainer - Northern E: Provincial Works Managers, Provincial Engineers
Civil Engineers - Northern Region 15 persons 2 weeks
Conduct Training (Islands) Trainer - Islands F: Provincial Works Managers, Provincial Engineers
Civil Engineers - Islands Region 15 persons 2 weeks
Item No. DESCRIPTION OF ACTIVITY PRESENTED GROUP TYPE TARGET GROUP PERIOD
BY
3rd. July
12th. July
28th. July
Completed By
PROPOSED ACTION PLAN FOR PAPUA NEW GUINEA
Item
No. DESCRIPTION OF ACTIVITIES MONTHS MONTHS
Months June July August Septembe October NovemberDecember January February March April May June
1 Introduction to Senior Management
Prepare and submitt report
Give an Overview on HDM-4
Finalise action plan and obtain approval
3 Implementation of Training
4 Government Support
Overview presentation of HDM - 4
5 Promotion
Overview presentation of HDM - 4
**Arbitrary
Action Plans
Will Commence on*:
• Orientation of key officials regarding HDM-4 framework and relevance of the July 2000
training programs
July 2000
• Establishment of Rapport with Government Agencies related to Road/Transport
Infrastructure Planning and Development and Private and Professional
Organizations (e.g., PICE, REAP, TSSP, Consulting Firms, etc.
• Development of the Training Courses mentioned above August 2000
• Identify Training Requirements (funding, etc.) August 2000
• Commencement of the Conduct of the Training Courses February 2001
• Formulation of Technical and Institutional Measures for the Sustainability of the February 2001
Training Courses
• Institutionalizing HDM-4 Training for Human Resource Development and Capability March 2001
Building
*Suggested
2000 2001
Activities
July August Sept. Oct. Nov. Dec. Jan Feb. March
1. Orientation of key officials regarding HDM-4
framework and relevance of the training
program
2. Establishment of Rapport with Government
Agencies related to Road/Transport Infra-
structure Planning and Development and
Private and Professional Organization
(e.g., PICE, REAP, TSSP, Consulting Firms)
3. Creation and Calibration of Road/Transport
Databases
4. Conduct Pilot Study on Ad Hoc Road
Projects for Country Case Studies to be
used in the Conduct of Training Programs
Courses
5. Development of the above mentioned
Training Courses
6. Identify Training Sources (funding, etc.)
- to be decided
7. Commencement of the Conduct of Training
Courses
8. Formulation of Technical and Institutional
Measures for the Sustainability of the
Training Courses
9. Institutionalizing HDM-4 Training for Human
Resource Development and Capability
Building
Tentative Action Plan
Proposed by Siriphan Jitprasithsiri, Thailand
For The HDM-4: Train the Trainer Course at IKRAM, Malaysia
23 June 2000
1. Identify their own learning styles and describe how an understanding of these styles and
the Learning Cycle can help them to prepare and deliver better training.
2. Define the main target groups who will need HDM-4 training in their country and carry
out an initial target group profiling exercise for each of these groups.
3. Carry out a training needs analysis of the groups and develop a draft outline training
programme for each of the main groups.
5. Prepare and successfully deliver to a peer group, an HDM4 training session selected from
the detailed training programme in 4 above.
8. Identify learning difficulties and design and use learning aids to enhance learning in these
areas
10. Develop an action plan for implementing HDM4 training in their individual country/areas
The purpose of the training programme is to introduce participants to the key factors and
techniques in training. Through this knowledge and techniques the key trainers will be better
prepared to pass on their technical and training skills and knowledge more effectively to
others when developing the HDM-4 training and dissemination programme in their respective
countries. This training component of the train the trainers programme lays emphasis on the
importance of preparing and structuring training session content to meet the pre-determined
needs of the trainees and the use of participatory methods suitable for adult learners.
At this initial stage, the project emphasises “face to face” trainer lead training programmes.
However, as the project progresses, a growing range of print and computer-based materials
will be developed which will allow more flexible learning approaches to be supported. These
will be structured to meet the identified information and training needs of those involved with
highway policy, planning and management and may include individual study, self supporting
groups, support networks and web bases information and reference sites
This notebook provides trainers with information and examples from which they can draw
upon in their future work. It is emphasised that the content is not a comprehensive text
covering every aspect of training but simply those aspects that have been identified as
relevant to training activities in this case..
The notes are divided into Units that have common themes. Each Unit is preceded by a brief
introductory text giving the aims and objectives of the UNIT and guidance on how the text
might be used.
The material and approaches proposed are best used within a participatory workshop format.
This is so that principles can be illustrated and applied during individual exercises and
assignments. These exercises and assignments should be based upon realistic working
conditions and technical issues likely to be experience by the trainers being trained.
The text is focused on the practical aspects of the learning process and it should be read in
conjunction with other available texts.
CONTENTS
Page No
UNIT 1
LEARNING STYLES
The Learning Cycle, assessment of learning styles and the implications of
individual styles for training strategies, content and methods.
UNIT 2
TARGET GROUP PROFILES
Gathering and using social, psychological, cultural, economic and technical
information in training planning.
UNIT 3
ADULT LEARNERS and PARTICIPATION
Adult learning and participatory approaches with adult learners.
UNIT 4
TRAINING NEEDS ANALYSIS
Identifying the training needs of individuals and groups.
UNIT 5
AIMS AND OBJECTIVES IN TRAINING
Using aims and objectives to enhance the construction and evaluation of
training programmes and sessions.
UNIT 6
TRAINING SESSION PLANNING
The planning of training sessions and selecting training approaches
UNIT 7
LEARNING AIDS
Designing and using simple learning aids
UNIT 8
TRAINING PROGRAMME PLANNING
Designing and planning in-country training programmes for different target
groups.
UNIT 9
COURSE EVALUATION
Improving feedback and formative evaluation methods in the project.
UNIT TWO
TARGET GROUPS
UNIT AIM
1. To introduce the concept of Target Group profiles so that trainers can prepare
training sessions that more accurately match the trainees needs and are conducted
within a supportive learning environment.
UNIT OBJECTIVES
1. State the main headings which can be used to construct a target group profile.
2. Complete a Profile for a specified target group.
3. Use the profile to explain how the technical content, training methods and learning
environment of a specified session would be influenced by the information
available.
Trainers Notes
This session takes the people focus concept of training developed in learning styles
into a more complex set of characteristics which can also have a major impact on the
training and learning environment. Often the Target Group Profiling and the Training
Needs Analysis which follows in UNIT 3 can be carried out together as they both,
under ideal conditions, involve face to face discussion and information collection. It is
strongly emphasised that both these processes are carried out in the field where the
intended trainees live and work. Desk based analysis is always a very poor second to
field work and desk analysis done from memory and second hand information
invariably results in second rate training.
1
TARGET GROUPS
By now it should be clear that in training, the technical content of a training
programme is only considered after we have clearly defined who the Target Group(s)
are.
Exercise
Consider this situation:
You have been asked to provide a short introductory presentation on
“The advantages of using HDM4 in the design and maintenance of the national road
network”
Your immediate and understandable response might be to start looking out data on the
economic considerations or various permutations on construction options, the scope
of the HDM 4 analytical capacity and the various calibration values needed for
different environments. However before you do all this it might be interesting for you
to stop and ask the question – Who is it that I will be making this presentation to ?
Consider the implications of he two groups below.
1 A group of 15 year old secondary school student doing a project on the national
road network
2 A group of senior government officials with responsibility for the development of
national road policy.
Clearly these two groups will have very different expectations from a session with the
same title
How would you structure presentations for these two groups and what diffence
in content would you make ?.
From this example it will be obvious that the planning of training must begin with the
focus being placed on people and only when they have been considered in detail can
we move to considering the content and the training methods we can adopt.
This Unit on Target Groups provides some suggestions for the main headings you can
use to structure a Target Group Profile. These might be:
2
Age range,
Issues
1 A wide range in ages in the group could suggest a wide range of experience and
different expectations in different age bands
2 More mature people may want more time for discussion and practice in new
methods and concepts They may also appear to be more resistant to change
3 Expectations may vary considerably across age ranges
4 .
Gender
Issues
1 Male or female or a mix in group?
2 Do men and women in the work place have equal job responsibilities?
3 Do men and women have equal access to work place resources?
4 Is training programme scheduled to accommodate domestic commitments?
5 Accommodation and facilities if course is residential
3
2 Do not assume that a practically experienced trainee also has a full understanding
of the theoretical concepts involved or vice versa ( see learning styles?)
To collect really accurate information the trainer should interview a cross section of
the target group and also those who have regular contact with them. e.g supervisors,
clients, subordinates etc.
Numbers to train.
For practical courses it is best to keep groups small - usually practical groups of more
than 10/12 are too big to be able to provide sufficient practice and equipment. Also
on practical training it is not easy to properly supervise larger groups and especially
where safety is an issue. Remember the only way to learn a practical skill is to be
able to have extensive hands on practice. Watching others is not sufficient.
Age of trainees
The age of trainees can influence the training approaches needed in a number of ways.
In general younger people are able to grasp new ideas and concepts more readily than
their older counterparts. However this should not be seen as implying that older
people are slow learners. It simply means that older people have more experience and
often have to compare these new ideas and knowledge against their existing
understandings and practices before considering the acceptance of new ones. This is
why trainers must allow plenty of time for explanation and discussion with older
groups. Remember also that older people do not like to “lose face” so when asking
questions or getting people to do practical tasks try to ensure that you present
questions that they can answer and the tasks in simple steps which they can grasp and
master with the minimum of mistakes. Adult learners are also much more concerned
with the immediate application and utility of the training content being presented. So
keep it relevant and applicable.
4
While, in general, there is no difference between teaching men and women, it is
important to remember that their background experience may well be different. A
good trainer (man or woman) will ensure that an accurate target group profile is
drawn up and that the training programme meets the needs of all participants.
Where there are considerable differences in experience and previous training between
men and women it may be appropriate to run separate courses. The primary guiding
rule is that the trainer must remain sensitive to the needs of all participants, not just
one particular sub-group.
A major difficulty experienced by many trainers is how to deal with a group which
contains a number of people with wide or specific expertise in the topic. A good
trainer will make use of such people during the course, calling upon them to
contribute to the training sessions and perhaps using them as group leaders in some of
the practical exercises.
It is important for the trainer to find out the qualifications and experience of the group
as early as possible. Some may have had very useful “hands-on” field experience, or
be especially knowledgeable on a particular sector of road construction or design.
Many younger and inexperienced trainers do not like to ask the assistance of someone
in the group as it might reflect badly on their position as trainer. Trainees, however,
rarely think like this. Trainers should take full advantage of any special experience
and knowledge available to them in the group. This can also considerably improve
the involvement and motivation of people who may otherwise feel the course is at too
low a level or not relevant to them.
Motivation
Interest and enthusiasm is infectious and they must start with the trainer. Good
preparation, a sound knowledge of the subject and a good understanding of the target
group, their experience and their problems in the technical area of the training will be
of great value.
The trainer’s own attitude will greatly influence the group. It should be remembered
that it takes a little time for trainer and trainees to get to know each other at the
beginning of a course. This is a time when a trainer can be anxious and nervous.
However, good preparation should make that initial period easier.
Most trainees will be keen to learn, providing the subject matter is relevant to their
situation and needs. There will always be a few who will need encouragement; this
can usually be achieved by actively drawing them into group activities or discussion.
5
processing with HDM4) can also be a hidden way of motivating the less enthusiastic
trainees.
In any group, there may be great variation and different characteristics and it is
important for the trainer to find out as much as possible about the group to be trained.
In this way, the training can be designed and delivered in a manner suitable for that
particular target group. A great deal of training is less effective than it might be
because those designing and delivering it do not take the trouble to identify their
target groups clearly.
Finally, trainers should remember that even the best trainees can be demotivated if the
domestic details of the course (transport, accommodation, meals, payment of daily
allowances, etc.) are badly organised.
6
Exercise Construct a profile for two different Target groups you will train in the use
of HDM4. Use the format given below
Age range,
Motivational issues
Notes
7
GROUP EXERCISE
Task
Using the Target Group profile form provided and your experience draw up a profile
of the group you have chosen
8
UNIT THREE
UNIT AIMS
1. To highlight some of the main issues in adult learning so that trainers can plan more
effective training sessions.
UNIT OBJECTIVES
1. Describe four ways in which Adult learners differ from younger learners.
2. Give three examples of how the trainer can utilise these differences positively in adult
learners
3. Give two examples of participatory teaching methods.
Trainers notes
Your work is concerned mostly with adults and they may have many different backgrounds
and experiences. Most adults already have experience and information related to the subject
being taught. In some cases the training being given may challenge or contradicting what the
trainee already knows and does. To make changes adults often have to UNLEARN what they
already have learned and experienced before they can adopt new methods and ideas. The
notes and tips in the text give direction to the trainer on how more effective participatory
training can be planned
1
ADULT LEARNING
Early learning theory was based on knowledge gained first from studying animals then
children. The term Pedagogy was used to mean the art and science of teaching children. In
early adult education adults were taught as children and because of this much adult education
was unsuccessful. This was often the case with adult literacy programmes.
We have since come to realise that adults learn in a very different way. In the 1960s and
1970s a number of American and European educators were conducting research into the way
adults learn. In 1970 Malcolm Knowles wrote a book entitled “The Modern Practice of Adult
Education: From Pedagogy to Andragogy”. He used the term Andragogy to mean the art
and science of helping adults to learn.
There are many ways in which adult learning differs from child learning. Knowles identified
four major differences.
1 Experience
Adults have a vast reservoir of experience that is a valuable resource for learning. If
these experiences are devalued or ignored then barriers to learning are often created.
2 Self Concept
Adults are independent, self-directing beings who like to exercise control over their
own learning.
3 Immediacy of Application
Adults are generally more motivated if their learning can be applied immediately,
rather than at some time in the distant future.
4 Social Roles
Adults are motivated if their learning is of benefit to them in one of their current roles
in life.
3 Adults need to be able to diagnose their own needs and to plan their own learning.
They may, however, need help with this.
4 Adults learners need to be given as much information about the programme, methods
and purpose of all aspects of a course in which they are participating. This way they
can take responsibility for their own learning.
5 Negotiation between the trainer and the learner should be an important feature of any
adult training programme.
2
6 Adults need to be able to explore new material and concepts in the light of their own
experiences.
8 Learning may not fall into neat artificially constructed categories: - a more holistic
approach may be needed.
3
PARTICIPATION
What is meant by the participatory approach to training?
Participatory training involves using training methods in which the trainer encourages
trainees to be active in the learning process and give feedback. The trainees share ideas,
experiences and opinions. Participatory training relies on people communicating with each
other.
The big advantage in training adults is that they already have extensive knowledge and
experience. However, because they are so busy they will only go to training if they think it is
relevant. A good way of showing adults that information is useful to them is to relate it to
their own situations and experiences. The best way to do this is to encourage them to
discuss their ideas, to identify their problems, to present their opinions and to share their
experience, i.e. PARTICIPATE! This also requires trainers to have a realistic and preferable
first hand experience of the practical conditions in which the training is to be applied.
Adults are more difficult to convince with just words. They will only adopt technology if
they see it working for themselves. The trainer must spend time showing adults how to do
something and then letting them practice for themselves. This is because they have many
past experiences that they measure new ideas against.
Adults are more sceptical than children, they need more convincing. They can be
conservative and resistant to change. This is particularly true of people in a vulnerable
position where they think new technology or methods may threaten their current position.
They may also have a fear that they may find new methods too difficult to learn. Therefore, it
is particularly important to encourage adults to actively participate during a training session
so that they can put the information into their own words and relate it to the own experiences.
It is not only in the area of knowledge that will need careful handling. Practical processes and
procedures may also be problematic as they are often related to changing ways in which
existing processes are done. Make sure that plenty of time is allowed for guided practice
within the training session. Early failures in practical sessions can demotivate and leave
trainees rejecting a new technique.
Traditionally, adults with little formal education learn by watching others (sitting by Nellie)
who are more experienced, by sharing information and skills, and by Learning by Doing.
They PARTICIPATE.
b) The trainer can draw on the experiences and abilities of the participants and relate the
subject to the real conditions of the trainees.
e) Training becomes more directly relevant to the trainees needs and experiences.
4
d) Problem solving is more effective using participatory techniques. Many heads are
better than one and the combined experience of the group becomes available to all of
them
e) Participation helps people to remember what they learn in training. For example, in
your own experience do you find it easier to recall ideas you have learnt in a
participatory session such as problem solving in group work and hands on practical
session or ideas presented to you in a lecture?
f) The trainer can get feed back from participating trainees which signals if they are
having difficulties in understanding during the session and so the programme can be
modified to meet the needs of trainees.
h) Contributions from trainees enrich the training session and can give trainees greater
self confidence The participants feel valued and on an equal level with the trainer.
This is very important in training adults.
i) Other trainees as well as the trainer can help trainees with difficulties. Trainees
providing help to other trainees also deepen their own learning by having to explain
concepts and procedures.
k) Participants and trainers begin to value each other’s opinions and experiences more
and both are learning.
1) Through participation, people learn to accept ambiguity and realise that there may be
many sides to an argument with not just one right answer.
Disadvantages
Of course every method can have disadvantages. Here are some of the problems that can
arise in using participatory techniques?
a) In a participatory approach the trainees take more control of what happens in training.
The trainer must be able to respond to the way in which trainees react in training and
be prepared to adapt. Some trainers find this frightening
b) Adults are status conscious and shy of making fools of themselves; they may be
hesitant to participate at first.
e) Adults may be out of practice at learning and so the trainer will have to carefully plan
how to involve them using participatory methods. This requires considerable skill and
an ability to encourage and motivate people. .
5
d) The trainer must plan ahead and be fully prepared in both the technical and methods
of the session. This often means more extensive planning than that needed for less
participatory teaching styles.
6
BARRIERS TO GOOD LISTENING
In participation and communications it is just as important to be able to listen effectively so
that as well as giving information you also receive it. You should try to listen with the same
attention and open mind that you hope people will listen to you with. In training it is
essential to develop good listening skills if you are to be a good communicator. Listed below
are some of the things that can happen to stop you hearing the messages.
`On-off' listening
This unfortunate habit in listening arises because most of us think about four times as fast as
the average person speaks. Sometimes we use this extra time to think of our personal
concerns, or interests and troubles, instead of listening and so often miss the point of the other
person’s comments.
'Glassy-eyed' listening
Sometimes we look at a person intently and seem to be listening although our minds are on
other things. We get glassy-eyed and a dreamy expression often appears on our faces. We
can tell when people look at us in this way when we are speaking. Similarly, others can see
the same in us.
'Fact' listening
Often as we listen to people, we try to remember the facts and repeat them over and over
again to drive them home. Frequently, as we do this, the speaker has gone on to new facts
and we become lost in the process.
'Pencil' listening
Trying to put down on paper everything the speaker says. Eye contact becomes more
difficult and the pencil and paper also act as barriers.
7
“HEARING IS NOT ALWAYS LISTENING”
IMPROVING YOUR LISTENING SKILLS
Be patient – the time you are prepared to give to listening may be the most valued part of
your contribution.
Leave your feelings behind - and don’t use “I know how you feel” because in most cases
you won’t.
Look at the speaker – What are their body actions telling you (non-verbal
communications). What is your body saying!?.
Don’t argue – either mentally or directly. Wait until the full story is heard and then you can
ask questions and make comments. Try to get a feeling about their personality, their likes and
dislikes, what motivates them, what values they hold.
Question your own assumptions before you respond. Avoid making instant judgements.
Most situations are not as straight forward as they first appear.
So listen accurately hear what is said, not what you think is being said.
8
EFFECTIVE USE OF QUESTIONS
This is one of the most widely used and most important participatory training methods. It
requires direct communication between the trainer and the trainees. It takes thought, practise
and planning to use questions as an effective teaching technique. In this technique the trainer
uses questions to guide the students thinking and processing of information.
An essential rule in using Questioning in this way is illustrated in a simple Poem by Kipling
The message is – donj’t use questions which only require a “Yes or No” answer. For
example,: Do you understand? – Is that right? As you begin to use questioning techniques
this advice will become clearer.
Types of Questions
Open - Several possible answers
Closed - One answer (right or wrong)
Examples
Q1 What is the capital city of France?
Closed - One answer (right or wrong)
Objective - Requires a factual answer.
9
Q2 What is the difference between a diesel engine and a petrol engine?
Open - Several possible answers
Objective - Requires a factual answer.
In a participatory session the Directed question is most effective as it allows the trainers to
spread the questions out and also grade the questions so that trainees who are not sure can be
encouraged by giving them easier questions to answer. This method allows the trainer to
nominate a particular person to answer. Remember that trainees will need to understand the
process if they are not to call out during named questions.
The use of questions is usually traditionally thought of as a means of testing or finding out
what the trainee knows. While this is one use’ the real skill of using questions is to lead
trainees to discover and work out how particular concepts are applied or why a particular
process is the best one to follow. Teaching through participatory question and answer
sessions can be very effective but it has to be carefully planned and needs a great deal of
practice to be used well
- answer it -
- put the question back to the trainee Reverse
- put the question back to another trainee or the whole group Relay
Try to avoid repeating the question as it encourages lazy listening in the group and poor
presentation skills by the trainee asking the question.
These are non-specific questions and are not very useful or effective questions because they
do not give the trainer much information. (feed back) and are usually answered with “yes”
For example, when you ask the question "Do you understand?” how do you interpret the
meaning of the answer "Yes"?
10
Trainees may answer "Yes" because:
(a) They do understand
(b) They do not want to appear stupid by saying no
(c) They do not really know what the trainer means by the question
(d) They are bored and want the trainer to move on to the next piece of information
Now, imagine that you had just explained the main ways to reduce high rates of deterioration
on a particular road type and you wanted to check whether the trainees had grasped the main
points.
Which one of these questions do you think is most useful for the trainer?
What will the trainer find out from the answers?
1. "Do you understand?" (The trainer will not be sure what the trainees have understood, if
anything).
2. "Can someone tell me one important factor leading to reduced rates of deterioration in
this case.?" (This answer will tell the trainer if the trainees can recall what has been said
and will show which points they think are most important).
3. "In your own experience, what would be the best solution to reducing pavement
deterioration rates in a situation where only hand labour is available.?"(The answer to this
question will show whether the trainees can apply the training matter to their own
situation).
4. "Can you explain to me in your own words how you would explain to a road maintenance
gang how to carry out this maintenance process. (The answer to this question will indicate
to the trainer whether the trainee can correctly recall the subject matter. It will show
whether the most important points have got across. It will also reinforce and clarify the
information for the rest of the group. (“The best way to learn a subject is to have teach
it”)
(b)
Right answer but not the answer that you wanted
Action
• Be careful to distinguish between a wrong answer and an answer that is right but not the
one you were expecting.
• Be careful to praise the participants and encourage them for giving a correct answer.
11
• If the answer is not the one you expect, you could say: "Yes, that is correct but can
anyone else think of any other answers?" or, "I hadn't thought of that, can you think of
anything else?"
(c)
Wrong answer
Action
• Why may this happen?
• Has the question been misunderstood?
• Has the not understood the information given in the session?
• Rephrase the question For example, at the end of a training session about estimating
pavement condition. The trainer might ask a trainee:
The answer you want is, "when the surface falls below X Y standards." However, the trainee
answers, "when you can see pot holes forming." This answer is wrong and also shows that
the trainee has not understood the purpose of preventative maintenance.
These activities are designed to determine what misunderstanding has occurred to cause the
wrong answer to be given and to determine what subsequent action will then need to be taken
to put things right by:
Supplying more information;
Clarifying the information
Having more discussion
Rephrasing the question;
When using the directed question, do not pause for too long between asking the question and
nominating the trainee. Trainees may need training in this technique - that is you may have
to explain that when you have asked the question, you do not want trainees to call out the
answer, as you will ask one of them by name.
12
Finally remember that asking questions involves risks so be prepared for surprises.
If you are asked a question which you do not know the answer to, it is best to say ”sorry I
don’t know” BUT then find out as soon as possible and give them the answer.
Some questions can also have different answers for different people
13
GROUPS
Groups are a normal and important element of most cultures and they can be formed for
many reasons and group work is an important area of activity in participatory learning.
In training it can be helpful to use groups to aid learning by allowing individuals to share
their experiences with others and to learn from each other. Groups allows more complex and
longer term exercises and assignments to be set.
Sometime groups do not work well and the following list may give you some clues why a
particular group does not work.
Unclear objectives
Unbalanced composition
Inappropriate leadership
Overbearing leadership
Low creativity
14
Summary of issues for working groups
Working Groups?
Purpose Membership and structure Co-operation & Development
Is the purpose clear? Voluntary participation The need to belong (Inclusion) Forming
Whose purpose is it? Group size exert influence (Control)
Group heterogeneity to give and receive approval and support
The need for clear owned goals Common values and principles (Affection)
The need to measure and monitor progress Members individual objectives - Overt and
The need to avoid aimlessness, drifting and covert Pecking order Storming
disintegration Performed leadership or decided by group Muscle flexing
Group processes and procedures clearly Testing of others
defined Shifting loyalties
Mutual accountability
Task and maintenance functions and roles Emergence of Norming
defined Cohesion
Norms
-meetings
-decision making
Leadership Performing
control and limitation mechanisms
Adjourning??)
15
OHP
Stages of group Development
Forming Collection of individuals
Each with own agenda
Little or no shared experience
!
Storming Personal values and principles challenged
Roles and responsibilities accepted or rejected
Conflict as personal agendas are followed
If successfully handled, group objectives and processes
are defined.
!
Norming Rules accepted
Shared goals established
Group and sub-grup settle down and become cohesive
!
Performing Group maturity and energy
Maximum productivity
New insights and solutions appear
!
Adjourning Loss (achievement?) of common goals
Sub groups form with separate agenda
Loss of interest and purpose
Members leave
Group ceases to function
"
16
OHP
Stages of group Development
Forming Lennon and McCartney didn’t know each other
and uncertain of each other’s skills
Members unable to commit to long term
! relationship
The Quarry Men, Johnny and the Moondogs,
The Silver Beatles
Storming Lennon and McCartney recognise each other
talent and begin to practice together.
The group struggle through long periods of
internal strife, drummers, managers and
guitarists quit.
!
Norming Something happens in Hamburg. New
techniques and songs are tried. Tight bonding
and cohesion; camaraderie “were just four parts
! of one” McCartney
Performing The group entered a period of unparalleled
productivity. Concert tours. television and films.
Millions of records sold. They stopped touring
! in 1966 but continued to sell albums until 1970
Adjourning Growing distance amongst members. Lennon
married Ono. McCartney turns more to his wife
Linda. Epstine dies of overdose McCartney
announces he is leaving “All things must pass”
"
17
Participation equals
" Discussion
" Questions - answering and
asking
" Looking Observing
" Explaining Describing
" Problem solving
" Carrying out practical processes
" Sharing ideas, opinions and
experience
" LISTENING
18
UNIT SIX
UNIT AIM
1 To introduce a structured approach to session planning so that trainers can prepare
sessions within a common framework/structure and can share session plans with
others.
UNIT OBJECTIVES
1 State the main parts of the session
2 Explain the purpose of major UNIT and their role in the learning process
3 Produce a session plan to the required format
Trainers Notes
The following text is intended as a framework for the major considerations when
planning a training session It draws upon and itegrates the materials covered in the
preceding Units. While the basic session framework is where the structure and
purpose of the session is developed. The trainer will in addition also need to prepare a
set of personal training notes which set out in detail the content of the session and
the methods and activities planned to achieve participation, group work, use of
learning aids and timing etc. The key to this process is that the session content and
methods are guided by the session aim and objectives. If the objectives have been
careful developed in the training plan then the individual trainer can plan a session to
meet these objectives. The point being that while individual trainers may take
different approaches, the outcomes should be the same. i.e. the objectives are met.
GUIDELINES FOR PREPARING TRAINING SESSIONS
This is a checklist for you to use in the planning process. However keep in mind that
each session while being planned individually should also recognise the training
previously covered and any training sessions following.
The “Proforma Planning Sheet” given below is a suggested layout but individual
trainers may decide to modify it in the light of their own experience and needs.
However the same general principles and headings should still be relevant. A good
guide to use is that if you gave your session plan to another trainer, He/she should be
be able to use it to deliver the same session.
The Proforma Planning sheet is only a summery sheet and must be supplemented with
a set of “Training notes” which set out the content notes needed and show how the
detailed training process is programmed
i.e. It covers detailed timing, when visual aids are to used, what leading question
might be used, when group work might be introduced and guide notes for group task
etc. For a new trainer these notes may be quite extensive and detailed so that they
don’t forget what is being taught and how they intend to implement the training
methods selected. As experience grows these will be reduced to headings and simple
short notes.
CHECKLIST
A Session Title Is it motivating and informative
2 have you written your Objectives with Action Verbs which clearly state what
the trainee will do?
3 consider what type of activity your objectives are indicating your trainees will
practice during the session:
a) KNOWLEDGE
b) APPLICATION
c) SYNTHESIS
d) EVALUATION
e) ATTITUDES
f) SKILLS
4 Are your aims and objectives closely matched to the training needs of your
TARGET GROUP?
Aim
To teach the contribution of unsealed roads to the national network and their
management so that participants can implement improved management programmes
and methods.
Objectives
The trainees will
Explain the contribution of these roads in the national network
Describe the main classifications and characteristics of each
Describe the main factors to be considered in developing an effective management
strategy for these roads.
Develop and defend a suitable management programme from a given HDM4 case
study.
B Introduction to Session
1 Can you use visual aids, anecdotes, real material to motivate your trainees?
3 Have you planned how you will motivate the trainees by INVOLVING them
early in the session?
4 Have you written down the QUESTIONS you will use to involve trainees in
the session?
5 Will you tell trainees the TITLE of the session and give an OVERVIEW of it?
6 Will you tell your trainees how this session fits with other training sessions?
1 Does your session develop from SIMPLE ideas to more COMPLEX ones?
7 Have you planned how you will use the white board. overhead projector and
power-point presentations
9 Have you estimated how long each part of the session will take?
D Conclusion of Session
1 have you planned how you will summarise the main points covered in the
session and test your objectives?
2 Have you written down the questions you will ask the trainees to test whether
the trainees have learned what you intended?
3 Will you tell trainees how the session will be followed up in future practical or
theory sessions?
E The Room
1 Will you CHECK that all the aids you will use are available and working?
2 Will you check that all your aids can be clearly seen from all parts of the room
If the answer to most of these questions is YES then you are more likely to motivate
your trainees, keep them interested and achieve your objectives, helping trainees to
learn and prepare for their future jobs.
TRAINING SESSION
SELF-EVALUATION CHECKLIST
5 4 3 2 1 0
HOW WELL DID I......? Very good satis- not very poorly not
well factory well applicable
1 link this session to other
sessions
10 handle problems of
inattention
11 ask questions
21 check learning
23 convey enthusiasm
Aim
To teach
Why
Objective(s)
The trainee will:
Development
Training points
Summary
Conclusion
The table below indicates with the word code where the question may be relevant.
This not a precise tool but it does make the trainer consider the content of the session
carefully and the answers will begin to reveal what needs to be included and the
possible structure of the session
Aim
To teach the factors influencing budget allocations for road maintenance and
improvement in startegic analysis
So that trainees can identify the key factorsand criteria involved and use them in
HDM-4 strategic planning applications.
Objective
The trainees will
describe the key elements and criteria to be used.in HDM.4
programme and produce an HDM4 report when supplied with a given set of
conditions and support information.
demonstrate their ability to explain and apply the results available in the report.
Example2
Title
Unsealed Roads
Improved title
The importance, types and management of unsealed roads in national networks
Aim
To teach the contribution of unsealed roads to the national network and their
management so that participants can implement improved management programmes
and methods.
Objectives
The trainees will
Explain the contribution of these roads in the national network
Describe the main classifications and characteristics of each
Describe the main factors to be considered in developing an effective management
strategy for these roads.
Develop and defend a suitable management programme from a given HDM4 case
study.
UNIT SEVEN
UNIT AIMS
To make you aware of the potential uses and types of simple learning aids available to
the trainer so that you can make better use of them and enhance the trainees learning
experience.
UNIT OBJECTIVES
Define three ways in which learning aids can improve trainee learning.
Design and use a learning aid in a practice training session to get over a particular
learning point
Define the learning difficulty and how the aid overcomes this.
Trainers notes
There is a wide range of Learning Aids available to the trainer and often the most
effective ones are designed and made by the trainer from locally available resources.
The primary aspect of learning aids is that they must be selected and used to meet a
specific purpose in the training session. The trainer needs to be clear on why they are
being used.
Too often trainers will show a long series of slides/power-point or a video which may
take up to 30 minutes to present. Unless the intention is to present an overview then it
is far better to select only those slides or video clips that are directly related to the
Session topic and Objectives. The most likely learning aids that will be used in the
computer based HDM-4 programme are text in the form of notes, summary lists,
graphs, formulae etc and pictures and diagrams frequently presented through power
point presentations, overhead transparencies or directly on the computer screen.
However this does not mean that that white boards and flip charts, etc are not of any
use here. In fact, the white board, and flipcharts can be extremely useful and flexible
when working with groups where feedback can be collected and presented in this
way. These are also more flexible when working in a participatory way when visual
presentations can be built up and modified in response to the responses and questions
of trainees.
Learning aids are not only a great help to the learner, they are also a useful way to
structure your training plan and session. They provide a useful focus when trying to
build new ideas and concepts with trainees. Careful use of questions centred on the
aid allows you to systematically integrate the trainee’s experience with the new
concepts and knowledge you are introducing. Finally, remember that we learn a great
deal more through our eyes than we do through our ears. Vision is a greatly under-
utilised channel in the learning process
PRODUCTION AND USE OF VISUAL AIDS
Introduction
It is worth remembering that just as people can misunderstand what is said they can
also misinterpret what they see. This means that you must choose and design visual
aids with considerable care
The three examples given at the end of this Unit illustrate how although we assume
that our eyes never deceive us that is not always the case
Example one shows how something that can be drawn cannot always exist in the real
world.
Example two illustrates how we learn to interpret ways in which we can use various
visual “language” (dotted lines in the box in this case) to make something clearer.
Example three shows how we may interpret the drawing (joining up the lines) when
the detail is incomplete. If you see the drawing upside down first you may not realise
it is a chair.
These examples illustrate how careful we should be in developing and using visual
aids. Never assume that everyone sees things in the same way as you do. So when
using visual aids always ask question about them “Can you describe the shape/ colour
of the road surface?” “What is the problem you can see at the edge of the pavement
area in the picture?” “Can you explain what this graph shows?” If you get an answer
you didn’t expect it could indicate that the trainee is not “seeing” the visual aid in the
way you had intended.
This Unit gives broad guidance on the use of some forms of visual aids. However,
ultimately their effective use depends on the skill of the trainer in diagnosing the areas
of learning difficulty and choosing an aid appropriate for use in that situation. The
illustrations, pictures and visual materials available in the Train the Trainer CD will
assist trainers in developing their training sessions, but they should also try to ensure
that they develop a wider range that are specifically relevant to their own local
conditions and culture.
Handouts
Handouts provide an effective way of providing trainees with information that can be
used for reference in the future. They can also save a great deal of time by cutting
down the amount of time trainees spend making notes. However, the trainer must
design handouts that are not simply a copy from a page in a book. A handout should
present the main points and important information of the session in a logical and
simple way that helps the trainee to recognise the structure of the training session
which will help them to remember and use the information in the future.
Handouts should be used during a training session, not simply given out at the end.
This can be achieved by asking questions on the content or organising small
discussion groups to work on a particular topic in the handout. Spaces can also be left
for key sentences or words to be filled in. This makes important information more
easily remembered.
Whiteboards
Boards are frequently used by trainers, but unfortunately, they can be used badly and
become a personal scribbling-pad for the trainer.
If you plan to use the board as your main presentation tool then try to ensure that the
main headings and key points are built up as the training session progresses. This
helps to emphasise the main elements of the session and provides the trainee with a
good structure to follow.
Boards come in many shapes and sizes, from very small portable ones that can be
used in practical field sessions to larger class boards. It is a good idea to split the
larger boards into sections and keeping part free for the sketches, diagrams,
calculations etc which may be needed during a training session.
Use colour to give emphasis and interest to notes and diagrams, but keep diagrams
and sketches very simple and use prepared Acetates and charts for more complicated
topics unless you are good at sketching. Use block capitals for main headings, lower
case underlined for sub-headings and, provided it is easily read, your normal
handwriting for text. Size of writing will depend on room/group size, but as a basic
rule, your letters should be not less than 25 mm high (approximately the distance from
the top of your thumb to the first joint).
Charts
Charts are one of the most flexible and reliable visual aids available to trainers who
conduct field-based training.
Advantages of charts
• Inexpensive, easily and quickly produced;
• Ideas for field-based training
• Easily transported and stored.
Making charts
Some rules on visibility
Lines drawn should be at least 3 mm thick
Lettering should be at least 25 mm high (length of top thumb joint)
Avoid drawing a lot of detail as it is normally unnecessary and often too small to be
seen from the normal viewing distance
The practical test of visibility is to view the chart (or any other visual aid) from the
position of the furthest trainees. Can you see it clearly?
Use of colour
Colour gives emphasis to important parts or shows parts which are related in purpose.
However, using too many colours can cause confusion, so try not to use more than
four colours.
Colour can be added with poster paints, felt-tipped pens or even coloured chalk.
However, filling large block of colour can be time-consuming and tedious,
particularly with felt tipped pens.
Design
Some trainers prefer to develop their own drawings, but this is normally unnecessary.
There are always plenty of examples and ideas to be found in booklets, leaflets and
manuals. As the HDM-4 programme develops it is hoped that more visual materials
will be available in training support materials and on the HDM-4 web site.
1 They fail to use the full visual areas available and so produce work which is
too small to be seen properly.
2 They include a lot of writing and labels on the diagrams and pictures . This
often reduces visual impact and adds rather than reduces confusion.
A good inexpensive screen can be made from a framed half-sheet of hard-board, the
roughened surface of which is given two coats of white emulsion paint. With two
hooks screwed into the top of the frame, it can be hung at an angle from the ceiling in
a corner of the training room. The angle ensures a square picture.
As with all visual aids, the lettering and details must be big enough to be visible from
the back of the room and the simplest way of verifying this is to go to the back and
see for yourself. When making an OHP sheet, first draft your diagram and any
lettering on a sheet of A4 lines paper, then place that under an acetate sheet and trace
it. Do not try to put too much information on one sheet.
When an OHP sheet is used for information only, or for presentation of data, it is
often helpful to the trainees if you reveal the contents of one point at a time by
covering the remainder of the slide with a piece of paper.
When using the OHP, you can stand facing the audience using a thin pen (not your
finger) to point out features on the slide. There is no need to turn your back on the
audience.
Summary
• Learning aids are intended to be tools for the trainer to use to make the
learning process more easy for the learner.
• Always make sure that the learning aid has a specific purpose in the learning
process.
• Learning aids are for use in the training process so don’t try to turn them into
detailed notes or make them over-complicated. They should complement the
training session an are not intended to stand alone.
• Make sure that everything in the aid is visible to all the trainees
• Use colour to make them more interesting
• Learning aids are useful when trying to link theory to practice.
• The use of learning aids should be carefully planned.
UNIT ONE
UNIT TITLE: - LEARNING STYLES
UNIT AIMS
1 To make you aware of your own learning styles and how your preferences can
influence your training style which in turn may favour some of your students
while disadvantaging others.
2 To introduce the Learning Cycle as a means of providing a framework for
planning training.
3 To emphasise the learner and their learning process as the primary focus in
teaching.
UNIT OBJECTIVES.
1. Complete the learning style inventories and defines your learning style
preferences
2. Give two examples of a learning style profile and state what strategies would be
needed by the teacher to strengthen the learning process in the examples presented
3. Describe a typical trainee behaviour, which would indicate the test profile given
Notes
This UNIT contains a Questionnaire, which allows the trainer to set an exercise,
focusing on individual styles of learning. By completing the questionnaire each
participant can plot his or her own learning style and from comparison with other can
see that everyone is different. It is essential to emphasise that there is no "best" profile
but rather that everyone can improve their learning skills by being aware of their
individual preferences
The results can then be firstly explained by reference to the “Learning Styles –
General Descriptions” It should then be related to the Learning Cycle diagram and the
Learning Cycle Overhead Transparency (OHT) given at the end of this Unit. This can
be used to demonstrate that the learning cycle (shown as four stages) is one that we
should try to copy. However, we all naturally tend to have preferences for two to
three stages only. The trainer should try to compensate for this by encouraging
students to engage in activities, which are less preferred by them. Also when planning
training sessions, trainers should try to provide activities which take the learner
through all four stages in the Learning Cycle
This questionnaire and diagnosis should not be taken too seriously as it indicates
tendencies rather than providing strict categories. What the exercise can do is to
highlight the importance of the trainer understanding that trainees do have differing
learning styles and the teaching must be able to accommodate these. It is also useful if
students can be encouraged to expand their own learning style to use the full range of
Finally trainers should be reminded that their own particular learning preferences
could favour students with similar profiles. Therefore trainers must remain conscious
of their own preferred styles when planning and conducting training. This means
taking students through the full learning cycle.
Example
This questionnaire is designed to find out your preferred learning style(s). Over the
years you have probably developed learning "habits" that help you benefit more from
some experiences than others. Since you may be unaware of this, this questionnaire
will help you pinpoint your learning preference so that you are in a better position to
select learning experiences that suit your style.
There is no time limit to this questionnaire. It will probably take you 10-15 minutes.
The accuracy of the results depends on how honest you can be. There are no right or
wrong answers. Use the Learning Style Answer sheet to record your answers. If you
agree more than you disagree with a statement put a tick by it (!). If you disagree
more than you agree put a cross by it (").
1 I have strong beliefs about what is right and what is wrong, good and bad.
2 I often act without considering the possible consequences.
3 I tend to solve problems using a step-by-step approach.
4 I believe that formal procedures and policies restrict people.
5 I have a reputation for saying what I think, simply and directly.
6 I often find that actions based on feelings are as sound as those based on careful
thought and analysis.
9 What matters most is whether something works in practice.
10 I actively seek out new experiences.
11 When I hear about a new idea or approach I immediately start working out how
to apply it in practice.
12 I am keen on self-discipline such as watching my diet, taking regular exercise,
sticking to a fixed routine etc.
13 I take pride in doing a thorough job.
14 I get on best with logical, analytical people and less well with spontaneous,
"irrational" people.
15 I take care over the interpretation of data available to me and avoid jumping to
conclusions.
16 I like to reach a decision carefully after weighing up many alternatives.
17 I'm attracted more to novel, unusual ideas than to practical ones.
18 I don't like disorganised things and prefer to fit things into a coherent patter.
19 I accept and stick to established procedures and policies so long as I regard
them as efficient way of getting the job done.
20 I like to relate my actions to a general principle.
21 In discussions I like to get straight to the point.
22 I tend to have distant, rather formal relationships with people at work.
23 I thrive on the challenge of tackling something new and different.
24 I enjoy fun-loving, spontaneous people.
25 I pay meticulous attention to detail before coming to a conclusion.
26 I find it difficult to produce ideas on impulse.
27 I believe in coming to the point immediately.
28 I am careful not to jump to conclusions too quickly.
29 I prefer to have many sources of information as possible - the more data to think
30 Flippant people who don’t take things seriously enough usually irritate me.
31 I listen to other people’s point of view before putting my own forward.
32 I tend to be open about how I’m feeling.
33 In discussions I enjoy watching the manoeuvrings of the other participants.
34 I prefer to respond to events on a spontaneous flexible basis rather than think
things out in advance.
35 I tend to be attracted to techniques such as network analysis, flow charts,
branching programmes contingency planning etc.
36 It worries me if I have to rush out a piece of work to meet a tight deadline.
37 I tend to judge people’s ideas on their practical merits.
38 Quiet, thoughtful people tend to make me feel uneasy.
39 I often get irritated by people who want to rush things.
40 It is more important to enjoy the present moment than to think about the past?
41 I think that decisions based on a through analysis of all the information are
sounder than those based on intuition.
42 I tend to be a perfectionist.
43 In discussions I usually produce lots of spontaneous ideas.
44 In meetings I put forward practical realistic ideas.
45 More often than not, rules are there to be broken.
46 I prefer to stand back from a situation and consider all the perspectives.
47 I can often see inconsistencies and weaknesses in other people’s arguments.
48 On balance I talk more than I listen.
49 I can often see better, more practical ways to get things done.
50 I think written reports should be short and to the point.
51 I believe that rational, logical thinking should win the day.
52 I tend to discuss specific things with people rather than engaging in social
discussion.
53 I like people who approach things realistically rather than theoretically.
54 In discussions I get impatient with irrelevancies and digressions.
55 If I have a report to write I tend to produce lots of drafts before settling on the
final version.
56 I am keen to try out things out to see if they work in practice.
57 I am keen to reach answers via a logical approach.
58 I enjoy being the one that talks a lot.
59 In discussions I often find that I am the realist, keeping people to the point and
avoiding wild speculations.
60 I like to ponder many alternatives before making up my mind.
61 In discussions with people I often find I am the most dispassionate and
objective.
62 In discussions I’m more likely to adopt a “low profile” than to take the lead and
do most of the talking.
63 I like to be able to relate current actions to a longer-term bigger picture.
64 When things go wrong I am happy to shrug it off and “put it down to
Circle your score from the score sheet for each category
This will show your most and least preferred styles
Activist Reflector Theorist Pragmatist Preference
20 20 20 20
19
18 19
17 19 Very Strong
16 19 18
15 18
14 17
13 18 16 17
12 17 15 16 Strong
16
11 15 14 15
10 14 13 14
9 13 12 13 Moderate
8
7 12 11 12
6 11 10 11
5 10 9 10 Low
4 9 8 9
3 8 7 8
7 6 7
6 5 6
2 5 4 5
4 4 Very Low
1 3 3 3
2 2 2
1 1 1
0 0 0 0
The general descriptions given below explain the tendencies associated with each
Category. No two people score the same but it does help the trainee and the trainer to
consider how they can improve their learning by attempting to strengthen their
performance in the columns showing lower scores. Read through these and consider
where your own scores place you in these descriptions. Remember these scores
simply indicate the strength of your tendency in any one field.
Fig 1
The Experience
(Exposure)
# $
% &
Obviously we all have our strengths and weaknesses and each learner’s style is a
combination of all four characteristics with some characteristics influencing
behaviour more than others.
For example
a) Someone who is a strong Activist and Pragmatist may be quick to pickup the
practical aspects in a learning environment but could experience difficulty if asked
to explain how something works, diagnose a problem or suggest changes in
procedures to improve performance.
b) Someone who is strong on Reflection and Analysis may not be able to pick up the
practical processes of a task quickly but would be more able to grasp and discuss
the underlying principles and their applications.
From these simple examples it will be seen that the trainer can provide a more
effective learning environment if trainees are encouraged to participate more fully in
those parts of the learning cycle which they normally tend to put less emphasis. It is
also important that the Teacher recognises that they also have to plan to give equal
emphasis to all four processes in the cycle irrespective of their own preferences.
Table 1 Table 2
Activist Reflector Theorist Pragmatist Preference Activist Reflector Theorist Pragmatist Preference
20 20 20 20 Very 20 20 20 20 Very
Strong Strong
19 19
18 19 18 19
17 19 17 19
16 19 18 16 19 18
15 18 15 18
14 17 14 17
13 18 16 17 13 18 16 17
12 17 15 16 Strong 12 17 15 16 Strong
16 16
11 15 14 15 11 15 14 15
10 14 13 14 Moderate 10 14 13 14 Moderate
9 13 12 13 9 13 12 13
8 8
7 12 11 12 7 12 11 12
6 11 10 11 Low 6 11 10 11 Low
5 10 9 10 5 10 9 10
4 9 8 9 4 9 8 9
3 8 7 8 Very Low 3 8 7 8 Very Low
7 6 7 7 6 7
6 5 6 6 5 6
2 5 4 5 2 5 4 5
4 4 4 4
1 3 3 3 1 3 3 3
2 2 2 2 2 2
1 1 1 1 1 1
0 0 0 0 0 0 0 0
Summary
This session has highlighted the importance of viewing each student as an individual.
It has also emphasised the teacher’s own learning style must be consciously
recognised and taken into account when planning and conducting teaching.
The linking of the concept of learning styles with the learning cycle demonstrates the
similarity of thinking in both these concepts. This similarity reinforce the idea that
teaching can and must be planned to ensure that the technical subject content and
structure and the teaching strategies and methods are assembled in such a way that
none of the parts of the learning process missed out or passed over.
This concept of structuring the learning process further reinforces the need for
teachers to structure material to satisfy the ways in which the student will have to
apply the knowledge and skills taught in their future employment and not being
restrained by the traditional structures of the subject being studied.
Conclusion
The principles established in this session must be used to reinforce more practical
sessions on planning, teaching methods and trainee participation.
These are:
The trainee is an individual and has to be seen in that context
Training methods must allow and encourage students to undertake the activities
identified in the learning cycle (telling and showing is not enough)
THE
EXPERIENCE
# $
PLAN THE REVIEW THE
NEXT STEP EXPERIENCE
% ANALYSE and
&
DRAW CONCLUSIONS
from the
EXPERIENCE
15 18
14 17
13 18 16 17
12 17 15 16 Strong
16
11 15 14 15
10 14 13 14
9 13 12 13 Moderate
8
7 12 11 12
6 11 10 11
5 10 9 10 Low
4 9 8 9
3 8 7 8
7 6 7
6 5 6
2 5 4 5
4 4 Very Low
1 3 3 3
2 2 2
1 1 1
OHT 3
UNIT AIM
To introduce a system of course evaluation which is simple to design and use and which can
give trainers useful feedback on which to base improvements in the session/programme.
UNIT OBJECTIVES
1 Identify the major areas in which formative feedback is useful to the trainer
2 Design a simple evaluation document for use in a course being taught by trainers
3 Use the evaluation, process the feedback data and make recommendations for
improvements.
Trainers notes
With any training course it is essential that it should be evaluated. A primary evaluation
criteria will be whether the trainees have achieved the objectives. However it is also essential
to have feedback from the trainees on how they found the course. A simple but useful
method is given in the UNIT text below. What is important is to find out if the trainees have
found the various parts of the course programme useful and where if any there has been parts
which have been less effective. By using the scheme given and particularly the conversion of
the results into a bar graph it is very easy to make comparisons between sessions and equally
importantly between the same course delivered to other groups.
The results of such evaluations are extremely useful in diagnosing where the trainers have to
review and improve units or sessions
Without such evaluation it is impossible to review or improve training. As part of the whole
programme it is important that every effort is made to monitor the training being given by
trainers you may be supervising.
Unit 9 1
COURSE EVALUATION
At this stage in the development of courses in HDM-4 for different groups it is important for
trainers to have early feedback on the groups opinion of the content of the course and how
useful they considered it to be to them in their future work
In the Train the Trainer Course you have just had, such a simple evaluation as given below
might be used. However, because of the duration of the course and large number of training
sessions it contained, many participants would have difficulty remembering all the sessions
and so a different method has been used. In this course the evaluation of each day of the
programme was carried out by the group feed-back sessions held each morning The process
was:
Each group was invited to provide a brief summary of the previous days programme as they
recalled it. This allowed the trainers to assess if the participants had generally grasped the
main concepts and issues relating to these sessions. Secondly groups were asked to highlight
any particular points that were either of particular interest to them or to draw attention to
areas where they felt they were not sure about or that they thought its delivery could be
improved. Tutors then summarised the group feed-back as they had perceived it and
discussed what action was needed.
This process is known as Formative evaluation as it allows participants and trainees to adjust
and change as the programme goes along. Summative evaluation is carried out at the end of a
programme and does not allow change to the existing programme but it can help with the
development of subsequent courses.
For the Summative evaluation on this course we are using two tools.
Forms A1 and A2 given at the end of this Unit allows us to make a comparison between your
entry level of expertise and your exit level as you perceive it. This information and your work
during the many practical exercises and sessions provides an accurate picture of the success
the programme has had in improving your skills and knowledge. However it is not an
examination and in most cases an examination approach is unnecessary and can be a negative
factor in adult courses.
To get a picture of individual trainee opinion on the programme it is possible to put out a
questionnaire that gathers opinion from each participant individually. The simple example
given below is one and for short courses this can be very useful. However as already
mentioned, in a three week programme with a wide range of topics and activities such a
method become less effective and produces large quantities of responses that may or may not
be significant.
In this case we are interested in using a method which will highlight major strengths and
areas in need of improvement and is in a form that is practical to manage. One method that
allows this filtering process to be used is the one outlined in Form C.
There is a wide range of evaluation and assessment tools available to the trainer and trainers
should choose those that best suit their needs. The main issue to be considered is that the
evaluation process in adult courses such as this one is not to test the participant. It is to ensure
that the training programme, its content and the methods used are continuously being
reviewed and adjusted to meet the learning needs of the participants.
Unit 9 2
EXAMPLE
Course evaluation for a short course
This approach may be useful in short course such as those being developed for “HDM-4 for
policy makers” (See Units Eight example programme)
Each Session or Topic received by the course participants is entered and participants provide
responses . This information will allow the training team to see which sessions have been
well received and which ones may need some attention.
From this information the trainer or training team will be able to consider where
improvements can be made to the content (and Methods) in future courses for groups with
similar needs.
Comments
Unit 9 3
SUMMARY OF THE INDIVIDUAL PARTICIPANT RESPONSE SHEETS
Enter the total number of responses taken from the individual participant sheets as below.
(with 20 participants each session line should add up to 20)
Session and Topic Very Useful Not
Useful Very
Useful
(3) (2) (1)
Session 1
Overview of Highway Management systems 7 12 1
Session 2
Development of Computer solutions 14 6
Session 3
HDM-4 use and analytical 17 3
Session 4
Highway information and its use in HDM-4 2 10 8
analysis
Session 5
HDM-4 A planner’s tool 11 9
Session 6
Packaging methods for specific products 16 3 1
Session 7
Adoption of HDM-4 19 1
Session 8
Demonstration 6 9 5
It is possible to make a graphical comparison by giving the score value in ( ) i.e. (3),(2) or (1)
for each entry in each column.
Example Session 6
Very useful =16 x (3) = 48
useful = 3 x (2) = 6
Not useful = 1 x (1) = 1
______
total = 55
If all the participants gave a "very useful" response then the total would be 60
The actual figure recorded in the example was 55 and this converts to 91% (55/60 x 100)
A graph of the results for the programme is below and indicates that session 4 needs attention
and sessions 1 and 8 may also benefit from some considered changes.
Unit 9 4
100
90
80
70
60
50
40
30
20
10
0
session session session session session session session session
1 2 3 4 5 6 7 8
Unit 9 5
FORM A
HDM-4 Train the Trainers Course
Kuala Lumpar. June 2000
Unit 9 6
FORM B
HDM-4 Train the Trainers Course
Kuala Lumpar. June 2000
Unit 9 7
FORM C
HDM-4 Train the Trainers Course
Kuala Lumpar. June 2000
Task B
Gather together as a whole course group and appoint a chairperson to manage the Task
Consider the responses of the groups and select THREE positive and THREE areas for
change from all the group opinions which everyone can agree with.
As a whole group, compose a short statement (100words) which you feel summarises the
group feelings about the course
Record and print out.
Unit 9 8
UNIT FOUR
UNIT AIMS
1. To introduce the concept of Qualitative and Quantitative criteria as a basis for
defining training needs.
UNIT OBJECTIVES
1. State the five criteria to use when carrying out training needs analysis in a work
environment.
2. State why it is important to use a wide range of information sources when
conducting training needs analysis.
3. Demonstrate the ability to complete the tables 3 and 4 successfully within a
practical exercise in the programme
Trainers notes
The text follows through the process in a logical way and is developed from the
Training Cycle diagram provided. The example tables provided models for guidance
and it is recommended that training needs analysis is best largely done as a practical
exercise in the real work situation. The Concept of the Training Gap is a useful place
to begin and also it must be remembered that training needs have to be related to
specific target groups. The quality of a training programme will rely largely upon the
quality of the training needs analysis . An analysis done sitting in the office will
invariable become a work of half remember facts and fiction. The analysis is best
carried out in the work place and with direct observation of the work process and
contact with those doing the job.
Finally the “five rules of training” are a very useful framework to use. If the training
being recommended does not meet these rules then it likely that it will fail and be a
waste of time for trainer and trainee alike.
The OHT on "Training" at the end of the UNIT is a reminder that training is only one
part of the equation and in defining training needs it is always necessary to establish
why a particular process is being followed. The boxes around the training box in the
OHT illustrate other factor that may have to be addressed before training is
considered
1
TRAINING NEEDS ANALYSIS
INTRODUCTION
Training Needs Analysis (TNA) is a a means of establishing the difference between
the present levels of knowledge, skills and attitudes in the individual and what the
work place requires for efficient and effective performance. We refer to this
difference between the present knowledge, skills and attitudes and the desired
knowledge, skills and attitudes as the:
It is worth remembering that training does not take place in a vacuum and there is
little point in providing expensive training if the problem is related to lack of
equipment or money, management structures, or politics etc. However it is usually a
combination of issues in which the provision of training is part of the solution.
Ultimately the criteria for successful training is that the FIVE output criteria for
training must be applied. If training is to be judged successful then it must achieve
one or more of the following improvements
Plus
Improved ENVIRONMENT ?
In the training needs analysis it is important to identify the Knowledge, Skills and
Positive Attitudes required to bridge the training gap identified. It is also important to
ensure that the trainer keeps the required balance between the needs of all three areas.
2
The above list illustrates the wide range of sources from which information can be
collected.
Exercise
Review the list above and draw up a list possible sources of information on training
needs for HDM4. Once you have done this rank them in order of descending order of
importance. When preparing your training programme you will have to make use of
these sources. Observing work in progress is just as important as sitting in a
comfortable chair having a discussion with the site and base office staff on their work.
All sources of information are important and useful but the most important ones are
direct observation of, and contact with, the person doing the job in their normal
working environments.
• The information being given and the practical skills and methods introduced
are compatible with the current systems being used;
• That it doesn't require equipment and materials that are not available.
TABLE 1 EXAMPLE
Preliminary Estimate Of Training Needs In Work Areas
3
EXERCISE (GROUPS)
Using Table one layout as an example, consider the HDM4 programme, identify a
main Area and some Processes. and try to describe the likely difficulties that the
target group may have. Select a target group and remember that if the target group
already have some experience with HDM3 or similar programmes, that experience
must be recognised in the training programme design
Task
Using this basic information make a first attempt at defining what difficulties the
target group might experience. Keep it basic and simple
Table 2 gives an example we can use to refine the process. This will require the
various processes associated with HDM4 to be accurately identified.
4
Table 2
POST: e.g. engineer, data collector, on site supervisor, policy maker, university student/
etc (choose only one category at a time)
Q: Do you/would you experience any problems with the following tasks/duties?
2 calibrating HDM-4
5 Interpreting HDM-4
financial reports
6
10
11
12
13
14
5
Table 2
Position:
Q: Do you/would you experience any problems with the following tasks/duties?
10
11
12
13
14
6
OHT
ANALYSIS
OF THE
TRAINING NEED
! "
EVALUATION PLANNING
OF THE THE
TRAINING TRAINING
# CARRYING OUT
$
THE
TRAINING
7
TRAINING NEEDS ANALYSIS
Policies
Organisation Politics
Management Rewards
Resources
8
Training should equal
FASTER
HIGHER QUALITY
GREATER QUANTITY
LESS EFFORT
COSTS LESS
(ENVIRONMENTAL IMPACT)
9
ASSESSING TRAINING NEEDS
(SOURCES OF INFORMATION)
% First hand observation and information
gathered from the work place.
10
UNIT FIVE
TITLE: AIMS AND OBJECTIVES IN TRAINING
UNIT AIM
1 To introduce the concept of aims and objectives as a planning and evaluation tool
so that planning and teaching can be accurately matched to target groups and
training needs
UNIT OBJECTIVES
1 State the main parts of an aim and objective
2 Explain the relationship between trainer and aim and trainee and objective
3 Illustrate through example how the choice of verbs can indicate the level of
performance required and the possible training strategies to be used.
Trainers notes
This UNIT is a key area for the trainer. As it was once said "If you do not have a
destination any road will do" The purpose of Aims and Objectives is to allow the
trainer and the trainees to have a clear destination. The Aim is to guide the trainer and
the objective is for the trainee to achieve. The text supplied gives guidance on the
writing of aims and objectives. There is also a list of verbs which can be consulted to
indicate the level of performance required by the student As the verbs used in
objectives moves from Knowledge to Evaluation the required student performance
will become more difficult to achieve as each level (1 to 6) builds upon the mastery of
the level below.
It can therefor be helpful in planning training to ensure that the training objectives are
progressively scaled from 1 to 6 as the training programme progresses. A similar
framework exists for the practical skills list of verbs.
It should be remembered that aims and objectives for a complete programme will be
phrased in more general terms and will then become more specific as the planning
process moves to specific training sessions. While objectives are there for students to
achieve the trainer should also remember that testing objectives has to be realist in
terms of time and equipment needed.
Finally the verb in the objective will provide some indication of what teaching
approaches will be needed. For example if the student has to "select" then part of the
training session will be giving information on the criteria needed to be able to select
and to provide a range of what is acceptable and what is not. e.g. ripeness of fruit. . It
will also be necessary to provide a demonstration and student practice in the selection
process. The advantage of using objective is that the teacher knows exactly what to
teach for the student to achieve the performance stated in the objective.
1
AIMS AND OBJECTIVES IN TRAINING AND TEACHING
Aims and objectives are widely used in training and teaching and their use has made a
major contribution to the development of effective training in the last thirty years.
It states what subject matter you intend to teach and why you feel justified in having
this intention. This justification is as important as the intention and helps ensure that
you:
In themselves the achievement of aims cannot be evaluated because there are external
factor which can influence the long term achievement of the Aim, you cannot
guarantee to achieve your aim. However you can test to ensure that all the component
parts of knowledge and skills have been mastered. The criteria for these tests are
specified in the objectives.
The objective describes what your trainee will be able to do which will demonstrate
that the teaching and learning process has succeeded. An objective based on new
abilities or behaviour is known as a behavioural objective.
• what the trainee will be able to do to show that learning has taken place;
• the conditions under which the trainee will demonstrate this ability e.g. time
after the training session, the location and limits on facilities or assistance;
2
An objective for the aim we have already seen could be:
The Trainee will:
From appropriate sources, prepare input data required for investigating optimum
road maintenance standards for paved roads and enter into HDM-4 programme.
Assign a range of maintenance standards to road sections and run HDM-4.
Interpret the results from HDM-4 results and recommend optimum maintenance
standards for paved roads.
Consider the alternative set of objectives below which you could apply to this case
What do you think are the main differences
Through the case study used trainees will
Review and comment upon the input data prior to entry.
Run HMD-4 successfully and achieve an output
Examine and discuss the key issues arising from the output.
Objectives must be realistic in terms of both what we can expect of a trainee at that
stage of training and what we will actually be able to observe and evaluate.
It is the most widely developed method of planning so it has been seen to work.
Trainers are encouraged to make sure that their planning is relevant to the needs of
their trainees.
They encourage trainers to make explicit the values underlying their training.
They provide a rational basis for evaluating training and help in the achievement of
success.
3
They set goals for the trainees.
They may slow down the process of change and development in training.
The more superficial objectives are easier to define and assess e.g. recall rather than
evaluation.
Writing true behavioural objectives for some worthwhile educational goals is far from
easy e.g. in the affective (attitude) domain.
In a training session the use of objectives may limit the responsiveness of the trainer
to unexpected events and opportunities.
Objectives tend to lead to the view that there is one correct sequence for teaching a
subject and one ideal method.
4
EDUCATIONAL OBJECTIVES IN THE COGNITIVE DOMAIN
This system defines in a hierarchical fashion different general, i.e. not subject-
specific, types of capability that may be developed in a student as a result of learning
experiences. These capabilities are described by reference to the actions that the
student may be able to perform which demonstrate the achievement of the various
levels of objective. The scheme therefore identifies and classifies the products of
learning experiences.
5 Synthesis Ability to put ideas together in new ways that reach beyond what
has been specifically taught
5
6 Evaluation Ability to judge the quality or value of material by reference to
appropriate criteria
judge support identify evaluate
defend avoid determine attack
select recognise choose criticise
6
Learning Style Score Sheet
15 18
14 17
13 18 16 17
12 17 15 16 Strong
16
11 15 14 15
10 14 13 14
9 13 12 13 Moderate
8
7 12 11 12
6 11 10 11
5 10 9 10 Low
4 9 8 9
3 8 7 8
7 6 7
6 5 6
2 5 4 5
4 4 Very Low
1 3 3 3
2 2 2
1 1 1
0 0 0 0
Participation equals
! Discussion
! Questions - answering and
asking
! Looking Observing
! Explaining Describing
! Problem solving
! Carrying out practical processes
! Sharing ideas, opinions and
experience
! LISTENING
3Participation.doc
TRAINING NEEDS ANALYSIS
Policies
Organisation Politics
Management Rewards
Resources
1
Training should equal
FASTER
HIGHER QUALITY
GREATER QUANTITY
LESS EFFORT
COSTS LESS
(ENVIRONMENTAL IMPACT)
2
ASSESSING TRAINING NEEDS
(SOURCES OF INFORMATION)
! First hand observation and information gathered
from the work place.
3
illusionunit7
What is it?
unit7
Where is the Ball?
Objectives
j - Cognitive
g Hierarchy
y
Synthesis Put ideas together in new ways that reach beyond what
has been taught
C
Comprehension
h i Expressing ideas in own words or transposing into
E
alternate forms.
M i l ti
Manipulation F
Follows instruction to perform skill
I am happy I am p
puzzled
? ?
?
?
Philippines
pp
• Climate : Tropical
Humid
• Roughness Age : 1.09
1 09
• Cracking Initiation : 0.833
• Cracking Progression : 1.2
• Roughness Progression: 11.22
• Factors from Tables 7.3-7.4 : 0.025,
1.2, 1.0
Roughness Comparison
18
16
14
12
10
IRI Before
IRI
IRI After
8
0
1998 2000 2002 2004 2006 2008 2010 2012 2014 2016
Year
Summary
• Calibrating HDM to suit the
environmental conditions in a
tropical-humid region or area
can p
predict the deterioration of
the road section under study.
F t
Factors: Roughness
R h Age(Kge)
A (K ) = 11.3 3
Roughness Progression(Kgp) = 0.7
crack initiation (Kci)= 1.5
15
crack progression( Kcp)=1.5
2020
2015
yee a r
2010
2005
2000
1
3
5
7
9
11
13
15
17
19
iri roughness
without vr with adjustment
20.00
15 00
15.00
10.00 iri values
5.00
0 00
0.00
0.00 5.00 10.00 15.00 20.00
Conclusions:
hdm4 default values
- predicts crack initiation early
- predicts faster road deterioration
3 years before the road fails (IRI
(IRI> 16)
• Use Papua
p New Guinea Case
Tropical Humid
R
Roughness
h A
Age EEnvironment
i (Kgc)
(K )
• Environmental Coefficient (m) = 0.025
• Km (Table 7.4)
7 4) = 11.0
0
• meff = mKm = 0.025
• Roughness age environment (Kgc) =
meff/0 023 = 11.087
meff/0.023 087
C ki IInitiation
Cracking i i i (Kci)
(K i)
• Kcp = 1/Kci
• No adjustment
j
100.00
80.00
International Group
60.00
40 00
40.00
Calibrated
20.00 Base
0.00
1995 2000 2005 2010 2015 2020
Year
Summary
Q 4.
Answer 4:
HDM-4
IIntroduction
t d ti to t Program
P
Analysis
Obj ti
Objective
• Method:
– Life cycle analysis
– Budget period analysis
• Objective
Obj ti function:
f ti
– Maximize NPV
– Maximize road network condition
Budget allocation
• Routine maintenance costs are based on
road lengths.
• Recurrent costs are based on pavement
p
surface condition.
• Periodic maintenance are based on
economic indicators.
• Improvement needs are based also on
economic indicators.
Budgeting Procedure
• When sufficient budget is available:
– Select all independent projects with NPV>0
– Select
S l t mutually
t ll exclusive
l i projectj t alternative
lt ti
with the highest NPV.
• When
Wh there
th is
i budget
b d t constraints:
t i t
– Select the independent project using the
highest NPV/Cost ratio
– Select mutually exclusive projects using the
i
incremental
t l NPV/Cost
NPV/C t
Prioritization
Project Cost NPV NPV/Cost Rank
A 150,000 35,000 0.233 2
B 150 000
150,000 45 000
45,000 0 300
0.300 1
C 200,000 40,000 0.200 3
D 250,000 37,000 0.150 6
E 500 000
500,000 70 000
70,000 0 140
0.140 7
F 500,000 77,000 0.154 5
G 500 000
500,000 78 000
78,000 0 156
0.156 4
Incremental Analysis
Project
j Cost NPV NPV/Cost Rank
A 20,000 4,869 0.2435 2
B 30,000 240 0.0080 8
C 20,000 5,184 0.2592 1
D 20,000 210 0.0105 7
E 40,000 6,401 0.1600 4
F 35,000 4,497 0.1285 5
G 20 000
20,000 11,200
200 0 0600
0.0600 6
H1 80,000 15,000 0.1875 3
H2 100,000 16,829 0.1683
D fi iti
Definition
• Program analysis is concerned
with the preparation of single or
multi-year work program for the
road network by maximizing the
economic return (INPV/Cost)
subject
bj t to
t budgetary
b d t constraints.
t i t
Thank You
D Basic Requirements
D Limitations of HDM-
HDM-4
z Data
D R
Requirements
i
É Computer
486
486--DX with processing speed 100 MH
Monitor and other accessories
É Environment
Windows (95/98 or advanced)
È Road
R d N
Network
kDData
È Classification of Roads
È Section
S i Number,
N b
È Links Identification,,
È Nodes Number
Vehicle Fleet Data
È Vehicle name,
name Vehicle class,
class Base Type,
Type
Category
È Traffic Data
È Vehicle type, Characteristics, Fleet
Composition Volumes
Composition, Volumes, Growth rates
rates, Axle
loading, Speed flow type and traffic flow
p
pattern
È Pavement
P t Condition
C diti Data
D t
È Cracking, Rutting, Potholes, Edge Break etc.
etc.
È Pavement Ride Quality Data
È Roughness and Skid Resistance
ÈWork Standards
È Maintenance Standards
(items of work defined for maintenance)
È Improvements Standards
È Unit Cost Data
È Economic Unit Price
È Financial Unit Price
È Climatic Data
È R i f ll Temperature,
Rainfall, T t Moisture
M it level,
l l Freeze
F
Index etc.
È Environment
E i t Data
D t
È Emissions, day night temperature etc.
HOWEVER
Precisely
HDM-4 Model is applicable with suitable
HDM-
calibration and for the set of said
The software is easy to Use and User friendly
Evaluation of many criteria at a time is possible
The Model could be easily modified calibrated
and simulated for local conditions
Standard Reports could be easily generated
Relatively less processing time is required for
analysis
l i
Project, Program and Network Level of analysis
Prioritization of different strategies and works
can be
b carried
i d out
Comprehensive plans for development and
maintenance
i t with
ith optimized
ti i d budgeting
b d ti can be b
done
H l iin decision
Helps d i i making ki withith scientific
i tifi andd
logical reasoning
Th reliability
The li bilit andd confidence
fid level
l l for
f the
th
result is high
Li it d resources andd information
Limited i f ti isi needed
d d
The development is based on the feedback
If applied properly, the software can help in
reducingg the highway
g y budget
g byy allocatingg funds
to the projects which have high economic return
The software is fully compatible with the
existing data base.
The development of HDMHDM-4 4 is based on the
authenticated and documented research work by
leading research organization of world
The software and the support material can
provide the student the basic principals involved
in planning and development of highways and
road
The students may y be having
g advantage
g of usingg
the software earlier and can help in carrier
buildingg
The academic use of HDM-4 can assist the
student in the exposure of practical working
HDM - 4
HIGHWAY DEVELOPMENT & MAINTENANCE
ASSIGNMENT OF
MAINTENANCE AND
IMPROVEMENT OPTIONS IN
HDM-4 ANALYSIS
Presented by:
Md. Amir Azam
Executive Engineer
(Maintenance),
LGED, Bangladesh.
Assignment of Maintenance
and Improvement Options
in HDM-4 Analysis
AIM
To teach how to assign maintenance
p
and improvement options
p in HDM-4
analysis:
4 so that the trainees can select the base option
p
and the other options to compare with, in
HDM-4 analysis.
Assignment of Maintenance and
Improvement Options in HDM-4
Analysis
y
OBJECTIVES
* The trainees will:
* Describe different maintenance and
improvement options available in HDM-
4.
* Demonstrate
D t t their
th i ability
bilit to
t use those
th in
i
HDM-4 analysis answering related
q
questions.
Output Data in HDM-4
Analysis
Road Network
* The Road Network folder provides the basic
facilities for network referencing within HDM-4.
* It is central to the operation of HDM-4 .
* It allows users to define different networks
and sub-networks
sub networks, and section of roads
which are the fundamental unit of analysis in
HDM-4 .
Input Data in HDM-4
A l i
Analysis
Vehicle Fleet
* Representation of vehicles is key data
requirement for HDM-4 .
* Vehicle Fleet folder provides facilities for storage
and retrieval of vehicle fleet details udes for
calculation of:
- vehicle speed
- operating cost
- travel
t a el time cost
- other vehicle effects.
Input Data in HDM-4
Analysis
HDM Configuration
* It is recognised that HDM-4
HDM 4 will be used in wide
range of environment.
* HDM Configuration folder provides the facilities
to costomise system operation to reflect the
local conditions.
Input Data in HDM-4
A l i
Analysis
[[Operation
p 1]+[Intervention
] [ criteria]+[Limits]+[Effects]+[Cost]="Works
] [ ] [ ] [ ] item 1"
HDM-4 Workspace
Input Data in HDM-4
A l i
Analysis
Questions?
Assignment of Maintenance
and Improvement Options
in HDM
HDM-4
4 Analysis
Conclusion:
Selection of alternatives,, and maintenance
and improvement options to the road
pp p
network is central for appropriate
selection of work standards to:
improve the quality of services,such as:
⌧relieving traffic congestion
⌧road safety
⌧road passibility
⌧accommodating increased axle load.
Thank You All
Use of HDM-4 & Its
Analytical Frame Work
T
Target
t Group
G
Roads
85%
Hi t
History off Roads
R d
and
Operations Programming
g g
Preparations
ROLE OF HDM-4 IN HIGHWAY
MANAGEMENT
• Planning
• Programming
• Preparations
• Operations
O i
HDM--4 SYSTEM ARCHITECTURE
HDM
HDM--4
HDM Standard
& Custom
Reports
Integrating HDM
HDM--4 within Road
Management Systems
Long Term
Road Strategic
Inventory
DATABASE HDM--4
HDM Road Plan
Bridges P li i
Policies
Condition Standards
Research
INPUTS OUTPUTS
NETWORK ELEMENTS
• Links
• Sections
– a length of road that is homogeneous in terms of
its physical attributes (basic unit of analysis)
• Network
• Sub-
S b-networks
Sub t k
PAVEMENT CONDITION
– Roughness surveys
– Deflections
TRAFFIC DATA
•VVolume
l AADT (MT andd NMT)
• Traffic Composition
p and ggrowth rates
• Categories - Normal, Diverted,
Generated
• Axle loading
• Traffic flow pattern
• Roadside friction
Calibration To Local Conditions
• Vehicle Fleet
• Unit Costs
• Works Standards
• Model Calibration
• Language
Works Standards
• Works types
– maintenance
– improvement
p
• Unit costs
• Construction
C t ti quality
lit
– material properties
– construction quality
– compaction,
compaction etc.
HDM-4 ANALYSES
z Strategy analysis
z For preparing medium to long term
planning estimates of funding needs for
road network development & maintenance
z Programme analysis
z For preparing single or multi-year work
programmes underd budget
b d t constraints
t i t
z Project analysis
z For evaluating the economic or
engineering viability of different road
investment projects
PRIORITISATION
• Under budget constraints:
– Independent projects by NPV/Cost
– Mutually exclusive projects by incremental
NPV/Cost
• When there is no budget constraint:
– Independent projects by NPV or IRR
– Mutually exclusive projects by maximum NPV
Summary or Conclusion
playing
l i an important
i t t role
l in
i building
b ildi
infrastructure
s uc u e
Modeling Non
Non--Motorized
Transport (NMT) in HDM
HDM--4
• 10 to
t 90 % off Traffic
T ffi ini China,
Chi India,
I di
Indonesia, and Bangladesh is NMT
• In Tokyo
Tokyo, about 30% of the total trip is
cyclist
• Wide range of intermediate modes of transport
between pedestrians and MT in Asia and hence
Transport
p systems
y in Asia are veryy efficient
NMT Vehicle Types in HDM-
HDM-4
• Pedestrians
• Bicycles
• C l Rickshaws
Cycle Ri k h
• Animal Carts
• Farm Tractors (not implemented)
Effects of NMT on Economic
Analysis
• MT Flow and Speed
• p
MT Operatingg Costs
• NMT Speed-Flow ( Partially Implemented)
• NMT Operating
p g Costs
• NMT Energy Consumption
• MT and NMT Safety Related Cost (NI)
• Road Deterioration and Maintenance (NI)
• NMT Travel Demand (NI)
Modeling Impact of NMT
Calculate
Operating Speed
For each NMT type
Calculate
Time and Operating Costs
Perform
Economic Analysis
and
Energy Balance Analysis
Conclusions
•Economic Viability
•Engineering Viability
off R
Road
d IInvestment
t t
projects/alternatives/options
TYPES OF PROJECTS
› Maintenance
› Rehabilitation
› Widening
› G
Geometric
i improvement
i schemes
h
› Pavement upgrading
› New Construction
KEY FACTORS
Structural performance of road pavements
Life cycle predictions of road deterioration,
deterioration
road works effects and costs
Road user costs and benefits
Economic comparison of project alternatives
Structural Performance of Road
Pavements
ƪTYPES CONDITIONS
Basic
B i performance
f under
d ƪTraffic
different conditions over ƪClimate
ƪ
time ƪWeather
ƪDesign standard
ƪConstruction standard
Road User Costs vs Benefits
User Costs Benefits
>Fuel,
F l Oil,
Oil Lubricants
L bi t L
Lower V.O.C
VO C
>Tyres, Spare Parts Shorter travel time
>Travel time
> Vehicle
V hi l accidents
id t
Life Cycle Predictions
• Road deterioration
• Road work deterioration effects
• Road user deterioration effects
• Socio-economic and environmental
deterioration effects
Economic Comparison
z Technicians
z 10 - 15 participants
Surfacing
Road Base
Data Base
Base HDM-4
///\\///\\
Subgrade
Introduction
10
IRI (m/k
8
6
4
2 S f
Surface Treatment
Asphalt Concrete
0
2000 2005 2010 2015 2020
Y
Year
Introduction
z Traffic Data
z Climatic Data
Types of Road Section Data
z Definition
z Road Inventory
z R d Geometry
Road G
z Pavement
z [Configuration Data]
Definition
z Pavement Type
z Length (m)
z Carriageway Width (m)
z Shoulder Width (m)
[also No. of Shoulder and Edge Step (mm)]
z Number of Lanes
Road Geometry
Routine works AMGB AMSB AMAB AMAP STGB STSB STAB STAP
R
Reseal
l STAP STAP / STAP STAP STGB STSB STAB STAP
STSB 1
Mill & replace to **AP **AP **AP **AP N/A **SB **AB **AP
intermediate
y
surface layer
Mill & replace to **GB **SB **AB **AP **GB **SB **AB **AP
base
Source: NDLI (1995
Summary
AIM:
To introduce to the participants the different
types of road deterioration that will aid in
treatment selection as a basis for establishing
maintenance and improvement standards
through the use of HDM-4
Types of Road Deterioration
OBJECTIVES:
The pparticipants
p will be able to find the basis of
selecting appropriate maintenance and
p
improvement standards.
The participants will be aware of the importance of
different types of road deterioration as a vital inputs
of HDM-4 in the selection of appropriate
maintenance and improvements standards
standards.
Types
yp of Road Deterioration
Overview of Road Types
• Road inventory
• Roughness survey
• Deflection survey
T
Types off Road
R dD Deterioration
t i ti
presented by
ERWANTO WAHYUWIDAYAT
INDONESIA
Kuala Lumpur
20 June 2000
2/3/2014 Erwanto 1
HIGHWAY INFORMATION & ITS
USE FOR HDM-4 ANALYSIS
OUTLINE OF PRESENTATION
2/3/2014 Erwanto 2
HIGHWAY INFORMATION & ITS
USE FOR HDM-4 ANALYSIS
AIMS
To teach the student by providing basic
knowledge in making use of highway
information for HDM-4 analysis, so that
the student could interpret the result
appropriately
OBJECTIVES
• The student will be able to describe in
broad manner of the requirement of
highway information for HDM-4 analysis
• The student is expected to be able to
explain
l i the
th potential
t ti l power andd outcome
t off
HDM-4 analysis and its use for decision
making
2/3/2014 Erwanto 3
THE ROLE OF INFORMATION
IN HIGHWAY MANAGEMENT
2/3/2014 Erwanto 4
CONCEPTUAL FRAMEWORK OF
HDM-4 STRUCTURE
2/3/2014 Erwanto 5
TYPE OF BASIC DATA REQUIRED
FOR HDM-4 ANALYSIS
• Road Data
• Road network
• Road geometry
g y
• Road condition
• Road strength
2/3/2014 Erwanto 7
SUMMARY
• Highway
g y information is one of the
key factor in the decision making
process within road transport sector
management
2/3/2014 Erwanto 8
CONCLUSION
• HDM
HDM-44 is
i a tool
t l that
th t could
ld assist
it
road administration and policy
maker in managing their road asset
i the
in h better
b way
2/3/2014 Erwanto 9
HIGHWAY INFORMATION & ITS
USE FOR HDM-4 ANALYSIS
THANK YOU
2/3/2014 Erwanto 10
Wh M
Why Manage??
Resource is Limited.
Limited
Managing
g g is Difficult
Budget
g Allocation
for
R dM
Road Maintenance
i t
Role of HDM
HDM-4
4
in
R dN
Road Network
k Pl
Planning
i
Process
Role of HDM-4 in
Road Network Planning Process
Outline
• Network Management Process
• The Planning Process
• Performance Indicators
• HDM-4
• Role of HDM-4
Network Management Process
PLANNING
PROGRAMMING
PREPARATIONS
OPERATIONS
Planning
Analysis
y of the road system
y as a whole.
Required
What-If Resources?
Strategies
Analysis
Benefits?
The Planningg Process
with HDM-4
Collected Data
What-If
What-
Required
q
Analysis Resources?
Strategies using
HDM--4
HDM Benefits?
Benefits
• Allows modelling and prediction of future
road and traffic conditions.
• Enables testingg of various road
improvements and maintenance strategies.
• Provide objective method to estimate
budget requirements over several years.
• Better estimates of Annual Maintenance
Requirements.
– Planned
Pl d Maintenance
M i t after
ft Conceiving
C i i
Summary
• 4 Stages in the
– Network Management Process
• Planning Stage is Critical
– What if no planning?
• Performance Indicators
– NPV = Benefits
fi - Cost
Conclusion
Role of HDM-4
HDM4
A DECISION
MAKING TOOL FOR
MANAGEMET
VEHICLE NAME
VEHICLE CATEGORY
VEHICLE CLASS
BASE TYPE
VEHICLE PCSE
FREE SPEED
FIELD SURVEY
AVARAGE OCCUPANCY
ECONOMIC AND FINANCIAL
COST
VEHICLE
TYRE
TYRE RETREADING
CREW WAGE
MAINTENANCE LABOUR
FREE SPEED
FIELD SURVEY
AVARAGE OCCUPANCY
ECONOMIC AND FINANCIAL
COST
INTEREST RATE
FUEL TYPE AND COST/LIT
LUBRICANT
AVERAGE ANNUAL KM RUN
AVARAGE ANNUAL HOUR RUN
% OF NON HOME WORK TRIPS
VALUE OF PASSENGER WORK AND NON WORK
TIME
VALUE OF GOODS HOLDING TIME
AXLE LOAD
TO TRANSFER VEHICLE LOAD TO
STANDARD AXLE LOAD
NOTE VEHICLE NAME, CATEGORY,
TYPE AND NUMBER OF AXLES
BRING ONE WHEEL OF EACH AXLE
OVER THE WEIGH BRIDGE AND
RECORD
AXLE LOAD (Cont..)
(Cont )
TRANSFER DATA TO A FREQUENCY
TABLE
GET DAMAGE FACTOR FOR EACH
VEHICLE
GET TOTAL AXLE LOAD FOR EACH
AXLE
GET EASL FOR EACH VEHICLE
TYPE
XLE LOAD SURVEY TABLE -1
VEHI AXLE- AXLE- AXLE- AXLE AXLE- AXLE-
CLE 1 2 3 4 5 6
BUS 2 4
(L)
TRUC 2.5 5.4
K(L)
BUS 2.5 6.5
(H)
TRUC 3.0 6.0 5.6
K(H)
SPEED MEASUREMENT
RADAR SPEEDOMETER
SELECTION OF SAMPLE VEHICLES
DIFFERENT HOURS OF SURVEY
AVOID OVERTAKING SITUATION
EXPERIENCED TECHNICIAN
REQUIRED FOR SPEED FLOW CURVE
CONCLUSIONS
COLLECTED DATA ON TRAFFIC AND
VEHICLE IS COMPULSORY INPUT TO
HDM
RELIABILITY OF MANAGEMENT
PLAN DEPENDS UPON ACCURATE
DATA
AADT, GROWTH RATE, EASL
INFLUENCE VOC, INTERVESION
STANDARDS FOR UNSEALED
ROADS IN MONGOLIA
Aims and Objectives
z GRAVEL
z IMPROVED EARTH
z UNIMPROVED EARTH
MONGOLIA - STEPPE TRACK
PRIORITIES FOR RURAL ROADS
2M i P
2.Main Procedure
d iin St
Strategy
t A
Analysis
l i
3 Th A
3.The Application
li ti off St
Strategy
t A
Analysis
l i
Annual
7.0 Budget
50%
6.0
ghness
80%
5.0
R oug
100%
4.0
Target
g
3.0
1998 1999 2000 2001 2002 2003 2004
Summary
W have
We h just
j covered:
d
• Flexible Pavement
• Composite
p Pavement
• Rigid Pavement
TYPICAL X-
X-SECTION OF ROAD PROJECT
13.300 M
C
L
4% 2% 2% 4%
AC SURFACING
AC BASE
AGGREGATE BASE
SUB BASE
What is Road Management
15%
50%
Maintenance
- Fuel Consumption
- Oil Consumption
- Tyres Wear & Tear
- Parts Replacement 10% 25%
- Vehicle Depreciation
- Travel Time
- Accidents
Key Statistics from Emerging Countries
• Transport sector is 5 to 10 per cent GDP
• Paved roads account for 12% of length and 90% non non-
urban traffic
• In some countries,
countries actual expenditure is 25% of need
• Every $ not spent on maintenance increases transport
costs
t by
b 4$
• Surfacing lives vary from less than 5 to 10 years
• Transport fuels and machinery account for 40% of
p bill
import
Weaknesses in Road Management
28%
Very Poor
(IRI>7.0)
47%
8%
Fair (3.5<IRI<5.0)
17%
7365 km
two-lane
equivalent Poor
network
t k (5 0<IRI<7 0)
(5.0<IRI<7.0)
Network Condition
RIDE QUALITY
NETWORK
Fair (8.0%)
Good (28.0%)
Poor (17.0%)
RSL > 5
(13%)
RSL 3 - 5
651
(5%) 229
Kms
K
Kms
2706
1483 Kms
Kms RSL <=
< 0
(53%)
RSL 0 - 3
(29%)
RIDE QUALITY & ROAD USER COST
30
25
20
15
10
0
1000 2000 3000 4000 5000 6000 7000
35
30
on
25
5
Rs. Billio
20
15
10
0
Year 1991-92 1992-93 1993-94 1994-95 1995-96 1996-97 1997-98 1998-99 1999-
2000
Demand
Allocation Years
Releases/Utilization
Funding Status
Demand, Allocation & Release
(Maintenance Budget)
7000
6000
5000
Rs. Million
n
4000
3000
R
2000
1000
0
8
9
8
00
-8
-8
-9
-9
-9
-9
-9
-9
-9
-9
-9
-9
87
88
89
90
91
92
93
94
95
96
98
-2
97
99
19
19
19
19
19
19
19
19
19
19
19
19
19
Demand
Allocation
Received Year
Minimum Budgetary Requirement
SCENARIO TO MAINTAIN 1998 CONDITION; MIN. 2-3 BILLION RS PER YEAR *
100%
85
8.5
90%
80%
7.5
70%
ughness (IRI))
ondition (%)
60% 6.5
Poor
50% Fair
Average Rou
Network Co
5.5 G d
Good
Avg.. IRI
40%
30% 4.5
A
20%
3.5
10%
0% 2.5
1998 1999 2000 2001 2002 2003 2004
Year
ORK ANALYSIS - ECONOMIC LOSS OF NON MAINTENANCE
4 Total Economic Loss : Rs. 38 Billion
4 Economic Loss due to N-5 Rs. 26 Billion
N-25
N-5
What SOLUTIONS can we suggest ?
Establish a dedicated and secure Road Fund
Prioritise on a rational and equitable basis
Adopt affordable standards
Implement appropriate management and
quality systems
Address transport management, regulation and
enforcement issues
Seek to improve road user behaviour and
vehicle
hi l standards
t d d
Overview of Development
of Computer-Based HDM
M d l and
Models d
Development of
HDM-4
HDM 4 Software
Siriphan 1
Manual Calculation
? + - * / % $
$ NPV ?
!!! IRR ?
NPV/C
/C ?
Siriphan 2
EXCEL-- Spread Sheet
EXCEL
Siriphan 3
User Friendly -- HDM
HDM-4
4
Siriphan 4
Objectives
Siriphan 6
Development of HDM
Models -- on PC
z HDM-PC (1989): A Micro Computer
Version of HDM-III
z HDM-Q ((1994):
) Incorp.
p Traffic
Congestion Effect
z HDM Manager (1994): Providing a
Menu-Driven Front End
Siriphan 7
HDM 4
HDM-4
Highway Development
and Management
Tool
(since 1995)
Siriphan 8
Basic Capabilities of
HDM Models
TAUFIK WIDJOYONO
KUALA LUMPUR
What is the real problem ???
Road deterioration
Budget short coming
Investment priority
p y
Regional budget
allocation
Budget accountability
Introduction
The Role of HDM
HDM-4
4 as a tool of Decision Support System
Time value
Economic parameter to determine
road investment
Discounted
Di t dBBenefit
fit andd costt ratio
ti
NPV
NPV/Cost
EIRR
Prioritization regarding regional
development
Economic consideration
(generated traffic )
Social Consideration
(increasing accessibility)
Towards Better Policy
PAVEMENT TYPES
FLEXIBLE PAVEMENT
SEMI-RIGID PAVEMENT
RIGID PAVEMENT
PAVEMENT STRUCTURAL
CONDITION EVALUATION
TYPES OF EQUIPMENT
COMPONENTS OF THE BENKELMAN
BEAM
PRINCIPLE OF REBOUND DEFLECTION
METHOD OF STRUCTURAL
CONDITION EVALUATION
EQUIPMENT FOR DATA
COLLECTION
• LOADED TRUCK
Wheel Load – 8160 kg.
Tyre
T P
Pressure – 5.6
5 6 kg/sq.cm
k /
Benkelman beam with dial gauge
Thermometer
Glycerol
Pavement Cutting Tools
Rapid Moisture Meter
Tyre Pressure Gauge
SELECTION OF HOMOGENEOUS
PAVEMENT SECTIONS
PAVEMENT SURFACE CONDITION DATA
CRACK AREA
PATCH AREA
RUT AREA
IRC PROCEDURE FOR
DEFLECTION
C O MEASUREMENT
AS
Chainage Do Di Df D
41.400
41 420
41.420
41.440
MARKING OF PAVEMENT SECTION
SECONDARY DATA
SUBGRADE MOISTURE
PAVEMENT TEMPERATURE
PAVEMENT COMPONENT LAYERS
SUBGRADE SOIL TYPE
ROAD DETERIORATION MODELS.
z SESSION AIMS
To teach the key variables affecting road deterioration modeling
in HDM-4.
To make the participant aware of the key reference document
z SESSION OBJECTIVES
To outline the principles behind modeling of bituminous pavement
deterioration.
To identify the various factors which affects bituminous pavement
performance.
To explain the basis of various pavement defects and condition
indicators which are modelled in HDM-4, and to highlight the most
important explanatory variables.
INTRODUCTION
Temperature
Classification
oisture
assification Tropical
p Sub-tropical
p Sub-tropical
p Temperate
p Temperate
p
hot cool cool freeze
z Affected by:
Pavement materials and thickness
S b d
Subgrade
Drainage regime and seasonality
z Adj
AdjusteddS Structurall N
Number(SNP)
b (SNP) isi usedd
Similar to modified structural number of HDM-
III/AASHO
But contribution of pavement layers/subgrade
weighted by depth.
EFFECT OF DRAINAGE
• INTRODUCTION
• HDM - 4 AND ITS APPLICATION
• STRATEGIC ANALYSIS
– PROCEDURES
– OUTPUTS
• CONCLUSION
65,000
COMPETING
KILOMETRES
SECTORS
BUDGET
CONSTRAINTS??
PUBLIC DEMANDS!!!
ROAD
CONDITION
TRAFFIC
HIGH MEDIUM LOW
GOOD 76 283 307
Fair 15 13 4
Poor 31 22 11
South 75 40 24
East/Central 63 50 12
10
8
R O U G H N E SS
6 50%
80%
4 100%
2
0
1997 1998 1999 2000 2001 2002 2003 2004 2005
YEAR
CALIBRATION
•Vehicle Fleet
•Unit Costs
•Road Deterioration and Works Effects
•Road User Effects
Wo r l d R o a d
Association
The Highway Development
and Management Series
volumes one - five
H D M - 4
ENGLISH
volume five
1 2 3 4 5
A Guide to Calibration and Adaptation
THE HIGHWAY DEVELOPMENT AND MANAGEMENT SERIES
H D M - 4
ENGLISH
volume four
Analytical Framework and Model Descriptions
THE HIGHWAY DEVELOPMENT AND MANAGEMENT SERIES
H D M - 4
ENGLISH
volume three
Software User Guide
THE HIGHWAY DEVELOPMENT AND MANAGEMENT SERIES
H D M - 4
ENGLISH
volume two
Applications Guide
THE HIGHWAY DEVELOPMENT AND MANAGEMENT SERIES
PYCCKий
FRANÇAIS
H D M - 4
ESPAÑOL
volume one
ENGLISH
Overview of HDM-4
THE HIGHWAY DEVELOPMENT AND MANAGEMENT SERIES
volume one
Overview of HDM-4
Henry G. R. Kerali
Association
mondiale
de la Route
Wo r l d R o a d
Association
1
THE HIGHWAY DEVELOPMENT AND MANAGEMENT SERIES
About This Manual
This Version 1.0 edition of Overview of HDM-4 provides a short executive summary
describing the HDM-4 system. Where necessary it highlights the differences between the
HDM-III and the HDM-4 models.
It is intended to be used by all readers new to HDM-4, particularly high-level management
within a road organisation.
The Overview of HDM-4 is one of five manuals comprising the suite of HDM-4
documentation (see Figure 1).
Overview of HDM-4
Volume 1
Applications
Applications Guide
Guide Software
Software User
User Guide
Guide
Volume
Volume 22 Volume
Volume 33
Analytical
Analytical Framework
Framework andand
Model Descriptions
Model Descriptions
Volume
Volume 44
A
A Guide
Guide to
to Calibration
Calibration and
and Adaptation
Adaptation
Volume
Volume 55
A short executive summary describing the HDM-4 system. It is intended to be used by all
readers new to HDM-4, particularly high level management within a road organisation.
! Applications Guide (Volume 2)
Overview of HDM-4 i
Version 1.0
About This Manual
Describes the analytical framework and the technical relationships of objects within the
HDM-4 model. It contains very comprehensive reference material describing, in detail,
the characteristics of the modelling and strategy incorporated in HDM-4. It is to be used
by specialists or experts whose task is to carry out a detailed study for a road
management organisation.
! A Guide to Calibration and Adaptation (Volume 5)
Suggests methods for calibrating and adapting HDM models (as used in HDM-III and
HDM-4), to allow for local conditions existing in different countries. It discusses how to
calibrate HDM-4 through its various calibration factors. It is intended to be used by
experienced practitioners who wish to understand the detailed framework and models
built into the HDM-4 system.
Notes:
1 Volumes 1, 2 and 3 are designed for the general user.
2 Volumes 4 and 5 will be of greatest relevance to experts who wish to obtain low level
technical detail. However, Volume 5, in particular, presents very important concepts,
which will be of interest to all users.
Overview of HDM-4 ii
Version 1.0
About This Manual
Describes the three analysis tools used to cater for different types of highway studies,
namely:
❏ Strategy analysis
❏ Programme analysis
❏ Project analysis
! Section 6 - HDM-4 Modules
Describes the modular structure and the main functions of the various modules:
❏ Data managers
❏ Models
! Section 7 - Data Requirements
Describes the user interface and compares its improvement to previous models.
! Section 9 - Nomenclature
This provides a reference to the Nomenclature and Glossary of terms used in the HDM-4
suite of documents.
! Section 10 - References
ISOHDM Products
The products of the International Study of Highway Development and Management Tools
(ISOHDM) consist of the HDM-4 suite of software, associated example case study databases,
and the Highway Development and Management Series collection of guides and reference
manuals. This Volume is a member of that document collection.
Customer contact
Should you have any difficulties with the information provided in this suite of documentation
please do not hesitate to report details of the problem you are experiencing. You may send an
E-mail or an annotated copy of the manual page by fax to the number provided below.
The ISOHDM Technical Secretariat welcomes any comments or suggestions from users of
HDM-4. Comments on the Overview of HDM-4 should be sent to the following address:
E-mail: isohdm@bham.ac.uk
Telephone: +44 - 121 - 414 6717 (5049)
Fax: +44 - 121 - 414 3675 (5060)
Post: ISOHDM Technical Secretariat.
School of Civil Engineering
The University of Birmingham
Edgbaston
Birmingham B15 2TT
United Kingdom
Change details
This is the first formal edition (Version 1.0) of the HDM-4 documentation.
Related documentation
HDM-4 documents
The Highway Development and Management Series Collection is ISBN: 2-84060-058-7, and
comprises:
Volume 1 - Overview of HDM-4, ISBN: 2-84060-059-5
Volume 2 - Applications Guide, ISBN: 2-84060-060-9
Volume 3 - Software User Guide, ISBN: 2-84060-061-7
Volume 4 - Analytical Framework and Model Descriptions, ISBN: 2-84060-062-5
Volume 5 - A Guide to Calibration and Adaptation, ISBN: 2-84060-063-3
Overview of HDM-4 iv
Version 1.0
About This Manual
Future documentation
The following documents will be issued at a later release:
Volume 6 - Modelling Road Deterioration and Works Effects, ISBN: 2-84060-102-8
Volume 7 - Modelling Road User and Environmental Effects, ISBN: 2-84060-103-6
Terminology handbooks
PIARC Lexicon of Road and Traffic Engineering - First edition. Permanent International
Association of Road Congresses (PIARC), Paris 1991. ISBN: 2-84060-000-5
Technical Dictionary of Road Terms - Seventh edition, English - French. PIARC Commission
on Terminology, Paris 1997. ISBN: 2-84060-053-6
Acknowledgements
The development of HDM-4 has been sponsored by several agencies, primarily:
! Asian Development Bank (ADB)
! Department for International Development (DFID) in the United Kingdom
! Swedish National Road Administration (SNRA)
! The World Bank
Overview of HDM-4 v
Version 1.0
About This Manual
Many other organisations and individuals in a number of countries have also contributed in
terms of providing information, or undertaking technical review of products being produced.
The study has been co-ordinated by the ISOHDM Technical Secretariat at the University of
Birmingham in the United Kingdom. A number of organisations participated in the research
including:
! Finnra
Responsible for developing deterioration relationships for cold climates, road safety,
environmental effects, and supporting HRG with system design.
All research organisations received support from local and regional staff, visiting experts and
external advisers, to ensure that a high standard of quality and international consensus was
achieved. A number of other countries and individuals have supported this work through
supplying expert advice and reviewing the products.
Copyright statement
These HDM-4 products have been produced by the International Study of Highway
Development and Management Tools (ISOHDM), sponsored by The World Bank, the Asian
Development Bank, the Department for International Development (UK), the Swedish
National Road Administration, and other sponsors. The HDM-4 products are jointly published
by The World Road Association (PIARC), Paris and The World Bank, Washington, DC.
Copyright © 2000 The World Road Association (PIARC) on behalf of the ISOHDM sponsors.
All rights reserved.
This copyright covers all documents and document components, computer software and data
delivered as components of the HDM-4 product, in any physical or electronic forms.
Overview of HDM-4 vi
Version 1.0
Part A Overview
Contents
Overview of HDM-4
1 Introduction
The Highway Design and Maintenance Standards Model (HDM-III), developed by the World
Bank, has been used for over two decades to combine technical and economic appraisals of
road projects, to prepare road investment programmes and to analyse road network strategies.
The International Study of Highway Development and Management (ISOHDM) has been
carried out to extend the scope of the HDM-III model, and to provide a harmonised systems
approach to road management, with adaptable and user-friendly software tools. This has
produced the Highway Development and Management Tool (HDM-4).
The scope of HDM-4 has been broadened considerably beyond traditional project appraisals,
to provide a powerful system for the analysis of road management and investment
alternatives. Emphasis was placed on collating and applying existing knowledge, rather than
undertaking extensive new empirical studies, although some limited data collection was
undertaken. Wherever possible, creative new approaches were developed for applying up-to-
date knowledge to the technical problems and management needs of different countries.
Figure 1.1 shows a view of the HDM-4 documentation suite comprising a series of five
volumes. This Overview document is Volume 1 of the series. It contains a short executive
summary describing the HDM-4 system. All readers new to HDM-4, particularly high level
management within a road organisation, should use this document.
Volumes 2 and 3 are considered as guides for users of the HDM-4 software, where user tasks
are documented and an understanding of the product can soon be learnt. Volume 2 is a task-
oriented guide describing typical examples of different types of analyses. Volume 3 describes
the HDM-4 software and is a general-purpose document that provides an understanding of the
software user interface.
Volumes 4 and 5 contain more detailed reference material that is not vital to getting started
using HDM-4, but is designed to provide detailed guidance to the more advanced users.
Volume 4 describes the analytical framework and the technical relationships incorporated in
HDM-4. It is to be used by specialists or experts whose task is to carry out a detailed study for
a road management organisation. Volume 5 describes methods for adapting and calibrating
HDM-4 in different countries. It is intended to be used by experienced practitioners who wish
to understand the detailed framework and models incorporated into the HDM-4 system.
Overview of HDM-4 1
Version 1.0
OVERVIEW
Overview of HDM-4
Volume 1
Applications
Applications Guide
Guide Software
Software User
User Guide
Guide
Volume
Volume 22 Volume
Volume 33
Analytical
Analytical Framework
Framework andand
Model Descriptions
Model Descriptions
Volume
Volume 44
A
A Guide
Guide to
to Calibration
Calibration and
and Adaptation
Adaptation
Volume
Volume 5 5
Overview of HDM-4 2
Version 1.0
OVERVIEW
2 Background
The HCM model highlighted areas where more research was needed to provide a model that
was more appropriate to developing country environments with additional relationships
specific to that environment.
Following this, TRRL, in collaboration with the World Bank, undertook a major field study in
Kenya to investigate the deterioration of paved and unpaved roads as well as the factors
affecting vehicle-operating costs in a developing country. The results of this study were used
by TRRL to produce the first prototype version of the Road Transport Investment Model
(RTIM) for developing countries (Abaynayaka, 1977). In 1976, the World Bank funded
further developments of the HCM at MIT that produced the first version of the Highway
Design and Maintenance Standards model (HDM) (Harral, 1979).
Further work was undertaken in a number of countries to extend the geographic scope of the
RTIM and HDM models:
! The Caribbean Study (by TRRL)
Investigated the effects of road geometry on vehicle operating costs (Morosiuk and
Abaynayaka, 1982; Hide, 1982)
! India Study (by the Central Road Research Institute - CRRI)
Overview of HDM-4 3
Version 1.0
OVERVIEW
Further developments of both models continued with the TRRL producing RTIM3 in 1993 to
provide a user-friendly version of the software running as a spreadsheet (Cundill and
Withnall, 1995), and in 1994, the World Bank produced two further versions of HDM:
! HDM-Q
Incorporating the effects of traffic congestion into the HDM-III program (Hoban, 1987).
! HDM Manager
Overview of HDM-4 4
Version 1.0
OVERVIEW
Each of these functions is carried out as a sequence of activities known as the management
cycle (Robinson et al., 1998) described in Section 3.2.
Planning
Planning involves the analysis of the road system as a whole, typically requiring the
preparation of medium to long term, or strategic, estimates of expenditure for road
development and preservation under various budget and economic scenarios. Predictions may
be made of road network conditions under a variety of funding levels in terms of key
indicators together with forecasts of required expenditure under defined budget heads. The
physical highway system is usually characterised at the planning stage by:
! Characteristics of the road network
The results of the planning exercise are of most interest to senior policy makers in the roads
sector, both political and professional. A planning unit will often undertake this work.
Programming
Programming involves the preparation, under budget constraints, of multi-year road work and
expenditure programmes in which sections of the network likely to require maintenance,
improvement or new construction, are selected and analysed. It is a tactical planning exercise.
Ideally, cost-benefit analysis should be undertaken to determine the economic feasibility of
each set of works. The physical road network is considered at the programming stage on a
link-by-link basis, with each link characterised by homogeneous pavement sections defined in
terms of physical attributes. The programming activity produces estimates of expenditure in
each year, under defined budget heads, for different types of roadwork and for each road
section. Budgets are typically constrained, and a key aspect of programming is to prioritise the
road works in order to find the best use of the constrained budget. Typical applications are the
Overview of HDM-4 5
Version 1.0
OVERVIEW
preparation of a budget for an annual or a rolling multi-year work programme for a road
network, or sub-network. Managerial-level professionals within a road organisation normally
undertake programming activities, perhaps within a planning or a maintenance department.
Preparation
This is the short-term planning stage where road schemes are packaged for implementation.
At this stage, designs are refined and prepared in more detail; bills of quantities and detailed
costing are made, together with work instructions and contracts. Detailed specifications and
costing are likely to be drawn up, and detailed cost-benefit analysis may be carried out to
confirm the feasibility of the final scheme. Works on adjacent road sections may be combined
into packages of a size that is cost-effective for execution. Typical preparation activities are
the detailed design of:
! An overlay scheme
! Road improvement works
Operations
These activities cover the on-going operation of an organisation. Decisions about the
management of operations are made typically on a daily or weekly basis, including the
scheduling of work to be carried out, monitoring in terms of labour, equipment and materials,
the recording of work completed, and use of this information for monitoring and control.
Activities are normally focused on individual sections or sub-sections of a road, with
measurements often being made at a relatively detailed level. Operations are normally
managed by sub-professional staff, including works supervisors, technicians, clerks of works,
and others.
As the management process moves from Planning through to operations, it will be seen that
changes occur to the data required. The data detail starts as a coarse summary but progressively
moves towards a detailed level (see Table 3.1).
Overview of HDM-4 6
Version 1.0
OVERVIEW
sections
Overview of HDM-4 7
Version 1.0
OVERVIEW
Policies
Classification
Performance
and
monitoring
standards
Highway Management
Information:
•Inventory
•Condition
•Resources
•Treatments
•Productivity
•Unit Costs
•Economic parameters
Finance and
Resources
Overview of HDM-4 8
Version 1.0
OVERVIEW
Table 3.2
Management functions and the corresponding HDM-4 applications
Budgeting system
Overview of HDM-4 9
Version 1.0
OVERVIEW
4 Analytical Framework
The HDM-4 analytical framework is based on the concept of pavement life cycle analysis.
This is applied to predict the following over the life cycle of a road pavement, which is
typically 15 to 40 years:
! Road deterioration
! Road work effects
! Road user effects
! Socio - Economic and Environmental effects
Overview of HDM-4 10
Version 1.0
OVERVIEW
Poor
Maintenance Standard
Roughness
Pavement
Performance
Curve
Rehabilitation
Good
The impacts of the road condition, as well the road design standards, on road users are
measured in terms of road user costs, and other social and environmental effects. Road user
costs comprise:
! Vehicle operation costs (fuel, tyres, oil, spare parts consumption; vehicle depreciation
and utilisation, etc.),
! Costs of travel time - for both passengers and cargo, and
! Costs to the economy of road accidents (that is, loss of life, injury to road users,
damage to vehicles and other roadside objects).
The social and environmental effects comprise vehicle emissions, energy consumption, traffic
noise and other welfare benefits to the population served by the roads. Although the social and
environmental effects are often difficult to quantify in monetary terms, they can be
incorporated within the HDM-4 economic analyses if quantified exogenously.
It should be noted that in HDM-4, road user effects can be calculated for both motorised
transport (motorcycles, cars, buses, trucks, etc.) and non-motorised transport (bicycles, human
powered tricycles, animal pulled carts, etc.).
Figure 4.2 illustrates the impact of road condition (represented in terms of the IRI) on the
costs of different modes of transport.
Overview of HDM-4 11
Version 1.0
OVERVIEW
Bus
Road User Costs ($/veh-km)
Heavy Truck
Pickup/utility
Car
Rickshaw
Figure 4.2 Effect of Road Condition on Vehicle Operating Costs for Rolling
Terrain
Road User Costs in HDM-4 are calculated by predicting physical quantities of resource
consumption and then multiplying these quantities by the corresponding user specified unit
costs. It is necessary to ensure that the vehicle resource quantities predicted are in keeping
with the range of values observed in the area of application. For further details, refer to A
Guide to Calibration and Adaptation - Volume 5.
Economic benefits from road investments are then determined by comparing the total cost
streams for various road works and construction alternatives against a base case (without
project or do minimum) alternative, usually representing the minimum standard of routine
maintenance. HDM-4 is designed to make comparative cost estimates and economic analyses
of different investment options. It estimates the costs for a large number of alternatives year-
by-year for a user-defined analysis period. All future costs are discounted to the specified base
year. In order to make these comparisons, detailed specifications of investment programmes,
design standards, and maintenance alternatives are needed, together with unit costs, projected
traffic volumes, and environmental conditions.
Overview of HDM-4 12
Version 1.0
OVERVIEW
5 HDM-4 Applications
For example, a road network matrix could be modelled using; three traffic categories (high,
medium, low), two pavement types (asphalt concrete, surface treatments), and three pavement
condition levels (good, fair, poor). In this case, it is assumed that the environment throughout
the study area is similar and that the road administration is responsible for one road class (for
example, main roads). The resulting road network matrix for this would therefore comprise (3
x 2 x 3 =) 18 representative pavement sections. There is no limit to the number of
representative pavement sections that can be used in a strategy analysis. The trade-off is
usually between a simple representative road network matrix that would give rather coarse
results compared against a detailed road network matrix with several representative sections
that could potentially provide more accurate results.
Strategy analysis may be used to analyse a chosen network as a whole, to prepare medium to
long range planning estimates of expenditure needs for road development and conservation
under different budget scenarios. Estimates are produced of expenditure requirements for
medium to long term periods of usually 5-40 years. Typical applications of strategy analysis
by road administrations would include:
! Medium to long term forecasts of funding requirements for specified target road
maintenance standards (see Figure 5.1a).
! Forecasts of long term road network performance under varying levels of funding (see
Figure 5.1b).
Overview of HDM-4 13
Version 1.0
OVERVIEW
! Optimal allocation of funds according to defined budget heads; for example routine
maintenance, periodic maintenance and development (capital) budgets (see
Figure 5.1c).
! Optimal allocations of funds to sub-networks; for example by functional road class
(main, feeder and urban roads, etc.) or by administrative region (see Figure 5.1d).
! Policy studies such as impact of changes to the axle load limit, pavement maintenance
standards, energy balance analysis, provision of NMT facilities, sustainable road network
size, evaluation of pavement design standards, etc.
A typical sample application of the HDM-4 Strategy Analysis is given in
Appendix A.
Overview of HDM-4 14
Version 1.0
OVERVIEW
$10m
6.0
$15m
5.0
4.0 $20m
Target
3.0
2000 2001 2002 2003 2004 2005 2006
Budget Allocations
7.0 Feeder
Average Roughness (IRI)
Roads
$30m/yr
6.0
Secondary
Roads
5.0 $35m/yr
4.0 Primary
Roads
$20m/yr
3.0
2000 2001 2002 2003 2004 2005 2006
Overview of HDM-4 15
Version 1.0
OVERVIEW
250
200
D e v e lo p m e n t
U S $ m /y e a r
150
Im p ro v e m e n t
P e r io d i c
100
R o u tin e
50
0
2000 2001 2002 2003
200
150
U S $ m /y e a r
P rim a ry
100 S e c o n d a ry
F eed er
50
0
2000 2001 2002 2003
Overview of HDM-4 16
Version 1.0
OVERVIEW
The HDM-4 programme analysis application may be used to prepare a multi-year rolling
programme, subject to resource constraints (see Figure 5.2a and Figure 5.2b). The
prioritisation method employs the incremental NPV/cost ratio as the ranking index, described
in more detail in Analytical Framework and Model Descriptions - Volume 4. This provides an
efficient and robust index for prioritisation purposes. Indices such as the NPV, economic rate
of return (ERR), or predicted pavement condition attributes (for example, road roughness) are
not recommended as ranking criteria. The incremental NPV/cost ratio satisfies the objective of
maximising economic benefits for each additional unit of expenditure (that is, maximise net
benefits for each additional $1 of the available budget invested).
A typical sample application of the HDM-4 programme analysis application is
given in Appendix B.
Overview of HDM-4 17
Version 1.0
OVERVIEW
Overview of HDM-4 18
Version 1.0
OVERVIEW
Project analysis may be used to estimate the economic or engineering viability of road
investment projects by considering the following issues:
! The structural performance of road pavements
! Life-cycle predictions of road deterioration, road works effects and costs
! Road user costs and benefits
! Economic comparisons of project alternatives
Typical appraisal projects would include the maintenance and rehabilitation of existing roads,
widening or geometric improvement schemes, pavement upgrading and new construction.
There are no fundamental changes to the philosophy of the system in this area, but improved
road deterioration relationships have been extended to cover a wider range of pavements and
the performance of materials in temperate and cold climates. Road user cost relationships
include impacts on road safety.
Typical examples of project analysis applications using HDM-4 are given in
Appendix C.
In terms of data requirements, the key difference between the strategy and programme
analyses, with that for project analysis, is in the detail at which data is defined. Use is made of
the concept of information quality levels (IQL) recommended by the World Bank (Paterson
and Scullion, 1990). Project level analysis data is specified in terms of measured defects
(IQL-II), whereas the specification for strategy and programme analyses can be more generic
(IQL-III). For example; for project level analysis, road roughness would be specified in terms
of the IRI value (m/km); but for strategy and programme analyses, roughness could be
specified as good, fair or poor. The relationship between IQL-II and IQL-III level data is
user-defined in the HDM Configuration depending on road class, pavement surface type and
traffic class.
Overview of HDM-4 19
Version 1.0
OVERVIEW
6 HDM-4 Modules
The overall structure of HDM-4 is illustrated in Figure 6.1. The three analysis tools (Strategy,
Programme and Project) operate on data defined in one of four data managers:
! Road Network
Defines the characteristics of the vehicle fleet that operate on the road network to be
analysed.
! Road Works
Defines maintenance and improvement standards, together with their unit costs, which
will be applied to the different road sections to be analysed.
! HDM Configuration
Defines the default data to be used in the applications. A set of default data is provided
when HDM-4 is first installed, but users should modify these to reflect local
environments and circumstances.
Technical analysis within the HDM-4 is undertaken using four sets of models:
! RD (Road Deterioration)
Simulates the effects of road works on pavement condition and determines the
corresponding costs.
! RUE (Road User Effects)
Overview of HDM-4 20
Version 1.0
OVERVIEW
! Technical models
File Core
CoreData
Data Model libraries
FileConverter
Converter Vehicle Models
VehicleFleets
Fleets can be used in
Road
RoadNetworks
Networks other systems
transfer Road
RoadWorks
transferdata
data Works
Projects
totoexternal
external Projects
systems Programmes
Programmes RDWE RUE SEE
systems Strategies
Strategies
External RDWE
RDWE:Road
:RoadDeteriorationand
DeteriorationandWorks
WorksEffects
Effects
Systems RUE : Road User Effects
RUE : Road User Effects
Databases, SEE
SEE: :Safety,
Safety,Energy
Energyand
andEnvironmental
EnvironmentalEffects
Effects
PMS, etc.
PMS
PMS: :Pavement
PavementManagement
Manageme System
ntSystem
Overview of HDM-4 21
Version 1.0
OVERVIEW
7 Data requirements
7.1 Overview
The HDM-4 applications have been designed to work with a wide range of data type and
quality. For example, pavement condition data collected by visual inspection according to
condition classes (for example, Very good, good, fair, poor condition) can be converted to the
HDM-4 model requirements prior to running any of the applications (see section below on
Importing and Exporting Data). Similarly, HDM-4 can work with very detailed measurements
of pavement condition if the data is available. This flexibility in data requirements should
permit all potential users with a variety of data to integrate HDM-4 into their road
management functions. For further details, refer to Applications Guide - Volume 2.
Comprise one or more sections over which traffic is reasonably constant. This is provided
for purposes of compatibility of the network referencing system with existing pavement
management systems.
! Nodes
Intersections which connect links or other points at which there is a significant change in
traffic, carriageway characteristics, or administrative boundaries.
All network data is entered using the Road Network folder, and facilities are also available for
editing, deleting and maintaining this data. The approach to network referencing is
considerably more flexible than that used in HDM-III, and is designed to handle a wide range
of external referencing conventions as might be used by other systems with which HDM-4
may need to interface.
Overview of HDM-4 22
Version 1.0
OVERVIEW
Overview of HDM-4 23
Version 1.0
OVERVIEW
8 User interface
A key objective for the development of HDM-4 is to provide a system that is user-friendlier
than the original HDM-III. This has been achieved by addressing the user interface design and
data requirements. The user interface has been improved by developing the system to run
under a standard Microsoft Windows environment. Many computer users are already familiar
with the Windows system and this should make learning HDM-4 more intuitive. A modular
system design has been adopted to enable different modules to be used relatively easily and to
facilitate future system operation on other non-Windows platforms. More details of system
issues are described in the Software User Guide - Volume 3.
Considerable attention has also been paid to the data that must be entered by users,
particularly because the total data requirement is greater than that in HDM-III in order to
allow for the extended facilities included in HDM-4. The concept of data hierarchy is used,
whereby default data is defined for many items in HDM Configuration, and users can choose
the extent to which defaults are used. The system can be used with a level of data entry detail
that is appropriate for particular applications. A comprehensive Help system is also provided.
Figure 8.1 illustrates the HDM-4 Workspace. Detailed instructions for the operation of the
individual modules are given in the Software User Guide - Volume 3.
Overview of HDM-4 24
Version 1.0
OVERVIEW
Overview of HDM-4 25
Version 1.0
OVERVIEW
9 Nomenclature
A nomenclature explaining any special numbering, markings, etc., that are associated with the
HDM-4 suite of documents is included in Part H of the Analytical Framework and Model
Descriptions - Volume 4.
A glossary of terms is included in Part I of the Analytical Framework and Model Descriptions
- Volume 4 describing the use of phrases found in the HDM-4 documentation.
Overview of HDM-4 26
Version 1.0
OVERVIEW
10 References
Overview of HDM-4 27
Version 1.0
OVERVIEW
Watanatada T., Harral C.G., Paterson W.D.O., Dhareshwar A.M., Bhandari A., and
Tsunokawa K., (1987)
The highway design and maintenance standards model volume 1: description of the
HDM-III model. The Highway Design and Maintenance Standards Series. Baltimore:
Johns Hopkins for the World Bank.
Overview of HDM-4 28
Version 1.0
STRATEGY ANALYSIS APPLICATION
A.1 Background
The national road network in a country comprises 4,267 km of paved roads and 3,145 km of
unpaved (gravel) roads. A recent pavement condition survey has shown that 22.3% of the
paved road network is in good condition, 36.2% in fair condition and 41.5% in poor condition.
Traffic surveys were also carried out in the previous year. Following an analysis of the data
available together with the characteristics of the national road network, it has been decided to
categorise the paved road network into three traffic classes (high, medium, low), and three
pavement condition classes (good, fair, poor), and the unpaved road network has been
categorised into two traffic categories only (medium and low). The resulting road network
matrix is summarised in Table A.1:
Table A.1
Road Network Length in each Matrix Category
Traffic categories
Paved roads: High Medium Low
(AADT > 4000) (4000>AADT>1000) (AADT < 1,000)
Good condition 234 km 306 km 410 km
A.2 Standards
The national road authority has a defined set of standards for road maintenance and road
network improvement in accordance with its own policies and those set by the government.
! The maintenance and improvement standards for paved roads include
❏ Widening all paved roads with volume to capacity ratio greater than 0.8.
❏ Rehabilitation (structural overlay) of all paved roads in poor condition.
❏ Resealing (surface dressing) paved roads when surface deterioration exceeds 30%.
❏ Reactive routine maintenance comprising patching potholes immediately, sealing
cracks, edge repairs, etc., as required.
❏ Routine maintenance to shoulders, drainage ditches, road markings and all roadside
furniture.
Overview of HDM-4 29
Version 1.0
STRATEGY ANALYSIS APPLICATION
❏ Upgrading to paved surface, all gravel roads carrying more than 250 vehicles per
day.
❏ Regravelling when the remaining gravel thickness falls below 50 mm.
❏ Grading gravel roads with medium traffic twice a year, and grading once a year for
gravel roads carrying less than 100 vehicles per day.
❏ Routine maintenance to shoulders, drainage ditches and all roadside furniture.
Overview of HDM-4 30
Version 1.0
STRATEGY ANALYSIS APPLICATION
12.0 25.0
Average Roughness (IRI)
4.0
10.0
2.0 5.0
0.0
0.0
2000 2005 2010 2015 2020
2000 2005 2010 2015 2020
200
3,000
2,500 150
(millions)
(millions)
2,000 100
1,500 50
1,000
0
2000 2005 2010 2015 2020
2000 2005 2010 2015 2020
Overview of HDM-4 31
Version 1.0
PROGRAMME ANALYSIS APPLICATION
B.1 Background
The national road authority has drawn up a long list of candidate road sections for periodic
maintenance and improvement over the next three years in Western Province. The long list of
candidate road sections follows a review of pavement condition surveys carried out by
consultants. The national road authority has a policy to prioritise the candidate projects and
select those that will be included in the periodic maintenance programme for the three-year
budget period. Given that the candidate projects are from the main road network only, the
objective is to prioritise according to the economic benefits that would be derived from each
candidate road project.
B.2 Standards
The standards for periodic maintenance and road improvement defined by the national road
authority require the following road works to be carried out:
! Road improvement standards
❏ Pave gravel roads with AADT greater than 150 vehicles per day.
❏ Widen roads with peak volume to capacity ratio greater than 0.85.
! Periodic maintenance standards
Overview of HDM-4 32
Version 1.0
PROGRAMME ANALYSIS APPLICATION
Overview of HDM-4 33
Version 1.0
PROGRAMME ANALYSIS APPLICATION
2000 MSE 203 km 80 - 90 10.0 1152 Overlay 20mm 2.51 11.03 34.18 0.899
2000 MSC 138 km 5 - 10 & 65 - 70 10.0 1271 Single Seal 2.41 3.80 37.98 0.999
2000 MAN 446 km 53.7 - 57.0 3.3 983 Overlay 20mm 2.40 3.31 41.29 1.086
2000 MSE 932 km 36.2 - 52.2 16.0 809 Overlay 20mm 2.19 17.64 58.93 1.551
2000 MSE 334 km 3.8 - 12.8 9.0 932 Overlay 20mm 1.98 9.92 68.85 1.812
2000 MAN 203 km 185 - 190 5.0 1248 Single Seal 1.72 1.90 70.75 1.862
2000 MAN 243 km 0 - 21 21.0 1009 Overlay 20mm 1.55 23.15 93.90 2.471
2000 MTS 549 km 0 - 1.0 1.0 2438 Overlay 20mm 1.37 1.10 95.00 2.500
2000 MSC 142 km 12.5 - 20 7.5 942 Single Seal 0.98 3.04 98.04 2.580
2000 MSW 131 km 10 - 16.5 6.5 428 Overlay 20mm 0.92 7.72 105.76 2.783
2000 MSW 905 km 128 - 170 42.0 717 Single Seal 0.92 15.96 121.73 3.203
2000 MSV 449 km 45 - 59.6 14.6 671 Single Seal 0.46 5.70 127.43 3.353
2000 MTN 748 km 30 - 60 30.0 589 Single Seal 0.21 11.40 138.83 3.653
…Continued
Overview of HDM-4 34
Version 1.0
PROGRAMME ANALYSIS APPLICATION
2000 MTN 830 km 8.6 - 20 11.4 680 Pavement Reconstruction 0.10 51.63 194.87 5.128
2000 MTN 805 km 145 - 152.6 7.6 844 Pavement Reconstruction 0.05 37.80 232.67 6.123
2001 MSW 901 km 180 - 330 60.0 1623 Overlay 20mm 3.16 66.15 298.82 7.864
2001 MSV 537 km 43 - 45 2.0 937 Overlay 20mm 1.42 2.21 301.03 7.922
2001 MSW 935 km 16 - 30 14.0 532 Overlay 20mm 0.33 15.44 316.46 8.328
2002 MSC 131 km 16.5 - 26 9.5 2253 Overlay 20mm 4.91 11.03 327.49 8.618
2002 MSC 141 km 28 - 49.5 21.5 2253 Overlay 20mm 4.88 24.26 351.74 9.256
2002 MSW 956 km 10 - 75 30.0 1733 Overlay 20mm 3.31 33.08 384.82 10.127
2002 MSC 155 km 50 - 55 5.0 1353 Overlay 20mm 2.61 5.51 390.33 10.272
2002 MSW 938 km 3.2 - 10 6.8 913 Overlay 20mm 1.30 7.72 398.05 10.475
2002 MSW 150 km 0.0 - 16.5 16.5 800 Overlay 20mm 0.99 18.74 416.79 10.968
2002 MAN 214 km 90 - 97.7 7.7 759 Overlay 20mm 0.86 8.82 425.61 11.200
2002 MAN 409 km 110 - 115 5.0 512 Overlay 20mm 0.28 5.51 431.12 11.345
This report was generated by Version 1.0 of HDM-4. Neither this report nor the application may be distributed without prior consent.
Overview of HDM-4 35
Version 1.0
PROGRAMME ANALYSIS APPLICATION
2000 MSE 203 km 80 - 90 10.0 1152 Overlay 20mm 2.51 11.03 34.18 0.899
2000 MSC 138 km 5 - 10 & 65 - 70 10.0 1271 Single Seal 2.41 3.80 37.98 0.999
2000 MAN 446 km 53.7 - 57.0 3.3 983 Overlay 20mm 2.40 3.31 41.29 1.086
2000 MSE 932 km 36.2 - 52.2 16.0 809 Overlay 20mm 2.19 17.64 58.93 1.551
2000 MSE 334 km 3.8 - 12.8 9.0 932 Overlay 20mm 1.98 9.92 68.85 1.812
2000 MAN 203 km 185 - 190 5.0 1248 Single Seal 1.72 1.90 70.75 1.862
2000 MAN 243 km 0 - 21 21.0 1009 Overlay 20mm 1.55 23.15 93.90 2.471
2000 MTS 549 km 0 - 1.0 1.0 2438 Overlay 20mm 1.37 1.10 95.00 2.500
2000 MSC 142 km 12.5 - 20 7.5 942 Single Seal 0.98 3.04 98.04 2.580
2000 MSW 131 km 10 - 16.5 6.5 428 Overlay 20mm 0.92 7.72 105.76 2.783
2000 MSW 905 km 128 - 170 42.0 717 Single Seal 0.92 15.96 121.73 3.203
2001 MSW 901 km 180 - 330 60.0 1623 Overlay 20mm 3.16 66.15 187.88 4.944
2001 MSV 537 km 43 - 45 2.0 937 Overlay 20mm 1.42 2.21 190.08 5.002
… Continued
Overview of HDM-4 36
Version 1.0
PROGRAMME ANALYSIS APPLICATION
2002 MSC 141 km 28 - 49.5 21.5 2253 Overlay 20mm 4.88 24.26 225.36 5.931
2002 MSW 956 km 10 - 75 30.0 1733 Overlay 20mm 3.31 33.08 258.44 6.801
2002 MSC 155 km 50 - 55 5.0 1353 Overlay 20mm 2.61 5.51 263.95 6.946
2002 MSW 938 km 3.2 - 10 6.8 913 Overlay 20mm 1.30 7.72 271.67 7.149
2002 MSW 150 km 0.0 - 16.5 16.5 800 Overlay 20mm 0.99 18.74 290.41 7.642
2002 MAN 214 km 90 - 97.7 7.7 759 Overlay 20mm 0.86 8.82 299.23 7.874
This report was generated by Version 1.0 of HDM-4. Neither this report nor the application may be distributed without prior consent.
Overview of HDM-4 37
Version 1.0
PROJECT ANALYSIS APPLICATION
C.1 Introduction
The project analysis application in HDM-4 can be used to carry out economic appraisal for a
wide range of project types. These include:
! Periodic maintenance and rehabilitation
❏ Overlays
❏ Resealing
❏ Pavement reconstruction
! Upgrading
❏ Dual carriageways
❏ Lane addition
❏ Carriageway widening
! Non-motorised traffic facilities
❏ Bicycle lanes
❏ Carriageway separation
Note that the above list is not exhaustive.
Overview of HDM-4 38
Version 1.0
PROJECT ANALYSIS APPLICATION
! Specify the maintenance and road improvements to be carried out together with the unit
costs.
! Run the HDM-4 Project Analysis application to determine the economic benefits.
! Review reports of the analyses conducted.
C.3 Output
There are several detailed summary reports produced by HDM-4. These include:
! Pavement deterioration and road works
❏ Summary of road user costs (vehicle operation, travel time and accidents)
❏ Traffic flow details
❏ Average travel speeds
❏ Traffic volume to capacity ratios
! Environmental effects
❏ Vehicle emissions
❏ Energy consumption
! Economic analysis results
Overview of HDM-4 39
Version 1.0
PROJECT ANALYSIS APPLICATION
financing a project and the financial profitability of the project. The economic feasibility of
the project is assessed by comparison against a base-line project alternative (that is, a without
project alternative). The project alternatives are:
! Without Project
Maintain the existing gravel road before upgrading to a bituminous pavement, followed
by maintenance of the bituminous pavement.
Overview of HDM-4 40
Version 1.0
PROJECT ANALYSIS APPLICATION
Alternative Description
3 3m +6.69
4 2 lanes +12.18
Overview of HDM-4 41
Version 1.0
PROJECT ANALYSIS APPLICATION
B C D
E
Key: Existing road sections
Proposed bypass (Section E)
Alternative Description
1 This alternative represents the base case situation without the bypass.
2 Construct Section E: A two-lane AMGB (Asphalt Mix on Granular Base) road, with
a two-year construction period (2000-2001), and opening in year 2002.
Overview of HDM-4 42
Version 1.0
PROJECT ANALYSIS APPLICATION
B 4,000 4,000
C 6,000 1,000
D 8,000 3,000
E n/a 5,000
C.6.4 Results
The Economic Analysis Summary indicates that construction of the bypass (Section E) would
be viable in economic terms. The most cost-effective alternative would be a 2-lane bypass as
summarised in the table below.
Overview of HDM-4 43
Version 1.0
A4 back covers 10/12/99 10:25 am Page 1
volume one
A short executive summary describing the HDM-4 system. It is intended to be used by all readers new to HDM-4,
particularly high-level management within a road organisation.
ISBN: 2-84060-059-5
PIARC, Registered in France These HDM-4 products have been produced by the International Study of Highway
Development and Management Tools (ISOHDM), sponsored by The World Bank, the
Registered Office Asian Development Bank, the Department for International Development (UK), the
The World Road Association (PIARC) Swedish National Road Administration, and other sponsors. HDM-4 is jointly published
La Grande Arche by The World Road Association (PIARC), Paris, and The World Bank, Washington, D.C.
Paroi Nord, Niveau 8
92055 La Defénse Cedex PIARC endeavours to ensure that the information in this document is correct and fairly
France stated but does not accept liability for any error or omission.
Tel: +33 1 41 02 05 84 The development of PIARC products and services is continuous and published
Fax: +33 1 49 00 02 02 information may not be up to date. It is important to check the current position with
Email: piarc.hdm4@attglobal.net PIARC. This document is not part of a contract or licence save insofar as may be
Web: http://hdm4.piarc.org expressly agreed.
Wo r l d R o a d
Association
2
THE HIGHWAY DEVELOPMENT AND MANAGEMENT SERIES
Part A
Overview
Overview of
of HDM-4
HDM-4
Volume
Volume 11
Applications
Applications Guide
Guide Software
Software User
User Guide
Guide
Volume
Volume 22 Volume
Volume 33
Analytical
Analytical Framework
Framework andand
Model Descriptions
Model Descriptions
Volume
Volume 44
A
A Guide
Guide to
to Calibration
Calibration and
and Adaptation
Adaptation
Volume
Volume 55
A short executive summary describing the HDM-4 system. It is intended to be used by all
readers new to HDM-4, particularly high level management within a road organisation.
! Applications Guide (Volume 2)
Applications Guide i
Version 1.0
About This Manual
Describes the analytical framework and the technical relationships used within the HDM-
4 model. It contains very comprehensive reference material describing, in detail, the
characteristics of the modelling and strategy incorporated in HDM-4. It is to be used by
specialists or experts whose task is to carry out a detailed study for a road management
organisation.
! A Guide to Calibration and Adaptation (Volume 5)
Suggests methods for calibrating and adapting HDM models (as used in HDM-III and
HDM-4), to allow for local conditions existing in different countries. It discusses how to
calibrate HDM-4 through its various calibration factors. It is intended to be used by
experienced practitioners who wish to understand the detailed framework and models
built into the HDM-4 system.
Notes:
1 Volumes 1, 2 and 3 are designed for the general user.
2 Volumes 4 and 5 will be of greatest relevance to experts who wish to obtain low level
technical detail. However, Volume 5, in particular, presents very important concepts,
which will be of interest to all users.
Applications Guide ii
Version 1.0
About This Manual
Applications
ApplicationsGuide
Guide
Introduction
Introduction
Part
PartAA
Concepts
Concepts
ofofAnalysis
Analysis
Part
PartBB
Project
Project Programme
Programme Strategy
Strategy
Analysis
Analysis Analysis
Analysis Analysis
Analysis
Chapter
ChapterB1B1 Chapter
ChapterB2
B2 Chapter
ChapterB3
B3
Research,
Research,
Policy
Policyand
and
Regulation
Regulation
Analyses
Analyses
Part
PartCC
Data
DataManagement
Management
Part
PartDD
Case
CaseStudies
Studies
Appendices
Appendices
Describes the overall procedure for project analysis in HDM-4. This allows users to
assess the physical, functional and economic feasibility of specified project
alternatives by comparison against a base case (without-project) alternative.
❏ Programme Analysis
Describes the concept of Programme Analysis. This deals with the preparation of a
multi-year rolling programme for a road network in which candidate investment
options are identified and selected, subject to resource constraints. Road networks
are analysed section by section and estimates are produced of road works and
expenditure requirements for each road section for each year of the funding period.
❏ Strategy Analysis
Describes the procedure for Strategy Analysis. This is concerned with the analysis of
a chosen network as a whole, for preparing long range planning estimates of
expenditure needs for road development and conservation under different budget
scenarios. The road network is characterised by lengths of road in different
categories defined by parameters such as road class, surface type, and pavement
condition or traffic flow. Estimates are produced of expenditure requirements for
medium to long term periods of 5-40 years.
! Part C - Research, Policy and Regulation Analyses
Describes typical applications of HDM-4 for Research, Policy and Regulation analyses.
! Part D - Data Management
Case studies for Project, Programme and Strategy analyses are supplied as appendices.
The Applications Guide will be updated with new case studies as future versions of
HDM-4 are released.
Applications Guide iv
Version 1.0
About This Manual
ISOHDM Products
The products of the International Study of Highway Development and Management Tools
(ISOHDM) consist of the HDM-4 suite of software, associated example case study databases,
and the Highway Development and Management Series collection of guides and reference
manuals. This Volume is a member of that document collection.
Customer contact
Should you have any difficulties with the information provided in this suite of documentation
please do not hesitate to report details of the problem you are experiencing. You may send an
E-mail or an annotated copy of the manual page by fax to the number provided below.
The ISOHDM Technical Secretariat welcomes any comments or suggestions from users of
HDM-4. Comments on the Applications Guide should be sent to the following address:
E-mail: isohdm@bham.ac.uk
Telephone: +44 - 121 - 414 6717 (5049)
Fax: +44 - 121 - 414 3675 (5060)
Post: ISOHDM Technical Secretariat.
School of Civil Engineering
The University of Birmingham
Edgbaston
Birmingham B15 2TT
United Kingdom
Change details
This is the first formal edition (Version 1.0) of the HDM-4 documentation.
Related documentation
HDM-4 documents
The Highway Development and Management Series Collection is ISBN: 2-84060-058-7, and
comprises:
Volume 1 - Overview of HDM-4, ISBN: 2-84060-059-5
Volume 2 - Applications Guide, ISBN: 2-84060-060-9
Volume 3 - Software User Guide, ISBN: 2-84060-061-7
Volume 4 - Analytical Framework and Model Descriptions, ISBN: 2-84060-062-5
Volume 5 - A Guide to Calibration and Adaptation, ISBN: 2-84060-063-3
Applications Guide v
Version 1.0
About This Manual
Future documentation
The following documents will be issued at a later release:
Volume 6 - Modelling Road Deterioration and Works Effects, ISBN: 2-84060-102-8
Volume 7 - Modelling Road User and Environmental Effects, ISBN: 2-84060-103-6
Terminology handbooks
PIARC Lexicon of Road and Traffic Engineering - First edition. Permanent International
Association of Road Congresses (PIARC), Paris 1991. ISBN: 2-84060-000-5
Technical Dictionary of Road Terms - Seventh edition, English - French. PIARC Commission
on Terminology, Paris 1997. ISBN: 2-84060-053-6
Acknowledgements
The development of HDM-4 has been sponsored by several agencies, primarily:
! Asian Development Bank (ADB)
! Department for International Development (DFID) in the United Kingdom
! Swedish National Road Administration (SNRA)
! The World Bank
Applications Guide vi
Version 1.0
About This Manual
Many other organisations and individuals in a number of countries have also contributed in
terms of providing information, or undertaking technical review of products being produced.
The study has been co-ordinated by the ISOHDM Technical Secretariat at the University of
Birmingham in the United Kingdom. A number of organisations participated in the research
including:
! Finnra
Responsible for developing deterioration relationships for cold climates, road safety,
environmental effects, and supporting HRG with system design.
All research organisations received support from local and regional staff, visiting experts and
external advisers, to ensure that a high standard of quality and international consensus was
achieved. A number of other countries and individuals have supported this work through
supplying expert advice and reviewing the products.
Copyright statement
These HDM-4 products have been produced by the International Study of Highway
Development and Management Tools (ISOHDM), sponsored by The World Bank, the Asian
Development Bank, the Department for International Development (UK), the Swedish
National Road Administration, and other sponsors. The HDM-4 products are jointly published
by The World Road Association (PIARC), Paris and The World Bank, Washington, DC.
Copyright © 2000 The World Road Association (PIARC) on behalf of the ISOHDM sponsors.
All rights reserved.
This copyright covers all documents and document components, computer software and data
delivered as components of the HDM-4 product, in any physical or electronic forms.
Contents
Part C RPRA
C1 Concepts of Research, Policy and Regulation Analyses C1-1
volume two
A short executive summary describing the HDM-4 system. It is intended to be used by all readers new to HDM-4,
particularly high-level management within a road organisation.
ISBN: 2-84060-059-5
PIARC, Registered in France These HDM-4 products have been produced by the International Study of Highway
Development and Management Tools (ISOHDM), sponsored by The World Bank, the
Registered Office Asian Development Bank, the Department for International Development (UK), the
The World Road Association (PIARC) Swedish National Road Administration, and other sponsors. HDM-4 is jointly published
La Grande Arche by The World Road Association (PIARC), Paris, and The World Bank, Washington, D.C.
Paroi Nord, Niveau 8
92055 La Defénse Cedex PIARC endeavours to ensure that the information in this document is correct and fairly
France stated but does not accept liability for any error or omission.
Tel: +33 1 41 02 05 84 The development of PIARC products and services is continuous and published
Fax: +33 1 49 00 02 02 information may not be up to date. It is important to check the current position with
Email: piarc.hdm4@attglobal.net PIARC. This document is not part of a contract or licence save insofar as may be
Web: http://hdm4.piarc.org expressly agreed.
Contents
A1 Introduction
1 Introduction to HDM-4 A1-1
1.1 Scope of HDM-4 A1-1
1.2 Background A1-1
1.3 Objectives of model development A1-3
1.4 Overview of HDM-4 A1-3
1.5 Input data A1-7
1.6 Applications A1-8
1.7 User interface A1-10
2 References A1-11
Applications Guide i
Version 1.0
PART A INTRODUCTION
Part A Introduction
A1 Introduction
1 Introduction to HDM-4
1.2 Background
Both the RTIM and HDM models were tested and used extensively in the field. Further
fieldwork was undertaken, as listed below, with the general aim of extending the geographic
spread of the results:
! Caribbean - by TRRL
Investigation of the effects of road geometry on vehicle operating costs (Morosiuk and
Abaynayaka, 1982; Hide, 1982).
! India - by the Central Road Research Institute
TRRL's experience of model use suggested that there was a need for a simpler model
than those, which already existed. The requirement was to obtain a model that was easier
to use by consultants working in the field and by users in developing countries. This
resulted in the development of RTIM2 (Parsley and Robinson, 1982). The World Bank,
on the other hand, saw the need for a more comprehensive model, and this led to the
development of HDM-III (Watanatada, 1987).
Both of these models were originally designed to operate on mainframe computers and,
as computer technology advanced, the University of Birmingham (Kerali, 1985)
produced a microcomputer version of RTIM2 for TRRL. Later, the World Bank
(Archondo-Callao and Purohit, 1989) produced a microcomputer version of HDM-III.
Further development of both models has since taken place. RTIM3 was produced in
1993 to provide an extremely user-friendly version of the software running as a
spreadsheet (Cundill and Withnall, 1995). In 1994, the World Bank produced two further
developments:
❏ HDM-Q
Produced to incorporate the effects of traffic congestion into the HDM program
(Hoban, 1987).
❏ HDM Manager
operating costs are now considerably less than those predicted by the models. Therefore the
relationships were updated to reflect the present state-of-the-art.
Whilst most applications of the various models have been in developing countries, in recent
years many industrialised countries have begun to make use of the model. This has resulted
in the need for additional facilities to be included, such as the need to deal effectively with
congestion, with temperate and cold climates, with a wider range of pavement types and
structures, and with environmental issues.
It is against this background that the development work has been undertaken.
For the economic appraisal of road maintenance, rehabilitation, upgrading and new
construction through life cycle analysis of proposed road investments (see Part B).
! Works programming
For the preparation of multi-year and rolling programmes for road network maintenance
and development to facilitate the preparation of medium term budgets for road network
maintenance and extension (see Part B).
! Strategic planning
For policy development, long term resource allocation plans and road network planning
(see Part B).
! Software environment
A user-friendly system, built around a set of modules with the capacity to cope with a
wide range of data requirements and user skill levels.
In order to meet wider system objectives, technical relationships were updated and calibrated
to best current knowledge; additional technical capabilities have been provided to deal with
traffic congestion, non-motorised vehicles, concrete pavements, drainage, environmental and
safety effects.
File Core
CoreData
Data Model libraries
FileConverter
Converter Vehicle Models
VehicleFleets
Fleets can be used in
Road
RoadNetworks
Networks other systems
transfer Road
RoadWorks
transferdata
data Works
Projects
totoexternal
external Projects
systems Programmes
Programmes RDWE RUE SEE
systems Strategies
Strategies
External RDWE
RDWE:Road
:RoadDeteriorationand
DeteriorationandWorks
WorksEffects
Effects
Systems RUE : Road User Effects
RUE : Road User Effects
Databases, SEE
SEE: :Safety,
Safety,Energy
Energyand
andEnvironmental
EnvironmentalEffects
Effects
PMS, etc.
PMS : Pavement Management
PMS : Pavement Management SSystem
ystem
Contains data defining the road network or sub-network that will be the basis of the
analysis.
! Vehicle Fleet
Contains the characteristics of the vehicle fleet that will be operating on the road network
being analysed.
! Road Works
Contains the specifications for maintenance and improvement standards that will be
applied to different road sections to be analysed.
! HDM Configuration
Contains all of the default data to be used with the data analysis; a set of default data will
be provided with the system, but users will be able to modify it to reflect local
circumstances.
period under a user-specified scenario of circumstances. The primary cost set for the life
cycle analysis includes the costs of capital investment, maintenance and vehicle operation, to
which travel time costs can be added as an option. The costs of accidents, and environmental
pollution may also be included in the analysis.
The broad concept of the life cycle analysis is illustrated in Figure A1.2. Interacting sets of
costs, related to those incurred by the road administration and those incurred by the road
user, are added together over time in discounted present values. Costs are determined by
first predicting physical quantities of resource consumption and then multiplying these
quantities by their unit costs or prices. Economic benefits are then determined by comparing
the total cost streams for various maintenance and construction alternatives with a base case
(do nothing or do minimum alternative), usually representing minimal routine maintenance.
The following models are contained within HDM-4:
! Road Deterioration (RD)
Determines effects of emissions and noise, and predicts numbers of road accidents and
amounts of energy consumption.
More comprehensive details of the HDM-4 models are provided in the Analytical
Framework and Model Descriptions.
HDM-4 is designed to make comparative cost estimates and economic analyses of different
investment options. It estimates the costs for a large number of alternatives year-by-year for
a user-defined analysis period, discounting the future costs. Rates of return, net present
values, and first year benefits can also be determined by HDM-4. In order to make these
comparisons, detailed specifications of investment programmes, design standards, and
maintenance alternatives are needed, together with unit costs, projected traffic volumes, and
environmental conditions.
The model simulates, for each road section, year-by-year, the road condition and resources
used for maintenance under each alternative, as well as the vehicle speeds and physical
resources consumed by vehicle operation. Physical quantities involved in construction,
maintenance and vehicle operation are estimated, and user-specified prices and unit costs
applied to determine financial and economic costs. Relative benefits are then calculated for
different alternatives, followed by present value and rate of return computations.
Details of the benefits and costs considered in HDM-4 are included in Part G of the
Analytical Framework and Model Descriptions.
Road geometry and surface Social and Levels of emissions and energy
texture, vehicle characteristics Environmental Effects used, and number of accidents
1.4.3 Outputs
The three main types of outputs produced by HDM-4 are as follows (see Section 1.6 for
details):
! Economic efficiency indicators – from analyses of individual road projects
The method used to represent vehicle fleets is considerably more flexible than that used in
HDM-III. Motorcycles and non-motorised vehicles can be included, and there is no limit on
the numbers or types of vehicles that can be specified. Multiple vehicle fleet data sets can be
set up for use in different analyses, and a wide range of default data are provided.
1.6 Applications
HDM-4 can assist with:
! Project analysis for detailed economic appraisal (see Section 1.6.1).
! Programme analysis for annual or rolling work program preparation (see Section 1.6.2).
! Strategic analysis for long term planning (see Section 1.6.2).
! Research and policy studies (see Section 1.6.3).
This is essentially the use for which the earlier generations of HDM systems were developed.
Project analysis is used to estimate the economic or engineering viability of road investment
projects by considering the following issues:
! Pavement structural performance.
! Life cycle predictions of road deterioration, maintenance effects and costs.
! Road user costs and benefits.
! Economic comparisons of project alternatives.
Typical appraisal projects would include the maintenance and rehabilitation of existing roads,
widening or geometric improvement schemes, pavement upgrading and new construction.
Case studies demonstrating the application of Project analysis are presented in Part B.
The philosophy of this analysis has not fundamentally changed from the HDM-III version, but
improved road deterioration relationships have been extended to cover a wider range of
pavements and the performance of materials in temperate and cold climates. Road user cost
relationships have been updated and extended to include the impacts of road accidents.
Environmental impacts assessment and energy balance analysis is also included.
The main difference between strategy and programme analysis is the way in which sections
are physically identified:
! Programme analysis
Deals with sections that are unique physical units identifiable from the road network
throughout the analysis.
! Strategy analysis
An important difference between strategic and programme analysis, and that for project
analysis, is in the detail at which data is defined. Use is made of the concept of Information
Quality Levels (IQL) recommended by the World Bank (Paterson and Scullion, 1990).
Project level analysis data is specified in terms of measured defects (IQL-II), whereas the
specification of data for strategic and programme analyses can be more generic (IQL-III).
For example:
This has been improved by developing the system from the outset to run on Personal
Computers operating under the Microsoft Windows environment. System design is
modular to enable different technical modules to be used relatively easily and to facilitate
future system operation on other non-Windows platforms, such as UNIX, Linux, etc.
More details of system issues are described in the Software User Guide.
! Data requirements
The total data input to HDM-4 is greater than in HDM-III to allow for the extended
facilities that are available. The concept of a data hierarchy is used, whereby defaults
are set for many data items in the HDM Configuration folder. Users may choose the
extent to which defaults are used, and run the system using a level of data entry detail
that is appropriate to their particular application.
A comprehensive Help system is being designed and will be available in a later release of the
HDM-4 system.
2 References
Abaynayaka, S.W., Morosiuk G., and Hide H., (1977)
Prediction of road construction and vehicle operating costs in developing countries.
Proceedings of the Institution of Civil Engineers. 62 (Part 1), 419-446.
Archondo-Callao, R. and Purohit R., (1989)
HDM-PC: user's guide. The Highway Design and Maintenance Standards Series.
Washington DC: The World Bank.
Archondo-Callao, R., (1994)
HDM Manager Version 3.0. Transportation Division, Transportation, Water & Urban
Development Department. Washington DC: The World Bank.
CRRI, (1982)
Road user cost study in India. Final Report, New Delhi: Central Road Research Institute.
Cundill, M.A., and Withnall S.J., (1995)
Road transport investment model RTIM3. In: TRB. Sixth International Conference on
Low-Volume Roads, Conference Proceedings 6, Volume 1. Washington DC: National
Academy Press, 187-192.
FHWA, (1994)
Highway economics requirements system. Washington DC: Federal Highway
Administration.
Geipot, (1982)
Research on the interrelationships between costs of highway construction, maintenance
and utilisation (PICR). Final report, 12 volumes, Brasilia, Brazil.
Harral, C.G., and others, (1979)
The highway design and maintenance standards model (HDM): model structure,
empirical foundations and applications. PTRC Summer Annual Meeting, University of
Warwick, 13-16 July 1979. London: PTRC Education and Research Services.
Hide, H., (1982)
Vehicle operating costs in the Caribbean: results of a survey of vehicle operators. TRRL
Laboratory Report 1031. Crowthorne: Transport and Road Research Laboratory.
Hoban, C.J., (1987)
Evaluating traffic capacity and improvements to geometry. Technical Paper Number 74.
Washington DC: The World Bank.
Kerali, H.R., Parsley L.L., Robinson R., and Snaith M.S., (1985)
Development of a microcomputer based model for road investment in developing
countries. In: CIVILCOMP. Proceedings of the Second International Conference on
Civil and Structural Engineering Computing, London, December 1985. London:
Institution of Civil Engineers.
Watanatada T., Harral C.G., Paterson W.D.O., Dhareshwar A.M., Bhandari A., and
Tsunokawa K., (1987)
The highway design and maintenance standards model volume 1: description of the
HDM-III model. The Highway Design and Maintenance Standards Series. Baltimore:
John Hopkins for the World Bank.
Contents
Applications Guide i
Version 1.0
PART C RESEARCH, POLICY AND REGULATION ANALYSES CONTENTS
Applications Guide ii
Version 1.0
PART B CONCEPTS OF ANALYSIS
1 Introduction
This chapter describes the overall procedure for project analysis in HDM-4. This allows users
to assess the physical, functional and economic feasibility of specified project alternatives by
comparison against a base case, or a without-project alternative. The key processes of the
analysis are the:
! Life cycle costing of pavement structural performance.
! Prediction of road deterioration.
! Estimation of road user costs (vehicle operating costs, travel time and accidents).
! Modelling of road works effects and the costs of these to the road administration.
! Calculation of economic or financial benefits from comparisons of the project
alternatives.
The aim is to determine which project alternative is most cost-effective.
Four case studies to demonstrate its application are described in Appendices A1, A2, A3, and
A4.
Project analysis is associated with the following types of road projects:
! Maintenance of existing roads
These works cover a wide range of maintenance techniques for different pavement types.
This type of works also includes those that arise when a road pavement has received
insufficient maintenance over its design life, or because the pavement was not built to the
standards of quality required by the original design (sometimes known as rehabilitation).
! Improvement of existing roads
These projects aim to provide additional capacity when a road is nearing the end of its
design life or because there has been an unforeseen change in use of the road. Works
include measures to improve the quality of service, such as relieving traffic congestion,
road safety, road passibility, the need to accommodate increased vehicle axle loads, etc.
Typical examples of road improvement projects are:
❏ Pavement reconstruction
❏ Pavement upgrading
The paving of gravel roads, the provision of concrete pavements in places where the
subgrade soils may be very poor etc.
❏ Road widening and geometric improvements
Includes widening of existing carriageways, the provision of additional lanes, dual-
carriageways, overtaking lanes, climbing lanes, and the betterment of shoulders.
❏ Realignment and pavement upgrading of the existing roadway
Combines the activities described in the above two items (Pavement upgrading and
road widening).
! New construction
Assesses the performance of a completed project to see if objectives set out during
appraisal have been met. Project evaluation requires project data that have been collected
and recorded in a systematic way throughout all stages of the project cycle. The
evaluation should result in specific recommendations about improving aspects of the
project design that can be used to improve ongoing and future planning.
Project
Project
Diverted
Divertedtraffic
traffic Project
Project Project
Project Project
Project etc
etc
alternative
alternative11 alternative
alternative22 alternative
alternative33
Generated traffic
Exogenous Benefits
Section
Section11 Section
Section22 Section
Section33 etc
etc
and Costs alternative
alternative22 alternative
alternative22 alternative
alternative22
Maintenance
Maintenance Maintenance
Maintenance Improvement
Improvement etc
etc
Standard
StandardAA Standard
StandardBB Standard
StandardYY
[Operation
[Operation1]1]++[Intervention
[Interventioncriteria]
criteria]++[Cost]
[Cost]==“Works
“Works1”
1”
[Operation
[Operation1]1]++[Intervention
[Interventioncriteria]
criteria]++[Cost]
[Cost]==“Works
“Works2”
2”
etc
etc
Name of the vehicle fleet that includes the vehicle types that use the road network.
The components of the project are then specified in detail.
! Conversion rate
The conversion factors between the output and vehicle fleet currency units, and that
between output and road works currency units used in the analysis.
2.3 Network
❏ Surface class
All types of surfacing, or one of: bituminous, concrete or unsealed surfacings.
❏ Structural adequacy
Pavements of all strength, or those of: poor, fair or good strength.
! Speed-flow type
All speed-flow types or one of the speed-flow types pre-defined in the Road Network
folder.
! Traffic volume
Road sections carrying any traffic volume, or those carrying one of: low, medium or
high traffic.
! Geometry
The selection criteria can be used to add new sections to those already selected, or to replace
those selected. At any time during the selection process, a list of those sections currently
selected can be displayed.
Specified for each vehicle type, for each specified growth period.
A traffic growth period can be added (or amended) by specifying the following:
! Start year of the traffic growth period
! Growth type for normal traffic
Apply standards to the section alternative, giving for each the date from which the
standard is effective.
! Generated traffic
These are benefits and costs that are not calculated within HDM-4. These are specified
using the following data:
❏ Description
❏ Starting year relative to the beginning of the road works
❏ Amounts of the costs or benefits in economic and/or financial terms
❏ Annual growth rate in terms of percentage increase, incremental amounts, or actual
amounts
The specified alternative can be edited to alter the description of the alternative, or to add or
change the standards that apply.
Measured in terms of new AADT, traffic composition and growth rates on all the
sections affected (see Part B of the Analytical Framework and Model Descriptions).
! Analysis by section
Analyses, individually, each of the road sections that make up the project. Several
alternatives can be defined for each section as shown in Table B1.1 (for example, three
alternatives for section A, four alternatives for section B, etc.), with one alternative
designated as the base case against which all the other alternatives will be compared.
Economic indicators (for example, NPV, IRR, BCR and FYB) are calculated for each
section alternative.
Table B1.1
Analysis by sections
Note: RM = Routine maintenance, and includes: patching, crack sealing, and drainage
works.
! Analysis by project
Table B1.2
Analysis by projects
Note: RM = Routine maintenance, and includes: patching, crack sealing, and drainage
works.
Describes the base case situation; that is, without a new road section.
! Project alternative 2
Includes a new section, E, a 2-lane AMSB (asphalt mix on stabilised base) road.
! Project alternative 3
Includes new section, E, a 4-lane STGB (surface treatment on granular base) road.
Table B1.3
Analysis involving new sections
Section B RM RM RM RM
Section C RM RM RM RM
Section D RM RM RM RM
Note: RM = Routine maintenance, and includes: patching, crack sealing, and drainage
works.
This project can be analysed to determine whether or not the construction of the new section is
cost-effective and what design/construction specifications to follow.
A new section can only be scheduled and not triggered by responsive intervention criteria.
A new section can be added by specifying the following:
! General
! Construction costs
Costs data consisting of total costs, duration of works, salvage value, and percentage of
total expenditure over each year of the works duration.
! Road works standards
Maintenance and/or improvement standards to apply to the section alternative, giving for
each the date from which the standard is effective.
! Normal traffic
! Section details
Includes any generated traffic resulting from the works option, in terms of the following:
❏ Description
❏ Starting year relative to the beginning of the improvement works
❏ Traffic volumes by vehicle type for the remainder of the analysis period
Further details on specifying generated traffic for a section alternative are discussed in
Part B of the Analytical Framework and Model Descriptions.
! Exogenous benefits and costs
Benefits and costs associated with the works option that are significant; they are specified
using the following data:
❏ Description
❏ Starting year relative to the beginning of the improvement works
❏ Monetary value of the cost or benefit in financial and economic terms
Standards can also be added or deleted.
The base case against which all other project alternatives are compared. Note that
for section analysis, the first section alternative is automatically assigned as the base
case.
❏ Discount rate (%)
Including whether or not costs are to be included and, if so, the unit costs for accidents in
terms of fatal, injury only, damage only, or all accidents.
! Energy balance analysis
! Vehicle emissions
! Acceleration effects
2.10.1 Introduction
Standard outputs that can be generated from HDM-4 are described. The outputs have been
categorised as follows:
! Tabulated input data
! Results of analyses
Input data
Printout of tabulated input data, with appropriate headings, is useful for checking, reviewing
and updating the information that will be used in a study. All key input data, which affect the
results of analyses significantly, should be printed to accompany the reports. Careful checking
of all input data, preferably verified on hard copy, is strongly advised.
Analysis results
The results of analyses to be performed within HDM-4 have been grouped as follows:
! Deterioration/works effects
! Road user effects
! Environmental effects
! Energy use
! Cost streams
! Others
! User-definable reports
Deterioration/works effects
This group includes the following types of reports:
! Timing of works
! Pavement condition versus time
! Pavement strength
Includes works quantities, average condition and traffic data (by alternative).
❏ Free speeds
❏ Average operating speeds
❏ Operating speeds by traffic flow periods
! Traffic flows
❏ Fuel consumption
❏ Lubricating oil consumption
❏ Tyre consumption
❏ Parts consumption
❏ Maintenance labour hours
❏ Crew hours
❏ Depreciation
❏ Interest
! Time resource consumption
❏ Delay hours
❏ Additional vehicle operating costs
! Road safety
! Summary reports
Environmental effects
The reports on environmental effects include the following:
! Vehicle emissions
❏ Particulates (Par)
❏ Lead (Pb)
! Net change in vehicle emissions
Energy use
The reports on energy use are available as follows:
! National energy use and global energy use
Cost streams
The reports of cost streams are divided into the following subgroups:
! Road agency costs
These costs will be reported under user-defined budget categories for each of the
following cost categories:
❏ Economic
❏ Financial
! Road user costs
User-definable reports
Users may specify their own desired types of reports, which are different from those presented
above. The user-definable reports are generated using proprietary report writer software
(Crystal Reports). For example users may wish to adopt the following criteria:
! Aggregation of analysis results
1 Introduction
1.1 Applications
Programme Analysis deals primarily with the prioritisation of a defined long list of candidate
road projects into a one-year or multi-year work programme under defined budget constraints.
It is essential to note that here, we are dealing with a long list of candidate road projects
selected as discrete segments of a road network. The selection criteria will normally depend
on the maintenance, improvement or development standards that a road administration may
have defined (for example from the output produced by the strategy analysis application).
Examples of selection criteria that may be used to identify candidate projects include:
! Periodic maintenance thresholds (for example, reseal pavement surface at 20% damage).
! Improvement thresholds (for example, widen roads when volume/capacity ratio exceeds
0.8).
! Development standards (for example, upgrade gravel roads to sealed pavements when the
annual average daily traffic exceeds 200 vehicles per day).
The above examples do not imply firm recommendations to be used by road authorities.
When all candidate projects have been identified, the HDM-4 programme analysis application
can be used to compare the life cycle costs predicted under the existing regimen of pavement
management (that is, the without project case) against the life cycle costs predicted for the
periodic maintenance, road improvement or development alternative (that is, with project
case). This provides the basis for estimating the economic benefits that would be derived by
including each candidate project within the budget timeframe.
It should be noted that the main difference between strategy analysis and programme analysis is
the way in which road links and sections are physically identified. Programme analysis deals
with individual links and sections that are unique physical units identifiable from the road
network throughout the analysis. In strategy analysis, the road system essentially loses its
individual link and section characteristics by grouping all road segments with similar
characteristics into the road network matrix categories.
For programme analysis, the problem can be posed as one of seeking that combination of road
work alternatives across a number of sections in the network that optimises an objective
function under budget constraint. If, for example, the objective function is to maximise the
Net Present Value (NPV), the problem can be defined as:
Select that combination of road work options for sections that maximises NPV
for the whole network subject to the sum of the treatment costs being less
than the budget available.
The HDM-4 programme analysis application may be used to prepare a multi-year rolling
programme, subject to resource constraints. The prioritisation method employs the
incremental NPV/cost ratio as the ranking index, described in more detail in Analytical
Framework and Model Descriptions. This provides an efficient and robust index for
prioritisation purposes. Indices such as the NPV, economic rate of return (ERR), or predicted
pavement condition attributes (for example, road roughness) are not recommended as ranking
criteria. The incremental NPV/cost ratio satisfies the objective of maximising economic
benefits for each additional unit of expenditure (that is, maximise net benefits for each
additional $1 of the available budget invested).
Road geometry and surface texture, Social and Levels of emissions and energy
vehicle characteristics Environmental Effects used, and number of accidents
! Postpone road works until the first year after the budget period.
This simplification is normal practice in work programming and it can be used to keep the
analysis simple and reduce run times because there is usually a large number of network
elements to be analysed. In this method, the work programme is also prepared separately for
each year in the budget period, and the results are summed up for the budget period.
Economic calculations are done by comparing investments made within the budget period
against deferring the action to the first year after the budget period as illustrated in Figure
B2.2. For example, an overlay applied in the first year of the budget period year would be
compared against the road work required after the budget period. The road works are not
necessarily the same, because the road section could be in worse condition after postponement
of the overlay, and this might no longer be applicable on the road section.
IRI
2 years
Deferred investment
Investment now
Years
The discounted costs and benefits of road works are evaluated only one year beyond the
budget period. Thus, there are two issues to be considered under this approach:
! Determination of the residual value of the investment
! Estimation of future benefits accruing due to the investment
A simplifying assumption is made that the deferred action will be less effective in improving
pavement condition as shown in Figure B2.2. It is also assumed that the pavement
performance after the next year will be similar for both alternatives. Therefore, there is no
need to calculate pavement deterioration and road user costs beyond the first year after the
budget period. This approach is a modification of that used in the United States Highway
Economics Requirements System (HERS) and in the UK Highway Agencies PMS (HAPMS).
The following two examples in Table B2.1 and Table B2.2 illustrate the above process.
Table B2.1
Example of one-year forward programme
S2 OL50 REC
S3 OL50 OL50
In Table B2.1, a one-year work programme for three sections is analysed. An overlay is
triggered for all sections, however, the budget would be exceeded. Two alternatives are
automatically analysed for each road section in order to calculate economic indicators for the
optimisation:
Alternative 1 Alternative 2
S1 Overlay at year 1 vs. Do nothing at year 1 and overlay at year 2
The second example in Table B2.2 illustrates a multi-year work programme in which an
overlay is triggered in the second year of the Budget Period. Technically the overlay could be
postponed beyond the budget period, but after that reconstruction would be needed.
The base case in this example is Alternative 2, and the economic benefits are calculated by
comparing Alternative 1 for each of the sections against alternative 2. The results are saved
for the optimisation process described in Section 4.
Table B2.2
Example of multi-year forward programme
Alternative 1 Alternative 2
(Road works applied within (Road works required
Sections the budget period) after the budget period)
S2 Overlay Reconstruction
S3 Resealing Overlay
S4 Overlay Reconstruction
To analyse the above problem, HDM-4 will automatically analyse the equivalent of two
project alternatives for each of the road sections. The with project alternative for each road
section is as follows:
1 Section 1
Resealing in the first year, then overlay in the third year of the Budget Period.
2 Section 2
! Input and output currencies, and conversion rate between the given currencies
! Discount rate
In the first case, the entire network is selected. In the second case, sections can be selected
through user-defined selection criteria. The parameters available for the selection criteria are
given below:
! Surface class
! Structural adequacy
! Road type
! Road geometry
! Traffic
! Capacity
The following are examples of the types of selections that might be made:
! All unpaved sections having AADT > 100
! All paved sections with width > 5.0 metres
! All paved sections having AADT > 5000
! Sub-network - North
The group option may be used to combine several sections into a group, which is treated as a
composite unit for analysis. A typical example is a group of consecutive road sections that
form a road link. This kind of analysis might also be used, for example, when a mobile
asphalt or a crushing plant is situated in a different location each year, and the sections near to
this are all to be treated at the same time. For analysis purposes, the results for individual
sections are combined to give a weighted average, which is then compared with other sections
(or groups).
The sub-network option may be used to combine several sections that could be treated as a
composite unit in the reporting of results. Sub-networks differ from links in such a way that
while links are analysed together, road sections in a sub-network are analysed separately and
only summarised for reporting.
Irrespective of the method used, a data set is created as a result of the selection process of the
form shown in Table B2.3.
The user is also able to add or remove one or more sections, but sections in a link or group
cannot be selected individually.
This process is similar to the selection of sections for Project Analysis. Only definitions of
links and sub-networks have been added.
Table B2.3
Example of selected sections
Scope: [Maintenance]
Section Group or link (optional)
S1 P1
S2 P2
S3 P2
S4 P2
S5 L1
S6 L1
S7 L1
If option (3) above is selected, either option (1) or (2) must follow it, unless it is the final year
of the analysis period.
This process is similar to the specification of traffic characteristics for Project Analysis.
upgrading a gravel road). For each of these, the intervention criteria, unit costs, applicable
limits and the effects of the works item must be specified.
This process is generally similar to the specification of Road Works Standards for Project
Analysis.
3 Work programming
The Work Programming process is divided into six phases:
1 Selection of maintenance or improvement standards to be used for different
sections (see Section 3.1)
Table B2.4
Example of assignment of standards
S3 U Ideal (U) - -
S5 B Ideal (B) - -
S6 B Ideal (B) - -
Table B2.5
Example of a tentative work programme
4 Optimisation
If the needed budget for each budget period is below the given budget constraints, no further
economic analysis is necessarily needed. However, if the budget needs are higher than the
available budgets, the user has two choices:
1 Revise maintenance and improvement standards and run the automatic assignment again
2 Execute an economic analysis and optimisation
Programme analysis provides two methods for budget optimisation, as follows:
1 Total enumeration (see Section 4.1)
2 Incremental NPV/cost ranking (see Section 4.2)
Further methods may be added in later versions of HDM-4.
If the number of sections to be analysed is less than 100, and there are no more than five
budget periods and 16 alternatives per section, total enumeration can be used. This will be
externally done in the EBM-HS model of HDM-III. If the above constraints are exceeded,
incremental benefit/cost ranking will used.
Either: maximise NPV or maximise the improvement in roughness; for example, NPV
! Constraints on resources for each budget period
where:
k a road section
m a investment alternative on a section
K the number of sections in the network analysed
km subscript denoting alternative m for section k
Mk the number of alternatives for section k
Xkm the zero-one decision variable,
Xkm = 1 if alternative m of investment unit k is chosen,
Xkm = 0 otherwise
m = 1,...,Mk
and
k = 1,...,K
OBJkm the objective function to be maximised which may be the discounted net
present value economic benefits, or the average reduction in roughness due to
the investment alternative
∑∑ R
K =1 m =1
kmqt X km ≤ TR qt , q = 1,...., Q; t = 1,...., T ...(4.2)
where:
mk
∑X
m =1
km ≤ 1, k = 1,...., K ...(4.3)
that is, for each section k, no more than one alternative can be implemented.
If M is the average number of alternatives for the sections, the problem then has KM (= K x
M) zero-one variables, QT (= Q x T) resource constraints and K interdependency constraints.
The parameters that define the problem size are K, M and QT. Depending on the solution
method used, different problem-size parameters determine whether the method is suitable for
the problem in terms of the computational effort needed.
The total enumeration method provides the user with an unconditionally optimal solution. It
computes the total net present values of all feasible programme selections, and chooses the
one with the highest value. The computational effort required for this may be considerable, so
the method is only feasible when the number of alternatives per investment unit is relatively
small.
Total enumeration is done externally in the EBM-HS software of HDM-III. The procedure is
as follows:
1 An input file for the EBM-HS is created in the Programme Analysis
(The format of this file is defined in the HDM-III EBM documentation)
2 The user performs the following:
(a) starts EBM-HS
(b) imports the file to EBM-HS
(c) runs EBM-HS
(d) exports the results to an output file
3 The output file is imported to the Programme Analysis for reporting
(NPV j - NPVi )
E ji =
(cost j - cost i ) ...(4.4)
where:
NPVk the net present value of investment alternative k
costk the investment cost of alternative k
i the index of the cheaper alternative
j the index of the more expensive alternative
The aim is to select sections successively starting with the largest NPV/cost ratio (Eji). Where
there is more than one investment option on any individual road section, it is assumed that the
user will have specified the alternative with the lowest discounted economic costs as the do
minimum alternative. It is important to note that if the user selects a do minimum alternative
which does not have the lowest discounted total economic cost, then all other options with
lower discounted total economic cost will be ignored.
The incremental method considers all options with higher discounted total economic cost, and
compares these incrementally against the ‘do minimum’ option. The procedure selects the
maximum NPV/Cost ratio for options that would be within budget limit. Table B2.6 shows an
example of the calculation of the incremental NPV/cost ratio.
Table B2.6
Example of Cost-effective Options List
Note: * - A section with two alternative standards with high incremental benefits
An incremental search technique is used to select the options starting with the highest
incremental NPV/cost ratios, ensuring that at any time there is no more than one option
selected per road section. The process continues until the budget is exhausted for each budget
period. The method is often referred to as the efficiency frontier, which is a line that joins
investments with the highest NPV along the cost axis in a plot of NPV against investment
cost. In essence, the method seeks out those options that are close to the boundary of the
frontier. The algorithm is illustrated in Figure B2.4, and is defined in the following steps:
1 Determine the pre-defined investment options for pre-selected sections and deduct the
financial costs of these options from the available budget in corresponding years.
Exclude these sections from any further optimisation.
2 Determine possible investment options for the remaining sections. If the life cycle
analysis option is being used, set the user-defined base alternatives as the do minimum
for each road section. For the multi-year forward programme, the do minimum option is
that with the delayed capital works.
3 If the total financial cost of the do minimum investment alternatives on each section is
greater than the available budget for any period, then the investment options or budget
constraints must be redefined.
4 Deduct the financial cost of the do minimum investments from the available budget to
determine the remaining budget for each period. Set the do minimum as the first Base
option for each section.
5 Calculate the incremental NPV/Cost ratio for all remaining section-options compared
against the Base option, and all other option pairs with higher economic cost. For
example, consider the following investment options for a particular section arranged in
the ascending order of discounted total economic costs:
options: A, B, C, D, E
The incremental NPV/Cost ratios for these are given by:
Eba Eca Eda Eea ; Ecb Edb Eeb ; Edc Eec ; Eed
6 Delete incremental NPV/cost ratios that are less than the user specified minimum
incremental value (MIV).
7 List the remaining incremental NPV/cost ratios in decreasing order (with the associated
section-option pair codes) and, within each incremental NPV/cost, in the order of
decreasing economic cost. For example, if Eeb = Edb then Eeb is ranked higher.
8 Select the next incremental NPV/cost ratio from the top of the list. If the lower cost
section-option is not the current Base Option for that section, continue selection until one
is found.
9 If the remaining budget is insufficient in any of the periods for the financial costs of
works required for the section-option selected in Step 8 above, then the selected option
should be rejected, and continue searching by repeating Step 8.
10 If the section-option can fit within the remaining budgets for all periods, deduct the net
increase in financial cost of capital works from all corresponding budget periods. Set the
Base option for this section to be that corresponding to the lower cost option for the
incremental NPV/Cost ratio chosen in Step 8. Providing that the remaining list is not
empty, return to Step 8.
The process described above continues until the budget is exhausted or there are no more
section-options remaining in the list. The resulting list of selected section-alternatives
constitutes the optimal work programme.
NPV
E
D
B
C A, B, C, D, E : Section-Alternatives
A
ECONOMIC COST
(relative to Base Option)
Figure B2.4 Efficiency frontier concept
Table B2.8 illustrates the output of this optimisation. In this example, the Resealing on section
N1-2 was pre-defined before optimisation and is therefore first on the list.
Table B2.7
Comparison of work programme alternatives
Note: * denotes a section with two alternative standards with high incremental benefits.
Table B2.8
Example of an optimised work program
1 Introduction
1.1 Concept
The concept of strategic planning of medium to long term road network expenditures requires
that a road organisation should consider the requirements of its entire road network asset.
Thus, strategy analysis deals with entire networks or sub-networks managed by one road
organisation. Examples of road networks include; the main (or trunk) road network, the rural
(or feeder) road network, urban (or municipal) road network, etc. Examples of sub-networks
include; all motorways (or expressways), all paved (or unpaved roads), different road classes,
etc.
In order to predict the medium to long term requirements of an entire road network or sub-
network, HDM-4 applies the concept of a road network matrix comprising categories of the
road network defined according to the key attributes that most influence pavement
performance and road user costs. Although it is possible to model individual road sections in
the strategy analysis application, most road administrations will often be responsible for
several thousand kilometres of roads, thereby making it cumbersome to individually model
each road segment. The road network matrix can be defined by users to represent the most
important factors affecting transport costs in the country. A typical road network matrix could
be categorised according to the following:
! Traffic volume or loading
! Pavement types
! Pavement condition
! Environment or climatic zones
! Functional classification (if required)
For example, a road network matrix could be modelled using; three traffic categories (high,
medium, low), two pavement types (asphalt concrete, surface treatments), and three pavement
condition levels (good, fair, poor). In this case, it is assumed that the environment throughout
the study area is similar and that the road administration is responsible for one road class (for
example, main roads). The resulting road network matrix for this would therefore comprise (3
x 2 x 3 =) 18 representative pavement sections. There is no limit to the number of
representative pavement sections that can be used in a strategy analysis. The trade-off is
usually between a simple representative road network matrix that would give rather coarse
results compared against a detailed road network matrix with several representative sections
that could potentially provide more accurate results.
Strategy analysis may be used to analyse a chosen network as a whole, to prepare medium to
long range planning estimates of expenditure needs for road development and conservation
under different budget scenarios. Estimates are produced of expenditure requirements for
medium to long term periods of usually 5-40 years. Typical applications of strategy analysis
by road administrations would include:
! Medium to long term forecasts of funding requirements for specified target road
maintenance standards (see Figure B3.5a).
! Forecasts of long term road network performance under varying levels of funding (see
Figure B3.5b).
! Optimal allocation of funds according to defined budget heads; for example routine
maintenance, periodic maintenance and development (capital) budgets (see
Figure B3.5c).
! Optimal allocations of funds to sub-networks; for example, by functional road class
(main, feeder and urban roads, etc.) or by administrative region (see Figure B3.5d).
! Policy studies such as impact of changes to the axle load limit, pavement maintenance
standards, energy balance analysis, provision of NMT facilities, sustainable road network
size, evaluation of pavement design standards, etc.
A typical sample application of the HDM-4 Strategy Analysis is given in
Appendix C.
1.2 Background
Strategy analysis is undertaken at infrequent intervals and is primarily a tool for resource
allocation by, or for, the top managers within a road agency. It is to be used for the analysis of
a chosen road network as a whole to prepare long-range planning estimates of expenditure
needs for road development and preservation under various budget scenarios. Strategy
analysis will be required to produce expenditure estimates for medium to long term periods of
between 5 to 40 years, under various budget constraints, maintenance standards, and road use
scenarios. It will predict key performance indicators of the road network such as average
roughness trends, surface distress, road user costs, safety levels and environmental effects.
When conducting a strategy analysis, the physical road network may be categorised into a
matrix defined by user-defined parameters such as road class, surface type, pavement
condition or traffic flow (Paterson, 1995). Strategy analysis utilises a model of the road
network (or sub-network) to simulate network-wide performance under specified road use
characteristics, maintenance standards, development standards and resource constraints.
The main difference between strategy and programme analysis is that programme analysis
deals with physical links and sections that can be identified from the road network. In strategy
analysis, the road system essentially loses its individual link and section characteristics as
these are grouped into representative sections to model the road network being analysed.
Thus, whereas in programme analysis the basic unit of the network is an identifiable road
section, the basic unit in strategy analysis is a representative section. Since sections are the
basic units of analysis in HDM-4 for both applications, similar types of economic analysis and
optimisation methods can be utilised for both modules.
$10m
6.0
$15m
5.0
4.0 $20m
Target
3.0
2000 2001 2002 2003 2004 2005 2006
Budget Allocations
7.0 Feeder
Average Roughness (IRI)
Roads
$30m/yr
6.0
Secondary
Roads
5.0 $35m/yr
4.0 Primary
Roads
$20m/yr
3.0
2000 2001 2002 2003 2004 2005 2006
250
200
D e v e lo p m e n t
U S $ m /y e a r
150
Im p ro v e m e n t
P e r io d i c
100
R o u tin e
50
0
2000 2001 2002 2003
200
150
U S $ m /y e a r
P rim a ry
100 S e c o n d a ry
F eed er
50
0
2000 2001 2002 2003
2 Determining long-term network performance trends - for expected long term funding
levels
The first objective can be solved by defining appropriate maintenance and improvement
standards, whilst the second objective is essentially an optimisation problem, with one or
several long-term budget constraints. In the second objective, one of two objective functions
may be used to select the best network development and/or preservation options that will
satisfy the constraints:
! Maximisation of benefits (NPV)
The choice of objective function depends on the particular decision situation. Benefit
maximisation is the standard approach and maximisation of the improvement in network
condition may be appropriate where the reduction in roughness can be used as a proxy of
reduction in vehicle operating costs.
It should be noted that the network development or preservation standards to be optimised are
user-defined and are not therefore the set of all possible standards for the network being
analysed.
As with programme analysis, a key difference between strategy and project analysis is in the
detail with which data is defined. Project level analysis data is specified in terms of measured
defects, often at information quality level II (IQL-II) (Paterson and Scullion, 1990), whereas
the specification for strategy analysis is more generic at IQL-III or IQL-IV. As an example,
roughness would be specified in terms of the international roughness index for project level
analysis, whereas for strategy analysis, the riding quality could be specified as good, fair or
poor. The relationship between IQL-II and IQL-III data is user defined depending on road
class, pavement class and traffic class as illustrated in Table B3.1. The default relationships
will be defined in the HDM system configuration (Wightman et al., 1994).
Table B3.1
Example of relationships between IQL-II and IQL-III data
Poor 8.0
Good 5.0
Poor 2.0
Good zero
Main Paved High Structural Fair Wide 10
adequacy cracks
Poor 30
Good 10
Fair Deflection 50
Poor 100
Good zero
Poor 30
etc.
Table B3.2
Sample matrix template for strategy analysis
etc. : : : :
The relationships between class names (for example, High, Medium, Low; or Good, Fair,
Poor) are defined in the HDM-4 Configuration.
to define a comprehensive set of vehicle types from which a few may be selected for the
analysis.
This process is similar to the selection of vehicle types in the Project Analysis module.
If option (3) is selected, either option (1) or (2) must follow it, unless it is the final year of the
analysis period.
This process is similar to the specification of traffic characteristics for Project Analysis.
In addition users can select the types of output required from a list of outputs produced by
HDM-4. (These include all of those currently produced by HDM-III and EBM.) In addition,
graphs can be created from selected types of output. Samples of typical outputs from strategy
analysis are included in Appendix C1 (case studies).
3 Strategy analysis
The strategy analysis process is divided into four phases:
1 Definition of investment options to be used (see Section 3.1)
Table B3.3
Example of definition of investment options
Label: [STR97]
Representative section Surf. Maintenance Improvement Future
class maintenance
S1 B Basic (B) - -
B Ideal (B) - -
U Basic (U) - -
B Ideal (B) - -
S4 B Ideal (B) - -
B Basic (B)
Section 3.1. As in the project analysis, the user has to define the base policy that is used in the
calculation of economic indicators. This is done in the Run Setup of the Strategy Analysis.
The results are saved for economic analysis and optimisation (see Analytical Framework and
Model Descriptions - Part G).
After life-cycle analysis, the options with the highest NPV (Net Present Value) are assigned
for each representative section. The result of this process is a tentative unconstrained strategy
for the road network, determined by the maintenance or improvement standards defined by the
user. This result reflects the overall needs for road works and it is illustrated in Table B3.4.
Table B3.4
Example of a Tentative Strategy (Unconstrained)
Label: [STR97]
Representative section Strategy NPV/C
S1 OL50 at 4.0 1,1
S3 Rs at 4 years 2,1
S4 Do nothing ---
S5 Do nothing ---
---
Annual
B. NEEDS 3000
B. CONSTRAINT 2300
If the needed budget for each budget period is below the given budget constraints, no further
economic analysis is necessarily needed, and the unconstrained solution can be used as an
optimal strategy. However, if the budget needs are higher than the available budgets, the user
has two choices:
1 Revise maintenance and improvement standards and run the life-cycle analysis again
2 Execute optimisation
3.3 Optimisation
Within strategy analysis, three objective functions are available (see Analytical
Framework and Model Descriptions - Part G):
1 Total enumeration
If the number of representative sections to be analysed is less than 100, and there are no more
than five budget periods and 16 alternatives per representative section, total enumeration can
be adopted. This will be externally done using the EBM-HS model of HDM-III. If the above
constraints are exceeded, incremental benefit/cost ranking must be used.
Both total enumeration and incremental benefit/cost ranking are similar to those in the
Programme Analysis (Kerali and Mannisto, 1999).
An alternative objective function is to minimise road agency costs in order achieve specified
target road network condition. In this case, HDM-4 selects from amongst the specified road
work standards, the combination which will minimise road agency costs in achieving the
target road network condition.
The results of the analysis are stored in the database. The structure of this is as shown in Table
B3.5.
Table B3.5
Structure of output from total enumeration optimisation
etc.
4 References
Federal Highway Administration (1994)
Highway Economic Requirements System, (HERS), FHWA, Washington D.C.
Kerali, H.R., and Mannisto V. (1999)
Optimisation methods for Programme and Strategy Analysis in HDM-4. Journal of the
Transportation Research Board, Record 1558, National Academy Press, Washington
DC.
Paterson, W.D.O. (1995)
A Concept Outline of the Application Analysis Levels and the Strategic Analysis
Application for the New Highway Development and Management Model (HDM-4).
(Published in Reference No. 5 below)
Paterson, W.D.O. and Scullion, T. (1990)
Information Systems For Road Management: Draft Guidelines On System Design And
Data Issues. Infrastructure and Urban Development Department Report INU77. The
World Bank, Washington DC.
Robinson, R. and Kerali, H.R. (1995)
Proceedings; Strategic and Programme Analysis Workshop, May 1995, School of Civil
Engineering, The University of Birmingham.
Watanatada T., Harral C. G., Paterson W.D.O., Dhareshwar A.M., Bhandari A., and
Tsunokawa K., (1987)
The Highway Design and Maintenance Standards Model - Volume 1: Description of the
HDM-III Model.
The Highway Design and Maintenance Standards Series Baltimore:
Johns Hopkins University Press for the World Bank
Wightman, D.C., et al, (1994)
Design Overview and Development Strategy, August 1994, School of Civil Engineering,
The University of Birmingham.
Contents
Applications Guide i
Version 1.0
PART C RPRA
Part C RPRA
1 Introduction
This chapter describes the procedure for Research, Policy and Regulation analyses within
HDM-4, and includes details of case studies.
1.1 Background
To be included in a subsequent edition of this document
Contents
5 References D1-28
D2 Data Hierarchy
1 Introduction D2-1
2 Data hierarchy D2-2
2.1 Aggregate and detailed data D2-2
Applications Guide i
Version 1.0
PART C RESEARCH, POLICY AND REGULATION ANALYSES CONTENTS
Applications Guide ii
Version 1.0
PART C RESEARCH, POLICY AND REGULATION ANALYSES CONTENTS
D5 Traffic Data
1 Introduction D5-1
2 Representation of traffic D5-2
2.1 Purposes of traffic data D5-2
2.2 Data types D5-3
3 Traffic categories D5-4
4 Traffic composition, volumes and growth rates D5-5
4.1 Basic features D5-5
4.2 Composition, volumes and growth rates D5-6
4.3 Input Data Definitions: Traffic Data D5-8
Applications Guide iv
Version 1.0
PART D DATA MANAGEMENT
D1 Customising HDM-4
1 Introduction
This chapter describes the data sets that can be customised under HDM Configuration. The
primary role of the data configuration system, within HDM-4, is the management of all of the
default data to be used within the analysis. A set of default data are provided with the HDM-4
software, but users may modify this to reflect local circumstances.
2 Overall structure
! Hourly traffic flow distribution (traffic flow pattern) (see Section 3.3)
An overview of the modelling logic is given in Part A1. A more detailed description of the
models is provided in the Analytical Framework and Model Descriptions and the A Guide to
Calibration and Adaptation.
The maximum number of vehicles that can pass a point, or traverse a road section, in one
hour (total both directions). Capacity values determine the shape of speed-flow curves
by establishing the ultimate capacity value.
! Free speed
The speed of each vehicle at zero (or very low) flow. It is unaffected by other traffic but
is affected by the physical characteristics of the road and other non-traffic factors. The
average free speed is calculated for each vehicle type.
! Speed at capacity
As traffic flows increase, average speeds for all vehicles converge towards the speeds of
the slowest vehicles in the stream, as passing becomes more and more restricted. As flow
approaches capacity, average speeds may fall even lower than slow vehicle free speeds,
due to small disturbances in the traffic stream. An estimate of average speed at capacity,
also known as jam speed at capacity, is needed to describe the speed-flow-capacity
relationship.
To model the effects of traffic congestion, the mixed traffic flows are converted into
equivalent standard vehicles. The conversion is based on the concept of Passenger Car Space
Equivalents (PCSE). This accounts only for the relative space taken up by the vehicle on the
road, and reflects the fact that the speed-flow model takes account explicitly of speed
differences of the various vehicles in the traffic stream. PCSE factors vary by road type, and
narrow roads have higher PCSE values than wide roads. Hoban et al. (1994) gives the concept
of PCSEs.
The following data are specified for each speed-flow type:
! Ultimate capacity, Qult (PCSE/lane/hr)
Where:
XQ1 = Q0/Qult
Q0 = Free flow capacity
Qult = Ultimate capacity
! Nominal capacity as a proportion of the ultimate capacity, XQ2
Where:
XQ2 = Qnom/Qult
Qnom = Nominal capacity
Qult = Ultimate capacity (=QLult * NLANES)
! Jam speed at capacity, Sult (km/hr)
The number of speed-flow types to be defined is at the discretion of the user. Table D1.1
gives typical examples of capacity and speed-flow data for different road widths. Since this
data applies to individual sections, it is important to ensure that the data relates to a single and
not a dual carriageway. This ensures consistency with the definition of a road section.
Table D1.1
Examples of capacity and speed-flow model parameters for different road types
Key:
Sult = Jam speed at capacity (km/h)
σmaxr = Maximum acceleration noise (m/s2)
the full year. Because congestion delays and costs are greatest during the highest-flow hours,
particular attention should be paid to these hours. These highest-flow hours should be divided
into periods of shorter duration.
The data that describes the hourly traffic flows and the volume-capacity ratios are required for
modelling congestion effects on vehicle speeds and vehicle operating costs. The key
parameters are as follows:
! Qp
The traffic flow, in PCSE per hour, during each traffic flow period p.
! VCRp
The number of flow periods into which the total number of hours in a year (8760) are
divided. Only one flow period needs to be specified for aggregate or network level
analyses. A uniform traffic composition and value of travel time across all flow-
frequency periods is assumed.
! The number of hours in each flow-frequency period, p (HRYRp)
Table D1.2
Examples of hourly traffic flow distribution data
The hourly traffic flow for each flow-frequency period is expressed as a proportion of AADT,
and is given by:
365 * PCNADTp
HVp = . . .(3.1)
100 * HRYR p
where:
Qp =
∑ HV
k =1
p * PCSE k * AADTk . . .(3.2)
where:
The volume-capacity ratio during each traffic flow period is expressed as follows:
Qp
VCRp = . . .(3.3)
Q ult
Default values of accident rates at intersections will not be provided in HDM-4 since this
would be unrealistic due to the wide variations in accident rates from country to country.
The number of intersection types to be defined is at the discretion of the user.
Note: Intersections are not implemented in the current HDM-4 software release.
Table D1.3
Detailed data related to moisture classification
Table D1.4
Detailed data related to temperature classification
Temperature range 15 50 40 45 60
(°C)
Percent of vehicles 0 0 0 10 20
with studded tyres
Percent of driving 0 0 0 10 20
times on snow
covered roads
Percent of driving 20 10 15 5 10
times on water
covered roads
Note:
* The Freezing index is measured in OC-days.
o
( C - days)
-5 5 1 5
-4 4 2 8
-3 3 4 12
-2 2 13 26
-1 1 15 15
>=0 0 330 0
Totals 365 66
The number of climate zones to be defined is at the discretion of the user. Each climate zone
will have a unique name, and will have been assigned detailed values (default or otherwise) to
its moisture classification and temperature classification.
The road surface class to which the traffic band applies (that is, bituminous, concrete or
unsealed).
! Mean AADT
The number of traffic bands to be defined for each road surface class is at the discretion of the
user. Table D1.5 shows the detailed traffic levels assumed related to different road surface
classes, with the following default levels:
! Low
! Medium
! High
Table D1.5
Definition of traffic bands by road surface class
LA B
A
C2
R1
A F1
R2 F2
B
R3
The number of geometry classes to be defined is at the discretion of the user. Table D1.6.
presents the detailed data assumptions relating to the default geometry classes for HDM-4.
The values given can be amended as required.
Table D1.6
Definition of geometry classes
4 Bendy and 15 2 75 3 80
gently
undulating
Structural adequacy
The strength of bituminous pavements is defined by their structural adequacy to carry traffic
loading. At aggregate data level, structural adequacy is defined in terms of qualitative
descriptors/measures such as good, fair, poor, etc. The detailed data values relating to these
are in terms of the Adjusted Structural Number of the Pavement (SNP). The detailed data
have been assigned by traffic level.
In HDM Configuration, the user should first define the number of qualitative measures of
structural adequacy that should be used, and their names. Then for each qualitative measure,
and for each of the pre-defined traffic bands, the user defines the mean value of the dry season
SNP.
The default data assumptions for structural adequacy of bituminous pavements in terms of
three qualitative descriptors (good, fair, and poor), for different traffic bands (that is, low,
medium, and high), are shown in Table D1.7.
Table D1.7
Definition of structural adequacy using mean SNP values
The HDM-4 definitions relating to Adjusted Structural Number of the Pavement (SNP) are
given in the Analytical Framework and Model Descriptions. Note that these parameters
represent further modification of the Modified Structural Number adopted by HDM-III.
Note also that Benkelman Beam and FWD deflection data can be used to derive SNP values
automatically by HDM-4.
The user should define several ranges (for example, 5) of SNP values by entering the
minimum and maximum values for each range. For each SNP range, and for each of the
eight HDM-4 pavement types (plus the option of all pavement types), the user should
define the following:
❏ most recent surfacing thickness (mm)
❏ previous/old surfacing thickness (mm)
❏ road base thickness - if base type is SB-Stabilised Base (mm)
In addition, the user should provide the following information to be used as defaults in all
analyses:
❏ surface material for AM (Asphaltic Mix) surface type
❏ surface material for ST (Surface Treated) surface type
❏ resilient modulus for stabilised roadbase type (GPa)
! Method 2 - HDM-4 default
Alternatively, SNP can be converted into pavement layer thicknesses as shown in Table
D1.8.
Table D1.8
Pavement Layer Thickness
Note:
Default strength coefficients for the asphalt and base materials are 0.35 and 0.15 respectively.
This method will be provided as default, together with the following information:
❏ surface material for AM will be asphalt concrete (AC)
❏ surface material for ST will be double bituminous surface dressing (DBSD)
❏ resilient modulus for stabilised bases, CMOD = 12 GPa
Structural adequacy
The definition of structural adequacy for concrete pavements is based upon the slab thickness,
traffic level, and either the modulus of rupture (MR28) or the percentage of steel, depending
on the pavement type.
In HDM-4 Configuration, first the user should define the number of qualitative measures of
structural adequacy that should be used, and their names. Then for each qualitative measure,
and for each of the pre-defined traffic bands, the user defines the detailed data as described
below:
! Jointed Plain Concrete Pavement (JP concrete pavement)
Table D1.9
Definition of structural adequacy of JP concrete pavement using slab thickness
in mm
Table D1.10
Definition of structural adequacy of JR concrete pavements using slab
thickness in mm
Table D1.11
Definition of structural adequacy of CR concrete pavements using slab
thickness in mm
Table D1.12
Default values of other variables
General
The modelling of the deterioration of unsealed roads requires detailed data on layer material
gradation and plasticity index. There is a wide variation in materials from place to place,
making the provision of sensible default materials in HDM-4 difficult. Users must therefore
define specific detailed data for layer materials that can be retrieved when aggregate data are
specified.
For unsealed roads, structural adequacy and surface condition are both represented by the
thickness of the surfacing material, gravel (see Section 3.9.2).
Surface material
The user should define a surface material by giving its name, and defining the following
detailed data:
! Maximum particle size (mm)
! Percent passing 2.0 mm sieve
! Percent passing 0.425 mm sieve
! Percent passing 0.075 mm sieve
! Plasticity index (%)
The default surface materials for unsealed roads that are included in HDM-4 are shown in
Table D1.13.
Table D1.13
Definition of surface materials for unsealed roads
4 Coral 21 64.3 49 25 13
gravels
angular
Subgrade material
The user should define a subgrade material by giving its name, and defining the following
detailed data:
! Maximum particle size (mm)
! Percent passing 2.0 mm sieve
! Percent passing 0.425 mm sieve
! Percent passing 0.075 mm sieve
! Plasticity index (%)
Table D1.14 shows the subgrade materials that have been provided as defaults in HDM-4.
These are based on Casagrande Soil Classification.
Table D1.14
Definition of subgrade materials for unsealed roads
2 Gravel-sand 17 46 34 23 17
mixtures with
excess of fines,
GF
The default data assumptions for ride quality of bituminous pavements, concrete pavements,
and unsealed roads in terms of four qualitative measures (good, fair, poor and bad), for
different road classes (that is, primary, secondary, and tertiary), are shown in Table D1.15,
Table D1.16 and Table D1.17, respectively.
Table D1.15
Default values for ride quality - bituminous roads
Secondary or Main 3 5 7 9
Tertiary or Local 4 6 8 10
Table D1.16
Default values for ride quality - concrete roads
Secondary or Main 3 5 7 9
Tertiary or Local 4 6 8 10
Table D1.17
Default values for ride quality - unsealed roads
Secondary or Main 6 9 12 15
Tertiary or Local 8 12 16 20
Table D1.18
Default distress values for surface condition - bituminous surface
Ravelling (%) 0 1 10 20 30
Table D1.19
Default distress values for surface condition - JP concrete pavements
Faulting (mm) 0 1 2 4 8
Table D1.20
Default distress values for surface condition - JR concrete pavements
Faulting (mm) 0 1 4 7 10
Table D1.21
Default distress values for surface condition - CR concrete pavements
For unsealed roads, surface condition and structural adequacy are both related to the traffic
level and are represented by the thickness of the gravel surfacing. The user can define the
number of qualitative measures of surface condition that should be used, and their names.
Then for each qualitative measure, and for each of the pre-defined traffic bands, the user can
define the mean values of the gravel thickness.
The default values of gravel thickness for unsealed roads are given in Table D1.22.
Table D1.22
Default gravel thickness (mm) for surface condition of unsealed roads
Low 100 50 25 0
Table D1.23
Default distress values for surface texture - ST pavements
Table D1.24
Default distress values for surface texture - AM pavements
For concrete pavements and unsealed roads, the user should define the default values of
texture depth. The default value used is 0.
The default values for construction quality relating aggregate to detailed data are shown in
Table D1.25.
Table D1.25
Definition of construction quality defaults
Relative compaction 97 91 91 85 85
For unsealed roads, the default construction method is mechanical. The user may change this
in HDM Configuration.
3.10.2 Age
The three surfacing ages (AGE1, AGE2, AGE3 and AGE4) required for modelling the
deterioration of bituminous pavements can be derived from the input last surfacing year (at
section level) as follows:
If the surface type is ST, then the above input is taken as AGE2; and:
AGE3 = AGE2 + 5
AGE3 = AGE4
AGE1 = AGE2
If the surface type is AM, then the above input (last surfacing year) is taken as AGE3, and:
AGE4 = AGE3 + 5
Defined as a percentage of total carriageway area; default value = 25% for ST surface
type, and 40% for AM surface type.
! Area of previous ‘wide structural cracking’ (PCRW)
3.11 Miscellaneous
3.11.1 Shoulders
Shoulder width is defined by section, with the two unsealed shoulders for each road section
being the default, and with a condition rating which is the same as that of the carriageway.
The default value of the difference in elevation between the carriageway and the shoulder,
ESTEP, is 15 mm.
The user may define the following default information in the configuration file:
! Number of shoulders
! Shoulder type
! ESTEP (mm)
3.11.2 Drainage
At the aggregate level, drainage condition can be defined by a qualitative measure (for
example, excellent, good, fair, poor, very poor, etc.) that represents the Drainage Factor
(DF). The values of DF are used in the modelling of bituminous pavement performance.
The user should define the number of drainage qualitative measures to be used, and their
descriptions. For each qualitative measure and for each drain type the user should specify the
value of DF.
The default data assumptions for drainage condition for bituminous pavements, in terms of
five qualitative measures (excellent, good, fair, poor and very poor), for different drain types
are shown in Table D1.26.
Table D1.26
Default DF values for drainage condition
The user can change the values of Annual Deterioration of Drainage Factor (ADDF) by
temperature type for each drain type. The default values of ADDF are given in Chapter C2 of
the Analytical Framework and Model Descriptions.
4.1 Currency
The currency to be used in HDM-4 is specified in HDM Configuration using the following
information:
! Name
! Symbol
! Symbol position
! 'Thousand' separator
5 References
Hoban C., Reilly W., and Archondo-Callao R., (1994)
Economic Analysis of Road Projects with Congested Traffic
Methods for Economic Evaluation of Highways Investments and Maintenance
Transport Division, Transportation, Water & Urban Development Department
The World Bank, Washington DC, USA
D2 Data Hierarchy
1 Introduction
This chapter considers the data requirements for the different HDM-4 analyses. The concepts
of key data, aggregate data and detailed data are discussed. The data hierarchy complies with
the concept of information quality levels whereby data is specified at different levels of detail
(tiers) for different purposes (Paterson and Scullion, 1990).
2 Data hierarchy
Table D2.1
Definition of traffic bands by road surface class
Within HDM-4, the user may define the number of aggregate level bands and the associated
detailed data. Default data sets (specifying aggregate descriptors and associated detailed
values) are given in Chapter D1.
Key network data comprises those attributes needed to define the physical and functional
characteristics of the network; for example, section length and width, speed-flow type and
traffic-flow pattern.
Key vehicle data comprises the vehicle physical characteristics, utilisation details and costs
(including vehicle, labour and time costs).
Works costs are key data, and include costs associated with Maintenance Works and
Improvement Works. A default set of costs for both works categories is held under Works
Standards/Default Works Costs. Maintenance Works are assumed to take place within a
specified calendar year. With Improvement Works, the percentage of costs incurred over each
year of the works period (between 1 and 5 years) is specified.
4 References
Paterson W.D.O., and Scullion T., (1990)
Information Systems for Road Management: Draft Guidelines on System Design and
Data Issues, Technical Paper, Report INU 77. Infrastructure and Urban Development
Department. The World Bank, Washington DC, USA
1 Introduction
This chapter discusses the concept of road network referencing used in HDM-4. A flexible
framework is provided to cater for a wide range of road network representations. The Road
Network folder holds the road network data that is used by all of the HDM-4 applications.
2 Basic concepts
2.1 General
Road network referencing is a system used for the locational positioning of data pertaining
to roads. It defines the technical rules and conventions employed to identify data on a
particular part of a road network. There are two types of locational referencing data:
! Data that describe the position and geometry of a road segment
! Attribute data that describe characteristics of the road or inventory associated with it
Attribute data are assigned to a location or segment of a road according to the segmentation
scheme used. Locational referencing and road segmentation together forms the basis for the
retrieval of road network information.
The establishment of a road network referencing system is principally a modelling process
that maps the real world on to its equivalent computer abstraction. It is essential to make a
distinction between objects and events that exist in the real world (that is, physical) and those
that are stored in the computer (that is, logical). The physical view is both locational and
quantifiable, whilst logical data are merely an abstraction (that is, a representation) of the real
world. Information is also needed about how objects and events are related to each other (that
is, logical relationship). Each entity or data item associated with a road network may have
both logical and physical attributes, together with the logical relationships between them.
Sections
A section is a length of road that is homogeneous in terms of its physical attributes. A road
section may be split further into sub-sections, normally for purposes of data collection.
Sub-sections
A sub-section is defined as a part of a road section that has strong consistency in particular
attributes at a higher degree of precision and reliability.
Network
A network is a collection of sections that are grouped together for administrative or analysis
purposes.
Sub-networks
A sub-network is defined as a smaller grouping of sections within a larger road network.
These may be selected according to common logical attributes such as pavement type, road
function, administrative boundary, etc.
Nodes
Within a road network, nodes mark points at which traffic characteristics change or road
features change significantly. Such points would include changes in traffic volume or
composition, significant changes in geometry, road layout, population centres or
administrative boundaries.
Links
A link is defined as a length of road joining two nodes. A link should have a consistent
geometric layout along which the volume of traffic flow is uniform (that is, traffic does not
vary substantially).
Note: Nodes Links are not implemented in the current release of the HDM-4 software.
Road networks can be represented or modelled in two ways:
! Physical networks
Physical networks comprise the geometric description of the roads within a region.
Physical networks usually contain locational and inventory information, and are
represented graphically using two-dimensional maps.
! Logical or administrative networks
! Traffic volume
Each logical section is identified by a unique name and ID, either of which may be used to
identify the section on the various HDM-4 output reports. The logical definition of the
network within the HDM-4 database must be matched by physical referencing on the ground.
This may take the form of road markings (for example, studs), concrete posts (for example,
kilometre posts) or bar codes, typically mounted on posts. It is important that the physical
referencing is maintained to facilitate the future location of sections.
3 Data levels
Facilities are provided for adding (new), modifying (edit attribute) and deleting attributes in
these data groups.
The data groups are summarised in sections 4.2 and 4.3. The specification of road network
data (for example, speed-flow type, traffic flow pattern, and climate zone) is described in
Chapter D1. Network element data (aggregate and detailed) are described in Section 5.
5.2 Sections
A list of parameters associated with a particular section, defined by their type and value
as may be appropriate in the data field format; examples are geometry, pavement
structure, climate, topography.
Road sections may be pre-defined as in the case where data are imported from an existing
road management system.
Table D3.1
Surface class and pavement type
Note:
Definitions of pavement types are given in the Analytical Framework and Model
Descriptions.
The speed-flowtype, traffic-flow pattern, and climate zone of the section are assigned as
described in Chapter D1.
The geometric characteristics of the section are assigned in terms of the following:
! Length of section (km)
❏ road class
❏ traffic
❏ geometry
! Pavement
❏ structural adequacy
! Condition
❏ ride quality
❏ surface condition
❏ surface texture
! History
❏ construction quality
These aggregate data types are described below. Detailed definitions are assigned in HDM
Configuration and are described in Chapter D1.
General
Road Class
It is usually convenient to classify sections in a network by road class according to a
functional hierarchy. Such a classification facilitates management or analysis of the road
network in a number of areas:
! Managing roads
Different road administrations or agencies may have the responsibility for managing
roads of different classes. Classification can be used to assist in identifying who is
responsible for what.
! Funding
Funding may also be allocated separately to roads of different classes, and the source of
funding may also differ. For example; main or trunk roads may be funded with monies
raised and allocated by central government, and district or village councils may fund
local roads.
! Road functions
Classification enables different road functions to be recognised. Main roads may have
the purpose of carrying long distance traffic whereas local roads may have the purpose of
providing only access to industrial, commercial, agricultural or domestic premises.
These functions should be defined in the policy framework of the road administration,
and different objectives, standards and intervention levels for works should be set up for
each road class.
! Types and numbers of traffic
The different functions of road classes are likely to be reflected in different types and
numbers of traffic being carried by each. The capacity that is provided by each road class
is also likely to differ. However, there may be road sections in the network that have
relatively low traffic but are of key strategic importance because of the places that they
link. The maintenance standards, intervention levels and priorities assigned to such
sections can be related to the road class, rather than the traffic level.
! Maintenance and improvement standards
The division of roads into classes enables users of HDM-4 to set up maintenance and
improvement standards, and priorities that are appropriate to each functional class.
Default values of road class are:
! Primary or trunk
! Secondary or main
! Tertiary or local
Traffic
For each road section, traffic level is specified in terms of annual average daily traffic
(AADT) flow. At aggregate data level, traffic volume can be defined, with the following
default levels:
! Low
! Medium
! High
Geometry
Horizontal and vertical curvature is defined in terms of various combinations of the following
default values:
! Horizontal curvature
! Vertical curvature
Pavement
Pavement details are defined in terms of the following attributes and default values:
! Structural adequacy
❏ good
❏ fair
❏ poor
Condition
Ride quality
Ride quality is an indication of the roughness of the road. It is an important parameter for
indicating road condition and maintenance needs, and for predicting vehicle operating costs.
It is defined in terms of the following default levels:
! Good
! Fair
! Poor
! Bad
Surface condition
This is an aggregate measure of several deterioration parameters. The following default
values are used:
! New
! Good
! Fair
! Poor
! Bad
Surface texture
This value gives an indication of the surface texture and skid resistance of the surface. The
default values are:
! Good
! Fair
! Slippery
History
Construction quality is specified under this heading with the following defaults:
! Good
! Fair
! Poor
Pavement defects.
! History
5.3 Nodes
The specification of nodes is an optional attribute of the road network representation in HDM-
4. Nodes are considered as entities (that is, separate data items) in all analyses. Typically
nodes are intersections or junctions where two or more road sections meet. In addition to this
are intermediate nodes that include the start and end of dual carriageways, administrative
boundaries, and the like. The start or end of a road (that is, a dead-end) is considered to be a
separate type of node. Thus, a road network may comprise three types of node:
! Intermediate
Positioned at some other point on the section where a locational position needs to be
referenced for some particular purpose.
! Intersection
Positioned at a dead end of a section, where there are no junctions with other sections.
A node should be specified by its name and a unique reference number through a textual
description and a label or ID. The ID should consist of a short alphanumeric code that
provides a key to the node. It is often helpful if the ID is set to the grid co-ordinates of the
node, since this may assist subsequent spatial location.
The intersection type can also be defined, with a value being assigned from those specified
earlier. For analytical purposes, it is assumed that only intersections require attributes to be
specified. The results calculated for nodes may be combined with those of the adjoining road
section, or taken individually. Each node may be assigned more than once to different
sections within the same road network.
Note: Nodes are not implemented in the current version of HDM-4.
6 Special considerations
Intermediate node
direction
local dualling
A A’ B C
single-carriageway double-carriageway
A B C
7 Applications issues
8 References
Robinson R., Danielson U., and Snaith M., (1998)
Road Maintenance Management Concepts and Systems. London and Basingstoke:
Macmillan Press.
1 Introduction
Vehicle Fleet data describes the characteristics of vehicles that use a road network. Vehicle
fleet characteristics are required in HDM-4 for the estimation of traffic flow and capacity,
vehicle operating costs (VOC), travel time costs, accident costs, and the evaluation of
environmental effects from vehicle emissions and traffic noise. Within HDM-4, vehicle fleet
data are held in the Vehicle Fleet folder.
This chapter describes the methodology used for the storage and retrieval of the vehicle
parameters to be used in HDM-4. A vehicle fleet representation framework has been provided
to suit the needs of many countries and will also satisfy the analytical requirements of HDM-
4.
2 Basic concepts
2.1 General
Road administrations normally only deal with the vehicle fleet used in their region or country.
However, international consultants often undertake road investment studies in many different
countries, and would therefore deal with several vehicle fleets. It is possible to deal with both
of these requirements for vehicle fleet representation within HDM-4. Several sets of vehicle
types may be defined corresponding to the requirements of the different users and types of
analyses.
A vehicle classification system is primarily a modelling process that attempts to represent the
diverse characteristics of the vehicles in a traffic mix. It uses a set of representative vehicles
for which a number of physical and performance characteristics are defined. The
representative vehicles may or may not exist physically. However, the parameters or
attributes specified for the representative vehicles should reflect the overall physical and
performance characteristics of the range of vehicles that use a road network.
Figure D4.1 illustrates the hierarchical structure of the relationships between vehicle
categories, vehicle classes and vehicle types.
Motorised
Motorised Categories
Motor
Motorcycles
cycles Passenger
Passengercars
cars Utilities
Utilities Trucks
Trucks Buses
Buses Classes
Motor
Motorcycles
cycles Small
Smallcar
car Medium
Medium Light
Lighttruck
truck Medium
Medium truck
truck
(1) Types
(1) (2)
(2) car
car (8)
(8) (9)
(9)
(3)
(3)
Large
Largecar
car
(4)
(4)
Heavy
Heavytruck
truck Articulated
Articulated truck
truck
(10)
(10) (11)
(11)
Light Light
Lightgoods Mini-bus Light
Light goods Mini-bus Lightbus
bus
delivery
delivery vehicle
vehicle (12) (13)
(12) (13)
vehicle
vehicle (6)
(6)
(5)
(5)
Four
Fourwheel
wheel Medium
Mediumbusbus Heavy
Heavybus
bus
drive
drive (14)
(14) (15)
(15)
(7)
(7)
Coach
Coach
(16)
(16)
❏ Motorcycles
❏ Cars
❏ Utilities
❏ Trucks
❏ Buses
! Non-motorised vehicles
❏ Animal Cart
❏ Bicycle
❏ Pedestrian
❏ Rickshaw
2.2.5 Trailers
Trailers have not been included as a separate class/type of vehicle. Vehicles with semi-trailers
or drawbar trailers should, therefore, be included as a separate vehicle type, with appropriate
values for the gross vehicle mass, number of wheels, and other parameters.
3 Data levels
Detailed data items are those used in the models or relationships built into HDM-4. These
include, for example, details of the physical characteristics, utilisation, performance, and other
calibration parameters. Such data will normally be determined following detailed
investigations of the particular vehicle fleet under study (see Section 5.4).
Default values are available for several items of data in both tiers.
Facilities are provided for adding (new), modifying (edit attribute) and deleting (remove)
attributes.
5 Vehicle data
5.2.1 Definition
A vehicle type can be defined or modified by specifying the following set of data:
! Name
A description or unique reference for each vehicle type within the vehicle fleet.
! Base vehicle type
! Description
A textual description that can be included to give more details relating to user-defined
vehicle types.
! Life method
This may be either Constant Life or Optimal Life, used for RUE analysis.
❏ tyre type
❏ base number of recaps per tyre carcass
❏ retread cost as a percentage of the new tyre cost
! Utilisation
5.3 Costs
Costs are specified in terms of economic and financial unit rates. Costs data are specified
under the headings:
! Vehicle resources (economic and financial costs)
! Overheads associated with operating the vehicle (lump sum per year)
! Optimal life
Acceleration noise from different sources and roughness at which maximum acceleration
noise occurs.
! Energy
Energy use parameters, also tyre weight and unladen vehicle weight.
Further details of calibration data are given in Part E of the Analytical Framework and Model
Descriptions.
6 Special considerations
7 Applications issues
D5 Traffic Data
1 Introduction
This chapter describes the traffic characteristics used in HDM-4. Section 2 summarises the
traffic data requirements for different HDM-4 applications, and lists the traffic data types.
The three basic traffic categories (normal, diverted and generated) are described in Section 3,
and their specification detailed in Section 4. Further details are given in Part B of the
Analytical Framework and Model Descriptions.
2 Representation of traffic
The required traffic data for this type of analysis is similar to that for project analysis, with
the exception that the data are at a more aggregate level. For example; traffic volumes may
be specified by vehicle classes, and several representative road sections could use the data.
! Strategy analysis
This requires the specification of an aggregate set of traffic data, that is representative of
a group of road sections being analysed. Traffic levels should be expressed in terms of
daily flows, and may be described as low, medium, or high. The composition of traffic
may be expressed as a percentage of daily flow for each vehicle class/type.
For the purposes of representing traffic characteristics both for project and network level
analyses, road links within a network must be categorised according to the following:
! Speed flow type
The speed flow type defines the parameters for capacity, speed-flow relationship shape,
width effects and passenger car space equivalents for different road widths (for example,
single lane road, two lane road, wide two lane road, four lane road).
! Traffic-flow pattern
The Traffic flow pattern describes the patterns of traffic flow along each road category;
commuter routes, for example, tend to have weekday peaks but low weekend traffic,
whereas recreational routes have a more peaked distribution.
! Non-motorised transport factor
! Roadside friction
This is a measure of the effect of roadside activity on traffic speeds; and includes the
effects of land use, roadside stalls, bus stops, parking, etc.
These are specified within the individual applications programs (that is, project analysis,
programme analysis and strategy analysis) as required by the purpose of the particular
study.
! Axle loading
This information is specified for each vehicle type in the Vehicle Fleet folder (see
Overview of HDM-4). Provision is made to derive most of the required parameters (for
example, equivalent standard axle load factors, etc.) from the user-specified data. Details
of axle loading data requirements are given in Part B of the Analytical Framework and
Model Descriptions.
! Road capacity and speed-flow relationships
These parameters are defined by the speed flow type within HDM-4 Configuration.
Further details are given in Chapter D1.
! Traffic-flow pattern
This defines the hourly flow frequency distribution over a year, and is used to estimate
hourly flow ranges from annual average traffic data. Traffic flow patterns are road
dependent and are specified for each road use category within HDM-4 Configuration.
Further details are given in Chapter D1.
3 Traffic categories
Traffic categories consist of the following traffic data types:
! Normal (see Section 4.2.1)
Normal traffic is defined as traffic that would pass along the project road if no investment
took place. It is specified for each road section within the applications programs.
! Diverted (see Section 4.2.2)
Diverted traffic is defined as traffic that changes from another route (or transport mode)
to the project road. It is specified together with the road investment option that causes it,
and is specified within the relevant applications programs.
! Generated (see Section 4.2.3)
Generated traffic is defined as additional traffic that occurs in response to the road
investment. It is specified together with the road investment option that induces it, and is
specified within the relevant applications programs.
The existing traffic volumes on the road being analysed are specified in terms of vehicle type
or class, depending on the kind of analysis to be performed. The value entered for each
vehicle type is expressed as the annual average daily traffic (AADT):
This constitutes the baseline flow for the analysis period. It is assumed that seasonal variations
in traffic flows have already been accounted for when estimating the AADT from traffic
counts carried out over shorter periods.
For project analysis, traffic composition data are specified for each section. For network
level analysis, several representative sets of traffic composition data can be specified, and
each is assigned to a group of road sections with similar traffic characteristics.
In most situations, traffic growth has a major effect on the level of benefits obtained. Changes
in traffic composition over time can be modelled by specifying different growth rates for each
vehicle type/class. Uncertainties always exist in estimating initial traffic, but there is even
greater uncertainty in forecasting future growth rates. Thus, it is recommended that sensitivity
analyses are always undertaken to evaluate the effects of different forecast growth rates.
Traffic volumes are derived from the baseline AADT and the composition data. The
computation of annual road user effects requires the AADT for each road section to be broken
down by vehicle type. For normal traffic, the annual average daily traffic for each vehicle type
(AADTkn) is obtained by multiplying the normal traffic AADTn by the composition data
defined for the vehicle type k in the base year. For generated traffic, AADTkg is either
specified directly or calculated, based on AADTkn and the traffic growth type used. Volumes
for normal and generated traffic are required as separate inputs into the Road User Effects
(RUE) model to enable the economic benefits of generated traffic to be determined.
The specification of traffic composition data for each section is carried out for each traffic
category as described in Sections 4.2.1, 4.2.2 and 4.2.3.
This is specified in vehicles per day at the traffic start year. It is permissible and
sometimes convenient to enter AADT for some year before the start of the analysis
period. However, it must be ensured that the traffic start year is always less than or
equal to the start year of the analysis period.
! Initial composition
The proportion of each representative vehicle that uses the road, as a percentage of the
AADT. For example; medium cars 10%; articulated trucks 5%; etc. These data items are
used to calculate the actual volumes, by vehicle type, for the applicable year.
! Future traffic
This data item is specified as a traffic growth period defined in terms of its start year and
by one of the following growth types:
❏ annual percentage increase.
❏ annual incremental increase in AADT.
❏ actual AADT, which must be followed by another traffic growth period, unless the
year in which it applies is the last year of the analysis period.
Then, depending on the selected traffic growth type, the growth rate is specified for each
representative vehicle. For example, light trucks increasing at 4% per year from 1995 to 2000,
if annual percentage increase was selected.
Several traffic growth periods may be defined, each with a minimum length of one year. It is
important to ensure that the defined traffic growth periods cover consistently each and every
year of the analysis period. Therefore, if only one growth period has been defined, this will be
assumed to apply to all successive analysis years.
The calendar year in which traffic diversion starts, typically this should coincide with the
year following the completion of the road works.
! New AADT
The annual average daily traffic in the start year for each road section.
! New composition
The traffic composition in the start year for each road section.
! Future traffic
A number of years, after the start of the road works, that generated traffic appears. Thus,
relative start year j means the jth year following that year in which the works are started
that cause the generated traffic.
! Future traffic
1 Introduction
This document describes the specification of Road Works Standards used within HDM-4.
The standards include the road maintenance and improvement standards that are followed by
road administrations in their network management and development activities. The standards,
which are pre-defined by users and stored in the Road Works Standards folder, can
subsequently be selected for use with a particular HDM-4 analysis.
2.1 Definition
Standards refer to the targets or levels of conditions and response that a road
administration aims to achieve. Road agencies set up different standards that
can be applied in practical situations in order to meet specific objectives
which are related to functional characteristics of the road network system.
In HDM-4, a standard is defined by a set of operations or works activities with
definite intervention criteria to determine when to carry them out. In general
terms, intervention levels define the minimum level of service that is allowed.
A number of maintenance and improvement standards can be defined within the Road Works
Standards folder, and these can be assigned to physical road sections for project and
programme analysis, or to representative road sections for strategy analysis.
For a given feature type (that is, carriageway, shoulders, non-motorised traffic (NMT) lanes,
miscellaneous and special), only one maintenance standard and/or improvement type will be
effective in any analysis year.
3 Maintenance standards
A detailed discussion of road maintenance works, the intervention criteria that determine
when they should be applied, their design, effects on the road characteristics, and the costs of
implementing each works activity is given in Part D of the Analytical Framework and Model
Descriptions.
The following sections describe the specification of constituent works for a Maintenance
Standard.
! Intervention
These effects are for responsive interventions only. The effects are specified in terms of
the percentage of the following types of distress repaired, based on the operation
type/works activity. Examples of distress type repairs specified are given below for the
following works activities:
❏ crack sealing
- transverse thermal cracking
- wide structural cracking
❏ patching
- severely damaged area, wide structural cracking, potholing and ravelling
❏ edge repair
- edge break
! Intervention
Unit rates (economic and financial) for the operation specified under General, and
associated preparatory works.
! Effects
❏ pavement type
❏ surface material type
❏ thickness of new surfacing (mm)
❏ AASHTO strength coefficient during the dry season
❏ relative compaction
❏ base thickness (for stabilised base only)
❏ resilient modulus (for stabilised base only)
❏ construction details in terms of defect indicators for bituminous surfacing (CDS) and
base (CDB)
! Intervention
3.4.1 Grading
Grading can be scheduled to take place at a fixed time interval or after a fixed amount of
traffic; alternatively, grading can be triggered by a roughness threshold. The works to which
the standard applies are defined by the following:
! General
! General
❏ Either:
- volume of gravel to add (m3/km/year)
or
- percentage of annual gravel loss to replace (%)
! Intervention
3.4.3 Regravelling/resurfacing
Regravelling can be scheduled to take place at a fixed interval based on the age of the gravel,
or can be responsive to roughness and gravel thickness. The works to which the standard
applies are defined by the following:
! General
Economic and financial costs of regravelling (per m3 and associated preparatory spot
regravelling (per m3).
! Effects
❏ derived, or
❏ user defined - initial roughness, IRI, (m/km)
Specified for:
4 Improvement standards
❏ partial widening
❏ lane addition
❏ realignment of bituminous roads
❏ upgrading of bituminous roads
❏ pavement reconstruction
! Unsealed carriageways
❏ partial widening
❏ lane addition
❏ realignment of unsealed roads
❏ upgrading of unsealed roads
An improvement standard can be set up or amended by defining data under the following
headings:
! General
! Design
! Intervention
! Costs
! Pavement
! Geometry
! Effects
These are each described in turn for each of the works types in the sub-sections associated
with Sections 4.2 and 4.3.
General
The works to which the standard applies are defined by the following:
! Name
❏ partial widening
! Duration of the works (years)
! Intervention type
Design
The design of the works is specified in terms of the following:
! Speed-flow type
! Road class
! Pavement type
For example; asphalt mix on an asphalt base, surface treatment on a granular base, etc.
! Increase in carriageway width (m)
Dry (brittle) surfacing, with binder content nominally 10% below optimal: CDS = 0.5
Normal surfacing, with binder content at optimal value: CDS = 1.0
Rich (soft) surfacing, with binder content nominally 10% above optimal: CDS = 1.5
CDB ranges between 0 (no construction defects) and 1.5 (several defects). Defects
considered are: poor gradation of base material, poor aggregate shape and poor
compaction.
! Resurfacing
Check on whether resurfacing the existing carriageway is part of the standard and, in
those cases where it is, then the following is specified:
❏ thickness of the new surfacing (mm)
❏ patching per m2
❏ edge repairs per m2
❏ crack sealing per m2
Intervention
Partial widening can be on a scheduled based upon a calendar year. Alternatively,
intervention can be based on the following criteria:
! Speed
Costs
Costs are specified in terms of the following:
! Unit rates (per km)
Pavement
Details of the pavement to be provided under an improvement standard are specified with the
following data:
! Details of the surfacing layers
Geometry
The following data under geometry can be specified for partial widening:
2
! Acceleration noise due to driver and road alignment (m/s )
❏ due to non-motorised transport, with a value between 0.6 (significant reduction) and 1.0
(no reduction)
❏ due to motorised transport, with a value between 0.4 (significant reduction) and 1.0 (no
reduction)
❏ due to roadside friction, with a value between 0.6 (significant reduction) and 1.0 (no
reduction)
Effects
The effects of an improvement intervention define the condition of the road after the
intervention has been made:
! Change in traffic flow pattern
! Pavement condition after works have been carried out
These effects are specified in terms of the percentage of the following types of distress
repaired:
❏ Roughness, IRI, (m/km)
❏ mean rut depth (mm)
❏ skid resistance, (SCRIM at 50 km/h)
❏ surface texture (mm)
In addition, the pavement deterioration factors for the initiation and progression of defects and
other calibration factors can be specified as optional data.
A detailed discussion of road improvement works, the intervention criteria that determine
when they should be applied, their design, effects on the road characteristics, and the costs of
implementing each works activity is given in Part D of the Analytical Framework and Model
Descriptions.
General
The specification is the same as for partial widening (see Section 4.2.1) in all areas except
for design. (Note however, that with partial widening the increase in width is limited to the
range 0-3 metres.)
Design
The design of the lane addition works is specified in terms of the following:
! Speed-flow type
! Road class
! Pavement type
General
The specification of this is similar to that for partial widening (see Section 4.2.1), with the
exception of design and geometry.
Design
The design of realignment works is specified in terms of the following:
! Speed-flow type
! Road class
! Pavement type
! Length adjustment factor
! Proportion of new construction
The ratio of new construction length to the section length after realignment works, with a
value between zero and one.
Geometry
The following data under geometry can be specified for realignment:
! Rise plus fall (m/km)
! Superelevation (%)
2
! Acceleration noise (m/s ) - due to driver and road alignment
❏ due to non-motorised transport, with a value between 0.6 (significant reduction) and 1.0
(no reduction)
❏ due to motorised transport, with a value between 0.4 (significant reduction) and 1.0 (no
reduction)
❏ due to roadside friction, with a value between 0.6 (significant reduction) and 1.0 (no
reduction)
General
Upgraded pavements can be constructed in either bituminous or concrete materials, and the
specification of the upgrading standard reflects this. The specification for this is similar to
partial widening (see Section 4.2.1), with the exception of design, pavement, geometry and
effects.
Design
The design of the improvement works is specified in terms of the following:
! Speed-flow type
! Road class
! Pavement type
! Length adjustment factor
❏ shoulder type
❏ load transfer efficiency (%)
! Drainage
❏ edge drains
❏ drainage coefficient
Pavement
Depending on the road surface class resulting from the upgrading, the pavement details are
specified as follows:
! Bituminous
Geometry
The following data under geometry can be specified for upgrading:
! Rise plus fall (m/km)
! Superelevation (%)
2
! Acceleration noise (m/s ) - due to driver and road alignment
❏ due to non-motorised transport, with a value between 0.6 (significant reduction) and 1.0
(no reduction)
❏ due to motorised transport, with a value between 0.4 (significant reduction) and 1.0 (no
reduction)
❏ due to roadside friction, with a value between 0.6 (significant reduction) and 1.0 (no
reduction)
Effects
The effects of an improvement intervention define the condition of the road after the
intervention has been made.
Bituminous roads
The following effects are specified:
! Change in traffic flow pattern
! Pavement condition
These effects are specified after the works have been carried out, in terms of the
following types of distress:
❏ Roughness, IRI, (m/km)
❏ mean rut depth (mm)
❏ skid resistance, (SCRIM at 50 km/h)
❏ surface texture depth (mm)
In addition, the pavement deterioration factors for the initiation and progression of defects and
other calibration factors can be specified under Section/Details/Surface Distress.
Concrete pavements
The following effects must be specified:
! Change in traffic flow pattern
! Pavement condition
These effects are specified after the works have been carried out, in terms of the
following type of distress:
❏ Roughness, IRI, (m/km)
In addition, detailed calibration factors can be specified under Section/Details/Surface
Distress.
❏ Pavement Type
following the improvement (reconstruction)
❏ Increase in carriageway width (up to 3 metres)
! Pavement
❏ Surface material
❏ Dry season Adjusted Structural Number (excluding subgrade contribution)
❏ Surface thickness (mm)
❏ Relative compaction
General
The works to which the standard applies are defined by the following:
! Name - of the improvement works
! Short code
! Improvement type
❏ partial widening
! Duration of the works (years)
! Intervention type
Design
The design of the improvement works is specified in terms of the following:
! Speed-flow type
! Road class
! Pavement type
Intervention
Partial widening can be on a scheduled based upon a calendar year. Alternatively,
intervention criteria can be based on the following parameters:
! Daily average volume capacity ratio
! Peak period volume capacity ratio
! Mean speed
! Minimum speed
! Two-way AADT
For both scheduled and responsive works, the time of intervention can be limited by
specifying the last applicable year in which the works can take place.
Costs
Costs are specified in terms of the following:
! Unit rates (per km)
Pavement
The following data can be specified:
! Material properties
Geometry
The following data under road geometry can be specified:
2
! Acceleration noise (m/s ) due to driver and road alignment
❏ due to non-motorised transport, with a value between 0.6 (significant reduction) and 1.0
(no reduction)
❏ due to motorised transport, with a value between 0.4 (significant reduction) and 1.0 (no
reduction)
❏ due to roadside friction, with a value between 0.6 (significant reduction) and 1.0 (no
reduction)
Effects
The effects of an improvement intervention define the condition of the road after the
intervention has been made. The following effects are specified:
! Change in traffic flow pattern
! Pavement condition after works have been carried out
General
Specification of lane addition to unsealed roads is identical to that for partial widening of
unsealed roads (see Section 4.3.1), with the exception of the specification under design.
Design
The design of the lane addition works is specified in terms of the following:
! Speed-flow type
! Road class
! Pavement type
! Additional number of lanes
General
Specification of realignment of unsealed roads is identical to that for partial widening of
unsealed roads (see Section 4.3.1), with the exception of the specification under design and
geometry.
Design
The design of realigned unsealed roads is specified in terms of the following:
! Speed-flow type
! Road class
! Pavement type
! length adjustment factor
! Proportion of new construction
The ratio of new construction length to the section length after realignment works, with a
value between zero and one.
Geometry
The following data under geometry can be specified for realignment:
! Rise plus fall (m/km)
! Superelevation (%)
2
! Acceleration noise (m/s ) due to driver and road alignment
❏ due to non-motorised transport, with a value between 0.6 (significant reduction) and 1.0
(no reduction)
❏ due to motorised transport, with a value between 0.4 (significant reduction) and 1.0 (no
reduction)
❏ due to roadside friction, with a value between 0.6 (significant reduction) and 1.0 (no
reduction)
General
Specification of the upgrading of unsealed roads is similar to that for partial widening of
unsealed roads (see Section 4.3.1), but there are exceptions under design, intervention,
pavement, geometry and effects. These are described below:
Design
The design of the improvement works is specified in terms of the following:
! Speed-flow type
! Road class
! Pavement type
! Length adjustment factor
Depending on the type of road resulting from the upgrading, the construction details are
specified as follows:
! Bituminous roads
Intervention
Partial widening can be on a schedule based upon a calendar year. Alternatively, intervention
can be based on the following criteria:
! Speed
❏ two-way AADT
❏ cumulative equivalent standard axle loads (CESAL)
❏ Peak period volume-capacity ratio
❏ daily average volume-capacity ratio
Pavement
Depending on the road surface class resulting from the upgrading, the pavement details are
specified as follows:
! Bituminous
Geometry
The following data under geometry can be specified for upgrading:
! Rise plus fall (m/km)
! Superelevation (%)
2
! Acceleration noise (m/s ) - due to driver and road alignment
❏ due to non-motorised transport, with a value between 0.6 (significant reduction) and 1.0
(no reduction)
❏ due to motorised transport, with a value between 0.4 (significant reduction) and 1.0 (no
reduction)
❏ due to roadside friction, with a value between 0.6 (significant reduction) and 1.0 (no
reduction)
Effects
The effects of an improvement intervention define the condition of the road after the
intervention has been made, and depend on the type of road resulting from the upgrading.
Bituminous roads
The following effects are specified:
! Change in Traffic flow pattern
! Pavement condition after works have been carried out
Concrete pavements
The following effects are specified:
! Changes in road use
! Pavement condition after works have been carried out
Contents
Applications Guide i
Version 1.0
APPENDICES CONTENTS
Applications Guide ii
Version 1.0
PROJECT ANALYSIS CASE STUDIES
Applications Guide 1
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
Applications Guide 2
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
❏ Select Reports
It is recommended that the user keep a record of all input data in hard copy format, noting the
sources of all information. This will assist when input data is checked (for example, on
HDM-4 reports) and outputs reviewed, possibly by different personnel.
Applications Guide 3
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
3.1 General
This Tab page confirms the project title and type of analysis required - project analysis in this
case, as the task is to compare, for all three road sections, the upgrade alternative against the
do-minimum alternative.
The project start year has been defined as 2000 with the analysis period (duration) specified as
20 years. The road network to be used for the project has been pre-defined under the name
Northern Province in the Road Network folder. Similarly the vehicle fleet has been pre-
defined in the Vehicle Fleet folder under the same name. The road sections (within the
Northern Province road network) and vehicles types (from the Northern Province vehicle
fleet) to be used for the analysis are selected under the Select Sections and Select Vehicles
Tabs respectively, as discussed below:
Finally the General Tab page confirms that the input and output currencies are both US
dollars.
Applications Guide 4
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
Table A1.1
Case study 1 - Summary of section attributes
Number of lanes 2 2 2
Number of shoulders 0 0 0
Geometry (before
upgrading)
Rise and Fall (m/km) 10 15 30
Condition 1998
Gravel Thickness 150 150 150
Notes:
Applications Guide 5
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
The Select Sections Tab page displays those sections of the Northern Province road network
that will be included in the analysis. For this case study, the three sections listed should have
a tick in the Include column to confirm their selection.
Each section has been assigned a unique Description and ID, either or both of which may be
used to identify the section on HDM-4 reports. In this case study, the section description
identifies towns at the ends of each section. It is helpful if the Section ID includes reference
to the road class and road number (denoted by B001 in this case study).
By double-clicking on one section in the Select Section Tab page, the characteristics of that
section can be reviewed under four Tabs (Definition, Geometry, Pavement, and Condition).
The corresponding Tab pages are reviewed below for one of the sections from this case study
(Section Description: Town A to Town B).
! Section/Definition
Lists basic characteristics such as speed-flow type and traffic-flow pattern, also
carriageway length and width. The pavement type is defined as Gravel (HDM-4
pavement type GRUP, Granular Unsealed Pavement).
Section pavement details are stored under the Pavement Tab.
Applications Guide 6
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
! Section/Geometry
Applications Guide 7
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
❏ Section/Geometry/Alignment screen
! Pavement
❏ Section/Pavement screen
Gives pavement layer descriptions and year of last regravelling. The material properties
can be reviewed under Section/Details/Material Gradation.
Applications Guide 8
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
! Condition
Confirms gravel thickness 150 mm and roughness 8 m/km (IRI) in1998. Note that, to
enable road condition to be modelled through the analysis period, condition data must be
specified for a date prior to the start of the analysis period. In this case study the
condition data refer to the end of 1998, which is prior to the analysis start year 2000.
❏ Section/Condition screen
Applications Guide 9
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
The tick in the Include column confirms the selection. This list defines the range of vehicle
types that can be assigned to each road section, although the full range need not be used. The
assignment of traffic by section is effected under the Define Normal Traffic Tab, discussed
later.
Vehicle attributes for a particular vehicle type can be reviewed by double-clicking on the
appropriate vehicle type descriptions. The vehicle attributes are held under four Tabs:
Definition, Basic Characteristics, Economic Unit Costs, and Financial Unit Costs.
An example of each Tab page for one vehicle type is shown below:
! Vehicle Attributes/Definition screen
Applications Guide 10
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
Applications Guide 11
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
Note that the vehicle attributes are defined for a representative vehicle within each vehicle
class. For example, under Basic Characteristics, the loading details (No of equivalent
standard axles and operating weight) are based on the average weight of vehicles in the
vehicle class.
Applications Guide 12
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
By double-clicking on the row representing a specific section, the traffic composition in the
specified year (by % of each vehicle type) can be reviewed, together with vehicle growth
rates. Thus for section A-B, the normal traffic details are as given below:
! Define Normal Traffic/Motorised screen
For this case study, only one growth period is indicated over the 20-year analysis period.
Several growth periods may be defined within the analysis period. The user is encouraged to
define future growth rates (by vehicle type) as accurately as possible. Note that negative
growth rates are permitted.
Applications Guide 13
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
3.5 Alternatives
The Alternatives Tab is displayed after selecting the Specify Alternatives button.
The two project alternatives considered in this case study are defined below. The first
alternative, Without Project, represents a continuation of current maintenance practice. The
second alternative, With Project, represents the implementation of the project to upgrade the
existing gravel road to paved standard.
! Without Project (see Section 3.7)
This project alternative comprises grading every six months; spot regravelling to replace
30% of material lost each year (if the gravel thickness falls below 100 mm, and gravel
resurfacing (whenever the gravel thickness falls below 50 mm).
Note that in a given year, if gravel resurfacing is triggered, this operation supersedes
grading and spot regravelling.
! With Project (see Section 3.8)
This project alternative includes upgrading the road in the year 2004 (duration of works
is two years for road sections B001-01 and B001-03, and one year for section B001-02).
Before upgrading, the existing gravel road will be maintained, by grading, every six
months and spot regravelling to replace 30% of material lost each year (if the gravel
thickness falls below 100 mm). The latter works will maintain some protection of the
subgrade prior to upgrading.
After upgrading, the road will receive routine maintenance in the form of crack sealing (if
wide structural cracking reaches 5%) and patching (if the severely damaged area reaches
5%).
The analysis period is 20 years, starting in the year 2000, and analysis is by project as
specified under the General Tab.
The Alternatives Tab page is split into two boxes. The upper box shows the names of the two
project alternatives that have been set up for this case study.
Applications Guide 14
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
The bottom box shows details, by section, of the Road Works Standards (assignments)
associated with each project alternative. Note that the assignments box shows the Road
Works Standards corresponding to the selected project alternative and section. For the
selected alternative, the Road Works Standards assigned to a different section can be reviewed
by selecting that section.
A schematic overview of the Project Alternatives and their constituent Road Works Standards
is shown in Figure A1.1. Note that Road Works Standards are sub-divided into Maintenance
Standards denoted by M , and Improvement Standards denoted by I .
The Road Works Standards and associated works assigned to each project alternative are
summarised in Table A1.2. A summary of the intervention limits for maintenance works is
given in Table A1.3. These Tables allow the user to check consistency of data across
alternatives.
Each Road Works Standard has an associated date, assigned at project level, which represents
the year from which the standard takes effect. For the Without Project alternative, each of
the three road sections has been assigned a Maintenance Standard named Gravel Road
Maintenance. This assignment commences in the year 2000, which is the first year of the 20-
year analysis period (2000-2019).
For the With Project alternative, each section has been assigned a set of Road Works
Standards as follows:
! M Maintenance Standard for gravel road prior to upgrading
! I Improvement Standard representing the upgrading works
Applications Guide 15
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
Figure A1.1 Case Study 1: Overview of project alternatives and road work
standards
Applications Guide 16
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
Table A1.2
Case Study 1: Details of road works standards for each project alternative
Applications Guide 17
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
Table A1.3
Case study 1: Intervention limits for maintenance works
M Gravel Road 2000 Grading S 2019 30 n/a 180 days 750 days 0 100,000
Maintenance
Spot regravelling R 2019 30 100 m3/km/yr n/a n/a 0 100,000
M Maintenance before 2000 Grading S 2003 30 n/a 180 days 10000 days 0 100,000
upgrading
Spot regravelling R 2003 30 100 m3/km/yr n/a n/a 0 100,000
M Crack sealing and 2006 Crack sealing R 2019 12.5 1500 n/a n/a 0 100,000
patching paved road
m2/km/yr
Notes:
n/a not applicable
* S = Scheduled intervention
R = Responsive intervention
Applications Guide 18
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
Project Alternative:
Without Project
M
Gravel Road Maintenance
Project Alternative:
With Project
M M
I
Maintenance before Crack sealing and
Upgrading works
upgrading patching paved roads
Figure A1.2 Schematic diagram showing details of Project Alternatives for Case
Study 1
Applications Guide 19
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
By double-clicking on Gravel Road Maintenance, the window for the specification of this
maintenance standard is displayed.
Applications Guide 20
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
This window shows the names and codes assigned to the Gravel Road Maintenance
Maintenance Standard and constituent Works Items. These names and codes appear on
certain HDM-4 reports so it is essential that they are unique within a Maintenance Standard.
The software checks for unique names and codes when Maintenance Standards are set up.
The window lists the three works items that are associated with the Gravel Road
Maintenance Maintenance Standard. As previously described, the Without Project
alternative applies grading every six months, spot regravelling to replace 30% material lost
each year (if the gravel thickness falls below 100 mm), and gravel resurfacing (if the gravel
thickness falls below 50 mm). The original gravel thickness was 150 mm (1998).
Note that in a given year, spot regravelling and grading are both permitted as these are
complimentary works. However if gravel resurfacing is triggered, this will supersede the
other works. Hence in a given year, the possible works under the Gravel Road Maintenance
Maintenance Standard are:
! Grading every six months, or
! Gravel resurfacing
The Works Items and Operations associated with the Maintenance Standard are shown in
Figure A1.3. Full details of the logical rules associated with the implementation of works are
given in Chapter D1 of the Analytical Framework and Model Descriptions.
Regravelling/
Operation/Works Activity: Grading Spot regravelling
Resurfacing
Note that, when setting up a Maintenance Standard, the user defines the name and code for the
Maintenance Standard and Maintenance Works. The name of the Operation/Works Activity
is selected from a pre-defined list. It is useful if the name/code assigned to the Works Item
also includes information on intervention (for example, grading every six months). It may
also be useful if the name given to the Maintenance Standard refers to the surface type (for
example, Gravel Road Maintenance).
Applications Guide 21
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
! Grading/General
The General Tab page for Works Item Grading is shown below:
This confirms the Name and Short Code of the Works Item. The operation type/works
activity (specified from a pre-defined list) is confirmed as grading, and intervention type
is scheduled.
! Grading/Design
The Design Tab page for grading specifies the compaction method as mechanical.
Applications Guide 22
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
! Grading/Intervention
The Intervention Tab page confirms the frequency at which the grading works will be
undertaken (that is, 180 days).
In general, scheduled grading works are timed from the last regravel year, specified
under Section/Pavement (1998 in this case study). However the works can only be
implemented after the date when the Maintenance Standard becomes effective (specified
at project level, 2000 for this case study).
Note that for scheduled works, the frequency of works can also be specified in terms of
traffic interval (that is, number of vehicles between successive gradings).
The Intervention screen also shows limits outside which the grading works would not
be triggered:
❏ Last year
The last year in which grading should be considered. For this Maintenance Standard
the last year for grading is set at 2019 (the last year in the analysis period).
This parameter could be used to ensure that a given works is not triggered too close
to the end of the analysis period.
❏ Maximum roughness
Sets the maximum roughness at which grading would be undertaken (30 m/km IRI).
❏ Interval
The minimum interval between successive grading operations has been set at 180
days (six months).
The maximum interval between successive grading operations, above which grading
will not be implemented, has been specified as 750 days (just over two years) to
allow for the fact that grading will not be triggered in a year when gravel resurfacing
Applications Guide 23
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
The unit costs for grading works (expressed in US dollars per km) can be reviewed under
the Costs Tab. Note that default costs for all works operations are specified under the
HDM Workspace option Road Works Standards/Default Works Costs & Energy
Consumption.
By clicking the OK button (on the Costs screen), the user is returned to the entry screen
for the Maintenance Standard Gravel Road Maintenance.
Applications Guide 24
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
! Spot regravelling/General
The General Tab page for the Works Item Spot regravelling is shown below:
This confirms the Name and Short Code of the Works Item. The operation type/works
activity is spot regravelling, with intervention type responsive.
! Spot regravelling/Design
Two design options for spot regravelling are given under the Design Tab.
Applications Guide 25
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
Either a specified volume of gravel can be added each year (specified as m3/km/yr), or a
specified percentage of the annual gravel loss is replaced (for example, 100% would
replace all gravel lost). For this case study the latter design option is adopted, replacing
30% gravel lost each year. Note that the amount of gravel actually applied to the road
section will be subject to user defined limits specified under Intervention.
! Spot regravelling/Intervention
The Intervention Tab page gives the responsive criterion that has been specified for spot
regravelling (that is, whenever gravel thickness falls below 100 mm). Note that with
some works activities, more than one intervention criterion can be specified; however
with spot regravelling, gravel thickness is the only parameter that is available.
The Intervention Tab page also gives limits affecting the spot regravelling works:
❏ Last year
The last year in which spot regravelling should be considered. For this case study,
the last year is set at 2019 (last year of analysis period).
❏ Maximum roughness
Sets the maximum roughness at which spot regravelling would be undertaken (30
m/km IRI).
❏ Maximum quantity
Sets an upper limit on the quantity of material to be used each year for spot
regravelling, specified as m3/km/year. This could reflect the upper limit considered
practicable before complete resurfacing should be considered, but should also take
account of availability of material and work capacity of the Roads Authority. For
this case study, the quantity specified is 100 m3/km/year.
Applications Guide 26
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
❏ AADT
This allows the user to specify the AADT range over which the works are
considered appropriate. For this case study, the range is specified as 0-100,000 to
avoid elimination of grading based on this parameter.
! Spot regravelling/Costs
The unit assigned to spot regravelling (expressed in US Dollars per cubic metre) can be
reviewed via the Costs Tab.
Applications Guide 27
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
! Gravel Resurfacing/General
The General Tab page, shown below, confirms that the Operation Type/Works Activity
is Regravelling/Resurfacing and that the Intervention Type is responsive (the intervention
criterion is defined under the Intervention Tab).
Applications Guide 28
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
! Gravel Resurfacing/Design
The Design Tab page for Gravel Resurfacing gives material properties (particle size
distribution and Plasticity Index) of the proposed surfacing material. In addition, the
final gravel thickness (or increase in gravel thickness) and compaction method are
specified. For this case study, a final gravel thickness of 150 mm is specified, using
mechanical compaction.
Applications Guide 29
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
! Gravel Resurfacing/Intervention
For this case study, the intention is to resurface only if the gravel thickness falls below 50
mm. This is reflected in the intervention criterion adopted. The implementation of
resurfacing is subject to the user-defined limits indicated under the Intervention Tab,
shown below:
Applications Guide 30
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
Note that, for gravel resurfacing, roughness could also be used as the intervention
criterion (for example, if the roughness is 15 m/km IRI or greater). Both parameters
(layer thickness and roughness) could be used together., If more than one criterion is
used with a given works item, a logical AND is assumed, and displayed on-screen (for
example, if gravel thickness falls below 50 mm AND roughness is 15 m/km IRI or
greater).
If the user wishes to trigger gravel resurfacing based on either layer thickness OR
roughness (logical OR), then two separate Works Items for gravel resurfacing would
be set up under the same Maintenance Standard with the appropriate intervention
criteria.
To summarise, examples of intervention options for responsive gravel resurfacing
are:
Intervention is based on:
! Gravel Thickness only
Applications Guide 31
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
The Intervention Tab page also shows the limits on the use of gravel resurfacing:
❏ Last year
The last year in which gravel resurfacing would be considered, set at 2017, year 18
of the 20 year analysis period.
❏ Maximum roughness
Sets the maximum roughness at which gravel resurfacing would be undertaken (30
m/km IRI).
❏ Interval
The minimum interval between successive resurfacings has been specified as three
years for this case study.
The user must also specify the maximum interval between successive resurfacing
operations, above which resurfacing will not be considered. As resurfacing is the
highest ranking maintenance operation available under this Maintenance Standard, a
value of 99 years is used to avoid elimination by this limit.
❏ AADT
For this case study, no limits on AADT are imposed (that is, minimum and
maximum values set at 0 and 100,000 respectively). In this case the minimum
interval specified above will override.
! Gravel Resurfacing/Costs
The unit cost assigned to gravel resurfacing (expressed as US Dollars per cubic metre)
can be reviewed under the Costs Tab.
Applications Guide 32
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
! Gravel Resurfacing/Effects
Note the Effects Tab with this Works Item, which gives the initial roughness value
immediately after the gravel resurfacing works are completed.
A user defined value of 3 m/km IRI is used for this case study.
The model calibration factors relating to gravel loss can be reviewed under
Section/Details. Details of the modelling logic for unsealed roads are given in Chapter
D4 of the Analytical Framework and Model Descriptions.
Applications Guide 33
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
Applications Guide 34
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
! Grading/Design
! Grading/Design
Applications Guide 35
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
! Grading/Costs
The intervention limits associated with the grading works are given on the Intervention
screen, and summarised in Table A1.3. Note that the Last Year is specified as 2003, the year
before the start of the upgrading works.
Applications Guide 36
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
! Spot regravelling/General
! Spot regravelling/Design
Applications Guide 37
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
! Spot regravelling/Intervention
! Spot regravelling/Costs
The intervention limits associated with the spot regravelling works are given on the
Intervention Tab page, and summarised in Table A1.3.
Applications Guide 38
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
Table A1.4
Section details before and after upgrading works
A 7 6 8
(excluding subgrade)
(US dollars/km)
Notes:
B = Before upgrading
A =After upgrading
Applications Guide 39
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
Confirms the improvement type upgrading, with duration of two years, and scheduled
intervention.
! Design
Applications Guide 40
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
The length adjustment factor is specified as 1.0 as no change in length is planned (that is,
no realignment). Similarly no increase in width is proposed.
Applications Guide 41
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
Gravel Surfacing 150mm max H3 = 150mm Imported granular sub-base (CBR 60%)
Subgrade Subgrade
Before Upgrading After Upgrading
Design: Pavement Type = STGB (Surface Treatment on Granular Base)
Pavement: Surface Material = Double Bitumen Surface Dressing
Dry Season Structural Number = SN = 1.84
Surface Thickness = 25mm
H2 and H3 not used directly by HDM-4 but included in costs
and contribute to SN
Figure A1.4 Case Study 1: Pavement design for Section B001-01 (Town A to Town B)
Applications Guide 42
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
The Construction Quality Indicators (CDS and CDB) are specified under Design/Edit
Construction Details. For this Improvement Standard, CDS = 1 and CDB = 0. Further
details of these parameters are given in Chapters C2 and D2 of the Analytical Framework and
Model Descriptions.
! Intervention
This confirms the scheduled implementation date as the year 2004. The Effective from
year assigned to the Improvement Standard (at project level) is also specified as 2004.
Note that the Effective from year could be specified differently to the implementation
year. For example, if no maintenance was needed for the year prior to the start of
improvement works, then the assignment (effective from) dates for standards would be:
M Maintenance before upgrading 2000 (start of analysis period)
(Note: Improvement Standard effective from 2003, works implementation is 2004. The
Improvement Standard takes precedence over the previous Maintenance Standard, so no
works would be triggered in 2003).
Applications Guide 43
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
! Costs
The costs (expressed in US Dollars per km) are spread across the two-year construction
period specified under the General Tab. 50% costs have been assigned to the year 2004,
and 50% to 2005.
! Pavement
This Tab page is used to specify details of the new pavement after upgrading.
Applications Guide 44
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
❏ Dry Season Structural Number, SN, is derived for the pavement layers indicated in
Figure A1.4 (that is, surface, granular base and sTabilised sub-base). Note that SN
does not include a contribution from the subgrade.
For road section B001-01, the value of SN has been derived from layer thicknesses
and coefficients as indicated below:
(mm)
Surfacing ST 25 0.2
where:
Recommended values for layer coefficients are given in Chapters C2 and D2 of the
Analytical Framework and Model Descriptions, which also gives full details of the
HDM-4 definitions relating to Structural Number.
Note that roadbase details are not requested as the proposed roadbase is not
sTabilised.
❏ The surface thickness refers to the thickness of the new surfacing that is 25 mm for
this road section (Double bitumen surface dressing).
❏ Relative compaction of pavement layers, taken as 97% on average.
Applications Guide 45
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
! Geometry
In this case study, the effect of the upgrading works on Geometry is summarised in
Table A1.4. There are no changes to the horizontal or vertical alignment. Note that the
posted speed limit (after upgrading) is 120 km/h, compared with 100 km/h before
upgrading.
These data were based on the data held for the gravel road prior to upgrading. Note that
the geometry data before upgrading is held in the Road Network folder under
Section/Geometry and Section/Details/Alignment.
! Effects
The screen confirms that the road condition after the improvement works will be derived
by the HDM-4 Works Effects model. The calibration factors can be reviewed under the
Applications Guide 46
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
Edit Detailed Calibration button. Further details of calibration of Road Works Effects
are given in Part D of the Analytical Framework and Model Descriptions.
Confirms the improvement type upgrading, with duration of one year, and scheduled
intervention.
Applications Guide 47
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
! Design
Confirms the Pavement Type after improvement as Surface Treatment on Granular Base
(STGB). The pavement design is described under Pavement below. The factors CDS
and CDB can be reviewed under Design/Edit Construction Details.
! Intervention
The construction period for this section is one year, and so all works costs are allocated
to the year 2004.
Applications Guide 48
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
! Pavement
The pavement design for the upgrading of Section B001-02 is given in Figure A1.5. For
this road section, the intention is to remove the existing gravel surfacing to formation
level, then sTabilise the existing subgrade material (to a depth of 150 mm) below
formation. This improved subgrade will be considered (for HDM-4 purposes) as a
sTabilised sub-base on top of which a granular roadbase layer (150 mm thick) will be
placed. A double bitumen surface dressing (thickness 25 mm) will be applied as
surfacing.
This design may be appropriate where the existing subgrade material is generally weak
but is suiTable (in terms of plasticity and particle size distribution) for strengthening by
the addition of a lime or cement sTabiliser. Note that sTabilisation would normally be
considered only if the cost of the process is less than the cost of replacing the existing
weak material with imported suiTable material.
The structural number, SN is derived for the pavement layers indicated in Figure A1.5
(that is, surface, granular base and sTabilised sub-base). Note that SN does not include a
contribution from the subgrade.
For road section B001-02, the value of SN has been derived from layer thicknesses and
coefficients as indicated below:
(mm)
Surfacing ST 25 0.2
Applications Guide 49
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
where:
Recommended values for layer coefficients are given in Chapters C2 and D2 of the
Analytical Framework and Model Descriptions, which also gives full details of the
HDM-4 definitions relating to Structural Number.
! Geometry
The geometric characteristics of road section B001-02 after upgrading are shown on the
Geometry Tab page. Note that the posted speed limit is changed from 80 km/h (before
upgrading) to 100 km/h.
! Effects
The condition after the upgrading works is derived by the HDM-4 Works Effects models.
Applications Guide 50
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
Gravel Surfacing 150mm max H2 = 150mm Imported granular roadbase (CBR 60%)
Subgrade
H3 = 150mm Cement sTabilised sub-base (CBR 60%)
Subgrade
Figure A1.5 Case Study 1: Pavement Design for Section B001-02 (Town B to Town C)
Applications Guide 51
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
Name and Short Code refer to Section B001-03. The construction period for the
proposed improvement works is two years.
! Design
Confirms the pavement type after improvement as Surface Treatment on STabilised Base
(STSB). The pavement design is described under Pavement below.
! Intervention
As specified with 50% works costs incurred in each of the construction years 2004 and
2005.
Applications Guide 52
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
! Pavement
For this road section, the proposed pavement design comprises a 150 mm sTabilised sub-
base placed on existing formation level with a 200 mm sTabilised road base and a double
surface dressing (Figure A1.6).
For road section B001-03, the value of SN has been derived from layer thicknesses and
coefficients as indicated below:
(mm)
Surfacing ST 25 0.2
where:
Applications Guide 53
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
Recommended values for layer coefficients are given in Chapters C2 and D2 of the
Analytical Framework and Model Descriptions, which also gives full details of the
HDM-4 definitions relating to Structural Number.
The layer coefficients are representative of the as new condition of the pavement. It is
recognised that the sTabilised layers will develop cracking. Details of the deterioration
models for sTabilised layers are given in the Analytical Framework and Model
Descriptions.
The proposed design includes a sTabilised sub-base and sTabilised base. As the roadbase
is sTabilised, its thickness (200 mm) and Resilient Modulus (15 GPa) are requested.
HDM-4 does not require these parameters for the sTabilised sub-base.
Note that use of a sTabilised roadbase gives a significant risk of reflection cracking
through the surfacing. The risk is effectively reduced if a granular layer is placed above
the sTabilised layer (for example, as for section B001-02 using a sTabilised sub-base and
granular roadbase).
! Geometry
The geometric characteristics of road section B001-03 after upgrading are shown on the
Geometry Tab page. Note that the posted speed limit is changed from 100 km/h (before
upgrading) to 120 km/h.
! Effects
The condition after the upgrading works is derived by the HDM-4 Works Effects models.
Applications Guide 54
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
Gravel Surfacing 150mm max H3 = 150mm Cement sTabilised sub-base (CBR 60%)
Subgrade
Before Upgrading After Upgrading
Figure A1.6 Case Study 1: Pavement design for Section B001-03 (Town C to Town D)
Applications Guide 55
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
Applications Guide 56
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
! General
This Tab page confirms the Name and Code assigned to this Works Item. The works
activity is specified as crack sealing and intervention type is responsive.
! Intervention
The Intervention Tab page confirms the responsive criterion (Wide structural cracking
affecting 5% or more of carriageway area over the section). The other criterion available
is based on the number of transverse thermal cracks per km).
Applications Guide 57
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
! Costs
The Costs Tab confirms the unit cost of the crack sealing works, (expressed in US
Dollars per square metre).
! Effects
! General
Applications Guide 58
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
This Tab page confirms the Name and Code assigned to this Works Item. The works
activity is specified as patching and intervention type is responsive.
! Intervention
The Intervention Tab page confirms the responsive criterion (when 5% of section
carriageway area is severely damaged). Note that other criteria could be defined, based
on potholing, ravelling, or wide structural cracking.
The user defined limits are:
! Costs
The Costs Tab confirms the unit cost of the patching works, (expressed in US Dollars per
square metre)
Applications Guide 59
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
! Effects
The Effects Tab confirms the percentage of distress to be repaired in terms of Potholing
(set at 100%). This is subject to the limits defined under Intervention above.
Note that the user may select one of the radio buttons shown. For this case study, the
responsive criterion is based on the number of potholes per kilometre, and this is
intended to trigger pothole patching.
Further details of road works effects are given in Chapter D2 of the Analytical
Framework and Model Descriptions.
Applications Guide 60
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
The costs and benefits of the Upgrade gravel road alternative will be compared with those
for the Maintain gravel road alternative, as defined under
Specify Alternatives/Alternatives (Section 3.5). Note that the HDM-4 Workspace may hold
data for any number of projects with associated network, fleet and work standards data. The
run that has been set up will consider only those project alternatives defined under
'Alternatives', with the specified selected sections and vehicles.
For this case study, accident costs, energy balance analysis, emissions calculations, and
acceleration effects are not included in the analysis.
Applications Guide 61
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
The user may then select Generate Reports to display the folders holding pre-defined report
options. The pre-defined report categories available with the current program release are:
! Deterioration/Work Effects
! Road User Effects
! Environmental Effects
! Cost Streams
For each project alternative, this report lists, by analysis year, the works description,
quantity and costs for each road section. A summary of total annual economic costs is
provided for each project alternative.
2 Timing of Works (by section)
This report lists similar details by section together with the summary of total annual
economic costs.
For this case study, the Timing of Works (by section) report is included in the Reports section
at the end of this chapter.
Considering the Without Project alternative, the report indicates that, at Section B001-01
(Town A-Town B), gravel resurfacing would be implemented in the years 2003, 2007, 2010,
2013 and 2016 (with associated preparatory spot regravelling). Spot regravelling would be
implemented every year except the first year and each year immediately following gravel
resurfacing. Grading is implemented every six months as scheduled. These works are
specified under the Maintenance Standard Gravel Road Maintenance described previously.
The With Project alternative for the same road section (B001-01) confirms that the upgrade
to paved standard would be implemented during 2004 and 2005. Before upgrading, the
existing road is graded at six monthly intervals between 2000 and 2003 with spot regravelling.
After upgrading, the paved road has no works specified until 2016, when patching is listed
each year between 2016 and 2019 (end of analysis period). These works are specified under
the three works standards:
! M Maintenance before upgrading
! I Pave Section B001-01 in 2004
! M Crack sealing and patching paved road
Applications Guide 62
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
The effect of these works on roughness is indicated in the Deterioration report and associated
chart showing roughness at the end of the year.
This plot shows the roughness progression during the analysis period for each project
alternative. Comparison with the Timing of Works (by section) report indicates that, before
upgrading, the regravelling works (listed as gravel resurfacing) have a significant impact on
roughness. Regravelling has been triggered when the roughness reaches a level of about 16 or
17 IRI. Note that intervention (regravelling) was requested if the thickness of gravel surfacing
was reduced to 50 mm.
After upgrading section B001-01, the roughness increases less dramatically. Patching is
triggered between 2016 to 2019, towards the end of the analysis period.
Note that the Improvement Standard specification included the option to define the effects of
the upgrading works in terms of roughness, mean rut depth, skid resistance and surface texture
(under the Effects Tab). For this case study, these values are derived by the HDM-4 Works
Effects model. An overview of the HDM-4 modelling logic is given in Chapter A1; full
details are given in Chapter D4 of the Analytical Framework and Model Descriptions.
Applications Guide 63
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A1 - PROJECT ANALYSIS CASE STUDY 1
(millions of US Dollars)
B001-01 Town A to Town B 0.62
Note that the traffic levels are lowest for Section B001-02 (negative NPV), also that
maximum benefits are derived for Section B001-03 which has highest traffic flow. A revised
project comprising sections B001-01 and B001-03 only would give an overall NPV of 2.08
(millions of US Dollars).
Applications Guide 64
Version 1.0
PROJECT ANALYSIS CASE STUDIES
Applications Guide 1
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
! Specify Alternatives
1.2.1 General
This Tab page confirms the project description, analysis type, analysis period and the pre-
defined Road Network and Vehicle Fleet.
This case study is presented as a section analysis. The road under study is represented by one
section, and the different rehabilitation proposals represent section alternatives. The road
section will be selected from the Northern Province network, stored in the Road Network
folder. The vehicles using the road will be selected from the Northern Province vehicle fleet,
stored in the Vehicle Fleet folder.
It is also possible to conduct the analysis by Project although we have only one road section.
The analysis by Project option would combine the results of selected sections into one
Project.
The analysis period is defined by a start year 2000, and a duration 20 years, (that is, 2000 -
2019).
Applications Guide 2
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
This Tab page indicates that one section will be included in the analysis. By double-clicking
on the Section Description reveals the Definition/Geometry/Pavement/Condition Tabs
where the section details are held.
! Definition
The Definition Tab gives details of basic section characteristics including road class,
speed-flow type and traffic flow pattern. The Pavement Type has been specified as
Surface Treatment on Asphalt Pavement (STAP). It is important that the correct
Pavement Type is specified, as the deterioration relationships held in HDM-4 are defined
by Pavement Type (see Table A2.1).
Details of the existing pavement are discussed under Pavement below:
Applications Guide 3
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
! Geometry
Details of the road section's geometry, including horizontal and vertical alignment, are
held under Section/Geometry and Section/Details/Alignment. The data indicates that
the road section under study is in rolling terrain (Rise + Fall = 20 m/km) and subject to a
speed limit of 100 kph.
Alignment data can be assessed from the existing mapping, if available, or measured by a
topographical survey.
Applications Guide 4
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
! Pavement
Note that the specification of Pavement Type (on the Section/Definition page) refers to
the current pavement construction and should be compatible with the Previous Works
details specified on the Pavement Tab page.
The Pavement Type is automatically updated immediately after any maintenance works.
The definition of bituminous Pavement Types based on surface and base types is given in
Table A2.1. (Table A2.2 contains descriptions of surface and base materials.) A
summary of Pavement Type resets after maintenance works is given in Table A2.3.
For the road section under study, the changes in Pavement Type since the date of the last
construction (1988) are summarised in Figure A2.1.
The last reconstruction was 1988, given by pavement type AMGB. After the application
of overlay in 1992, Table A2.3 shows that:
After surface dressing (reseal) in 1995 (last resurfacing), Table A2.3 gives:
Applications Guide 5
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
Figure A2.1 Case Study 2: Details of existing pavements and pavement type
resets
Applications Guide 6
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
Table A2.1
Generic HDM-4 bituminous pavement types
Note: AM and ST surfacings on concrete pavements, that is, AMCP and STCP, are
modelled in HDM-4 as concrete pavement types in the rigid pavement sub-model.
The abbreviations in Table A2.1 are described in Table A2.2.
Applications Guide 7
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
Table A2.2
Descriptions of surface and base materials
Applications Guide 8
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
Table A2.3
Pavement type resets after maintenance works
Mill & replace **AP **AP **AP **AP N/A **SB **AB **AP
to intermediate
surface layer
Mill & replace **GB **SB **AB **AP **GB **SB **AB **AP
to base
Notes:
1 The pavement type depends on the critical thickness (Hmin) of the existing
bituminous surfacing that is user-definable in HDM Configuration.
** Indicates that these two characters are dependent on the specific works activity (or
operation) and the surface material.
N/A Not applicable.
The Pavement Tab page shows the current pavement strength, defined by the Adjusted
Structural Number of the pavement (Adjusted SNP). The Adjusted SNP incorporates a
weighting factor that reduces the contribution from the sub-base and subgrade, as
described in the Analytical Framework and Model Descriptions. (Note that the Adjusted
SNP is not the same as the Modified Structural Number used in HDM-III).
The following HDM-4 conventions should be noted:
1 SNP refers to the full structural number including contributions from the sub-base
and subgrade.
2 SN refers to the structural number comprising contributions from the surfacing,
roadbase and sub-base only.
For this case study, the Structural Number (SN) was entered directly together with the
subgrade CBR using Option (1) on the Section/Pavement screen. By clicking Option 1,
the calculated SNP value is displayed at the top of the screen.
Applications Guide 9
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
Table A2.4
Ranking of road works applicable to the carriageway
Inlay 11 per m2
Edge-repair1 22 per m2
Routine
Patching1 22 per m2
Pavement
Crack sealing1 22 per m2
Note that Option (2) allows the user to calculate the SNP value with the help of the SNP
Calculation Wizard, which can calculate SNP based on deflections or layer thickness
and coefficients.
! Condition
The condition of the road in 1998 can be reviewed under the Condition Tab page.
Applications Guide 10
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
For project analysis, this data should preferably be collected as part of a detailed
condition survey. However, users can set up aggregate data specifying detailed values
under HDM Configuration. Default values are shown in Chapter D2.
Applications Guide 11
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
1.2.5 Alternatives
The Alternatives Tab is displayed after selecting the Specify Alternatives button. The
Alternatives screen is split into two boxes. The upper box shows the names of the six section
alternatives that have been set up for this case study.
The bottom box shows details of the Road Works Standards (assignments) associated with
each section alternative.
The six section alternatives considered for this case study are defined below. The analysis
period is 20 years (from year 2000 - 2019).
Applications Guide 12
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
Alternative Description
Applications Guide 13
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
4 Mill 75 mm and replace 75 mm if Roughness >= 6 IRI AND Cracking >= 20%
carriageway area
5 Mill 75 mm and replace 100 mm if Roughness >= 6 IRI AND Cracking >= 20%
carriageway area
Note that each alternative includes routine pavement maintenance in the form of:
! Crack sealing if area of wide structural cracking > = 5%
! Patching if Severely Damaged Area >= 10%
The Road Works Standards and associated works assigned to each alternative are summarised
in Table A2.5. In this case study, each alternative is defined by one Maintenance Standard.
Note that an identical set of routine pavement works (crack sealing and patching) is common
to each Maintenance Standard. To be considered for implementation during any analysis
year, the routine works and rehabilitation works must be specified within the same
Maintenance Standard. (If routine works and rehabilitation works are specified as two
separate Maintenance Standards with the same effective from date, the software accepts only
one standard, that is, that which is assigned to the alternative first).
The intervention limits for the Works Items are summarised in Table A2.6.
Applications Guide 14
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
Table A2.5
Case Study 2: Details of road works standards for each project alternative
Applications Guide 15
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
Table A2.6
Case Study 2: Intervention limits for maintenance works
Notes:
n/a = not applicable; * S = Scheduled intervention, R = Responsive intervention
Applications Guide 16
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
Note that for certain works items, the unit costs of preparatory works are entered separately on
the Works Item/Costs Tab page. In Case Study 2, this applies to the following works items.
Patching
50 mm overlay
Edge repair
3 Patching
Crack Sealing
Patching
Crack Sealing
Where HDM-4 does not offer a separate facility for entry of preparatory costs, such costs (if
any) should be included in the costs of the Works Item.
Issues relating to the Maintenance Works Standard for each alternative are discussed below:
Applications Guide 17
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
The works included in this Maintenance Standard are the routine pavement works (crack
sealing and patching) and reconstruction. The works are listed in the opening screen for
the Maintenance Standard:
The specification for each of these works can be reviewed by double-clicking the
appropriate works item name.
❏ Works Item: Crack Sealing and Patching
The specifications for these works, included in all alternatives for this case study, are
similar to those described for Case Study 1 within the Maintenance Standard Crack
Sealing and Patching Paved Road (see Appendix A1).
❏ Works Item: Reconstruct at IRI 12 and 40% damage
The specification for Reconstruction is held under the Tabs:
General/Design/Intervention/Costs/Effects
Note that the Works Activity (Pavement Reconstruction) is specified under General.
The Design Tab shows the proposed new Pavement Type as Asphalt Mix on
Granular Base (AMGB) together with pavement details required by HDM-4.
Note that the details under Design refer to the new pavement construction. The
requested structural number refers to the layers above the subgrade.
The reconstruction Costs to be specified include all costs associated with the works.
The Tab pages specifying the Reconstruction Works Item are shown below:
Applications Guide 18
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
Applications Guide 19
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
Applications Guide 20
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
The works included in the Maintenance Standard are the routine pavement works (crack
sealing and patching) and a 50 mm thick overlay (condition responsive, based on IRI).
The works are listed in the opening screen for the Maintenance Standard:
The specification for the overlay works can be reviewed by double-clicking the
appropriate works item name.
Note that two works items for overlaying are specified. Both are condition responsive,
one based on roughness, the other on structural cracking. This means that either
roughness OR structural cracking can trigger overlaying. The Works Items are identical
Applications Guide 21
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
except for the Intervention Tab page. The Works Item based on roughness is reviewed
below:
❏ Maintenance Works: Overlay 50 mm at IRI 6
The specification for the Overlay works can be reviewed under the Tabs:
General/Design/Intervention/Costs/Effects
The Works Activity (overlay dense-graded asphalt) is specified under General. The
overlay material, thickness and dry season strength coefficient are specified under
Design. The Construction defect indicators for the bituminous overlay (CDS) is also
specified.
Note that if overlay is implemented, the Pavement Type will be automatically reset
according to Table A2.3:
The Tab pages specifying the Overlay Works are reviewed below:
Applications Guide 22
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
Applications Guide 23
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
Applications Guide 24
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
The works included in this Maintenance Standard are the same as for Alternative 2 plus
an extra works item, Surface Dressing (SD).
The specification for the Surface Dressing Works Item can be reviewed by double-
clicking the appropriate works item name:
❏ Maintenance Works: Surface Dressing at SFC <= 0.4
The specification for Surface Dressing can be reviewed under the Tabs:
General/Design/Intervention/Costs/Effects
The Works Activity is specified as Surface Dressing Single (General Tab), with
responsive intervention based on the skid resistance (Intervention Tab). Note that a
double surface dressing is also available, and shape correction may be applied to
either single or double surface dressings. Shape correction adds to the cost, but has
a more beneficial effect in terms of roughness (details are given in Chapter D2 of the
Analytical Framework and Model Descriptions).
For this case study, intervention is based on the skid resistance. Additional criteria
that may be used with surface dressing are cracking, ravelling, texture depth, and
total damaged area.
Note that the unit costs of surface dressing works and preparatory works are
specified under Costs.
The pavement type (STAP) will be reset automatically according to Table A2.3 if
either the surface dressing or overlay works is implemented.
For example:
Applications Guide 25
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
The Tab pages specifying the Surface Dressing Works are given below:
Applications Guide 26
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
Applications Guide 27
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
In addition to routine pavement works (crack sealing and patching) this Maintenance
Standard includes Mill and Replace (condition responsive based on IRI).
The specification for Mill and Replace can be reviewed under the Tabs:
General/Design/Intervention/Costs/Effects
Applications Guide 28
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
The Tab pages specifying the Mill and Replace Works for Alternative 4 are
reviewed below:
Applications Guide 29
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
Applications Guide 30
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
Applications Guide 31
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
This Maintenance Standard is similar to that specified for Alternative 4, except in this
case a 100 mm thick surfacing is added after milling out 75 mm.
As with Alternative 4, the Pavement Type after implementation of the specified Mill &
Replace works would be given by:
The Tab pages specifying the Mill and Replace Works for Alternative 5 are reviewed
below:
Applications Guide 32
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
Applications Guide 33
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
Applications Guide 34
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
In addition to the routine pavement works (crack sealing and patching), this Maintenance
Standard includes Inlay, that is condition responsive, based on the mean rut depth.
The specification for the Inlay works can be reviewed under the Tabs:
General/Design/Intervention/Costs/Effects
Applications Guide 35
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
Applications Guide 36
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
Applications Guide 37
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
Applications Guide 38
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
This screen confirms the base alternative for economic analysis (that is, the first alternative
entered), and that the discount rate is 6%.
For this case study, accident costs, energy balance emissions and acceleration effects are not
included in the analysis.
Applications Guide 39
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
2000 Overlay at 15% cracking including preparatory patching and edge repair
2007 Overlay at 15% cracking
2015 Overlay at IRI 6
Observations The absence of routine pavement works between overlay applications
should be investigated by comparing the intervention criteria with the
progression of defects on the Deterioration Summary report.
Applications Guide 40
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
The works proposed by these alternatives are summarised below. Note the different
thickness of new surfacing (after milling) for alternatives 4 and 5.
Applications Guide 41
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
(US$ millions)
1 Routine and Reconstruct 18.8 (base)
2 Routine + 50 mm overlay 18.0 78.1
3 Routine + SD + 50 mm overlay 56.0 50.6
4 Routine + Mill 75 mm & Replace 75 mm 21.2 92.5
5 Routine + Mill 75 mm & Replace 100 24.8 86.5
mm
6 Routine + Inlay 10.2 86.9
Note that the significant condition parameters available as intervention criteria for Case Study
2 are:
! Roughness (IRI)
May be used as an intervention criteria for Reconstruction, Overlay, and Mill & Replace
! Total (unpatched) damaged area (ADAMR)
May be used as an intervention criterion for Reconstruction and Surface Dressing, and
defined by:
where:
Applications Guide 42
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A2 - PROJECT ANALYSIS CASE STUDY 2
* Note :
where:
where:
Note also that transverse thermal cracks (no/km) may be adopted as an intervention
criterion for crack sealing
! Severely Damaged Area (ADAMS)
Applications Guide 43
Version 1.0
PROJECT ANALYSIS CASE STUDIES
Applications Guide 1
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A3 - PROJECT ANALYSIS CASE STUDY 3
! Specify Alternatives
1.2.1 General
This screen confirms the project description, analysis type, analysis period and the pre-defined
Road Network and Vehicle Fleet.
This case study is presented as a project analysis. The road under study is represented by one
section, 10 km long, and the different widening proposals represent project alternatives. The
analysis period is defined by the start year 2000 and duration 20 years (that is, 2000 - 2019).
Applications Guide 2
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A3 - PROJECT ANALYSIS CASE STUDY 3
The screen confirms that only one road section will be included in the analysis. By double-
clicking on the section description reveals the Definition/Geometry/Pavement/Condition
Tabs which gives access to the section details.
The following details have particular relevance to this case study; they are concerned with the
effects of road widening.
! Definition
The speed-flow type of the existing road is specified as a two lane road. The existing
carriageway width is 7 m, with two traffic lanes. If lane addition is implemented
(alternative 4 in this case study), the width of each additional lane will be calculated as
Applications Guide 3
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A3 - PROJECT ANALYSIS CASE STUDY 3
the existing carriageway width divided by the existing number of traffic lanes (that is, 7/2
= 3.5 m for this case study).
The speed-flow type defines the capacity of the existing road. The capacity may be
increased by Improvement Works such as partial widening (by up to 3 m) or lane
addition (by 1 - 9 lanes). The capacity of the road after improvement is defined by the
speed-flow type specified within the Improvement Standard. Default values of the
ultimate capacity for several speed-flow types are given in Table A3.1. Further details of
speed-flow model parameters are given in Chapter D5, and in Part B of the Analytical
Framework and Model Descriptions.
Table A3.1
Ultimate capacity for selected speed-flow types
The traffic flow (AADT) on the road section is confirmed as 15,000 in 1998. Details of
AADT, traffic composition and growth rate, as defined for this project case study, will be
reviewed under Define Normal Traffic.
Note also that for this road section there is some speed reduction due to roadside
activities or motorised/non-motorised transport. The speed reduction factors are
specified under Section/Details/Alignment.
Applications Guide 4
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A3 - PROJECT ANALYSIS CASE STUDY 3
! Pavement
The pavement type has been defined as Asphalt Mix on Granular Base (AMGB) under
the General Tab. The dates of Previous works listed under the Pavement Tab indicates
that the pavement type is unchanged since the original construction in 1988.
The Structural Number (SN), estimated as 3.3, represents the existing pavement layers
above the subgrade. The SNP value includes the subgrade (8% CBR) contribution. Note
that for the improvement works (partial widening or lane addition), a value of 3.3 has
also been specified for the Structural Number, SN, (under Improvement/Pavement).
Applications Guide 5
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A3 - PROJECT ANALYSIS CASE STUDY 3
! Condition
This confirms the condition levels defined for 1998. For project appraisal, this data
would normally be collected as part of a detailed condition survey.
The vehicles selected from the pre-defined Northern Province fleet are shown on this screen.
Note that Non-Motorised Transport (NMT) is included in the analysis (NMT box ticked at top
of Select Vehicles Tab page). Individual vehicle attributes can be reviewed by double-
clicking on the appropriate vehicle type description.
Applications Guide 6
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A3 - PROJECT ANALYSIS CASE STUDY 3
Applications Guide 7
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A3 - PROJECT ANALYSIS CASE STUDY 3
1.2.5 Alternatives
The four different alternatives considered for this case study are defined below. The analysis
period is 20 years (from years 2000 to 2019).
Applications Guide 8
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A3 - PROJECT ANALYSIS CASE STUDY 3
Alternative Description
The key data relating to the alternative widening scenarios are given in Table A3.2 and Table
A3.3.
Table A3.2
Summary of widening alternatives
Notes:
1 All data for Section ID A003-01, length 10 km
2 Alternative 1 represents the base case (7 m wide two-lane commuter road)
As noted previously, the speed-flow type determines the capacity of the road section. The
base case (Alternative 1) has been assigned the speed-flow type for a two-lane road. The
same speed-flow type has been assigned to Alternative 2. With Alternatives 3 and 4, the
capacity can be expected to increase due to the revised speed-flow types specified under the
respective Improvement Standards.
The various widening alternatives have an impact on the speed reduction due to NMT, MT,
and roadside friction. For the existing road section, these data are specified under
Applications Guide 9
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A3 - PROJECT ANALYSIS CASE STUDY 3
Section/Details/Alignment. For the widening alternatives, the data are specified under
Improvement/Geometry. The values assigned for each alternative are given in Table A3.3.
Table A3.3
Speed limit and speed reduction factors
The Road Works Standards and associated works assigned to each alternative are summarised
in Table A3.4. Each widening assignment is scheduled to start in year 1 of the analysis
period (year 2000), with a construction period of two or three years.
Each widening alternative includes a Maintenance Standard (Routine + 50 mm overlay)
providing condition responsive routine pavement works and condition responsive overlay,
effective from the year following completion of the widening works. This Maintenance
Standard includes the following works items.
1 Crack Sealing if Area of Wide Structural Cracking > = 5% Carriageway Area
2 Patching if Severely Damaged Area > = 5% Carriageway Area
3 50 mm overlay if Roughness > 6 IRI OR if Area of Structural Cracking >= 15%
carriageway area
These maintenance works define the base alternative.
Applications Guide 10
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A3 - PROJECT ANALYSIS CASE STUDY 3
Table A3.4
Details of road works standards for each project alternative
This case study is concerned with the effects of road widening. Note that the base case road
geometry can be reviewed under Section/General, as discussed previously under Select
Sections. Relevant details of each alternative are discussed below:
Applications Guide 11
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A3 - PROJECT ANALYSIS CASE STUDY 3
Details of each works activity can be reviewed by double-clicking the appropriate works
name. The details are similar to those described for Case Study 2, Alternative 2.
The intervention criteria can be reviewed via the Intervention Tab associated with each works
activity.
Applications Guide 12
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A3 - PROJECT ANALYSIS CASE STUDY 3
Applications Guide 13
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A3 - PROJECT ANALYSIS CASE STUDY 3
Applications Guide 14
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A3 - PROJECT ANALYSIS CASE STUDY 3
Alternative 2: Widening by 1 m
! I Partial Widening by 1 m (Effective from year 2000)
! M Routine + 50 mm overlay (Effective from year 2002)
Details of the Improvement Standard (partial widening) are given under seven Tabs:
General/Design/Intervention/Costs/Pavement/Geometry/Effects
Applications Guide 15
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A3 - PROJECT ANALYSIS CASE STUDY 3
Applications Guide 16
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A3 - PROJECT ANALYSIS CASE STUDY 3
Applications Guide 17
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A3 - PROJECT ANALYSIS CASE STUDY 3
The improvement type (partial widening) and duration are specified under the General Tab.
The increase in width, together with the pavement type (of the partial widening) and speed-
flow type for the widened road is specified under Design. For this alternative the speed-flow
type after widening has been specified as a two-lane road (the same as the base case), as the
widening is only 1m. The pavement type (of the partial widening) is Asphalt Mix on Granular
Base (AMGB), which is the same as the existing pavement (Section/Pavement). Note that
the construction quality indicators (CDS and CDB) of the partial widening can be reviewed
via Design/Edit Construction Details. The main works costs and associated preparatory
costs are held under the Costs Tab, and are spread across the two-year construction period.
Applications Guide 18
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A3 - PROJECT ANALYSIS CASE STUDY 3
Details of the surfacing used for widening (material and thickness) and the Structural Number
(SN) of the partial widening are specified under Pavement. For improvements, SN will
normally be assessed using design layer thicknesses and as new layer coefficients.
Alternative 3: Widening by 3 m
! I Partial Widening by 3 m (Effective from year 2000)
! M Routine + 50 mm overlay (Effective from year 2002)
This alternative is similar to Alternative 2, except that the widening is 3 m, specified under
Improvement Standard/Design. The speed-flow type after widening is a wide two-lane road,
which has increased capacity over the original two-lane road (see Table A3.1). In addition,
speed reduction due to NMT is less significant (that is, higher factor in Table A3.3).
Applications Guide 19
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A3 - PROJECT ANALYSIS CASE STUDY 3
Note that on the Design Tab page the speed-flow type after improvement has been specified
as four lane road, which has improved capacity over the two lane road specified for the base
case (see Table A3.1). In addition, there is no speed reduction due to NMT, MT, or roadside
friction.
The cost of the improvement works and associated preparatory works are specified on the
Costs Tab page.
Applications Guide 20
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A3 - PROJECT ANALYSIS CASE STUDY 3
The Setup Run screen confirms that the base alternative for economic analysis is Alternative
1, and that the discount rate is 6%.
For this case study, accident costs, energy balance emissions and acceleration effects are not
included in the analysis.
Alternative 1 2 3 4
Widening None 1m 3m 2 lanes
Baseline IRI (year) 6.14 (2000) 5.98 (2002) 5.38 (2002) 4.64 (2003)
Applications Guide 21
Version 1.0
PROJECT ANALYSIS CASE STUDIES APPENDIX A3 - PROJECT ANALYSIS CASE STUDY 3
Note that with Alternatives 2-4, the roughness value immediately following the widening
works is based on the full carriageway width, and hence the baseline IRI value decreases as
the widening (with no defects immediately after construction) increases.
The Timing of Works report should be read in conjunction with the Deterioration Summary
report, which indicates the progression of certain defects throughout the analysis period.
Carriageway
width after NPV
Alternative Widening widening
(millions of US$)
(m)
2 1m 8
3 3m 10
4 2 lanes 14
Applications Guide 22
Version 1.0
CONCEPTS OF ANALYSIS
Zone centroid 1
A
C D
B
E
Key: Zone centroid 2
Existing road sections
Proposed bypass (Section E)
Applications Guide 1
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX A4 - PROJECT ANALYSIS CASE STUDY 4
A4.2) predict a transfer of heavy vehicles from Sections A, C and D to Section E once the
bypass (Section E) is constructed.
Table A4.1
AADT values by project alternative
Notes:
1 Bypass opening year is 2002 for Alternative 2, 2003 for Alternative 3, and 2004 for
Alternative 4.
2 Data for Sections A-D may be reviewed via Specify Alternatives/Edit
Alternative/Diverted Traffic.
3 Data for Section E may be reviewed via Specify Alternatives/Edit Alternative/Edit New
Section.
4 Shaded data included for information, and is not entered to HDM-4.
5 n/a = not applicable
Applications Guide 2
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX A4 - PROJECT ANALYSIS CASE STUDY 4
Table A4.2
Traffic composition and growth rates
Without bypass:
Sections A,C,D Sections B,E
Vehicle type Sections A-D (1998)
Compo- Annual Compo- Annual Compo- Annual
sition growth sition growth sition growth
(%) rate (%) (%) rate (%) (%) rate (%)
Medium bus 5 2 5 2 5 2
Heavy truck 5 2 3 2 10 2
Medium truck 10 2 7 2 14 2
Medium car 60 2 70 2 50 2
Mini-bus 5 2 5 2 5 2
Artic truck 5 2 0 0 6 2
Applications Guide 3
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX A4 - PROJECT ANALYSIS CASE STUDY 4
❏ Select Vehicles
❏ Define Normal Traffic
! Specify Alternatives (see Section 1.2.2)
The General screen confirms the project description, analysis type, analysis period
(2000-2019) and the pre-defined Road Network and Vehicle Fleet. Note that the
Applications Guide 4
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX A4 - PROJECT ANALYSIS CASE STUDY 4
construction of a new road section can only be modelled as a section alternative within a
project alternative (analysis by project).
! Select Sections
The four road sections (Sections A-D) included in the analysis are shown in the Select
Sections screen. Section details can be accessed by double-clicking the section
description. The section representing the bypass (Section E) is not included. Section E
represents a proposed new section and is defined under specific project alternatives (see
later). Note that, even after the analysis, Section E is not automatically added to the
Road Network folder, as the section does not actually exist; however the user may add it
to the appropriate Network after construction.
! Select Vehicles
The vehicle types that may be assigned to the sections (including the new section) are
given under Select Vehicles.
Applications Guide 5
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX A4 - PROJECT ANALYSIS CASE STUDY 4
Traffic data (AADT, vehicle composition and growth rates) for Sections A, B, C and D are
defined under Define Normal Traffic. The vehicle composition and growth rates for a
particular section can be reviewed by double-clicking the appropriate row on the above Tab
page. Note that traffic data for the proposed bypass (Section E) are assigned separately under
the appropriate Project Alternatives.
The four project alternatives considered for this case study are displayed above.
Alternative 1 represents the base case, that is, existing road sections A-D without the bypass.
Alternatives 2, 3 and 4 include the bypass (Section E), with carriageway width and pavement
Applications Guide 6
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX A4 - PROJECT ANALYSIS CASE STUDY 4
construction as defined in Table A4.3. The screen above indicates the maintenance standard
assigned to Section A under Alternative 1 (base case).
Table A4.3
Details of proposed bypass by project alternative
Number of lanes 2 2 4
The maintenance standards associated with each project alternative are summarised in Table
A4.4. For each section, the same maintenance standard has been assigned to each constituent
section alternative. This allows the effect of different bypass carriageway widths to be
assessed directly.
All alternatives include condition responsive routine maintenance (crack sealing and
patching), as well as a condition responsive overlay.
Applications Guide 7
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX A4 - PROJECT ANALYSIS CASE STUDY 4
Table A4.4
Summary of maintenance standards by project alternative
Notes:
1 RM = condition responsive Routine Pavement Maintenance: includes patching and crack
sealing (section denoted by subscript).
2 OVL = condition responsive 50mm overlay (section denoted by subscript).
3 New section AMGB = pavement type (Asphalt Mix on Granular Base).
4 ( 2-lane) = speed-flow type given in parentheses.
The Road Works Standards and associated works assigned to each alternative are summarised
in Table A4.5. Each alternative includes a Maintenance Standard (Routine + 50 mm overlay)
providing condition responsive routine works and condition responsive overlay. This is
effective from the year 2000 for the base case. The ‘effective from’ date for Section E is the
year of opening of the bypass (applies Alternatives 2, 3 and 4).
Note that, for Section E, the ‘effective from’ date could also have been specified as the year
2000, giving the same resulting works. This is because the maintenance standard would not
be applied during the construction period. (Based on the works ranking hierarchy, new section
construction supersedes works associated with the maintenance standard).
The Maintenance Standard includes the following works:
! Crack Sealing if Area of Wide Structural Cracking > = 5% Carriageway Area.
! Patching if Severely Damaged Area > = 5% Carriageway Area.
! 50 mm overlay if Roughness > 6 IRI.
! 50 mm overlay if ‘All Structural Cracking’ > = 15% Carriageway Area.
Applications Guide 8
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX A4 - PROJECT ANALYSIS CASE STUDY 4
Table A4.5
Details of road works standards for each project alternative
Applications Guide 9
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX A4 - PROJECT ANALYSIS CASE STUDY 4
This shows the maintenance standard assigned to Section E under Alternative 2. The
assignments for a particular section can be reviewed by selecting that section. The pre-
defined maintenance standards were assigned via the Assign Maintenance button.
Applications Guide 10
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX A4 - PROJECT ANALYSIS CASE STUDY 4
The following screen is displayed by clicking the Edit New Section button on the
Alternative Details screen (see previous page).
This gives Section E details that have been specified for Alternative 2:
❏ Details of Bypass
General Name of new section (Section E).
Start year of construction of Section E is 2000.
Costs Economic and financial costs specified as shown (US Dollars per km).
Duration of construction is 2 years.
Allocation of construction cost within construction period is 50% each
year.
Applications Guide 11
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX A4 - PROJECT ANALYSIS CASE STUDY 4
! Normal Traffic
The following screen is displayed by clicking the Normal Traffic button on the New
Construction Section Option screen (see previous page).
This screen defines the AADT and vehicle composition using the bypass in its opening year
(2002 for Alternative 2). The traffic data for this case study are summarised in Table A4.1
and Table A4.2.
! Section Details
The following screen is displayed by clicking the Section Details button on the New
Construction Section Option screen.
Applications Guide 12
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX A4 - PROJECT ANALYSIS CASE STUDY 4
Once Section E has been defined for a specific project alternative via the New
Construction Section Option screen, it is listed in the Alternative Details screen.
The top button on this screen allows the user to specify Diverted Traffic. In the context
of this case study, "Diverted traffic" refers to traffic using sections A,B C and D after the
opening of the bypass.
Applications Guide 13
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX A4 - PROJECT ANALYSIS CASE STUDY 4
The Diverted Traffic screen indicates the initial AADT (1998) on sections A-D, and the
new "diverted" AADT values in 2002 (the bypass opening year under Alternative 2).
The vehicle composition and growth rates for a particular section (in 2002) may be
reviewed by double-clicking the relevant cell under the "New AADT in 2002" column or
by selecting the appropriate cell and clicking the Edit Diversion Details button.
! Diverted Traffic Details
The screen below shows the vehicle composition and growth rates (in 2002) for Section
A under Alternative 2.
Applications Guide 14
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX A4 - PROJECT ANALYSIS CASE STUDY 4
Applications Guide 15
Version 1.0
Part A Appendices
Contents
Applications Guide i
Version 1.0
CONCEPTS OF ANALYSIS
1.1 Introduction
This Case Study demonstrates the application of Programme Analysis to the production of a
rehabilitation programme for a road network in the Western Province of a country. The study
demonstrates the Life Cycle Analysis method described in Chapter B2.
The objective here is to prepare a prioritised list road projects from a selected group of road
sections that have been identified to be in poor condition. In practice, this would follow some
condition survey of the road network and the application of pavement condition trigger levels
or thresholds to identify those sections in need of some form of rehabilitation.
The Life Cycle analysis method used for this Case Study is the same as that used in Project
Analysis and described in Chapter B1.
To demonstrate this Programme analysis method, the Case Study is presented in the following
steps:
Step 1 : Overview of the case study.
Step 2 : Review of the case study input data.
Step 3 : Run programme analysis with the life-cycle option
Step 4: Examine the results of the work programme
Applications Guide 1
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX B1 - PROGRAMME ANALYSIS CASE STUDY 1
To review the case study data, open the Programmes folder and select the Western Province
Rehabilitation (Life Cycle Analysis) study as illustrated in Figure B1.1.
Applications Guide 2
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX B1 - PROGRAMME ANALYSIS CASE STUDY 1
Assign Standards
! Specify Alternatives Tab
Generate Programme
! Perform Run Tab
Generate Reports
! Reports Tree
1.4.1 General
The General Tab can be used to define the study description, analysis type, analysis period
(start year and duration), the currency conversion rates and the discount rate to be used in the
economic analysis. This Tab also shows the selected Road Network and Vehicle Fleet.
Applications Guide 3
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX B1 - PROGRAMME ANALYSIS CASE STUDY 1
This Case Study involves Life Cycle analysis carried out over a fifteen year analysis period
starting from 2000 to 2014.
Applications Guide 4
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX B1 - PROGRAMME ANALYSIS CASE STUDY 1
1.5.1 Alternatives
The Life Cycle analysis method requires a comparison to be made between a base case (do
minimum) alternative against a project case (do something) alternative. Details of at least
two alternatives can be specified in this Tab.
For this Case Study, two alternatives have been defined for each road section; a base case
alternative and a rehabilitation alternative. The base case comprises routine pavement
maintenance only (crack sealing and patching potholes only). The rehabilitation alternative
includes routine maintenance, resealing, overlays and reconstruction for bituminous
pavements. Details of the maintenance standards are summarised in Table B1.1.
Applications Guide 5
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX B1 - PROGRAMME ANALYSIS CASE STUDY 1
Table B1.1
Details of Maintenance Standard (Brehab) assigned to bituminous road
sections
Resealing Single Surface Dressing Total damaged area >= 30% of pavement 16
surface area
Note: * Ranking of road works from Chapter D2 of Analytical Framework and Model
Descriptions
If more than one of these activities are triggered in a given year, the works item that is highest
in the hierarchy of road works (that is, with lowest ranking value) will be applied.
Note that routine pavement works (for example, patching and crack sealing) have the same
ranking, and therefore both can be applied in the same year. (The hierarchy is defined by the
ranking of road works described in Chapter D2 of Analytical Framework and Model
Descriptions).
Applications Guide 6
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX B1 - PROGRAMME ANALYSIS CASE STUDY 1
Table B1.2
Details of alternatives for life-cycle analysis
Alternative Surface Maintenance Standard Description Code Effective Maintenance Works Code
Class from year
Base Case Bituminous Patching and Crack Sealing Broutine 2000 Pothole Patching PATPOT
Rehabilitation Bituminous P&CS, Reseal, Overlay & Reconstruct Brehab 2000 Pothole Patching * PATPOT
Reseal RSL30
Overlay OV5IRI
Reconstruct REC12
Notes:
1 Details of the intervention criteria for road work activities are given in Table B1.1.
2 The routine pavement works in the Rehabilitation alternative (identified by *) are identical to those in the Base Case.
Applications Guide 7
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX B1 - PROGRAMME ANALYSIS CASE STUDY 1
For life-cycle analysis, the Base alternative for economic analysis is specified in the Run
Setup screen. This can be displayed by selecting the Run Setup button within the Perform
Run Tab.
On completing Run Setup, press the Start button to commence the economic analysis, which
produces an unconstrained work programme. This contains, for each road section, the
alternative with the highest NPV (refer to Chapter B2). This represents an unconstrained
programme, with the total financial cost given in the cumulative cost column.
The ideal budget required for periodic maintenance in year 2000 is approximately US$ 22
million.
Applications Guide 8
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX B1 - PROGRAMME ANALYSIS CASE STUDY 1
Details of the budget optimisation options are specified by pressing the Optimisation Setup
button. For this case study, optimisation is by incremental benefit/cost ranking (see Chapter
B2). The parameters to be specified are given below:
Minimum incremental value = 0.1
Efficiency frontier Zone = 75%
Applications Guide 9
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX B1 - PROGRAMME ANALYSIS CASE STUDY 1
Applications Guide 10
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX B1 - PROGRAMME ANALYSIS CASE STUDY 1
Year Section Length AADT Surface Work Description NPV/Cost Financial Cumulative
(km) Class Costs Costs
2000 MTS 549 km 0 - 1.0 1.0 2521 Bituminous Overlay 50mm 3.38 0.13 0.13
2000 MSC 112 km 35 - 80 21.0 1770 Bituminous Overlay 50mm 1.85 2.65 2.77
2000 MAN 203 km 254 - 262.8 8.8 1781 Bituminous Overlay 50mm 1.83 1.13 3.91
2000 MSC 141 km 28 - 49.5 21.5 2154 Bituminous Resealing SBSD 1.28 0.92 4.83
2000 MSC 131 km 16.5 - 26 9.5 2154 Bituminous Resealing SBSD 1.14 0.42 5.25
2000 MTS 754 km 0 - 1.8 1.8 1074 Bituminous Overlay 50mm 1.11 0.25 5.50
2000 MSW 959 km 0.1- 5, 10- 15.4 10.3 1265 Bituminous Overlay 50mm 1.10 1.26 6.76
2000 MSE 203 km 80 - 90 10.0 1191 Bituminous Overlay 50mm 0.76 1.26 8.02
2000 MAN 409 km 105 - 110 5.0 1076 Bituminous Overlay 50mm 0.60 0.63 8.65
2000 MTN 836 km 0 - 13 13.0 926 Bituminous Overlay 50mm 0.60 1.64 10.29
2000 MAN 243 km 0 - 21 21.0 1044 Bituminous Overlay 50mm 0.53 2.65 12.94
2000 MTN 805 km 145 - 152.6 7.6 873 Bituminous Overlay 50mm 0.41 1.01 13.94
2000 MSC 138 km 5 - 10 & 65 - 70 10.0 1315 Bituminous Resealing SBSD 0.21 0.42 14.36
This report was generated by version 1.0 of HDM-4. Page -1 of 1
Applications Guide 11
Version 1.0
CONCEPTS OF ANALYSIS
1.1 Introduction
This Case Study demonstrates the application of Programme Analysis to the production of a
rehabilitation programme for a road network in the Eastern Province of a country. The study
demonstrates the Multi-Year Forward Programme method described in Chapter B2.
The objective is again to prepare a prioritised list road projects from a selected group of road
sections that have been identified to be in poor condition. In practice, this would follow some
condition survey of the road network and the application of pavement condition trigger levels
or thresholds to identify those sections in need of some form of rehabilitation.
This Case Study demonstrates the application of the Multi-year Forward Programme method
for the production of a rehabilitation programme for a road network. The only difference with
the life cycle analysis data set is the way that the maintenance standards are assigned to road
sections. The multi-year forward programme method requires one maintenance standard to be
assigned to each road section, while life-cycle analysis requires at least two standards per
section to permit economic analysis. The multi-year forward programme method assigns a
default do minimum alternative which corresponds to deferring capital road works (periodic
maintenance and improvements) to the first year after the budget period.
To demonstrate the multi-year forward programme method, the Case Study is presented in the
following steps:
Step 1 : Overview of the case study data.
Step 2 : Review of the case study data.
Step 3 : Run multi-year forward programme analysis
Step 4: Examine the analysis results
Applications Guide 1
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX B2 - PROGRAMME ANALYSIS CASE STUDY 2
To review the case study data, open the Programmes folder and select the Eastern Province
Rehabilitation (Forward Programme) study as illustrated in Figure B2.1.
Applications Guide 2
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX B2 - PROGRAMME ANALYSIS CASE STUDY 2
The data used for this Case Study has already been created and therefore the procedure
described below does not include the creation of a new study.
The procedure is similar to that of the Life Cycle method and involves defining details of the
programme analysis to be carried out, assignment of work standards, generation of the work
programme, optimisation under budget constraints, and finally generation of reports. This
procedure should normally be followed sequentially, although iteration between the above
tasks can be done in order to review previously specified data.
The data is arranged in groups (corresponding to the tasks) and, within these, the data are
presented in different Tabs as summarised below:
Applications Guide 3
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX B2 - PROGRAMME ANALYSIS CASE STUDY 2
Select Sections
The road network under study is represented by 21 sections (one section with unsealed surface
class and all others with bituminous surfacing).
It is assumed that the long list of candidate road sections from Eastern Province was drawn up
following some pavement condition survey. The list would normally contain only those road
sections deemed to require some form of periodic maintenance or rehabilitation during the
next budget period. Consequently, all road sections have been selected for the analysis.
However, if the road network had contained details of all road sections in Eastern Province, it
is possible to select only those that should be included in the work programme for the next
budget period. Alternatively, all road sections in the database could be analysed, particularly if
this involves a long budget period. The selected sections define a sub-network for which a
prioritised programme is required.
It is recommended that work programmes should be prepared for short-term budget periods,
for example less than five years.
Details of each road section can be edited from within this Tab (press the View/Edit Network
button).
Select Vehicles
The vehicles selected from the pre-defined Provincial Vehicle Characteristics fleet are
indicated below. Individual vehicle attributes may be reviewed by double-clicking on the
appropriate vehicle type description (or by selecting the section and clicking the Edit Vehicle
button).
Applications Guide 4
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX B2 - PROGRAMME ANALYSIS CASE STUDY 2
Define Traffic
This Tab defines the traffic composition and growth rates on the selected road sections. The
initial traffic composition and growth rates (by vehicle type) can be reviewed by double-
clicking the section description (or by selecting the section and clicking the Section Traffic
Details button).
This Tab specified the traffic composition and annual growth rates that are applicable on each
road section for each vehicle type in this analysis. Note that the traffic details can be copied
from one road section to others, thereby permitting each road section to have a different road
set of traffic growth characteristics, if required.
Work programming
For Multi-Year Forward Programme, one maintenance standard (representing the
rehabilitation alternative) is assigned to each road section, or, if an improvement is proposed,
this is assigned together with associated maintenance standards before and after the
improvement is applied. For Multi-Year Forward Programme, the base alternative is
automatically assigned as the works needed in the first year following the defined budget
period (with no works during the budget period).
In this Case Study, one maintenance policy has been assigned to all paved road sections and
one to unsealed sections based on the Surface Class (that is, bituminous or unsealed) as
follows:
Bituminous sections: M P&CS, Reseal, Overlay & Reconstruct (short code Brehab)
where P&CS = Patching and Crack Sealing
Unsealed section: M Grade, Spot regravel & Resurface (short code Grehab)
Each maintenance standard comprises a set of condition responsive works activities and
establishes a maintenance policy for the road sections to which the standard is assigned. For
this Case Study, one standard has been set up for each surface class. In practice this may be
further refined by assigning different maintenance standards based on, for example, road class
or traffic level, both of which indicate the relative importance of the section..
Details of the works activities and condition responsive intervention criteria associated with
each maintenance standard are summarised in Table B2.1 and Table B2.2. The details may be
reviewed under Maintenance Standards in HDM-4 Workspace shown in Figure B2.1.
Applications Guide 5
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX B2 - PROGRAMME ANALYSIS CASE STUDY 2
Applications Guide 6
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX B2 - PROGRAMME ANALYSIS CASE STUDY 2
Table B2.1
Details of Maintenance Standard assigned to bituminous road sections
short code: Brehab
Note: * Ranking of road works from Chapter D2 Analytical Framework and Model
Descriptions.
If more than one of these activities are triggered in a given year, the works item that is highest
in the hierarchy of road works (that is, with lowest ranking value) will be applied. Note that
the routine pavement works (patching and crack sealing) have the same ranking, and both can
be applied in the same maintenance year. (The hierarchy is defined by the ranking of road
works in Chapter D2 of Analytical Framework and Model Descriptions - Modelling Road
Works Effects).
Table B2.2
Details of Maintenance Standard assigned to unsealed road sections
short code: Grehab
Note: * Ranking of road works from Chapter D2 of Analytical Framework and Model
Descriptions
The ranking order assigned to these road work items implies that Resurfacing has the highest
priority. Spot regravelling and Grading have equal ranking, and therefore both of these can be
performed in a given analysis year.
Details of the maintenance works items can be reviewed under Maintenance Standards in
HDM-4 Workspace shown in Figure B2.1.
Applications Guide 7
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX B2 - PROGRAMME ANALYSIS CASE STUDY 2
Perform Run
For Multi-Year Forward Programme Analysis, the user can choose to perform the run either
with or without economic analysis. Economic analysis is required before budget optimisation.
This can be selected using the Run Setup button.
The Run Setup button also allows the user to select Energy Balance and Emissions models,
to model acceleration effects, and to include accident costs in the economic analysis.
When Run Setup is completed, the Work Programme (prior to budget optimisation) is
generated by pressing the Start button. Once completed, the user is strongly advised to Save
the generated programme, especially before viewing any reports.
Applications Guide 8
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX B2 - PROGRAMME ANALYSIS CASE STUDY 2
Unconstrained Programme
The generated Work Programme can be viewed on screen prior to performing a budget
optimisation.
Define Budget
The budget optimisation provides a facility to select road sections that can be included within
a specified budget in order to maximise the economic benefits. Given that the selected road
sections from Eastern Province would require US$ 80 million in over the three-year period
from 2000 to 2002 for periodic maintenance, it is now possible to select the optimal set of
road sections if less than 100% of the required budget is available. Assume that the amount of
money available for the three-year period has been set at US$ 60 million.
Optimised Programme
After performing the budget optimisation, a revised work programme is produced. This has a
total cost that is within the specified budget. The road sections are listed chronologically (by
year) in accordance with the incremental benefit/cost ranking index described in Chapter B2.
Applications Guide 9
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX B2 - PROGRAMME ANALYSIS CASE STUDY 2
Bituminous P&CS, Reseal, Overlay & Reconstruct (Brehab) 1999 Reseal (RSL30)
Overlay (OV5IRI)
Rehabilitation
Reconstruct (REC12)
Grading (GRADE) *
Unsealed Grade, Spot Regravel & Resurface (Grehab) 1999 Spot Regravel (SPRG) *
Resurface (RESURF)
Notes:
1 Details of intervention criteria for works activities are given in Table B2.1 and Table B2.2.
2 Routine pavement works in rehabilitation alternative (identified by *) are identical to those in base alternative.
Applications Guide 10
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX B2 - PROGRAMME ANALYSIS CASE STUDY 2
Year Section Road Class Length AADT Surface Class Work Description NPV/C Financial Cumulative
(km) Costs Costs
2000 Gaborone-Lobatse Section 2 Primary or Trunk 34.0 1334 Bituminous Resealing SBSD 128.46 1.37 1.37
2000 Gaborone-Pilane Primary or Trunk 30.0 1516 Bituminous Resealing SBSD 4.06 1.21 2.57
2000 Pilane-Rasesa Secondary or Main 7.0 459 Bituminous Resealing SBSD 3.38 0.28 2.85
2000 Gaborone-Lobatse Section 1 Primary or Trunk 60.0 1814 Bituminous Resealing SBSD 3.34 2.41 5.27
2000 Serule-Francis Town Primary or Trunk 98.0 1119 Bituminous Resealing SBSD 3.08 3.94 9.21
2000 Mosetse-Nata Secondary or Main 68.0 177 Bituminous Resealing SBSD 3.00 2.73 11.94
2000 Gaborone-Tlokweng Primary or Trunk 10.0 2586 Bituminous Overlay 50mm 1.63 1.21 13.15
2000 Sebina - Mosetse Secondary or Main 79.0 202 Bituminous Resealing SBSD 1.53 3.18 16.32
2000 Francis Town - Selebi Phikwe Primary or Trunk 85.0 1226 Bituminous Resealing SBSD 1.33 3.42 19.74
2000 Palapye-Serule Primary or Trunk 70.0 963 Bituminous Overlay 50mm 0.77 8.44 28.18
2000 Lobatse-Jwaneng Section 1 Primary or Trunk 120.0 812 Bituminous Overlay 50mm 0.38 15.55 43.73
2000 Jwaneng-Sekoma Section 1 Secondary or Main 80.0 607 Bituminous Overlay 50mm 0.34 9.65 53.38
2000 Monametsana-Mahalapye Secondary or Main 142.0 446 Bituminous Overlay 50mm 0.33 17.13 70.51
2000 Gravel Road Tertiary or Local 10.0 114 Unsealed Resurface 0.18 0.29 70.80
2000 Sekoma-Morwamosu Secondary or Main 79.0 313 Bituminous Overlay 50mm 0.17 8.53 79.33
2002 Rasesa-Monametsana Secondary or Main 15.0 270 Bituminous Resealing SBSD 0.22 0.60 79.93
Applications Guide 11
Version 1.0
Part A Appendices
Contents
Applications Guide i
Version 1.0
CONCEPTS OF ANALYSIS
General
This screen confirms the study description, optimisation method, analysis period (start year
and duration) and the names of the pre-defined Road Network and Vehicle Fleet. For this
case study the selected optimisation method is maximisation of benefits (NPV).
Select Sections
The list of representative sections is displayed under the Select Sections Tab.
The road network under study has been modelled by eleven representative sections. The
representative sections for bituminous (paved) roads have been based on traffic volume
(High, Medium or Low) and road condition (Good, Fair and Poor), giving a total of nine
sections.
The section ID (5 letters) for the bituminous sections is derived as shown by the following
example:
For PHTGC:
P Paved
HT High Traffic
GC Good Condition
Unsealed (gravel) roads have been classified by traffic volume alone, since condition may
change rapidly in any given year. For this case study, no gravel sections are subjected to high
traffic, so the gravel network is represented by two sections only, one for medium traffic and
one for low traffic. The section ID (3 letters) is based on pavement type and traffic level as
indicated below:
For GLT:
G Gravel
LT Light traffic
For GMT:
G Gravel
MT Medium traffic
The use of a matrix template to define the characteristics of the road network in terms of
representative sections is discussed in Section 2.2 of Chapter B3. While this is useful in
identifying groups of road sections with specific characteristics, the definition of properties
for each representative section (for example, based on the average or weighted average of
values within each group) is currently undertaken manually.
Double-clicking on a specific Section Description (or selecting the section and clicking the
View/Edit Section button) reveals the General/Geometry/Pavement/Condition Tabs that
give access to the section's details. Some relevant characteristics of the representative sections
are summarised in Table C1.1and Table C1.2. Note that, for both bituminous and gravel
sections, the condition must be defined at the start of the analysis period. As the analysis
period commences in the year 2000, this means that the condition data must be defined for the
year 1999 (that is, end of 1999) or earlier.
Table C1.1
Details of representative sections for bituminous roads
Section ID PHTGC PHTFC PHTPC PMTGC PMTFC PMTPC PLTGC PLTFC PLTPC
Length (km) 234 392 437 306 483 615 410 670 720
Traffic level High High High Medium Medium Medium Low Low Low
AADT 6200 5240 5180 2500 2300 2060 1400 1150 970
AADT year 1998 1998 1998 1998 1998 1998 1998 1998 1998
Condition (1999) Good Fair Poor Good Fair Poor Good Fair Poor
Roughness (IRI) 2.5 4.0 6.5 2.5 4.4 6.5 3 4.5 6.1
Skid resistance (SCRIM 50 kph) 0.7 0.5 0.4 0.7 0.5 0.4 0.7 0.5 0.3
Drainage Excellent Fair Poor Excellent Good Fair Good Fair Poor
Pavement type AMAP AMAP AMAP STGB STGB STGB STGB STGB STGB
…Continued
Applications Guide 5
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX C1 - STRATEGY ANALYSIS CASE STUDY 1
Section ID PHTGC PHTFC PHTPC PMTGC PMTFC PMTPC PLTGC PLTFC PLTPC
Old surf thick (mm) 50 50 50 50 50 0 0 0 0
- rehabilitation (overlay) 1996 1996 1996 1991 1991 1991 1991 1991 1991
- resurfacing (reseal) 1996 1996 1996 1996 1996 1991 1991 1991 1991
- preventative treatment 1996 1996 1996 1996 1996 1991 1991 1991 1991
SN after last treatment 2.70 2.70 2.70 2.07 2.07 2.26 1.67 1.67 1.67
Applications Guide 6
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX C1 - STRATEGY ANALYSIS CASE STUDY 1
Table C1.2
Details of representative sections for gravel roads
Condition (1998)
Gravel thickness (mm) 100 70
A tabulated data summary is useful in checking consistency of data across sections and also
within a section. This includes checking that the condition classification is compatible with
the condition parameters, and, with bituminous pavements, that the pavement type (defined
under Section/General) is compatible with the pavement details (surfacing material type,
surfacing thicknesses and dates of previous works) defined under Section/Pavement.
Remember that the dates of previous works refer to HDM-4 Works Types. Hence the date of
last resurfacing refers to Works Type resurfacing which includes works activities such as
surface dressing and slurry seal. Works activities such as overlaying, mill and replace, and
inlay are classed as Works Type rehabilitation and would be recorded under date of last
rehabilitation. The bituminous pavement types are reviewed below.
Bituminous pavement types
Table C1.1 indicates Pavement Type AMAP for the bituminous sections carrying high traffic
levels (PHTGC, PHTFC and PHTPC). This is based on an original construction of 50 mm
AC on a granular roadbase (that is, pavement type AMGB) in 1991 (date of last
reconstruction), subsequently overlaid with 50 mm AC in 1996 (date of last rehabilitation).
This results in the updated Pavement Type AMAP, and the dates of last resurfacing and last
preventative treatment are set to 1996 (equal to date of last rehabilitation). Note the most
recent surfacing thickness set at 50 mm (that is, overlay) and previous thickness is also 50 mm
(original construction).
Pavement Type STAP is indicated for the sections identified as PMTGC and PMTFC. This is
based on an original construction of 50 mm AC on a granular roadbase (pavement type
AMGB) in 1991 (date of last reconstruction), subsequently surface dressed in 1996 (date of
last resurfacing). This gives the updated pavement type STAP. The date of the last
rehabilitation is set to 1991 (equal to date of original construction), and the date of last
preventative treatment is set to 1996 (equal to date of last resurfacing). Note the most
recent surfacing thickness is 25 mm (double surface dressing), and previous surfacing
thickness is 50 mm (original construction).
The road section identified as PMTPC has pavement type AMGB that represents the original
construction of 50 mm AC on a granular roadbase in 1991 (date of last reconstruction). Note
that the dates of last rehabilitation, resurfacing and preventative treatment are also set to 1991
(equal to date of last reconstruction). In this case the most recent surfacing thickness is set to
50 mm (original construction), while the previous surfacing thickness is zero.
Applications Guide 7
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX C1 - STRATEGY ANALYSIS CASE STUDY 1
The three sections with low traffic level (PLTGC, PLTFC and PLTPC) have pavement type
STGB, representing the original construction of double surface dressing on a granular
roadbase in 1991 (date of last reconstruction). The dates of last rehabilitation, resurfacing and
preventative treatment are also set at 1991. The most recent surfacing thickness is set to 25
mm (double surface dressing), while the previous surfacing thickness is zero.
Note that the Adjusted Structural Number (SN or SNP as appropriate) assigned to each section
represents the condition of the pavement immediately after the most recent treatment (that is,
date of last preventative treatment as defined under the Pavement Tab). For example, the SN
value assigned to section PHTGC above represents Pavement Type AMAP (specified under
the Definition Tab) immediately after the overlay application in 1996. The value of SN
(surfacing, roadbase and sub-base only) can be quickly approximated as indicated below,
based on the original AASHTO relationship:
SN = 0.0394
∑a h i i ...(1.1)
where:
ai layer coefficient
hi layer thickness (mm)
Note the reduced value of layer coefficient (0.2) applied to the old surfacing due to its
condition.
Table C1.3
Layer coefficients and thicknesses
hi ai
1996 AC Overlay 50 0.4 20
1991 AC 50 0.2 10
Total 67.5
Giving:
This value has been assigned to section PHTGC under the Section/Pavement Tab.
The subgrade contribution (added to SN to give SNP) is computed by the HDM-4 software.
Note that this approximation does not take account of the reduced contribution from the sub-
base, but is a much quicker manual calculation. Details of HDM-4 Structural Number
concepts are given in Chapter C1 of the Analytical Framework and Model Descriptions.
Applications Guide 8
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX C1 - STRATEGY ANALYSIS CASE STUDY 1
Select Vehicles
The vehicles selected from the pre-defined Western Province fleet are indicated above.
Individual vehicle attributes may be reviewed by double-clicking on the appropriate vehicle
type description (or by selecting the section and clicking the Edit Vehicle button).
This screen confirms the traffic volume (AADT) on the selected road section in the given
year. The initial traffic composition and growth rates (by vehicle type) can be reviewed by
double-clicking the section description (or by selecting the section and clicking the
Section/Traffic/Details button). The initial vehicle compositions are summarised in Table
C1.4.
Note that the initial composition is defined for same year as the AADT. The annual growth
rates may be specified for several periods, but must at least be specified for a period
Applications Guide 9
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX C1 - STRATEGY ANALYSIS CASE STUDY 1
commencing from the analysis period start year (that is, 2000 in this case study). For example
the AADT for section PHTGC is specified for 1998, and the initial vehicle composition
applies to that year. The annual growth rates have been defined only for the period
commencing 2000, and in this case are also applied between 1998 and 2000.
Applications Guide 10
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX C1 - STRATEGY ANALYSIS CASE STUDY 1
Table C1.4
Details of initial vehicle composition and growth rates
Traffic level
Leyland bus 10 3 10 3 0 0
Mercedes heavy truck 10 3 25 3 10 3
Nissan pickup 15 3 20 3 20 3
Tata 7.5t lorry 15 3 0 0 10 3
Toyota Corolla car 40 3 30 3 40 3
Toyota Hiace minibus 10 3 15 3 20 3
Total (%) 100 100 100
Note: Traffic details (AADT & associated year, initial composition and associated growth rates) are assigned by section.
Applications Guide 11
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX C1 - STRATEGY ANALYSIS CASE STUDY 1
By clicking the View Details box, the user may review the assignment of Road Works
Standards by Alternative and Section.
Alternatives
For this case study, four investment alternatives are considered for each of the representative
sections, ranging from the provision of routine pavement maintenance only (Alternative 1) to
an ideal maintenance case (Alternative 4). For each investment alternative, Road Works
Standards have been pre-defined as shown in Table C1.5 and Table C1.6.
Table C1.5
Assigned Road Works Standards - short codes
Gravel (unpaved)
1. Grade & Spot regravel G1 G1
Applications Guide 12
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX C1 - STRATEGY ANALYSIS CASE STUDY 1
With bituminous road sections, maintenance standards have been assigned by alternative and
traffic class. For gravel road sections, a maintenance standard has been defined for each
investment alternative, and an improvement standard has been introduced to Alternative 4
(that is, upgrading gravel roads with medium traffic followed by paved road maintenance, and
widening gravel roads with low traffic).
(Note that Road Authorities may wish to assign standards by Road Class, which may be
represented in some countries by Surface Class alone. With condition responsive
maintenance intervention, the interval between maintenance applications will clearly be
influenced by the traffic volume.)
For bituminous (paved) sections, the Maintenance Standard short code is derived from the
pavement surfacing, traffic class, and investment alternative, for example:
PHT1 is based on:
P Paved
HT High Traffic
1 Alternative 1
For gravel sections, the short code for Maintenance Standards is derived from the pavement
surfacing material and the investment alternative:
For example:
The definition of investment alternatives for each of the representative sections is indicated in
Table C1.6. This shows the short code identifier associated with each of the pre-defined Road
Works Standards (that is, Maintenance Standards and Improvement Standards).
Applications Guide 13
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX C1 - STRATEGY ANALYSIS CASE STUDY 1
Table C1.6
Definition of investment alternatives for representative sections
2 PHT2 - -
PHTGC B H
3 PHT3 - -
4 PHT4 - -
1 PHT1 - -
2 PHT2 - -
PHTFC B H
3 PHT3 - -
4 PHT4 -
1 PHT1 - -
2 PHT2 - -
PHTPC B H
3 PHT3 - -
4 PHT4 - -
1 PMT1 - -
2 PMT2 - -
PMTGC B M
3 PMT3 - -
4 PMT4 - -
1 PMT1 - -
2 PMT2 - -
PMTFC B M
3 PMT3 - -
4 PMT4 - -
1 PMT1 - -
2 PMT2 - -
PMTPC B M
3 PMT3 - -
4 PMT4 - -
1 PLT1 - -
2 PLT2 - -
PLTGC B L
3 PLT3 - -
4 PLT4 - -
…Continued
Applications Guide 14
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX C1 - STRATEGY ANALYSIS CASE STUDY 1
2 PLT2 -
PLTFC B L
3 PLT3 - -
4 PLT4 - -
1 PLT1 - -
2 PLT2 - -
PLTPC B L
3 PLT3 - -
4 PLT4 - -
1 G1 - -
2 G2 - -
GMT U M
3 G3 - -
4 G1 GU4 PLT3
1 G1 - -
2 G2 - -
GLT U L
3 G3 - -
4 G1 GW4 G2
Note: Details of Road Works Standards identified by the above short codes are given in
Table C1.9 and Table C1.11 for bituminous and gravel sections respectively.
Maintenance Standards for bituminous sections
The Maintenance Standards for bituminous sections are reviewed first. These are based on
different combinations of the following works activities/operations.
Table C1.7
Works activities considered for bituminous sections
Crack sealing 22
If more than one of these activities is triggered in a given year, the works item that is highest
in the hierarchy of road works (that is, with lowest ranking value) will be applied. Note that
the routine pavement works (patching and crack sealing) have the same ranking, and both can
be applied in the same maintenance year. (The hierarchy is defined by the ranking of road
works shown in Chapter D2 of the Analytical Framework and Model Descriptions.
Applications Guide 15
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX C1 - STRATEGY ANALYSIS CASE STUDY 1
! Alternative 1
Comprises patching and crack sealing only. This represents the base case (or do
minimum) alternative. In this case study the intervention levels for patching vary from
10 potholes/km for sections with high traffic, to 50 potholes/km for sections with low
traffic (see Table C1.9).
! Alternative 2
Table C1.8
Pavement reconstruction details for bituminous sections
Note: M/Design refers to the Design Tab under the appropriate Maintenance Standard
Details of the Maintenance Standards associated with each alternative are summarised in
Table C1.9. The details may be reviewed under Maintenance Standards in HDM-4
Workspace or under Specify Standards Assignments/Alternatives/Edit Standards.
Applications Guide 16
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX C1 - STRATEGY ANALYSIS CASE STUDY 1
Table C1.9
Details of investment alternatives for bituminous sections
Investment Maintenance Standard Condition responsive criteria Units Intervention levels by traffic class
alternative
Works item High Medium traffic Low traffic
traffic
1 Patch & crack seal (PHT1) (PMT1) (PLT1)
Pothole patching Potholing no/km >=10 >=25 >=50
…Continued
Applications Guide 17
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX C1 - STRATEGY ANALYSIS CASE STUDY 1
Notes:
1 Short code identifiers for Maintenance Standards shown in parentheses, for example (PHT1)
2 SBSD = Single Bitumen Surface Dressing
Applications Guide 18
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX C1 - STRATEGY ANALYSIS CASE STUDY 1
Table C1.10
Works activities considered for gravel sections
Spot regravelling 7
M Resurfacing Regravelling 6
Lane addition 4
The order of ranking of the Maintenance works items gives regravelling highest priority (that
is, with lowest ranking value). Spot regravelling and Grading have equal ranking, and both of
these can be performed in a given analysis year. Details of precedence rules for road works
are given in Chapter D2 of the Analytical Framework and Model Descriptions.
! Alternative 1
Comprises grading and spot regravelling, both condition responsive. This represents the
base case (do-minimum alternative).
! Alternative 2
Includes three works items, grading, spot regravelling and regravelling (resurfacing), all
condition responsive. Grading is triggered by roughness, while spot regravelling and
regravelling are both triggered by the thickness of gravel surfacing. This alternative aims
to maintain a reasonable thickness of gravel surfacing and so provide continuous
protection to the pavement subgrade.
! Alternative 3
Includes the same works items as Alternative 2, but the intervention levels are set to
provide a higher level of serviceability that should trigger the works more frequently.
! Alternative 4
Introduces upgrading to a paved standard for sections with medium traffic (representative
section GMT), and widening for sections with low traffic (section GLT). This represents
the ideal investment alternative. Suitable maintenance is provided before and after the
respective improvements. Note that the respective Improvement Standards are effective
from the year 2003, with works scheduled to start in the year 2004. This means that no
maintenance will be applied during the year 2003, the year prior to improvement.
Details of the Road Works Standards for gravel road sections are given in Table C1.11. The
details can be reviewed under Maintenance Standards in HDM-4 Workspace or under:
Specify Standards Assignments/Alternatives/Edit Standards
Applications Guide 19
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX C1 - STRATEGY ANALYSIS CASE STUDY 1
Table C1.11
Details of investment alternatives for gravel sections
Alt Sections M/I Road Works Standard Effective Responsive Criteria [M] Details (design)
from year
Works Item Start Date/Duration [I]
Spot regravelling GThk <= 100 mm Replace 10% annual material loss
Spot regravelling GThk <= 100 mm Replace 10% annual material loss
Spot regravelling GThk <= 125 mm Replace 10% annual material loss
…Continued
Applications Guide 20
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX C1 - STRATEGY ANALYSIS CASE STUDY 1
Alt Sections M/I Road Works Standard Effective Responsive Criteria [M] Details (design)
from year
Works Item Start Date/Duration [I]
I GU4: Pave section GMT in 2004 2003 Start year 2004 Upgrade to STGB with 2 m widening
Cracking >= 5%
I GW4: Widen section GLT in 2004 2003 Start year 2004 Widen by 3 m
Notes:
Column 1 Alt = Investment Alternative
Column 3 M/I = Maintenance or Improvement Standard
Column 6 GThk = Gravel thickness
Applications Guide 21
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX C1 - STRATEGY ANALYSIS CASE STUDY 1
Work Programme
On completing Run Setup, press the Start button to commence the economic analysis, which
produces a costed work programme containing, for each representative section, the alternative
with the highest NPV (refer to Section 3.2.3 of Chapter B2). This represents an unconstrained
programme, with the total financial cost given in the cumulative cost column. The road works
assigned to each section represent the optimum maintenance and improvements that should be
applied to the road network in accordance with the specified standards.
Applications Guide 22
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX C1 - STRATEGY ANALYSIS CASE STUDY 1
For this case study, two budget scenarios are considered, one at 75% of funding requirements
(from Step 3), and another at 50%. For each of these, optimisation has been based on
maximisation of benefits, and the long term performance trends predicted.
Define Budget
The budget for the defined budget period has been set at 900 million US dollars.
Details of the budget optimisation set-up are specified on this screen. For this case study,
optimisation is by incremental benefit/cost ranking (see Section 4.2 of Chapter B2). The
parameters to be specified are given below:
Minimum incremental value = 0
Efficiency frontier zone = 95
Work Programme
After performing the budget optimisation, a revised work programme is produced. This has a
total cost that falls within the specified budget. The road sections are listed in priority order in
accordance with the incremental benefit/cost ranking procedure described in Section 4.2 of
Chapter B2.
Applications Guide 23
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX C1 - STRATEGY ANALYSIS CASE STUDY 1
Define Budget
The budget for the defined budget period has been set at 450 million US dollars, representing
approximately 50% of the funding requirements from Step 3.
Work Programme
After performing the budget optimisation, the revised work programme is produced which
shows the road sections listed in priority order in accordance with the incremental benefit/cost
ranking procedure.
Applications Guide 24
Version 1.0
CONCEPTS OF ANALYSIS APPENDIX C1 - STRATEGY ANALYSIS CASE STUDY 1
Applications Guide 25
Version 1.0
volume three
Software User Guide
David C. Wightman
Eric E. Stannard
Jo M. Dakin
Association
mondiale
de la Route
Wo r l d R o a d
Association
3
THE HIGHWAY DEVELOPMENT AND MANAGEMENT SERIES
ABOUT THIS MANUAL
Overview
Overview of
of HDM-4
HDM-4
Volume
Volume 11
Applications
Applications Guide
Guide Software
Software User
User Guide
Guide
Volume
Volume 22 Volume 3
Volume 3
Analytical
Analytical Framework
Framework and
and
Model
Model Descriptions
Descriptions
Volume
Volume 44
A
A Guide
Guide to
to Calibration
Calibration and
and Adaptation
Adaptation
Volume
Volume 55
A short executive summary describing the HDM-4 system. It is intended to be used by all
readers new to HDM-4, particularly high level management within a road organisation.
! Applications Guide (Volume 2)
Describes the analytical framework and the technical relationships of objects within the
HDM-4 model. It contains very comprehensive reference material describing, in detail,
the characteristics of the modelling and strategy incorporated in HDM-4. It is to be used
by specialists or experts whose task is to carry out a detailed study for a road
management organisation.
! A Guide to Calibration and Adaptation (Volume 5)
Suggests methods for calibrating and adapting HDM models (as used in HDM-III and
HDM-4), to allow for local conditions existing in different countries. It discusses how to
calibrate HDM-4 through its various calibration factors. It is intended to be used by
experienced practitioners who wish to understand the detailed framework and models
built into the HDM-4 system.
Notes:
1 Volumes 1, 2 and 3 are designed for the general user.
2 Volumes 4 and 5 will be of greatest relevance to experts who wish to obtain low level
technical detail. However, Volume 5, in particular, presents very important concepts,
which will be of interest to all users.
ISOHDM Products
The products of the International Study of Highway Development and Management Tools
(ISOHDM) consist of the HDM-4 software suite, associated example case study databases,
and the Highway Development and Management Series collection of guides and reference
manuals. This volume is a member of that document collection.
Customer contact
Should you have any difficulties with the information provided in this suite of documentation
please report details of the problem by E-mail, or an annotated copy of the manual page by fax
to the number provided below.
The ISOHDM Technical Secretariat welcomes any comments or suggestions from users of
HDM-4. Comments on the Software User Guide should be sent to the following address:
E-mail: isohdm@bham.ac.uk
Telephone: +44 - 121 - 414 6717 (5049)
Fax: +44 - 121 - 414 3675 (5060)
Post: ISOHDM Technical Secretariat.
School of Civil Engineering
The University of Birmingham
Edgbaston
Birmingham B15 2TT
United Kingdom
Change details
This is the first formal edition (Version 1.0) of the HDM-4 documentation.
Related documentation
HDM-4 documents:
The Highway Development and Management Series Collection is ISBN: 2-84060-058-7, and
comprises:
Volume 1 - Overview of HDM-4, ISBN: 2-84060-059-5
Volume 2 - Applications Guide, ISBN: 2-84060-060-9
Volume 3 - Software User Guide, ISBN: 2-84060-061-7
Volume 4 - Analytical Framework and Model Descriptions, ISBN: 2-84060-062-5
Volume 5 - A Guide to Calibration and Adaptation, ISBN: 2-84060-063-3
Future documentation
The following documents will be issued at a later release:
Volume 6 - Modelling Road Deterioration and Works Effects, ISBN: 2-84060-102-8
Volume 7 - Modelling Road User and Environmental Effects, ISBN: 2-84060-103-6
Terminology handbooks
PIARC Lexicon of Road and Traffic Engineering - First edition. Permanent International
Association of Road Congresses (PIARC), Paris 1991. ISBN: 2-84060-000-5
Technical Dictionary of Road Terms - Seventh edition, English - French. PIARC Commission
on Terminology, Paris 1997. ISBN: 2-84060-053-6
Acknowledgements
The development of HDM-4 has been sponsored by several agencies, primarily:
! Asian Development Bank (ADB)
! Department for International Development (DFID) in the United Kingdom
! Swedish National Road Administration (SNRA)
! The World Bank
Many other organisations and individuals in a number of countries have also contributed in
terms of providing information, or undertaking technical review of products being produced.
The study has been co-ordinated by the ISOHDM Technical Secretariat at the University of
Birmingham in the United Kingdom. A number of organisations participated in the research
including:
! Finnra
! SNRA
Responsible for developing deterioration relationships for cold climates, road safety,
environmental effects, and supporting HRG with system design.
All research organisations received support from local and regional staff, visiting experts and
external advisers, to ensure that a high standard of quality and international consensus was
achieved. A number of other countries and individuals have supported this work through
supplying expert advice and reviewing the products.
Copyright statement
These HDM-4 products have been produced by the International Study of Highway
Development and Management Tools (ISOHDM), sponsored by The World Bank, the Asian
Development Bank, the Department for International Development (UK), the Swedish
National Road Administration, and other sponsors. The HDM-4 products are jointly published
by The World Road Association (PIARC), Paris and The World Bank, Washington, DC.
Copyright © 2000 The World Road Association (PIARC) on behalf of the ISOHDM sponsors.
All rights reserved.
This copyright covers all documents and document components, computer software and data
delivered as components of the HDM-4 product, in any physical or electronic forms.
Contents
1 About HDM-4
2 Installing HDM-4
2.1 Installing HDM-4 for the first time 2
2.2 Installing an HDM-4 update 3
2.3 Installing an HDM-4 upgrade 4
3 Running HDM-4
3.1 Working with the HDM-4 Workspace window 6
4 Road Networks
4.1 Key concepts 7
4.2 Software overview 7
4.3 Creating new Road Networks with no Sections 9
4.4 Creating Road Networks based on existing networks 10
4.5 Viewing / editing Road Networks 10
4.6 Deleting Road Networks 11
4.7 Renaming Road Networks 11
4.8 Exporting Road Networks 12
4.9 Importing Road Networks 12
4.10 Adding Sections based on aggregate data 13
4.11 Adding Sections based on existing Sections 14
4.12 Editing Section attributes 15
4.13 Specifying which Sections are displayed in the Road Network window 17
4.14 Editing Section details directly from the Road Network window 18
4.15 Deleting Sections 18
5 Vehicle Fleets
5.1 Key Concepts 20
5.2 Software overview 21
5.3 Creating Vehicle Fleets 22
7 Project analysis
7.1 Key concepts 49
7.2 Software overview 49
8 Programme analysis
8.1 Key concepts 73
8.2 Software overview 73
8.3 Creating Programmes 75
8.4 Creating Programmes based on existing Programmes 76
8.5 Viewing / editing Programmes 77
8.6 Deleting Programmes 77
8.7 Renaming Programmes 78
8.8 Exporting Programmes 78
8.9 Importing Programmes 79
8.10 Selecting Sections for a Programme 79
8.11 Selecting Vehicle Types to include in a Programme 81
8.12 Defining normal traffic 82
8.13 Life-cycle analysis 83
8.14 Multi-year forward Programme analysis 87
8.15 Generating a work programme 88
8.16 Optimisation using budget constraints 90
8.17 Generating reports 91
9 Strategy analysis
9.1 Key concepts 93
11 Configuring HDM-4
11.1 Traffic Flow Patterns 115
11.2 Speed Flow Types 120
11.3 Climate Zones 125
11.4 Currencies 130
11.5 Section aggregate data and tables 133
The button functions are also available from the Workspace menu.
B
C E
A. Each row displays a separate Section.
B. These buttons allow you to add, delete and edit Sections.
C. The status bar shows information about the currently active control.
D. The title bar shows the name of the Road Network and the category
of data currently on display.
E. You can change the size of the window by dragging the Resize Box.
You can edit most (but not all) Section attributes here in the spreadsheet. To
edit attributes that are not available in the spreadsheet, double-click a Section.
The resulting Section dialog box provides access to all Section attributes.
Type the name of the new Road Network in the text box and click OK.
The Road Network window is displayed.
You can now add Sections to the blank Road Network (see Sections 4.10
and 4.11).
When you have finished working with the Road Network, click Save before
closing the window. If do not click Save, your changes will be lost.
The new Road Network is added to the Workspace window. To view the new
copy, see Section 4.5.
You are prompted to confirm the deletion. Click Yes to delete the Road
Network, or No to cancel the delete action.
If you click Yes, the deleted Road Network is no longer displayed in the
Workspace window.
! When you delete a Road Network, all corresponding Sections are also
deleted.
! Projects, Programmes and Strategies that use the deleted Road Network
will be affected. You will be informed if this is the case.
! You cannot reverse a deletion.
3 Select the based on aggregate data option button and click OK.
The New Section From Aggregate Data dialog box is displayed.
The fields in this dialog box vary according to the calculation method
you selected.
8 Complete the fields as required.
9 Click Next.
The Step 3 of 3 tab-page is displayed. It shows the calculated SNP
results for the calculation method you selected.
! When you delete a Section, all corresponding data are also deleted.
! Studies that use the deleted Section will be affected. You will be
informed if this is the case.
! You cannot reverse a deletion.
You can edit most (but not all) attributes here in the spreadsheet. To edit
attributes that are not available in the spreadsheet, you can double-click a
Vehicle Type. The resulting Vehicle Attributes dialog box provides access to
all vehicle attributes.
3 Type the name of the new Vehicle Fleet and choose a unit cost Currency
from the drop-down list. This is the Currency in which unit costs should
be specified for all Vehicle Types that are added to the fleet.
4 Click OK.
The Vehicle Fleet window is displayed. You can now add Vehicle Types
to the blank Vehicle Fleet (see Section 5.10 and 5.11).
When you have finished working with the Vehicle Fleet, click Save before
closing the window. If do not click Save, your changes will be lost.
To edit the Vehicle Types, see Section 5.12.
You are prompted to confirm the deletion. Click Yes to delete the
Vehicle Fleet, or No to cancel the delete action.
If you click Yes, the deleted Vehicle Fleet is no longer displayed in the
Workspace window.
! When you delete a Vehicle Fleet, all corresponding Vehicle Types are
also deleted.
! Studies that use the deleted Vehicle Fleet will be affected. You will be
informed if this is the case.
! You cannot reverse a deletion.
4 Enter a unique name for the Vehicle Fleet and click OK.
In step 4, above, you are informed if a Vehicle Fleet of the same name already
exists. You must then decide whether to overwrite the existing Vehicle Fleet,
to provide an alternative name for the Vehicle Fleet that is being imported, or
to cancel the import operation.
1 In the Workspace window, expand the Vehicle Fleets folder and select
a Vehicle Fleet.
2 Click Export.
The Export Vehicle Fleet As dialog box is displayed.
3 Select the Motorised option button and select a base Vehicle Type from
the drop-down list.
The drop-down list contains default motorised Vehicle Types and user
defined Vehicle Types for the current Vehicle Fleet. Initial data for the
new Vehicle Type is derived from the base Vehicle Type.
4 Click OK.
The Vehicle Attributes dialog box is displayed.
3 Select the Non-motorised option button and select a base Vehicle Type
from the drop-down list.
The drop-down list contains four default non-motorised Vehicle Types
and user defined Vehicle Types for the current Vehicle Fleet. Initial data
for the new Vehicle Type is derived from the base Vehicle Type.
4 Click OK.
The NM Vehicle Attributes dialog box is displayed.
9 Click Finish to accept the calculated values and insert them in the Basic
Characteristics tab-page of the Vehicle Attributes dialog box.
5 Enter the axle load equivalency exponent (LE) in the edit box.
6 Click Next.
The Step 2 of 3 tab-page is displayed.
12 Click Finish to accept the calculated ESAL factor and insert it in the
Basic Characteristics tab-page of the Vehicle Attributes dialog box.
13 When you have finished working in the Vehicle Attributes dialog box,
click OK to save any changes.
5.14 Editing Vehicle Type attributes directly from the Vehicle Fleet
window
You can edit Vehicle Type details in the Vehicle Fleet window spreadsheet.
1 In the Workspace window, open a Vehicle Fleet.
The Vehicle Fleet window is displayed.
2 To select the group of attributes, do one of the following:
❏ In the main HDM-4 window, click the View menu
❏ Right-click anywhere in the Vehicle Fleet window
The following groups of attributes are listed in the lower half of the
menu:
❏ Definition
❏ Basic Characteristics
❏ Economic Unit Costs
❏ Financial Units Costs
The active attribute group is marked with a ".
3 Select an attribute group from the menu.
The columns in the Vehicle Fleet window change according to the
attribute group you selected. You can edit the tabular information in the
Vehicle Fleet window.
In step 3 above, if the attribute group you require is greyed out, select a
different vehicle category (Motorised, Non-motorised) first, then select the
attribute group.
4 When you have finished working with the Maintenance Standard, click
OK to save any changes and close the dialog box.
You are prompted to confirm the deletion. Click Yes to delete the
Maintenance Standard, or No to cancel the delete action.
If you click Yes, the deleted Maintenance Standard is no longer
displayed in the Workspace window.
You are prompted to confirm the deletion. Click Yes to delete the Work
Item, or No to cancel the delete action.
If you click Yes, the deleted Work Item is no longer displayed in the
Maintenance Standard dialog box.
! When you delete a Work Item, all corresponding data are also deleted.
! Studies that use the deleted Work Item will be affected. You will be
informed if this is the case.
! You cannot reverse a deletion.
3 Enter a name and short code for the new Improvement Standard. These
must be unique.
4 Work through the tab-pages and complete the necessary fields. The
number of tab-pages and their content changes according to the
improvement type you choose in the General tab-page.
5 Click OK to save the Improvement Standard and close the dialog box.
The new Improvement Standard is added to the Workspace window. To
view it, see Section 6.14.
You are prompted to confirm the deletion. Click Yes to delete the
Improvement Standard, or No to cancel the delete action.
If you click Yes, the deleted Improvement Standard is no longer
displayed in the Workspace window.
! When you delete an Improvement Standard, all corresponding data are
also deleted.
! Studies that use the deleted Improvement Standard will be affected. You
will be informed if this is the case.
! You cannot reverse a deletion.
The following table shows how the workflow buttons guide you through the
process of defining and running a Project analysis.
! By Project
Define Project
Details Determines the best alternative for the group of Sections
analysed.
! By Section
Define Normal 7 Define the traffic composition and expected growth for each
Traffic Section / Vehicle Type.
Run Analysis 10 Run the analysis. If vital information is missing, you are
informed, and must supply the necessary information so that
HDM-4 can complete the run. The time required to perform
the analysis depends on the complexity of the Project.
Generate Reports Select Reports 11 Generate reports and examine the results of the run.
3 Enter a name for the new Project, and select a Road Network and Vehicle
Fleet.
4 Click OK.
The Project dialog box is displayed. The Define Project Details
workflow button is selected and the General tab-page is active.
When you have finished working with the Project, click OK to save any
changes.
In steps 1 and 2 above, you can alternatively, double-click the Project in the
Workspace window.
You are prompted to confirm the deletion. Click Yes to delete the
Project, or No to cancel the delete action.
If you click Yes, the deleted Project is no longer displayed in the
Workspace window.
You can only export one Project at a time. If you want to export a number of
Projects one after the other, select a different folder for each Project. Do not
rename any of the files.
4 Click Select.
The Selected Sections dialog box is displayed. The Sections that match
the selection criteria are listed.
! The Include NMT in analysis option is only enabled if there are non-
motorised vehicle types in the selected fleet.
! You can exclude non-motorised Vehicle Types from the analysis by
clearing the Include NMT in analysis check box.
! You can edit/view the attributes of the selected Vehicle Type. Click
View/Edit Vehicle to display the Vehicle Attributes dialog box.
! You can edit/view the selected Vehicle Fleet. Click View/Edit Fleet to
display the Vehicle Fleet window.
3 Click anywhere in a Section row and click Edit Section Traffic Details.
The Normal Traffic Details dialog box is displayed. The Vehicle Types
that you selected for the Project are listed. Motorised Vehicle Types and
non-motorised Vehicle Types are separated into two tab-pages.
! In step 3 above, tick / clear the View Details check box to display / hide
alternatives, and assigned Maintenance and Improvement Standards.
! In the Section Alternative Details dialog box, you can edit an
alternative name by selecting it and clicking Edit Alternative.
! In the Section Alternatives Details dialog box, you can delete an
alternative by selecting it and clicking Delete Alternative. The
alternative no longer appears in the list.
3 Enter the year from which the Improvement Standard will be considered
for implementation, and select an Improvement Standard.
Road improvements may have associated benefits, costs, and traffic
effects. To specify exogenous benefits and costs for an Improvement
Standard, see Section 7.13.4. To define generated traffic for an
Improvement Standard, see 7.13.5.
4 Click OK.
The dialog box is closed, returning you to the Section Alternatives
Details dialog box. When you select an alternative, the standards that
have been assigned to it are shown in the Assignments column.
! You can make alterations to the standard assignments you have made.
Select the Section, select the assignment, and click Edit Assignment.
! You can remove a standard assignment. Select the Section, select the
assignment, and click Remove Assignment.
! You can copy standard assignments from one Section to another. Select
the Section that you want to copy from and click Copy Assignment.
Select the Section that you want to copy to, and click Paste
Assignment.
3 Enter a description and the number of years from the start of the
improvement work until benefits and costs will start.
4 Click Append Benefit and Append Cost to add benefit and cost rows to
the spreadsheet.
5 Click Add Period to add a growth period column to the spreadsheet.
Add the necessary number of growth periods to reflect the growth rate of
the benefits and costs. You must define a growth period for the second
year of benefits and costs.
6 Enter initial amounts and growth rates for the benefits and costs you have
added.
! You can make changes to the growth period definition by clicking
anywhere in the period column and clicking Edit Period.
! You can delete periods by clicking anywhere in the period column and
clicking Delete Period.
! You can delete benefits and costs by clicking anywhere in the benefit or
cost row and clicking Delete Ben/Cost.
5 Enter a start year for the period, select a growth type and click OK.
A traffic growth period column is added to the spreadsheet.
6 Enter appropriate traffic growth values for the Vehicle Types and click
OK.
If you do not click OK before closing the dialog box, any changes you have
made will be lost.
! You can make changes to the traffic growth period definition. Click
anywhere in the period column and click Edit Period.
! You can delete traffic growth periods. Click anywhere in the period
column and click Delete Period.
4 Enter the year from which the Maintenance Standard will be considered
for implementation, select a Maintenance Standard, and click OK.
The dialog box is closed, returning you to the Alternative Details dialog
box. When you select a Section, the standards that have been assigned to
the Section/alternative combination are shown in the Assignments
column.
4 Enter the year from which the Improvement Standard will be considered
for implementation, and select an Improvement Standard.
Road improvements may have associated benefits, costs, and traffic
effects. To specify exogenous benefits and costs for an Improvement
Standard, see Section 7.13.4. To define generated traffic for an
Improvement Standard, see Section 7.13.5.
5 Click OK.
The dialog box is closed, returning you to the Alternative Details dialog
box. When you select a Section, the standards that have been assigned to
the Section/alternative combination are shown in the Assignments
column.
! You can make alterations to standard assignments you have made. Select
the Section, select the assignment, and click Edit Assignment.
! You can remove a standard assignment. Select the Section, select the
assignment, and click Remove Assignment.
! You can copy standard assignments from one Section to another. Select
the Section that you want to copy from and click Copy Assignment.
Select the Section that you want to copy to and click Paste Assignment.
3 Enter a description and period start year. The start year cannot be before
the second year of analysis.
4 Click OK to return to the Traffic Diversion for Alternative dialog box.
A column is added to the spreadsheet for the new diversion period.
5 Click a Section row in the diversion period column.
6 Click Edit Diversion Details.
The Traffic Diversion Details dialog box is displayed. The Vehicle
Types that you selected for the Project are listed. Motorised and non-
motorised Vehicle Types are separated onto two tab-pages.
7 Select the growth type and enter the initial traffic compositions and
growth.
8 Click OK.
9 Repeat steps 5 to 8 for each Section that you want to define diverted
traffic for.
2 Enter a name for the new Section, the start year, and the duration of
construction. Complete the other fields as necessary.
3 Click Normal Traffic.
The Normal Traffic Details dialog box is displayed. The Vehicle Types
that you selected for the Project are listed. Motorised Vehicle Types and
non-motorised Vehicle Types are separated into two tab-pages.
7 Work through the tab-pages and enter the necessary values. See Section
4.12.1 for details of how to use the SNP calculation wizard in the
Pavement tab-page.
8 When you have finished defining the new Section, click OK to save the
changes and return to the New Construction Section Option dialog
box.
9 Click OK to save the changes and return to the Alternative Details
dialog box.
The new Section is added to the list. The icon that is used for a Section
added in this way is slightly different to the icon used for Sections
already included in the Project.
! You can make alterations to a new Section. Select the Section and click
Edit New Section.
! To remove a new Section, select the Section and click Remove New
Section.
4 Click Start.
As the Project is analysed, messages are displayed in the list box and the
status of the analysis is shown below the list box.
! If vital information has not been supplied, you are warned and the
analysis run stops. The message tells you which information is missing.
Supply the information and start the run again.
! If non-vital information has not been supplied, you are warned, but the
run continues.
! At the end of the analysis, you can view a complete warning message.
Select the message in the list box and click View.
! You can abort an analysis at any time by clicking Abort.
The following table shows how the workflow buttons guide you through the
process of defining and running a Programme analysis.
Alternatives (for a The steps depend on the analysis mode you are using:
Life Cycle
Analysis) For a Life Cycle Analysis, define the set of alternatives to be
analysed.
Specify Standards
8
Assignments Work Programming For a Multi-year Forward Programme, define the
(for a Multi-year Maintenance Standards and Improvements for each Section.
Forward
Programme)
Generate Reports Select Reports 16 Generate reports and examine the results of the run
3 Enter a name for the new Programme, and select a Road Network and
Vehicle Fleet.
4 Click OK.
The Programme window is displayed. The Define Programme Details
workflow button is selected and the General tab-page is active.
❏ Double-click a Programme
The Programme window is displayed. The Define Programme Details
workflow button is selected and the General tab-page is active.
3 When you have finished working with the Programme, click Save to
save any changes.
You are prompted to confirm the deletion. Click Yes to delete the
Programme, or No to cancel the delete action.
The deleted Programme is no longer displayed in the Workspace
window.
! When you delete a Programme, associated alternatives, traffic and run
settings are also deleted.
! You cannot undo a deletion.
2 Click Add New and enter a name and description for the group.
4 Enter the year from which the Maintenance Standard will be considered
for implementation, select a Maintenance Standard, and click OK.
The dialog box is closed, returning you to the Alternative Details dialog
box. When you select a Section, the standards that have been assigned to
the Section/alternative combination are shown in the Assignments
column.
4 Enter the year from which the Improvement Standard will be considered
for implementation, and select an Improvement Standard.
Road improvements may have associated benefits, costs, and traffic
effects. To specify exogenous benefits and costs for an improvement, see
Section 7.13.4. To define generated traffic for an improvement, see
Section 7.13.5.
5 Click OK.
The dialog box is closed, returning you to the Alternative Details dialog
box. When you select a Section, the standards that have been assigned to
the Section/alternative combination are shown in the Assignments
column.
! You can remove a standard assignment. Select the Section, select the
assignment, and click Remove Assignment.
! You can copy standard assignments from one Section to another. Select
the Section that you want to copy from and click Copy Assignment.
Select the Section that you want to copy to and click Paste Assignment.
4 Complete the fields according to the type of analysis you are doing:
7 Click Start.
! If vital information has not been supplied, you are warned and the
analysis run stops. The message tells you which information is missing.
Supply the information and start the run again.
! If non-vital information has not been supplied, you are warned, but the
run continues.
! At the end of the Programme analysis, you can view a complete warning
message. Select the message in the list box and click View.
! You can abort a Programme analysis at any time by clicking Abort.
The following table shows how the workflow buttons guide you through the
process of defining and running a Strategy analysis.
Define Strategy General 1 Specify the Road Network that contains the Sections to be
Details analysed.
! Maximise NPV
! Maximise dIRI
Define Normal 7 Define the traffic composition and expected growth for the
Traffic Section/Vehicle Type combinations.
Specify Standards Alternatives 8 Define the alternatives to be analysed. Some alternatives may
Assignments include diverted and/or generated traffic, exogenous benefits
and costs.
Generate Reports Select Reports 16 Generate reports and examine the results of the run.
3 Enter a name and select a Vehicle Fleet. Select one of the option buttons:
❏ Create new network matrix—allows you to create representative
Sections for the Strategy analysis.
❏ Use existing network—allows you to choose Sections from an
existing Road Network. Select the Road Network that you want to
use from the drop-down list.
4 Click OK.
The Strategy window is displayed. The Define Strategy Details
workflow button is selected and the General tab-page is active.
If you chose to select Sections from an existing Road Network, the
Select Sections tab-page is displayed (see Section 9.11).
When you have finished working with the Strategy analysis, click Save
before closing the window. If do not click Save, your changes will be lost.
The new Strategy analysis is added to the Workspace window. To view it,
see Section 9.5.
In step 2 above, if the Strategies folder or any existing Strategy analysis is
selected in the Workspace window, the Strategy item will be automatically
selected in the New HDM-4 Item dialog box, and you will only need to click
OK.
3 When you have finished working with the Strategy analysis, click OK to
save any changes.
In steps 1 and 2 above, you can double-click the Strategy analysis in the
Workspace window.
You are prompted to confirm the deletion. Click Yes to delete the
Strategy analysis, or No to cancel the delete action.
The deleted Strategy analysis is no longer displayed in the Workspace
window.
3 Click anywhere in a Section row and click Edit Section Traffic Details.
The Normal Traffic Details dialog box is displayed. The Vehicle Types
that you selected for the Strategy analysis are listed. Motorised Vehicle
Types and non-motorised Vehicle Types are separated into two tab-
pages.
9 You can select the work programme for a Section and exclude it from
budget optimisation. Click the Section row and click Manual
Assignment. The row is shown in italics.
! If vital information has not been supplied, you are warned and the
analysis run stops. The message tells you which information is missing.
Supply the information and start the run again.
! If non-vital information has not been supplied, you are warned, but the
run continues.
! At the end of the Strategy analysis, you can view a complete warning
message. Select the message in the list box and click View.
! You can abort a Strategy analysis at any time by clicking Abort.
The Print and Export controls are explained in the following sections.
2 Specify whether you want to print all the pages in the report, or a range.
You can print several copies of the report and have them collated.
3 Click OK.
4 Select the report template that you want to add, and the folder in which
you would like it to appear. Enter a name for the report.
5 Click OK.
To remove a user-defined report, select it and click Remove Report.
A number of default road uses are defined in HDM-4. Each has an associated
set of flow period data. Changing the selection of road use for a Traffic Flow
Pattern causes the corresponding flow period default data to be adopted. To
reset the flow period data to the default values for the currently selected road
use, click Defaults.
5 Click OK to save the Traffic Flow Pattern and close the window.
The new Traffic Flow Pattern is added to the Workspace window. To
view it, see Section 11.1.4.
! In step 2 above, if the Traffic Flow Patterns folder or any existing
Traffic Flow Pattern is selected in the Workspace window, the Traffic
Flow Pattern item will be automatically selected in the New HDM-4 Item
dialog box, and you will only need to click OK.
! Using more flow periods improves accuracy but slows down analysis
time.
You are prompted to confirm the deletion. Click Yes to delete the Traffic
Flow Pattern, or No to cancel the delete action.
If you click Yes, the deleted Traffic Flow Pattern is no longer displayed
in the Workspace window.
! When you delete a Traffic Flow Pattern, all corresponding data are also
deleted.
! There may be Sections which use the Traffic Flow Pattern to be deleted.
These Sections will be affected by the deletion. You will be informed if
this is the case.
! You cannot reverse a deletion.
All the Traffic Flow Patterns are exported irrespective of the item that you
select in the Traffic Flow Patterns folder.
11.2.3 Creating Speed Flow Types based on existing Speed Flow Types
If you want to create a Speed Flow Type that is similar to an existing one, you
can copy the existing type and edit the copy as necessary.
1 In the Workspace window, expand the Configuration folder.
2 Expand the Speed Flow Types folder.
3 Select the Speed Flow Type that you wish to copy.
4 To copy the selected Speed Flow Type, do one of the following:
❏ Click Copy in the Workspace window
❏ With the Workspace window active, select Copy from the
Workspace menu
You are prompted to confirm the deletion. Click Yes to delete the Speed
Flow Type, or No to cancel the delete action.
If you click Yes, the deleted Speed Flow Type is no longer displayed in
the Workspace window.
! When you delete a Speed Flow Type, all corresponding data are also
deleted.
! There may be Sections which use the Speed Flow Type to be deleted.
These Sections will be affected by the deletion. You will be informed if
this is the case.
! You cannot reverse a deletion.
2 Click Export.
The Export Speed Flow Type As dialog box is displayed.
For each of the two categories, aggregate level parameters are provided.
These are Temperature Classification and Moisture Classification. HDM-4
supports a fixed set of Temperature Classifications, each of which has an
associated set of defaults for the temperature parameters. Changing the
Temperature Classification selected for a Climate Zone causes the
corresponding set of temperature data to be adopted.
Similarly, HDM-4 supports a fixed set of Moisture Classifications, each of
which has an associated set of defaults for the moisture parameters. Changing
the Moisture Classification currently selected for a Climate Zone causes the
corresponding set of moisture defaults to be adopted.
To reset the default values for Temperature and Moisture Classifications,
click Defaults.
You are prompted to confirm the deletion. Click Yes to delete the
Climate Zone, or No to cancel the delete action.
! When you delete a Climate Zone, all corresponding data are also deleted.
! There may be Sections which use the Climate Zone to be deleted. These
Sections will be affected by the deletion. You will be informed if this is
the case.
! You cannot reverse a deletion.
2 Click Export.
The Export Climate Zone As dialog box is displayed.
11.4 Currencies
3 Click Add.
A row is added to the table.
4 Enter a description and symbol for the Currency. Select a symbol
position from the drop-down list.
5 Click OK to save the Currency definition and close the dialog box.
2 Click Export.
The Export Currency As dialog box is displayed.
! Traffic volume
! Road class
! Geometry class
! Construction quality
! Structural adequacy
! Ride quality
! Surface condition
! Surface texture
The parameter values that you define (for example good, fair, bad) correspond
to detailed values for related parameters in the Section Aggregate Tables
in the Configuration folder.
Each HDM-4 system can, therefore, have a different definition of what
constitutes good, bad or fair to suit the local conditions.
When you select a parameter value in the New Section from Aggregate Data
dialog box, the corresponding detailed values from the Section Aggregate
Tables are used when you run an analysis.
! You can also delete a parameter value. Select the value and click Delete.
! You can rename a parameter value. Select the value and click Rename.
! You can also access the Section Aggregate Tables by expanding the
Configuration folder and double-clicking Section Aggregate Tables.
All the tables are displayed in tab-pages in the resulting dialog box. The
Bituminous layers table can only be accessed in this way.
volume three
A short executive summary describing the HDM-4 system. It is intended to be used by all readers new to HDM-4,
particularly high-level management within a road organisation.
ISBN: 2-84060-059-5
PIARC, Registered in France These HDM-4 products have been produced by the International Study of Highway
Development and Management Tools (ISOHDM), sponsored by The World Bank, the
Registered Office Asian Development Bank, the Department for International Development (UK), the
The World Road Association (PIARC) Swedish National Road Administration, and other sponsors. HDM-4 is jointly published
La Grande Arche by The World Road Association (PIARC), Paris, and The World Bank, Washington, D.C.
Paroi Nord, Niveau 8
92055 La Defénse Cedex PIARC endeavours to ensure that the information in this document is correct and fairly
France stated but does not accept liability for any error or omission.
Tel: +33 1 41 02 05 84 The development of PIARC products and services is continuous and published
Fax: +33 1 49 00 02 02 information may not be up to date. It is important to check the current position with
Email: piarc.hdm4@attglobal.net PIARC. This document is not part of a contract or licence save insofar as may be
Web: http://hdm4.piarc.org expressly agreed.
Wo r l d R o a d
Association
4
THE HIGHWAY DEVELOPMENT AND MANAGEMENT SERIES
Part A
Overview
Overview of
of HDM-4
HDM-4
Volume
Volume 11
Applications
Applications Guide
Guide Software
Software User
User Guide
Guide
Volume
Volume 22 Volume
Volume 33
Analytical
Analytical Framework
Framework andand
Model Descriptions
Model Descriptions
Volume
Volume 44
A
A Guide
Guide to
to Calibration
Calibration and
and Adaptation
Adaptation
Volume
Volume 55
A short executive summary describing the HDM-4 system. It is intended to be used by all
readers new to HDM-4, particularly high level management within a road organisation.
! Applications Guide (Volume 2)
Describes the analytical framework and the technical relationships of objects within the
HDM-4 model. It contains very comprehensive reference material describing, in detail,
the characteristics of the modelling and strategy incorporated in HDM-4. It is to be used
by specialists or experts whose task is to carry out a detailed study for a road
management organisation.
! A Guide to Calibration and Adaptation (Volume 5)
Suggests methods for calibrating and adapting HDM models (as used in HDM-III and
HDM-4), to allow for local conditions existing in different countries.
Notes:
1 Volumes 1, 2 and 3 are designed for the general user
2 Volumes 4 and 5 are only to be used by experts who wish to obtain low level technical
detail
Analytical
AnalyticalFramework
Framework
and
andModel
Model
Descriptions
Descriptions
Introduction
Introduction
Part
PartAA
Traffic
Traffic
Part
PartBB
RD
RDModel
Model WE
WEModel
Model RUE
RUEModel
Model SEE
SEEModel
Model
Part
PartCC Part
PartDD Part
PartEE Part
PartFF
Economic
EconomicAnalysis
Analysis
RD = Road Deterioration Part
PartGG
Nomenclature
Nomenclature
WE = Road Works Effects Part
PartHH
The nine parts in the Analytical Framework and Model Descriptions contain the following
information:
! Part A - Introduction
Describes the traffic characteristics used in HDM-4, and provides details of the traffic
data that is required.
! Part C - Road Deterioration Models
❏ C3 Concrete Pavements
Describes the specifications used to model rigid (or Portland) cement concrete
pavements. This includes details of the different types of rigid concrete pavement
construction that are modelled in HDM-4.
❏ C4 Unsealed Roads
Describes the specifications used in the unsealed Road Deterioration models. It
provides an overview of the HDM-4 Road Deterioration modelling logic, and the
relationships and default coefficient values for each of the distresses modelled.
! Part D - Road Works Effects
❏ D1 Types of Works
Describes the types of road works and their effects on road pavements, including the
calculation of physical quantities of road works and the corresponding costs.
The three types of pavements considered are:
1 Bituminous pavements (see Chapter D2)
2 Concrete pavements (see Chapter D3)
3 Unsealed roads (see Chapter D4)
❏ D2 Bituminous Pavements
Describes the detailed modelling of Road Works Effects for bituminous pavements.
This includes methods of defining works activities and intervention criteria, the
calculation of physical quantities of road works and their corresponding costs, and
the effects of road works on pavement characteristics and road users. Road works
are grouped into the following classes:
" Routine maintenance
" Periodic maintenance
" Special works
" Improvement works
" Construction works
❏ D3 Concrete Pavements
Describes the detailed modelling of Road Works Effects for concrete pavements.
This includes methods of defining works activities (maintenance and rehabilitation)
and intervention criteria, and classification of the road works. The following three
types of concrete pavements are analysed:
1 Jointed Plain Concrete Pavements
2 Jointed Reinforced Concrete Pavements
3 Continuously Reinforced Concrete Pavements
The following classes of road works are performed:
" Routine maintenance
" Restorations
" Overlays
" Reconstruction
❏ D4 Unsealed Roads
Describes the detailed modelling of road works effects for unsealed roads, including
the methods of defining works activities and intervention criteria, the calculation of
physical quantities of road works and their corresponding costs, and the modelling
of the effects of road works on pavement characteristics and road users. Unsealed
road works are grouped into the following classes:
" Maintenance
" Improvement works
" Construction works
! Part E - Road User Effects
❏ E1 Introduction
Provides an overview of the implementation of Road User Effects (RUE) models for
calculating motorised vehicle operating costs and travel time, including an overview
of the HDM-4 representative vehicle framework and the default representative
vehicle types.
❏ E2 Vehicle Speeds and Operating Costs
Describes the implementation of Road User Effects (RUE) models for calculating
motorised vehicle speeds, operating costs and travel time. It provides an overview of
the modelling concepts and logic, the relationships used, and the default parameter
values for each of the RUE components.
❏ E3 Non-Motorised Transport
Discusses the use of non-motorised modes of transport and their effect on the
motorised transport. The impact of non-motorised transport on other road users and
road characteristics is included.
❏ E4 Road Safety
Describes the implementation of road safety models through a series of look-up
tables of accident rates with user defined accident costs.
! Part F - Social and Environmental Effects
❏ F1 Introduction
Gives an overview of the modelling logic used to quantify the energy consumed by
both motorised and non-motorised vehicles, together with the emissions produced
by motorised vehicles.
❏ F2 Energy Balance Analysis
Describes the energy implications of alternative transport projects and policies. The
models estimate separately the global and national consumption of non-renewable
energy by motorised vehicles, the consumption of renewable energy by non-
motorised vehicles, and energy use in roadworks.
❏ F3 Vehicle Emissions
Describes the models used to quantify vehicle emissions. The document presents the
emissions models together with default parameters for the 16 standard vehicle types.
Details of the input data and the output reports are also given.
! Part G - Economic Analysis
Describes how the annual costs streams calculated by the HDM-4 components are
compared to determine the benefits and costs associated with a road investment. Standard
discounting methods are then applied to calculate key economic indicators such as net
present values, internal rates of return and benefit/cost ratios. Investment optimisation
methods are also described.
! Part H Nomenclature
Provides details of the units and numbering system used within the HDM-4 suite of
documents.
! Part I Glossary
ISOHDM Products
The products of the International Study of Highway Development and Management Tools
(ISOHDM) consist of the HDM-4 suite of software, associated example case study databases,
and the Highway Development and Management Series collection of guides and reference
manuals. This Volume is a member of that document collection.
Customer contact
Should you have any difficulties with the information provided in this suite of documentation
please do not hesitate to report details of the problem you are experiencing. You may send an
E-mail or an annotated copy of the manual page by fax to the number provided below.
The ISOHDM Technical Secretariat welcomes any comments or suggestions from users of
HDM-4. Comments on the Analytical Framework and Model Descriptions should be sent to
the following address:
E-mail: isohdm@bham.ac.uk
Telephone: +44 - 121 - 414 6717 (5049)
Fax: +44 - 121 - 414 3675 (5060)
Post: ISOHDM Technical Secretariat.
School of Civil Engineering
The University of Birmingham
Edgbaston
Birmingham B15 2TT
United Kingdom
Change details
This is the first formal edition (Version 1.0) of the HDM-4 documentation.
Related documentation
HDM-4 documents:
The Highway Development and Management Series Collection is ISBN: 2-84060-058-7, and
comprises:
Volume 1 - Overview of HDM-4, ISBN: 2-84060-059-5
Volume 2 - Applications Guide, ISBN: 2-84060-060-9
Volume 3 - Software User Guide, ISBN: 2-84060-061-7
Volume 4 - Analytical Framework and Model Descriptions, ISBN: 2-84060-062-5
Volume 5 - A Guide to Calibration and Adaptation, ISBN: 2-84060-063-3
Future documentation
The following documents will be issued at a later release:
Volume 6 - Modelling Road Deterioration and Works Effects in HDM-4,
ISBN: 2-84060-102-8
Volume 7 - Modelling Road User and Environmental Effects in HDM-4,
ISBN: 2-84060-103-6
Terminology handbooks
PIARC Lexicon of Road and Traffic Engineering - First edition. Permanent International
Association of Road Congresses (PIARC), Paris 1991. ISBN: 2-84060-000-5
Technical Dictionary of Road Terms - Seventh edition, English - French. PIARC Commission
on Terminology, Paris 1997. ISBN: 2-84060-053-6
Acknowledgements
The authors would like to express their appreciation to all those who have contributed to
developing this manual, in particular, but not limited to, the following people:
Economic Analysis
! Vesa Mannisto of Inframan Oy (Finland)
Contributed to investment optimisation methods.
Development of HDM-4
The development of HDM-4 has been sponsored by several agencies, primarily:
! Asian Development Bank (ADB)
! Department for International Development (DFID) in the United Kingdom
! Swedish National Road Administration (SNRA)
! The World Bank
Many other organisations and individuals in a number of countries have also contributed in
terms of providing information, or undertaking technical review of products being produced.
The study has been co-ordinated by the ISOHDM Technical Secretariat at the University of
Birmingham in the United Kingdom. A number of organisations participated in the research
including:
! Finnra
Responsible for developing deterioration relationships for cold climates, road safety,
environmental effects, and supporting HRG with system design.
All research organisations received support from local and regional staff, visiting experts and
external advisers, to ensure that a high standard of quality and international consensus was
achieved. A number of other countries and individuals have supported this work through
supplying expert advice and reviewing the products.
Note: Within each Part/Chapter there will be an Acknowledgements chapter. Each listing
contains references to specific documentation for the topics described in the Part/Chapter.
Copyright statement
These HDM-4 products have been produced by the International Study of Highway
Development and Management Tools (ISOHDM), sponsored by The World Bank, the Asian
Development Bank, the Department for International Development (UK), the Swedish
National Road Administration, and other sponsors. The HDM-4 products are jointly published
by The World Road Association (PIARC), Paris and The World Bank, Washington, DC.
Copyright © 2000 The World Road Association (PIARC) on behalf of the ISOHDM sponsors.
All rights reserved.
This copyright covers all documents and document components, computer software and data
delivered as components of the HDM-4 product, in any physical or electronic forms.
Contents
Volume 4 - Analytical Framework & Model
Descriptions
Part A Overview
A1 Introduction A1-1
Part B Traffic
B1 Traffic characteristics B1-1
Part H Nomenclature
H1 Nomenclature H1-1
Part I Glossary
I1 Glossary I1-1
volume four
A short executive summary describing the HDM-4 system. It is intended to be used by all readers new to HDM-4,
particularly high-level management within a road organisation.
ISBN: 2-84060-059-5
PIARC, Registered in France These HDM-4 products have been produced by the International Study of Highway
Development and Management Tools (ISOHDM), sponsored by The World Bank, the
Registered Office Asian Development Bank, the Department for International Development (UK), the
The World Road Association (PIARC) Swedish National Road Administration, and other sponsors. HDM-4 is jointly published
La Grande Arche by The World Road Association (PIARC), Paris, and The World Bank, Washington, D.C.
Paroi Nord, Niveau 8
92055 La Defénse Cedex PIARC endeavours to ensure that the information in this document is correct and fairly
France stated but does not accept liability for any error or omission.
Tel: +33 1 41 02 05 84 The development of PIARC products and services is continuous and published
Fax: +33 1 49 00 02 02 information may not be up to date. It is important to check the current position with
Email: piarc.hdm4@attglobal.net PIARC. This document is not part of a contract or licence save insofar as may be
Web: http://hdm4.piarc.org expressly agreed.
Contents
A1 Introduction
1 Structure of this manual A1-1
2 Economic appraisal of road projects A1-5
2.1 Purpose A1-5
2.2 Transport cost components A1-6
3 Economic appraisal method A1-10
3.1 Project alternatives A1-10
3.2 Discounting A1-10
4 Applications of economic appraisal A1-16
4.1 Financial and economic costs A1-16
4.2 Project types A1-16
4.3 Sensitivity and risk analysis A1-16
4.4 Prioritisation A1-18
Analytical
AnalyticalFramework
Framework
and
andModel
Model
Descriptions
Descriptions
Introduction
Introduction
Part
PartAA
Traffic
Traffic
Part
PartBB
RD
RDModel
Model WE
WEModel
Model RUE
RUEModel
Model SEE
SEEModel
Model
Part
PartCC Part
PartDD Part
PartEE Part
PartFF
Economic
EconomicAnalysis
Analysis
RD = Road Deterioration Part
PartGG
Nomenclature
Nomenclature
WE = Road Works Effects Part
PartHH
A1 Introduction
Describes the traffic characteristics used in HDM-4, and provides details of the traffic
data that are required.
! Part C - Road Deterioration Models
Volume 1
Overview
Volume 2 of HDM-4 Volume 3
Applications Software
Guide User Guide
Volume 4
Analytical
Framework
and Model
Descriptions
Glossary &
Nomenclature
Volume 5
A Guide to
Calibration
and
Adaptation
❏ C3 Concrete Pavements
Describes the implementation of Road Deterioration models for concrete pavements.
It provides an overview of the modelling framework, a brief analysis of the concrete
pavement types and distresses considered, and a description of the models.
❏ C4 Unsealed Roads
Describes the specifications for modelling unsealed Road Deterioration. It provides
an overview of the modelling logic and the relationships for each of the distresses
modelled.
! Part D - Road Works Effects
❏ D1 Types of Works
Describes the classification of road works and the generic framework for modelling
works quantities, costs and effects for the three road surface classes bituminous,
concrete and unsealed.
❏ D2 Bituminous Pavements
Describes the detailed modelling of Road Works Effects for bituminous pavements.
The methods of defining works activities and intervention criteria, the calculation of
physical quantities of works and the costs to road administration, and works effects
on road characteristics and road use are discussed for different road works classes.
❏ D3 Concrete Pavements
Describes the detailed modelling of Road Works Effects for concrete pavements.
The methods of defining works activities and intervention criteria, the calculation of
physical quantities of works and the costs to road administration, and works effects
on road characteristics and road use are discussed for different works classes.
❏ D4 Unsealed Roads
Describes the detailed modelling of road works effects for unsealed roads.
The methods of defining works activities and intervention criteria, the calculation of
physical quantities of works and the costs to road administration, and works effects
on road characteristics and road use are discussed for different works classes.
! Part E - Road User Effects
❏ E1 Introduction
Describes the Road User Effects (RUE) components modelled in HDM-4. It
provides an overview of the HDM-4 representative vehicle framework, which
considers both motorised vehicles and non-motorised transport.
❏ E2 Vehicle Speeds and Operating Costs
Describes the Road User Effects models for calculating motorised vehicle speeds,
operating costs and travel time.
❏ E3 Non-Motorised Transport
Discusses the use of non-motorised modes of transport and its effects on motorised
transport. Since non-motorised modes of transport account for the vast majority of
the movement of people and goods in many developing countries it was essential to
include this model in HDM-4.
❏ E4 Road Safety
Describes the specification for implementing road safety analysis. The model allows
users to define a series of look-up tables of accident rates.
❏ F1 Introduction
Introduces the modelling of energy balance and vehicle emissions.
❏ F2 Energy Balance Analysis
The energy used in the road transport sector forms a significant share of the total
energy consumption in most countries. It is therefore important for planners and
decision-makers to be able to understand the energy implications of alternative
transport projects, strategies and policies.
❏ F3 Vehicle Emissions
Describes the implementation of Vehicle Emissions analysis. The document presents
the emissions model together with default parameters for the 16 HDM-4 standard
vehicle types.
! Part G - Economic Analysis
Describes how HDM-4 is used to determine the benefits and costs associated with a road
investment, and how these are used in economic analysis and optimisation procedures.
! Part H - Nomenclature
Lists details about the units and numbering methods used within the HDM-4 suite of
documents.
! Part I - Glossary
2.1 Purpose
The purpose of road investment appraisal is to select projects with high economic returns.
The decision of whether to invest in roads or in some other infrastructure development is not
the primary objective of road investment appraisal since in most cases such a decision will
already have been made. The purpose of an economic appraisal of road projects therefore is
to determine how much to invest and what economic returns to expect. The size of the
investment is determined by the costs of construction and annual road maintenance. The
economic returns are mainly in the form of savings in road user costs due to the provision of a
better road facility. These three costs constitute what is commonly referred to as the total
(road) transport cost or the whole life cycle cost.
Three primary objectives of road investment appraisal can therefore be identified as follows:
! To determine the appropriate size of investment and the returns expected from this
investment,
! To determine the appropriate geometric and structural design standards for the size of
investment in order to obtain the expected returns,
! To determine the relative priorities for investment among competing road projects when
there is a budget constraint, and
! To assess the economic and socio-economic impact of investments in roads such as the
improvement to the community of industrial, agricultural, educational, and health
services.
The assessment of the socio-economic benefits of road investments however is difficult to
quantify in monetary terms. This is usually done separately after an economic appraisal has
been carried out using a road investment appraisal model.
The primary function of a road investment appraisal model is therefore to calculate the costs
of road construction, road maintenance and road user costs for a specified analysis period.
This is accomplished by modelling the interrelationships between the environment,
construction standards, maintenance standards, geometric standards and road user costs.
A road investment appraisal model may be used to assist with the selection of appropriate
road design and maintenance standards which minimise the total transport cost. The effect of
providing better road standards on the components of the total transport cost is illustrated in
Figure A1.2. If a low standard road is constructed, high maintenance and road user costs can
be expected resulting in a high total transport cost despite the low cost of construction.
Conversely, if a high standard road is constructed, the very high cost of construction will
outweigh the low maintenance and road user costs. A road investment appraisal model can
therefore be used to search for the road design or maintenance standard for which the total
transport cost is a minimum. This alternative is represented by the dotted vertical line in
Figure A1.2. The interaction between the transport cost components however is more
intricate than is suggested in Figure A1.2. For example a high construction cost will not
necessarily lead to lower road user costs, for example, the construction of a very wide but
structurally weak road.
When planning investments in the roads sector, it is necessary to evaluate all costs associated
with the proposed project. These include construction costs, maintenance and rehabilitation
costs, road user costs, and all other external or exogenous costs or benefits that can be directly
attributed to the road project. It is normal to consider such costs or benefits over an extended
analysis period usually equal to a longer than the expected life of the road, hence the term
'life-cycle' cost analysis. The costs of construction, routine and periodic maintenance are
usually borne by the agency or authority in charge of the road network. However, road user
costs are borne by the community at large in the form of vehicle operating costs (VOC), travel
time costs, accident costs and other indirect costs.
Optimum
Construction Maintenance
& Road User
A road investment model simulates the interaction between pavement construction standards,
maintenance standards and the effects of the environment and traffic loading in order to
predict the annual trend in road condition. This, together with the geometric standards of the
road, has a direct effect on vehicle speeds and on the costs of vehicle operation and accident
rates on the road.
haulage and compaction. The cost of pavement construction is a function of the number,
thickness and type of pavement layers, and includes the costs of shoulders and kerbs.
A sizeable percentage of road construction costs can be attributed to the costs of bridges and
the provision of adequate drainage facilities. Road construction will usually involve the
setting up of temporary site camps and the transportation of equipment, materials and
manpower to the construction site. The costs of these activities together with the profit
margin for the contractor and consultants fees are usually grouped into an overhead cost
which may be specified as a fixed cost per kilometre or as a percentage of the total cost of
construction. At the end of the analysis period, a salvage value may be specified,
representing the percentage of total costs incurred on permanent structures such as
embankments, cuttings, bridges and drainage facilities.
In HDM-4, the costs of construction or any other major road improvements, are user
specified.
BI = 55 × QI …(2.1)
A new road has to be assigned a structural number and an initial roughness. The high
dependence of road user costs on roughness, and of roughness progression on the pavement
structural number, suggests that the results of an economic analysis will be affected by the
values assigned to these two variables at the beginning of the analysis.
distance. Unit costs for the various resources are specified by the user in order to calculate the
annual total costs of vehicle operation. Vehicle depreciation is considered to be a function of
the predicted travel time and of the level of vehicle utilisation.
3.2 Discounting
It is necessary to discount the transport costs in each year of an analysis period to their value
in the base year. This is carried out to reflect the time value of money represented by the
opportunity cost of the capital invested in a road project. Discounting is performed by
multiplying the cost in a given year by the discount factor for that year. Discount factors are
derived from the equation below:
r -N
D.F. = (1 + ) …(3.1)
100
where:
Table A1.1 illustrates the principle of discounted cash flow analysis (DCF) applied to a gravel
road which will be paved after one year. The paved road has a design life of 10 years after
construction. The economic comparisons are based on the totals of the discounted costs. This
is referred to as the Present Value (PV) of costs. For example, in Table A1.1, the PV of the
Without Project alternative is $ 23.8 million, and the PV of the With Project alternative is
$20.2 million, both at 12% discount rate. If the example given in Table A1.1 represents a real
road project, the alternative chosen for implementation would depend on the economic criteria
used to compare the alternatives. The most commonly used criteria for selecting projects are
the net present value (NPV), the internal rate of return (IRR) and the Benefit Cost Ratio
(BCR).
0 (Base) 1.0000 0.8 1.5 2.3 2.3 0.8 1.5 2.3 2.3
1 0.8929 0.9 1.6 2.5 2.2 5.8 0.4 1.0 7.2 6.4
NPV
BCR = +1 …(3.2)
C
From the example given in Table A1.1, the estimated BCR would be 1.55.
! Road investments often take up a large proportion of national income, therefore any
failure will be expensive,
! Alterations during implementation can be very expensive and prohibitive and should
therefore be avoided by selecting the most suitable alternative at the outset,
! To determine impact of possible changes (for example, to the environment, socio-
economic) on the overall viability of road projects and plan for these accordingly.
The formal method of economic appraisal is only one step in the process of quantifying risk.
Some of the causes of uncertainty can be assessed by additional analyses such as;
! Sensitivity analysis
❏ This is applied to study the effects of changes in one parameter (for example,
construction cost, or traffic growth rate) on the overall viability of a road project.
! Scenario analysis
❏ This is carried out to determine the broad range of parameters that together would
affect the viability of the road project. For example, a review of government long-
term policies could yield alternative economic growth rates which would affect both
traffic growth rates and the cost of road construction.
! Risk analysis
4.4 Prioritisation
There are often situations when the budget available for road projects will not be sufficient to
undertake all projects shown to have a positive return (that is, projects with positive NPV). In
such situations, a formal method of selecting projects to be included within the budget can be
applied. Capital budgeting or rationing can be applied to a group of projects that meet either
of the following conditions:
! Projects that are independent of each other (for example, list of road projects from
different parts of the country),
! Mutually exclusive projects (that is, projects that are alternatives to each other) when
only one alternative can be selected.
The NPV capital budgeting rules can be applied in both situations where sufficient funds are
available, and also when there is a budget constraint. The rules are summarised below:
! When sufficient funds are available to undertake all projects;
❏ Select all independent projects with NPV > 0
❏ Select mutually exclusive project alternatives with the highest NPV
! When capital rationing is to be applied due to shortage of funds;
❏ Select independent projects with the highest NPV/Cost ratio
❏ Select mutually exclusive projects using the incremental NPV/Cost method
described below.
The incremental analysis is used to test whether the ratio of the increase in NPV to the
increase in costs between alternative mutually exclusive projects is greater than a specified
marginal ratio. The formula is defined as:
(NPV2 - NPV1 )
IBCR = …(4.1)
(C 2 - C1 )
where:
If the above ratio is greater than a specified marginal value, then the project alternative is
included among those to be funded. The marginal value is usually determined from the BCR
of the road project at the budget boundary.
Further details of the above method are given in the Applications Guide (Volume 2).
Contents
B1 Traffic characteristics
1 Introduction B1-1
2 Representation of traffic B1-1
2.1 Purposes of traffic data B1-1
2.2 Data types B1-2
3 Traffic categories B1-3
4 Traffic composition, volumes and growth rates B1-4
4.1 Basic features B1-4
4.2 Composition, volumes and growth rates B1-5
5 Axle loading B1-9
5.1 The need for axle loading data B1-9
5.2 Vehicle axles B1-9
5.3 Equivalent standard axle load factors B1-9
5.4 Light and heavy vehicles B1-10
5.5 Cumulative traffic loading B1-11
6 Road capacity and speed-flow relationships B1-13
6.1 Basic concepts B1-13
6.2 Data to be specified B1-15
7 Hourly flow-frequency distribution data B1-17
7.1 Basic concepts B1-17
7.2 Data to be specified B1-17
7.3 Limits on traffic flows B1-20
8 References B1-22
Analytical
AnalyticalFramework
Framework
and
andModel
Model
Descriptions
Descriptions
Introduction
Introduction
Part
PartAA
Traffic
Traffic
Part
PartBB
RD
RDModel
Model WE
WEModel
Model RUE
RUEModel
Model SEE
SEEModel
Model
Part
PartCC Part
PartDD Part
PartEE Part
PartFF
Economic
EconomicAnalysis
Analysis
RD = Road Deterioration Part
PartGG
Nomenclature
Nomenclature
WE = Road Works Effects Part
PartHH
Part B Traffic
B1 Traffic characteristics
1 Introduction
This chapter describes the traffic characteristics used in HDM-4, and provides details of the
traffic data that are required. These include the methods used for calculating future traffic and
axle loading together with the approaches for modelling traffic congestion. The traffic data are
used in all four sets of HDM-4 models (Road deterioration, Works effects, Road user effects
and Social and environmental effects, see Figure B).
As travel demand varies both in time and space, traffic data should be representative averages
for a road section. The representation of traffic needs to be at an appropriate level of detail in
accordance with the type of analysis (project, programme or strategy) to be performed.
A list of research documents referenced from this chapter is given in Section 8.
2 Representation of traffic
The required traffic data for this type of analysis is similar to that for project analysis,
with the exception that the data is at a more aggregate level (see Applications Guide). For
example; traffic volumes may be specified by vehicle classes, and the data could be used
by several representative road sections.
! Strategy analysis
This requires the specification of an aggregate set of traffic data, that is representative of
a group of road sections being analysed. Traffic levels should be expressed in terms of
daily flows, and may be described as low, medium, or high. The composition of traffic
may be expressed as a percentage of daily flow for each vehicle class/type.
For the purposes of representing traffic characteristics both for project and network level
analyses, road sections within a network must be categorised according to the following:
! Road types
For example, single lane roads, four-lane roads, motorways, etc. This data is used to
determine the parameters for capacity, speed-flow relationship shape, width effects and
passenger car space equivalents for each road type.
! Traffic-flow pattern
For example; commuter, recreational, inter-city, etc. This data is necessary for describing
the patterns of traffic flow along each road category; commuter routes, for example, tend
to have weekday peaks but low weekend traffic, whereas recreational routes have a more
peaked distribution.
! Non-motorised transport factor
Measures the effect of roadside activity on traffic speeds; this includes the effects of land
use, roadside stalls, bus stops, parking, access points, etc.
! Motorised transport factor
Specified within the individual applications programs (that is, project analysis,
programme analysis and strategy analysis) as required by the purpose of the study being
performed.
! Axle loading (see Section 5)
Specified for each vehicle type in the Vehicle Fleet folder (see Overview of HDM-4).
Provision is made to derive most of the required parameters (for example, equivalent
standard axle load factors, etc.) from the user-specified data.
! Road capacity and speed-flow relationships (see Section 6)
Defined according to the road type within the HDM-4 Configuration (see Overview of
HDM-4).
! Hourly flow-frequency distribution data (see Section 7)
Required to estimate hourly flow ranges from annual average traffic data. These traffic
data types are road dependent and are specified for each road use category within HDM-4
Configuration (see Overview of HDM-4).
The specification for each of these data sets is discussed in Sections 3 to 7 (as referenced above).
3 Traffic categories
Traffic is separated into the following categories in order to assess benefits, TRRL Overseas
Unit (1988):
! Normal (see Section 4.2.1)
Normal traffic is defined as traffic that would pass along the project road if no investment
took place, including normal growth. It is specified for each road section within the
applications programs.
! Diverted (see Section 4.2.2)
Diverted traffic is defined as traffic that changes from another route (or transport mode)
to the project road, but still travels between the same origin and destination (this is
termed reassigned traffic in transport modelling). It is specified together with the road
investment option that causes it, and is specified within the relevant applications
programs.
! Generated (see Section 4.2.3)
Generated traffic is defined as additional traffic that occurs in response to the road
investment (this includes redistributed traffic as defined in transport models). Generated
traffic arises either because a journey becomes more attractive because of a cost or time
reduction, or because of the increased development that is brought about by a road
investment. It is specified together with the road investment option that induces it, and is
specified within the relevant applications programs.
These categories are each treated separately in an economic analysis (see Part G).
The existing traffic volumes on the road being analysed are specified in terms of vehicle type
or class, depending on the kind of analysis to be performed. The value entered for each
vehicle type is expressed as the annual average daily traffic (AADT):
This constitutes the baseline flow for the analysis period. It is assumed that seasonal variations
in traffic flows have already been accounted for when estimating the AADT from traffic
counts carried out over shorter periods.
For project analysis, traffic composition data is specified for each section. For network level
analysis, several representative sets of traffic composition data can be specified, and each is
assigned to a group of road sections with similar traffic characteristics.
In most situations, traffic growth has a major effect on the level of benefits obtained.
Specifying different growth rates for each vehicle type/class can effect changes in traffic
composition over time. Uncertainties always exist in estimating initial traffic, but there is even
greater uncertainty in forecasting future growth rates. Thus, it is recommended that sensitivity
analyses be always undertaken on the effects of different forecast growth rates.
Traffic volumes are derived from the baseline AADT and the composition data. The
computation of annual road user effects requires the AADT for each road section to be broken
down by vehicle types. For normal traffic, the annual average daily traffic for each vehicle
type (AADTkn) is obtained by multiplying the normal traffic AADTn by the composition data
defined for the vehicle type k in the base year. For generated traffic, AADTkg is either
specified directly or calculated, based on AADTkn and the traffic growth type used. Volumes
for normal and generated traffic are required as separate inputs into the Economic Analysis
module (see Part G) to enable the economic benefits of generated traffic to be
determined.
The specification of traffic composition data for each section is carried out for each traffic
category as described in Sections 4.2.1, 4.2.2 and 4.2.3.
It is permissible and sometimes convenient to enter AADT, in vehicles per day, for some
year before the start of the analysis period. However, it must be ensured that the traffic
start year is always less than or equal to the start year of the analysis period.
! Initial composition
The proportion of each representative vehicle that uses the road, as a percentage of the
AADT. For example; medium cars 10%; articulated trucks 5%; etc. These data items are
used to calculate the actual volumes, by vehicle type, for the applicable year.
! Future traffic
This is specified as a traffic growth period defined in terms of its start year and by one of
the following growth types:
❏ annual percentage increase (p)
( y −1)
p
AADTy = AADTstartyear 1 + . . .(4.1)
100
where:
where:
❏ actual AADT, which must be followed by another traffic growth period, unless the
year in which it applies is the last year of the analysis period
Then, depending on the selected traffic growth type, the growth rate is specified for each
representative vehicle. For example, light trucks increasing at 4% per year from 2000 to
2005, if annual percentage increase was selected.
Several traffic growth periods may be defined, each with a minimum length of one year.
It is important to ensure that the defined traffic growth periods cover consistently each
and every year of the analysis period. Therefore, if only one growth period has been
defined, this will be assumed to apply to all successive analysis years. A combination of
different growth types may be used to specify future traffic over the analysis period.
The calendar year in which traffic diversion starts, typically this should coincide with the
year following the completion of the road works.
! New AADT
The annual average daily traffic in the start year for each road section.
! New composition
The traffic composition in the start year for each road section.
! Future traffic
The number of years, after the start of the road works, that generated traffic appears.
Thus, relative start year j means the jth year following the start year of the works that
cause the generated traffic.
! Future traffic
(p + q)
AADTtotal( y ) = AADTtotal( y −1) 1 + ...(4.4)
100
p
AADTnorm( y ) = AADTnorm( y −1) 1 + ...(4.5)
100
where:
p
AADTgen( y ) = AADTnorm( y ) ...(4.6)
100
5 Axle loading
Defined as the total number of axles of all vehicles traversing a given road section in a
given year.
! Numbers of equivalent standard axle loads (ESA)
Combines the damaging effects of the full spectrum of axle loading in a common
damage-related unit. ESA is considered on each road section, for each year of the
analysis period.
T * NUM_AXLES k
YAX k = k ...(5.1)
ELANES * 10 6
YAX =
∑ YAX
k =1
k ...(5.2)
where:
YAX annual total number of axles of all vehicle types (millions per lane)
Tk annual traffic volume of vehicle type k, (k = 1, 2, . . , K)
NUM_AXLESk number of axles per vehicle type k
ELANES effective number of lanes for the road section
The effective number of lanes ELANES is used to model the effect of traffic load distribution
across the width of paved roads. This may either be specified by the user or be taken by
default as equal to the number of lanes (NLANES) for the road section.
may be specified directly by the user, or calculated from axle load information defined in the
Vehicle Fleet folder.
For each vehicle type, ESALFk is computed using information on the different damaging
effects of various axle configurations. For each type of axle group j a standard load, SAXLj, is
used to determine the loading ratio. The expression for calculating ESALF is as follows
Watanatada et al. (1987):
LE
Ik Jk
AXL kij
ESALFk =
∑ 100 ∑
i =1
Pki
j =1
SAXL
j
...(5.3)
where:
ESALFk equivalent standard axle load factor for vehicle type k, in equivalent standard
axle loads
Ik the number of subgroups i (defined in terms of load range) of vehicle type k (i
= 1, 2, . . ., Ik). Note that i may represent each individual vehicle
Pki percentage of vehicles in subgroup i of vehicle type k. If i represents each
individual vehicle then Pki = 100 (%)
LE axle load equivalency exponent (default = 4.0)
Jk the number of single axles per vehicle of type k (j = 1, 2, . . ., Jk)
AXLkij the average load on axle j of load range i in vehicle type k (tonnes)
SAXLj the standard single axle load of axle group type j, for example, 6.60 tonnes for
single-wheel single axle, 8.16 tonnes for dual-wheel single axle, etc.
The factor ESALFk is therefore an average over all vehicles of type k, loaded and unloaded, in
both directions on the given road section.
The annual total number of equivalent standard axles is calculated as:
K
YE4 =
∑ ELANES
T * ESALF
k =1
* 10
k k
6
...(5.4)
where:
YE4 annual total number of equivalent standard axles (millions per lane)
All other parameters are as previously defined in Sections 5.2 and 5.3.
Heavy vehicles (ADH) are specified in terms of vehicles per day for each year of the analysis
period.
The modelling of the changes in pavement skid resistance requires the specification of the
flow of heavy commercial vehicles per lane per day (QCV).
ADH
QCV = ... (5.5)
ELANES
where:
The modelling of changes in pavement texture depth requires the specification of the annual
number of equivalent light vehicle passes (∆NELV) over the road section. This is calculated
from the following expression:
where:
The number of vehicles with studded tyres is required for modelling pavement rutting during
freezing seasons.
365 * ST * AADT y * 10 −5
PASS = ...(5.7)
NTFD
where:
PASS annual number of vehicle passes with studded tyres in one direction (measured
in 1000s)
AADTy annual average daily traffic (AADT) in year y (veh/day)
ST percentage of annual number of vehicle passes with studded tyres (%)
NTFD number of traffic flow directions
The cumulative number of equivalent standard axle loads since the last rehabilitation,
reconstruction or new construction works (NE4) is given by:
∑
AGE 3
where:
NE4 cumulative number of equivalent standard axle loads since last rehabilitation,
reconstruction or new construction (millions/lane)
YE4y number of equivalent standard axle loads in year y (millions/lane)
AGE3 number of years since last rehabilitation, reconstruction or new construction
(years)
The maximum number of vehicles that can pass a point, or traverse a road section, in one
hour (total both directions). Capacity values determine the shape of speed-flow curves
by establishing the ultimate capacity value.
! Free speed
The speed of each vehicle at zero (or very low) flow. It is unaffected by other traffic but
is affected by the physical characteristics of the road and other non-traffic factors (see
Part E). The average free speed is calculated for each vehicle type.
! Speed at capacity
As traffic flows increase, average speeds for all vehicles converge towards the speeds of
the slowest vehicles in the stream, as passing becomes more and more restricted. As flow
approaches the ultimate capacity, average speeds may fall even lower than slow vehicle
free speeds, and any small disturbances in the traffic stream causes a stop-and-go
situation. An estimate of average speed at ultimate capacity, also known as jam speed,
is needed to describe the speed-flow-capacity relationship.
S1
S2
Speed km/h
S3
Snom
Sult
Qo Qnom Qult
Flow in PCSE/h
Qo the flow level below which traffic interactions are negligible in PCSE/h
Qnom nominal capacity of the road (PCSE/h)
Qult the ultimate capacity of the road for stable flow (PCSE/h)
Snom speed at the nominal capacity (km/h)
Sult speed at the ultimate capacity, also referred to as jam speed (km/h)
S1 to S3 free flow speeds of different vehicle types (km/h)
PCSE passenger car space equivalents (see below)
Notes: The basic PCSE values have been used as defaults for the HDM-4 standard vehicle
types (see Part E)
Qo
XQ1= ...(6.1)
Qult
Qnom
XQ2 = ...(6.2)
Qult
Table B1.2 Examples of capacity and speed-flow model parameters for different
road types
Since these data apply to individual road sections, it is important to ensure that the data relates
to a single and not a dual carriageway. This ensures consistency with the definition of a road
section as used in HDM-4.
The data in Table B1.1 describing the capacity of road sections are specified for each road
type:
! Ultimate capacity per lane (QLult) (PCSE/lane/hr). The ultimate capacity for the road
section Qult = QLult*NLANES
! Free flow capacity as a proportion of the ultimate capacity (XQ1)
Free speed value for each vehicle type is determined internally using the model described in
Part E.
The speed at nominal capacity is estimated to be 85% of the free speed of the slowest vehicle
in the traffic stream.
The maximum acceleration noise (σmaxr) represents the maximum standard deviation of
acceleration for each road type. This is required for modelling the effect of speed change
cycles (that is, speed variations along the road) on vehicle operating costs. In addition to
driver behaviour, speed fluctuations are sensitive to road geometry, road condition, the
presence of NMT, roadside friction, intersections, etc. (See Part E).
F lo w P e rio d s
M ean Peak
H o u rly F lo w -fre q u e n c y N e x t to P e a k
F lo w d istrib u tio n c u rv e
M e d iu m flo w
(% A A D T )
N e x t to L o w
O ve rn ig h t
C u m u la tiv e n u m b e r o f h o u rs in th e Y e a r (8 7 6 0 )
The number of flow periods into which the total number of hours in a year (8760) are
divided. Only one flow period needs to be specified for aggregate or network level
analyses. A uniform traffic composition and value of travel time across all flow-
frequency periods is assumed.
! The number of hours in each flow-frequency period, p (HRYRp) and
either
! Hourly traffic flow in each period as a proportion of AADT (HVp)
or
! The percentage of AADT in each flow-frequency period, p (PCNADTp)
Table B1.4 shows typical values of hourly flow-frequency distribution data for the three road-
use categories with hourly flows expressed as a percentage of AADT.
The parameter PCNADT is converted to hourly traffic flow as a proportion of AADT using
the following expression:
365 * PCNADTp
HVp = ...(7.1)
100 * HRYR p
where:
The data that describes the hourly traffic flows and the volume-capacity ratios are required for
modelling congestion effects on vehicle speeds and vehicle operating costs. Therefore the key
parameters are as follows:
! Qp
The traffic flow, in PCSE per hour, during traffic flow period p.
! VCRp
Qp =
∑ HV
k =1
p * PCSE k * AADTk ...(7.2)
where:
The volume-capacity ratio during each traffic flow period is expressed as follows:
Qp
VCR p = ...(7.3)
Qult
where:
AADTy ≤ AADTylim
and:
24 * Qult
AADTylim = K
...(7.4)
∑ (PCSE
k =1
k * MIX k )
where:
Since the proportion of each vehicle type in the traffic stream (MIXk) may vary each year, and
the ultimate road capacity Qult may also change due to a works intervention, the value of
AADTylim is calculated for each year of the analysis period. A check on the AADT limit is
carried out within the traffic module for every analysis year. Necessary adjustments to the
values of the traffic data are then made before passing on the data for use in the subsequent
modules. If the projected AADTy is higher than the AADT limit for the road, AADTy will be
set equal to AADTylim, and the analysis in year y will be performed using AADTylim. In these
circumstances, a printed warning is included in the HDM-4 outputs.
In reality, changes in the hourly flow-frequency distribution take place on a continuous basis
as traffic volume on a road increases with time. As a result, adjustments to the values of the
flow-frequency distribution parameters HVp and HRYRp are required. However, for the
purposes of analyses, the hourly flow-frequency distribution is only changed under one of the
following conditions:
1 When a spill-over from one high flow period to the next period occurs; that is, when the
hourly traffic flows computed for a high flow period exceeds the ultimate capacity of the
road, the excess flows will spill-over into the next highest flow period, so that some peak
spreading occurs. In this circumstance, new values of HVp are calculated, but HRYRp
values remain unchanged in order to simplify the analysis.
2 An intervention occurs that changes road users' pattern of journey time (for example, an
increase in road capacity that reduces traffic congestion).
8 References
Hoban C., Reilly W., and Archondo-Callao R., (1994)
Economic Analysis of Road Projects with Congested Traffic
Methods for Economic Evaluation of Highways Investments and Maintenance
Transport Division, Transportation, Water & Urban Development Department
World Bank, Washington D.C., USA
TRRL Overseas Unit, (1988)
A guide to road project appraisal, Road Note 5
Transport and Road Research Laboratory
Crowthorne, Berkshire, UK
Watanatada T., Harral C. G., Paterson W.D.O., Dhareshwar A.M., Bhandari A., and
Tsunokawa K., (1987)
The Highway Design and Maintenance Standards Model - Volume 1: Description of the
HDM-III Model.
The Highway Design and Maintenance Standards Series Baltimore:
Johns Hopkins University Press for the World Bank
Contents
C2 Bituminous Pavements
1 Introduction C2-1
2 Model framework and logic C2-2
2.1 Classification and concepts C2-2
2.2 Computational logic C2-5
3 Pavement strength C2-11
3.1 Adjusted structural number C2-11
3.2 Seasonal and drainage effects C2-13
3.3 User input options C2-16
4 Construction quality C2-18
5 Crack modelling C2-20
5.1 Structural cracking C2-20
5.2 Transverse thermal cracking C2-27
C3 Concrete Pavements
1 Introduction C3-1
2 Modelling framework and logic C3-2
2.1 Concrete pavement structure C3-3
2.2 Concrete pavement distress modes C3-5
2.3 Primary modelling parameters C3-9
2.4 Computational procedure C3-10
C4 Unsealed Roads
1 Introduction C4-1
2 Modelling logic C4-2
2.1 Classification, concepts, and logic C4-2
2.2 Primary model parameters C4-3
2.3 Basic computational procedure C4-5
2.4 Initialisation of variables C4-6
3 Road roughness C4-7
3.1 General C4-7
3.2 Roughness progression C4-7
3.3 Effect of compaction on roughness progression C4-9
Analytical
AnalyticalFramework
Framework
and
andModel
Model
Descriptions
Descriptions
Introduction
Introduction
Part
PartAA
Traffic
Traffic
Part
PartBB
RD
RDModel
Model WE
WEModel
Model RUE
RUEModel
Model SEE
SEEModel
Model
Part
PartCC Part
PartDD Part
PartEE Part
PartFF
Economic
EconomicAnalysis
Analysis
RD = Road Deterioration Part
PartGG
Nomenclature
Nomenclature
WE = Road Works Effects Part
PartHH
1 Introduction
Road deterioration is broadly a function of the original design, material types, construction
quality, traffic volume, axle load characteristics, road geometry, environmental conditions,
age of pavement, and the maintenance policy pursued.
HDM-4 includes relationships for modelling Road Deterioration (RD) and Road Works
Effects (WE). These are used for the purpose of predicting annual road condition and for
evaluating road works strategies. The relationships should link standards and costs for road
construction and maintenance to road user costs through road user cost models. In HDM-III
these relationships were combined into a single module called the Road Deterioration and
Maintenance Effects (RDME), as described by Watanatada et al. (1987). In HDM-4, this
module has been separated in order to address properly the expanded scope of modelling Road
Deterioration and Works Effects. The analysis now includes:
! Physical environments (climatic zones)
Encompassing cold (freeze/thaw) climates, very high temperatures and a very wide range
of temperature variations such as desert conditions, and very high moisture regimes and
arid conditions.
! Rigid/concrete and semi-rigid pavements, and a wide range of flexible pavements.
! Models for the following distresses:
This chapter deals with the pavement classification system used, and describes the RD
modelling approach for the different road surface classes considered in HDM-4. The key
variables that affect road deterioration (in particular those that are associated with climate and
environment) are also discussed. A comprehensive classification of climate in terms of
temperature and moisture is also given. The modelling of Road Works Effects is described in
Part D.
2 Pavement classification
A versatile framework of pavement classification is required to cater for the expanded scope
of RD and WE analysis. A system of classifying pavements has therefore been formulated
which uses broad definitions of road surfacing and roadbase types as illustrated in Table C1.1
(NDLI, 1995).
Block BRLC BR BR LC LC
SAUP SA SA
Key:
RC Rolled Concrete
FC Fibre Concrete
PC Porous Concrete
LT Lateritic Gravel
QZ Quartzitic Gravel
Note:
Asterik (*) indicates that different types of material or construction pattern may be defined.
Subdivides the paved category into bituminous, concrete and block surfaces; together
with the unsealed class there are thus four classes that are used to define the family of
distress models used for performance modelling.
! Pavement type
Integrates surface and roadbase types. Each type is designated by a four-character code,
combining the surface and roadbase type codes.
! Surface type
There are eight generic types, including those which allow for overlays of asphalt on
concrete and vice versa. Each base type is designated by a two-character code.
! Surface material
Defines more specific surface types (for example, different types of asphalt mixes).
These are user definable.
! Base material
Absolute models predict the condition (or distress) at a particular point in time as a function of
the independent variables. Incremental models give the change in condition from an initial
state as a function of the independent variables.
The families of pavement performance models used are based on the road surface classes:
Skid resistance
Roughness
4.1.1 Classification
It is necessary for the user to define climatic and environment information as per Table C1.3
and Table C1.4:
Per-humid High rainfall, or very many wet-surface days > 100 > 2400
4.1.2 Precipitation
The Mean Monthly Precipitation (MMP) is used in the modelling of bituminous pavement
deterioration and unsealed road deterioration, and is expressed in mm/month. The annual
average precipitation (PRECIP) is used in the modelling of concrete pavement deterioration
and is expressed in inches/year.
where:
FI =
∑ ABS[MIN(TEMP,0)]
i =1
...(4.2)
where:
FI freezing index
TEMP temperature (oC)
Ndays number of days in one freezing season
Note that FI is only required as input data for the two temperate temperature zones, and are
used in modelling the performance of concrete pavements.
4.2 Traffic
The primary traffic-related variables that affect road deterioration include the number and
types of vehicles using the road, and axle loading characteristics of the different vehicle types.
Details of the variables required are described in Part B, and the way in which they are used in
the various deterioration relationships is defined in the appropriate Sections of this document.
5 References
LAST (1996)
Modelling road design and maintenance effects for pavements in HDM-4
Final Report, FICEM, Latin American Study Team, International Study of Highway
Development and Management Tools, Santiago, Chile
NDLI (1995)
Modelling Road Deterioration and Maintenance Effects in HDM-4
Final Project Report, Asian Development Bank, RETA 5549
N. D. Lea International Limited, Vancouver, Canada
Paterson W. D. O., (1987)
Road Deterioration and Maintenance Effects: Models for Planning and Management
The Highway Design and Maintenance Standards Series
World Bank, Johns Hopkins Press, Baltimore, USA
Watanatada T., Harral C.G., Paterson W.D.O., Dhareshwar A.M., Bhandari A., and
Tsunokawa K., (1987)
The Highway Design and Maintenance Standards Model - Volume 1 Description
World Bank, Johns Hopkins Press, Baltimore, USA
C2 Bituminous Pavements
1 Introduction
This chapter describes the detailed modelling of bituminous pavement deterioration in
HDM-4 (see Figure C2.1).
Road
RoadDeterioration
Deterioration
Models
Models
The development of the road deterioration models was based on three previous documents:
1 Riley and Bennett (1995 & 1996) - based on Paterson (1987)
2 Watanatada et al. (1987)
3 NDLI (1995)
A series of formal workshops followed. These were held at the University of Birmingham
(UoB), UK, in April 1996, December 1996 and October 1997. Informal meetings were then
held in Washington, the University of Birmingham and the Transport Research Laboratory
(TRL), UK that supplemented these workshops.
As a result of these discussions, major changes were regularly proposed to the road
deterioration models which were presented in various versions of the fourth and fifth draft
specifications (Morosiuk, 1996 & 1998a). Following on from the beta testing of the HDM-4
software in November 1998, a sixth then a seventh draft of specifications were produced. A
further workshop was held at TRL in June 1999 to resolve outstanding issues. The decisions
reached at this workshop have been incorporated in the eighth draft version of the
specifications. The main contributors to the specifications were Paterson (IBRD), Morosiuk
(TRL), Riley (Riley Partnership), Odoki and Kerali (UoB).
Section 2 provides an overview of the Road Deterioration modelling framework. This is
followed by the relationships and default coefficient values for each of the distresses to be
modelled. The model coefficient values are stored in data files instead of having them hard
coded into the model. This facilitates local calibration and adaptation. The HDM-4 model
has more calibration factors than the previous HDM-III model.
A list of research documents referenced from this chapter is given in Section 14.
AC CRS
GB AMGB
HRA GM
PMA AB AB AMAB
RAC CS
AM SB AMSB
CM LS
PA TNA
AMAP
SMA AP FDA
Xx
CAPE CRS
GB STGB
DBSD GM
SBSD AB AB STAB
SL CS
ST SB STSB
PM LS
Xx TNA STAP
AP
FDA
Note: The modelling of AM and ST surfacings on concrete pavements, that is, AMRB and STRB, is
not included in this software release.
PA Porous Asphalt
AM Asphalt Mix
PMA Polymer Modified
Asphalt
SL Slurry Seal
The available models are constructed from different factors. Many are created from surface
and base types, whilst others are created from surface materials. Accordingly, the modelling
is done in terms of surface material and base type, even though base materials can be
specified. HDM-4 contains default coefficient values for the bituminous pavement types given
in Table C2.3.
Routine works AMGB AMSB AMAB AMAP STGB STSB STAB STAP
Mill & replace to **AP **AP **AP **AP N/A **SB **AB **AP
intermediate
surface layer
Mill & replace to **GB **SB **AB **AP **GB **SB **AB **AP
base
Notes:
1 The pavement type will depend on the critical thickness (Hmin) of the existing
bituminous surfacing, that is user-definable in HDM-4 Configuration
** Indicates that this, two character variable, is dependent on the specific works activity
(that is, operation
N/A not applicable
These are defined in Table C2.5, and can be considered under the following three categories:
! Surfacing distress
where:
Measure Definition
Area (of distress) Sum of rectangular areas circumscribing manifest distress (line cracks are
assigned a width of 0.5 m), expressed as a percentage of the carriageway area
Narrow cracking Interconnected or line cracks of 1-3 mm crack width (equivalent to AASHTO
Class 2)
Wide cracking Interconnected or line cracks of 3 mm crack width or greater, with spalling
(equivalent to AASHTO Class 4)
Indexed cracking Normalised sum of AASHTO Classes 2 to 4 cracking weighted by class, see
Section 5.3
Pothole Open cavity in the road surface with at least 150 mm diameter and at least 25
mm depth
Edge-break Loss of bituminous surface material (and possibly base materials) from the edge
of the pavement
Rut depth Maximum depth under 2 m straightedge placed transversely across a wheelpath
Mean Texture depth The average depth of the surface of a road surfacing expressed as the quotient
of a given volume of standardised material [sand (sand patch test), glass
spheres] and the area of that material spread in a circular patch on the surface
being tested, (PIARC, 1997)
Skid resistance Resistance to skidding expressed by the sideways force coefficient (SFC)
measured using the Sideways Force Coefficient Routine Investigation Machine
(SCRIM)
Road condition
Pavement and side-drain condition data at the beginning of the first analysis year or the first
year after construction are required. The data items for surfacing and deformation distress
modes and surface texture, are as described in Table C2.5.
The pavement conditions at the end of the year (that is, before road works) are predicted as
follows:
where:
Pavement history
The required data items refer to pavement ages, and these are related to the previous
maintenance, rehabilitation and construction works carried out on the pavement.
There are four variables defining the age of the pavement used in the models; AGE1, AGE2
AGE3 and AGE4. These variables are defined as follows:
1 AGE1 is referred to as the preventive treatment age. It is defined as the time, in number
of years, since the latest preventive treatment, reseal, overlay (or rehabilitation),
pavement reconstruction or new construction activity.
2 AGE2 is referred to as the surfacing age. It is defined as the time, in number of years,
since the latest reseal, overlay, pavement reconstruction or new construction activity.
3 AGE3 is referred to as the rehabilitation age. It is defined as the time, in number of
years, since the latest overlay, pavement reconstruction or new construction activity.
4 AGE4 is referred to as the base construction age. It is defined as the time, in number of
years, since the latest reconstruction that involves the construction of a new base layer
or new construction activity.
Traffic
The required traffic data are the flow of all vehicle axles (YAX) and the flow of equivalent
standard axle loads (YE4), both expressed on an annual basis in millions per lane. The data are
calculated for each analysis year based on the user-specified trafficand vehicle characteristics.
The annual average traffic speed and the average speed of heavy vehicles are also required
in some RD relationships
3 Calculate the amount of change in each surfacing distress mode during the analysis year
in the following order:
(a) Cracking
(b) Ravelling
(c) Potholing
(d) Edge-break
4 Check that the total damaged and undamaged carriageway surface area equals 100%
based on the limits defined for each distress mode, and determine the amount of each
surfacing distress mode at the end of the year and the average value for the year
5 Compute the change in each deformation distress mode during the year, and determine
the amount of the distress mode at the end of the year and the average value for the year
6 Compute the change in each surface texture distress mode during the year, and determine
the amount of the distress mode at the end of the year and the average value for the year
7 Store results for use in subsequent modules (that is, RUE, WE, SEE) and in the following
analysis year, and for reporting
3 Pavement strength
SNBASU s = 0.0394
∑ ais hi
i =1
...(3.2)
b 0 exp ( −b 3 z j ) b 1exp ( − (b 2 + b 3 )z j )
m + −
∑ − b3 (b 2 + b 3 )
SNSUBA s = 0.0394 a js ...(3.3)
0
b exp ( −b z
3 j−1 ) b 1exp ( − (b 2 + b 3 )z j−1 )
j =1 +
− b3 (b 2 + b 3 )
[
SNSUBG s = [b 0 - b 1exp (- b 2 z m )] [exp(- b 3 z m )] 3.51 log10 CBR s - 0.85 (log10 CBR s ) − 1.43
2
]
...(3.4)
where:
The values of the model coefficients b0 to b3 are given in Table C2.6 and the values of the
layer coefficients ai and aj are given in Table C2.7.
Pavement type b0 b1 b2 b3
All pavement types 1.6 0.6 0.008 0.00207
GB CBR > 70, cemented sub-base ai = 1.6 (29.14 CBR - 0.1977 CBR2 +
0.00045 CBR3) 10-4
Notes:
1 The table reproduces information from the source, with the exception of the granular
sub-base coefficient
2 If the user quotes a CBR value for a stabilised (lime or cement) layer, the
corresponding granular coefficient should be used
3 Unconfined Compressive Strength (UCS) is quoted in MPa at 14 days
4 MR30 is the resilient modulus by the indirect tensile test at 30 °C
5 CBR is the California Bearing Ratio
Equation 3.4 above predicts negative values for the subgrade contribution below CBR = 3.
This is different to HDM-III where the values were set to 0 and reflects the detrimental impact
of weak subgrades on pavement performance.
where:
f
fs = ...(3.6)
[(1 - d) + d(f p )]1/p
and:
Distress Model p
Cracking Initiation of structural cracking 2.0
If only one season’s SNP value is available then the following relationship (Riley, 1996a)
should be used to calculate the wet/dry season SNP ratio. This relationship will also be used
to calculate the wet/dry season SNP ratio for each year of the analysis period, taking into
account changes in the drainage factor and the amount of cracking.
[1 - exp(a 0 MMP )]
f = K f 1 - (1 − a 2 DFa )(1 + a 3 ACRA a + a 4 APOTa ) ...(3.7)
a1
where:
Table C2.9 Default coefficient values for the seasonal SNP ratio
Coefficient a0 a1 a2 a3 a4
The drainage factor, DF, is a continuous variable whose value can range between 1 (excellent)
and 5 (very poor), depending on the type of drain (Paterson, 1998). The user will be required
to input the type of drain (as listed in Table C2.10) and the condition of the drain as excellent,
good, fair, poor or very poor.
Surface lined 1 3
V-shaped – hard 1 4
Shallow – hard 2 5
Shallow – soft 2 5
The minimum (excellent) and maximum (very poor) values for DF suggested for various
types of drain are given in Table C2.10. The value of DF for drains in a good, fair or poor
condition is determined by linearly interpolating between these values.
In some instances there will be an absence of drains. In situations where a drain is required the
value of DF ranges between 3 and 5, whereas in situations where a drain is unnecessary a
value of 1 for DF is suggested.
The condition of the drains will deteriorate unless they are well maintained, for example,
through routine maintenance. The incremental annual change in DF due to deterioration is
given in Equation 3.8 below: (the change in DF due to maintenance, ∆DFw, is detailed in Part
D - Road Works Effects).
and:
(DFmax - DFmin )
ADDF = ...(3.9)
Drain Life
where:
Drain life has been expressed as a function of the terrain as given below. The proposed
default coefficient values a0 and a1 are given in Table C2.11 (Morosiuk, 1998b) for the
climatic categories classified by moisture (see Chapter C1 - Section 4).
where:
a0 a1 a0 a1 a0 a1 a0 a1 a0 a1
No drain - but 3.5 -0.0029 2.5 -0.0027 2 -0.0033 1.5 -0.0044 1.5 -0.0044
required
No drain - not 50 0 50 0 50 0 50 0 50 0
required
The relationships used to convert Benkelman beam deflections (DEF) to SNP values are
based on those in HDM-III (Paterson, 1987) and are given below:
Base is not cemented
−0.63
SNPs = 3.2 (DEFs ) + dSNPK ...(3.11)
Base is cemented
−0.63
SNPs = 2.2 (DEFs ) + dSNPK ...(3.12)
and:
dSNPK = 0.0000758 {MIN (63, ACX a ) HSNEW + MAX [MIN (ACX a − PACX,40 ),0] HSOLD}
...(3.13)
where:
DEFs benkelman beam rebound deflection under 80 kN axle load, 520 kPa
tyre pressure and 30°C average asphalt temperature for season s
(mm)
dSNPK reduction in adjusted structural number due to cracking
ACXa area of indexed cracking at the start of the analysis year (% of total
carriageway area)
HSNEW thickness of the most recent surfacing (mm)
PACX area of previous indexed cracking in old surfacing (% of total
carriageway area); that is, 0.62 (PCRA) + 0.39 (PCRW)
HSOLD total thickness of previous underlying surfacing layers (mm)
Some models need Benkelman beam deflection values. Where these are not user input,
DEF values are derived from SNP values using relationships originated from those in
HDM-III, that is:
❏ Base is not cemented
−1.6
DEFs = 6.5 (SNPK s ) ...(3.14)
❏ Base is cemented
−1.6
DEFs = 3.5 (SNPK s ) ...(3.15)
and:
where:
The central FWD deflection at 700 kPa is used as the equivalent Benkelman beam
deflection. The equations in Option 2 are then used to calculate SNP
4 Construction quality
Poor construction quality results in greater variability in material properties and performance.
HDM-4 does not provide a method of modelling proportions of road that are classified as
good, fair and poor, so only an average level of construction defects is usually included. The
construction defects indicators (CDS and CDB) used in the deterioration models are described
below.
The relative compactions of the base, sub-base and selected subgrade layers (COMP) is
important in predicting the initial densification of rut depth. Paterson (1987) gives an
equation to calculate COMP, but it is proposed that users are also able to estimate it based on
the values in Table C2.12.
The initiation (and in some cases progression) of certain distresses is more accurately
attributed to problems in material handling, preparation, or construction than to structural
weakness in the pavement. In HDM-III, a construction quality code (CQ) was used in the
crack initiation and ravelling models. However, in HDM-4 the construction defects are input
through two indicators:
! CDS
CDS is a factor indicating the general level of binder content and stiffness relative to the
optimal material design for the specified bituminous mixture. It is used as an indicator to
illustrate whether a bituminous surfacing is prone to cracking and ravelling (low value of
CDS), or prone to rutting through plastic deformation (high value of CDS).
CDS is a continuous variable, generally ranging in value between 0.5 and 1.5 as shown in
Table C2.13. Intermediate values are chosen by judgement. This may involve back-analysis
to verify that the plastic deformation and cracking predictors are valid (see A Guide to
Calibration and Adaptation).
Rich (Soft) Nominally about 10% above design optimal binder content 1.5
For potholing, the base construction defects indicator (CDB) is used. CDB is a continuous
variable ranging between 0 (no construction defects) and 1.5 (several defects). The type of
defects that should be considered in setting a value of CDB is given in Table C2.14.
5 Crack modelling
Cracking is one of the most important distresses in bituminous pavements. Fatigue and ageing
have been identified as the principal factors which contribute to cracking of a bituminous
pavement layer. The propagation of cracking is accelerated through the embrittlement
resulting from ageing and the ingress of water, which can significantly weaken the underlying
pavement layers.
There are two types of cracking considered in HDM-4:
! Structural cracking
This is effectively load and age/environment associated cracking (see Section 5.1).
! Transverse thermal cracking
...(5.2)
! Other bases
YE4
ICA = K cia CDS 2 a 0 exp a 1 SNP + a 2 + CRT ...(5.3)
SNP 2
YE4
a 0 expa 1 SNP + a 2 2
+ CRT
ICA = K cia CDS 2 MAX SNP
...(5.4)
PCRW
* MAX1 − a ,0 , a 4 HSNEW
3
YE4
a 0 exp a 1 SNP + a 2
MAX SNP 2
ICA = K cia CDS 2 + CRT ...(5.5)
PCRA
* MAX1 − ,0 , a 4
a3
where:
The proposed default coefficient values a0 to a4 for the initiation of All cracking are given in
Table C2.15, and those of a0 to a2 for the initiation of Wide cracking are given in Table C2.16.
Table C2.15 Default coefficient values for initiation of All structural cracking
models
STAB All except SL, >0 5.4 4.21 0.14 -17.1 20 0.12
CAPE
Table C2.16 Default coefficient values for initiation of Wide structural cracking
models
CRP
dACA = K cpa
CDS
[
Z A (Z A a 0 a 1 δt A + SCA )
a1 1/a1 − SCA ] ...(5.7)
Progression of All structural cracking commences when δtA > 0 or ACAa > 0
where:
if ACA a > 0 δtA = 1 otherwise δt A = MAX {0, MIN [(AGE2 − ICA ), 1]}
Y = [a 0 a1 Z A δt A + SCA a1 ] ...(5.8)
! if Y < 0
then:
dACA = K cpa
CRP
(100 − ACA a ) ...(5.9)
CDS
! if Y ≥ 0
then:
dACA = K cpa
CRP
(
Z A Y 1/a1 − SCA ) ...(5.10)
CDS
then:
dACA = K cpa
CRP
CDS
(
100 - c 1
1/a1
− ACA a ) ...(5.11)
where:
CRP
dACW = K cpw
CDS
[
Z w (Z w a 0 a 1 δt w + SCW )
a1 1/a1 − SCW ] ...(5.13)
where:
Progression of Wide structural cracking commences when δtW > 0 or ACW a > 0
where:
if ACWa > 0 δtW = 1 otherwise δt w = MAX { 0, MIN [(AGE2 − ICW ), 1] }
The initiation of Wide structural cracking is constrained so that it does not commence before
the area of All structural cracking (ACAa) exceeds 5% as follows:
δtW = 0 if ACA a ≤ 5 and ACWa ≤ 0.5 and δtW > 0
If patching of Wide structural cracking was performed in the previous analysis year, reducing
the area of Wide cracking to below 1% but with the area of All structural cracking remaining
at over 11% at the start of the current analysis year (that is, ACWa ≤1 and ACAa > 11), then
the rate of progression of Wide structural cracking is assumed to begin not at the low initial
rate, but at a higher rate similar to the rate before patching.
For this situation a temporary value of Wide structural cracking, ACWtemp is defined to be 5%
less than ACAa; that is:
ACWtemp = ACAa – 5 if ACWa ≤ 1 and ACAa > 11
This value is then used as the temporary value of ACWa for the computation of dACW in that
analysis year.
dACW is computed each analysis year as follows:
if ACWa ≥ 50 ACWa ≥ 50 then: zw = -1
otherwise: zw = 1
[
SCW = MIN ACWa , (100 - ACWa ) ]
Y = [a 0 a 1 Z w δt w + SCW a1 ] ...(5.15)
! if Y < 0
then:
dACW = K cpw
CRP
MIN [(ACA a + dACA − ACWa ), (100 − ACWa )] ...(5.16)
CDS
! if Y ≥ 0
then:
CRP
MIN [(ACA a + dACA − ACWa ), Z w (Y
dACW = K cpw 1/a1 − SCW )] ...(5.17)
CDS
CRP
dACW = K cpw [ (
MIN (ACA a + dACA − ACWa ), 100 − c1
1/a1
− ACWa )]
CDS
...(5.18)
where:
and:
dACA incremental change in area of All structural cracking during the analysis
year (% of total carriageway area)
dACW incremental change in area of Wide structural cracking during the analysis
year (% of total carriageway area)
ACAa area of All structural cracking at the start of the analysis year
ACWa area of Wide structural cracking at the start of the analysis year (% of total
carriageway area)
δtA fraction of analysis year in which All structural cracking progression
applies
The proposed default coefficient values a0 and a1 for the progression of All cracking and those
for the progression of Wide cracking are given in Table C2.17.
Table C2.17 Default coefficient values for progression of All and Wide
structural cracking
Teq 50 7 50 50 7
...(5.22)
MIN(a 0 PNCT, (PNCT - NCTa )),
dNCT = K cpt
1
MIN(NCTeq − NCTa ), MAX
δt T
CDS
2 NCTeq (AGE3 - ICT - 0.5 )
,0
(Teq )2
...(5.23)
A transverse thermal crack is assumed to traverse the full width of the carriageway. Thus the
area of transverse thermal cracking is given by:
dNCT
dACT = ...(5.24)
20
where:
The default coefficient values a0 to a2 for the initiation, and those of a0 for the progression of
transverse thermal cracks, are given in Table C2.20.
a0 a1 a2 a0
All pavement types except STGB and STSB 1.0 -1.0 0.02 0.25
where:
where:
6 Ravelling
Ravelling is the progressive loss of surface material through weathering and/or traffic
abrasion. The occurrence of ravelling varies considerably among different regions and
countries according to construction methods, specifications, available materials, and local
practice. Ravelling is a common distress in poorly constructed, thin bituminous layers such as
surface treatment, but it is rarely seen in high quality, hot-mix asphalt.
The construction defects indicator for bituminous surfacings, CDS, (see Section 4) is used as a
variable in the ravelling models. The initiation model is basically as proposed by Paterson
(1987), with CDS replacing the original construction quality variable CQ. The progression
model is also based on that proposed by Paterson (1987) but with a traffic variable introduced
as proposed by Riley (1999).
6.1 Initiation
Ravelling is said to occur on a given road section when 0.5% of the carriageway surface area
is classified as ravelled. The initiation is given as:
where:
The proposed default coefficient values a0 to a1 for the ravelling initiation model is given in
Table C2.21.
6.2 Progression
The general form of the model for the progression of ravelling is given below:
K vp 1
dARV = [
Z (Z( a 0 + a 1 YAX)a 2 δt v + SRV )
a2 1/a2 − SRV ] ...(6.2)
RRF CDS
2
if ARVa ≥ 50 then: z = -1
otherwise: z = 1
[
Y = (a 0 + a 1 YAX)a 2 Z δt v + SRV a2 ] ...(6.3)
! if Y < 0
then:
K vp 1
dARV = 2
(100 − ARVa ) ...(6.4)
RRF CDS
! if Y ≥ 0
then:
K vp 1
Z(Y
dARV = 1/a2 − SRV )
...(6.5)
RRF CDS 2
then:
K vp 1
dARV =
RRF CDS 2
(
100 − c 1
1/a2
− ARVa ) ...(6.6)
and:
where:
dARV change in area of ravelling during the analysis year (% of total carriageway
area)
ARVa area of ravelling at the start of the analysis year (% of total carriageway area)
δtv fraction of analysis year in which ravelling progression applies
AGE2 pavement surface age since last reseal, overlay, reconstruction or new
construction (years)
Kvp calibration factor for ravelling progression
IRV time to ravelling initiation (years)
Pavement type a0 a1 a2
All pavement types 0.6 3.0 0.352
7 Potholing
Potholes usually develop in a surface that is either cracked, ravelled, or both. The presence of
water accelerates pothole formation both through a general weakening of the pavement
structure and lowering the resistance of the surface and base materials to disintegration.
The potholing models use the construction defects indicator for the base, CDB, as a variable
(see Section 4). In the models, potholing is expressed in terms of the number of pothole units
of area 0.1 m2. The volume of each of these pothole units is assumed to be 10 litres (that is,
100 mm in depth). The relationships for the initiation and progression of potholing have been
modified from those given in the NDLI (1995) and Riley (1996b).
7.1 Initiation
Initiation of potholes due to cracking only arises once the total area of Wide structural
cracking (ACW) exceeds 20%. Ravelling initiated potholes arise when the ravelled area
(ARV) exceeds 30%.
(1 + a 1HS )
IPT = K pi * a 0 ...(7.1)
(1 + a 2 CDB ) (1 + a 3 YAX ) (1 + a 4 MMP )
where:
IPT time between the initiation of Wide structural cracking or ravelling and the
initiation of potholes (years)
HS total thickness of bituminous surfacing (mm)
CDB construction defects indicator for the base
YAX annual number of axles of all motorised vehicle types in the analysis year
(millions/lane)
MMP mean monthly precipitation (mm/month)
Kpi calibration factor for pothole initiation
The values for IPT are calculated separately for potholing due to cracking and due to
ravelling. The separation between these two mechanisms of potholing is maintained
throughout the analysis with the progression being modelled differently for potholes due to
cracking, due to ravelling and due to the enlargement of existing potholes.
The proposed default coefficient values a0 to a4 for the pothole initiation model is given in
Table C2.23.
Cause of Pavement a0 a1 a2 a3 a4
pothole initiation type
AMGB, STGB 2.0 0.05 1.0 0.5 0.01
Cracking
All except GB bases 3.0 0.05 1.0 0.5 0.01
7.2 Progression
Pothole progression arises from potholes due to cracking, ravelling and the enlargement of
existing potholes. The progression of potholes is affected by the time lapse between the
occurrence and patching of potholes. A time lapse factor (TLF) has been introduced as an
indicator of the response time to patching potholes (Odoki, 1997; and Riley, 1997).
The annual incremental increase in the number of pothole units due to each of these three
distresses is calculated as:
! If at the start of the first year of the analysis period ARVa = 0, then potholing progression
from ravelling commences when:
! If at the start of the first year of the analysis period 0 < ACWa ≤ 20, then potholing
progression from wide cracking commences when ACWa > 20
! If at the start of the first year of the analysis period 0 < ARVa ≤ 30, then potholing
progression from ravelling commences when ARVa > 30
! If at the start of the first year of the analysis period ACWa > 20, then potholing
progression from wide cracking commences immediately
! If at the start of the first year of the analysis period ARVa > 30, then potholing
progression from ravelling commences immediately
! Potholing progression from enlargement commences if NPTa > 0 at the start of an
analysis year
The total annual increase in the number of pothole units per kilometre of road length is given
by:
dNPT =
∑ dNPT
i =1
i ...(7.3)
where:
dNPTi additional number of potholes per km derived from distress type i (Wide
structural cracking, ravelling, enlargement) during the analysis year
ADISi the percentage area of Wide structural cracking at the start of the analysis year,
or the percentage area of ravelling at the start of the analysis year, or number
of existing potholes per km at the start of the analysis year
TLF time lapse factor (see Table C2.25)
dNPT total number of additional potholes per km during the analysis year
Kpp calibration factor for pothole progression
1 month 0.06
2 months 0.12
3 months 0.20
4 months 0.28
6 months 0.43
12 months 1.00
8 Edge-break
Edge-break can be defined as the loss of surface and base materials at the pavement edge,
caused by shear failure and attrition. This commonly arises on narrow roads with unsealed
shoulders, where vehicle wheels pass on or close to a pavement edge.
The measure for edge-break that the user provides as input into the model, and the
corresponding output data, will be in square metres per km, and not in cubic metres per km.
The value in square metres is then multiplied internally by 100 mm (same depth as potholes)
to obtain the volume of edge-break in cubic metres for modelling purposes.
The edge-break model predicts that edge-break occurs on roads with a carriageway width of
up to a user defined maximum width of CWmax. The default value of CWmax is 7.2 metres and
an upper limit of CWmax has been set to 7.5 metres (that is, no edge-break is predicted for
roads with a carriageway width greater than 7.5 metres).
The edge-break model is as follows:
and:
CWmax - CW
PSH = MAXMINMAX a 3 + a 4 CW, , 1, 0 ...(8.2)
a5
where:
The proposed default coefficient values a0 to a5 for the edge-break model are given in Table
C2.26.
Pavement type a0 a1 a2 a3 a4 a5
AMGB 50 -1 0.2 2.65 -0.425 10
! An upper limit of 18% is imposed on the area of edge-break. The upper value of 18% is
based on the assumption that edge-break will not extend beyond 0.5 metres from either
edge of a pavement of 5.5 metres width.
9.2.1 Edge-break
AVEB b = MIN [ 18, (AVEB a + dAVEB ) ] ...(9.1)
where:
AVEBb area of edge-break at the end of the analysis year (% of total carriageway area)
AVEBa area of edge-break at the start of the analysis year (% of total carriageway
area)
dAVEB unadjusted increase in the area of edge-break during the analysis year (% of
total carriageway area)
Equation 9.1 above requires that the volume of edge-break, VEB, be converted into an area of
edge-break measured as a percentage of total carriageway area. The area of edge-break
expressed as a percentage of the total carriageway area, AVEB, is obtained from the following
expression:
VEB
AVEB = ...(9.2)
CW
where:
Thus, by substituting VEB in Equation 9.2 above with VEBa the value of AVEBa is obtained,
and by substituting VEB with dVEB, the value of dAVEB is obtained;
where:
VEBa volume of edge-break per km at the start of the analysis year (m3/km)
dVEB unadjusted increase in volume of edge-break per km during the analysis year
(m3/km)
9.2.2 Potholes
APOTb = MIN [10, (APOTa + dAPOT )] ...(9.3)
where:
APOTb area of potholes at the end of the analysis year (% of total carriageway area)
APOTa area of potholes at the start of the analysis year (% of total carriageway area)
dAPOT unadjusted increase in the area of potholes during the analysis year (% of total
carriageway area)
Equation 9.3 above requires that the number of potholes per km be converted into area of
potholes in per cent of total carriageway area. The area of potholes expressed as a percentage
of total carriageway area (APOT) is obtained from the expression:
(NPT ) (STDAPOT )
APOT = ...(9.4)
10(CW )
where:
Thus, by substituting NPT in Equation 9.4 above with NPTa the value of APOTa is obtained,
and by substituting NPT with dNPT the value of dAPOT is obtained;
where:
NPTa number of potholes per km at the start of the analysis year (no/km)
dNPT unadjusted increase in the number of potholes per km during the analysis year
9.2.3 Cracking
where:
ACRAb total area of cracking at the end of the analysis year (% of total carriageway
area); that is, ACAb + ACTb
ACRAa total area of cracking at the start of the analysis year (% of total carriageway
area); that is, ACAa + ACTa
dACRA unadjusted increase in total area of cracking during the analysis year (% of
total carriageway area); that is, dACA + dACT
AVPDb AVEBb + APOTb
dAVEBCR increase in area of edge-break arising from cracked area during the analysis
year (% of total carriageway area)
dAPOTCR increase in area of potholes arising from cracked area during the analysis year
(% of total carriageway area)
ACAb area of All structural cracking at the end of the analysis year (% of total
carriageway area)
ACAa area of All structural cracking at the beginning of the analysis year (% of total
carriageway area)
dACA unadjusted increase in area of All structural cracking during the analysis year
(% of total carriageway area)
ACTb area of transverse thermal cracking at the end of the analysis year (% of total
carriageway area)
ACTa area of transverse thermal cracking at the start of the analysis year (% of total
carriageway area)
dACT unadjusted increase in area of transverse thermal cracking during the analysis
year (% of total carriageway area)
then:
otherwise:
dAVEBCR = 0
and:
where:
∆AVEB adjusted increase in the area of edge-break during the analysis year (% of
total carriageway area)
VBCR user definable percentage of dAVEB arising from cracked areas (default =
20)
! if dNPT > 0
∆NPTc (STDAPOT )
dAPOTCR = ...(9.7)
10 (CW )
otherwise:
dAPOTCR = 0
dNPTc
∆NPTc = ∆NPT ...(9.8)
dNPT
and:
where:
∆NPT adjusted total increase in the number of potholes per km during the
analysis year
∆NPTc adjusted increase in the number of potholes per km derived from Wide
structural cracking during the analysis year
dNPTc unadjusted increase in the number of potholes per km derived from Wide
structural cracking during the analysis year (see Section 7)
NPTb total number of potholes per km at the end of the analysis year
NPTa total number of potholes per km at the start of the analysis year
! if ACRAa > 0
(dACA - dAPOTCR )
∆ACA = ∆ACRA ...(9.11)
(dACRA − dAPOTCR )
otherwise:
∆ACA = 0
and:
ACTa
q = 1 - ...(9.12)
ACRA a
and:
where:
∆ACRA adjusted increase in total area of cracking during the analysis year (% of
total carriageway area)
∆ACA adjusted increase in area of All structural cracking during the analysis
year (% of total carriageway area)
...(9.14)
where:
ACWb area of Wide structural cracking at the end of the analysis year (% of total
carriageway area)
ACWa area of Wide structural cracking at the start of the analysis year (% of total
carriageway area)
dACW unadjusted increase in area of Wide structural cracking during the analysis
year (% of total carriageway area)
then:
otherwise:
ACTb = 0
9.2.4 Ravelling
(100 - AVPC b ),
ARVb = MAX 0, MIN
(ARVa + dARV − dAVEBRV − dAPOTRV − dACRARV )
...(9.16)
where:
ARVb area of ravelling at the end of the analysis year (% of total carriageway area)
ARVa area of ravelling at the start of the analysis year (% of total carriageway area)
dARV unadjusted increase in area of ravelling during the analysis year (% of total
carriageway area)
AVPCb AVEBb + APOTb + ACRAb
dAVEBRV increase in area of edge-break arising from ravelled area during the analysis
year (% of total carriageway area)
dAPOTRV increase in area of potholes arising from ravelled area during the analysis
year (% of total carriageway area)
dACRARV increase in area of cracking arising from ravelled area during the analysis
year (% of total carriageway area)
then:
otherwise:
dAVEBRV = 0
where:
VBRV user definable percentage of dAVEB arising from ravelled area (default =
20)
then:
∆NPTr (STDAPOT )
dAPOTRV = ...(9.18)
10(CW )
otherwise:
dAPOTRV = 0
and:
dNPTr
∆NPTr = ∆NPT ...(9.19)
dNPT
where:
∆NPTr adjusted increase in the number of potholes per km derived from ravelling
during the analysis year
dNPTr unadjusted increase in the number of potholes per km derived from
ravelling during the analysis year (see Section 7)
then:
otherwise:
dACRARV = 0
where:
CRV user definable percentage of dACRA arising from the ravelled area
(default = 10)
where:
ADAMRb total non-patched damaged surface area at the end of the analysis year (%
of total carriageway area)
where:
ADAMSb severely damaged surface area at the end of the analysis year (% of total
carriageway area)
10 Rut depth
Rutting is defined as the permanent or unrecoverable traffic-associated deformation within
pavement layers which, if channelised into wheelpaths, accumulates over time and becomes
manifested as a rut (Paterson, 1987).
Rut depth modelling is performed after the values of all the surface distresses (that is,
cracking, ravelling, potholing and edge-break) at the end of the year have been calculated.
The rut depth model is based on four components of rutting:
! Initial densification (see Section 10.1)
The rut depth at any time is the sum of the four components.
For HDM-4 the rut depths have been standardised to a 2.0 m straight-edge. Since HDM-III
was based on a 1.2 m straight-edge, the default model coefficients have been changed
accordingly.
(
RDO = K rid a 0 YE4 10 6 )(a + a
1 2 DEF )
SNP a3 COMP a 4
...(10.1)
where:
The proposed default coefficient values a0 to a4 for the initial densification model is given in
Table C2.27.
Pavement type a0 a1 a2 a3 a4
AMGB, AMAB, AMSB, STGB, STAB, STSB 51740 0.09 0.0384 -0.502 -2.30
AMAP, STAP 0 0 0 0 0
Initial densification only applies to new construction or reconstruction that involves the
construction of a new base layer (that is, from when AGE4 = 0), for a period of time of one
year. AGE4 is defined as follows:
(
∆RDSTuc = K rst a 0 SNP a1 YE4 a 2 COMP a3 ) ...(10.2)
[
∆RDSTcrk = K rst a 0 SNP 1 YE4 2 MMP 3 ACX a
a a a a4
] ...(10.3)
then:
! if ACRA > 0
then:
where:
ACXa area of indexed cracking at the beginning of the analysis year (% of total
carriageway area)
SNP average annual adjusted structural number of the pavement
YE4 annual number of equivalent standard axles (millions/lane)
Krst calibration factor for structural deformation
The proposed default coefficient values a0 to a4 for the structural deformation models are
given in Table C2.28.
Pavement type a0 a1 a2 a3 a4
Without cracking All pavement types 44950 -1.14 0.11 -2.3
After cracking All pavement types 0.0000248 -0.84 0.14 1.07 1.11
where:
The proposed default coefficient values for the plastic deformation model are given in Table
C2.29.
Surface type a0 a1 a2
AM 2.46 -0.78 0.71
ST 0 -0.78 0.71
where:
∆RDW incremental increase in rut depth due to studded tyres in the analysis year
(mm)
PASS annual number of vehicle passes with studded tyres in one direction (1000s)
S average traffic speed (km/h)
SALT variable for salted or unsalted roads (2 = salted; 1 = unsalted)
W road width (m) (carriageway plus total shoulder width)
Krsw calibration factor for surface wear
The proposed default coefficient values a0 to a4 for the surface wear model is given in Table
C2.30.
Pavement type a0 a1 a2 a3 a4
All pavement types 0.0000248 1.0 -0.46 1.22 0.32
otherwise:
where:
∆RDM incremental increase in total mean rut depth in both wheelpaths in the
analysis year (mm)
RDO rutting due to initial densification in the analysis year (mm)
∆RDST incremental increase in structural deformation in the analysis year (mm)
∆RDW incremental increase in wear by studded tyres in the analysis year (mm)
The total rut depth, RDMb, at any given time is given as:
where:
RDMb total mean rut depth in both wheelpaths at the end of the analysis year (mm)
RDMa total mean rut depth in both wheelpaths at the start of the analysis year (mm)
where:
RDSb rut depth standard deviation at the end of the analysis year (mm)
RDMb mean rut depth at the end of the analysis year (mm)
The standard deviation of rut depth at the beginning of the analysis year (RDSa) is also
calculated from Equation 10.11 above with RDMa replacing RDMb.
11 Roughness
The roughness model consists of several components of roughness (cracking, disintegration,
deformation and maintenance). The total incremental roughness is the sum of these
components. The surface distress values used in predicting roughness are those that have
been adjusted so that the total damaged surface area plus the undamaged area equals 100%.
11.1 Structural
The structural component of roughness relates to the deformation in the pavement materials
under the shear stresses imposed by traffic loading. It is given by:
−5
∆RI s = a 0 exp (m K gm AGE3 ) (1 + SNPK b ) YE4 ...(11.1)
and:
and:
where:
The default values for the environmental coefficient m are given in Table C2.31.
11.2 Cracking
The incremental change in roughness due to cracking is given by:
where:
∆RIc incremental change in roughness due to cracking during the analysis year (IRI
m/km)
∆ACRA incremental change in area of total cracking during the analysis year (% of
total carriageway area)
11.3 Rutting
The incremental change in roughness due to variation of rut depth is given by:
where:
∆RIr incremental change in roughness due to rutting during the analysis year (m/km
IRI)
∆RDS incremental change in standard deviation of rut depth during the analysis year
(mm) (= RDSb – RDSa)
11.4 Potholing
The potholing effect depends upon the number of vehicles that actually hit the potholes, which
in turn depends upon the traffic volume and the freedom to manoeuvre. A freedom to
manoeuvre variable (FM), ranging between 0 and 1, is used and is predicted using Equation
11.6 below:
∆NPT * TLF
a2
∆RI t = a 0 (a 1 − FM) NPTa * TLF + − NPTa
a2
...(11.7)
2
a2
∆NPT
∆RI t = a 0 (a 1 − FM) * ∆NPT * NPTa + ...(11.8)
2
where:
FM freedom to manoeuvre
CW carriageway width (m)
AADT annual average daily traffic (veh/day)
∆RIt incremental change in roughness due to potholing during the analysis year (IRI
m/km)
∆NPT incremental change in number of potholes per km during the analysis year
NPTa number of potholes per km at the start of the analysis year
TLF time lapse factor (see Table C2.25)
11.5 Environment
The environmental component of roughness is due to factors which include temperature and
moisture fluctuations, and also foundation movements (for example, subsidence). It is given
by:
where:
∆RIe incremental change in roughness due to the environment during the analysis
year (IRI m/km)
RIa roughness at the start of the analysis year (IRI m/km)
m environmental coefficient
Kgm calibration factor for the environmental component
where:
∆RI total incremental change in roughness during the analysis year (IRI m/km)
Kgp calibration factor for roughness progression
The default coefficient values for the various roughness components are given in Table C2.32.
The roughness of the pavement at the end of the analysis year is given by:
where:
RIb roughness of the pavement at the end of the analysis year (IRI m/km)
RIa roughness of the pavement at the start of the analysis year (IRI m/km)
a0 upper limit of pavement roughness, specified by the user (default = 16 IRI
m/km)
The annual average roughness of the pavement for a given analysis year is calculated as:
where:
RIav annual average roughness of the pavement for the analysis year (IRI m/km).
This is the roughness used in the Road User Effects model.
where:
∆TD incremental change in sand patch derived texture depth during analysis year
(mm)
ITD initial texture depth at construction of surfacing (mm)
TDa texture depth at the beginning of the analysis year (mm)
∆NELV number of equivalent light vehicle passes during the analysis year (one heavy
truck or heavy bus is equal to 10 NELV; light vehicles equal 1)
Ktd calibration factor for texture depth
The proposed default coefficient values for a0 for the texture depth model are given in Table
C2.33. This table also includes values for the initial texture depth (ITD) which is used as
default when resetting pavement surface type. These can be replaced by user definable
values.
CM 0.7 0.005
PA 1.5 0.008
SL 0.7 0.006
PM 1.5 0.008
The texture depth at the end of the analysis year is given by the following relationship:
where:
The annual average texture depth for a given analysis year is calculated as follows:
where:
TDav annual average texture depth for the analysis year (mm). This is the texture
depth used in the Road User Effects model.
where:
∆SFC50 incremental change in sideway force coefficient during the analysis year,
measured at 50 km/h
∆QCV annual incremental increase in the flow of commercial vehicles (veh/lane/day)
Ksfc calibration factor for skid resistance
The proposed default coefficient values for a0 for the skid resistance model are given in Table
C2.34.
CM -0.663 x 10-4
PA -0.663 x 10-4
SL -0.663 x 10-4
PM -0.663 x 10-4
The skid resistance measured at 50 km/h at the end of the analysis year is given by the
following expression:
where:
SFC50b sideway force coefficient, measured at 50 km/h, at the end of the analysis year
SFC50a sideway force coefficient, measured at 50 km/h, at the start of the analysis year
∆SFC50 incremental change in sideway force coefficient, measured at 50 km/h, during
the analysis year
The annual skid resistance value for a given analysis year is calculated as follows:
where:
SFC50av annual average sideway force coefficient, measured at 50 km/h, for the
analysis year
where:
The user needs to define a value of SFC50 in order for skid resistance modelling to be
performed. This will also need to be supplied after maintenance treatments.
13 Calibration factors
The deterioration models contain calibration factors to facilitate local calibration. These
factors have default values of 1.0 and are summarised in Table C2.35.
Edge-break Keb
14 References
Cenek P.D., and Griffith-Jones G., (1997)
Incremental Forms for Modelling Texture Depth and Friction Parameters in HDM-4
Communication to the ISOHDM
Works Central Laboratories, New Zealand
Djarf L., (1995)
Road deterioration and maintenance effects models in cold climates. Swedish Road
Research Institute, Sweden
Morosiuk G., (1996, 1998a & 1999)
Specifications for the HDM-4 Road Deterioration Model – Fourth, Fifth, Sixth, Seventh
and Eighth Drafts, ISOHDM, UK
Morosiuk G., (1998b)
Notes on the coefficient values for estimating drain life
Communication to the ISOHDM
Morosiuk G., (1998c)
Derivation of a new rut depth model for the structural deformation component in HDM-4
TRL Unpublished Project Report PR/ORC/610/98
Transport Research Laboratory, Crowthorne, UK
NDLI, (1995)
Modelling Road Deterioration and Maintenance Effects in HDM-4. Final Report Asian
Development Bank Project RETA 5549
N.D. Lea International, Vancouver
Odoki J.B., (1997)
Notes on the effects of patching on the progression of potholing. Communication to the
ISOHDM
Odoki J.B., (1998)
Notes on the logic for total damaged and undamaged carriageway surface areas
ISOHDM Working Paper, University of Birmingham, UK
Parkman C.C., and Rolt J., (1997)
Characterisation of pavement strength in HDM-III and possible changes for HDM-4
TRL Unpublished Project Report PR/ORC/587/97
Transport Research Laboratory, Crowthorne, UK
Paterson W.D.O., (1987)
Road Deterioration and Maintenance Effects
World Bank Publications
Washington, D.C., USA
Paterson W.D.O., (1998)
Notes on modelling the deterioration of drainage
Communication to the ISOHDM
C3 Concrete Pavements
1 Introduction
The prediction models for concrete pavement deterioration included in HDM-4 are based on
research carried out by the Latin American Study Team, in Chile (1996). The research
involved a comprehensive bibliographical review of the existing concrete pavement
deterioration models. The models implemented in HDM-4 are mainly based on previous work
carried out by SHRP (1993), Al-Omari and Darter (1994), Lee and Darter (1994) and ERES
Consultants (1995).
The process of model selection considered the following aspects:
! Validity range of the model
This chapter describes the Road Deterioration (RD) models for concrete pavements included
in HDM-4. It commences with an overview of the modelling framework, followed by a brief
analysis of the concrete pavement types and distress modes considered, and finally a complete
description of the models.
It should be noted that the RD models for concrete pavements are basically absolute models
(as opposed to incremental models used for bituminous pavements). The models have been
developed in imperial units. However, for consistency within HDM-4 user-interface, the data
required for modelling is initially input in metric. This data is then converted into imperial
units for use in the relationships, and then the results of calculations are converted back into
metric for reporting purposes.
There are different sets of deterioration models for concrete pavements included in HDM-4
that are based on pavement surface type and construction type (see Table C3.2). Calibration
parameters have also been provided to account for variations in surface material and to
facilitate local adaptation of the models.
Table C3.2 Surface types upon which the concrete RD models are based
! Phase 1
Refers to the time after the pavement has received a major maintenance or has been
reconstructed.
Phase 1 models are described in this chapter; Phase 2 models are discussed in Chapter D3.
Joint spacing
3 - 6 m
Slab Aggregate
Interlock
Base
Joint spacing
3 - 6 m
Dowels
Joint spacing
10 - 20 m
Slab
Dowels
Base
Cracks separation
Slab
Base
Reinforcement steel
0,6 - 0,8 % of area
Each of these distress modes is described in the following sections (2.2.1 - 2.2.6).
2.2.1 Cracking
There are three types of cracking distress that are commonly identified on concrete
pavements:
1 Transverse cracking
2 Longitudinal cracking
3 Durability cracks
Transverse cracks are predominantly perpendicular to the central axis of the road (see Figure
C3.5). They manifest three severity levels, according to SHRP (1993):
! Low
Cracks with a width of less than 3 mm, without visible spalling or faulting; or well
sealed, with a non-determinable width.
! Medium
Cracks with a width between 3 and 6 mm, or with spalling less than 75 mm, or faulting
less than 6 mm.
! High
Cracks with a width greater than 6 mm, or spalling greater than 75 mm, or faulting
greater than 6 mm.
Distress Distress
width width
A B C D
Longitudinal Joint
C
L
C D
Transv. Transv.
Joint Joint
A B Traffic
Slab
Shoulder
Transverse cracking may have significant impact on the riding quality and are therefore
modelled in HDM-4.
Longitudinal cracks are predominantly parallel to the axis of the road. Durability cracks are
fine cracks, slightly spaced, and often occur adjacent to joints, cracks, or free edges. They
begin in the slab corner as a group of obscure cracks just as in the surrounding area.
A B
Longitudinal Joint
C
L
Transv.
Transv. Joint
Joint
A B
Traffic
Slab
Shoulder
Spalling of less than 75 mm of distress width, measured from centre of the joint, with or
without loss of material.
! Medium
Spalling of between 75 and 150 mm of distress width, measured from centre of the joint,
with loss of material.
! High
Spalling of greater than 150 mm of distress width, measured from the centre of the joint,
with loss of material.
A B C D
Crack Joint
Joint
Transv. Transv. Transv.
Joint Joint Joint
Low Sev.:
1,8 m Low Sev.: C D
2m Moder. Sev.:
2,5 m
High Sev.:
1,5 m Traffic
A B
Shoulder
2.2.4 Failures
This distress is a major defect that occurs in Continuously Reinforced (CR) concrete
pavements. Located failures include loosening and breaking of reinforcement steel and
transverse crack spalling. Failures are measured in number per mile (or km).
Many of the maintenance activities in CR concrete pavements are directly related to failures.
Often, it is necessary to estimate the quantity of these distresses for the purpose of preventive
designs and rehabilitation planning.
PSR Condition
0-1 Very poor
1-2 Poor
2-3 Fair
3-4 Good
2.2.6 Roughness
This is a measure of the deviations of a surface from a true planar surface with characteristic
dimensions that affect vehicle dynamics, ride quality, dynamic loads and drainage (ASTM E-
867-82A) - typically in the ranges of 0.1 to 100 m wavelength and 1 to 100 mm amplitude.
International Roughness Index (IRI), is the reference measure expressing roughness as a
dimensionless average rectified slope statistic of the longitudinal profile and defined in Sayers
et al. (1986).
Road condition
The pavement and side-drain condition data at the beginning of the first analysis year or the
first year after construction are required as inputs. The data for pavement condition is as
described above in Section 2.2.
The average pavement condition indicators in a given analysis year (that is, before road
works) are predicted using absolute models. Absolute models predict condition (or distress) at
a particular point in time as a function of the independent variables, and can be represented as
follows:
where:
Pavement history
The required data items refer to pavement age and the year of previous major maintenance
and construction works carried out on the pavement.
Traffic
The required traffic data are the annual flow of equivalent standard axle loads (ESAL) and the
cumulative equivalent standard axle loads (NE4), both expressed in millions per lane. These
data are calculated for each analysis year based on the user-specified traffic and vehicle
characteristics.
8 Convert the required outputs into metric for use in the RUE and SEE modules and for
reporting
3 Structural characteristics
3.1 Introduction
This section describes the principal pavement structural data that are necessary to predict the
deterioration of concrete pavements. These include the following:
! Properties of the materials (see Section 3.2)
where:
The value of the Modulus of elasticity of concrete used in the pavement deterioration
models is 5,000,000 psi
! Modulus of Rupture of concrete (MR28)
Stresses in concrete pavements are mainly caused by the effects of traffic and
environmental action. The Modulus of Rupture is a measure of the concrete flexural
strength in providing a sustained resistance to the stresses. During the useful life of the
pavement the stress levels may exceed the Modulus of Rupture at certain points, causing
fatigue damage and cracking in the slabs.
The Modulus of Rupture measured after 28 days and denoted by MR28 can be
determined using AASHTO T97 and ASTM C78 procedures, or estimated from the
compressive strength of concrete, as follows:
where:
The Modulus of Rupture can also be estimated using the modulus of elasticity of
concrete, which can be obtained from Falling Weight Deflectometer (FWD) test results
or from laboratory testing. The empirical equation (Foxworthy, 1985) for estimating the
Modulus of Rupture is as follows:
E
MR = 43.5 * c + 488.5 ...(3.3)
10 6
where:
The deterioration models for concrete pavements consider the modulus of rupture (MR)
in the long term. The long-term value is estimated by increasing the modulus of rupture
at 28 days (MR28) by 11%.
! Thermal coefficient of concrete (α)
The thermal coefficient of expansion is used to determine the warping (or curling)
stresses produced in a concrete pavement when it is subjected to a temperature difference
between the top and the bottom of the slab. The stresses are greatest at the edges of the
slab, and may result in slab cracking usually near its mid point.
The thermal coefficient of expansion varies with such factors as water to cement ratio,
concrete age, richness of the mix, relative humidity, and the type of aggregate in the mix.
Table C3.5 shows typical values of the Thermal coefficient of concrete according to the
type of aggregate. A value of 5.5 x 10-6 per ºF is commonly used in concrete pavement
analysis.
Sandstone 6.5
Gravel 6.0
Granite 5.3
Basalt 4.8
Limestone 3.8
Concrete pavement slabs are subjected to daily variations in temperature and humidity
with associated expansion or contraction effects. A slab with unrestricted movement in
the horizontal direction would not develop stresses under the effects of expansion and
contraction. However, in real site conditions there exists some resistance between the
slab and the underlying base.
Shrinkage in concrete pavements is caused by the loss of water in the drying process. The
drying shrinkage coefficient (γ) is used in the evaluation of the opening and closing of the
joints caused by variations in the mean temperature to which the slabs are subjected.
The shrinkage and strength of concrete are strongly dependent upon the water to cement
ratio. High values of water to cement ratio will reduce the strength of the concrete and
increase the drying shrinkage potential. Shrinkage can therefore be considered to be
inversely proportional to the indirect tensile strength of the concrete. Table C3.6 can be
used to estimate the drying shrinkage coefficient of concrete.
400 0.0006
500 0.00045
600 0.0003
For most cement treated materials, the value of µ normally varies between 0.10 and 0.25,
with 0.15 generally accepted as a representative value.
! Modulus of elasticity of dowel bars (Es)
Dowel bars may be used to transfer (or distribute) load across discontinuities such as
transverse joints. The value of the Modulus of elasticity of load transfer dowels assigned
within the deterioration model is 2.9*107 psi (or 2.0*105 MPa).
! Modulus of elasticity of bases (Ebase)
The stiffness of base influences the overall behaviour of concrete pavements, mainly as a
result of the support provided to the slabs. Drainage effects also have a significant
influence on the behaviour of the base, as discussed below. A more rigid base will
generally provide better support to the slab, and this should reduce the occurrence of
faulting at transverse joints. However, a very rigid base may increase the warping effect
induced by changes in temperature or humidity, and transverse cracking will increase.
Table C3.7 gives typical values of Modulus of elasticity for different base types.
(in psi)
Granular (GB) 25,000
The effects of a stabilised base is considered in the cracking model, see Section 4.
! Modulus of subgrade reaction (KSTAT)
The modulus of reaction of a material is an elastic constant that defines the stiffness of
the material or resistance to deformation, under certain loading conditions. The Modulus
of subgrade reaction (KSTAT) is defined by Equation 3.4 below:
RPRESS
KSTAT = ...(3.4)
DEF
where:
The value of KSTAT can be determined through the plate load test, where the deflection
is the displacement of a circular plate of 30 inches diameter subjected to a static pressure.
KSTAT is expressed in pounds per cubic inch (pci). The assumption associated with the
determination of the value of KSTAT is that the plate is in complete contact with the
subgrade soil and that the soil is elastic.
This is equivalent to the saturation time (T50) given in Table C3.8, and the associated values
for the drainage coefficient Cd are given in Table C3.9 (AASHTO, 1986).
Table C3.8 Relationship between drainage time and the quality of drainage
Good 1 day
Fair 7 week
Bad 1 month
Table C3.9 Recommended values of the drainage coefficient (Cd) for concrete
pavements
Table C3.10 shows a simplified matrix that can be used to estimate the drainage coefficient
(FWHA, 1995).
Table C3.10 Modified AASHTO simplified matrix for drainage coefficient (Cd)
Dry (arid) 0.95 - 1.05 0.90 - 1.10 0.90 - 1.15 1.00 - 1.10
Yes Wet (humid) 1.00 - 1.10 0.75 - 0.95 0.90 - 1.10 1.05 - 1.15
Dry (arid) 1.10 - 1.20 0.95 - 1.15 1.10 - 1.20 1.15 - 1.20
Notes:
DEFunld
LT = * 100
...(3.5)
DEFload
where:
The efficiency of load transfer is used in the calculation of the maximum bearing stress of the
dowel-concrete system. Theoretically, if a dowel is 100% efficient it is capable of assigning
half of the applied load to each adjacent slab. However, a reduction in load transfer efficiency
would occur over the pavement life, either due to the loss of bond in the zone where the load
transfer device is imbedded in the concrete slab or due to the deterioration of the aggregate
interlock mechanism. Generally, the reduction in load transfer efficiency increases as traffic
loads increase since aggregate load transfer decreases with load repetitions. The reduction in
the load transfer efficiency can be assumed to be around 5% to 10%, therefore the value of LT
used in the deterioration model is 45%.
STRESS unld
LTE sh = * 100
...(3.6)
STRESS load
where:
If tied concrete shoulders are provided in the original pavement construction, a value for
LTEsh = 20% should be used. If the shoulders are provided on an existing pavement the value
of LTEsh should be taken as 10%.
4 Cracking
The HDM-4 cracking model considers transverse cracking in concrete pavements due to high
stress levels in the slabs or defects originating from material fatigue. The stresses are caused
generally by the combined effect of thermal curling, moisture-induced curling and traffic
loading.
Separate relationships are given for predicting the amounts of transverse cracking over the life
cycle of Jointed Plain concrete pavements and Jointed Reinforced concrete pavements. The
models are deterministic and predict the expected average deterioration based on the input
variables.
100
PCRACK = Kjp c * ...(4.1)
1 + 1.41 * FD -1.66
where:
∑N
n tg
FD = ...(4.2)
tg
tg=1
where:
According to Miner’s theory, transverse cracking is expected to occur when the cumulative
fatigue damage (FD) approaches 1.0.
! Temperature gradients
The variations in concrete pavement temperature over the year can be represented by a
distribution of temperature gradients. The average temperature gradient is defined as the
difference between the temperature at the top and at the bottom of the slab divided by the
slab thickness. A positive gradient indicates the top of the slab is warmer than the
bottom, which normally occurs during the daytime. Negative gradient condition typically
occurs during the cooler hours of the evening. In all types of climate the positive
temperature gradients occur with greater frequency than the negative temperature
gradients.
Owing to the difficulty that may be experienced in obtaining field data on temperature
gradient distribution, a default data set based on climate zones is provided in HDM-4 as
illustrated in Table C3.11.
Temperature Frequency
difference
(FREQ)
∆T) in °F
(∆ Dry with Dry without Wet with Wet without
freezing freezing freezing freezing
-8 0.086660 0.073237 0.090494 0.086209
Note: The frequencies do not add up to 1.0 because the data relative to temperature
differences of negative 8 ºF do not provide significant (meaningful) information to the
concrete models.
There are factors other than temperature that cause curling (which may be concave) in
slabs. A correction to the difference in temperature measured in the slab is applied
according to climate zones as follows, Eisenmann and Leykauf (1990):
a1 * (SLABTHK - 2 )
∆Ts = ∆T - a0 + ...(4.3)
SLABTHK 3
where:
∆Ts adjusted difference in temperature at the top and bottom of the slab (ºF)
∆T difference between the temperature measured at the top and bottom of the
slab (ºF )
(= Ttop - Tbottom)
SLABTHK slab thickness (inches)
a0 and a1 model coefficients based on climate zones
Climate type a0 a1
Dry with freezing 6.29 436.36
The total traffic loading since the construction of pavement is distributed over the
temperature gradients as follows:
NE4
n tg = * FREQ tg ...(4.4)
LCR tg
where:
The lateral coverage ratio of traffic (LCR) is simply a measure of the likelihood of the
wheel loading passing through the critical edge location. The edge loading location is
considered critical for jointed plain concrete pavements as this is the location of the
maximum stress and will be the point of crack initiation.
Assuming that the average location of vehicle wheels is 22 inches from the edge of the
slab, with a standard deviation of 8.4 inches, the following regression equation can be
used to calculate LCR:
SIGMA tg
SR tg = ...(4.6)
MR
where:
The maximum number of load repetitions to the failure of concrete slab (N) applied
during temperature gradient tg depends on the induced stress level, and is calculated
through the law of fatigue as follows:
where:
where:
The stress at the slab edge produced by traffic loads, for each temperature gradient, is
expressed as follows:
where:
Edge stress in the slab is calculated using Westergaard’s equation (Westergaard, 1948)
for a circular load, in which the load application radius for a simple axle is replaced by
the equivalent radius for a single axle dual wheel, as follows:
3 * (1 + µ ) * P E c * SLABTHK 3 4µ 1 - µ a eq
σe = ln + 1.84 -
+ + 1.18 * (1 + 2µ )
π (3 + µ ) * SLABTHK 2 4
100 * KSTAT * a eq 3 2 l
...(4.10)
where:
µ Poisson’s ratio
P total load applied by each wheel of a single-axle dual wheel (lb), default =
9000
SLABTHK slab thickness (inches)
Ec modulus of elasticity of concrete (psi)
KSTAT modulus of subgrade reaction (pci)
aeq equivalent load application radius for a dual-wheel single axle (inches)
radius of relative stiffness of the slab-foundation system (inches)
The equivalent load application radius (aeq) is calculated from the following expression:
2 2
SP a SP SP a
0.909 + 0.339485 * a + 0.103946 * l - 0.017881 * a - 0.045229 * a * l
a eq
=
a 3 3 2 3
+ 0.000436 * SP - 0.301805 * SP * a + 0.034664 * SP + 0.001 * SP * a
a a l l a l
...(4.11)
aeq equivalent load application radius for a dual-wheel single axle (inches)
a load application radius for a single-wheel single axle, in inches. This is
given by the square root of (P/π*p)
p tyre pressure (psi)
SP spacing between central wheels of dual wheel single axle (inches)
l radius of relative stiffness of the slab-foundation system (inches)
The radius of relative stiffness of the slab-foundation system is calculated from the
following expression:
0.25
E c * SLABTHK 3
l= ...(4.12)
12 * (1 − µ ) * KSTAT
2
where:
In pavement sections with concrete shoulders or other forms of edge support (such as
kerb), load stress should be multiplied by the edge support adjustment factor, calculated
as follows:
100
fES = ...(4.13)
100 + LTE sh
where:
In sections with widened lanes, the load stress should be multiplied by an adjustment
factor, calculated as follows (Benekohal et al., 1990):
2 3
0.013211 * l
+ 0.386201 *
a a + 0.053891 * a
f WL = 0.454147 + - 0.24565 *
DW DW DW DW
...(4.14)
where:
COEF * E c * α * ∆Ts
σ curl = ...(4.15)
2
where:
∆Ts adjusted difference in temperature at the top and bottom of the slab (ºF)
The curling stress coefficient (COEF) in Equation 4.15 above is obtained from Equation 4.16
below developed by Westergaard (1926) and Bradbury (1938):
and:
12 * JTSPACE
λ= ...(4.17)
l* 8
where:
The load induced stresses and curling stresses cannot be added directly since curling
produces a debonding effect between the slab and the base. This effect is taken into
account by the regression coefficient (Rtg) in Equation 4.8 above.
The regression coefficient is calculated for each temperature gradient using the following
equation:
(
86.97 * Y 3 - 1.051 * 10 -9 * E c * dT * KSTAT + 1.7487 * dT * Y 2 )
(
R = − 1.068 - 0.387317 * dT - 1.84 * 10 -11 2 2
* E c * dT * KSTAT + 8.16396 * dT * Y )
(
+ 1.062 - 1.5757 * 10 -2 * dT - 8.76 * 10 -5 * KSTAT + (1.17 - 0.181 * dT) * 10 -11 * E * dT * KSTAT
c
)
...(4.18)
The intermediate parameters Y and dT in Equation 4.18 above are expressed as follows:
12 * JTSPACE
Y= ...(4.19)
100 * l
dT = α * ∆Ts * 10 5 ...(4.20)
where:
∆Ts adjusted difference in temperature at the top and bottom of the slab (ºF)
2 * (SLABTHK - NAXIS)
f SB = ...(4.21)
EFFETHK
where:
The location of the neutral axis (NAXIS) and the effective slab thickness (EFFETHK)
are calculated from Equations 4.22 below and 4.23 below, respectively:
E base
2
0.5 * SLABTHK + E * BASETHK * (SLABTHK + 0.5 * BASETHK )
NAXIS = c
E base
SLABTHK + * BASETHK
E c
...(4.22)
0.5
E * BASETHK
EFFETHK = SLABTHK 2 + BASETHK 2 * base
E c * SLABTHK
...(4.23)
where:
...(4.24)
where:
Note that this model does not use the spacing between joints to predict crack deterioration.
The model highlights the following:
! Crack deterioration increases with pavement age and traffic.
! A significant increase in the quantity of longitudinal steel reinforcement (greater than
0.15%) reduces the number of deteriorated cracks.
! Crack deterioration occurs at a higher rate in cold and wet climates (higher values of MI).
! Stabilised bases gives less cracking than non-stabilised bases.
! Crack deterioration can be reduced by using higher strength concrete or by increasing
slab thickness.
5 Faulting
Faulting is caused by the loss of fine material under a slab and the increase in fine material
under nearby slabs. This flow of fine material is called pumping, and is caused by the
presence of high levels of free moisture under a slab carrying heavy traffic loading. The
effects of thermal and moisture-induced curling and lack of load transfer between slabs
increase pumping.
The HDM-4 pavement deterioration model considers faulting in jointed plain concrete
pavements (with and without load transfer dowels) and jointed reinforced concrete pavements
as described in Sections 5.1, 5.2 and 5.3.
...(5.1)
where:
The following design characteristics can be used to reduce faulting in jointed plain concrete
pavements without load transfer dowels:
! Provision of better drainage conditions
Since this model predicts the average faulting, it is recommended that the critical level of
intervention should be set fairly low (at around 0.07 in) to provide some safety factor. In
situations where faulting in JP concrete pavements without load transfer dowels is predicted to
be excessive, the use of load transfer dowels has to be considered.
...(5.2)
where:
The value of the maximum concrete bearing stress (BSTRESS) has a significant impact on the
predictions of the faulting model, and is calculated as follows (Heinrichs et al., 1989):
where:
The relative stiffness of the dowel-concrete system (BETA) is obtained from Equation 5.4
below:
0.25
Kd * DOWEL
BETA = ...(5.4)
4 * E s * INERT
where:
α * TRANGE
OPENING = 12 * CON * JTSPACE *
+γ ...(5.5)
2
where:
where:
INERT moment of Inertia of the transverse section of the dowel bar (in4)
DOWEL dowel diameter (inches)
Based on model results, the following design characteristics can reduce faulting:
! Use of load transfer dowels of greater diameter to reduce the stress levels at the
dowel-concrete support system
! Provision of better drainage conditions
Faulting in JP concrete pavements is more likely to occur in cold and wet climates than in
warm climates.
...(5.7)
where:
The following design characteristics can reduce faulting in JR concrete pavements and JP
concrete pavements with dowels:
! Use of load transfer dowels of greater diameter to reduce the stress levels at the
dowel-concrete support system
! Provision of better drainage conditions
Note that the model does not include the effects of shoulder types.
6 Spalling
Transverse joint spalling is the cracking or breaking of the edge of the slab up to a maximum
of 0.6 metres from the joint. Spalling generally does not extend through the whole thickness
of the slab, but intercepts the joint at an angle. Transverse joint spalling can be caused by a
variety of factors including:
! Presence of incompressible materials
The presence of incompressible materials in the joint which produces excessive stress in
the joint. This produces a fracture or detachment of the joint edges when the slab expands
in warm conditions.
! Disintegration of concrete under high traffic loads
! Improper consolidation of the concrete in the joint
! Wrongly designed or built load transfer system
The HDM-4 model considers medium and high severity spalled transverse joints.
...(6.1)
where:
The following observations can be made regarding the behaviour of the JP concrete pavement
model:
! The transverse joint spalling increases with the pavement age to the second power
! Pre-formed seals are more effective in reducing joint spalling than other types of
seals considered
! Transverse joints without seals, show a great quantity of spalling
! Liquid sealant gives a better performance than the silicon seals
! An increase in joint spacing increases the percentage of spalled joints
! An appropriate dowel protection against corrosion reduces joint spalling
The model also shows that the ageing of JP concrete pavements has a significant effect on
joint spalling. Effective maintenance strategies that include joint cleaning and resealing at
regular intervals can reduce the ageing effect, and therefore, reduce joint spalling
significantly.
where:
The following observations can be made regarding the behaviour of the JR concrete pavement
model:
! The percentage of spalled joints increases rapidly with the age of the pavement
! Transverse joint spalling is more likely to occur in cold climates than in warm
climates
! An increase in joint spacing results in an increase of the percentage of spalled
joints
! Protecting dowels against corrosion will reduce spalling
! The use of a stabilised dense base increases joint spalling
! Pre-formed seals reduce spalling in pavements with stabilised bases
7 Failures
This is the main deterioration mode that occurs in Continuously Reinforced concrete
pavements. Located failures include loosening and breaking of reinforcement steel and
transverse crack spalling. These are caused by high tensile stresses induced in the concrete
and reinforcement steel by traffic loading and changes in environmental factors. Most
maintenance activities carried out on CR concrete pavements are directly related to failures.
Failures in continuously reinforced concrete pavements are predicted using Equation 7.1
below (Lee et al., 1991):
...(7.1)
where:
FAIL number of failures per mile in the more trafficked lane (number/mile)
SLABTHK slab thickness (in)
PSTEEL percentage of longitudinal reinforcement steel (%)
NE4 cumulative equivalent standard axle load (ESALs) since pavement
construction (millions per lane)
AB 1 if base type is asphaltic
0 in other cases
SB 1 if base type is cement stabilised
0 in other cases
GB 1 if base type is granular
0 in other cases
CHAIRS 1 if chairs are used for installation of the reinforcement
0 if tubes are used
Kcrf calibration factor (default = 1.0)
Slab thickness and percentage of reinforcement steel have a significant effect on the number
of failures in CR concrete pavements. Installation of reinforcement steel with chairs produces
fewer failures than installation using tubes. The use of stabilised or asphaltic bases further
reduces the risk of failures.
8 Serviceability loss
Present Serviceability Rating (PSR) is a subjective user rating of the existing ride quality of
pavement condition. PSR has been correlated with various roughness indicators, such as slope
variance and IRI. It is a reflection of the user response to pavement condition.
where:
FAULT * 5280
TFAULT = ...(8.2)
JTSPACE
where:
It should be noted that cracking has a very significant influence on the predictions of the PSR
model.
where:
9 Roughness
where:
The total number of cracked slabs per mile (TCRACKS) is calculated as:
PCRACK * 5280
TCRACKS = ...(9.2)
JTSPACE * 100
where:
RI t = Kjrr * − log e
0.2 * PSR t
...(9.3)
0.0043
where:
RI t = Kcrr * − log e
0.2 * PSR t
...(9.4)
0.0043
where:
10 Calibration factors
The deterioration models contain calibration factors to facilitate local calibration. The
calibration factors have default values of 1.0 and are summarised in Table C3.13.
11 References
The AASHO Road Test, (1962)
Report 5 - Pavement research, Special Report No. 61E
The Highway Research Board
Washington DC, USA
AASHTO, (T22-92) American Association of State Highway and Transportation Officials
Standard Method of Test for Compressive Strength of Cylindrical Concrete Specimens
AASHTO, (T140-92) American Association of State Highway and Transportation Officials
Standard Method of Test for Compressive Strength of Concrete Using Portions of Beams
Broken Inflexure (ASTM C116)
AASHTO, (T97) American Association of State Highway and Transportation Officials
Standard Method of Test for Time of Setting of Concrete Mixtures by Penetration
Resistance (ASTM C403-90)
AASHTO, (1986) American Association of State Highway and Transportation Officials
Guide for Design of Pavement Structures
Washington DC, USA
AASHTO, (1993) American Association of State Highway and Transportation Officials
Guide for Design of Pavement Structures
Washington DC, USA
ASTM C39, American Society for Testing Materials (1994)
Standard Test Method for Compressive Strength of Cylindrical Concrete Specimens
ASTM C78, American Society for Testing Materials (1994)
Standard Test Method for Flexural Strength of Concrete (Using Simple Beam with
Third-Point Loading)
ASTM C469
ASTM E-867-82A
Al-Omari B., and Darter M.I., (1994)
Relationships Between IRI and PSR
Transportation Research Board, 73rd Annual Meeting
Washington DC, USA
Benekohal R.F., Hall K.T., and Miller H.W., (1990)
Effects of Lane Widening on Lateral Distribution of Truck Wheels
Transportation Research Record 1286
Transportation Research Board
Washington D.C.
C4 Unsealed Roads
1 Introduction
This chapter describes the detailed modelling of unsealed road deterioration (see Figure C4.1).
Road
RoadDeterioration
Deterioration
Models
Models
The HDM-4 Road Deterioration model for unsealed roads is based on the specifications given
in the HDM-III documentation by Watanatada et al. (1987) reproduced with the approval of
the World Bank. Minor modifications have been made in the text and to the models by
incorporating calibration factors to facilitate local calibration and adaptation. The background
of the model is given in Paterson (1987).
A list of research documents referenced from this chapter is given in Section 7.
2 Modelling logic
reverts to that of an earth road. Upon gravel resurfacing, all unpaved roads become gravel
roads by definition of the new surfacing layer.
Deterioration is predicted using the properties of the surfacing layer, whether that is gravel or
subgrade, as it is defined for the analysis year. Thus the user must specify the physical
properties of both gravel surfacing and subgrade for unpaved roads.
The pavement classification framework for defining the modelling of unsealed road
deterioration and works effects is shown in Table C4.1.
NDLI (1995) give definitions of the characteristics used to define different types of pavements
into the above framework and alternative terminology applied to the same pavement materials
(see also Chapter C1).
Variable Definition
ADH the average daily heavy vehicle traffic (GVW ≥ 3,500 kg) in both directions (veh/day)
ADL the average daily light vehicle traffic (GVW < 3,500 kg) in both directions (veh/day)
D95j the maximum particle size of the material, defined as the equivalent sieve opening through which
95% of the material passes (mm)
P075j the amount of material passing the 0.075 mm sieve (or ASTM No. 200 sieve) (% by mass)
P425j the amount of material passing the 0.425 mm sieve (or ASTM No. 40 sieve) (% by mass)
P02j the amount of material passing the 2.0 mm sieve (or ASTM No. 10) (% by mass)
QIMINj the minimum roughness of the material (either estimated in Section 3.4 or specified) (QI)
QIMAXj the maximum roughness of the material (either estimated in Section 3.2 or specified) (QI)
by the rise plus fall variable, RF). Roughness progression, and in particular the maximum
roughness, is influenced by both characteristics. In material loss prediction the horizontal
curvature affects the rate of traffic-induced material whip-off and the gradient interacts with
rainfall in causing erosion. Cross-sectional geometry, including crown, camber and
superelevation, were not measured in the study and are discussed in the following section. The
average shoulder width (SW) is used to compute the amount of gravel used in spot
regravelling and gravel resurfacing. The variables RF, C and SW are defined in Table C4.2.
4 Compute the depth of material loss during the analysis year as a function of the traffic
volume, monthly rainfall, and road geometry and the attributes of the gravel (if gravel
road) or the subgrade (if earth road) (see Section 4).
5 Store the results for later use in the RUE (see Part E) and WE (see Part D) models and in
the evaluation and reporting phase.
If the analysis year is neither the first year of the analysis period nor an
improvement/construction opening year;
2 From the existing section characteristics data
3 Road roughness
3.1 General
The roughness of unpaved roads increases through the shear, mechanical disintegration, and
erosion of the surfacing material caused by traffic and surface water runoff. Roughness levels
are usually 4 to 15 m/km IRI (50 to 200 QI) although lower levels sometimes occur with fine
materials. Roughness in excess of 13 m/km IRI (180 QI) is usually related to depressions,
potholes or transverse erosion gullies, and levels above 22 m/km IRI (300 QI), which
correspond to numerous wheel-sized potholes, are very rare and usually apply only on short
sections or unclassified tracks. The roughness modelled for economic analysis is the profile in
the wheel-paths of the traffic, since this generates the vehicle operating costs. The location of
the wheel-paths tends to vary when roughness reaches high levels as vehicles seek to
minimise the dynamic impact. Hence the prediction of roughness progression must take this
self-regulating tendency into account. On account of the high variability of:
! Material properties,
! Drainage,
! Surface erosion, and
! High roughness levels of unpaved roads,
prediction errors tend to be large, in the order of 1.5 to 2.5 m/km IRI (20 to 32 QI) standard
error, or equivalent to 95 percentile confidence intervals of 20% to 40%.
A number of different model forms have been applied to roughness progression and to the
effects of maintenance grading (Hodges et al., 1975; Visser, 1981; Paterson, 1987). As the
objective of policy analyses can be satisfied by computation of the average roughness
resulting from a specified policy, the model selected for predicting roughness was one which
both represented the progression and grading phases of the roughness cycle realistically and
also permitted a closed-form solution. The model form and its derivation are described in
detail elsewhere (Paterson, 1987). The primary principles and parameter estimates are
described in Sections 3.2 - 3.6.
Although the IRI roughness measure, or other compatible measure, could have been used in
the following relationships (see Sections 3.2 - 3.6) because many parameters are
dimensionless, the nomenclature of QI is used for roughness for internal consistency with the
remainder of the model.
where:
( )
c = - 0.461 + 0.0174 * ADL + 0.0114 * ADH - 2.87 * AADT * MMP * 10 -5 * 10 −3
…(3.3)
where:
[
QIMAX j = MAX 279 - 421 (0.5 - MGD j ) + 0.22 * C − 9.930 * RF * MMP * 10 −3 , 150
2
]
…(3.4)
where:
if P425j = 0, then:
MGD j = 1 …(3.5)
P075 j
MGD j = …(3.6)
P425 j
where:
Note: The standard error of this prediction on the original database was 1.5 m/km IRI
(19.8 QI).
[ (
c' = c * MIN 1, 0.25 * t * MAX 1, n 0.33 )] …(3.7)
where:
and thus:
c'
…(3.8)
b' = exp365 *
n
where:
b', c' are the values of b and c (see Section 3.2) when mechanical compaction is in
effect
where:
QIMIN j = MAX {10, MIN [100, 4.69 * D95 j * (1 − 2.78 * MG j )]} …(3.11)
where:
P075 j
log e
95
MG075 j = …(3.14)
log e 0.075
D95 j
otherwise:
MG075 j = 0.3
P425 j
log e
95
MG425 j = …(3.15)
log e 0.425
D95 j
otherwise:
MG425 j = 0.3
P02 j
log e
95
MG02 j = …(3.16)
log e 2.0
D95 j
otherwise:
MG02 j = MG425 j
Note: The standard error of this prediction on the original database was 2.4 m/km IRI (31.6
QI).
Case 1: if (t*n) ≥ 1
The average roughness during year t (QIavg) is computed as follows:
y * NS
QIavg = QIMAX j * (1 - y ) + …(3.17)
n
(b - 1) * n
y= …(3.18)
365 * c
[ ] (
k 1 − (a * b )n
n * k + 1 − (a * b ) * QIa −
n )
(1 − a * b )
NS = …(3.19)
(1 − a * b)
k = (1 - a) * QIMIN j + a * (1 − b ) * QIMAX j …(3.20)
where:
For the first year of analysis period when t = 1, QIa = QIo (the value specified by the
user).
! Subsequent analysis year
For any subsequent analysis year t, QIa = QIb (= roughness at end of the previous year t-
1, as given below in QI).
In any given analysis year t, the roughness at the end of the year (QIb) is given by:
QIb = (a * b ) * QIa +
n [
k * 1 − (a * b )
n
] …(3.21)
(1 − a * b)
where:
All the parameters are as defined previously.
[exp(365 * c ) - 1]
QIavg = QIMAX j - (QIMAX j - QI a ) * …(3.22)
365 * c
where:
All the parameters are as defined previously.
[QIMAX j * (1 − b ) + QIMIN j * (1 − a ) * b]
QIH = …(3.24)
(1 − a * b)
where:
and:
[QIMAX j - QIMIN j ]
QIlta = QIMAX j + (1 - a ) * (1 − b ) * …(3.27)
[(1 - a * b ) * log e b]
where:
All the parameters are as defined previously.
These relationships are illustrated in Watanatada et al. (1987).
4 Material loss
From the Brazil-UNDP study the following relationship for predicting the annual quantity of
material loss as a function of monthly rainfall, traffic volume, road geometry and
characteristics of the gravel (if a gravel road) and the subgrade (if an earth road) was
obtained (Paterson, 1985):
(
MLA = K gl * 3.65 3.46 + 2.46 * MMP * RF * 10 -4 + KT * AADT ) …(4.1)
where:
0.969 * C
0.022 + + 3.42 * MMP * P075 j * 10 - 6
57300
KT = K kt * MAX 0, …(4.2)
- 9.2 * MMP * PI j 10 - 6 - 1.01 * MMP * 10 - 4
where:
5 Passability
Passability is the quality of the road surface that ensures the safe passage of vehicles. In the
vehicle operating cost model, provision has been made to determine the economic impact of a
partial reduction in passability through factors augmenting the operating costs of the various
vehicle types (see Part E). This augmentation comes into effect when the gravel surfacing
thickness drops below a minimum, and relates to the risk of the subgrade material being
impassable.
The user however must determine endogenously whether passability will be a problem in the
subgrade material, because no physical estimation of it is made within the model. The
following criteria from Visser (1981) are adequate for ensuring passability and surface
stability:
1 Passability
is a function of the shear strength of the saturated material, and is satisfactory when:
2 Surfacing stability
P075 ≥ 14 …(5.2)
where:
6 Calibration factors
The deterioration models contain calibration factors to facilitate local calibration. These
factors have default values of 1.0 and are summarised in Table C4.3.
7 References
GEIPOT, (1982)
Research on the Interrelationships between Costs of Highway Construction, Maintenance
and Utilisation (PICR)
Final Report 12 Volumes
Brasilia, Brazil
Hodges J.W., Rolt J., and Jones T.E., (1975)
The Kenya Road Transport Cost Study: Research on Road Deterioration
Report LR 673. Crowthorne, England
Department of the Environment, Transport and Road Research Laboratory
NDLI, (1995).
Modelling Road Deterioration and Maintenance Effects in HDM-4
Final Report Asian Development Bank Project RETA 5549
N.D. Lea International, Vancouver
Paterson W.D.O., (1987)
Road Deterioration and Maintenance Effects
World Bank Publications, Washington, D.C., USA
Visser A.T., (1981)
An Evaluation of Unpaved Road Performance and Maintenance
Ph.D. thesis. Austin, Texas
University of Texas at Austin, department of Civil Engineering
Watanatada T., Harral C.G., Paterson W.D.O., Dhareshwar A.M., Bhandari A., and
Tsunokawa K., (1987)
The Highway Design and Maintenance Standards Model - Volume 1: Description of the
HDM-III Model
World Bank Publications, Washington, D.C., USA
Contents
D1 Types of roadworks
1 Introduction D1-1
2 Road works classification D1-3
2.1 Works categories D1-4
2.2 Works classes D1-4
2.3 Works types D1-5
3 Modelling framework D1-9
3.1 Road feature types D1-9
3.2 Works standards D1-9
3.3 Intervention criteria D1-10
3.4 Works design D1-11
3.5 Works duration D1-11
3.6 Unit costs of works D1-14
3.7 Works effects D1-14
4 Costs to road administration D1-15
4.1 Annual costs and budget categories D1-15
4.2 Cost spread over works duration D1-15
4.3 Salvage value D1-16
5 References D1-17
D2 Bituminous Pavements
1 Introduction D2-1
2 Modelling logic D2-2
2.1 Overall computational procedure D2-2
2.2 Hierarchy of works D2-2
2.3 Pavement types reset D2-4
3 Routine maintenance D2-6
3.1 Patching D2-6
D3 Concrete Pavements
1 Introduction D3-1
2 Modelling logic D3-3
2.1 Overall computational procedure D3-3
2.2 Hierarchy of works D3-3
2.3 Road characteristics reset D3-5
3 Routine maintenance D3-6
4 Preventive treatment D3-7
4.1 Load transfer dowels retrofit D3-7
4.2 Tied concrete shoulders retrofit D3-8
4.3 Longitudinal edge drains retrofit D3-10
4.4 Joint sealing D3-12
5 Restoration works D3-15
5.1 Slab replacement D3-15
5.2 Full depth repairs D3-19
D4 Unsealed Roads
1 Introduction D4-1
2 Modelling logic D4-2
2.1 Overall computational procedure D4-2
2.2 Hierarchy of works D4-2
3 Maintenance works D4-4
3.1 Periodic grading D4-4
3.2 Spot regravelling D4-6
3.3 Gravel resurfacing D4-7
3.4 Routine-miscellaneous maintenance D4-9
4 Improvement works D4-11
4.1 Widening D4-11
4.2 Realignment D4-16
5 Construction works D4-21
5.1 Upgrading D4-21
5.2 Dualisation D4-22
5.3 New section D4-23
6 References D4-24
Analytical
AnalyticalFramework
Framework
and
andModel
Model
Descriptions
Descriptions
Introduction
Introduction
Part
PartAA
Traffic
Traffic
Part
PartBB
RD
RDModel
Model WE
WEModel
Model RUE
RUEModel
Model SEE
SEEModel
Model
Part
PartCC Part
PartDD Part
PartEE Part
PartFF
Economic
EconomicAnalysis
Analysis
RD = Road Deterioration Part
PartGG
Nomenclature
Nomenclature
WE = Road Works Effects Part
PartHH
D1 Types of roadworks
1 Introduction
The HDM-4 system includes relationships for modelling Road Deterioration (RD) (see Part C)
and Road Works Effects (WE). In HDM-III these were combined into a single module called
the RDME, Watanatada et al. (1987). In HDM-4, RD and WE have been separated in order to
meet the requirements to trigger the extended number and types of road works using a wider
range of criteria based on threshold values of road user effects parameters.
Roadworks modelling in the context of the HDM-4 system imply the following:
! Defining roadworks in a logical manner to achieve a target standard
! Timing of works over the analysis period
! Calculation of the physical quantities or the amounts of works to be undertaken
! Estimating the costs of works which is part of the cash flow analysis used for economic
analysis and budget preparation
! Resetting/changing one or more of the characteristics that define the road as a result of
implementing the roadworks
Thus, the WE module is used to estimate road agency resource needs for road preservation
and development. These needs are expressed in terms of the physical quantities and the
monetary costs of works to be undertaken. Road agency costs are used together with road user
costs and social and environmental costs to determine the economic viability of different road
investment options and strategies.
This chapter describes the classification of road works and the generic framework for
modelling road works effects in HDM-4. The detailed modelling logic for the different types
of works are described separately for each of the three road surface classes (see Figure D1.1)
as follows:
1 Bituminous pavements (see Chapter D2)
Road
RoadWorks
WorksEffects
Effects
Development works aim to expand road network capacity, provide stronger pavement,
and improve road geometric characteristics in order to minimise the total cost of road
transportation and mitigate environmental impacts.
2.2.1 Maintenance
There are three works classes under maintenance:
1 Routine maintenance (see Section 2.3.1)
2.2.2 Development
Development works are divided into two classes as follows:
1 Improvement (see Section 2.3.4)
Comprises works that aim to provide additional capacity when a road is nearing the end
of its design life or because there has been an unforeseen change in use of the road. These
include measures of improving quality of service on existing roads such as relieving
traffic congestion, road safety, road passability, etc.
2 Construction (see Section 2.3.5)
Includes all other works that are not modelled endogenously in HDM-4, for example,
vegetation control, line-marking, road signs repair, guard rail repair, etc.
All routine works are considered under the recurrent budget (by default).
For bituminous pavements this is the addition of a thin film of surfacing to improve
surface integrity and waterproofing that does not increase the strength of the pavement.
This group includes:
❏ fog sealing
❏ rejuvenation
For concrete pavements preventive treatments include:
❏ joint sealing
❏ load transfer dowels retrofit
❏ longitudinal edge drains retrofit
❏ tied concrete shoulders retrofit
Preventive treatments are considered under capital budget (by default).
! Resurfacing or Restoration
The addition of thick surfacings, or the removal of part of the existing pavement and the
addition of layers to restore or improve structural integrity and to increase the strength of
the pavement. For bituminous pavements this group includes:
❏ open-graded asphalt overlay
❏ dense-graded asphalt overlay
❏ rubberised asphalt overlay
❏ mill-and-replace
❏ inlays
Rehabilitation of concrete pavements includes:
❏ bonded concrete overlay
❏ unbonded concrete overlay
Rehabilitation is considered under capital budget (by default).
! Reconstruction
The removal of part (both bound and unbound layers) or all of the existing pavement
layers and the construction of a new pavement. This group includes:
❏ strengthening by multiple-layer overlays thicker than 125 mm
❏ granular overlays
❏ recycling of base
❏ membrane-interlayer overlays
❏ reconstruction of concrete pavements
Reconstruction is considered under capital budget (by default).
! Emergency
Comprises works undertaken to clear a road that has been cut or blocked. This group
includes:
❏ traffic accident removal
❏ clearing debris
❏ repairing washout/subsidence
Emergency works is considered under special budget head (by default).
! Winter
Comprises works undertaken to prevent the formation of ice or to remove snow from the
pavement. This group includes:
❏ salting/gritting
❏ snow removal
Winter works is considered under special budget head (by default).
Note: Winter works are not modelled endogenously in this release.
Full pavement reconstruction with the allowance for minor widening has been included
under improvement works since it is unlikely to be funded from the maintenance budget.
Full reconstruction is considered under capital budget (by default).
! Widening
Comprises works that retain the existing pavement, but increase width throughout the
length of the section. This group includes:
❏ partial widening
❏ lane addition
Widening is considered under capital budget (by default).
! Realignment
Comprises works that changes the road geometry for part of a section, but that retains
some of the existing pavement structure. This group includes:
❏ local geometric improvements
❏ junction improvements
Realignment works is considered under capital budget (by default).
! Off-Carriageway Works
Comprises improvement works that is carried out off-carriageway on road features like
shoulders, NMT lanes, and drainage structures. This group includes:
❏ shoulders addition or upgrading
❏ side drain improvement
Involves changing the pavement surface class and improving geometric characteristics of
an existing road section. This group includes upgrading:
❏ An unsealed road to bituminous or rigid concrete pavement
❏ A bituminous pavement to rigid concrete pavement
❏ A low grade bituminous pavement to a high grade bituminous pavement
❏ An earth road to a gravel road
Pavement upgrading is considered under capital budget (by default).
! New section
Comprises works to create a new pavement in an entirely new location. This group
includes:
❏ dualisation of existing road sections
❏ new link construction
New section works are considered under capital budget (by default).
3 Modelling framework
The above distinction provides a logical framework that enables a detailed modelling of road
works effects on each of the feature types to be performed separately.
2 Improvement/construction standards
For a given road feature, only one maintenance standard and/or improvement type will be
effective in any analysis year.
The operation type plus the following information define a works standard:
! Intervention criteria (see Section 3.3)
A standard may therefore comprise several operations each with definite intervention criteria
to determine when to carry them out. The following sections (Sections 3.3 to 3.7) discuss
briefly the information that is required to define a works standard. A detailed discussion of
standards is given in the Applications Guide.
Limits
In addition to the above criteria, user-specified limits can be defined in order to simulate
realistic policies, Watanatada et al. (1987). For example, it is recognised that practical
considerations would normally preclude the situation of periodic maintenance being applied
immediately before a major overlay or reconstruction was planned. Also due to the economic
implications, different maintenance standards would be applied to a road section at different
thresholds of traffic volume. Provision is therefore made to specify:
! Minimum and maximum intervals between successive works in order to comply with
budget or other constraints
! Maximum applicable roughness beyond which the works activity must not be done
! Minimum and maximum AADT threshold within which a standard is applicable on the
road section
! Last applicable year for the works
! Maximum annual quantity of the works
A works activity will be triggered when any one or a combination of the user-specified criteria
has been met. When more than one works activity meets the criteria for being applied in a
given analysis year, the works hierarchy established for each road surface class and feature
type is used to prioritise and select the highest placed activity. The details of these modelling
logic are discussed for each road surface class in Chapters D2, D3 and D4.
It is assumed that the works are carried out at the end of the analysis year in which it has
been scheduled or triggered, and its effects become effective at the beginning of the
following analysis year. All maintenance works are considered under this category.
Note:
Cay = Variable at the beginning of year y
Road
Cby = Variable at the end of year y
variable
Cb1
Responsive Cbo = Ca1
intervention
Cao
Cb4
Cb2 = Ca3
Ca2
0 1 2 3 4 Years
Scheduled
intervention
Percent
total
costs
100
0 1 2 3 4 Years
Figure D1.2 Modelling of road condition and costs for road works of one year
duration
! For works of duration greater than one year (see Figure D1.3)
Road characteristics are assumed to remain the same from the end of the first year of
works until the end of the year of works completion. The effects of works become
effective at the beginning of the analysis year following the year of works completion.
Note:
Cay = Variable at the beginning of year y
Road Cby = Variable at the end of year y
variable
Responsive
Cb1 = Ca2 Cb2 = Ca3 Cb3
Cbo = Ca1
intervention
Cao
Cb4 = Ca5
Ca4
0 1 2 3 4 Years
Scheduled
intervention
Percent
total
costs
40
30
0 1 2 3 4 Years
Figure D1.3 Modelling of road condition and costs for road works of duration
greater than one year (for example, 3 years)
! Cost per kilometre (for example, upgrading a gravel road to a bituminous surface road)
Unit costs are used to multiply the physical quantities (or amount) of works in order to obtain
the total cost required to implement the operation. The amount of each road works is
computed for each year of the analysis period in which the works activity is applicable.
where:
RACj annual total road agency cost incurred under investment option j (currency)
CAPj annual road agency capital (or periodic) cost incurred under investment option
j (currency)
RECj annual road agency recurrent (or routine) cost incurred under investment
option j (currency)
SPECj annual road agency special cost incurred under investment option j (currency)
For each investment option (or alternative), the annual total agency costs will be reported by
works activities and by budget heads or categories.
The total agency costs are incurred in the analysis year in which the works is scheduled
or triggered by the specified responsive criteria. It is assumed that all maintenance works
fall under this category (see Figure D1.2).
! For works of duration greater than one year
The total agency costs are broken-down into annual cost streams according to user-
specified percentages. The first percentage of the total costs is incurred in the analysis
year in which the works is scheduled or triggered. Subsequent annual costs are incurred
in each of the following years of works (see Figure D1.3).
The annual total costs to road administration are used in economic analyses of different
investment options (see Part G).
SALVA j =
∑∑ SALVA
s =1 w =1
jsw ...(4.2)
where:
In the economic analysis of investment options, the salvage value SALVAj is considered as a
benefit in the last year of analysis period (see Part G).
5 References
Paterson W.D.O., (1995)
Classification of Road Works - ISOHDM Working Paper
Communication to ISOHDM Secretariat, University of Birmingham, UK
Watanatada T., Harral C.G, Paterson W.D.O., Dhareshwar A.M., Bhandari A., and
Tsunokawa K., (1987)
The Highway Design and Maintenance Standards Model - Volume 1 Description
World Bank, John Hopkins University Press
Watanatada T., Harral C.G., Paterson W.D.O., Dhareshwar A.M., Bhandari A., and
Tsunokawa K., (1987)
The Highway Design and Maintenance Standards Model - Volume 2 User's Manual
World Bank, John Hopkins University Press
D2 Bituminous Pavements
1 Introduction
This chapter describes the detailed modelling of Road Works Effects for bituminous
pavements (see Figure D2.1).
Road
RoadWorks
WorksEffects
Effects
The methods of defining works activities and intervention criteria, the calculation of physical
quantities of works and the costs to road administration, and works effects on road
characteristics and road use are discussed for the following works classes:
! Routine maintenance (see Section 3)
The modelling logic described comprises the overall computational procedure, the
hierarchical ranking of works activities and pavement type resets after works. The background
of the modelling logic is given in Watanatada et al. (1987).
A list of research documents referenced from this chapter is given in Section 8.
2 Modelling logic
Inlay 11 per m2
Edge-repair1 22 per m2
Routine
Patching1 22 per m2
Pavement
Crack sealing1 22 per m2
Note:
1
Routine pavement works (that is, crack sealing, patching, edge-repair) have the same
ranking, and all of them can be performed in the same analysis year.
An improvement, or construction works, of a fixed specification is applied to a given road
section only once during the analysis period. This rule applies particularly to improvement
works that have been defined as responsive to the levels of user-specified intervention criteria
based on road user effects parameters.
Routine pavement works, defined by the user, can be applied as separate operations in each
year, or used to repair some distresses before applying the higher-priority works (for example,
preventive treatment, resealing, or overlays). In the former case, routine pavement works are
performed every year in which no periodic maintenance works are applied. In the latter case,
routine pavement works are considered to be an integral part of the periodic maintenance
works, and are referred to as preparatory works. Although preparatory works are
automatically triggered and performed together with the periodic maintenance works, the
amount and cost of each of the operations involved are modelled and reported separately.
Drainage works are applied in any given analysis year, if specified by the user, regardless of
the hierarchy for carriageway works activities given in Table D2.1. Improvement of side
drains takes priority over routine drainage maintenance should both works be applicable in an
analysis year.
The operations that apply to shoulders and non-motorised transport (NMT) lanes are also
performed in any analysis year, if specified by the user, regardless of the works hierarchy
described above. Shoulder or NMT lane improvement works takes priority over shoulders
repair or NMT lane repair, respectively.
For all road feature types, if more than one works activity of the same operation type (for
example, different specifications of overlay) are applicable in an analysis year, the one with
the highest cost takes priority over the others.
The works activities whose effects on pavement performance are not modelled endogenously
(for example, emergency works, winter maintenance, and routine- miscellaneous works) are
applied in a given analysis year, if specified by the user, regardless of any works hierarchy.
Mill & replace **AP **AP **AP **AP N/A **SB **AB **AP
to intermediate
surface layer
Mill & replace **GB **SB **AB **AP **GB **SB **AB **AP
to base
Notes:
1 The pavement type depends on the critical thickness (Hmin) of the existing
bituminous surfacing that is user-definable in HDM Configuration
** Indicates that these two characters are dependent on the specific works activity (or
operation) and the surface material
N/A Not applicable
All overlays, including thin overlay AC, HRA, PMA, RAC, CM, SMA, PA
Mill and replace SBSD, DBSD, CAPE, SL, PM, AC, HRA, PMA, RAC,
CM, SMA, PA
Notes: For the definition and full description of the abbreviated surface materials refer to
Chapter C2.
3 Routine maintenance
The routine maintenance works on bituminous roads whose effects on pavement performance
are modelled, comprises the following operations:
! Patching (see Section 3.1)
Other routine maintenance works (for example, vegetation control, repairs to road
appurtenances, etc.) are considered in the analyses only in terms of their costs to the road
administration.
3.1 Patching
Patching is used to repair the following surface distresses:
! Potholing
! Wide structural cracking
! Ravelling
The user may specify patching to repair the individual surface distresses, or a combination of
all the three distresses above, which is referred to as the severely damaged area (ADAMS). If
more than one kind of patching works is applicable in any analysis year then patching works
specified to treat ADAMS will override those specified to treat the individual distresses.
The user may specify a fixed percentage of the severely damaged area (Pdam) to be
patched and a maximum annual quantity of patching (in square metres per kilometre).
The fixed amount might, for example, reflect the maximum resources available from the
road authority when averaged over all roads within the road class. The amount of
patching performed is computed as the lesser of the specified maximum amount and the
severely damaged area to be repaired.
! Option 2: Condition - Responsive
The user may specify the percentage of the severely damaged area (Pdam) which is to be
patched in each year and impose a limit of the maximum annual quantity of patching.
The intervention criteria for this case will be the level of ADAMS.
The total area patched is given by the following expression:
where:
When patching is performed, the unpatched damaged area is reduced by the amount of
patching as described by the pseudo code given in Equation 3.3 below. It is assumed that
potholing, wide structural cracking, and ravelling have priorities in that order, and no
patching is performed to fix these individual distress-areas until those of higher priorities are
completely repaired.
The percentage repaired of each distress is calculated as follows:
START
If APAT ≥ 10*APOTbw*CW
Ppt = 100
LIMCW = APAT - 10*APOTbw*CW
If LIMRV ≥ 10*ARVbw*CW
Prv = 100
Else
10 * LIMRV
Pr v = (if ARVbw = 0, set Prv = 0)
ARVbw * CW
End if
Else
10 * LIMCW
Pcw = (if ACW bw = 0, set Pcw = 0)
ACWbw * CW
Prv = 0
End if
Else
10 * APAT
Ppt =
APOTbw * CW
Pcw = 0
Prv = 0
End if
END
...(3.3)
where:
The user may specify a fixed percentage of potholing area (Ppt) to be repaired and
impose a limit of the maximum annual quantity of patching.
! Option 2: Condition - Responsive
The user may specify the percentage of potholing area to be patched (Ppt) and impose a
limit of the maximum annual quantity of patching. The intervention criteria to use in this
case will be the level of potholing only.
For both options the total area patched is given by Equation 3.1 above, and the parameter ASP
is calculated as follows:
where:
Ppt percentage of potholing area to be patched, input by the user (default = 100%)
then:
10 * APATlim
Ppt = ...(3.5)
APOTbw * CW
otherwise:
Ppt equals the user-specified value (or the default value = 100%)
The user may specify a fixed percentage of wide structural cracking area (Pcw) to be
patched and impose a limit of the maximum annual quantity of patching.
! Option 2: Condition - Responsive
Under this option, the user may specify the percentage of wide structural cracking area to
be patched (Pcw) and impose a limit of the maximum annual quantity of patching. The
intervention criteria to use in this case will be the level of wide structural cracking only.
For both options; the total area patched is given by Equation 3.1 above, and the parameter
ASP is calculated as follows:
where:
Pcw percentage of wide structural cracking area to be patched, input by the user
(default = 100%)
10 * APATlim
Pcw = ...(3.7)
ACWbw * CW
otherwise:
Pcw equals the user-specified value (or the default value = 100%)
The user may specify a fixed percentage of ravelled area (Prv) to be patched and impose
a limit of the maximum annual quantity of patching.
! Option 2: Condition - Responsive
Under this option, the user may specify the percentage of ravelled area to be patched
(Prv) and impose a limit of the maximum annual quantity of patching. The intervention
criteria to use in this case will be the level of ravelling only.
For both options, the total area patched is given by Equation 3.1 above and the parameter ASP
is calculated as follows:
where:
Prv percentage of ravelled area to be patched, input by the user (default = 100%)
10 * APATlim
Prv = ...(3.9)
ARVbw * CW
otherwise:
Prv equals the user-specified value (or the default value = 100%)
where:
The change in the number of potholes per km due to works (∆NPTw) is given by:
Cracking
where:
ACWaw area of wide structural cracking after patching works (% of total carriageway
area)
ACAaw area of all structural cracking after patching works (% of total carriageway
area)
ACAbw area of all structural cracking before patching works (= ACAb) (% of total
carriageway area)
ACAb area of all structural cracking at the end of the year (% of total carriageway
area)
ACXaw area of indexed cracking after patching works (% of total carriageway area)
ACRAaw total area of cracking after patching works (% of total carriageway area)
ACTbw area of transverse thermal cracking before patching works (= ACTb) (% of
total carriageway area)
ACTb area of transverse thermal cracking at the end of the year (% of total
carriageway area)
∆ACWw reduction in area of wide structural cracking due to patching works
(= ∆ACWpat) (% of total carriageway area)
Ravelling
where:
Roughness
Roughness after patching is calculated as follows:
APAT
RI aw = RIbw - MINa0 * ∆CRX w + ∆RI t - a1 * MIN ,10 , (a2 - RIbw )
(10 * CW )
...(3.18)
and:
where:
∆NPTw * TLF
a2
∆RI t = a 0 (a 1 − FM) NPTa * TLF +
− NPTa
a2
...(3.21)
2
a2
∆NPTw
∆RI t = a 0 (a 1 − FM) * ∆NPTw * NPTa + ...(3.22)
2
where:
By specifying the percentage of the area with transverse thermal cracking (Pcrt) and/or
the percentage of the area with wide structural cracking (Pcrw) to be sealed in each year,
and imposing a limit of the maximum area. The intervention criteria to use in this case
will be the level of transverse thermal cracking and/or the level of wide structural
cracking.
The carriageway area sealed is computed as follows:
where:
ACWbw area of wide structural cracking before crack sealing (% of total carriageway
area)
Pcrt percentage of transverse thermal cracking area to be sealed, input by the user
(%)
Pcrw percentage of wide structural cracking area to be sealed, input by the user (%)
The values of ACWbw, ACAbw, and RIbw to be used in Equations 3.24 above, 3.30 below, 3.31
below and 3.34 below are obtained as follows:
if both crack sealing and patching are specified to be performed in a given analysis year, it
is assumed that patching takes priority over crack sealing in reducing the area of wide
structural cracking, thus:
where:
∆ACWpat reduction in area of wide structural cracking due to patching works (% of total
carriageway area). This is calculated using Equation 3.13 above, with the
subscript w replaced with pat
ACAbw adjusted area of all structural cracking before crack sealing (after patching) (%
of total carriageway area)
RIbw adjusted roughness before crack sealing works (after patching) (IRI m/km)
∆RIpat roughness after patching works (IRI m/km). This is calculated using Equation
3.19 above
otherwise:
ACWbw = ACWb
ACA bw = ACA b
RIbw = RIb
The total area of crack sealing (TACSL), in square metres, is given by the product of ACSL
and the section length (L), in kilometres. The total cost of crack sealing performed on the
entire road section is computed by multiplying TACSL by the user-specified unit cost per
square metre.
Cracking
When crack sealing is performed, it is assumed that the treatment of transverse thermal
cracking takes priority over that of wide structural cracking, and no crack sealing is performed
to fix wide structural cracking until transverse thermal cracking is completely repaired.
The areas of cracking are reduced by the amount of sealing as follows:
ACSL
ACTaw = ACTbw - MIN Pcrt * ACTbw * 10 - 2 , ...(3.28)
(10 * CW )
ACSL
∆ACW w = MAX0, − [ACTbw − ACTaw ] ...(3.29)
(10 * CW )
where:
ACTaw area of transverse thermal cracking after crack sealing works (% of total
carriageway area)
ACTbw area of transverse thermal cracking before crack sealing works (= ACTb) (% of
total carriageway area)
∆ACWw reduction in area of wide structural cracking due to crack sealing (% of total
carriageway area)
ACWaw area of wide structural cracking after crack sealing works (% of total
carriageway area)
ACWbw adjusted area of wide structural cracking before crack sealing (% of total
carriageway area)
ACAbw adjusted area of all structural cracking before crack sealing (% of total
carriageway area)
ACAaw area of all structural cracking after crack sealing works (% of total carriageway
area)
ACXaw area of indexed cracking after crack sealing works (% of total carriageway
area)
ACRAaw total area of cracking after crack sealing works (% of total carriageway area)
Roughness
Crack sealing has some little effect on roughness, which is computed as follows:
where:
3.3 Edge-repair
The user may specify edge-repair in one of the following manners:
! Option 1: Scheduled
By specifying the percentage area of edge-break to be repaired in each year (Pver) and
imposing a limit of the maximum annual quantity of edge-repair. The intervention criteria
to use will be the level of edge-break.
The area of edge-repair is computed as follows:
{ [
AVERP = MIN AVERPlim , Pver * AVEB bw * CW * 10 -1 ]} ...(3.36)
where:
The product of AVERP and the section length (L), in kilometres, gives the total area of edge-
repair (TAVER) in square metres. The total cost of edge-repair performed on the entire road
section is computed by multiplying TAVER by the user-specified unit cost per square metre.
AVERP
AVEB aw = AVEB bw - ...(3.37)
(CW * 10 )
where:
Drainage works is performed when the level of drain deterioration measured by drainage
factor (DF) exceeds the user-specified intervention criteria (see Chapter C2).
The unit costs of drainage works is specified in terms of currency per km per year. The annual
costs are obtained by multiplying the length of the road by the unit cost.
and:
where:
The drainage factor after works is used to compute the adjusted structural number of
pavement (SNP).
4 Periodic maintenance
The periodic maintenance works on bituminous roads comprises the following:
! Preventive treatment (see Section 4.1)
A fixed interval between successive treatments is specified (for example, three years),
and treatment is applied whenever the surfacing preventive treatment age (AGE1)
exceeds this interval.
! Option 2: Condition - Responsive
Preventive treatment is applied at the first signs of cracking or ravelling distress and is
constrained by the user-specified limits of the minimum and maximum allowable
preventive treatment intervals, in years. Preventive treatment is not applied if:
ACRAb ≥ 5,
ARVb ≥ 5, or
NPTa > 0
even if the maximum allowable interval has been exceeded (as may occur in the first
analysis year of an old pavement).
where:
The total area treated (TAPVT), in square metres, is given by the product of APVT and the
section length (L), in kilometres. The total cost of preventive treatment performed is
computed by multiplying TAPVT by the user-specified unit cost per square metre
When preventive treatment is performed, any surface distress (which is minimal) is reset
to zero.
! Pavement strength
Preventive treatment has the effects of delaying the initiation of cracking and ravelling.
These effects are considered in the models through changes in the cracking retardation
time (CRT) and ravelling retardation factor (RRF) as described below:
❏ The cracking retardation time after preventive treatment is reset as follows:
where:
RRFaw = 1 ...(4.4)
where:
The default values for CRM, CRTMAX, RRM and RRFMAX for rejuvenation and fog
sealing are given in Table D2.4 and Table D2.5, respectively.
A fixed interval between successive reseals is specified, and the resealing is applied
whenever the surfacing age (AGE2) exceeds this interval.
! Option 2: Condition - Responsive
Resealing is applied when the levels of pavement defect or roughness exceeds the user-
specified values.
For both options (1 and 2), resealing is not performed if the surfacing age (AGE2) is less than
the user-specified minimum applicable interval. However, resealing is always performed if
AGE2 exceeds the user-specified maximum allowable interval between reseals. In all cases,
resealing is not performed if the user-specified last applicable year or maximum applicable
roughness has been exceeded.
A resealing works is specified using the following:
! New surfacing thickness
! Layer strength coefficient
! Surface material, and
where:
Preparatory works
If the area of wide structural cracking is larger than 20%, or the areas of potholes, transverse
thermal cracking or of edge-break are not zero at the end of the year, preparatory works of
the following amount is assumed to be carried out along with resealing:
! Patching
where:
! Crack Sealing
where:
! Edge-repair
where:
The total area resealed (TARSL), in square metres, is obtained from the product of ARSL and
the section length (L), in kilometres. The total cost of resealing is obtained by multiplying
TARSL by the user-specified unit cost per square metre. The additional areas and costs of
preparatory patching, crack sealing and of edge-repair are reported separately under patching,
crack sealing and edge-repair, respectively.
Pavement type after resealing works is reset as defined in Table D2.2 and Table D2.3
above.
! Surfacing layer thickness
HS aw = HS bw + HSNEWaw ...(4.10)
where:
The thickness of the old, underlying bituminous surfacing after works is given by:
where:
! Pavement strength
To take account of the net strengthening of the pavement due to both maintenance and
cracking, the pavement strength parameters are updated through the following steps:
❏ Calculate the dry season SNP after works as follows:
where:
❏ Calculate the parameter f, defined as the ratio of wet season SNP to dry season
SNP using DFaw and ACRAaw as detailed in Chapter C2
❏ Calculate the parameter fs as specified for the different Road Deterioration models,
see Chapter C2
❏ Calculate the annual average adjusted structural number of pavement (SNP)
The Benkelman beam deflection after works is given by:
−1.6
SNPaw
DEFaw = DEFbw * ...(4.13)
SNPbw
where:
The construction defect indicator for bituminous surfacing (CDS) is reset to a user
specified value. If this is not specified, a good construction quality is assumed, with the
CDS value set to 1.0.
! Pavement surface distresses
Resealing works resets surface distresses to zero, and thereafter the pavement condition
is considered to be new.
! Rutting
where:
! Roughness
The effects of resealing works on roughness are user-specified. If these are not specified,
the roughness is calculated for each works activity in the following manner:
First, the roughness value at the end of the year is adjusted to take account of
preparatory works as follows:
where:
The value of ∆RIt is computed using Equations 3.20 above to 3.22 above with ∆NPTw
taken as equal to NPTbw (= NPTb), since all existing potholes should be patched.
The adjusted roughness after preparatory works (RIap) is then used to compute the final
roughness after resealing works as follows (NDLI, 1995):
❏ Single and double bituminous surface dressing without shape correction
where:
where:
RI aw = RIap - MAX { 0, MIN [a0 * Hsc * RIap , a1 * Hsc * MAX (0, (RIap - a2 ))]}
...(4.22)
where:
Resealing resets texture depth and skid resistance to the user-specified values. If these are
not specified, texture depth after works (TDaw) is reset to the default value of initial
texture depth given in Table D2.6; skid resistance after works (SFCaw) is reset to 0.6 for
all types of reseals.
HRA 0.7
PMA 0.7
AM RAC 0.7
CM 0.7
SMA 0.7
PA 1.5
SBSD 2.5
DBSD 2.5
ST CAPE 0.7
SL 0.7
PM 1.5
! Previous cracking
The area of previous cracking (PCRA and PCRW) is updated to equal the cracking in the
current surfacing before resealing, and a weighting (w) of the cracking in the previous
surfacing, as follows:
❏ if CRAi bw ≥ PCRi bw
where:
The number, per km, of previous transverse thermal cracking (PNCT) is also reset in
the same way as for all and wide structural cracking using Equations 4.23 above,
4.24 above and 4.25 above, except for the following definitions:
PCRiaw number of previous transverse thermal cracking after works (per km)
CRAibw number of transverse thermal cracking before works (per km)
PCRibw number of previous transverse thermal cracking before works (per
km)
! Pavement age
The surfacing age (AGE2) and the preventive treatment age (AGE1) is reset to zero after
resealing works.
4.3 Overlay
Overlay works can be defined in one of the following manners:
! Option 1: Scheduled
The overlay of fixed specifications is applied whenever the rehabilitation age (AGE3)
equals or exceeds a fixed time interval specified by the user.
! Option 2: Condition - Responsive
The overlay of fixed specifications is applied when the levels of the user-specified
intervention criteria are met.
For both options (1 and 2), overlay is not performed if the rehabilitation/overlay age (AGE3)
is less than the user-specified minimum applicable overlay-interval. An overlay is also not
performed if either the preventive treatment age (AGE1) or the surfacing age (AGE2) is less
than the respective minimum preventive treatment or resurfacing intervals. If the minimum
intervals have not been defined within any of the standards that have been assigned to the road
section, the following default values (in years) are used: preventive treatment - 2 years,
resealing - 4 years.
An overlay is always performed when AGE3 exceeds the user-specified maximum allowable
overlay interval. In all cases, an overlay is not performed if the last applicable year or
maximum applicable roughness has been exceeded.
Overlay is specified using the following:
! New surfacing thickness.
! Layer strength coefficient.
! Surface material.
! Construction defect indicator for bituminous surfacing (CDS).
where:
Before overlay is performed, it is often necessary to carry out some preparatory works. It is
assumed that the following amounts of preparatory works is carried out along with overlay:
! Patching
where:
! Edge-repair
The pavement type after overlay is reset as detailed in Table D2.2 and Table D2.3. The
surface material after works is user-specified.
! Surface thickness
The total surface thickness after works is obtained in the following manner:
HS aw = HS bw + HSNEWaw ...(4.28)
where:
The thickness of the old, underlying bituminous surfacing after works is calculated using
Equation 4.11 above.
! Pavement strength
The pavement strength parameters are updated to take account of the net change in
pavement strength due to the new overlay and the underlying cracks (if any), as follows:
where:
❏ The wet season adjusted structural number of pavement (SNPwaw) and the annual
average adjusted structural number of pavement (SNPaw) after works are then
calculated as discussed in Section 4.2.
The Benkelman beam deflection after works is computed using Equation 4.13 above.
! Construction defects indicators
The construction defect indicator for bituminous surfacing CDS is reset to a user
specified value. If these are not specified, a good construction quality is assumed, with
the CDS value set to 1.0.
! Pavement surface distresses
Overlay works resets surface distresses to zero, and thereafter the pavement condition is
considered to be new.
! Rutting
The effects of overlay on rutting are user-specified. If these are not specified, the mean
rut depth is calculated as follows:
where:
! Roughness
The effects of overlay on roughness are user-specified. If these are not specified, the
roughness value after works is calculated in the following manner:
The roughness value at the end of the year is adjusted to take account of preparatory
works as follows:
where:
The value of ∆RIt is computed using Equations 3.20 above to 3.22 above with ∆NPTw
taken as equal to NPTbw (= NPTb), since all existing potholes should be patched.
The adjusted roughness after preparatory works (RIap) is then used to compute the final
roughness after overlay as follows (NDLI, 1995):
where:
After overlay, the texture depth and the skid resistance are reset to the user-specified
values. If these are not specified, texture depth after works (TDaw) is reset to the default
value of initial texture depth given in Table D2.6; skid resistance after works (SFCaw) is
reset to 0.5 for all overlays.
! Previous cracking
The amounts of previous cracking (PCRA, PCRW, and PNCT) are reset using Equations
4.23 above and 4.24 above. The weighting factor (w) of the cracking in the previous
surfacing is computed as follows:
❏ For asphalt roadbase (AB), asphalt pavement (AP) and granular roadbase (GB):
HSNEWbw
w = MAX , 0.6 ...(4.33)
HSOLD aw
HSNEWbw
w = MAX , 0.6 ...(4.34)
(HSOLD aw + HBASE )
where:
w weight used for averaging the cracking in the old and new surfacing
layers
HBASE thickness of the base layer in the original pavement (required only for SB
roadbase types) (mm)
HSOLDaw thickness of old surfacing after works (mm)
HSNEWbw thickness of the most recent surfacing before works (mm)
! Pavement age
The rehabilitation age (AGE3), the surfacing age (AGE2), and the preventive treatment
age (AGE1) are reset to zero after overlay works.
The mill and replace of fixed specifications is applied whenever the rehabilitation age
(AGE3) equals or exceeds a fixed time interval specified by the user.
! Option 2: Condition - Responsive
The mill and replace of fixed specifications is applied when the levels of the user-
specified intervention criteria are met.
For both options (1 and 2), a mill and replace works is not performed if the rehabilitation age
(AGE3) is less than the user-specified minimum applicable interval, or if the last applicable
year has been exceeded. A mill and replace works is always performed when AGE3 exceeds
the user-specified maximum allowable interval.
The following information is required to specify mill and replace works:
! New surfacing thickness
! Layer strength coefficient
! Surface material
! Depth of milling
! Construction defect indicator for bituminous surfacing
where:
The total amount of mill and replace works (TAMR), in square metres, is obtained from the
product of AMR and the section length (L), in kilometres. The total cost of mill and replace
works is obtained by multiplying TAMR by the user-specified unit cost per square metre.
After mill and replace, the pavement type is reset as detailed in Table D2.2 and Table
D2.3. The surface material after works is user-specified.
The following conditions determine which kind of mill and replace activity is being
applied:
if MILLD ≥ HS bw then it is a mill and replace to roadbase
! Surface thickness
where:
The thickness of the old, underlying bituminous surfacing after works is given by:
where:
! Pavement strength
The pavement strength parameters are updated to take account of the net change in
pavement strength due to the new bituminous surfacing and the depth of milling as
follows:
❏ If MILLD ≤ HSNEWbw , then the dry season adjusted structural number of
pavement is given by:
...(4.38)
❏ If HSNEWbw < MILLD ≤ HS bw , then the dry season adjusted structural number
of pavement is calculated as follows:
...(4.39)
❏ If MILLD > HS bw , then the dry season adjusted structural number of pavement
is given by:
where:
The wet season adjusted structural number of pavement (SNPwaw) and the annual average
adjusted structural number of pavement (SNPaw) after works are calculated using the
same procedure as discussed in Section 4.2.
The Benkelman beam deflection after works is calculated using Equation 4.13 above.
The construction defect indicator for bituminous surfacing (CDS) is reset to a user
specified value. If these are not specified, a good construction quality is assumed, with
the CDS value set to 1.0.
Mill and replace works resets surface distresses to zero, and thereafter it is assumed that
the pavement behaves as if new.
! Rutting
Mill and replace resets rutting to zero by default, unless the user specifies a different
value.
! Roughness
The effects of mill and replace on roughness are user-specified. If these are not specified,
the following default values are used:
For AM surface type, RIaw = 2.0 (IRI m/km)
For ST surface type, RIaw = 2.8 (IRI m/km)
! Texture depth and skid resistance
After mill and replace, the texture depth and the skid resistance are reset to the user-
specified values. If these are not specified, texture depth after works (TDaw) is reset to the
default value of initial texture depth given in Table D2.6; skid resistance after works
(SFCaw) is reset to the following default values:
For surface type AM, SFCaw = 0.5
For surface type ST, SFCaw = 0.6
! Previous cracking
The amounts of previous cracking (PCRA, PCRW, and PNCT) are reset as follows:
If MILLD < HSNEWbw , then the amounts of previous cracking are reset as follows:
if CRAi bw ≥ PCRi bw
MILLD
wf = w * 1 - ...(4.43)
HSNEWbw
The weighting (w) of the cracking in the previous surfacing is computed as follows:
❏ For asphalt base (AB), asphalt pavement (AP) and granular roadbase (GB)
HSNEWbw
w = MAX , 0.6 ...(4.44)
HSOLD aw
HSNEWbw
w = MAX , 0.6 ...(4.45)
(HSOLD aw + HBASE )
if MILLD ≥ HSNEWbw
then:
HS bw − MILLD
wg = MAX 0, ...(4.47)
HSOLD bw
Equation 4.47 above assumes that if MILLD = HSbw any cracking in the roadbase is
rectified before surfacing.
All the parameters used in Equations 4.41 above to 4.47 above are as defined previously
under resealing and overlay.
! Pavement age
The rehabilitation/overlay age (AGE3), the surfacing age (AGE2), and the preventive
treatment age (AGE1) is reset to zero after a mill and replace works.
4.5 Inlays
Inlay is a special works activity, considered under Rehabilitation, that is normally applied to
treat rutting along wheelpaths (and this involves some milling of the existing pavement
layers). It is therefore not scheduled, but defined as a condition-responsive works, in which
an inlay of fixed specifications is applied when the level of pavement defect exceeds the user-
specified values.
Inlay is not performed if the user-specified last applicable year or the maximum applicable
roughness has been exceeded.
To define an inlay works, the percentage of total carriageway area to be repaired and the
construction quality factors should be specified.
where:
It is likely that patching, crack sealing and edge-repair will be carried out along with inlay
works. The modelling logic assumes therefore that first inlay is performed, and then patching
and crack sealing repairs the remaining areas of potholes and cracking. The amounts of
patching, crack sealing and edge-repair that may be performed are computed as follows:
! Patching
Pin
∆ACWpat = MAX 0.1 * (ACWb − 20 ) * 1 − , 0 ...(4.50)
100
where:
! Crack sealing
It is assumed that crack sealing repairs the entire remaining area of transverse thermal
cracking and wide structural cracking as follows:
Pin
ACSL = 10 * CW * Pacw * ACWbcs + Pact * ACTb * 1 - ...(4.51)
100
where:
! Edge-repair
where:
The wet season adjusted structural number of pavement (SNPwaw) and the annual average
adjusted structural number of pavement (SNPaw) after works is then calculated as discussed in
Section 4.2 above for resealing works.
The Benkelman beam deflection after works is computed using Equation 4.13 above.
! Construction defects indicators
The construction defect indicator for bituminous surfacing (CDS) is reset to a user
specified value. If these are not specified, a good construction quality is assumed, with
a CDS value of 1.0.
! Pavement surface distresses
Pin
∆ACWinlay = ACWb * 1 - ...(4.55)
100
where the parameters Pacw and Pact are as defined in Equation 4.51 above
The amounts of indexed cracking (ACXaw) and total area of cracking (ACRAaw) are
reset after inlay works using Equations 3.32 above and 3.33 above, respectively.
❏ Ravelling is reset as follows:
where:
❏ Edge-break is reset to zero (that is, AVEBaw = 0), due to the edge-repair works that
is assumed to be performed during inlay works.
! Rutting
The effects of inlay on rutting are user-specified. If these are not specified, the mean rut
depth is calculated as follows:
where:
! Roughness
Inlay reduces roughness by treating rutting, all the potholes, transverse thermal cracking
and wide structural cracking. The effect of inlay on roughness is user-specified. If these
are not specified, the values of roughness after works is computed as follows:
where:
The value of ∆RIt is computed using Equations 3.20 above to 3.22 above with ∆NPTw
taken as equal to NPTbw (= NPTb), since all existing potholes should be patched.
! Texture depth and skid resistance
Inlay resets texture depth and skid resistance in the same way as described for mill and
replace in Section 4.3.2 above.
! Previous cracking
The amounts of previous cracking (PCRA, PCRW, and PNCT) is reset as follows:
PCRi aw = 1 -
Pin
* PCRi bw ...(4.62)
100
where:
! Pavement age
AGEi aw = 1 -
Pin
* AGEi bw ...(4.63)
100
where:
AGEiaw AGE type i (i = 1, 2 or 3) after inlay works (years). (This value should be
adjusted to an integer)
AGEibw AGE type i before inlay works (years)
4.6 Reconstruction
Pavement reconstruction refers to all works that require the re-specification of the surfacing
and roadbase types.
Reconstruction may be specified either as a maintenance standard or as an improvement
standard if the works involve minor widening of the carriageway.
Note: the latter is permitted only for the sake of convenience in the modelling logic as
discussed in Section 6.1.
Reconstruction may be specified in one of the following ways:
! Option 1: Scheduled
A reconstruction works of fixed specifications is applied when the levels of the user-
specified intervention criteria, based on pavement condition and traffic loading, are met.
Reconstruction is not performed if the last applicable year has been exceeded.
For both options (1 and 2), reconstruction is not performed if:
AGE4 is less than the minimum reconstruction interval
AGE1, AGE2 and AGE3 are less than their respective minimum intervals
It is a construction-opening year
! Surface material
! Surfacing thickness
! Structural number of pavement (SN) of the layers above the subgrade
! Relative compaction
! Construction defect indicators
For stabilised roadbases, the base thickness and the resilient modulus are also required.
where:
where:
The total cost of reconstructing the entire road section is obtained from the product of
TARCON and the unit cost per square metre, or from the product of the section length (L) and
the unit cost per kilometre.
where:
where:
! Construction quality
The construction defect indicators for bituminous surfacing (CDS), and the construction
defect indicator for the roadbase, (CDB), are reset to user specified values. If these are
not specified, a good construction quality is assumed, with the CDS value set to 1.0 and
the CDB value set to 0.
! Pavement surface distresses
Surface distresses (that is, edge-break, potholing, cracking and ravelling) are all reset to
zero.
! Rutting
Reconstruction resets texture depth and skid resistance in the same way as described for
mill and replace in Section 4.4.2.
! Previous cracking
The area of previous cracking (PCRA, PCRW, and PNCT) is reset to zero.
The cracking retardation time, ravelling retardation factor and the other road deterioration
factors are also reset to user-specified values.
! Pavement age
The pavement ages: AGE1, AGE2, AGE3, and AGE4 are all reset to zero.
5 Special works
The effects of the following special works on pavement performance are not modelled
endogenously, and therefore only their costs can be considered in an analysis:
! Emergency works
These works types are scheduled at a fixed interval of time (minimum of one-year), and are
performed on an annual basis. If specified, the works activities are applied in a given analysis
year regardless of the works hierarchy given in Table D2.1.
Their unit costs are specified in terms of currency per kilometre per year, and the annual costs
are obtained by multiplying the road section length (L) by the unit cost.
6 Improvement works
Road improvement works aim to provide additional capacity and comprises the following:
! Partial widening
! Lane addition
! Realignment
! Non-carriageway improvement works
For example, improvement or addition of shoulders, NMT lanes, and side drains.
Although it is not an improvement works, reconstruction that involves minor widening of the
carriageway is modelled using the logic framework for improvement works.
where:
where:
The total cost (CSTCON) of reconstructing the entire road section is obtained from the
product of TARCON and the unit cost per square metre, or from the product of the section
length and the unit cost per kilometre.
The salvage value is given by:
where:
where:
The effective number of lanes after works is reset to a new value ELANESaw, which should be
input by the user. If this is not specified ELANESaw will be reset to the value of NLANES
(that is, the number of lanes for the road section) see Part B Section 5.2.
6.2 Widening
There are two operations included under widening, lane addition and partial widening. The
major distinction is that partial widening does not increase the number of lanes. Both widening
operations do not alter the road alignment.
Widening works can be specified in one of the following ways:
! Option 1: Scheduled
A widening works of fixed specifications is applied when the levels of the user-specified
intervention criteria, based on road user effects parameters, are met. Widening is not
performed if the last applicable year has been exceeded.
The following information is required to specify a widening works:
! New road type
! Road class
! Increase in carriageway width for partial widening
! Additional number of lanes and increase in carriageway width for lane addition
! Pavement type for the entire section
! Pavement details of the widened area of the carriageway
! Construction quality indicators
! Whether or not the existing carriageway is to be provided with a new surfacing
where:
The total area of widening over the entire section is given by the expression:
where:
TAWDN total widened area of carriageway for the entire section (m2)
L road section length (km)
The cost of widening (CSTWDN) is obtained from the product of TAWDN and the unit cost
per square metre, or from the product of the section length and the unit cost per kilometre.
The salvage value is given by:
where:
Additional works
It is likely that widening works will include re-surfacing or at least repairs of the severely
damaged area of the existing carriageway. The additional works that may be required are
modelled as described below. (Note that re-surfacing or re-surfaced has been used to denote
the provision of a new bituminous surfacing and to distinguish it from resurfacing as a type
of roadwork).
Case1: re-surfacing the existing carriageway
If the existing carriageway is to be re-surfaced either by an overlay or resealing (as
determined from the user-specified new pavement type after works), the amount of re-
surfacing works for the entire road section is given by:
where:
TANSF would be denoted as TAOVL, and the total cost of overlay is obtained from the
product of TAOVL and the unit cost of overlay per square metre.
Before overlay is performed, it is often necessary to carry out some preparatory works.
The amount of preparatory works to be performed for the entire road section is
computed as follows:
❏ Patching
where:
The total cost of preparatory patching is obtained from the product of TAPAT and
the unit cost of patching per square metre.
❏ Edge-repair
where:
The total cost of edge-repair is obtained from the product of TAVER and the unit
cost of edge-repair per square metre.
! The existing carriageway is resealed
ANSF would be denoted by TARSF, and the total cost of resealing is obtained from the
product of TARSF and the unit cost of resealing per square metre.
Before resealing is applied, the amount of preparatory works that may be required for
the entire road section is computed as follows:
❏ Patching
where:
It is assumed that 100% of transverse thermal cracking is sealed, and the amount of
crack sealing performed is given by:
where:
The total cost of crack sealing is obtained from the product of TACSL and the unit
cost of crack sealing per square metre.
❏ Edge-repair
The total area of edge-repair is calculated using Equation 6.10 above. The total cost
of edge-repair is obtained from the product of TAVER and the unit cost of edge-
repair per square metre.
Assuming that 100% of severely damaged area is patched, the amount of patching is
computed as follows:
where:
! Crack sealing
The total amount of crack sealing is calculated using Equation 6.12 above.
! Edge-repair
Assuming that all the edge-break area is repaired, the amount of edge-repair works is
given by Equation 6.10 above.
The total cost of widening works is the sum of carriageway widening cost and the cost of
additional works comprising re-surfacing of the existing carriageway and preparatory
works. The amount and cost of re-surfacing works are reported separately under resealing or
overlay. Similarly, the amounts and costs of patching, of crack sealing and edge-repair are
reported separately under patching, crack sealing and edge-repair, respectively.
In economic analysis, it is assumed that the cost of all these additional works is incurred in the
last construction year.
where:
For partial widening, the increase in carriageway width (∆CW) is specified directly by
the user. For lane addition, the increase in carriageway width is user-specified, if
this is not specified the increase will be given by:
ADDLN * CWbw
∆CW = ...(6.15)
NLANES bw
where:
For lane addition, the number of lanes after widening works (NLANES aw) is equal
to the number of lanes before works (NLANESbw) plus the user-specified additional
number of lanes (ADDLN).
The effective number of lanes after works is reset to a new value ELANESaw, which
should be input by the user. If this is not specified, ELANESaw will be reset to the value of
NLANESaw (that is, the number of lanes for the road section following widening) see Part
B Section 5.2.
! Thickness of surfacing layers
❏ Case 1: re-surfacing the existing carriageway
where:
(CWbw * HS bw )
HSOLD aw = ...(6.17)
CWaw
where:
where:
(CWbw * HSOLD bw )
HSOLD aw = ...(6.19)
CWaw
where:
! Pavement strength
The dry season adjusted structural number of pavement is reset to the weighted
average of that of the existing carriageway and that of the widened part of carriageway,
as follows:
(CWbw + ∆CW )
0.2
SNPdaw = SNPdexcw * SNPdww *
(CWbw * [SNPdww ] 5 + ∆CW * [SNPdexcw ] 5 )
...(6.20)
where:
SNPdaw dry season adjusted structural number of pavement after widening works
SNPdbw dry season adjusted structural number of pavement before widening
works
SNPdww dry season adjusted structural number of pavement of the widened part of
the carriageway
SNdww user-specified structural number of pavement (of the layers above the
subgrade) of the widened part of the carriageway. This may be computed
in a similar manner to that described for SNnew in Equation 4.66 above
SNSGs subgrade contribution to the structural number of pavement, calculated
using Equation 4.67 above
dSNPK reduction in the adjusted structural number of pavement due to cracking
aresf strength coefficient of the re-surfacing layer on the existing carriageway
where:
! Surface material
If the existing carriageway is to be re-surfaced, the surface material after works is reset to
that specified for the widening works. This is based on the assumption that the same
surfacing material is used for the re-surfacing.
If the existing carriageway is not to be re-surfaced, the surface material after works is
reset as follows:
If CWbw is greater than ∆CW, the surface material after widening works is
reset to that of the existing carriageway.
Otherwise the surface material after widening works is reset to that of the widened
part of the carriageway.
! Construction quality
The construction defect indicators for bituminous surfacing (CDS) and the construction
defect indicator for the roadbase (CDB) are reset to a weighted average computed in the
following manner:
where:
If the existing carriageway is to be re-surfaced, the amounts of all surface distresses after
widening works are reset to zero.
If the existing carriageway is not to be re-surfaced, the areas of edge-break, potholing,
transverse thermal cracking, wide structural cracking and ravelling after widening works
are all reset to zero. The area of all structural cracking and the total area of cracking are
calculated as follows:
where:
! Rutting
The mean rut depth is reset to a user-specified value. If this is not specified, the mean rut
depth is calculated as follows:
(CWbw * a0 * RDMbw )
RDM aw = ...(6.28)
CWaw
where:
! Roughness
After widening works, roughness is reset to a user-specified value. If this is not specified,
the value of roughness is obtained as follows:
❏ If the existing carriageway is to be re-surfaced the following values are used as
defaults:
For AM surface type: RIaw = 2.0 (IRI m/km)
For ST surface type: RIaw = 2.8 (IRI m/km)
❏ If the existing carriageway is not to be re-surfaced, it is assumed that patching and
crack sealing that may be performed on the existing carriageway would affect the
roughness after widening works as follows:
where:
RIaw roughness after widening works (IRI m/km)
RIbw roughness before widening works (= RIb) (IRI m/km)
RIn user-specified roughness for new construction, (default = 2.0 for AM, and
2.8 for ST)
RIap roughness after patching and crack sealing (IRI m/km)
∆CRXw reduction in the area of indexed structural cracking, (= ACXb - ACXaw)
The value of ∆RIt is computed using Equations 3.20 above to 3.22 above by taking
∆NPTw as equal to NPTbw (= NPTb), since all existing potholes should be patched.
! Texture depth and skid resistance
After widening, texture depth and skid resistance is reset to user-specified values. If these
are not specified, the values of texture depth and skid resistance after works are obtained
in the following manner:
❏ The existing carriageway is to be re-surfaced
Texture depth after works (TDaw) is reset to the default value of initial texture depth
given in Table D2.6; and skid resistance after works (SFCaw) is reset to the following
default values:
For surface type AM: SFCaw = 0.5
For surface type ST: SFCaw = 0.6
where:
! Previous cracking
(CWbw * CRAi bw )
PCRi aw = ...(6.33)
CWaw
HSNEWbw
w = MAX , 0.6 ...(6.35)
HSOLD aw
HSNEWbw
w = MAX , 0.6 ...(6.36)
(HSOLD aw + HSBASE )
where:
(CWbw * PCRi bw )
PCRi aw = ...(6.38)
CWaw
(CWbw * AGE4 bw )
AGE4 aw = (returns an integer value) ...(6.39)
CWaw
(CWbw * AGEi bw )
AGEi aw = (for i = 3 or 4) ...(6.40)
CWaw
❏ If the existing carriageway is not to be re-surfaced, the pavement ages are calculated
as follows:
(CWbw * AGEi bw )
AGEi aw = ...(6.41)
CWaw
where:
The cracking retardation time, ravelling retardation factor and the other road deterioration
factors (that is, the K factors) are also reset to user-specified values.
! Speed factors
These are the speed limit, speed enforcement factor, roadside friction factor, non-
motorised transport and motorised transport speed reduction factors and acceleration
noise, which depend primarily on the individual road section.
! Traffic flow pattern (road use)
The data that describes the hourly traffic flow distribution are also reset to a user-
specified type.
6.3 Realignment
Refers to local geometric improvements of road sections, which may also result in a reduction
of the road length. It is assumed that the carriageway width remains unaltered when a
realignment works is performed.
Realignment works can be defined as follows:
! Option 1: Scheduled
A realignment works of fixed specifications is applied when the levels of the user-
specified intervention criteria, based on road user effects parameters, are met.
Realignment is not performed if the last applicable year has been exceeded.
A realignment works is specified using the following:
! New road type
! Road class
! Proportion of new construction - defined as the ratio of new construction length to the
section length after realignment works
! Length adjustment factor
! Road geometry for the whole section
! Pavement details of the new realigned segments
! Construction quality indicators
! Whether or not the non-realigned segments (or parts) of the existing carriageway is
provided with a new surfacing (that is, re-surfaced)
L aw = L bw * LF ...(6.43)
where:
The cost of realignment (CSTREAL) is obtained from the product of REAL and the user-
specified unit cost per kilometre.
The salvage value is given by:
where:
Additional works
It is likely that realignment works will include re-surfacing or at least repair of the severely
damaged area of the existing carriageway segments that are not to be realigned. These
additional works are modelled as described below.
Case 1: re-surfacing the non-realigned segments
If the non-realigned segments of the existing carriageway are to be re-surfaced either by an
overlay or resealing, the amount of re-surfacing works for the entire road section is given by:
where:
! If an overlay were provided over the non-realigned parts of the existing carriageway,
TANSF would be denoted as TAOVL, and the total cost of overlay is obtained from the
product of TAOVL and the unit cost of overlay per square metre.
Before overlay is performed, the amounts of preparatory works to be performed for the
entire road section are computed as follows:
❏ Patching
where:
The total cost of preparatory patching is obtained from the product of TAPAT and
the unit cost of patching.
❏ Edge-repair
where:
The total cost of edge-repair is obtained from the product of TAVER and the unit
cost of edge-repair.
! If the non-realigned parts of the existing carriageway were resealed, ANSF would be
denoted by TARSF, and the total cost of resealing is obtained from the product of
TARSF and the unit cost of resealing per square metre.
Before resealing is applied, the amounts of preparatory works to be performed for the
entire road section are computed as follows:
❏ Patching
...(6.48)
where:
The total cost of preparatory patching is obtained from the product of TAPAT and
the unit cost of patching per square metre.
❏ Crack sealing
where:
The total cost of crack sealing is obtained from the product of TACSL and the unit
cost of crack sealing per square metre.
❏ Edge-repair
The total area of edge-repair is calculated using Equation 6.47 above. The total cost
of edge-repair is obtained from the product of TAVER and the unit cost of edge-
repair per square metre.
Case 2: no re-surfacing of the non-realigned segments
If the non-realigned parts of the existing carriageway are not to be re-surfaced, it is assumed
that the following amounts of patching, crack sealing and edge-repair are performed along
with the realignment works:
! Patching
Assuming that 100% of severely damaged area is patched, the amount of patching is
computed as follows:
where:
! Crack sealing
The total amount of crack sealing is calculated using Equation 6.49 above.
! Edge-repair
Assuming that all the edge-break area is repaired, the amount of edge-repair works is
given by Equation 6.47 above.
The total cost of realignment works is the sum of carriageway realignment cost and the cost of
additional works comprising re-surfacing of the existing carriageway and preparatory
works. The amount and cost of re-surfacing works are reported separately under
resealing or overlay. Similarly, the amounts and costs of patching, of crack sealing and edge-
repair are reported separately under patching, crack sealing and edge-repair,
respectively.
In economic analysis, it is assumed that the costs of all these additional works are incurred in
the last construction year.
The new length of the road section after realignment is given by Equation 6.43 above.
! Thickness of surfacing layers
❏ Case 1: re-surfacing non-realigned segments
where:
where:
If the non-realigned parts of the existing carriageway are not to be re-surfaced, the
thickness of new surfacing after realignment works is obtained as follows:
where:
where:
! Pavement strength
The dry season adjusted structural number of pavement is reset to the weighted
average of the structural number of the non-realigned parts of the existing carriageway
and that of the newly constructed segments, as follows:
where:
−1.6
SNPaw
DEFaw = DEFbw * ...(6.59)
SNPbw
where:
! Surface material
If the existing carriageway is to be re-surfaced, the surface material after works is reset to
that specified for the realignment works. This is based on the assumption that the same
surfacing material is used for the re-surfacing.
If the existing carriageway is not to be re-surfaced, the surface material after works is
reset as follows:
if Pconew is less than 0.5, the surface material after works is reset to that of
the existing carriageway.
otherwise the surface material after works is reset to that of the realigned parts of the
carriageway.
! Construction quality
The construction defect indicators for bituminous surfacing (CDS) and the construction
defect indicator for the roadbase (CDB) is reset to a weighted average computed as
follows:
where:
If the non-realigned parts of the existing carriageway are to be re-surfaced the surface
distresses (that is, edge-break, potholing, cracking and ravelling) are all reset to zero.
If the non-realigned parts of the existing carriageway are not to be re-surfaced, the area
of edge-break, potholing, transverse thermal cracking, wide structural cracking and
ravelling are reset to zero. The area of all structural cracking and total area of cracking
after realignment works are reset as follows:
where:
The mean rut depth is reset to a user-specified value. If this is not specified, the mean rut
depth is calculated as follows:
where:
! Roughness
where:
RIaw roughness after realignment works (IRI m/km)
RIbw roughness before realignment works (= RIb) (IRI m/km)
RIn user-specified roughness for realignment new construction,
(default = 2.0 for AM, and 2.8 for ST)
RIap roughness after patching and crack sealing (IRI m/km)
The value of ∆RIt is computed using Equations 3.20 above to 3.22 above by taking
∆NPTw as equal to NPTbw (= NPTb), since all existing potholes should be patched.
! Texture depth and skid resistance
After realignment, texture depth and skid resistance is reset to user-specified values. If
these are not specified, texture depth and skid resistance after works are obtained in the
following ways:
❏ If the non-realigned segments of the existing carriageway are to be re-surfaced,
texture depth after works (TDaw) is reset to the default value of initial texture depth
given in Table D2.6; and skid resistance after works (SFCaw) is reset to the following
default values:
For surface type AM: SFCaw = 0.5
For surface type ST: SFCaw = 0.6
❏ If the non-realigned segments of the existing carriageway are not to be re-surfaced,
the values of texture depth and skid resistance after works are computed as follows:
where:
! Previous cracking
If the non-realigned parts of the existing carriageway are to be re-surfaced, the amount of
previous cracking (PCRA, PCRW, and PNCT) is reset as follows:
❏ if CRAi b ≥ PCRi bw
HSNEWbw
w = MAX , 0.6 ...(6.70)
HSOLD aw
HSNEWbw
w = MAX , 0.6 ...(6.71)
(HSOLD aw + HBASE )
If the re-surfacing is a reseal (that is, surface type ST), the value of w is given by:
where:
If the non-realigned parts of the existing carriageway are not to be re-surfaced, the
amount of previous cracking (PCRA, PCRW, and PNCT) is reset as follows:
❏ If the non-realigned parts of the existing carriageway are not to be re-surfaced, the
pavement ages are calculated as follows:
where:
! Deterioration factors
The cracking retardation time, ravelling retardation factor and the other road deterioration
factors (that is, the K factors) are also reset to user-specified values.
! Speed factors
These are the speed limit, speed enforcement factor, roadside friction factor, non-
motorised transport and motorised transport speed reduction factors, and acceleration
noise which depend primarily on the individual road section.
! Traffic flow pattern (road use)
The data that describes the hourly traffic flow distribution are also reset to a user-
specified type.
A junction improvement works is specified by the new intersection type (to which a different
set of accident rates apply), the duration of works and the cost involved in terms of currency
per year.
6.5.1 Shoulders
The modelling of shoulders improvement works is not included in this release. It will be
implemented when the specifications are finalised. The improvement works for shoulders
includes the following:
! Shoulders addition
! Shoulders upgrading
An NMT lane upgrading works of fixed specifications is performed when the user-
specified intervention criteria, based on road user effects parameters, are met. Upgrading
is not performed if the last applicable year has been exceeded.
The following information is required to specify upgrading NMT lanes:
! New road type
! New surface type
It is assumed that the geometry details are the same as that of the existing road section.
If performed, the total amount of works is given in terms of the total length of NMT lanes
constructed as follows:
where:
The total cost of NMT lane upgrading is obtained from the product of NMTLaw and the user-
specified unit cost per lane per kilometre.
The effects of upgrading NMT lane is reflected by specifying new values for the speed
reduction factors, and resetting the data that describe the resulting traffic flow pattern.
An NMT lane(s) of fixed specifications is constructed when the levels of the user-
specified intervention criteria, based on road user effects parameters, are met.
Construction is not performed if the last applicable year has been exceeded.
The addition of new NMT lanes is specified by the number of lanes, the road type and the
pavement surface type. It is assumed that the geometry details are the same as that of the road
section.
If performed, the total amount of works is given in terms of the total length of NMT lanes
constructed as follows:
where:
The total cost of NMT lane construction is obtained from the product of NMTLaw and the
user-specified unit cost per lane per kilometre.
The effects of NMT lane addition should be reflected by specifying new values for the speed
reduction factors, and resetting the data that describe the resulting traffic flow pattern.
The total cost of side drain improvement is obtained from the product of road section length
and the unit cost given in terms of currency per kilometre per year.
7 Construction
Construction works comprises the following:
! Upgrading (see Section 7.1)
7.1 Upgrading
This operation involves both pavement upgrading and geometric improvements of an existing
road. Normally pavement upgrading would change the existing surface class to another
surface class of a higher performance grade. For example, a bituminous pavement road may
be upgraded to a rigid concrete pavement road. The modelling framework also allows for the
upgrading of a lower-grade bituminous pavement to a higher-grade bituminous pavement.
Upgrading works can be specified in one of the following ways:
! Option 1: Scheduled
An upgrading works of fixed specifications is applied when the levels of the user-
specified intervention criteria, based on road user effects parameters, are met. Upgrading
is not performed if the last applicable year has been exceeded.
An upgrading works is specified by the:
! Road type
! Road class
! Road geometric characteristics
! Pavement details
! Length adjustment factor
! Increase in width
! Additional number of lanes
! Construction quality indicators
LUPGRD = L aw ...(7.1)
L aw = LF * L bw ...(7.2)
where:
The total cost of upgrading (CSTUPGRD) is obtained from the product of LUPGRD and the
user-specified unit cost per kilometre.
The salvage value is given by:
where:
The new length of the road section after upgrading is given by Equation 7.2 above.
! Carriageway width
where:
The number of lanes after upgrading (NLANESaw) is equal to the number of lanes before
works (NLANESbw) plus the user-specified additional number of lanes (ADDLN).
The effective number of lanes after works is reset to a new value ELANESaw, which
should be input by the user. If this is not specified, ELANESaw will be reset to the value of
NLANESaw (that is, the number of lanes for the road section following upgrading) see
Part B Section 5.2.
! Pavement characteristics
The new pavement characteristics are reset according to the new road surface class
specified by the user as described in Sections 7.1.3 and 7.1.4.
The dry season adjusted structural number of pavement (SNPdaw) is computed using the
specified structural number of pavement (SNnew) from Equations 4.66 above and 4.67
above. The Benkelman beam deflection after upgrading works will be given as:
−1.6
DEFaw = a0 * (SNPaw ) ...(7.5)
where:
After upgrading, the pavement condition is new, and all distress values are therefore set
to zero.
! Rutting
Roughness after upgrading is set to a user-specified value. If this is not specified, the
following default values are used:
For AM surface type: RIaw = 2.0 (IRI m/km)
For ST surface type: RIaw = 2.8 (IRI m/km)
! Texture depth and skid resistance
After upgrading, texture depth and skid resistance is reset to the user-specified values. If
these are not specified, texture depth after works (TDaw) is reset to the default value of
initial texture depth given in Table D2.6; and skid resistance after works (SFCaw) is reset
to the following default values:
For surface type AM: SFCaw = 0.5
For surface type ST: SFCaw = 0.6
! Pavement ages
After upgrading, all the pavement age parameters (that is, AGE1, AGE2, AGE3 and
AGE4) are reset to zero.
! Previous cracking
The deterioration factors for modelling the performance of the new pavement are reset to
user-specified values.
! Speed factors
These are the speed limit, speed enforcement factor, acceleration noise, roadside friction
factor, non-motorised transport factor, and the speed reduction factor due to motorised
transport which depend primarily on the individual road section.
! Traffic flow pattern (road use)
The data that describes the hourly traffic flow distribution are also reset to a user-
specified type.
7.2 Dualisation
The dualisation of an existing road section can be either scheduled or triggered by responsive
intervention criteria.
Note: The modelling of dualisation has not been included in this release.
construction (CSTNEW) is obtained from the product of NEWCON and the user-specified
unit cost per kilometre.
The salvage value is given as:
where:
8 References
NDLI, (1995)
Modelling Road Deterioration and Maintenance Effects in HDM-4. Final Report Asian
Development Bank Project RETA 5549. N.D. Lea International, Vancouver, Canada
Riley M., (1995)
Framework for Defining the Effects of Works on Road parameters in HDM-4.
Communication to the ISOHDM
Watanatada T., Harral C.G., Paterson W.D.O., Dhareshwar A.M., Bhandari A., and
Tsunokawa K., (1987a)
The Highway Design and Maintenance Standards Model: Volume-1 Description. The
World Bank, John Hopkins University Press
Watanatada T., Harral C.G., Paterson W.D.O., Dhareshwar A.M., Bhandari A., and
Tsunokawa K., (1987b)
The Highway Design and Maintenance Standards Model: Volume- 2 User's Manual
World Bank, John Hopkins University Press
D3 Concrete Pavements
1 Introduction
This chapter describes the detailed modelling of road works effects for rigid concrete roads in
HDM-4. It is based on the specification documents prepared by the Latin American Study
Team in Chile (LAST, 1995; and LAST, 1996).
The methods of defining works activities and intervention criteria, the calculation of physical
quantities of works and the costs to road administration, and works effects on road
characteristics and road use are discussed for the following works classes:
! Routine maintenance
! Periodic maintenance
! Special works
! Construction works
The modelling logic described comprises the overall computational procedure, the
hierarchical ranking of works activities and pavement type resets after works. The background
of the modelling logic is given in LAST (1996).
Table D3.1 shows the maintenance works activities for the different pavement surface types
considered in HDM-4.
Table D3.1 Maintenance works for concrete pavements
Notes:
a indicates that the works activity is applicable to the pavement surface type
JP Jointed plain
JR Jointed reinforced
CR Continuously reinforced
2 Modelling logic
The modelling of concrete pavement performance is considered in two separate phases:
! Phase 1
Refers to the time after the pavement has received a major maintenance or has been
reconstructed.
This chapter discusses the Phase 2 modelling logic. Phase 1 modelling logic is described in
Chapter C3.
Table D3.2 shows the hierarchy of works activities that are applicable to the carriageway for
JP concrete pavements.
Note:
* Works activity can be applied together with slab replacement or partial depth repair
in the same analysis year
Table D3.3 shows the hierarchy of works activities that are applicable to the carriageway for
JR concrete pavements.
Note:
* Works activity can be applied together with full depth repair in the same analysis
year
Table D3.4 shows the hierarchy of works activities that are applicable to the carriageway for
CR concrete pavements.
The works activities whose effects on pavement performance are not modelled endogenously
in HDM-4 (for example, emergency works, winter maintenance, and routine maintenance
works) will be applied in a given analysis year, if specified by the user, regardless of any
works hierarchy.
3 Routine maintenance
The routine maintenance activities on concrete roads include vegetation control, drain
cleaning, road marking, repair to road appurtenances, and sign posts. These are considered in
the analyses only in terms of their costs to the road administration. The effect of drainage on
pavement performance is accounted for through the drainage coefficient (Cd), as defined by
AASHTO (1993).
Routine maintenance works should be scheduled at a fixed interval of time (a minimum of
one-year), and will be performed on an annual basis. If specified by the user, routine
maintenance works will be applied in any analysis year regardless of the works hierarchy.
The unit cost should be specified in terms of currency per kilometre per year, and the annual
cost of routine maintenance is obtained from the product of the road section length and the
unit cost.
4 Preventive treatment
The preventive maintenance works on concrete pavements comprises the following:
! Load transfer dowels retrofit (see Section 4.1).
! Tied concrete shoulders retrofit (see Section 4.2).
! Longitudinal edge drains retrofit (see Section 4.3).
! Joint sealing (see Section 4.4).
5280 * CW
LDOW = ...(4.1)
1.6093 * JTSPACE
where:
The product of LDOW and the section length (L) in kilometres gives the total length of joints
fitted with load transfer dowels (TLDOW) in metres. The total cost of retrofitting load transfer
dowels is computed by multiplying TLDOW by the user-specified unit cost per metre length.
The effects
Load transfer restoration has the ability to increase the structural capacity of a pavement.
After works, the pavement type will be reset to JP concrete pavement with dowels, and the
progression of cracking, faulting and spalling will be computed using the models for JP
concrete pavements with dowels. If the dowel diameter is less than 20 mm, the model for JP
concrete pavement without dowels will be used. The user should also specify whether or not
the dowels fitted are protected against corrosion.
Figure D3.1 shows the effect of retrofitting load transfer dowels on faulting in JP concrete
pavements without dowels.
0.40
0.35
0.30
Average Joint Faulting (in)
0.25
0.20
0.15
0.10
Original
0.05
Dowels Placement
0.00
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00
Figure D3.1 Retrofitting load transfer dowels effect on transverse joint faulting
in JP concrete pavements without dowels
TCS = L ...(4.2)
where:
The total cost of retrofitting tied concrete shoulders is computed by multiplying TCS by the
user-specified unit cost per kilometre.
The effects
Retrofitting tied concrete shoulders can increase the structural capacity of a pavement by
decreasing the critical stresses at the concrete edge and by decreasing the deflections at the
corners.
The value of the parameter LTEsh used in the equations for predicting cracking and faulting
will be updated as described in Table D3.5.
Figure D3.2 shows the effect of retrofitting tied concrete shoulders on cracking in JP concrete
pavements without dowels.
100%
90%
Load transfer efficiency (stress), LTEsh = 10 %
80%
Transverse cracking of slabs [%]
70%
60%
50%
40%
30%
Original
20%
Concrete shoulder, LTEsh = 10
10%
0%
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00
Cumulative ESAL (millions per lane)
Figure D3.3 shows the effect of retrofitting tied concrete shoulders on faulting in JP concrete
pavements without dowels.
0.40
0.35
0.30
Average Joint Faulting (in)
0.25
0.20
0.15
0.10
..... Original
0.05 Concrete Shoulders
0.00
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00
Figure D3.4 shows the effect of retrofitting tied concrete shoulders on faulting in JP concrete
pavements with dowels.
0.40
0.35
0.30
Average Joint Faulting
0.25
0.20
0.15
0.10
Original
0.05 Diamond Grinding
0.00
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00
Cumulative ESAL (millions per lane)
Retrofitting longitudinal edge drains can be scheduled at a fixed point in time defined by the
calendar year.
RED = L ...(4.3)
where:
The total cost of retrofitting longitudinal edge drains is computed by multiplying RED by the
user-specified unit cost per kilometre.
The effects
Since many pavement distresses can be attributed to water, removing it decreases the
opportunity for deterioration and increases the pavement’s life.
After retrofitting longitudinal edge drains, the parameter Cd (drainage coefficient) used in the
equation for calculating faulting should be reset to the user-specified value. The maximum
value of Cd is 1.25.
Figure D3.5 shows the effect of retrofitting longitudinal edge drains on faulting in JP concrete
pavements without dowels.
0.40
0.35
0.30
Average Joint Faulting (in)
0.25
0.20
0.15
0.10
Original, Cd = 0.5
0.05 Lateral Drains, Cd =1.0
0.00
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00
Figure D3.6 shows the effect of retrofitting longitudinal edge drains on faulting in JP concrete
pavements with dowels.
0.40
0.35
0.30
Average Joint Faulting (in)
0.25
0.20
0.15
0.10
Original, Cd = 0.5
0.05
Lateral Drains, Cd =1.0
0.00
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00
Cumulative ESAL (millions per lane)
Joint sealing is applied at a fixed point in time defined by the calendar year.
! Option 2: Scheduled
A fixed interval between successive seals is specified, and the joint sealing is applied in
an analysis year t defined as follows:
where:
t analysis year, defined relative to the start year of the analysis period
interval a fixed interval between successive joint sealings (years)
tprev number of years since last joint sealing or start year of analysis period
5280 * CW
LJSL = + (1000 * NLNGJTS ) ...(4.6)
1.6093 * JTSPACE
where:
The product of LJSL and the section length (L) in kilometres gives the total length of joints
sealed (TLJSL) in metres. The total cost of joint sealing is computed by multiplying TLJSL
by the user-specified unit cost per metre length.
The effects
Joint sealing has an effect only on the progression of spalling. This effect is modelled through
the parameters AGE and seal type used in the spalling equation. Following joint sealing, the
seal type will be reset to the user-specified type and AGE will be reset to zero.
Figure D3.7 shows the effect of joint sealing on spalling in JP concrete pavements without
dowels.
16.00
14.00
12.00
10.00
Joints Spalling [%]
8.00
6.00
4.00
Age [years]
Figure D3.8 shows the effect of joint sealing on spalling in JR concrete pavements.
100
90
80
70
60
Joint Spalling [%]
50
40
30
20 Original
80% Joint Sealing
10
0
0 5 10 15 20 25 30
Age [years]
5 Restoration works
The following four generic types of restoration works are modelled:
1 Slab replacement
4 Diamond grinding
A slab replacement works is performed at a fixed point in time defined by the calendar
year.
! Option 2: Scheduled
A fixed interval between successive slab replacement works is specified, and the slab
replacement works is applied in an analysis year t defined by Equations 4.4 above and
4.5 above.
! Option 3: Condition - Responsive
Slab replacement is performed when the level of pavement defect (cracking) exceeds the
user-specified value.
In all cases, slab replacement will not be performed if the user-specified last applicable year or
maximum applicable roughness has been exceeded.
A slab replacement works will be specified using the percentage of damaged slabs to be
replaced.
5280 * (NLNGJTS + 1)
NTSLBS = ...(5.3)
JTSPACE
where:
The product of SLB and the section length (L) in kilometres gives the total slab replacement
area (TSLB) in square metres. The total cost of slab replacement is computed by multiplying
TSLB by the user-specified unit cost per square metre.
The effects
The immediate effects of slab replacement on cracking, spalling and faulting are computed as
described below. The long-term effect on pavement distress progression is modelled by
applying weighting factors based on the proportions of the new and old slabs.
Cracking
The amount of cracking remaining after works is given by:
where:
PCRACK bw * NTSLBS
CRKSLBS = ...(5.5)
100
where:
Figure D3.9 shows the effect of slab replacement on cracking in JP concrete pavements
without dowels.
100%
90%
80%
Transverse cracking of slabs [%]
70%
60%
50%
40%
30%
20% Original
40% Slabs Replacement
10%
0%
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00
Cumulative ESAL [millions per lane]
Spalling
The amount of spalling remaining after works is given by:
where:
Faulting
The amount of faulting remaining after works is given by:
where:
Figure D3.10 shows the effect of slab replacement on faulting in JP concrete pavements
without dowels.
0.40
0.35
0.30
Average Joint Faulting (in)
0.25
0.20
0.15
0.10
Original
0.05 Slab replacement
0.00
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00
Cumulative ESAL [millions per lane]
Figure D3.11 shows the effect of slab replacement on faulting in JP concrete pavements with
dowels.
0.40
0.35
0.30
Average Joint Faulting (in)
0.25
0.20
0.15
0.10
Original
0.05
Slabs replacement
0.00
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00
A full depth repair works is performed at a fixed point in time defined by the calendar
year.
! Option 2: Scheduled
A fixed interval between successive full depth repair works is specified, and the full
depth repair is applied in an analysis year t defined by Equations 4.4 above and 4.5
above.
! Option 3: Condition - Responsive
Full depth repair is performed when the level of pavement defects (cracking, spalling and
failures) exceeds the user-specified values.
In all cases, full depth repair will not be performed if the user-specified last applicable year or
maximum applicable roughness has been exceeded.
The amounts of works and effects of full depth repair are calculated depending on the
pavement surface type, as described below.
(CKAREA + SPAREA )
FDR = ...(5.8)
1.6093
where:
where:
where:
1.5 * 5280
NJTS = ...(5.11)
JTSPACE
where:
The product of FDR and the section length (L) in kilometres gives the total area of full depth
repair (TFDR) in square metres. The total cost of full depth repair is computed by multiplying
TFDR by the user-specified unit cost per square metre.
The effects
Following full depth repair, pavement defects will be reduced as detailed below.
! Cracking
DCRACK aw = 1 -
PCTCKS
* DCRACK bw ...(5.12)
100
where:
Figure D3.12 shows the effect of full depth repair on cracking in JR concrete pavements.
25.00
Transverse Cracking [Nº / mi]
20.00
15.00
10.00
5.00
Original
50 % Full Depth Repair
0.00
0.00 10.00 20.00 30.00 40.00 50.00 60.00
Age [years]
! Spalling
SPALL aw = 1 -
PCTSPL
* SPALL bw ...(5.13)
100
where:
Figure D3.13 shows the effect of full depth repair on spalling in JR concrete pavements.
100
90
80
70
Joint spalling (%)
60
50
40
30
20
Original
10
50% Full Depth Repair
0
0 5 10 15 20 25 30
Age [years]
Figure D3.13 Full depth repair effect on joint spalling in JR concrete pavements
! Faulting
where:
where:
Figure D3.14 shows the effect of full depth repair on faulting in JR concrete pavements.
0.25
0.20
0.15
Average Faulting [in]
0.10
0.05 Original
0.00
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00
Cumulative ESAL [millions per lane]
where:
The product of FDR and the section length (L) in kilometres gives the total area of full depth
repair (TFDR) in square metres. The total cost of full depth repair is computed by multiplying
TFDR by the user-specified unit cost per square metre.
The effects
Following full depth repair, pavement defects will be reduced as follows:
! Failures
FAIL aw = 1 -
PCTFAIL
* FAIL bw ...(5.17)
100
where:
Figure D3.15 shows the effect of full depth repair on failures in CR concrete pavements.
35.00
30.00
25.00
Failures [ Nº /mi]
20.00
15.00
10.00
5.00 Original
50% Full depth repair
0.00
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00
Cumulative ESAL [millions per lane]
A partial depth repair works is performed at a fixed point in time defined by the calendar
year.
! Option 2: Scheduled
A fixed interval between successive partial depth repair works is specified, and the partial
depth repair is applied in an analysis year t defined by Equations 4.4 above and 4.5
above.
! Option 3: Condition - Responsive
Partial depth repair is performed when the level of pavement defect (cracking) exceeds
the user-specified value.
In all cases, resealing will not be performed if the user-specified last applicable year or
maximum applicable roughness has been exceeded.
where:
The product of PDR and the section length (L) in kilometres gives the total length of partial
depth repair (TPDR) in metres. The total cost of partial depth repair is computed by
multiplying TPDR by the user-specified unit cost per metre.
The effects
Following partial depth repair, spalling will be reduced as follows:
SPALL aw = 1 -
PREPJTS
* SPALL bw ...(5.19)
100
where:
Figure D3.16 shows the effect of partial depth repair on spalling in JP concrete pavements
without dowels.
16.00
14.00
12.00
10.00
Joints Spalling [%]
8.00
6.00
4.00
Diamond grinding is performed at a fixed point in time defined by the calendar year.
! Option 2: Scheduled
A fixed interval between successive diamond grinding is specified, and the grinding
works is performed in an analysis year t defined by Equations 4.4 above and 4.5 above.
! Option 3: Condition - Responsive
Diamond grinding is performed when the level of pavement defect (faulting and
roughness) exceeds the user-specified values.
In all cases, diamond grinding will not be performed if the user-specified last applicable year
or maximum applicable roughness has been exceeded.
where:
The product of DGR and the section length (L) in kilometres gives the total area of diamond
grinding (TDGR) in square metres. The total cost of diamond grinding is computed by
multiplying TDGR by the user-specified unit cost per square metre.
The effects
It is considered that the first application of diamond grinding will not reduce slab thickness.
However, when performed for the second (or more time) diamond grinding will reduce the
slab thickness as follows:
where:
Diamond grinding will remove all faulting, that is, faulting is reset to zero and thereafter the
pavement distress modes will be modelled using the reduced slab thickness.
Roughness after diamond grinding will be computed using the Phase 1 model based on the
amounts of distresses remaining.
Figure D3.17 shows the effect of diamond grinding on faulting in JP concrete pavements
without dowels.
0.40
0.35
Average Joint Faulting (in)
0.30
0.25
0.20
0.15
0.10
Original
0.05 Diamond Grinding
0.00
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00
Cumulative ESAL [millions per lane]
Figure D3.18 shows the effect of diamond grinding on faulting in JP concrete pavements with
dowels.
0.40
0.35
0.30
Average Joint Faulting (in)
0.25
0.20
0.15
0.10
Original
0.05 Diamond Grinding
0.00
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00
Cumulative ESAL [millions per lane]
6 Rehabilitation
A concrete pavement may require a new surfacing layer either because the riding quality has
become unacceptable or for structural strengthening. There are two basic types of concrete
overlays on existing concrete pavements that are modelled in HDM-4 under rehabilitation
works:
! Bonded concrete overlays
! Unbonded concrete overlays
A fixed interval between successive bonded concrete overlay is specified, and the overlay
is performed in an analysis year t defined by Equations 4.4 above and 4.5 above.
! Option 3: Condition - Responsive
Bonded concrete overlay is performed when the level of pavement defect (cracking,
failures and/or roughness) exceeds the user-specified values.
In all cases, bonded concrete overlay will not be performed if the user-specified last
applicable year or maximum applicable roughness has been exceeded.
where:
Before concrete overlay is applied, it is often necessary to carry out some preparatory
works. For JP concrete pavements, this is equal to the amount of slab replacement to be
performed (SLB), computed using Equations 5.1 above to 5.3 above.
The total amount of bonded concrete overlay (TBOL) in square metres is obtained from the
product of BOL and the section length (L) in kilometres. The total cost of bonded concrete
overlay is obtained by multiplying TBOL by the user-specified unit cost per square metre. The
additional areas and cost of preparatory works will be reported separately under slab
replacement.
The effects
The pavement type after bonded concrete overlay will not change. The seal type for JP
concrete pavements will be changed to the user-specified type.
! Slab thickness
Bonded concrete overlay will reset the concrete slab thickness as follows:
where:
! Cracking
The amount of cracking remaining after bonded concrete overlay works is computed
using Equation 5.4 above. Thereafter, cracking progression is predicted using the Phase 1
RD model with the new slab thickness (SLABTHKaw) and the new values of the
maximum permissible number of equivalent standard axle load repetitions during each
temperature gradient (Ntg). The cumulative fatigue damage (FD) is maintained and used
in the calculation of the new Ntg values (see Section 4 of Chapter C3).
Figure D3.19 shows the effect of bonded concrete overlay on cracking in JP concrete
pavements.
90%
80%
Same fatigue
70% Slab thickness changes
Determination of the number of admissible axles N
60%
Transverse Cracking [%]
50%
40%
30%
20%
Original
10% Overlay
0%
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00
Cumulative ESAL [millions per lane]
! Faulting
Faulting after bonded concrete overlay will be reset to zero. Thereafter, the progression
of faulting will be computed using the Phase 1 RD model. The value of NE4
(cumulative traffic loading in equivalent standard axle loads) will be reset to zero.
Figure D3.20 shows the effect of bonded concrete overlay on faulting in JP concrete
pavements without dowels.
0.20
0.18
0.16
Average Joint Faulting (in)
0.14
0.12
0.10
0.08
0.06
0.04
Original
0.02 Overlay
0.00
0.00 5.00 10.00 15.00 20.00 25.00 30.00
Cumulative ESAL [millions per lane]
Figure D3.21 shows the effect of bonded concrete overlay on faulting in JP concrete
pavements with dowels
0.18
0.16
0.14
0.12
Average Joint Faulting [in]
0.10
0.08
0.06
0.04
Original
0.02 Bonded Overlay
0.00
0.00 5.00 10.00 15.00 20.00 25.00 30.00
Cumulative ESAL [millions per lane]
! Spalling
Spalling after bonded concrete overlay will be reset to zero. Thereafter, spalling
progression will be computed using the Phase 1 RD model with the parameter pavement
age (AGE) counted since the time of overlay.
Figure D3.22 shows the effect of bonded concrete overlay on spalling in JP concrete
pavements with dowels.
16.00
14.00 Original
Overlay
12.00
10.00
Joint Spalling [Nº /mi]
8.00
6.00
4.00
2.00
0.00
0 5 10 15 20 25 30
Age [years]
! Roughness
Roughness after bonded concrete overlay will be reset to a new value, and its progression
computed using the Phase 1 RD model based on the amounts of cracking, spalling, and
faulting remaining.
The effects
! Slab thickness
where:
! Cracking
The adjusted slab thickness to account for the effect of the remaining cracking is given
by:
where:
SLABTHKadj adjusted slab thickness to account for the effect of the remaining
cracking (mm)
SLABTHKbw slab thickness before bonded concrete overlay (mm)
DCRACKaw number of deteriorated transverse cracks per mile remaining after
bonded concrete overlay (mm)
The maximum number of deteriorated transverse cracks per mile (MAXCKS) is given
by:
3 * 5280
MAXCKS = ...(6.5)
JTSPACE
where:
25.00
/ ]
Transverse Cracking Deterioration [Nº mi
20.00
Original
Bonded Overlay
15.00
10.00
5.00
0.00
0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00
Cumulative ESAL [millions per lane]
! Faulting
Faulting will be reset to zero. Thereafter the progression of faulting will be computed
using the Phase 1 RD model. The value of NE4 (cumulative traffic loading in
equivalent standard axle loads) will be reset to zero.
Figure D3.24 shows the effect of bonded concrete overlay on faulting in JR concrete
pavements.
0.18
0.16
Original
0.14
Average Faulting [in]
Bonded Overlay
0.12
0.10
0.08
0.06
0.04
0.02
0.00
0.00 5.00 10.00 15.00 20.00 25.00 30.00
Spalling after bonded concrete overlay will be reset to zero. Thereafter spalling
progression will be computed using the Phase 1 RD model with the parameter pavement
age (AGE) counted since the time of overlay.
Figure D3.25 shows the effect of bonded concrete overlay on spalling in JR concrete
pavements.
50
45 Original
Overlay
40
35
30
Joint Spalling [%]
25
20
15
10
0
0 5 10 15 20 25 30
Age [years]
! Roughness
Roughness after bonded concrete overlay will be reset to a new value, and its progression
is computed using the Phase 1 RD model based on the PSR value.
The effects
Following bonded concrete overlay the slab thickness will be reset as expressed by Equation
6.2 above.
! Failures
The number of failures after bonded concrete overlay will be reset to zero.
Figure D3.26 shows the effect of bonded concrete overlay on failures in CR concrete
pavements.
160.00
140.00 Original
Bonded Overlay
120.00
Failures [Nº /mi]
100.00
80.00
60.00
40.00
20.00
0.00
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00
! Roughness
Roughness after bonded concrete overlay will be reset to a new value, and its progression
is computed using the Phase 1 RD model based on the PSR value.
A fixed interval between successive unbonded concrete overlays is specified, and the
overlay is performed in an analysis year t defined by Equations 4.4 above and 4.5 above.
! Option 3: Condition - Responsive
Unbonded concrete overlay is performed when the level of pavement defect (cracking,
failures and/or roughness) exceeds the user-specified values.
In all cases, unbonded concrete overlay will not be performed if the user-specified last
applicable year or maximum applicable roughness has been exceeded.
where:
The total amount of unbonded concrete overlay (TUOL) in square metres is obtained from the
product of UOL and the section length (L) in kilometres. The total cost of unbonded concrete
overlay is obtained by multiplying TUOL by the user-specified unit cost per square metre.
The effects
After unbonded concrete overlay, the roadbase type will change to rigid/concrete base (RB).
The seal type for JP and JR concrete pavements will be changed to the user-specified types. It
is considered that the pavement will behave as new and the performance will be modelled
using the Phase 1 RD models. Roughness after unbonded concrete overlay will be reset to the
user-specified value.
7 Reconstruction
Pavement reconstruction refers to all works that require the re-specification of part or the
entire pavement structure and characteristics, which involves the removal and replacement of
the surfacing, roadbase and sub-bases. The re-specification of concrete pavement
reconstruction also allows for the adjustment to roadside geometry and safety features, and to
improve or add drainage structures.
Reconstruction may be specified in one of the following ways:
! Option 1: Scheduled
A reconstruction works of fixed specifications is applied when the levels of the user-
specified intervention criteria, based on pavement condition, are met. Reconstruction will
not be performed if the last applicable year has been exceeded.
where:
The total area of reconstruction (TREC) in square metres is obtained from the product of REC
and the road section length (L) in kilometres. The total cost of pavement reconstruction is
obtained by multiplying TREC by the user-specified unit cost per square metre.
The effects
After reconstruction, the pavement type will be reset to the new type specified by the user.
The pavement will behave as new and the performance will be modelled using the Phase 1 RD
models. Roughness will be reset to the user-specified value. The calibration factors for
modelling pavement deterioration will also be reset to user-specified values.
8 Special works
The effects of the following special works on pavement performance are not modelled
endogenously in HDM-4, and therefore only their costs can be considered in an analysis:
! Emergency works
For example, repairing washout/subsidence, clearing debris, traffic accident removal, etc.
! Winter maintenance
These works will be scheduled at a fixed interval of time (minimum of one-year interval), and
will be performed on an annual basis. If specified, these works activities will be applied in a
given analysis year regardless of the works hierarchy.
Their unit costs will be specified in terms of currency per kilometre per year, and the annual
costs will be obtained by multiplying the road section length (L) by the unit cost.
9 New construction
The construction of a new section can only be scheduled and not triggered by responsive
intervention criteria. In a project analysis, a new road section (that is, a new link) can be
specified as a section alternative within a selected project alternative.
The required components for defining a new section are as follows:
! Road section data
All the data items that are required to define a road section in HDM-4. The user can
specify these items using aggregate data.
! Traffic data
❏ diverted traffic
traffic that is diverted to the new section (link) from the nearby routes and other
transport modes.
❏ generated traffic
additional traffic that occurs in response to the new investment.
! Construction costs and duration
! Exogenous benefits and costs
! Maintenance standards
10 References
AASHTO, (1993) American Association of State Highway and Transportation Officials
Guide for Design of Pavement Structures
Washington D.C., USA
LAST, (1995)
Concrete pavement performance equations
Latin American Study Team
International Study of Highway Development and Management Tools
Santiago, Chile
LAST, (1996)
Modelling road design and maintenance effects for pavements in HDM-4
Final Report, FICEM, Latin American Study Team
International Study of Highway Development and Management Tools
Santiago, Chile
D4 Unsealed Roads
1 Introduction
This document describes the detailed modelling of Road Works Effects for unsealed roads
(see Figure D4.1).
Road
RoadWorks
WorksEffects
Effects
The methods of defining works activities and intervention criteria, the calculation of physical
quantities of works and the costs to road administration, and works effects on road
characteristics and road use are discussed for the following works classes:
! Maintenance (see Section 3)
The modelling logic described comprises the overall computational procedure, the
hierarchical ranking of works activities and pavement type resets after works.
A list of research documents referenced from this chapter is given in Section 6.
2 Modelling logic
An unsealed road is considered to comprise two layers, a gravel surfacing and a subgrade. A
gravel road has both layers, but an earth road has a zero thickness of gravel surfacing and its
surface characteristics are those of the subgrade. When a gravel road loses all of its gravel
surfacing, then its classification reverts to that of earth road. Upon gravel resurfacing, all
unpaved roads become gravel roads by definition of the new surfacing layer. The background
of the modelling logic is given in Watanatada et al. (1987) and Paterson (1987).
Notes:
1 Spot regravelling and grading have the same ranking, and both of them can be
performed in the same analysis year
are applied in a given analysis year, if specified by the user, regardless of any works
hierarchy.
3 Maintenance works
The maintenance of unsealed roads comprises the following operations:
! Periodic grading (see Section 3.1)
Maintenance of gravel surfacing is accounted each analysis year through the surfacing
thickness and the net change from material loss, spot regravelling and gravel resurfacing
maintenance. The material loss from earth roads, although computed, is accounted only for the
purpose of predicting spot regravelling quantities and is otherwise ignored, Watanatada et al.
(1987).
Grading works is performed at a fixed time interval in days between successive grading.
! Option 2: Traffic-responsive
Grading is performed after a maximum allowable roughness value has been exceeded.
DG = DG' ...(3.2)
If DG' ≤ DGMIN
DG = DGMIN ...(3.3)
where:
DGMAX the maximum allowable time interval between successive grading, in days,
specified by the user as an option or equal to the default value of 10,000 days
DG' the number of days between successive grading determined from the traffic or
roughness parameter
DGMIN the minimum applicable time interval between successive grading, in days,
specified by the user as an option or equal to the default value of 5 days
VEHG
DG' = ...(3.4)
AADT
(QIMAX j − QIMAX 0 )
DG' = * log e
1
...(3.5)
c [QIMAX j − (1 − a ) * QIMIN j − a * QIMAX 0 ]
where:
VEHG the traffic interval between successive grading, in vehicles, specified by the
user
QIMAXj the maximum roughness of material j (QI)
QIMINj the minimum roughness of material j (QI)
QIMAXO the maximum allowable roughness specified by the user (QI)
a and c are model parameters which are defined in Chapter C4, Sections 3.2, 3.4 and
3.5
If no grading is specified, the long-term average roughness (QIlta) is equal to the maximum
roughness, as follows:
Note that if the historic maintenance of the section has been nil-grading over several years,
then the existing roughness is the best estimate of the average roughness and the user can
provide this by specifying QIMAX endogenously with a value equal to the existing
roughness.
A fixed number of cubic metres per kilometre (VGS) per year is applied.
! Option 2: Responsive
A percentage of gravel or subgrade material loss in the current analysis year is replaced
subject to a maximum limit per year. In this option VGS is computed as follows:
where:
VGS the in-place volume of material added due to the spot regravelling
(m3/km)
Pmla percentage of annual material loss to be replaced, specified by the user
(%)
MLA annual material loss (mm)
CW carriageway width (m)
SW shoulder width (m)
where:
TVGS Total amount of spot regravelling for the road section (m3)
L road section length (km)
The cost of spot regravelling is computed as the product of TVGS and the unit cost of material
per cubic metre.
For gravel roads, the thickness of the gravel layer is increased to reflect the volume of
material added, according to the following formula (trapezoidal rule):
VGS
∆THGS = ...(3.9)
(CW + SW )
where:
Spot regravelling is predicted to reduce the average roughness on the assumption that the
gravel is applied in the major depressions and potholes that have appeared in the surface
in the upper ranges of roughness. Roughness levels above 190 QI (15 m/km IRI) are
invariably associated with the presence of visible birdbath type depressions or potholes,
which become larger or more frequent as the roughness level increases, and these can be
effectively patched, with high benefits, by spot regravelling. Over the roughness range of
150 to 190 QI (that is, 11 to 15 IRI m/km), such patchable birdbath depressions are
frequently observed but not always present so that, in this range, spot regravelling may
not always be effective. For example, spot regravelling is not effective maintenance on
corrugations or on runoff-induced surface erosion, which conditions commonly induce
roughness levels within this range. At roughness levels below 150 QI (that is, 11 IRI
m/km) spot regravelling is considered to be ineffective on roughness. This logic is
defined in the following algorithm, adopting the roughness to volume of depression ratio
as equal to 2 QI per m3/lane/km), allowing for the spot regravelling to be only 60%
effective (that is, 1.2 QI per m3/lane/km), and adopting an average effective lane width of
3 m:
[
QI avg(bw) - 150
QIavg (aw ) = MAX150, QI avg(bw ) − MIN1,
] * 3.6 * VGS
...(3.11)
40 CW
where:
! Option 1: Scheduled
Gravel resurfacing is applied when the gravel age (GAGE) equals or exceeds the fixed
time interval specified by the user.
! Option 2: Responsive
where:
The volume of gravel added per km is computed according to the following trapezoidal
formula:
Preparatory works
If roughness before works (RIbw) is greater than 11.6 IRI m/km, it is assumed that the
following amount of spot regravelling is performed before gravel resurfacing:
where:
The additional cost of spot regravelling is obtained by multiplying the product of VGS and the
section length (L) by the user-specified unit cost per cubic metre.
When gravel resurfacing is performed the pavement type is set to gravel (GRUP)
regardless of the previous surface type.
! Gravel age
The thickness of the gravel surfacing is increased according to the formula below:
If the final gravel thickness is specified:
where:
THG0 the final gravel thickness after gravel resurfacing, specified by the user
(mm)
∆THG the increase in the gravel thickness due to gravel resurfacing, specified by
the user (m)
! Material properties
The existing surface material is changed to the material specified by the user (which may
still be of the same attributes as the existing). The surface material attributes (P075,
P425, P02, D95, PI, QIMIN and QIMAX) are replaced either by the new values provided
by the user, or by the default values from the previous gravel attributes.
! Roughness
Roughness after regravelling is reset to a user specified value. If this is not specified, the
roughness after works is reset to the minimum allowable value (QIMIN) converted into
IRI m/km units.
cost of drainage maintenance should be included, when it is normally done. Otherwise, some
allowance due to the lack of drainage, for example, in the form of frequent road closures,
washouts, etc., should be incorporated in the economic analysis.
When specified by the user, the total annual cost of routine-miscellaneous maintenance is
obtained from the product of the section length (L) and the unit cost per kilometre per year.
4 Improvement works
Improvement works for unsealed roads comprises the following:
! Widening (see Section 4.1)
4.1 Widening
The operations included under widening are lane addition and partial widening. The
difference between the two is that partial widening does not increase the number of lanes. It is
considered that these operations do not alter the road alignment, hence there is no change in
section length.
Widening works can be defined in one of the following ways:
! Option 1: Scheduled
where:
The total area of widening over the entire road section is given by:
where:
The cost (CSTWDN) of widening the section is obtained from the product of TAWDN and
the user-specified unit cost per square metre, or from the product of the section length L and
the unit cost per kilometre.
The salvage value is given as:
where:
Additional works
It is likely that widening works will involve resurfacing, or repair to the severely damaged
area of the existing carriageway. The required additional works are modelled as described
below:
! Case 1: resurfacing the existing carriageway
where:
where:
The cost of regravelling the existing carriageway is obtained from the product of TVGR
and the user-specified unit cost per cubic metre.
! Case 2: no resurfacing of the existing carriageway
where:
For partial widening the increase in carriageway width (∆CW) is specified directly by the
user. For lane addition, the increase in carriageway width is user-specified. If this
is not specified the increase in width is given by:
(ADDLN * CWbw )
∆CW = ...(4.7)
NLANES bw
where:
For lane addition, the number of lanes after widening works (NLANES aw) is equal
to the number of lanes before works (NLANESbw) plus the user-specified additional
number of lanes (ADDLN).
! Gravel thickness
where:
and the pavement type is set to gravel (GRUP) regardless of the previous pavement
type
If the existing carriageway is not to be regravelled:
where:
❏ The increase in gravel thickness over the existing carriageway due to spot
regravelling (∆THGS) is obtained as follows:
If pavement type is gravel (GRUP):
After widening, the surface material properties (SMPi) are reset as follows:
❏ If the existing carriageway is to be regravelled, all the surface material properties are
reset to those of the new gravel material.
❏ If the existing carriageway is not to be regravelled
where:
! Roughness
Roughness after widening (RIaw) is reset to a user-specified value. If this is not specified,
reset RIaw to the minimum allowable roughness (QIMIN) converted into IRI m/km units.
! Gravel age
(CWbw * GAGE bw )
GAGE aw = ...(4.12)
CWaw
where:
GAGEaw gravel age after widening works (years) (Note: returns an integer
value)
GAGEbw gravel age before widening works (years)
After widening works, the material loss calibration factors (that is, Kgl and Kkt) are reset
to user-specified values.
! Speed factors
These are the speed limit, speed enforcement factor, roadside friction factor, non-
motorised transport and motorised transport speed reduction factors, and acceleration
noise which depends primarily on the individual road section.
! Traffic flow pattern (road use)
The data that describes the hourly traffic flow distribution is also reset to a user-specified
type.
4.2 Realignment
Refers to local geometric improvements of existing roads, which may also result in a
reduction of the road length. It is assumed that the carriageway width remains unaltered when
a realignment works is performed.
Realignment works can be defined as follows:
! Option 1: Scheduled
L aw = L bw * LF ...(4.14)
where:
The cost of realignment (CSTREAL) is obtained from the product of REAL and the user-
specified realignment unit cost per kilometre.
The salvage value is given by:
where:
Additional works
It is assumed that the following amounts of additional works would be carried out together
with the realignment works:
! Case 1: resurfacing the non-realigned segments
If the non-realigned parts of the existing carriageway are to be resurfaced, the amount of
regravelling works is given as follows:
( )
VGR = THG aw - THG bw * (CW + SW ) ...(4.16)
where:
The total amount of regravelling the non-realigned parts of the existing carriageway is
given by:
where:
The cost of regravelling the non-realigned parts of the existing carriageway is obtained
from the product of TVGR and the user-specified unit cost per cubic metre.
! Case 2: no resurfacing of the non-realigned segments
where:
If roughness before works (RIbw) is greater than 11.6 IRI m/km, the amount of spot
regravelling performed on the non-realigned parts of the existing carriageway is
computed as follows:
where:
The cost of spot regravelling the non-realigned parts of the existing carriageway is
calculated from the product of TVGS and the user-specified unit cost per cubic
metre.
The total cost of realignment works is the sum of realignment construction cost and the cost of
additional works comprising regravelling of the non-realigned parts of the existing
carriageway, or grading and any spot regravelling performed. The additional costs and the
amounts of works should be reported separately under regravelling, grading or spot
regravelling.
In economic analysis it is assumed that these additional costs are incurred in the last
construction year.
The new length of the road section after realignment is given by Equation 4.14 above.
! Gravel thickness
where:
❏ The gravel thickness over the non-realigned parts of the existing carriageway after a
realignment works is obtained as follows:
If the non-realigned parts of the existing carriageway are to be regravelled
and the pavement type is set to gravel (GRUP) regardless of the previous pavement
type.
If the non-realigned parts of the existing carriageway are not to be regravelled
where:
❏ The increase in gravel thickness over the non-realigned parts of the existing
carriageway due to spot regravelling (∆THGS) is obtained as follows:
If pavement type is gravel (GRUP):
∆THGS is computed using Equation 3.9 above
If pavement type is earth (EAUP):
∆THGS is set to zero
! Surface material properties
After realignment works, the surface material properties (SMPi) are reset as follows:
❏ If the non-realigned parts of the existing carriageway are to be regravelled, all the
surface material properties are reset to those of the new gravel material
❏ If the non-realigned parts of the existing carriageway are not to be regravelled
where:
! Roughness
Roughness after realignment works (RIaw) is reset to a user-specified value. If this is not
specified, reset RIaw to the minimum allowable roughness (QIMIN) converted into IRI
units of m/km.
! Gravel age
where:
GAGEaw gravel age after realignment works (returns an integer value, in years)
GAGEbw gravel age before realignment works (years)
After realignment, the material loss calibration factors (that is, Kgl and Kkt) are reset to
user-specified values.
! Speed factors
These are the speed limit, speed enforcement factor, roadside friction factor, non-
motorised transport and motorised transport speed reduction factors, and acceleration
noise which depend primarily on the individual road section.
! Traffic flow pattern (road use)
The data that describes the hourly traffic flow distribution is also reset to a user-specified
type.
5 Construction works
Construction works for unsealed roads comprises the following:
! Upgrading (see Section 5.1)
5.1 Upgrading
An unsealed road can be upgraded to a bituminous or concrete pavement. It is also possible to
upgrade an earth road to a gravel road, although both are of the same surface class.
Upgrading works can be defined as follows:
! Option 1: Scheduled
LUPGRD = L aw ...(5.1)
where:
The total cost of upgrading (CSTUPGRD) is obtained from the product of LUPGRD and the
user-specified unit cost per kilometre.
The salvage value is given as:
where:
The new carriageway width after upgrading is calculated using Equation 4.6 above.
The increase in carriageway width is either specified directly by the user, or calculated
using Equation 4.7 above.
The number of lanes after upgrading works (NLANESaw) is equal to the number of lanes
before works (NLANESbw) plus the user-specified additional number of lanes (ADDLN).
! Speed factors
These are the speed limit, speed enforcement factor, roadside friction factor, non-
motorised transport and motorised transport speed reduction factors, and acceleration
noise which depend primarily on the individual road section.
! Traffic flow pattern (road use)
The data that describes the hourly traffic flow distribution is also reset to a user-specified
type.
5.2 Dualisation
The dualisation of an existing road section can be either scheduled or triggered by responsive
intervention criteria.
Note: The modelling of dualisation works is not included in this release.
All the data items that are required to define a road section in HDM-4. The user is able
to specify these data items in aggregate terms.
! Traffic data
❏ diverted traffic - traffic that is diverted from the nearby routes and other transport
modes
❏ generated traffic – additional traffic that occurs in response to the new investment
! Construction costs, duration and salvage value
! Exogenous benefits and costs
! Maintenance and improvement standards
6 References
Paterson W.D.O., (1987)
Road Deterioration and Maintenance Effects
World Bank Publications, Washington D.C.
Watanatada T., Harral C.G., Paterson W.D.O., Dhareshwar A.M., Bhandari A., and
Tsunokawa K., (1987)
The Highway Design and Maintenance Standards Model - Volume 1 Description
World Bank, John Hopkins University Press
Watanatada T., Harral C.G., Paterson W.D.O., Dhareshwar A.M., Bhandari A., and
Tsunokawa K., (1987)
The Highway Design and Maintenance Standards Model - Volume 2 User's Manual
World Bank, John Hopkins University Press
Contents
E1 Overview
1 Introduction E1-1
2 Vehicle classification system E1-2
3 Vehicle speeds and operating costs E1-5
4 Non-motorised transport E1-6
5 Road safety E1-7
6 Total road user costs E1-8
7 References E1-9
E3 Non-Motorised Transport
1 Introduction E3-1
2 Modelling concepts and logic E3-2
2.1 NMT vehicle types E3-2
2.2 Modelling issues E3-3
2.3 Data requirements E3-4
2.4 Computational logic E3-4
3 Impact of NMT on motorised transport E3-6
3.1 Impact on MT speed E3-6
3.2 Impact on MT operating costs E3-6
4 NMT speeds E3-7
4.1 Factors influencing NMT speeds E3-7
4.2 The speed model E3-7
4.3 VROUGH E3-8
4.4 VGRAD E3-9
4.5 Resistance to motion E3-10
5 NMT time and operating costs E3-14
5.1 Travel time cost E3-14
5.2 Operating cost E3-15
5.3 Capital cost E3-15
5.4 Repair and maintenance cost E3-16
5.5 Crew cost E3-17
5.6 Energy cost E3-17
5.7 Overheads E3-18
6 Estimation of economic benefits E3-19
7 References E3-20
E4 Road Safety
1 Introduction E4-1
2 Modelling logic E4-2
Analytical
AnalyticalFramework
Framework
and
andModel
Model
Descriptions
Descriptions
Introduction
Introduction
Part
PartAA
Traffic
Traffic
Part
PartBB
RD
RDModel WE RUE
RUEModel SEE
SEEModel
Model WEModel
Model Model Model
Part
PartCC Part Part
PartEE Part
PartFF
PartDD
Economic
EconomicAnalysis
Analysis
RD = Road Deterioration Part
PartGG
Nomenclature
Nomenclature
WE = Road Works Effects Part
PartHH
E1 Overview
1 Introduction
The modelling of Road User Effects (RUE) in HDM-4 (see Figure E1.1) comprises analysis
of the following:
! Motorised vehicle (MT) speed, operating costs and travel time (see Figure E1.2)
! Non-motorised transport (NMT) speed and operating costs (see Figure E1.3)
This chapter gives an overview of the HDM-4 vehicle classification system; and describes the
different RUE components considered in HDM-4.
Road
RoadUser
UserEffects
Effects
Overview Vehicle
VehicleSpeeds
Speeds Non-Motorised
Non-Motorised Road
RoadSafety
Safety
Overview
and
andOperating
Operating Transport
Transport
Chapter
ChapterE1
E1 Chapter
ChapterE4
E4
Costs
Costs Chapter
ChapterE3
E3
Chapter
ChapterE2
E2
Individual representative vehicle types, for which a set of RUE relationships has been
provided.
Figure E1.2 shows the hierarchical representation of motorised vehicles into categories,
classes and types (NDLI, 1995). Figure E1.3 shows a similar hierarchical representation of
non-motorised transport (PADECO, 1996).
Motorised
Motorised Categories
Motor
Motorcycles
cycles Passenger
Passengercars
cars Utilities
Utilities Trucks
Trucks Buses
Buses Classes
Motor
Motorcycles Small
Smallcar Medium Light
Lighttruck Medium
cycles car Medium truck Medium truck
truck Types
(1)
(1) (2)
(2) car
car (8)
(8) (9)
(9)
(3)
(3)
Large
Largecar
car
(4)
(4)
Heavy
Heavytruck
truck Articulated
Articulated truck
truck
(10)
(10) (11)
(11)
Light Light
Lightgoods Mini-bus Light
Light goods Mini-bus Lightbus
bus
delivery
delivery vehicle
vehicle (12) (13)
(12) (13)
vehicle
vehicle (6)
(6)
(5)
(5)
Four
Fourwheel
wheel Medium
Mediumbusbus Heavy
Heavybus
bus
drive
drive (14)
(14) (15)
(15)
(7)
(7)
Coach
Coach
(16)
(16)
Fleet
Fleetrepresentation
representation
Categories
Motorised
Motorisedtransport
transport Non-motorised
Non-motorised
(MT)
(MT) transport
transport(NMT)
(NMT)
Pedestrian
Pedestrian Bicycle
Bicycle Cycle Animal
Cycle Animalcart
cart Farm
Farmtractor
tractor Classes
rickshaw
rickshaw
Pull
Pulltype
type Push
Pushtype
type Horse
Horsecart
cart Ox
Oxcart
cart Types
Pedestrian
Pedestrian Bicycle
Bicycle Small
Small Large
Large
Travel time is considered in terms of passenger-hours during working and non-working time,
and cargo holding hours. Travel time costs are expressed more appropriately only in economic
terms. Additional costs due to impassability of seriously damaged unsealed roads are also
included in the total amount of motorised road user cost.
4 Non-motorised transport
Non-motorised transport (NMT) modes such as bicycles, cycle rickshaws, animal carts, and
pedestrians play a major role in moving passengers and freight in many countries (see Chapter
E3). The use of NMT is increasing in some regions mainly because of their affordability,
flexibility and cost-effectiveness in providing low cost transportation. Furthermore, the
increasing focus on efficiency in energy use, and the environmental impacts arising from the
ever-increasing use of motorised transport (MT), has highlighted the need for better provision
of NMT facilities. This has led to the recognition that the full range of transport needs in many
countries would not be catered for adequately by MT alone. Therefore, investment policies in
the road transport sector should include NMT issues.
A formal method has been developed for calculating the operating costs incurred by NMT on
roads and thereby for estimating the benefits derived by NMT from road improvements
(Odoki and Kerali, 1999). The presence of NMT can influence the speed of motorised
transport, thereby affecting the operating costs of motorised vehicles. In addition, policies
such as road improvements influence the costs and benefits to both motorised and non-
motorised road users.
5 Road safety
The HDM-4 system allows users to define a series of look-up tables for accident rates. These
are basically broad, macro descriptions of the expected accident rates defined according to a
particular set of road and traffic attributes (for example, road type, traffic level and flow
pattern, presence of NMT, and geometry class). This tabular approach to implementing road
safety analysis was recommended (ISOHDM, 1995) following a detailed review of various
road safety studies, modelling and analysis methods. For each road type or intersection type,
users are required to specify the accident rate for each severity (that is, fatal, injury or damage
only), in terms of the numbers of accidents per 100 million vehicle-kilometres. When a road is
improved (for example, providing separate NMT lanes, and widening of road shoulders) a
new set of accident rates can be specified based on data observed for roads with similar traffic
flow and geometric characteristics. Thus, it is possible to analyse the change in total numbers
of accidents and the costs resulting from the improvement.
The annual road user cost for each investment option is given by:
where:
7 References
ISOHDM, (1995)
Predicting changes in accident rates in developing countries following modifications in
road design
International Study of Highway Development and Management
School of Civil Engineering, The University of Birmingham
Kerali H.G.R., Odoki J.B., and Wightman D.C., (1994)
Vehicle Fleet Representation, Draft Working Paper
International Study of Highway Development and Management Tools
School of Civil Engineering, The University of Birmingham
NDLI, (1995)
Modelling Road User Effects in HDM-4 - Final Report
Asian Development Bank Project RETA 5549. International Study of Highway
Development and Management Tools
N.D. Lea International, Vancouver
Odoki J.B., and Kerali H.G.R., (1999)
Modelling Non-motorised Transport in HDM-4 - TRB, Paper No. 991129
Transportation Research Board, 78th Annual Meeting, Washington D.C., USA
PADECO Co. Ltd., (1996)
Non-Motorised Transport (NMT) Modelling in HDM-4 - Draft Final Report (second
Version)
International Study of Highway Development and Management Tools. Transport
Division, The World Bank, Washington D.C., USA
Watanatada T., Harral C.G., Paterson W.D.O., Dhareshwar A.M., Bhandari A., and
Tsunokawa K., (1987)
The Highway Design and Maintenance Standards Model - Volume 1 Description
The World Bank, John Hopkins University Press
1 Introduction
This chapter describes the implementation of Road User Effects (RUE) models for calculating
motorised vehicle speeds, operating costs and travel time (see Figure E2.1). It provides an
overview of the modelling concepts and logic, and a description of the relationships and
default parameter values for each of the RUE components and HDM-4 standard representative
vehicles. For further details and background to the equations refer to NDLI (1995) and
Watanatada et al. (1987a).
Road
RoadUser
UserEffects
Effects
Overview Vehicle
VehicleSpeeds
Speeds Non-Motorised
Non-Motorised Road
RoadSafety
Safety
Overview
and
andOperating
Operating Transport
Transport
Chapter
ChapterE1
E1 Chapter
ChapterE4
E4
Costs
Costs Chapter
ChapterE3
E3
Chapter
ChapterE2
E2
After describing the modelling concepts and logic the chapter is divided into four modules as
follows:
! Module A (Sections 3 and 4)
The basic data for each of the 16 representative vehicle types are also given in Table E2.1.
These values were estimated from a variety of sources, as reported in NDLI (1995).
Includes the road alignment data, speed limit, roadside friction factor, section length,
width and the number of lanes.
! Speed-flow relationship
Includes road capacity and the parameters that determine the vehicle operating speeds
and flow characteristics at different traffic flow levels.
! Traffic flow pattern
Includes parameters that describe the road use in terms of the hourly traffic flow
distribution, and are used to determine the traffic flow in passenger car space equivalents
per hour (PCSE/h) for each traffic flow period.
! Road condition
Comprises data on the annual average surface roughness and texture depth for the road
section. These are obtained from the output data of the Road Deterioration module.
! Traffic
Comprise traffic volumes specified in terms of AADT (annual average daily traffic),
traffic composition and growth for each road section. These data should be obtained from
the traffic models.
! Unit costs
Includes costs of vehicle resources, (for example, fuel cost per litre, crew wages, new
vehicle price, tyre cost, etc.) and value of time. The vehicle resource data should be
defined in economic and financial terms, but the value of time should be specified only in
economic terms.
The following speed components are calculated for a given road section:
(a) Free speed of each vehicle type
(b) Congested speeds by vehicle type - these are the operating speeds at different traffic
flow levels
(c) Annual average operating speed of each vehicle type
(d) Annual average traffic speed – this is the weighted average speed of all the vehicles
in a traffic stream.
2 Compute quantities of vehicle operating resources
in the following order:
(a) Fuel
(b) Lubricating oil
(c) Tyre
(d) Spare parts
(e) Maintenance labour hours
(e) Capital costs (comprises depreciation and interest)
(f) Crew hours
(g) Overhead
3 Calculate travel time
in terms of passenger-hours during working and non-working time, and cargo holding
hours
4 Cost vehicle resources and travel time
by applying unit costs to the predicted quantities of the resources consumed
5 Calculate increased operating cost
due to reduced passability on seriously damaged unsealed roads
6 Summarise and store data
for use in subsequent analysis and for reporting
Vehicle Type Description Abbre- Fuel Number Number of Aero- Projected Tare Operating
Number viation type of axles wheels dynamic frontal area weight weight
drag Coeff.
(m2) (t) (t)
Notes:
3 Free speeds
This is defined as the speed of each vehicle travelling on uncongested wide sections of road in
the environment under investigation. Free speeds are required for determining the operating
speeds of each vehicle type on a given road section under different flow characteristics.
The modelling of free speed described in this section is focused on the individual
representative vehicle types.
σ 2
exp
2
VS ku = β
1 1 1 1 1
1 β 1 β 1 β 1 β 1 β
+ + + +
VDRIVEu VBRAKEu VCURVE VROUGH VDESIR
...(3.1)
where:
VSku the predicted steady-state speed for the uphill segment (m/s)
VDRIVEu the speed limited by gradient and used driving power for the uphill segment
(m/s) (see Section 3.2.1)
VBRAKEu the speed limited by gradient and used braking power for the uphill segment
(m/s) (see Section 3.2.2)
The model parameter β determines the shape of the assumed Weibull distribution of the
constraining speeds. When β approaches zero, the mean speed would be equal to the
minimum of the five constraining speeds. The greater the value of β the further away the
predicted mean speed will be from the constraining speed.
As described in Watanatada et al. (1987), the model parameter σ is an estimate of the
standard error of residuals in the estimation, which involves a logarithmic transformation. The
numerator of Equation 3.1 above gives the value of the bias correction factor.
σ2
exp
2
VS kd = β
1 1 1 1 1
1 β 1 β 1 β 1 β 1 β
+ + + +
VDRIVEd VBRAKEd VCURVE VROUGH VDESIR
...(3.2)
where:
VSkd the predicted steady-state speed for the downhill segment (m/s)
VDRIVEd the speed limited by gradient and used driving power for the downhill segment
(m/s)
VBRAKEd the speed limited by gradient and used braking power for the downhill segment
(m/s)
σ β (kW) (kW)
7.2
Sk = ...(3.3)
1 1
VS +
ku VS kd
where:
3.2.1 Limiting speed based on road gradient and engine power (VDRIVE)
The limiting speed due to driving power is related to the used driving power and the road
gradient through the balance of forces in the absence of acceleration. It is calculated by
solving the following cubic equation, which is based on the hypothesis that the vehicle is
driven at steady-state speed on a smooth, straight road:
where:
The used driving power is generally less than the rated power of the engine. The forces
opposing motion, under the hypothesis given above, are the aerodynamic resistance, gradient
resistance and rolling resistance (see Section 5.2).
The parameters z0 and z1 are calculated as follows:
where:
The mass density of air is required for computing the aerodynamic resistance, and is given by
St. John et al. (1978) as:
(
RHO = 1.225 * 1 - 2.26 * ALT * 10 -5 )4.255 ...(3.5)
where:
ALT road altitude, defined as the elevation of the road section above the mean
sea level (m)
Rolling resistance is calculated as a function of tyre and pavement characteristics, and climatic
factors.
The tyre factor CR1 depends upon the tyre type as follows:
CR1 = 1.0 if TYRE_TYPE = Radial
CR1 = 1.3 if TYRE_TYPE = Bias-ply
The pavement dependent coefficient of rolling resistance CR2 is calculated as:
...(3.6)
where:
Table E2.3 gives the default values for the various rolling resistance model parameters.
CR_B_a1
b12 =
WHEEL_DIA
CR_B_a2 * NUM_WHEELS
b13 =
(WHEEL_DIA )2
where:
CR_B_a2
The default values for tyre parameters CR_B_a0 to CR_B_a2 are given in Table E2.4.
The rolling resistance depends upon the percentage of time travelled on snow covered
(PCTDS) and water covered (PCTDW) roads:
The average road section gradient GR is estimated from the following expression taken from
Watanatada et al. (1987a):
RF
GR ± ...(3.7)
1000
where:
Thus, solving the cubic equation with GR = + [RF/1000] would yield the value for VDRIVEu,
and solving it with GR = - [RF/1000] would yield the value for VDRIVEd.
The solution of the cubic equation is through Descartes’ rule of signs:
z1
z2 =
(3 * z0)
1000 * PDRIVE
z3 =
(2 * z0)
DT = z2 3 + z3 2
if DT > 0 :
VDRIVE = 3 DT + z3 − 3 DT - z3
2π 4π
VDRIVE = MAX r * cos(z ), r * cos z + , r * cos z +
3 3
where:
1 − 2 * z3
z= arccos
3 z2 * r
r = 2 * − z2
The default values for the VDRIVE model parameters are given in Table E2.2.
Table E2.4 Parameters for calculating aerodynamic, rolling and inertial resistance
3.2.2 Limiting speed based on road gradient and braking capacity (VBRAKE)
For uphill segments, the value of VBRAKE is infinite, (that is, the speed on upgrades is not
limited by the braking power).
VBRAKEu = ∞ ...(3.8)
The speed on downgrades is dependent upon the length of gradient. Once the gradient length
(GL) exceeds a critical value, the brakes are used to reduce the speed. Below this critical
gradient there is no effect of downgrade on speed.
The critical gradient length CGL is calculated as follows:
where:
Thus, the following two conditions need to be analysed for the downhill segment:
If GL < CGL
VBRAKEd = ∞ ...(3.10)
If GL > CGL,
VBRAKEd is obtained by solving the following cubic equation, which is formulated using the
mechanistic principle of balancing forces:
where:
z1 =
b11 * CR2 * FCLIM * NUM_WHEELS
+ b12 * CR1 * CR2 * FCLIM * WGT_OPER + WGT_OPER * g * GR
1
GL = ...(3.12)
NUM_RF
where:
NUM_RF average number of road rise and fall per kilometre (minimum value = 0.1)
The solution of the above cubic equation is through Descartes’ rule of signs:
z1
z2 =
3 * z0
-1000 * PBRAKE
z3 =
2 * z0
DT = z2 3 + z3 2
if DT ≥ 0
VBRAKEd = ∞
else if DT < 0
1 −2 * z3
z= arccos
3 z2 * r
r = 2 * - z2
4π
VBRAKEd = r * cos z +
3
The default values for the VBRAKE model parameters are given in Table E2.2 (see Section
5.1.2).
where:
The average radius of road curvature, R, is estimated from the following expression taken
from Watanatada et al. (1987a):
180,000
R= ...(3.14)
π * MAX(18 , C)
π
where:
The default values for the limiting curve speed parameters are given in Table E2.5.
where:
ARV the average rectified velocity of suspension motion of the standard Opala-
Maysmeter vehicle in response to roughness (mm/s)
V the vehicle speed (m/s)
ARS the average rectified slope (mm/m)
The limiting speed due to the effect of road roughness is calculated as:
ARVMAX
VROUGH = ...(3.16)
VROUGH a0 * R1av
where:
The default values for the roughness effects parameters are given in Table E2.5 (see Section
5.2.3).
The desired speed in the absence of posted speed limits is calculated as:
where:
XNMT speed reduction factor due to non-motorised transport for the section
(dimensionless, in the range 0.6 - 1; default = 1.0)
VDESMUL desired speed multiplication factor (dimensionless, in the range 0.85 - 1.3;
default = 1.0). The basic model is for two lane roads. This factor is used to
adapt the model for single lane roads and multi-lane highways
Since desired speeds are affected by road widths, the speed values (VDES) need to be
adjusted for these effects. The adjustment is based on the work carried out by Hoban et al.
(1994), which assumes that there is critical width (CW1) below which speeds will be
unaffected by carriageway width. Between this minimum speed (VDESMIN) and the desired
speed on two-lane roads (VDES2), there is a linear increase in the speed. On roads wider than
two-lanes, Yuli (1996) showed that there is a continued increase in speed, but at a much lower
rate.
The values of VDES are adjusted as follows:
if CW ≤ CW1 VDES = VDESMIN
where:
The speed/width slope between CW1 and CW2 is calculated from the expression:
( VDES2 - VDESMIN )
VDES_a3 =
(CW2 - CW1)
The relationship between the minimum desired speed and the desired speed on two-lane roads
is given as:
where:
VDES_a2 the ratio of the desired speed on a single lane road to the desired speed on a
The default values for the desired speed model parameters are given in Table E2.6.
(m/s)
A complete set of desired speed parameters is required for each of the four road surface
classes: bituminous, block, concrete, and unsealed.
The actual desired speed is the minimum of the desired speed and the enforced speed limit:
where:
S k = 3.6 * VS ku ...(3.19)
For downhill sections, the steady-state speed VSkd is calculated as explained in Section 5.1.2.
The average steady-state free speed (km/h) is given as:
S k = 3.6 * VS kd ...(3.20)
S1
S2
Speed km/h
S3
Snom
Sult
Qo Qnom Qult
Flow in PCSE/h
Qo the flow level below which traffic interactions are negligible in PCSE/h
Qnom nominal capacity of the road (PCSE/h)
Qult the ultimate capacity of the road for stable flow (PCSE/h)
Sult speed at the ultimate capacity, also referred to as jam speed (km/h)
Snom speed at the nominal capacity (km/h)
S1 to S3 free flow speeds of different vehicle types (km/h)
PCSE passenger car space equivalents (see Part B)
The model predicts that below a certain volume there are no traffic interactions and all
vehicles travel at their free speeds. Once traffic interactions commence, the speeds of the
individual vehicles decrease until the nominal capacity where all vehicles will be travelling at
the same speed, which is estimated as 85% of the free speed of the slowest vehicle type. The
speeds can then further decrease towards the ultimate capacity beyond which unstable flow
will arise.
The values of the key parameters that define the speed-flow relationship vary depending upon
the road type and width (see Part B), and are specified by the user. The free speeds of different
vehicle types (S1, S2, etc.) and the speed at nominal capacity (Snom) are computed internally
as described in Section 3 and Section 4.2, respectively.
For a given road section and for each analysis year, the modelling of traffic congestion is
carried out for each traffic flow period. This is done separately for the uphill segment and the
downhill segment, and the results are combined to give an average for a round trip over the
road section. Further aggregation of the results is performed to obtain annual average values
for each vehicle type and for all the vehicles using the road section.
The computational procedure can be summarised as follows:
1 Define the speed-flow relationship
by determining the speed at nominal capacity and using the free speeds calculated, for:
(a) Uphill segment
(b) Downhill segment
2 For each vehicle type k and for each traffic flow period p calculate:
(a) Steady-state congested speed for the uphill segment, the downhill segment, and for a
round trip (see Section 4.2)
(b) Steady-state operating speed (see Section 4.3)
(c) Acceleration noise - a measure of the severity of speed changes (see Section 4.6)
3 For each vehicle type k calculate the annual average operating speed
(see Section 4.4)
4 Calculate the annual average traffic speed
(see Section 4.5)
(a) For all vehicles
(b) For heavy commercial vehicles only
where:
The uphill segment speed, VU, at each traffic flow period p and flow Qp is calculated as
follows:
! for Qp < Qo
VU kp = VS ku ...(4.2)
! for Qo ≤ Qp ≤ Qnom
where:
VUkp steady-state congested speed of vehicle type k during period p for the uphill
segment (m/s)
Qp traffic flow during period p (PCSE/h)
CALBFAC speed calibration factor (default = 1.0, range 0.1 to 10). This is a road type
dependent translation factor for the speed-flow model
Here, the speeds Sult (km/h) are converted into VSult (m/s) by dividing them by 3.6.
The computed uphill segment speeds (VUkp) are used in the calculation of fuel consumption
(see Section 5).
where:
The downhill segment speed, VD, (m/s) at each traffic flow period p and flow Qp (PCSE/h) is
calculated as follows:
! for Qp < Qo
VD kp = VD kd ...(4.7)
! for Qo ≤ Qp ≤ Qnom
These speeds (VDkp) are used in the calculation of fuel consumption (see Section 5).
where:
VDkp steady-state congested speed of vehicle type k during period p for the downhill
segment (m/s)
Qp traffic flow during period p (PCSE/h)
CALBFAC speed calibration factor (default = 1.0, range 0.1 to 10)
7.2
S kp = ...(4.11)
1 1
+
VU VD
kp kp
where:
Skp average steady-state congested speed (km/h) for vehicle type k for traffic flow
period p
SS kp = S kp * SPEEDBIAS ...(4.12)
where:
where:
COV coefficient of speed variation within the traffic stream (default = 0.15)
These speed values (SSkp) are used for the calculation of vehicle utilisation, crew hours,
passenger travel time, and cargo transit time and for reporting purposes.
n
∑
p =1
HRYR p * HVp * SS kp
SS kav = ...(4.14)
n
∑
p =1
HRYR p * HVp
where:
∑ SS
k =1
kav
S= ...(4.15)
K
where:
The annual average speed of heavy vehicles is required for the modelling of pavement rutting
due to plastic deformation, and is calculated as follows:
KH
∑ SS
kH=1
kav
Sh = ...(4.16)
KH
where:
4.6.1 Concepts
The speed-flow model presented in Figure E2.2 shows that as flows increase, there is an
increase in vehicle interactions and a decrease in speeds. The interactions are accompanied by
an increase in the frequency and magnitude of vehicle accelerations and decelerations. Under
ideal conditions drivers would maintain a steady-state speed without any decelerations and
accelerations. However, this is not possible in reality since drivers are forced to adjust their
speeds in response to traffic congestion, road alignment, pavement surface condition, the
presence of NMT and other roadside activities.
The standard deviation of accelerations, referred to as the acceleration noise, gives an
indication of the severity of speed changes. Low values of acceleration noise indicate that
there are minor speed changes, large values indicate major speed changes. For each road
section, it is considered that the additional vehicle operating costs due to speed change cycles
is proportional to the magnitude of acceleration noise.
The total acceleration noise for a vehicle type k operating on a road section during traffic flow
period p is considered in two components:
1 Natural acceleration noise
σa kp = (σan k
2
+ σat kp
2
) ...(4.17)
where:
σakp total acceleration noise (m/s2) for vehicle type k in flow period p
σatkp traffic induced acceleration noise for vehicle type k in flow period p (m/s2)
3 Roadside friction
4 Non-motorised transport
The total natural acceleration noise (σank) for each vehicle type k is given by the following
expression:
[
σan k = MAX 0.1, MAX(σ adral k , σ asfk , σanmt k , σ airi k
2 2 2 2
)] ...(4.18)
where:
σadralk natural acceleration noise due to driver behaviour and road alignment (m/s2)
(default value = 0.1)
σasfk acceleration noise due to roadside friction
The driver behaviour acceleration noise (σadr) and the road alignment acceleration noise
(σaal) are combined into a single value (σdral) as it is difficult to differentiate between these
two components. The other three components of natural noise are modelled as linear
functions. This is done using the following relationships:
RIAMAX k
σairi k = MINRIAMAX k , RIav * ...(4.21)
AMAXRIk
where:
XFRI Speed reduction factor due to road side friction for the road section (0.6 to
1)
FRIAMAXk maximum acceleration noise due to side friction (default value = 0.20 m/s2
for all vehicle types)
XNMT Speed reduction factor due to non-motorised transport for the road section
(0.6 to 1)
NMTAMAXk maximum acceleration noise due to non-motorised transport (default value
= 0.40 m/s2 for all vehicle types)
RIAMAXk maximum acceleration noise due to roughness (default value = 0.30 m/s2 for
all vehicle types)
RIav average roughness for a road section (IRI m/km)
AMAXRIk roughness at which maximum acceleration noise RIAMAXk occurs (default
value = 20 for all vehicle types) (IRI m/km)
1.04
σat kp = σatmax k ...(4.22)
(
1 + exp a0 + a1 * VCR p )
where:
Qp
VCR p = ...(4.23)
Q ult
where:
The maximum traffic acceleration noise for each vehicle type can be determined by taking
measurements of total acceleration noise under severely congested conditions. This gives the
maximum total acceleration noise on the road section, which includes both natural
acceleration noise and the maximum traffic induced acceleration noise. The maximum traffic
acceleration noise for each vehicle type k is then calculated as:
σatmax k = ( 2
σamax k − σan k
2
) ...(4.26)
where:
where:
σamaxvk maximum acceleration noise for a vehicle type k, input by the user in
Vehicle Fleet (default = 0.75 m/s2 for all vehicle types, with a range 0.1 -
0.75)
σamaxgr the gradient adjusted maximum acceleration noise for a road section (m/s2)
To account for the changes in driver behaviour due to gradients effect, the maximum
acceleration noise for a road type (σamaxr) is adjusted using the following expression:
where:
Section 5.4 describes the application of acceleration noise within the RUE models.
! Driver behaviour
Changes in any of these parameters as a result of a project will result in a change in vehicle
operating costs.
For a given road section and for each analysis year, vehicle resource consumption is modelled
by considering the operation of each vehicle type under the conditions of each traffic flow
period, and the results are aggregated into annual totals. The prediction of vehicle resource
consumption is done for each vehicle type in the following order:
1 Fuel consumption (see Section 5)
5 Fuel consumption
! Engine drag – this is the power required to overcome internal engine drag (or friction)
! Accessory power – this is the power required to run the vehicle accessories such as the
cooling fan, power steering, air conditioner, alternator, etc.)
For each vehicle type, fuel consumption is calculated for each traffic flow period separately
for the uphill segment and for the downhill segment, the results are then averaged for a round-
trip over the road section.
The computational procedure can be summarised as follows:
1 For each vehicle type k and for each traffic flow period p calculate:
(a) Total power requirements of the engine for the uphill segment and the downhill
segment (see Section 5.2)
(b) Fuel-to-power efficiency factor for the uphill segment and the downhill segment
(see Section 5.3)
(c) Instantaneous fuel consumption for the uphill segment and the downhill segment.
This combines the steady-state fuel consumption with the additional fuel
consumption due to speed change cycles (see Section 5.4).
(d) Specific fuel consumption for a round trip over the road section (see Section 5.5)
2 For each vehicle type k calculate the annual average fuel consumption
(see Section 5.6)
where:
IFCkpu instantaneous fuel consumption of vehicle type k during traffic flow period
p (ml/s)
IDLE_FUELk idle rate of fuel consumption of vehicle type k (ml/s)
ZETAkpu uphill fuel-to-power efficiency factor of vehicle type k (ml/kW/s)
The expression steady-state implies that speed-change cycles are not considered. Thus, the
steady-state instantaneous fuel consumption for the uphill segment is calculated using
Equation 5.1 above with the value of dFUEL set to zero.
where:
IFCkpd instantaneous fuel consumption of vehicle type k during traffic flow period
p (ml/s)
IDLE_FUELk idle rate of fuel consumption of vehicle type k (ml/s)
ZETAkpd downhill fuel-to-power efficiency factor of vehicle type k (ml/kW/s)
PTOTkpd downhill total power requirement for steady-state motion (kW)
dFUELkpd additional fuel consumption factor due to vehicle speed-change cycles
The steady-state instantaneous fuel consumption for the downhill segment is calculated using
Equation 5.2 above with the value of dFUEL set to zero.
(FTR kp * Vkp )
PTR kp = ...(5.3)
1000
where:
PTRkp tractive power for vehicle type k during traffic flow period p (kW)
FTRkp total resistance to steady-state motion experienced by vehicle type k during
traffic flow period p (N)
Vkp Speed of vehicle type k during traffic flow period p (m/s)
! For calculating the tractive power (PTRkpu) for the uphill segment
where:
The subscripts u and d denote the uphill and downhill segments, respectively.
The components of total resistance to steady state motion are calculated as given below:
! Aerodynamic resistance
The default parameter values to calculate aerodynamic resistance for each vehicle type
are given in Table E2.4.
! Gradient resistance
The gradient resistance to motion is calculated separately for the uphill and downhill
segments using the following expression:
FG = WGT_OPER * g * GR ...(5.7)
where:
All the parameters are as previously defined.
For the uphill segment FGu use the positive value of GR, and for the downhill segment
FGd use the negative value of GR.
! Rolling resistance
(
FR = FCLIM * CR2 * b11 * NUM_WHEELS + CR1 * b12 * WGT_OPER + CR1 * b13 * Vkp
2
)
...(5.8)
2
WGT_OPER * V 2
MAX 0, − WGT_OPER * g * e
kp
R
FCV = ...(5.9)
(NUM_WHEELS * CS )
1000
where:
WGT_OPER
2
+ CS_a2 *
CS_a1 * WGT_OPER
CS = Kcs * CS_a0 +
NUM_WHEELS NUM_WHEELS
...(5.10)
where:
Kcs 1 1 1 1
! Inertial resistance
The inertial resistance is not included in the total tractive power for steady-state motion.
It is considered in the modelling of speed-change cycle effects described in Section 5.4.
The inertial resistance is calculated as follows:
EMRAT_a2
EMRAT = EMRAT_a0 + EMRAT_a1 * atan
...(5.12)
3
Vkp
where:
ACC 2
vehicle acceleration in m/s
Note that the inertial resistance to motion is considered to be zero under steady-state
conditions.
The default parameter values to calculate inertial resistance for each vehicle type are given in
Table E2.4.
where:
PENGACCSkp total engine and accessories power of vehicle type k during traffic flow period
p (kW)
Kpea calibration factor for total engine and accessories power (default 1.0)
PRATk the maximum rated engine power of vehicle type k (kW)
RPMkp engine speed (rev/min)
RPM_IDLE idle engine speed (rev/min)
RPM100 the engine speed calculated at 100 km/h (rev/min)
PACCS_a0 the ratio of engine and accessory drag to rated engine power when travelling
at 100 km/h
PACCS_a1 a model parameter
The parameter PACCS_a1 is related to the idle fuel consumption rate. It is calculated from the
user-supplied value for IDLE_FUEL as follows:
(100 - PCTPENG )
a = ZETAB * EHP * Kpea 2 * PRAT *
100
c = - IDLE_FUEL
PACCS_a1 =
(- b + b 2 - 4ac ) ...(5.14)
2a
where:
The engine speed (RPM) depends upon the vehicle speed, and it is calculated using Equations
5.15 below to 5.17 below with Vkp = VUkp for the uphill segment and Vkp = VDkp for the
downhill segment:
! If Vkp ≤ 5.6 m/s
! If Vkp ≤ RPM_a3
2
RPM kp = RPM_a0 + RPM_a1 * Vkp + RPM_a2 * Vkp ...(5.16)
else
RPM100 =
(RPM_a0 + RPM_a1 * RPM_a3 + RPM_a2 * RPM_a3 2 ) * 27.8 ...(5.19)
RPM_a3
The values for the above model are given in Table E2.8.
PTR kp
PTOTkp = + PENGACCS kp ...(5.20)
EDT
else
where:
PTOTkp total power requirement for steady-state motion by vehicle type k during
traffic flow period p (kW)
PTRkp total tractive power of vehicle type k during traffic flow period p (kW)
EDT drivetrain efficiency
Vehicle Engine speed model parameters Idle Idle Base fuel Decrease Rated Efficiency Engine and
number engine fuel efficiency in engine of the Accessories Power
speed rate efficiency power drivetrain
RPM_a0 RPM_a1 RPM_a2 RPM_a3 RPM_ IDLE_ ZETAB EHP PRAT EDT PACCS_ PCTPENG
IDLE FUEL a0
...(5.22)
...(5.23)
The mean speed for each vehicle type k and traffic flow period p is given by:
SU kp = 3.6 * VU kp ...(5.24)
The mean speed for each vehicle type k and traffic flow period p is given by:
SD kp = 3.6 * VD kp ...(5.25)
The total acceleration noise (σakp) for each vehicle type k and for each traffic flow period p
used for determining dFUEL is calculated as described in Section 4.6.
where:
The fuel consumption (litres per 1000 vehicle-kilometres) is thus given by the expression:
where:
The traffic stream is comprised of vehicles travelling at different speeds and, thus, different
fuel consumption rates. Since fuel consumption is non-linear with speed, the mean fuel
consumption does not correspond to the fuel consumption at the mean speed. The fuel
adjustment factor (FUELBIAS) is therefore applied to correct the bias introduced to the
analysis through the use of the mean speed instead of individual vehicle speeds. It is given by
the following expression:
∑ HRYR
p =1
p * HVp * FC kp
FC kav = n
...(5.31)
∑ HRYR
p =1
p * HVp
where:
where:
OILCAP
OILCONT = ...(6.2)
DISTCHNG
where:
The values in Table E2.9 are the defaults for oil consumption model values.
Light goods and delivery vehicle, mini-bus, 4WD 7500 5.0 0.0028
The annual average oil consumption (litres per 1000 vehicle-kilometres) is given by:
where:
7 Tyre consumption
(a) The circumferential, lateral and normal forces acting on a tyre for the idealised
uphill and downhill segments (see Section 7.2.2)
(b) The tyre energy for the uphill segment and the downhill segment (see Section 7.2.1)
(c) Tyre consumption per 1000 vehicle-kilometres for a round trip over the road section
(see Section 7.3)
2 For each vehicle type k calculate the annual average tyre consumption
(see Section 7.5)
1 + 0.01 * RREC k * NR k
EQNTkpu = + 0.0027 ...(7.1)
DISTOTk pu
where:
EQNTkpu number of equivalent new tyres consumed per 1000 veh-km for each wheel
during traffic flow period p
RRECk retread cost as a percentage of new tyre cost (default = 15 for all vehicle types)
NRk the number of retreads per tyre carcass
DISTOTkpu total distance travelled in the uphill direction by the tyre (1000s of kilometres)
during traffic flow period p
where:
NR0k base number of recaps specified by the user (default = 1.30 for all vehicle
types)
RImod modified value of the average road roughness (m/km), see Section 7.4
The total distance travelled by a tyre carcass during traffic flow period p is given by:
VOL k
DISTOTkpu = (1 + NR k ) ...(7.3)
TWTkpu
where:
TWTkpu the rate of tread wear (dm3/1000 veh-km) during traffic flow period p
VOLk volume of wearable rubber (dm3), see Table E2.10
1 + 0.01 * RREC k * NR k
EQNTkpd = + 0.0027 ...(7.4)
DISTOTk pd
where:
EQNTkpd number of equivalent new tyres consumed per 1000 veh-km for each wheel
during traffic flow period p
DISTOTkpd total distance travelled in the downhill direction by the tyre (1000s of
kilometres) during traffic flow period p
The total distance travelled by a tyre carcass during traffic flow period p is given by:
VOL k
DISTOTkpd = (1 + NR k ) ...(7.5)
TWTkpd
where:
The default values for C0tc and Ctcte are given in Table E2.10.
TE kp =
(CFT kp
2
+ LFTkp
2
) ...(7.7)
NFT
where:
where:
FCVkp
LFTkp = ...(7.9)
NUM_WHEELS
where:
! Normal force
WGT_OPER * g
NFT = ...(7.10)
NUM_WHEELS
where:
where:
TCkp number of tyres consumed per 1000 veh-km by vehicle type k during traffic
flow period p
MODFACkp tyre life modification factor for vehicle type k during traffic flow period p
VEHFACk vehicle type modification factor (see Table E2.11)
TYPEFAC tyre type modification factor (see Table E2.12)
CONGFACkp Congestion effects modification factor for vehicle type k during traffic flow
period p
The default values of these modification factors are supplied in Section 7.4.
The life of a single tyre of vehicle type k operating in traffic flow period p, TLIFE, (1000s
kilometres) is given as:
1000
TLIFE kp = ...(7.13)
0.5 * (EQNTkpu + EQNTkpd )
[EQNTkpu * NUM_WHEELS ]
TC kp = ...(7.14)
MODFAC kp
For downhill sections, the total tyre consumption expressed in terms of equivalent number of
new tyres per 1000 vehicle-kilometres for each vehicle type k, for each traffic flow period p, is
calculated from the expression:
[EQNTkpd * NUM_WHEELS ]
TC kp = ...(7.15)
MODFAC kp
Tyre life (TLIFEkp) is calculated using Equation 7.13 above with the denominator replaced
with EQNTkpu for the uphill segment and EQNTkpd for the downhill segment.
The values of adjusted average roughness (RImod) used in Equation 7.2 above, and of
VEHFAC used in Equation 7.12 above are given in Table E2.11.
(RImod)
1, 2, 3, 4, 5, 6, 7, 8, 12, 13 = RIav 2.0
9, 14 =7 1.0
The values of tyre type modification factor, TYREFAC, used in Equation 7.12 above is
given in Table E2.12.
The values of congestion effects factor (CONGFACkp) used in Equation 7.2 above are
obtained as follows:
❏ for Qp < Qo
CONGFAC kp = 1 ...(7.16)
❏ for Qo ≤ Qp ≤ Qnom
CONGFAC kp = 1 −
0.3 * (Qp - Qo)
...(7.17)
(Qnom - Qo)
CONGFAC kp = 0.7 −
0.2 * (Qp - Qnom)
...(7.18)
(Qult - Qnom)
∑ HRYR
p =1
p * HVp * TC kp
TC kav = n
...(7.19)
∑ HRYR
p =1
p * HVp
where:
TCkav annual average number of tyres consumed per 1000 veh-km by vehicle type k
HRYRp the number of hours in traffic flow period p
HVp the hourly traffic flow in period p expressed as a proportion of AADT
TCkp number of tyres consumed per 1000 veh-km by vehicle type k during traffic
flow period p
8.1 Utilisation
Vehicle utilisation is expressed in terms of the annual kilometreage driven during the annual
working time. Working time as defined by Hine (1996) is the time spent undertaking the
essential tasks of making a complete round trip, in normal circumstances. This excludes time
spent idle, when the driver is eating, sleeping or otherwise resting, but includes time spent
driving, loading, unloading and refuelling. In certain circumstances it may be appropriate to
include the time that the vehicle must spend waiting to move forward in a queue.
where:
AKM0 baseline average number of kilometres driven per year , input by the user
(km/year)
AKMVi average number of kilometres driven per vehicle of age i per year, input by the
user (km/year)
PCTVi percentage of vehicles of age i in the fleet (for i = 1, 2, ..., n)
where:
HRWK0 baseline average number of vehicle working hours per year, input by the user
(hrs/year)
HRWKVi number of working hours per vehicle of age i per year
PCTVi percentage of vehicles of age i in the fleet (for i = 1, 2, ..., n)
The user chooses which of the two methods should be used for calculating vehicle parts
consumption and for modelling capital costs.
LIFEKM0 * LIFEKMPCT
LIFEKM = ...(8.3)
100
where:
The baseline average vehicle service life is calculated from the expression:
where:
AKM0 baseline average number of kilometres driven per vehicle per year, input by
the user (km/year)
LIFE0 baseline average vehicle service life in years, input by the user
The optimal life as a percentage of the user defined baseline vehicle service life is given by:
100
LIFEKMPCT =
( )
...(8.5)
1 + exp a0 * RI adj a1
where:
RIadj the adjusted road roughness (IRI m/km), see Section 9.2
a0, a1 regression coefficients. The default values (for all vehicle types) are as
follows:
a0 = -65.8553
a1 = -1.9194
The default values for vehicle utilisation parameters are shown in Table E2.13.
9 Parts consumption
(a) The adjusted road roughness to be used in the model (see Section 9.2)
(b) The age in terms of cumulative number of kilometres driven (see Section 9.3)
2 For each vehicle type k and for each traffic flow period p calculate:
(a) The incremental change in parts consumption due to speed change cycles (see
Section 9.4)
(b) The parts consumption per 1000 vehicle-kilometres as a fraction of replacement
vehicle price
3 For each vehicle type k calculate the annual average parts consumption
(see Section 9.5)
The parts consumption for each vehicle type k, and for each traffic flow period p, is calculated
as follows:
...(9.1)
where:
PCkp parts consumption per 1000 veh-km, expressed as a fraction of the average
new (or replacement) vehicle price, NVPk,
CKM average cumulative number of kilometres driven per vehicle type (km)
KP the age exponent in parts consumption model
RIadj the adjusted road roughness (IRI m/km)
CPCONk incremental change factor in parts consumption due to vehicle speed change
cycles effects (default value = 0.10)
dFUELkp additional fuel consumption factor due to vehicle speed-change cycles (that is,
accelerations and decelerations)
The default values for the model parameters are shown in Table E2.14, and are based on the
default CKM values shown.
RIadj = [
MAX RIav , MIN(RI0, RIMIN + a2 * RIa3 ) ] ...(9.2)
where:
RI_SHAPE
a2 = RI0
...(9.4)
RI_SHAPE
RI0
RI0
a3 = ...(9.5)
RI_SHAPE
where:
where:
CKMk average cumulative number of kilometres driven per vehicle type k (km)
AKM0k baseline average number of kilometres driven per vehicle type k per year ,
input by the user (km/year)
LIFE0k baseline average service life of vehicle type k in years, input by the user
where:
3 User specifies the age spectrum distribution and the percentage of vehicles
If the user specifies the age spectrum distribution along with the percentage of vehicles at
each age i, then the value of CKMk is calculated from the following expression:
n
CKMk =
∑ AKMV * VEHAGE
i=1
i
100
* PCTV i i
...(9.8)
where:
The CKMk value is used in Equation 9.1 above to calculate parts consumption PCkp per 1000
vehicle-km during each traffic flow period p. Note that the value of CKMk is the same for all
the traffic flow periods.
where:
∑ HRYR
p =1
p * HVp * PC kp
PC kav = n
...(9.10)
∑ HRYR
p =1
p * HVp
where:
PCkav annual average parts consumption per 1000 veh-km, expressed as a fraction of
the average new vehicle price
HRYRp the number of hours in traffic flow period p
HVp hourly traffic flow in period p expressed as a proportion of AADT
PCkp parts consumption per 1000 veh-km by vehicle type k during traffic flow
period p, expressed as a fraction of the average new vehicle price
10 Labour hours
2 For each vehicle type k calculate the annual average labour hours
The number of labour hours per 1000 vehicle-kilometres is calculated for each vehicle type k,
and for each traffic-flow period p as follows:
(
LH kp = K0lh * a0 * PC kp
a1
) + K1lh ...(10.1)
where:
LHkp number of labour hours per 1000 veh-km of vehicle type k in traffic flow
period p
PCkp parts consumption per 1000 veh-km of vehicle type k during traffic flow
period p, expressed as a fraction of average new vehicle price
a0 constant term of the maintenance labour model
a1 parts exponent of the maintenance labour model
K0lh rotational calibration factor (default value = 1.0)
K1lh translational calibration factor (default value = 0)
The default values for the labour hours model parameters are shown in Table E2.15, and are
based on the default CKM values given in Table E2.14.
∑ HRYR
p =1
p * HVp * LH kp
LH kav = n
...(10.2)
∑ HRYR
p =1
p * HVp
where:
LHkav annual average number of labour hours per 1000 veh-km of vehicle type k
HRYRp number of hours in traffic flow period p
HVp the hourly traffic flow in period p expressed as a proportion of AADT
LHkp number of labour hours per 1000 veh-km for vehicle type k during traffic flow
period p
11 Capital costs
where:
CAPCSTkp capital cost per 1000 veh-km incurred during traffic flow period p
DEPCSTkp depreciation cost per 1000 veh-km incurred during traffic flow period p
INTCSTkp annual interest cost per 1000 veh-km incurred during traffic flow period p
The depreciation and interest costs are computed separately as described below.
! Depreciation cost is given as:
where:
DEPkp depreciation cost factor per 1000 veh-km for traffic flow period p
NVPLTk average new (or replacement) vehicle price less tyres
For each vehicle type k, the average new vehicle price less tyres (to avoid double counting) is
calculated from the expression:
where:
where:
INTkp annual interest cost factor per 1000 veh-km for traffic flow period p
NVPk average new (or replacement) vehicle price less tyres
11.2 Depreciation
Vehicle depreciation arises mainly due to use, time/ageing and technical obsolescence. There
are two methods of calculating depreciation costs based on vehicle service life:
1 Constant life method (see Section 11.2.2)
Both methods calculate depreciation cost over the service life of a vehicle using a straight-line
method. The residual value at the end of the vehicle’s service life is deducted from the vehicle
price before calculating depreciation.
where:
RVPLTPCT residual vehicle price less tyres at the end of its service life (%)
RIav average road roughness (IRI m/km)
a2 minimum residual value of the vehicle (%) (default = 2)
a3 maximum residual value of the vehicle (%) (default value = 15)
a4 average roughness, IRI, below which the maximum value arises (default
value = 5)
where:
For passenger cars the depreciation cost factor per 1000 vehicle-kilometres, DEPkp, for each
traffic flow period is calculated as follows:
! If PP (that is, the percentage of vehicle use on private trips) is greater than 50:
where:
Note that the value of DEPk is the same for all the traffic flow periods.
11.3 Interest
Interest costs are the opportunity cost of vehicle ownership. These consist of the income that
would have been received had the capital invested in the vehicle been invested elsewhere.
The interest cost factor per 1000 vehicle-kilometres INTkp for each traffic flow period is
calculated as follows:
1000 * AINVk
INTkp = ...(11.9)
2 * SS kp * HRWK0 k * 100
where:
AINVk annual interest charge on the purchase cost of the average new vehicle type k
(%)
∑ HRYR
p =1
p * HVp * CAPCSTkp
CAPCSTkav = n
...(11.10)
∑ HRYR
p =1
p * HVp
where:
CAPCSTkav average capital cost per 1000 veh-km for vehicle type k
HRYRp number of hours in traffic flow period p
HVp hourly traffic flow in period p expressed as a proportion of AADT
CAPCSTkp capital cost per 1000 veh-km for vehicle type k during flow period p
12 Crew hours
where:
CHkp number of hours per crew member per 1000 veh-km for traffic flow period p
PPk percentage of vehicle use on private trips (%)
SSkp vehicle operating speed (km/h) during traffic flow period p
∑ HRYR
p =1
p * HVp * CHkp
CHkav = n
...(12.2)
∑ HRYR
p =1
p * HVp
where:
CHkav average number of hours per crew member per 1000 veh-km for vehicle type
k
HRYRp the number of hours in traffic flow period p
HVp the hourly traffic flow in period p expressed as a proportion of AADT
CHkp number of hours per crew member per 1000 veh-km for vehicle type k
during traffic flow period p
13 Overhead costs
where:
OCkp overhead cost per 1000 veh-km for vehicle type k incurred during traffic
flow period p
OAk overhead cost per year, for vehicle type k, input by the user
PPk percentage of vehicle use on private trips
SSkp vehicle operating speed (km/h) during traffic flow period p
∑ HRYR
p =1
p * HVp * OC kp
OC kav = n
...(13.2)
∑ HRYR
p =1
p * HVp
where:
OCkav annual average overhead costs per 1000 veh-km for vehicle type k
HRYRp the number of hours in traffic flow period p
HVp the hourly traffic flow in period p expressed as a proportion of AADT
OCkp overhead costs per 1000 veh-km for vehicle type k during traffic flow period
p
1000 * PAX k * Wk
PWH kp = ...(14.1)
100 * SS kp
where:
PWHkp number of working passenger-hours per 1000 veh-km, for vehicle type k
during traffic flow period p
PAXk number of passengers (non-crew occupants) in vehicle type k
Wk percentage of passengers on work-purpose journey (%)
SSkp vehicle operating speed (km/h) during traffic flow period p
where:
PNHkp number of non-working passenger-hours per 1000 veh-km for vehicle type k
during traffic flow period p
PAXk number of passengers (non-crew occupants) in vehicle type k
Wk percentage of passengers on work-purpose journey (%)
∑ HRYR
p =1
p * HVp * PWH kp
PWH kav = n
...(14.3)
∑ HRYR
p =1
p * HVp
where:
PWHkav annual average number of working passenger-hours per 1000 veh-km, for
vehicle type k
HRYRp the number of hours in traffic flow period p
HVp the hourly traffic flow in period p expressed as a proportion of AADT
PWHkp number of working passenger-hours per 1000 veh-km, for vehicle type k
during traffic flow period p
∑ HRYR
p =1
p * HVp * PNH kp
PNH kav = n
...(14.4)
∑ HRYR
p =1
p * HVp
where:
1000
CARGOH kp = ...(15.1)
SS kp
where:
CARGOHkp annual number of cargo holding hours per 1000 veh-km, for vehicle type k
during traffic flow period p
SSkp vehicle operating speed (km/h) during traffic flow period p
The annual average number of cargo holding hours per 1000 vehicle-kilometres is calculated
as follows:
n
∑ HRYR
p =1
p * HVp * CARGOH kp
CARGOH kav = n
...(15.2)
∑ HRYR
p =1
p * HVp
where:
CARGOHkav annual average number of cargo holding hours per 1000 veh-km, for vehicle
type k
HRYRp the number of hours in traffic flow period p
HVp the hourly traffic flow in period p expressed as a proportion of AADT
CARGOHkp number of cargo holding hours per 1000 veh-km, for vehicle type k during
traffic flow period p
GH
FPASS k = 1 + (FPLIM k - 1) * MAX 0, 1 − ...(16.2)
GHMIN
where:
Note that the physical explanation for increasing FPASS for gravel thickness less than the
minimum is that there is greater risk of weak spots and of increased vehicle costs, in this range
than for roads with adequate gravel cover thickness.
The factor FPLIM ranges in value, from:
1 for subgrade materials with soaked CBR greater than 10% (that is, fully
passable)
to
3 for heavy vehicles on soft soils
A default value of 1.0 is used which the user may override. By definition, CPASS is zero for
paved roads, and GH is zero for earth roads.
KCOST jk * L j
TRIPCOST jk = ...(17.1)
1000
where:
TRIPCOSTjk cost per vehicle-trip over the road section under investment alternative j
KCOSTjk cost per 1000 vehicle-kilometres of vehicle type k under section alternative j
Lj road section length under investment alternative j (km)
The annual average costs per vehicle trip are required as inputs for economic analysis and
comparisons of different investment options. They are used for calculating the net benefits of
each pair of section options to be compared (see Part G).
Tyres Number of equivalent new tyres per 1000 vehicle- Cost per tyre
kilometres, TC
Parts Proportion of new vehicle cost per 1000 vehicle- Cost of new vehicle
kilometres, PC
Depreciation Fraction of new vehicle cost less tyres per 1000 Cost of new vehicle minus cost of
vehicle-kilometres, DEP tyres
Interest Fraction of new vehicle cost per 1000 vehicle- Cost of new vehicle
kilometres, INT
Passenger working Working passenger-hours per 1000 vehicle- Value per hour of passenger
time kilometres, PWH working time
Passenger non- Non-working passenger-hours per 1000 vehicle- Value per hour of passenger non-
working time kilometres, PNH working time
Cargo holding time Vehicle-hours per 1000 vehicle-kilometres, Cargo holding cost per vehicle-hour
CARGOH
18 References
ISOHDM Publications, (1994 - 1996)
International Study of Highway Development and Management Tools
University of Birmingham, UK
NDLI, (1995)
Modelling Road User Effects in HDM-4
Final Report to the Asian Development Bank (RETA: 5549)
N. D. Lea International Ltd., Vancouver, Canada
Bennett C.R., and Greenwood I.D., (1996)
Specifications for the HDM-4 Road User Effects Model. Third and fourth Draft
International Study of Highway Development and Management Tools
University of Birmingham, UK
Bennett C.R., (1996b)
Modelling Capital Costs HDM-4
Report to the International Study of Highway Development and Management Tools
University of Birmingham, UK
Bennett C.R., (1996c)
Modelling the effects of traffic congestion on fuel consumption in HDM-4: Overview of
methodology and specifications
Bennett C.R., (1998)
Structured Testing of the HDM-4 Road User Effects Model for Australia
Highway and Traffic Consultants Ltd., UK
Biggs D.C., (1988)
ARFCOM – Models for Estimating Light to Heavy Vehicle Fuel Consumption
Research Report ARR 152, Australian Road Research Board, Nunawading, Australia
du Plessis H.W., editor (1989)
An Investigation of vehicle Operating Cost Relationships for use in South Africa
NITRR Report DPVT-C96.1, CSIR, Pretoria
Harrison R., and Aziz S., (1998)
HDM-4 Tyre Relationships, Memorandum to ISOHDM Secretariat
Centre for Transportation Research, The University of Texas at Austin, USA
Hine J. L., (1996)
Briefing Paper on Vehicle Utilisation and Service Life for the December 1996 Workshop
on HDM-4, Unpublished Project Report, PR/ORC/590/96 R6472
For International Study of Highway Development and Management Tools
Overseas Unit, TRL, Crowthorne Berkshire, UK
Hoban C., Reilly W., and Archondo-Callao R., (1994)
Economic Analysis of Road Projects with Congested Traffic
World Bank Publications. Washington D.C.
E3 Non-Motorised Transport
1 Introduction
This chapter describes the implementation of Road User Effects (RUE) models for calculating
non-motorised vehicle speeds, operating costs and travel time (see Figure E3.1). It provides an
overview of the modelling concepts and logic, a description of the relationships, and default
parameter values for each of the RUE components and the HDM-4 representative Non-
Motorised Transport (NMT) types.
Road
RoadUser
UserEffects
Effects
Overview Vehicle
VehicleSpeeds
Speeds Non-Motorised
Non-Motorised Road
RoadSafety
Safety
Overview
and
andOperating
Operating Transport
Transport
Chapter
ChapterE1
E1 Chapter
ChapterE4
E4
Costs
Costs Chapter
ChapterE3
E3
Chapter
ChapterE2
E2
Users can define their own set of NMT vehicles within each class by calibrating the default
NMT class. Table E3.1 gives the default values of the key NMT characteristics required for
the analyses.
The calculation of NMT speeds utilises the performance and physical size of these NMT
types. The operating costs are calculated separately according to the utilisation of the NMT
vehicle (for example, private or commercial use, passenger or freight use).
(Bullock)
Wheel type Pneumatic Pneumatic Wood
Number of wheels, 2 3 2
NUM_WHEELS
Number of passengers, 1 3 0 1
PAX
These improvements would affect the performance characteristics of NMT; thereby affecting
NMT user costs and benefits in terms of the following:
! Travel speed and time
! Wear and tear of NMT vehicles and components
! Fares/user charges
! Degree of conflicts with MT traffic
! Accident rates
In modelling terms, the effects on NMT that need to be considered in the economic analysis of
road investments can be quantified separately as follows (PADECO, 1996):
2 MT operating costs
The first six effects (1 to 6) are modelled in HDM-4, and are discussed further in the
following sections. The NMT effects on road deterioration and maintenance, and NMT travel
demand are currently not considered.
NMT type
Bicycle Rickshaw Animal cart Pedestrian
0.10 0.10 0.07 0.10
4 NMT speeds
! Inclement weather
However, to capture the effects of all these factors would necessitate the formulation of a
complex NMT speed model and calibration procedure. A simplified speed model has
therefore been adopted, based on the minimum limiting velocity approach used in HDM-4
(Watanatada et al., 1987b). The steady-state speed of NMT vehicles is regarded as the
minimum of potential speed constraints generated by the interaction of road severity factors
with relevant characteristics of the vehicle. The limiting factors include road roughness,
desired speed, and the road gradient. With the exception of NMT traffic volume, all of the
factors given above have been considered either explicitly or implicitly in the speed model or
in the calculation of forces opposing motion and their impact on energy use.
The average speed for a round trip is calculated as the harmonic mean of the uphill and
downhill speeds from the expression:
7.2 * XMT
Sk = …(4.3)
1 1
VS + VS
ku kd
where:
For the analysis of one-way traffic, the average speed is calculated as follows:
! For uphill segments
The benefits to NMT users of improving traffic flow conditions, for example by separating
NMT and MT traffic flows can be assessed by changing the value of the parameter XMT in
Equations 4.3 above to 4.5 above.
4.3 VROUGH
The limiting speed due to road roughness and the associated ride severity on NMT is
estimated from:
where:
Table E3.3 shows typical default parameter values for the NMT speed-roughness model. It is
expected that a proportion of NMT traffic will use part of the carriageway and the shoulders
(if provided). This will need to be specified, so that carriageway roughness is applied to the
former and the shoulder roughness applied to the latter. The NMT benefits arising from
improvement of road surface condition can be assessed in terms of increased operating speeds
given by Equation 4.6 above.
(Bullock)
Desired speed on km/h 21.26 18.60 3.83 5.11
paved roads VDESp)
4.4 VGRAD
The effect of gradient on NMT speed is calculated separately for uphill and downhill travel.
For the uphill direction, the limiting speed due to gradient effects is given by:
where:
For the downhill direction, the limiting speed depends on the critical gradient (CRGR). It is
assumed that below the critical gradient, there is no effect of downgrade on NMT speed. The
downhill speed is therefore calculated as follows:
if: abs(GR ) > abs(CRGR )
otherwise:
where:
CRGRk the critical gradient for NMT type k (default = -0.04 for all NMT types, range:
-0.15 < CRGR < 0)
The average road section gradient (GR) is estimated from the following expression
(Watanatada et al., 1987a):
RF
GR = ± ...(4.10)
1000
where:
Table E3.3 includes typical default parameter values for the NMT speed-gradient model. The
benefits to NMT of improving the vertical alignment of roadway, measured in terms of
increased speeds, can be estimated using Equations 4.7 above and 4.8 above.
FG = WGT_OPER * g * GR ...(4.11)
where:
For the uphill travel direction, FGu is calculated using the positive value of GR, and for the
downhill travel direction, FGd is calculated using the negative value of GR. The subscripts u
and d denote uphill and downhill, respectively.
[ (
FR = FCLIM * CR2 * b1 * NUM_WHEELS + CR1 * b2 * WGT_OPER + b3 * v 2 )]
...(4.12)
where:
where:
Table E3.4 gives the default values for the various rolling resistance model parameters. The
diameter and number of wheels influence the rolling resistance parameters as follows:
WD_a1
b2 = ...(4.15)
WHEEL_DIA
WD_a2 * NUM_WHEELS
b3 = ...(4.16)
WHEEL_DIA 2
where:
The effect of inclement weather is incorporated in the rolling resistance model in a similar
manner to that for motorised vehicles. The rolling resistance factor depends upon the
percentage of time travelled on water covered roads (PCTDW), and if applicable, on snow
covered roads (PCTDS) as follows:
where:
TOCk total time and operating cost of NMT type k per veh-km
TMCk travel time cost of NMT type k (cost/km)
VOCk operating cost of NMT type k (excluding pedestrians) (cost/km)
where:
PAXVk
PAXC k = ...(5.3)
Sk
where:
PAXVk average hourly value of passenger time for NMT type k (cost/h), and is equal
to the number of passengers per vehicle (PAX) multiplied by the value of
passenger time (PTV)
Sk annual average speed of NMT type k (km/h)
CAGVk
CARGC k = ...(5.4)
Sk
where:
CAGVk average hourly value of cargo holding time for NMT type k (cost/h)
where:
where:
The depreciation cost per km, DEPC, is calculated using the expression:
PCHC k
DEPC k = ...(5.7)
LIFE0 k * AKM k
where:
Interest cost per km, INTC, is calculated from the following expression:
PCHC k * AINVk
INTC k = ...(5.8)
2 * 100 * AKM k
where:
AINVk the annual interest charge on the purchase cost of NMT type k (%)
For each analysis year, the annual kilometres travelled, AKM, is calculated as follows:
where:
where:
Table E3.5 gives the default values of the repair and maintenance cost model calibration
coefficients.
Table E3.5 Default parameter values for NMT repair and maintenance cost
model
Pedestrians
CRWVk
CRWC k = ...(5.12)
Sk
where:
CRWCk average crew wages per hour for NMT type k (cost/hr)
Sk annual average speed of NMT type k (km/hr)
where:
Energy consumption per NMT veh-km, ENUSD, is calculated for the uphill and downhill
directions of traffic flow using Equations 5.14 below and 5.15 below, respectively:
The average energy consumption for a round trip is then calculated using Equation 5.16
below, which incorporates an energy factor to account for energy used to overcome other
forces opposing motion:
where:
Walking
Replogle (1992) estimates the energy used in walking as 1.8 kJ/km/kg. This is equivalent to
144 kJ/km for an 80kg person. Thus, the average energy consumption used by walking is
estimated from the following expression:
5.7 Overheads
This covers all other cost elements including; taxes, administration, insurance,
parking/garaging, and any overheads associated with the crew (for example, training,
uniforms, etc.). The overhead cost per NMT vehicle-km, OVHD, is calculated using the
expression:
OHC k
OVHD k = ...(5.18)
S k * HRWK0 k
where:
where:
UTOCjsk annual average NMT total time and operating cost per vehicle-trip over the
road section s for vehicle type k under investment option j
TOCjsk total time and operating cost of NMT type k for section s under investment
option j (cost/km)
Ljs road section length under investment option j (km)
The types of outputs for Non-Motorised Transport are similar to that for Motorised Transport.
The outputs include the following:
! NMT speeds
! NMT traffic flows
! Time and operating cost
The amounts of energy consumption are considered as renewable energy in Energy Balance
analysis within Social and Environmental Effects module (see Part F).
7 References
Bennett C.R., (1996)
The New HDM-4 Model
Proceedings, Combined 18th ARRB
Transport Research Conference and Transit New Zealand Symposium, Christchurch,
New Zealand, Part 4 (Asset Management)
Bester C.J., (1981)
Fuel Consumption of Highway Traffic
Ph.D. Dissertation, University of Pretoria, South Africa
Biggs D.C., (1988)
ARFCOM – Models for Estimating Light to Heavy Vehicle Fuel Consumption
Research Report ARR 152
Australian Road Research Board, Nunawading, Australia
Cenek P.D., (1994)
Rolling Resistance Characteristics of New Zealand Roads
Transit New Zealand Research Report PR3-001
Wellington, New Zealand
CRRI, (1985)
Traffic Simulation Modelling Study: Part 1 – Development of Simulation Models
Central Road Research Institute, New Delhi, India
Greenwood I.D., and Bennett C.R., (1996)
The Effects of Traffic Congestion on Fuel Consumption
Road & Transport Research, Journal of Australian and New Zealand Research and
Practice, ARRB Transport Research, South Victoria, Australia, Vol. 5, No. 2, June 1996
Hoban C. J., (1987)
Evaluating Traffic Capacity and Improvements to Road Geometry
World Bank Technical Report, Paper Number 74
World Bank, Washington D.C., USA
Kerali H.G.R., Odoki J.B., and Collings S., (1997)
Energy Balance Framework for Road Transport Analysis
January 1998, Paper No. 980819
Transportation Research Board, Washington D.C., USA
NDLI (1995)
Modelling Road User Effects in HDM-4 - Final Report
Asian Development Bank Project RETA 5549
International Study of Highway Development and Management Tools
N.D. Lea International, Vancouver
Odoki J.B., and Kerali H.G.R., (1999)
Modelling Non-motorised Transport in HDM-4 - TRB, Paper No. 991129
Transportation Research Board, 78th Annual Meeting, Washington D.C., USA
E4 Road Safety
1 Introduction
This chapter describes the specification for Road Safety analysis in HDM-4 (see Figure E4.1).
The HDM-4 system allows users to define a series of look-up tables for accident rates. These
are basically broad, macro descriptions of the expected accident rates which can be defined in
several different ways according to a particular set of road and traffic attributes (for example,
road type, traffic level and flow pattern, presence of Non-Motorised Transport (NMT), and
geometry class). This tabular approach to implementing road safety analysis in HDM-4 was
recommended (ISOHDM, 1995), following a detailed review of various road safety studies,
modelling and analysis methods.
Road
RoadUser
UserEffects
Effects
Overview Vehicle
VehicleSpeeds
Speeds Non-Motorised
Non-Motorised Road
RoadSafety
Safety
Overview
and
andOperating
Operating Transport
Transport
Chapter
ChapterE1
E1 Chapter
ChapterE4
E4
Costs
Costs Chapter
ChapterE3
E3
Chapter
ChapterE2
E2
2 Modelling logic
An accident is considered as fatal if death occurs within a fixed period (for example, 31
days) following the accident. The fixed period may vary from one country to another.
! Injury
ACCYR
ACCRATE = ...(2.1)
EXPOSURE
where:
The annual exposure is expressed in terms of one hundred million vehicle-kilometres as:
365 * AADT * L
EXPOSSEC = ...(2.2)
10 8
! Road intersections
The annual exposure is expressed in terms of one hundred million vehicles as:
365 * AADTin
EXPOSINT = ...(2.3)
10 8
where:
For each road type or intersection type, the user is required to enter the rate for each accident
severity (that is, fatal, injury or damage only), in terms of the number of accidents per 100
million vehicle-kilometres or per 100 million vehicles. For analyses at the aggregate data
level, the user has the option of entering a single rate for All accident types. This value is
equal to the sum of the values of the rates for the three accident types. When a road is
improved (that is, providing separate NMT lanes, widening of road shoulders) a change to the
road type or intersection type occurs, and different accident rates need to be defined for the
new road type. Thus, it is possible to analyse the change in accident rates and costs resulting
from improving a road section or an intersection node.
Note: Modelling of intersection nodes is not included in this release.
The annual average daily traffic (AADT) on a road section, in vehicles per day, (or the
total AADT entering an intersection) and the growth rates.
! Length of road section
! Accident rates
Different accident rates may apply in different analysis years depending upon the
changes in road type (or intersection type) resulting from the road works carried out over
the analysis period.
! Unit costs of accidents
The unit costs can be defined either by accident types, or a weighted mean value may be
specified for All accident types.
5 Store results for economic analysis, comparisons and the reporting phase
where:
where:
ACj = ∑ ACCOST
i
i ...(2.6)
where:
where:
where:
∆ACCYR(m -n)i the annual difference in the number of accidents (for accident type i)
3.3 Outputs
The standard reports for safety analysis are supplied in Applications Guide. These include:
1 Annual number of accidents, by accident type, for each investment option
3 Net annual streams of accident numbers for each pair of investment options
4 References
ISOHDM, (1995)
Contents
F3 Vehicle Emissions
1 Introduction F3-1
2 Emissions model F3-2
2.1 Types of pollutants F3-2
2.2 The relationships F3-2
3 Modelling logic F3-8
Analytical
AnalyticalFramework
Framework
and
andModel
Model
Descriptions
Descriptions
Introduction
Introduction
Part
PartAA
Traffic
Traffic
Part
PartBB
RD
RDModel WE
WEModel RUE SEE
SEEModel
Model Model RUEModel
Model Model
Part
PartCC Part
PartDD Part Part
PartFF
PartEE
Economic
EconomicAnalysis
Analysis
RD = Road Deterioration Part
PartGG
Nomenclature
Nomenclature
WE = Road Works Effects Part
PartHH
F1 Overview
1 Introduction
The Social and Environmental Effects (SEE) models (see Figure F1.1) are for the analysis of:
! Energy balance (see Section 2)
It is widely recognised that energy and environmental effects need to be considered in the
assessment of alternative investment policies and projects. By adopting projects and policies
that minimise total life cycle energy use and vehicle exhaust emissions, related benefits such
as reduced vehicle operating costs, reduced pollution, reduced dependency on imports of
energy and reductions in the balance of payments deficits can be maximised. Planners and
decision makers need to be able to understand the energy implications and environmental
impacts of alternative road transport projects and policies.
Social
Socialand
and
Environmental
EnvironmentalEffects
Effects
Overview Energy
EnergyBalance Vehicle
Overview Balance VehicleEmissions
Emissions
Analysis
Analysis
Chapter
ChapterF1
F1 Chapter
ChapterF3
F3
Chapter
ChapterF2
F2
The assessment of alternative investment policies and projects requires that the various impact
measures be translated or reduced into common policy-sensitive units, which can be
considered under a framework of multi-criteria analysis. It is expected that the evaluation of
other aspects of environmental impacts that arise from road transport (for example, noise
pollution, damage to crops and buildings, etc.) will be included in a later release.
2 Energy balance
The energy used in the road transport sector forms a significant share of the total energy
consumption in most countries. The impact of road investment strategies and projects, and of
road transport policies on energy use has become an important aspect of the appraisal process.
The appraisal of road transport projects is based primarily on the assessment of economic
benefits which are estimated by comparing the total discounted costs calculated for a base
case against that for the project case. This is essentially an economic balance analysis. A
similar analytical framework is implemented for comparing the total energy used by different
modes of road transport for the base case and project case. The proposed Energy Balance
framework is used to calculate:
! Life-cycle energy consumption
Total life-cycle energy consumption at both project and network level analyses of road
investment policies.
! Differences in consumption
This permits comparisons to be made between different modes of road transport and thereby
influence policy on long term investment in the road sector (Kerali et al., 1998).
The analytical framework, for assessing the energy implications of road investment projects
and strategies, considers a number of parameters. These have been classified into three energy
use categories:
! Energy used by motorised vehicles
The energy consumption associated with the use of motor vehicles can be split into two
broad categories:
❏ energy used to produce and deliver both the fuel and vehicle
❏ energy used to power and operate the vehicle
Together, the energy used in each of these categories constitutes the full life-cycle energy
consumption. The energy consumed by motorised vehicles depends on a wide range of
factors, related to:
❏ vehicle size
❏ weight
❏ design
❏ age
❏ road characteristics and condition
❏ traffic characteristics
The energy used by motorised vehicles is generally from non-renewable sources. The
following resource components, used in vehicle operation are included in the energy
balance analysis:
❏ fuel consumption
❏ lubricating oil consumption
❏ tyre wear
❏ vehicle parts consumption
! Energy used by non-motorised vehicles
Non-motorised modes of transport (NMT) account for the vast majority of the movement
of people and goods in many countries. For this reason the inclusion of NMT in the
appraisal of transport projects and policies in developing countries is essential. For
example the presence of NMT can influence the speed of motorised transport, thereby
affecting the operating costs of motorised vehicles. In addition, policies such as road
improvements influence the costs and benefits to both motorised and non-motorised road
users. The energy used by NMT is calculated separately for animal-drawn carts, cycle-
rickshaws, bicycles and pedestrians.
! Energy used during the construction and maintenance of road networks
This is a significant aspect of the complete energy balance picture for road transport
investments. Thus, when comparing the energy implications of alternative policy or
project alternatives, it is important that this type of energy use is considered.
The energy balance framework implemented provides an efficient and neutral mechanism for
assessing the benefits of road transport investments. This avoids the possible distortion in
economic balance methods that may be influenced by fuel prices. Whilst some of the
parameters have had to be estimated from a desk study, the concept of energy balance
framework can be applied to road investment appraisal using a similar method to that used in
economic analyses.
3 Vehicle emissions
The objective of modelling vehicle emissions is to assess the effects (Hammerstrom, 1995), in
terms of pollutant quantities and changes in the following:
! Road characteristics
! Traffic congestion
! Vehicle technology
The model predicts the different components of vehicle exhaust emissions as a function of
fuel consumption. Fuel consumption is a function of vehicle speed, which in turn depends on
road characteristics and the characteristics of the vehicle itself. Thus, it is possible to analyse
the change in emissions effects as a result of implementing different road maintenance and
improvement options, or the implications of major changes to the vehicle fleet (for example,
due to improved vehicle technology). The different components of emissions modelled are:
! Hydrocarbon
! Carbon monoxide
! Nitrous oxide
! Sulphur dioxide
! Carbon dioxide
! Particulates
! Lead
In this release, the effects of exhaust emissions are not costed for inclusion in economic
analysis of road investments, only the net differences in the quantities of pollutants are
assessed for each pair of investment options. It is intended that the scope of vehicle emissions
modelling will be extended (Collings and Watkiss, 1998) to include impacts on:
! Air quality
! Health
! Damage costs
! Global warming
4 Applications scope
The analysis of energy balance and environmental impacts has two levels of application:
1 Project level analysis
Allows users to compare the life-cycle energy and emissions implications of a range of
project alternatives with the base line, do-minimum, alternative. The results of these
comparisons will assist in deciding which road investment alternatives should be
implemented.
2 Network level analysis
Enables decision makers to understand the energy and emissions implications of broader
transport policy objectives which have an impact on specific road networks (for example,
urban roads) as well as at the national level (for example, rural roads).
5 References
Collings S.A., and Watkiss P.R., (1998)
Development of Environmental Impacts and Energy Balance Models for HDM-4
Final project report for the Department for International Development, UK
Hammerstrom U., (1995)
Proposal for a Vehicle Exhaust Model in HDM-4
ISOHDM Supplementary Technical Relationships Study Draft Report
Swedish National Road Administration, Road and Traffic Management Division,
Borlange, Sweden
Kerali H.R., Odoki J.B., and Collings S.A., (1998)
Energy Balance Framework for Road Transport Analysis
Transportation Research Board, Paper no. 980819, Washington D.C., USA
1 Introduction
Social
Socialand
and
Environmental
EnvironmentalEffects
Effects
Overview Energy
EnergyBalance Vehicle
Overview Balance VehicleEmissions
Emissions
Analysis
Analysis
Chapter
ChapterF1
F1 Chapter
ChapterF3
F3
Chapter
ChapterF2
F2
The HDM-4 system can be used for assessing technical, economic, social and environmental
impacts of road investment. The assessment of alternative investment policies and projects
requires that the various impact measures be translated or reduced into common policy-
sensitive units. Examples of policy-sensitive units include monetary value, amount of time,
safety level, quantity of pollutant, and energy used.
Economic analyses use monetary values of resources in order to derive economic indicators
that serve in decision making on different alternatives for road investments. However, the
results of economic analyses depend to some extent on the relative monetary values placed on
the different components of vehicle resources. For example, a project whose justification is
primarily dependent on savings in petroleum consumption shows relatively less economic
benefits in an oil rich country than it would in a country which imports its fuel at a high cost.
Viewing the benefits of such projects in terms of their efficiency in energy use would place
them on an equal footing, and this yields more useful information to planners and decision-
makers.
This chapter describes a methodology for assessing the energy implications of different road
transport projects and policies within the context of the HDM-4 model. The assessment
technique is referred to here as Energy Balance analysis.
The project analysis allows users to compare the life-cycle energy implications of a range of
project alternatives with the base line, do-nothing alternative. The results of these
comparisons assist in the decision of which investment project alternatives to implement. For
example; the decision over which maintenance techniques to use for a particular stretch of
road is influenced by the trade off between:
! Energy benefits
The energy benefits gained from lower vehicle fuel consumption, oil use and tyre wear,
due to the improved flow characteristics of the road, and
! Energy requirements
2 Analytical requirements
Vehicles have been categorised as Motorised Transport (MT) and Non-Motorised Transport
(NMT). The development of a model framework for assessing the energy implications of road
investment projects and strategies requires the consideration of a number of parameters.
These have been broadly classified into three energy use categories:
1 Energy used by motorised vehicles (see Section 2.1)
3 Energy used during road construction and maintenance (see Section 2.3)
Sections 2.1, 2.2 and 2.3 describe the parameters that need to be considered for each category
of energy use.
Together, the energy used in each of these categories encompasses the full life-cycle energy
consumption. Table F2.1 shows the various life-cycle stages that could be considered in
determining the full life-cycle energy use.
Table F2.1 Energy use stages associated with vehicle production and use
! Processing
! Fuel distribution
! Processing
! Component manufacture
! Component transportation
! Assembly
! Vehicle distribution
! Oil consumption
! Tyre wearing
! Distribution
The prediction of a MT vehicle’s energy consumption under specific driving conditions can
be considered in two ways; either by using:
1 A mechanistic approach based on physical and mechanical first principles, or
2 Actual measured data to give energy use factors for a range of vehicles operating
under different conditions
The former mechanistic approach provides a greater degree of accuracy but has far greater
data requirements, whereas the energy use factor approach can be applied relatively easily.
Fuel consumption
The fuel consumption model follows a purely mechanistic approach that allows:
! The flexibility to model individual vehicle and road characteristics
! The ability to alter the model as new technologies are introduced
The mechanistic approach has been well researched, providing a theoretical and experimental
justification for all of the numerical assumptions that are made for each of the parameters. The
standard default vehicles are based on those using petrol and diesel fuel.
Suitable models for other vehicle technologies using different fuel types (for example,
electricity, LPG, CNG, etc.) are still being developed. For analyses involving comparison with
these vehicles, the energy consumption of alternative fuelled vehicles can be estimated by
applying scale factors which relate the calculated energy consumption of petrol or diesel
vehicles to the energy consumption of various alternative fuelled vehicles.
The fuel consumption model does not include the treatment of cold starts on fuel
consumption. In order to provide the needed mixture strength for satisfactory combustion at
the start of a journey, a large excess of fuel must be delivered to compensate for the
condensing losses due to the cold engine. This is more of a problem for petrol vehicles than
for diesel.
For the majority of road project appraisals the cold start issue is not considered to be
important, with the assumption being made that all journeys are run under hot engine
conditions. However, the influence of cold starting on fuel consumption may be of particular
significance in relation to some transport policies, particularly in urban areas where there are a
great number of shorter trips.
To overcome the problem of cold starts, scale factors relating the calculated fuel consumption
under normal hot conditions to fuel consumption under cold conditions can be used. The
proportion of the journey, which is run under cold conditions, can be estimated using the
CORINAIR methodology (Eggleston et al., 1993) which takes into account the average trip
length of the vehicle journeys and the average ambient temperature.
The average fuel use per vehicle kilometre for a full trip is calculated from the sum of the cold
start fuel use and hot conditions fuel use, weighted by the proportion of journey run cold and
hot, respectively.
Different transport fuels have different calorific values. Thus in order to compare like with
like the average fuel use per vehicle kilometre should be converted to energy factors per
vehicle kilometre using the energy content values given in Table F2.2.
Diesel 38.7
LPG1 25.5
2
CNG 40
Ethanol 23.9
Methanol 18.1
Biodiesel 32.8
Notes:
1 Assumes 90% propane, 10% butane
2 Units are MJ/m3
Tyre wear
The tyre consumption of a vehicle is proportional to the energy requirements (see Watanatada
et al., 1987). It is calculated using a model based on slip energy theory.
Tyre consumption, in terms of number of tyres per 1000 vehicle kilometres, is calculated for
light and heavy vehicles. These figures can be converted to an energy value using a
conversion factor of 32 GJ/tonne of tyres (Department of Trade and Industry, 1996). The
weight of tyres can be estimated using the factors given in Table F2.3.
12 Mini-bus 4.0
16 Coach 11.2
Fuel production
Figure F2.2 illustrates the stages that must be considered for petrol and diesel fuel production.
The energy use values from each stage of the fuel cycle are calculated on an-energy delivered
basis, and then aggregated to give the total energy use per unit of energy delivered for the full
fuel-cycle. In this way energy use for fuel production can be calculated per vehicle kilometre.
Feed
Feedstock
stock Processing
Processing Distribution
Distribution End-use
End-use
Petrol and
Pipeline Oil refinery diesel road Diesel vehicle
tankers
Gas
Crude oil Service
separation Petrol vehicle
tanker station
plant
The energy use associated with each of the fuel production stages (shown in Figure F2.2 for
petrol and diesel) varies considerably from country to country. However, if no local data
exists and neither do the resources to carry out a fuel-cycle analysis, then default data such as
that shown in Table F2.5 could be used to estimate the energy use associated with fuel
production. Fuel production energy use on a vehicle kilometre basis is derived from the
energy consumption during vehicle use.
Diesel 0.122
1
LPG 0.122
CNG 0.061
2
Electric 2.857
Biomethanol 0.514
Bioethanol 0.510
Biodiesel 0.655
Notes:
1 Assumes 40% of LPG comes from refineries and 60% is directly extracted
2 Assumes a higher calorific value (HCV) average generating efficiency of 35%
Vehicle manufacture
Data concerning the energy used to manufacture and deliver a motor vehicle is very difficult
to obtain due to the commercial sensitivities of the motor manufacturing industry. A recent
survey of what data is available revealed a typical value of 100 GJ per medium-sized car with
a weight of 1 tonne (ETSU, 1995). This represents approximately 15% of total life-cycle
energy use and is therefore a significant part of the overall energy balance.
To estimate the energy used to manufacture other vehicle types, a first order approximation
can be obtained by scaling 100 GJ by the ratio between the weight of a medium sized car and
the weight of the other vehicle. Based on this approach Table F2.4 shows some estimates of
energy use for vehicle production for each of the 16 standard vehicle types. In addition, it
shows estimates of total life-time vehicle kilometres which can be used to calculate default
values for vehicle production energy use on a per vehicle kilometre basis. The user can input
their estimates of vehicle service life in the Vehicle Fleet folder.
The energy used in the production process of each vehicle can simply be divided by the
average total lifetime vehicle kilometres to give vehicle production energy use per vehicle
kilometre.
Data on energy use by animals is sparse. As a broad assumption, a relationship derived for
human walking activity may be used:
where:
θ slope gradient
Table F2.6 shows a range of energy use factors derived using this formula. The animal and
load weights have been chosen to reflect those typically seen in certain parts of the world. For
example the weight of a laden cart pulled by two oxen (common in more humid zones of West
Africa (Starkey, 1993) is estimated to be 1000kg (Dennis, 1995). An ox could be assumed to
weigh 400kg. For comparison a donkey pulling a 100 kg laden cart, common in the drier
zones of West Africa, weighs around 150 kg (Starkey, 1993). A pack animal, commonly used
in hilly areas, would tend to carry loads in the region of 30-70 kg.
The values in Table F2.6 have assumed a rolling resistance for carts of 0.04 (Dennis, 1995),
an average speed of 6.4 km/h and a zero gradient.
Table F2.6 Typical energy use factors for animal transport (kJ/km)
2.2.2 Cycling
There is over 800 million bicycles worldwide (United Nations, 1993). Of these, approximately
400 million are in Asia, with 300 million in China (Replogle, 1992).
Published figures regarding the energy used in cycling vary widely. Naturally, the specific
energy use depends upon the weight of the rider and bicycle, the friction due to the cycle and
the speed of motion. Table F2.7 shows a range of cycling energy use factors quoted by Lewis
(1995).
Weight Speed
(kg) (km/h)
19.2 24 27.2 30.4
2.2.3 Walking
Replogle (1992) estimates the energy used in walking as:
1.8 kJ/km/kg
This is equivalent to 144 kJ/km for an 80-kg person (see Chapter E3).
However, for the purpose of energy balance analysis, such a detailed treatment is not
appropriate. Instead, a framework is implemented using aggregate level data for the energy
used in performing each of the different types of road works modelled. For example, an
average representative value of energy used per cubic metre of overlay can be specified. This
value is then multiplied by the total quantity of overlay performed on the road section to give
the total energy used. It is also necessary to distinguish between labour intensive works and
mainly mechanised works.
The average energy use factors are combined with the estimated total annual vehicle-
kilometres operated by each mode to give total annual energy use for the policy or measure
being considered. These totals are then used to give a range of indicator results for
comparative analysis.
3.2.1 Fuel
Vehicle fuel use factors are split into factors for hot and cold operation. The fuel use per
vehicle kilometre under hot conditions comes as output from the RUE (Road User Effects)
model.
The cold start fuel use is related to the basic hot fuel use, the ambient temperature and the
average vehicle trip length. The level of cold start fuel use is related to the hot fuel use by
factor CRAT, which is a function of the ambient temperature. This is expressed as follows:
where:
FCOLDk cold start fuel use for vehicle type k (litres per km)
CRATk cold start ratio at a given ambient temperature
FHOTk hot fuel use for vehicle type k (litres per km)
where:
FCkav annual average fuel consumption for vehicle type k (litres per 1000 vehicle-
km)
The cold start ratios (CRAT) are for passenger cars of petrol and diesel technologies, but are
also applicable to two-wheelers and vans. These ratios can be estimated from the following
CORINAIR relationships (Eggleston, 1993):
where:
Buses and trucks are usually considered to operate permanently under hot conditions, to a
good approximation, since their average trip lengths are very high.
This fuel use under cold conditions only occurs in the initial stages of a journey. The
proportion of any journey run cold is calculated as follows:
where:
Taking into account the proportion of the journey run cold, the average vehicle fuel use per
kilometre for the full trip can be calculated as follows:
where:
FAVEk average fuel use for vehicle type k (litres per kilometre)
The average vehicle fuel use, FAVE, is converted to energy use factor per kilometre by
applying the energy content of fuel given in Table F2.2. Thus,
where:
where:
3.2.3 Tyre
The annual average number of equivalent new tyres consumed per 1000 vehicle-kilometres
output from the RUE (Road User Effects) model is converted to energy use factor using the
energy conversion factor given in Section 2.1.1 as follows:
where:
where:
and:
ENVPROD k
ENVPk = …(3.11)
LIFEKM k
where:
ENVPRODk total energy used in the production of vehicle type k (MJ) (see Table F2.4)
LIFEKMk the predicted vehicle service life (km)
where:
EGLICYk annual average life cycle energy use factor for vehicle type k (MJ/km)
FPf fuel production factor of fuel type f (MJ per MJ of fuel used), defaults are
shown in Table F2.5
The annual average global energy use per passenger-kilometre is calculated as:
EGLICYk
EGPAXKM k = …(3.13)
PAX k
where:
EGPAXKMk annual average global energy use per passenger-kilometre for vehicle type k
(MJ/passenger-km)
PAXk average number of passengers per vehicle type k
EGLICYk
EGGDSKM k = …(3.14)
PAYLD k
where:
EGGDSKMk annual average global energy use per tonne-km for vehicle type k (MJ/tonne-
km)
PAYLDk average payload per vehicle type k (tonnes)
The average payload for each vehicle type k is calculated from the difference between the
average operating weight and the tare weight as follows:
where:
ENLICYk annual average national life cycle energy use factor for vehicle type k (MJ/km)
PNPk proportion of parts for vehicle type k produced within the country (as a
fraction)
PNVk proportion of vehicle type k produced within the country (as a fraction)
PNFf proportion of fuel type f produced within the country (as a fraction)
The annual average national energy use per passenger-kilometre is calculated as:
ENLICYk
ENPAXKM k = …(3.17)
PAX k
where:
ENPAXKMk annual average national energy use per passenger-km for vehicle k
(MJ/passenger-km)
PAXk average number of passengers per vehicle type k
ENLICYk
ENGDSKM k = …(3.18)
PAYLD k
where:
ENGDSKMk annual average national energy use per tonne-km for vehicle type k
(MJ/tonne-km)
PAYLDk average payload per vehicle type k (tonnes)
where:
EGLICYk annual average global life cycle energy use factor for NMT type k (MJ/km)
ENUSDk average energy consumption for NMT type k (Joules/km)
ENLICYk annual average national life cycle energy use factor for NMT type k (MJ/km)
where:
The annual total global energy use is then the sum of the energy use for each vehicle type k
(of MT and NMT, for k = 1, 2, …, K) plus the energy used for road works performed on the
particular section(s) in that analysis year:
K
EGTOT = ∑ EGLOB
k =1
k + ENROAD …(3.22)
where:
The energy used for road construction and maintenance in the given analysis year is calculated
from the following equation:
W
ENROAD = ∑ QTY
w =1
w * WEFw …(3.23)
where:
The total global energy used over the analysis period for each investment option is given by
the expression:
Y
GLOENGY = ∑ EGTOT
y =1
y …(3.24)
where:
GLOENGY total global energy use over the analysis period (MJ)
EGTOTy total global energy use in analysis years y (y= 1, 2, . . Y) (MJ)
where:
The annual total national energy use is then the sum of the energy use for each vehicle type k
(of both MT and NMT, for k = 1, 2, …, K) plus the energy used for road works on the
section(s) involved in that analysis year:
K
ENTOT = ∑ ENAT
k =1
k + ENROAD …(3.26)
where:
The total national energy used over the analysis period for each investment option is given by
the expression:
Y
NATENGY = ∑ ENTOT
y =1
y …(3.27)
where:
NATENGY total national energy use over the analysis period (MJ)
ENTOTy total national energy use in analysis years y (y= 1, 2, . . Y) (MJ)
ERNWi = ∑ EYi
k∈NMT
k …(3.28)
where:
EYik annual energy use for NMT vehicle type k (MJ) (that is, EGLOBk or ENATk)
i global (g) or national (n)
The total renewable energy (RNWTEi) used over the analysis period for each investment
option is obtained by summing ERNWi over the years.
The annual non-renewable energy is calculated as:
ENONRWi = ∑ EYi
k∈MT
k …(3.29)
where:
The total non-renewable energy (NORNTEi) used over the analysis period for each
investment option is obtained by summing ENONRWi over the years.
Indicators are computed and implemented in the reports for each comparison of investment
options m and n. The indicators are changes in the:
! Annual average global and national life-cycle energy use factors
! Total global and national non-renewable energy use over the analysis period
4 References
Dennis R., (1995)
IT Transport, Personal Communication
Department of Trade and Industry, (1996)
Digest of United Kingdom Energy Statistics, 1996
HMSO, London, UK
Eggleston H.S., Gaudioso D., Gorissen N., Jourmard R., Rijkeboer R.C., Samaras Z., and
Zierock K-H., (1993)
CORINAIR working group on emissions factors for calculating 1990 emissions from
road traffic - Volume 1: Methodology and emissions factors, B4-3045 (91) 10PH
Commission of the European Communities, Brussels, Belgium
ETSU, (1995)
Life-Cycle Analysis of Motor Fuel Emissions - Final report to COST 319 Sub-group
A4.C. ETSU Ref: RYCA/18691001/Issue 1
ETSU, (1996)
Alternative Road Transport Fuels - A Preliminary Life-Cycle Study for the UK
HMSO, London, March 1996
Hughes P.S., (1992)
A Strategy for Reducing Emissions of Greenhouse Gases from Personal Travel in Britain
PhD Thesis Open University, 1992
Lewis C.A., (1995)
Energy Use in Bicycle and Animal-Drawn Transport in Developing Countries
ETSU Working Paper 18400304, 1995
Maibach M., Seiler B., and Seiler P.D., (1995)
Okoinventar Transporte
Verlag Infras, Zurich
Replogle M., (1992)
Non-Motorised Vehicles in Asian Cities
World Bank Technical Paper 162
Washington D.C., 1992
Starkey P., (1993)
Animal-Powered Transport in Africa
Appropriate Technology, Vol.20 No.1, June 1993 pp 9-10
United Nations, (1993)
Energy Efficiency in Transportation
Alternatives for the Future, 1993
Watanatada T., Harral C.G., Paterson W.D.O., Dhareshwar A.M., Bhandari A., and
Tsunokawa K., (1987)
The Highway Design and Maintenance Standards Model - Volume 1 Description
World Bank, John Hopkins University Press
F3 Vehicle Emissions
1 Introduction
This chapter describes the implementation of Vehicle Emissions analysis in HDM-4 (see
Figure F3.1). The objective of modelling vehicle emissions is to assess the effects, in terms of
pollutant quantities, of changes in road characteristics, traffic congestion, and vehicle
technology.
Social
Socialand
and
Environmental
EnvironmentalEffects
Effects
Overview Energy
EnergyBalance Vehicle
Overview Balance VehicleEmissions
Emissions
Analysis
Analysis
Chapter
ChapterF1
F1 Chapter
ChapterF3
F3
Chapter
ChapterF2
F2
The emissions model is based on the model proposed by Hammerstrom (1995). Originally a
variety of units were used. The relationships and model coefficients have been adjusted so that
all emissions predictions are in terms of grammes per vehicle-kilometre, Bennett (1996).
The model predicts the different components of vehicle exhaust emissions as a function of
fuel consumption and speed. Fuel consumption is a function of vehicle speed, which in turn
depends on road characteristics and the characteristics of the vehicle itself. Thus, it is possible
to analyse the change in emissions effects as a result of implementing different road
maintenance and improvement options, or when there are major changes to the vehicle fleet
using the road network (for example, due to improved vehicle technology).
In this release, the effects of exhaust emissions are not costed for inclusion in economic
analysis of road investments, only the net differences in the quantities of pollutants are
assessed for each pair of investment options.
2 Emissions model
6 Particulates (Par)
7 Lead (Pb)
where:
The default values of the model parameters are given in Table F3.1.
2 Carbon monoxide
where:
3 Nitrous oxide
...(2.3)
where:
4 Sulphur dioxide
where:
5 Carbon dioxide
where:
Table F3.3 Model parameters for carbon dioxide, particulates and lead
emissions
6 Particulates
where:
where:
3 Modelling logic
For each section (investment) option and for each analysis year, the quantities of each
component of exhaust emissions are computed separately for each vehicle type k and for each
traffic flow period p. The annual total quantities of emissions (by component) are obtained by
summing over all the vehicle types.
The annual traffic volume during each flow period (vehicles per year).
! Length of road section
! Vehicle speeds
The instantaneous fuel consumption, for each vehicle type, in each traffic flow period is
calculated within the RUE module.
! Vehicle service life and model parameters
where:
EYRikp the annual quantity of emission component i from vehicle type k in traffic flow
period p (tonnes)
Tpk the annual traffic volume of vehicle type k in traffic flow period p (vehicles per
year)
Lj length of road section under investment option j (km)
EMikp the average quantity of emission component i (g/1000 veh-km), from vehicle
type k during traffic flow period p
The value of the average quantity of emission (EMikp) is obtained from the following
expression:
where:
Eikpu the quantity of emission component i (g/veh-km) for the uphill trip on the
section. It is calculated using the respective equations for each emission
component given in Section 2.2 (see Equations 2.1 above to 2.7 above), using
the following parameters (also see Part E):
IFC taken as IFCkpu - the instantaneous fuel consumption (ml/s) for the
uphill road segment
SPEED taken as SUkp (km/h)
Eikpd the quantity of emission component i (g/veh-km) for the downhill trip on the
section. It is calculated using the respective equations for each emission
component given in Section 2.2 (see Equations 2.1 above to 2.7 above), using
the following parameters:
IFC taken as IFCkpd - the instantaneous fuel consumption (ml/s) for the
downhill road segment
SPEED taken as SDkp (km/h)
For the analysis of one-way, the values of EMikp are obtained as follows:
! One-way uphill
! One-way downhill
The annual average quantities of vehicle emissions (by component i) per 1000 vehicle-
kilometres is given by the following expression:
n
∑ HRYR
p =1
p * HVp * EMi kp
EAVi k = n
...(3.5)
∑ HRYR
p =1
p * HVp
where:
The annual quantities of emissions (by component i) for each vehicle type k using the road
section under investment option j is calculated from the following expression:
n
EYRi jk = ∑ EYRi
p =1
kp ...(3.6)
where:
EYRijk the annual quantity of emissions of component i by vehicle type k for section
option j (tonnes)
The total annual quantities of emissions (by component i) for all vehicles using the road
section are calculated from the following expression:
n
EYRi j = ∑∑ EYRi
k p =1
kp ...(3.7)
where:
EYRij the annual quantity of emissions of component i (tonnes), for section option j
where:
∆EYRi(m -n) the annual net difference in the quantity of emissions component i
By component i and by vehicle type k for each traffic flow period p (EMikp). These
quantities will be reported for each section option j.
! Annual average quantities of vehicle emissions (g/1000 veh-km)
By component i and by vehicle type k (EAVijk). These quantities will be reported for
each section option j.
! Annual quantities of vehicle emissions (tonnes)
By component i and by vehicle type k (EYRijk). These quantities will be reported for each
section option j.
! Annual total quantities of vehicle emissions (tonnes)
5 References
ISOHDM Publications, (1994 - 1996)
International Study of Highway Development and Management Tools
University of Birmingham, UK
Hammerstrom U., (1995)
Proposal for a Vehicle Exhaust Model in HDM-4
ISOHDM Supplementary Technical Relationships Study Draft Report
Swedish National Road Administration
Road and Traffic Management Division
Borlange, Sweden
Bennett C.R., (1996)
HDM-4 Emissions Model - Draft Specifications
International Study of Highway Development and Management Tools
University of Birmingham, UK
NDLI, (1995)
Modelling Road User Effects in HDM-4 - Final Report
Asian Development Bank RETA 5549
N. D. Lea International, Vancouver
Contents
Analytical
AnalyticalFramework
Framework
and
andModel
Model
Descriptions
Descriptions
Introduction
Introduction
Part
PartAA
Traffic
Traffic
Part
PartBB
RD
RDModel WE
WEModel RUE
Model Model RUEModel
Model SEE
SEEModel
Model
Part
PartCC Part
PartDD Part
PartEE Part
PartFF
Economic
EconomicAnalysis
Analysis
RD = Road Deterioration Part
PartGG
Nomenclature
Nomenclature
WE = Road Works Effects Part
PartHH
G1 Economic Analysis
1 Introduction
HDM-4 caters for three applications levels commonly used in decision making within the road
sub-sector. The different applications, which are described in more detail in the Applications
Guide, are:
1 Strategic planning
For estimating medium and long-term budget requirements for the development and
preservation of a road network under various budgetary and economic scenarios.
2 Programme analysis
For preparing single or multi-year work programmes under budget constraints, in which
those sections of the network likely to require maintenance, improvement, or new
construction, are identified in a tactical planning exercise.
3 Project analysis
For estimating the economic or engineering viability of different road investment projects
and associated environmental effects. Typical projects include the maintenance and
rehabilitation of existing roads, widening or geometric improvement schemes, pavement
upgrading and new road construction.
For all the three applications, the underlying operation of HDM-4 is based on the concept of
life cycle analysis under a user-specified scenario of circumstances. This involves the analysis
of pavement performance, road works effects and costs, together with estimates of road user
costs and environmental effects, and economic comparisons of different project alternatives.
This chapter describes how HDM-4 is used to determine the benefits and costs associated with
a road investment, and how these are applied in economic analysis and optimisation
procedures to find the best use of available resources.
2 Background
2.2 Optimisation
The purpose of the Strategy and Programme applications is to calculate the economic benefits
derived from maintenance or improvement options, and to select the set of investments to be
made on a number of road sections within a network which will optimise an objective
function.
Programme analysis is concerned with short to medium term planning and preparation where
budget levels are known with reasonable certainty and the objective is to select a set of road
sections and road works within the budget constraint.
Strategy analysis involves the analysis of an entire road network (or sub-network). The
objective is either to determine which types of road works should be applied in order to
maximise economic benefits, or it may also be applied to determine the budget required for a
given long term target road network condition. Thus, the problem can be posed as one of
searching for the combination of investment alternatives that optimises the objective function
under a budget constraint or a road network condition constraint. Note that the set of
investment options to be optimised is user-defined and is not the set of all possible options for
the particular network; hence the problem is not true optimisation since all possible solutions
are not normally considered. Note also that the investment options on any one road section are
mutually exclusive.
The three alternative objective functions provided for the Strategy and Programme
applications are:
This option is used when the problem can be defined as the selection of a combination of
investment options applied on several road sections which maximises the NPV (net
present value) for the whole network subject to the sum of the financial costs being less
than the budget available.
2 Maximisation of the improvement in network condition
The roughness reduction on each road section multiplied by the section length
(∆IRI*Length) is used instead of NPV. Consequently, the arithmetic procedure is similar
to that used for maximisation of economic benefits.
3 Minimise costs of road works to achieve a given target road network condition
This option is used mainly in the Strategy analysis application. The target road condition
defined in terms of the long-term average roughness (IRI) over the whole analysis period
must be specified for each road section. The optimisation procedure is then reduced to a
simple selection of the road work options for which the average IRI (over the analysis
period) is equal to or just below the target IRI and has the lowest total financial cost.
For example, vehicle operating costs, savings in travel time, accident costs.
2 Quantified benefits and costs not expressed in monetary terms
For example, road safety, pollution from vehicle emissions and traffic noise.
3 Non-quantified benefits and costs
The cost of works is derived from the product of the physical quantities involved in the
activity and the unit cost. These are determined for each road section and investment option,
and for each year of the analysis period. The resulting costs are assigned to budget categories
that are user-definable. The following default categories are used in HDM-4:
! Capital (or periodic)
! Recurrent (or routine)
! Special
Budget constraints can be applied separately to each category when required by the economic
analysis and optimisation.
❏ Overheads
! Travel time costs
These costs include passenger travel time costs, and cargo holding time costs.
! Non-motorised transport (NMT)
These costs are evaluated both in monetary and non-monetary terms, and separated into
several different types (for example, fatal, injury, and damage only, and all accidents). Note
users are allowed the flexibility to include or exclude accident costs from an economic
analysis.
These unit costs are specified by the user (see Part D).
! Road user costs
These unit costs include vehicle resources, travel-time values, and road accident
resources (see Part E).
In most cases, unit costs are specified in units-per-quantity. However, some costs are specified
as a proportion of other costs, or as a lump sum.
In addition to calculating economic costs, financial costs are also computed if the user gives
appropriate inputs (for example, unit costs in financial terms).
4 Outline methodology
Road geometry and surface Social and Levels of emissions and energy
texture, vehicle characteristics Environmental Effects used, and number of accidents
4.3 Models
Total life-cycle conditions and costs of sections or road networks can be simulated over a
user-defined period into the future. The inter-dependence between the costs incurred by the
road administration and the road user is recognised, and models are used to predict cost
streams under the various headings.
The models incorporated in HDM-4 contain technical relationships for the following
purposes:
1 Calculation of traffic volumes and flows, and vehicle loading over the road section.
2 Prediction of road deterioration, and works effects and costs, that are incurred in response
to traffic flows, time and the surrounding environment.
3 Prediction of the costs of road use incurred as road condition and traffic flow change over
time.
4 Prediction of accident rates as a function of the road and traffic characteristics, and the
evaluation of accident costs.
5 Evaluation of vehicle emissions and energy use due to different road investment projects.
6 Economic analysis by comparison of the impacts or effects of different road investment
project alternatives.
Enabling economic comparisons to be made for each pair of investment options, using
the effects and costs calculated over the analysis period for each option, and indicates
that generated and diverted traffic levels may vary depending on the investment option
considered.
2 Costs
How annual costs to the road administration and to the road users are calculated for
individual road section options.
3 Optimisation procedures
These are performed after economic benefits of all the section options have been
determined.
The pseudo code that represents the outer analysis loop is given below:
START
Define input data
Loop for each option
Loop for each section
Loop for each analysis year
Calculate traffic over the road section
Calculate total non-discounted net benefits over all the sections (see Figure
G1.2c)
Calculate total discounted net benefits over all the sections (see Figure G1.2c)
Calculate total environmental effects and energy used over all the sections (see
Figure G1.2c)
End loop
Calculate economic indicators (NPV, IRR, BCR, and FYB see Section 5.3.1 and
Figure G1.2c)
End Loop
Perform budget optimisation (for strategy and programme analysis)
Output results
END
LOOP
For each option
LOOP
For each section
LOOP
For each analysis year
CALCULATE
This year’s traffic
over the road section
STORE
Results for evaluation
and reporting phase
Yes
More
Years?
No
Yes
More
Sections?
No
Yes
More
Options?
No
See Figure G1.2c
Economic Analysis and
Comparisons
Optimisation
procedures
Output Results
MODEL
Annual effects and costs
CALCULATE
Road Deterioration
[RD Module]
CALCULATE
Road User Effects
(Speeds, VOC, time costs,
NMT, accident costs)
[RUE Module]
CALCULATE
Road Works Effects
[WE Module]
CALCULATE
Social and Environmental Effects
Social
(emissions
and Environmental
and energy balance)
Effects
(emissions andModule]
[SEE energy balance)
ADD
This year’s exogenous
benefits and costs
Return
LOOP
For each comparison
LOOP
For each analysis year
SET NPVsr = 0
Total net effects TNEsn (1, . . .,n) = 0
LOOP
For each section
CALCULATE
non-discounted Net Benefits
CALCULATE
Discounted Net Benefits at different
discount rates (r)
CALCULATE
Net effects (n)
Yes
More
Sections?
No
CALCULATE
Total net benefits and effects
Yes
More
Years?
No
CALCULATE
Total net effects
Yes
More
comparisons?
No
Return
The procedure for calculating annual road agency costs and road user effects for individual
section options is illustrated in Figure G1.2b and summarised by the following steps:
1 Calculate road deterioration - in the RD module (see Part C)
VOC, travel time costs, NMT time and operating costs, and accident costs - in the RUE
module (see Part E).
3 Calculate quantities and costs for road works - in the WE module (see Part D)
For example, emissions and energy use - in the SEE module (see Part F).
5 Add exogenous benefits and costs
Figure G1.2c illustrates the inner analysis loops for economic analysis and comparison of each
pair of road alternatives.
5 Economic analysis
∆C (m−n)i =
∑C
s
mis - ∑C
s
nis
...(5.1)
where:
∆RAC (m −n ) = ∑ ∆C
i
(m −n)i ...(5.2)
where:
These cost differences provide a relative measure of the increase in costs to the road
administration, of implementing investment option m over base option n.
The difference in the salvage values of works performed under investment options m and n is
a component of the net economic benefits to be included in the last year of the analysis period
(see Section 5.2.4), and is given as:
where:
Vehicle operating benefits due to normal and diverted traffic is calculated as follows:
∆VCN (m−n) =
∑ VCN - ∑ VCN
s
ns
s
ms
...(5.4)
VCNns = ∑ TN
k
nsk * UCnsk ...(5.5)
VCN ms = ∑ TN
k
msk * UC msk ...(5.6)
∆VCG (m−n) =
∑∑ {0.5 * [TG
s k
msk + TG nsk ] * [UC nsk − UC msk ]}
...(5.7)
The summations are over all the motorised vehicle types (k = 1, 2, ..., K) specified by the
user, and all road sections (s = 1, 2, ... ., S) being analysed.
The annual saving in vehicle operating costs is given by the expression:
where:
∆VCN(m-n) vehicle operating benefits due to normal and diverted traffic of investment
option m relative to base option n
VCNjs annual vehicle operating cost due to normal and diverted traffic over the
road section s with investment option j
TNjsk normal and diverted traffic, in number of vehicles per year in both
directions on road s, investment option j, for vehicle type k
UCjsk annual average operating cost per vehicle-trip over road section s, for
vehicle type k under investment option j (where j = n or m)
VCGjs annual vehicle operating cost due to generated traffic over road section s
with investment option j
Vehicle travel time benefits due to normal and diverted traffic are calculated as follows:
∆TCN (m−n) =
∑ TCN - ∑ TCN
s
ns
s
ms
...(5.9)
TCN ns = ∑ TN
k
nsk * UTnsk ...(5.10)
TCN ms = ∑ TN
k
msk * UTmsk ...(5.11)
Vehicle travel time benefits due to generated traffic are calculated as follows:
∆TCG (m−n) =
∑∑ {0.5 * [TG
s k
msk + TG nsk ] * [UTnsk − UTmsk ]}
...(5.12)
The annual savings in travel time costs are given by the expression:
where:
∆TCN(m-n) travel time benefits due to normal and diverted traffic of investment
option m relative to base option n
TCNjs annual vehicle travel time cost due to normal and diverted traffic over
road section s with investment option j
UTjsk annual average travel time cost per vehicle-trip over the road section s, for
vehicle type k, under investment option j (where j = n or m)
TCGjs annual vehicle travel time cost due to generated traffic over road section s
with investment option j
∆TCG(m-n) travel time benefits due to generated traffic of investment option m
relative to base option n on the given road section in the given year
∆TTC(m-n) savings in travel time costs due to total traffic of investment option m
relative to base option n
Non-Motorised Transport (NMT) time and operating benefits due to normal and diverted
traffic are calculated as follows:
∆TOCN(m−n) =
∑ TOCN - ∑ TOCN
s
ns
s
ms
...(5.14)
TOCNns = ∑ TN
k
nsk * UTOC nsk ...(5.15)
TOCNms = ∑ TN
k
msk * UTOC msk ...(5.16)
NMT time and operating benefits due to generated traffic are calculated as follows:
∆TOCG (m−n) =
∑∑ [0.5 * (TG
s k
msk + TG nsk ) * (UTOC nsk − UTOC msk )]
...(5.17)
The summations are over all the NMT types (k = 1, 2, ..., K) specified by the user, and all
road sections (s = 1, 2, ... ., S) being analysed.
The annual savings in NMT time and operating costs are given by the expression:
where:
∆TOCN(m-n) NMT time and operating benefits due to normal and diverted traffic of
investment option m relative to base option n
TOCNjs annual NMT time and operating costs due to normal and diverted
traffic over the road section s with investment option j
TNjsk NMT normal and diverted traffic, in number of vehicles per year in
both directions on road s investment option j, for vehicle type k
UTOCjsk annual average NMT time and operating cost per vehicle-trip over road
section s, for vehicle type k, under investment option j (where j = n or
m)
TOCGjs annual NMT time and operating costs due to generated traffic over
road section s with investment option j
TGjsk NMT generated traffic, in number of vehicles per year in both
directions on road s, for vehicle type k, due to investment option j
∆TOCG(m-n) NMT time and operating benefits due to generated traffic of investment
option m relative to base option n
∆NMTOC(m-n) annual savings in NMT time and operating costs due to total traffic of
investment option m relative to base option n
The benefits from reduction in total accident costs are given by the expression:
where:
The annual savings in road user costs are given by the expression:
∆RUC (m −n ) = [∆VOC (m-n) + ∆TTC (m-n) + ∆NMTOC (m-n) + ∆ACC (m-n) ] ...(5.20)
where:
where:
∆NEXBy(m-n) the annual net exogenous benefits of investment option m relative to base
option n, in year y
EXBjy exogenous benefits for investment option j, in year y, (where j = n or m)
where:
In the last year of the analysis period, the net economic benefits of implementing option m
relative to option n is calculated as:
...(5.23)
where:
NBY(m-n) net economic benefit of investment option m relative to base option n in the
last year of the analysis period Y, and the parameters on the right-hand side
are as defined earlier, but with subscript Y added to indicate the last year of
the analysis period
∑ [1+ 0.01 * r]
NB y(m-n)
NPV(m −n) = (y -1)
...(5.24)
y =1
where:
The higher the NPV, the greater the benefits from investment option m relative to base option
n. If there are no budget constraints, then the choice between the two alternative investments
should be based on NPV. Larger investments will tend to have larger NPVs.
This equation is solved for r° by evaluating the NPV at 5 percent intervals of discount rates
between -95 and +900 percent, and determining the zero(es) of the equation by linear
interpolation of adjacent discount rates with NPV of opposite signs. Depending on the nature
of the net benefit stream, NBy(m-n), it is possible to find one solution, multiple solutions, or
none at all.
The IRR gives no indication of the size of the costs or benefits of an investment; it acts as a
guide to the profitability of the investment - the higher the better. If the computed IRR is
larger than the planning discount rate, then the investment is economically justified.
NPV(m-n)
BCR (m-n) = +1
Cm
...(5.26)
where:
NPV(m-n) discounted total net benefit of investment option m relative to base option n.
This is the Net Present Value at discount rate r
Cm discounted total agency costs of implementing investment option m
If the NPV(m-n) is zero, then (NPV/C)(m-n) is zero. These ratios give an indication of the
profitability of investment option m relative to base option n at a given discount rate. These
measures eliminate the bias of NPV towards larger project options but, like the IRR, they give
no indication of the size of the costs or benefits involved.
100 * NB y °(m-n)
FYB (m-n) =
∆TCC (m−n)
...(5.27)
where:
FYB gives a rough guide to project timing: if it is greater than the discount rate, then the
project should go ahead; otherwise it should be delayed until it satisfies the criterion.
6 Optimisation
The two methods for budget optimisation provided for road works programming and network
strategic analysis are:
1 Total enumeration
where:
∑∑ R
s =1 m =1
smqt X sm ≤ TR qt , q = 1,...., Q; t = 1,...., T ...(6.2)
where:
Ms
∑m =1
X sm ≤ 1, s = 1,...., S ...(6.3)
that is, for each road section s, no more than one alternative can be implemented.
If M is the average number of alternatives for the roads, the problem then has SM (= S x M)
zero-one variables, QT (= Q x T) resource constraints and S interdependency constraints. The
parameters that define the problem size are S, M and QT. Depending on the solution method
used, different problem-size parameters determine whether the method is suitable for the
problem in terms of the computational effort needed.
The total enumeration method provides the user with an unconditionally optimal solution. It
computes the total net present values of all feasible programme selections, and chooses the
one with the highest value. The computational effort required for this may be considerable, so
the method is only feasible when the number of alternatives per investment unit is relatively
small.
NPV - NPV
j i
E = ...(6.4)
cost - cost
ji
j i
where:
In Equation 6.4 above, the incremental NPV/cost can be replaced by the incremental
∆IRI*Length/cost where ∆IRI*Length is the weighted average change in roughness obtained
by comparing the project alternatives using IRI instead of NPV.
The objective of the incremental method is to select road sections successively starting with
the largest NPV/cost ratio (Eji), since this maximises the NPV (net present value) for any
given budget constraint. Where there is more than one investment option on any individual
road section, that with the lowest discounted investment costs is designated the base case
alternative. This method considers all possible options, and compares these incrementally
starting against the base case, by using the incremental algorithm to select the combination
that maximises the selected objective function.
An incremental search technique is used to select the options with successively lower
incremental NPV/cost ratios, ensuring that at any time there is no more than one option per
road section. The process continues until the budget is exhausted for each budget period. The
method is often referred to as the efficiency frontier, which is a line that joins investments
with the highest NPV along the cost axis in a plot of NPV against investment cost (Harral and
Faiz, 1979). In essence, the method seeks out those options that are close to the boundary of
the efficiency frontier. The algorithm is illustrated in Figure G1.3, and is defined in the
following steps:
1. Determine the pre-defined investment options for pre-selected sections and deduct the
financial costs of these options from the available budget in corresponding years. Exclude
these sections from any further optimisation.
2. Determine possible investment options for the remaining sections. If the life cycle
analysis option is being used, set the user-defined base alternatives as the do minimum
for each road section. For the multi-year forward programme, the do minimum option is
that with the delayed capital works.
3. If the total financial cost of the do minimum investment alternatives on each section is
greater than the available budget for any period, then the investment options or budget
constraints must be redefined.
4. Deduct the financial cost of the do minimum investments from the available budget to
determine the remaining budget for each period. Set the do minimum as the first Base
option for each section.
5. Calculate the incremental NPV/Cost ratio for all remaining section-options compared
against the Base option, and all other option pairs with higher economic cost. For
example, consider the following investment options for a particular section arranged in the
ascending order of discounted total economic costs:
options: A, B, C, D, E
The incremental NPV/Cost ratios for these are given by:
Eba Eca Eda Eea ; Ecb Edb Eeb ; Edc Eec ; Eed
6 Delete incremental NPV/cost ratios that are less than the user specified minimum
incremental value (MIV).
7 List the remaining incremental NPV/cost ratios in decreasing order (with the associated
section-option pair codes) and, within each incremental NPV/cost, in the order of
decreasing economic cost. For example, if Eeb = Edb then Eeb is ranked higher.
8 Select the next incremental NPV/cost ratio from the top of the list. If the lower cost
section-option is not the current Base Option for that section, continue searching until one
is found.
9 If the remaining budget is insufficient in any of the periods for the financial costs of
works required for the section-option selected in Step 8 above, then the selected option
should be rejected, and continue searching by repeating Step 8.
10 If the section-option can fit within the remaining budgets for all periods, deduct the net
increase in financial cost of capital works from all corresponding budget periods. Set the
Base option for this section to be that corresponding to the lower cost option for the
incremental NPV/Cost ratio chosen in Step 8. Providing that the remaining list is not
empty, return to Step 8.
The process described above continues until the budget is exhausted or there are no more
section-options remaining in the list. The resulting list of selected section-alternatives
constitutes the optimal work programme.
NPV
E
D
B
C A, B, C, D, E : Section-Alternatives
A
ECONOMIC COST
(relative to Base Option)
Figure G1.3 Efficiency frontier concept
7 References
Harral, C.G. and Faiz, A. (1979)
The highway design and maintenance standards model (HDM): model structure,
empirical foundations and applications. PTRC Summer Annual Meeting, University of
Warwick, 13-16 July 1979. PTRC Education and Research Services, London, UK
PIARC, (1991)
Methods for Selecting Road Investment, Economic and Finance Committee of PIARC,
Paris, France
TRRL Overseas Unit, (1988)
A Guide to Road Project Appraisal. Transport and Road Research Laboratory Overseas
Road Note 5, Crowthorne UK
Contents
Analytical
AnalyticalFramework
Framework
and
andModel
Model
Descriptions
Descriptions
Introduction
Introduction
Part
PartAA
Traffic
Traffic
Part
PartBB
RD
RDModel WE
WEModel RUE
Model Model RUEModel
Model SEE
SEEModel
Model
Part
PartCC Part
PartDD Part
PartEE Part
PartFF
Economic
EconomicAnalysis
Analysis
RD = Road Deterioration Part
PartGG
Nomenclature
Nomenclature
WE = Road Works Effects Part
PartHH
Part H Nomenclature
H1 Nomenclature
To be included in a subsequent edition of this document.
Contents
Analytical
AnalyticalFramework
Framework
and
andModel
Model
Descriptions
Descriptions
Introduction
Introduction
Part
PartAA
Traffic
Traffic
Part
PartBB
RD
RDModel WE
WEModel RUE
Model Model RUEModel
Model SEE
SEEModel
Model
Part
PartCC Part
PartDD Part
PartEE Part
PartFF
Economic
EconomicAnalysis
Analysis
RD = Road Deterioration Part
PartGG
Nomenclature
Nomenclature
WE = Road Works Effects Part
PartHH
Part I Glossary
I1 Glossary
To be included in a subsequent edition of this document.
Wo r l d R o a d
Association
5
THE HIGHWAY DEVELOPMENT AND MANAGEMENT SERIES
Part A
Overview of HDM-4
Volume 1
Applications
Applications Guide
Guide Software
Software User
User Guide
Guide
Volume
Volume 22 Volume
Volume 33
Analytical
Analytical Framework
Framework andand Model
Model
Descriptions
Descriptions
Volume
Volume 44
A
A Guide
Guide to
to Calibration
Calibration and
and Adaptation
Adaptation
Volume
Volume 55
A short executive summary describing the HDM-4 system. It is intended to be used by all
readers new to HDM-4, particularly high level management within a road organisation.
! Applications Guide (Volume 2)
Describes the analytical framework and the technical relationships of objects within the
HDM-4 model. It contains very comprehensive reference material describing, in detail,
the characteristics of the modelling and strategy incorporated in HDM-4. It is to be used
by specialists or experts whose task is to carry out a detailed study for a road
management organisation.
! A Guide to Calibration and Adaptation (Volume 5)
Suggests methods for calibrating and adapting HDM models (as used in HDM-III and
HDM-4), to allow for local conditions existing in different countries. It discusses how to
calibrate HDM-4 through its various calibration factors. It is intended to be used by
experienced practitioners who wish to understand the detailed framework and models
built into the HDM-4 system.
Notes:
1 Volumes 1, 2 and 3 are designed for the general user.
2 Volumes 4 and 5 will be of greatest relevance to experts who wish to obtain low level
technical detail. However, Volume 5, in particular, presents very important concepts,
which will be of interest to all users.
Describes the need for model calibration, different levels of calibration based on
available time and resources.
! Chapter 3 - Reliability concepts
Addresses bias and precision of model predictions, input data accuracy, and ways of
assessing the reliability of HDM predictions.
! Chapter 4 – Sensitivity of HDM
Presents the results of analyses with HDM to identify the sensitivity of HDM’s output to
changing the input data or model parameters.
! Chapter 5 – Adapting data to the model
A discussion of how one adapts local data to the HDM model. Appendices A to H
address the specifics of how one collects data for use in HDM.
! Chapter 6 – RUE model calibration
Presents details on the calibration of the HDM road user effects model.
! Chapter 7 – RDWE calibration
Presents details on the calibration of the HDM road deterioration and maintenance effects
model.
Defines unit cost data for RUE and works effects (WE).
! Appendix G - Economic data
Describes the two approaches (the t Distribution and the normal distribution) that can be
used. The approach is dependent upon the number of samples.
! Appendix I - Survival curve analysis
Describes HDM Tools, a set of software applications, which is designed to assist in the
calibration of HDM-4.
Figure 2 shows the process readers should follow in using this manual.
Calibrate Selected
Establish Calibration
Setup RUE Model
Level
HDM Model Parameters
(Chapter 2)
(Chapter 6)
Calibrate Selected
Review Reliability
RDWE Model
Concepts
Parameters
(Chapter 3)
(Chapter 7)
Based on Time,
Resources, Type of Establish Input Data
Analysis Select Most for HDM Model
Critical Parameters (Appendix C to G)
(Chapter 4)
Establish Appropriate
Level of Data Detail
(Chapter 5)
ISOHDM products
The products of the International Study of Highway Development and Management Tools
(ISOHDM) consist of the HDM-4 suite of software, associated example case study databases,
and the Highway Development and Management Series collection of guides and reference
manuals. This Volume is a member of that document collection.
Customer contact
Should you have any difficulties with the information provided in this suite of documentation
please do not hesitate to report details of the problem you are experiencing. You may send an
E-mail or an annotated copy of the manual page by fax to the number provided below.
The ISOHDM Technical Secretariat welcomes any comments or suggestions from users of
HDM-4. Comments on A Guide to Calibration and Adaptation should be sent to the
following address:
E-mail: isohdm@bham.ac.uk
Telephone: +44 - 121 - 414 6717 (5049)
Fax: +44 - 121 - 414 3675 (5060)
Change details
This is the first formal edition (Version 1.0) of the HDM-4 documentation.
Related documentation
HDM-4 documents
The Highway Development and Management Series Collection is ISBN: 2-84060-058-7, and
comprises:
Volume 1 - Overview of HDM-4, ISBN: 2-84060-059-5
Volume 2 - Applications Guide, ISBN: 2-84060-060-9
Volume 3 - Software User Guide, ISBN: 2-84060-061-7
Volume 4 - Analytical Framework and Model Descriptions, ISBN: 2-84060-062-5
Volume 5 - A Guide to Calibration and Adaptation, ISBN: 2-84060-063-3
Future documentation
The following documents will be issued at a later release:
Volume 6 - Modelling Road Deterioration and Works Effects, ISBN: 2-84060-102-8
Volume 7 - Modelling Road User and Environmental Effects, ISBN: 2-84060-103-6
Terminology handbooks
PIARC Lexicon of Road and Traffic Engineering - First edition. Permanent International
Association of Road Congresses (PIARC), Paris 1991. ISBN: 2-84060-000-5
Technical Dictionary of Road Terms - Seventh edition, English - French. PIARC Commission
on Terminology, Paris 1997. ISBN: 2-84060-053-6
Acknowledgements
The work on this Guide to Calibration and Adaptation was initially sponsored by N.D. Lea
International Ltd. of Canada. Additional work was sponsored by the Asian Development
Bank, HTC Infrastructure Management Ltd., and finally the ISOHDM Study through PIARC.
The authors would like to express their appreciation to all those who assisted in developing
this manual:
! Ian Greenwood of Opus International Consultants Ltd. (N.Z.)
Wrote the HDM Tools calibration applications.
! Rodrigo Archondo-Callao of the World Bank
Assisted in generating elasticities for the HDM-III model.
! Technical comments were received from a number of reviewers including, but not
limited to:
❏ Rodrigo Archondo-Callao
❏ Clive Daniels
❏ Chris Hoban
❏ Michael J. Riley
❏ Robert Thurlow
The development of HDM-4 has been sponsored by several agencies, primarily:
! Asian Development Bank (ADB)
! Department for International Development (DFID) in the United Kingdom
! Swedish National Road Administration (SNRA)
! The World Bank
Many other organisations and individuals in a number of countries have also contributed in
terms of providing information, or undertaking technical review of products being produced.
The study has been co-ordinated by the ISOHDM Technical Secretariat at the University of
Birmingham in the United Kingdom. A number of organisations participated in the research
including:
! Finnra
Responsible for developing deterioration relationships for cold climates, road safety,
environmental effects, and supporting HRG with system design.
All research organisations received support from local and regional staff, visiting experts and
external advisers, to ensure that a high standard of quality and international consensus was
achieved. A number of other countries and individuals have supported this work through
supplying expert advice and reviewing the products.
Copyright statement
These HDM-4 products have been produced by the International Study of Highway
Development and Management Tools (ISOHDM), sponsored by The World Bank, the Asian
Development Bank, the Department for International Development (UK), the Swedish
National Road Administration, and other sponsors. The HDM-4 products are jointly published
by The World Road Association (PIARC), Paris and The World Bank, Washington, DC.
Copyright © 2000 The World Road Association (PIARC) on behalf of the ISOHDM sponsors.
All rights reserved.
This copyright covers all documents and document components, computer software and data
delivered as components of the HDM-4 product, in any physical or electronic forms.
Contents
Volume 5 - A Guide to Calibration and
Adaptation
1 Introduction 1
2 Calibration issues 4
2.1 The need for calibration 4
2.1.1 Model development considerations 5
2.1.2 Case examples 5
2.2 Input data 7
2.3 Calibration 7
2.3.1 Introduction 7
2.3.2 Level 1 - Application 9
2.3.3 Level 2 - Calibration 11
2.3.4 Level 3 - Adaptation 11
2.4 Report scope 11
3 Reliability concepts 13
3.1 Introduction 13
3.2 Bias and precision 13
3.3 Correction factors 15
3.4 Input data accuracy 15
3.5 Limits on bias 17
3.6 Assessing the reliability of HDM predictions 18
3.6.1 Pavement performance: simulation of the past 18
3.6.2 Pavement performance: controlled studies 19
3.6.3 Road User Effects: controlled survey 20
3.6.4 Road User Effects: tariff survey 20
3.6.5 Road User Effects: fleet survey 22
4 Sensitivity of HDM 24
4.1 Introduction 24
4.2 Road User Effects 25
4.3 Sensitivity classes 30
4.3.1 High impacts, Class S-I (> 0.5) 30
4.3.2 Moderate impacts, Class S-II (0.2 - 0.5) 30
7 RDWE calibration 92
7.1 Bituminous paved road deterioration 92
7.2 Level 1 - Basic application 93
7.2.1 Roughness-age-environment adjustment factor: S-I 93
7.2.2 Cracking initiation adjustment factor: S-I 96
7.2.3 Cracking progression adjustment factor: S-I 98
7.2.4 Rut depth progression adjustment factor: S-I 98
7.2.5 General roughness progression factor: S-III 100
9 References 115
1 Introduction
The Highway Development and Management Model (HDM), originally developed by the
World Bank, has become widely used as a planning and programming tool for highway
expenditures and maintenance standards. HDM is a computer model that simulates physical
and economic conditions over the period of analysis, usually a life cycle, for a series of
alternative alternatives and scenarios specified by the user.
HDM is designed to make comparative cost estimates and economic evaluations of different
construction and maintenance options, including different time-staging alternatives, either for
a given road project on a specific alignment or for groups of links on an entire network. It
estimates the total costs for a large number of alternative project designs and maintenance
alternatives year by year, discounting the future costs if desired at different postulated
discount rates so that the user can search for the alternative with the lowest discounted total
cost.
Three interacting sets of costs (related to construction, maintenance and road use) are added
together over time in discounted present values, where the costs are determined by first
predicting physical quantities of resource consumption and then multiplying these by unit
costs or prices.
The broad concept of HDM is illustrated in Figure 1.1. The user defines a series of
alternatives that describe different investment and preservation options for the road. The
investments influence the condition of pavement over time and road maintenance costs. The
pavement and traffic conditions have an influence on Road User Effects (RUE). The model
predicts traffic speeds and the consumption of the RUE components, such as fuel, tyres etc.
Multiplying these by the unit costs of the individual components gives the RUE over time.
Comparing the cost outputs from various investment alternatives allows assessment of the
relative merits, cost savings and benefits of the different alternatives using economic
principles.
Aternative n
Aternative 3
Process
Process
Aternative 2
Process
Aternative 1 Process Economic
Economic
Comparison
Comparison
Process
Process Total Costs
Pavement condition Road maintenance Total Costs
over time costs over time Net Present Value (NPV)
Net Present Value (NPV)
Internal Rate of Return (IRR)
Internal Rate of Return (IRR)
Vehicle operating
cost component
Vehicle operating
consumption as a
costs over time
function of road
condition
As illustrated in Figure 1.2, HDM consists of a series of sub-models that address different
aspects of the analysis. Each of these sub-models requires certain input data and each
produces its own output. In order to apply the model correctly, one needs to ensure that HDM
is given the appropriate input data and has been suitably calibrated.
This report presents guidelines for the quality assurance of HDM applications through control
of data quality and calibration of the HDM model. It describes the range of options for data
collection and input to HDM, as well as accuracy considerations. The sensitivity of HDM to
the input data is used to establish those data items that are most critical in the analysis.
Similarly, the report discusses how one calibrates HDM through its various calibration factors
and the sensitivity of the model to these factors. It presents three levels of effort to achieve
calibration based on the available time and resources.
The focus of the report is on HDM-4, although many of the broad principles are applicable to
the earlier releases HDM-III and HDM-95.
• Vehicle Operating Costs (VOC) are the costs of vehicle operation. They are
comprised of fuel, tyres, maintenance, labour, depreciation, interest, lubricants,
crew, cargo and travel time.
• Road User Effects (RUE) are the VOC along with safety, vehicle emissions,
noise and energy balance.
• Road Deterioration and Works Effects (RDWE) encompass pavement
deterioration and the effects of works improvements. The latter include
maintenance as well as improvements such as widening.
Road geometry and surface texture, Social and Levels of emissions and energy
vehicle characteristics Environmental Effects used, and number of accidents
2 Calibration issues
Predictive
Input Data Analysis Results
Models
This Report
Calibration of the HDM model focuses on the two primary components that determine the
physical quantities, costs and benefits predicted for the analysis, namely:
! Road User Effects (RUE) - comprised of vehicle operating costs (VOC), travel time,
safety and emissions, and
! Road deterioration and works effects (RDWE) - comprised of the deterioration of the
pavement and the impact of maintenance activities on pavement condition and the future
rate of pavement deterioration.
The number and variety of data items required to feed structured models are potentially
large and complex. The HDM modelling chose a middle path, avoiding parameters that
require research-level tests and, for road user effects, setting some parameters to have
default values that the user could change when analytical resources permitted.
! Non-modelled effects
Even though the field experiments covered wide ranges of conditions, and even though
state-of-the-art theory was applied in the model’s development, there remain some
factors that could not be introduced, because:
❏ they were not measured, or
❏ would have made the model’s input too complex, or
❏ their effects could not be determined within the ranges observed
Also, there have been advances in vehicle technology in the period since the studies were
done which have lowered consumption and operating costs. For these reasons, some
calibration of the HDM model to local conditions is both sensible and desirable.
180
2-Axle Truck 5-Axle Truck 7-Axle Truck
160
140
120
Crew
Cost in cents/km
100 Interest
Depreciation
80 Maintenance
Tires
40
20
0
Observed HDM Observed HDM Observed HDM
Chesher and Harrison (1987) also point to the need for local calibration through their
comparisons of the predictions of the models developed in the various road users cost studies.
The relationships not only show significant variations in the magnitudes of the various costs,
but also in their sensitivity to pavement and operating conditions. Finlayson and du Plessis
(1991), in a comprehensive report containing a number of different papers on VOC, illustrate
the importance of calibrating VOC equations to local conditions and discuss how it was done
for Southern Africa.
For road deterioration, environmental conditions have a strong influence, affecting
deterioration rates by factors of up to 2 or 3 between the extremes of hot arid and cold wet
climates. Local construction materials, practices and quality also affect deterioration rates and
the effectiveness of maintenance. Both influences can be controlled through calibration
factors in the inputs to HDM, in addition to the main traffic and pavement input parameters.
This report presents the ways in which the level of confidence of the HDM predictions can be
raised through differing levels of effort that can be matched to the level of application. The
advantages of a universal model such as HDM are that it can be used quickly with relatively
little or no investment in extensive empirical and statistical research, and that it represents a
comprehensive techno-economic framework. Enhancing the reliability thus involves first
ensuring the validity of input data and primary parameter values, and second, refining but not
reconstructing the predictive relationships to conform to local data. A third stage may be
undertaken which sees relationships replaced by new ones developed from local research.
An analogy of the sea is useful to illustrate the roles of data and calibration. Data
determine the order and magnitude of costs and effects, so these must be of the
same order as baseline information, much like the depth of the sea and other
attributes like density. Calibration ensures that the height of the waves will be
correct under the influence of winds, currents and depths of water.
Calibration
This report deals with both issues. In the context of HDM-4 data are those items
which, when you install the model, require input data. This includes unit costs, road
attributes, maintenance alternatives, etc. Where default values are supplied, these
are model parameters or coefficients that can be calibrated.
The main text of the report deals with calibration of model parameters and
coefficients; Appendix C to Appendix G deals with how one establishes the input
data for HDM-4.
2.3 Calibration
2.3.1 Introduction
Calibration differs from input data since calibration is aimed at adjusting the model
predictions. As shown in Appendix A, the HDM RUE and RDWE sub-models contain a large
number of parameters that can be adjusted. It will be noted from Appendix A that HDM-4
has many more calibration factors than HDM-III. This is because many values that were hard
coded in the HDM-III source code can now be altered in HDM-4.
The degree of local calibration appropriate for HDM is a choice that depends very much on
the type of application and on the resources available to the user. For example, in planning
applications the absolute magnitude of the RUE and road construction costs need to match
local costs closely because alternative capital projects with different traffic capacities or route
lengths are evaluated on the comparison of the total road transport costs. In road maintenance
programming, on the other hand, the sensitivity of RUE to road conditions, particularly
roughness, and all the road deterioration and maintenance predictions are the most important
aspects.
There are three levels of calibration for HDM, which involve low, moderate and major
levels of effort and resources, as follows:
! Level 1 - Basic Application (see Section 2.3.2)
Determines the values of required basic input parameters, adopts many default values,
and calibrates the most sensitive parameters with best estimates, desk studies or minimal
field surveys.
! Level 2 - Calibration (see Section 2.3.3)
Undertakes major field surveys and controlled experiments to enhance the existing
predictive relationships or to develop new and locally specific relationships for
substitution in the source code of the model.
In terms of effort, these three levels can be viewed as weeks, months and years. An analyst
should be able to undertake a Level 1 calibration in about one-week. For a Level 2 calibration
there is an increase in the amount of effort required so it will take at least a month. Level 3
calibrations require a long-term commitment to basic data collection so their extent spans for
a year or more.
Every HDM analysis requires at least a Level 1 calibration.
Figure 2.3 illustrates this concept of increasing effort and increasing resources. It must be
appreciated that there is a direct relationship between the time and effort expended in setting
up HDM and the reliability and accuracy of its output.
Time Required
Experimental
Years
Surveys and
Research
Months
Field Surveys
Weeks
Desk Studies
Resources
Limited Moderate Significant Required
As shown in Table 2.1, a related concept of a hierarchy of calibration with increasing levels
of activity was suggested by Curtayne et al. (1987) in regard to VOC. They noted that
calibration activities could be based on both primary and secondary data sources. Primary
sources are based on direct comparisons between costs and highway characteristics whereas
secondary data provide information about the economic or operating conditions of a region.
The calibration of VOC results varies from the selection of available relationships on the
basis of few data to the estimation of local relationships using especially collected data.
The following sections outline the components of Level 1 to Level 3 calibrations.
Appendix A gives the data items and model parameters that should be investigated in
calibration. It should be noted that not all the items need to be considered, Chapter 4 presents
the results of a sensitivity analysis with HDM which shows which are the most important data
and parameters in terms of influencing the model predictions. These are where the emphasis
should be focused with less sensitive items receiving attention only if time and resources
allowing.
Table 2.1
RUE calibration data sources and hierarchy of resources
Manufacturer's
literature
Consultant's reports
Operator
organisations
Road condition
inventories
Survey calibration Small-scale Confirm intercept Estimation of tyre Estimate new local
research collecting values and assess and depreciation relations, especially
cost data from slope magnitudes costs for maintenance
companies costs
operating over a Determine vehicle Confirm slope
good range of utilisation by age values for total Compare predicted
highway and road condition costs RUE with rates
characteristics
Personnel with
industry knowledge
Rates survey
Route classification
Trained personnel
Analytical
capabilities
As shown in Appendix A, the following input data should be obtained for a Level 1
calibration:
While HDM often calls for a wide range of input data and calibration parameters, but only the
most important need to be established for use with a Level 1 calibration, so the HDM default
values should be used almost exclusively.
Some data items can be estimated with reasonable accuracy using short-term counts, for
example the hourly distribution of traffic volume, but the reliability is greatly enhanced by
collecting data over more sites over a longer period.
Fundamental research considers the relationships used in HDM. This consists of structured
field surveys and experimental studies conducted under local conditions which lead to
alternative relationships. For example, alternative functions may be developed for predicting
fuel consumption or new pavement deterioration and maintenance effects functions for
different pavement types. Such work requires a major commitment to good quality, well-
structured field research and statistical analysis over a period of several years. Pavement
deterioration research is a particularly long-term endeavour, typically requiring a minimum of
5 years.
1987) detailing the original research. These contain material on experimental design, theory,
model forms, analytical methods and empirical limitations.
3 Reliability concepts
3.1 Introduction
The objective of an HDM analysis is to model roads. This entails predicting the deterioration
of the pavement under time and traffic, the road user effects, and the effects of maintenance
on the pavement condition and rate of deterioration. As with any model, HDM is a
representation of reality. How well the model predictions reflect reality is dependent upon a
combination of the:
! Validity of the underlying HDM relationships
! Accuracy and adequacy of the input data
! Calibration factors used in the analysis
Since the underlying HDM relationships have proven to be robust and applicable in a number
of countries, the reliability of most HDM analyses depends on the input data and the
calibration factors.
This chapter addresses reliability issues in the context of HDM. It gives an overview of the
commonly encountered problems and describes statistical methods to improve reliability.
A systematic difference that arises between the observed and predicted values. For
example, if the predictions are always 10 per cent lower than the observed data. The
formal definition of bias is the difference between the mean predicted and mean
observed values.
! Precision
A measure of how closely the observed and predicted data are to each other1. It is
represented by the reciprocal of the variances (σobs2/σpre2); that is, it is reflected by the
scatter when plotting the observed versus predicted data. Precision is influenced the
inherent stochastic variations of most natural processes, measurement and observational
errors, and unexplained factors omitted from relationships in the model.
Figure 3.1 illustrates both of these concepts for four scenarios:
1
Precision here should not be confused with precision of a measurement which refers to the closeness of
repeated measurements to each other, for example, the weight is 10 kg + 0.1 kg.
In Figure 3.1 the shaded ellipse represents observed data which has been plotted against
predicted data. The solid line at 45° is the line of equality, where the observed and predicted
are equal.
Data
Predicted
Predicted
Data
Observed Observed
Predicted
Data
Data
Observed Observed
When there is low bias, the data will fall around, or close to, the line of equality. As
illustrated in the Figure 3.1-A (low bias and high precision), when there is high precision
there is little scatter in the data. However, a reduction in precision sees an increase in the
scatter and, thus, the standard deviation (Figure 3.1-B (low bias and low precision)).
When there is high precision, the change from low to high bias sees the slope of the observed
data systematically different to the line of equality. This is illustrated in Figure 3.1-C (high
bias and high precision). This figure also shows what happens when there additionally is a
systematic difference, the slope of the observed versus predicted line does not pass through
the origin.
The most difficult situation is illustrated in Figure 3.1-D (high bias and low precision). In this
instance it is often difficult to verify the reasonableness of the model since the differences
could be equally due to the poor precision as to the bias.
Mean Observed
CFrot = …(3.1)
Mean Predicted
This is referred to as the rotation correction factor because, as illustrated in Figure 3.2-A, the
predictions are rotated down to where they correspond to the observed data.
The translation factor is used when there is a constant difference between the observed and
predicted values across all conditions (see Figure 3.2-B). An example of this is where the
vehicle operating costs are overestimated due do overheads being improperly included. In this
instance the correction factor is:
Figure 3.2-C also shows the third, and common, scenario where there is a combination of
rotation and translation.
Mean Sampled
Bias = …(3.3)
Mean Population
A Rotation B Translation
Translation
Rotation
Observed = Predicted Observed = Predicted
Predicted
Predicted
Data Data
Observed Observed
Translation
Rotation and
C Translation
Rotation
Observed = Predicted
Predicted
Data
Observed
As illustrated in Figure 3.3, when the mean of the sample is different to the mean of the
population, the sample is biased. If there is the same mean but different variances, there is a
problem with the precision (see Figure 3.4).
In quantifying input data the objective is to ensure that the distribution of the sampled data is
similar as that for the population. This is achieved by using standard statistical sampling
techniques that ensure that there is a sufficient sample size to limit random sampling errors.
Appendix H describes the method for establishing the required sample sizes for different
levels of confidence.
frequency
Bias
Population Sampled
value
frequency Precision
Sampled
Population
value
by the model to acceptable levels. Level 2, being of a higher standard, is intended to produce
less residual bias in the model’s predictions than a Level 1 calibration.
A Level 3 calibration aims to improve both precision and bias by reconfiguring one or more
relationships within the model. It thus requires a comprehensive scope and database in order
to adequately detect and determine the many factors, effects and interactions.
! The ages of pavements were difficult to accurately estimate which meant that the traffic
loadings were inaccurate.
! It was difficult to estimate the post-construction roughness.
! The pavement strength was estimated from construction records. However, the variations
in roughness along a section of road indicated that there were local variations in strength
which were not reflected in the construction records.
! Variations in traffic growth over the life of the road were not accurately captured.
The sites selected for the study should cover the full range of pavement types and strengths
within the country. For each pavement type, pavements covering the complete range of
strengths should be included. If there are major differences in the climate over the country,
the experiment should be designed to account for this.
The data to be collected will depend upon the objectives of the study. For an adequate
calibration the following items need to be collected as a minimum:
! Roughness
! FWD/Benkelman beam deflection
! Cracking
! Rut depth
In selecting the sections it should be appreciated that if the pavements have been properly
designed for the traffic level it should be difficult to observe traffic loading effects as these
have been catered for in the design. One should therefore try and use pavements under-
designed for their traffic levels.
It is important that the test sections be continually monitored and that all the data items are
collected at the same time. In a number of studies this was not done and it created problems
with the subsequent analysis.
It is recommended that a vehicle mounted roughness meter (RTRRMS) not be used in
measuring roughness. This is because the measurement errors with such instruments are of
the same or greater magnitude to the incremental changes in roughness over time. These
meters are also prone to calibration problems that may lead to additional errors. The
roughness should thus be measured using one of the direct profiling techniques used for
roughness meter calibration (for example, Dipstick, Walking Profilometer). This will ensure
that the data are of the greatest possible accuracy.
It is important to monitor crack initiation and progression. Because of this, the test sections
should not have any maintenance done to them over the course of the study. This is often
difficult to arrange but is very important, particularly insofar as monitoring crack progression
is concerned.
1
One major problem lies in being able to accurately establish the amount of tyre wear. Since this is a small
amount weighing the tyre before and after a set of measurements often does this. However as described in
Transit (1997), even when using the most controlled conditions significant errors may arise.
between observed costs and the VOC predictions from the RTIM2, HDM-III Brazil and
HDM-III India VOC models with observed costs.
There are two components which are reflected in the tariff, the:
! Non-productive cost (for example, loading/unloading, repairs, looking for work)
where:
a1
RATE = a0 + …(3.5)
DIST
where:
NDLI (1997) used medium truck tariffs to check VOC predictions in India. Data were
collected from nine operators in the form of the tariffs charged for a standard load to travel
various distances. The data were quite variable, reflecting the different charges by operators
for trips of the same length. A regression analysis of the data gave the following equations:
562
RATE = 6.62 + …(3.7)
DIST
The analysis indicated that the marginal financial cost for truck travel was 6.6 Rupees/km.
The model was run using the financial unit costs instead of the economic costs. The predicted
total cost was 6.2 Rupees/km, a difference of 6 per cent. To evaluate the predicted costs
against the raw tariff data the total trip cost was calculated using the same constant cost of
562 Rupees/trip as from the tariff data. The results are plotted in Figure 3.5 against the
original tariff data (NDLI, 1997). It was concluded that the model was giving a reasonable
representation of the observed costs.
For passenger cars, taxi rates are often a ready source of data for evaluating the model
predictions. The rates need to be adjusted for the value of driver time.
Regression to
6000 tariff data
5000
Tariff in Rs/trip
4000
2000
1000
0
0 100 200 300 400 500 600 700 800
Trip Length in km
Figure 3.5 Comparison of predicted and tariff survey trip costs in Gujarat
du Plessis and Schutte (1991) contain a number of papers describing several studies
conducted by the CSIR into South Africa that involved fleet surveys. These provide good
guidance on how to approach a fleet study and analyse the data.
du Plessis et al. (1991) studied a single bus company since this eliminated any inter-company
effects, although it was noted that the seven local depots had some latitude with their
maintenance practices to reflect local conditions. The company operated on a range of terrain
and road types. The data were grouped into depots in flat, rolling and mountainous terrain and
average attributes were established for each depot (for example, roughness, and operation on
different road types).
It was noted that “any direct attempt to relate costs to roughness may yield misleading results
because of other factors such as bus age and maintenance policy that may play a role”. The
study was therefore initiated at each depot where the actual policies were investigated. This
also allowed for discussions with the local maintenance managers that provided valuable
insights into local conditions that were influencing the costs. For example, it was found that
the tyre consumption was influenced by the angularity of gravel through casing penetrations.
The data were used to develop relationships that were compared to HDM. The results broadly
confirmed the validity of the HDM predictions.
Findlayson and du Plessis (1991) conducted a similar study into trucks that developed
maintenance, tyre and depreciation relationships. These also gave similar trends to HDM.
4 Sensitivity of HDM
4.1 Introduction
It is important for users to be aware of the general level of sensitivity of the model to each
parameter1 so that appropriate emphasis can be given to important parameters and less
emphasis to second or third order effects. The influences of individual parameters differ
according to the particular parameter, the particular result being considered, and the values
assigned to other parameters in the particular analysis. The sensitivity of results to variations
in a parameter therefore varies somewhat under different circumstances.
Sensitivity analyses were conducted with the HDM RUE2 and RDWE sub-models so as to
determine the levels of sensitivity and to rank them3. Sensitivity was quantified by the impact
elasticity, which is simply the ratio of the percentage change in a specific result to the
percentage change of the input parameter, holding all other parameters constant at a mean
value.
For example, if a 10 per cent increase in traffic loading causes a 2.9 per cent increase in
roughness developed after 15 years, the impact elasticity term of traffic loading for that
roughness result is 0.29. If there were a 2.9 per cent decrease, the value would be -0.29.
As described by Mrawira et al. (1998), there are different approaches which can be used for
undertaking sensitivity analyses. The approach used here is the traditional ceteris paribus
method: changing a single factor while holding all others constant. The alternative approach,
using factoral experiments which combine all the levels of one factor with all levels of all
other factors, were not used due to the large number of combinations to consider. Thus, the
analysis here does not consider factor interactions. Mrawira et al. (1998) describes the results
of using a factor approach for an HDM sensitivity analysis.
On the basis of the analyses, four classes of model sensitivity have been established as a
function of the impact elasticity. The higher the elasticity, the more sensitive the model
predictions. These classes are listed in Table 4.1. Throughout the remainder of this report the
terms S-I to S-IV will be used to refer to the various sensitivity classes.
1
In order to run HDM it is necessary to supply the basic input data and values for the model coefficients; for
simplicity both of these are termed parameters.
2
As described in Bennett (1999), the analyses were done using a stand-alone version of the HDM-4 RUE
sub-model developed for this purpose.
3
The authors gratefully acknowledge the assistance of Rodrigo Archondo-Callao of the World Bank who
performed the RUE sensitivity analysis for HDM-III.
Table 4.1
HDM sensitivity classes
Practitioners, as a guide to where their efforts should be directed, should use the results of
these sensitivity analyses. Those data items or model coefficients with moderate to high
impacts (S-I and S-II) should receive the most attention. The low to negligible impact (S-III
and S-IV) items should receive attention only if time or resources permit. One usually
assumes the default HDM values for S-III and S-IV items since these will generally give
adequate results.
The former is important in situations such as when there are different route lengths. The
effects of operating conditions on RUE are important when you are comparing changes to
road condition.
As described in Bennett (1999), sensitivity analyses were conducted which considered both of
these issues. The first set of analyses considered the sensitivity of the total VOC on a per-km
to changes in the input data. The second considered the effects of roughness changes. For the
latter, the sensitivity of the difference in the total VOC between 3 and 5 IRI m/km to changes
in the input data was tested.
In order to cover the range of conditions encountered in different studies, the analyses were
done with unit cost data adapted from actual HDM studies from Australia, India, and
Thailand. These three data sets represent countries with markedly different levels of
economic development. Using these different studies meant that the relative contribution of
each VOC component was different thereby influencing the sensitivities. This ensured that
the results are broadly representative of the full range of conditions where the model will be
applied.
The individual data items were varied under a range of operating conditions and their impact
elasticities (see Table 4.1) were determined. On the basis of the results of these analyses, the
variables were assigned to the different sensitivity classes shown in Table 4.2 for HDM-4.
For reference purposes, the results from a similar analysis with HDM-III are given in Table
4.3.
Passenger Time
Roughness
(h/1000 km)
Crew Time
(h/1000 km)
Cargo TIme
(h/1000 km)
Power
Fuel Consumption
Requirements
(l/1000 km)
(kW)
Tyre Consumption
(Tyres/1000 km)
Oil Consumption
(l/1000 km)
Parts Consumption
(%NVP/1000 km)
Labour Hours
(h/1000 km)
Figure 4.1 HDM-III/95 Road User Effects components and their interactions
Traffic Influenced
Traffic Volume
Speed
and Capacity
(km/h)
Non-Motorised
Transport
Road Width,
Geometry, Passenger Time
Pavement Texture, (h/1000 km)
Speed Limit
Crew Time
Roughness
(h/1000 km)
Cargo Time
Acceleration Noise
(h/1000 km)
Power
Fuel Consumption Oil Consumption
Requirements
(L/1000 km) (l/1000 km)
(kW)
Tyre Consumption
(Tyres/1000 km)
Residual Value
Figure 4.2 HDM-4 Road User Effects components and their interactions
Table 4.2
Sensitivity classes for HDM-4 RUE model1
Notes:
1 The variables listed here are defined in Bennett and Greenwood (1999).
2 This is the highest class found over all common applications.
3 These are the variables important in determining the total RUE.
4 These are the variables important in determining the effect of roughness on RUE.
Table 4.3
Sensitivity classes for HDM-III RUE model1
Notes:
1 The variables listed here are defined in Watanatada et al. (1987a).
2 This is the highest class found over all common applications.
3 These are the variables important in determining the total VOC.
4 These are the variables important in determining the effect of roughness on VOC.
Structural
Crack Initiation Crack Progression Rutting
Deformation
Potholing Roughness
Ravelling Environmental
Ravelling Initiation
Progression Effects
80
60
40
20
0
0 5 10 15 20 25
Age in years
Table 4.4
Sensitivity classes for RDWE variables
Structural • • •
number 1/
Modified structural • • •
number1/
S-I > 0.50 Traffic volume •
3/
Deflection • • •
Roughness • •
Annual loading • • •
Age • •
Cracking progression •
factor
Potholing area • •
Subgrade compaction • •
Ravelling area •
Ravelling factor •
Notes:
1 Only one of the structural parameters is required, the other two are optional. When the
structural capacity is high relative to the traffic loading these parameters are in class S-II.
5.1 Introduction
Data represent the particular information that describes the case being analysed and are the
specific inputs to the model that are required of the user. Examples include traffic volume and
composition, road geometry, pavement type and condition, unit costs, etc. These data items
are thus fundamentally different from the internal parameters and coefficients of the
underlying simulation model which need calibration; an error in a data item means that the
model is analysing a different case from the one intended, much like entering a wrong
destination in a flight instruction.
For example, the accuracy of input data can have substantial impact on the timing of future
interventions, sometimes more important than the deterioration rate. This is because HDM
uses incremental models and the existing condition is the start point for the modelling. This is
illustrated in Figure 5.1 which shows for the same intervention criteria a difference in the
initial condition has markedly different times for future interventions.
Condition Confidence
Interval
Existing
Condition Criteria
for
1 Intervention
Condition Condition
2 1 2
Trigger Trigger
Interval Interval
Time
Just as information can be either imprecise or very precise, so the HDM model can be
operated with either very simple information or much more detailed information. However,
by the nature of simulation models and software, the input parameters of the model are fixed
and explicit. So this chapter describes how the user adapts various levels of locally available
data (whether simple or complex) to the required input parameters, and later reconverts the
results into local formats, where desired.
Finally, we look at how the user handles the related issues of accuracy and approximation.
These often have significant cost implications for the operation of a road management system
but also have an impact on the reliability of results and decisions.
System Performance
IQL-5 Performance
Monitoring
Planning and
IQL-4 Structure Condition
Performance Evaluation
Programme Analysis or
IQL-3 Ride Distress Friction
Detailed Planning
Project Level or
IQL-2 Detailed Programme
Project Detail or
IQL-1
Research
In road management, five levels have been identified for general use, as defined in Table 5.1:
! IQL-1
! IQL-2
A simple level of detail (simpler than IQL 1 or IQL 2), typically two or three attributes,
which might be used for large production uses like network-level survey or where
simpler data collection methods are appropriate.
! IQL-4
A summary or key attribute which has use in planning, senior management reports, or
alternatively in low effort data collection.
! IQL-5
Represents a top level such as key performance indicators, which typically might
combine key attributes from several pieces of information. Still higher levels can be
defined when necessary.
Table 5.1
Classification of Information Quality Level and detail
2 A level of detail sufficient for comprehensive programming models and for standard design
methods. For planning, would be used only on sample coverage. Sufficient to distinguish
the performance and economic returns of different technical options with practical
differences in dimensions or materials. Standard acquisition methods for project-level data
collection. Would usually require automated acquisition methods for network surveys and
use for network-level programming. Requires reliable institutional support and resources.
3 Sufficient detail for planning models and standard programming models for full network
coverage. For project design, would suit elementary methods such as catalogue-type with
meagre data needs, and low-volume road/bridge design methods. Able to be collected in
network surveys by semi-automated methods or combined automated and manual methods.
Table 5.3 serves to introduce the concept. At IQL-1, pavement condition is described by
twenty or more attributes. At IQL-2, these would be reduced to 6-10 attributes, one or two
for each mode of distress. At IQL-3, this reduces to 2-3, namely roughness, surface distress,
and texture or skid resistance. At IQL-4, this reduces to one attribute, Pavement condition (or
state or quality) which may have been measured by class values (good, fair, poor) or by an
index (for example, 0-10). An IQL-5 indicator would combine pavement quality with other
measures such as structural adequacy, safety aspects, and traffic congestion; that is
representing a higher order information such as road condition.
Local data can be adapted for use with HDM by determining a conversion for transforming
the data into the HDM parameters, and vice-versa. The basic approach for adapting local data
to HDM is first to transform the local data (from whichever IQL they may have) into IQL2
input parameters, and later to transform outputs from IQL-2 to the user’s desired output form,
typically IQL-3 or IQL-4. Additional data collection would only be required when essential
information items were missing. The approach is summarised as follows, and illustrated in
Figure 5.3.
1 The local data items are sorted into groups that relate to the HDM input parameter
groups (these groups are addressed in the following four sections).
2 Identify and record the IQL of each local data group (for information purposes), note that
where the IQL of the data appears to be mixed it would be worthwhile to consider a
separate review to normalise them around the most appropriate IQL.
3 If the local data is IQL-1, then the items will need to be combined or aggregated into the
HDM inputs, usually this is a process of selection rather than conversion since the HDM
inputs are directly measurable parameters such as area. These transformations are
external to the model.
4 For local data of IQL-2, there may be a need for conversion of measurement units to the
international units used in HDM, this is external to the model.
5 For local data of IQL-3, the information is less detailed than the HDM inputs; that is, one
attribute needs to be subdivided into two or more attributes. Thus, it is only possible to
estimate what the other attribute values might be and the answer is not unique. The
approximations will use an average or mid-point value, assuming that values above and
below will compensate each other and the final result will be reasonably accurate. The
conversion can be made outside HDM. However, for those items that have an internal
transformation built into HDM-4, the conversion will need to be checked against real
local data to ensure that it is reasonable and to adjust the conversion as needed.
6 For local data of IQL-4, the information is even less detailed than HDM and the
approximation is somewhat greater, but the procedures for adaptation are similar to those
for IQL-3. This is unimportant if the data item is in Sensitivity classes S-III or S-IV, but
deserves review if the data item is S-I or S-II, in which case consideration should be
given to changing the data collection to a more detailed IQL.
7 For adaptation of the output data to local items and terminology, the reverse process may
need to be applied. The same conversion formula should be used (in reverse) to ensure
that the model’s results will be reflected correctly in the local terminology. Adaptation
is generally less of a problem for output data since there are fewer variations in the IQL-
3 and IQL-4 definitions.
IQL - 1
A IQL - 2
HDM
HDM Standard
IQL - 2 Detailed
Analysis Reports
Data Entry
IQL - 3
B
IQL - 3 HDM Class
Data Internal Transformations
Options
IQL - 4
IQL - 4
120
100
80
Area of Ravelling (%)
E-0
E-1
E-2
60 E-3
E-4
E-5
Means
40
20
0
0 1 2 3 4 5 6
"Enduit" (E) score
Table 5.2
Conversions between condition units and HDM-III input parameters - example
from Niger office of road management
Cracking (Fissuration) F 0 1 2 3
Deformation D 0 1 2 3
(deformation)
Notes:
1 The incidence of potholes is recorded only as 4 or 5.
2 Predominantly only wide cracks are visible from a moving vehicles. If the cracks are
widely spaced, these values should be reduced applying a factor of 0.4.
3 The rut depths were generally indiscernible (< 5 mm), even for high classes of
deformation.
! IQL-2 - the actual level of direct inputs into HDM-III and HDM-4
! IQL-3 - class-type information for the key attributes within each data group
The more detailed level, IQL-1, for each parameter is not shown for reasons of clarity but
comprise the fundamental parameters that describe the detailed attributes of an IQL-II data
parameter.
Table 5.3 shows the recommended grouping and aggregation of data items. Details of
definitions, classification ranges and conversion relationships are given in Appendix F .
Table 5.3
IQL examples for road data
5.5.4 Environment
There are 13 parameters defining aspects of environment for the various pavement types at
IQL-2. These can be approximated and estimated by three class parameters at IQL-2B:
! Rainfall class
! Climate class
! Cold climate classification
At IQL-3 and IQL-4, these can all be summarised by one climate classification with several
classes.
Table 5.4
IQL examples for traffic data
Volume growth
(%/yr) – class 1
Volume growth Volume growth (%
Volume growth (%/yr.) (all classes) AADT/yr)
(%/yr) – class n
Duration of F-F
period 1 (h)
Percent AADT in
F-F period n
Safety To be completed
Emissions To be completed
6.1 Introduction
The RUE model calibration focuses on ensuring that the key RUE model parameters and
calibration factors are appropriate for the conditions under which the model is to be applied.
As described in Chapter 2, there are three levels of calibration that entail different levels of
resources and time. These are:
! Level 1 - Basic application (see Section 6.3)
Determines the values of required basic input parameters, adopts many default values,
and calibrates the most sensitive parameters with best estimates, desk studies or minimal
field surveys.
! Level 2 - Calibration (see Section 6.4)
Undertakes major field surveys and controlled experiments to enhance the existing
predictive relationships or to develop new and locally specific relationships for
substitution in the source code of the model.
Figure 6.1 shows the recommended priorities for the RUE calibration. This shows the data
which are required, of first and second priority, and which should have the defaults assumed
for.
Assume
Required Priority 1 Priority 2
Defaults
Mass
Power
Capacity
Tyre Parameters
Level 1 Unit Costs Speed
Depreciation
All Others
Service Life
Parameters
Utilisation
Fuel
Speed
Power
Level 2 Capacity
Service Life
All Others
PCSE
Utlilisation
Priorities
Depend on
Level 3 Resources and
Objectives
This chapter commences with a discussion of selecting representative vehicles. This is then
followed by procedures for calibrating the model parameters to each of the three levels.
If adequate data are available, the medium and heavy trucks should be further disaggregated
since these vehicles tend to have the widest variations in the RUE and the greatest impact on
pavement performance.
Mass
The vehicle mass influences the vehicle speeds, fuel and tyre consumption and, through the
associated heavy vehicle damage factor, has a major impact on the rate of pavement
deterioration.
The influence of mass on RUE is not a major issue for pavements in flat terrain. However, the
presence of gradients will result in a major increase in fuel consumption and this increase is
proportional to the vehicle mass.
Table 6.1
Representative vehicles adopted in different HDM-III studies
where:
With this approach, for container trucks standard container weights of 2 t and 4 t can be
assumed.
∑
4
AX i
VDFVEH k = …(6.2)
SX i
i=1
∑ VDFVEH
k =1
k
VDF = …(6.3)
z
where:
It is common to divide the stream into similar vehicle class, for example, medium, heavy and
articulated trucks, and calculate a VDF for each class.
The standard axle loads (SXi for different configurations are (Watanatada et al., 1987c):
Introduction
HDM-4 and HDM-95 uses the speed-flow model proposed by Hoban (1987) and shown in
Figure 6.2. This requires the user to provide five key parameters:
! Qnom - the nominal capacity of where all vehicles are travelling at the same speed
Table 6.2 gives the default HDM parameter values. In the discussion that follows the term
capacity is applied to the ultimate capacity, unless otherwise noted.
S1
S2
Speed in km/h
S3
Snom
Sult
Qo Qnom Qult
Flow in pcse/h
Table 6.2
Default HDM speed-flow model parameters
Estimating capacity
The capacity of a facility is defined as the maximum hourly flow rate at which vehicles can
reasonably be expected to traverse a point or uniform section of road under prevailing road,
traffic and control conditions. While often considered to be a single fixed value, there are
always inherent variations due to factors such as driver behaviour, vehicle performance and
prevailing conditions. One should therefore calculate an average capacity that will
encompass these variations.
It is beyond the scope of this report to address capacity issues in detail and readers should
refer to reports such as the Highway Capacity Manual (HCM) for a fuller discussion (TRB,
1997). McLean (1989) gives a good discussion of capacity issues and provides insight into
the background behind the two-lane highway values adopted in the HCM.
For a Level 1 calibration the values in Table 6.2 should be adopted as defaults. However, it
must be appreciated that these are ideal capacities and they must therefore be adjusted to
reflect typical operating conditions. For example, the HCM reduces the ideal capacity to
account for different directional splits (for example, 60/40 instead of the ideal 50/50) and the
presence of no-passing zones. These reductions can be quite significant, reducing the ideal
capacity by 20 per cent or more.
Table 6.3 to Table 6.5 give capacity reduction factors from the Highway Capacity Manual
(TRB, 1997) for two-lane highways. These factors should be applied in a Level 1 calibration,
albeit with caution as they may not be entirely appropriate for conditions outside of those
from which they were developed. The ideal capacity is multiplied by these factors to reduce
it to a value that reflects actual operating conditions. Multi-lane factors are given in TRB
(1997).
Table 6.3
Two-lane highway capacity reduction factors - no passing zones
Table 6.4
Two-lane highway capacity reduction factors - directional split
Table 6.5
Two-lane highway capacity reduction factors - lane and shoulder width
Note: Where shoulder width is different on each side use the average width.
! Physical life - is the period which the vehicle exists (even if it is not being used)
! Economic life - is the period which the vehicle is economically profitable to operate
With HDM, the service life is of interest. It is used by HDM in calculating the depreciation
costs of vehicles that can have a significant impact on the RUE.
In HDM-4 the user needs to define the expected service life in kilometres for a vehicle
operating on a smooth pavement. This value is then used to determine the effect of roughness
on service life when using the Optimal Life technique. This expected service life is the
distance at which it becomes appropriate to scrap the vehicle.
There are a number of different techniques available for calculating the service life and an
overview of these may be found in Winfrey (1969). For a Level 1 calibration the ages of a
sample of vehicles should be obtained, either from a small survey or by sampling
advertisements of vehicles for sale. Daniels (1974) indicates that the service life will be
double the mean age. This was also found to be the case in New Zealand (Bennett, 1985)
where several different techniques were tested for estimating the service life.
kilometreage of a sample of used vehicles. Dividing the kilometreage by the age gives the
average kilometreage of the vehicle over its life. Problems may be encountered with older
vehicles wherein the odometer may have cycled beyond 100,000 km (or miles) or the ages
may be in error. However, if a sufficiently large sample of data is obtained these problems
will be minimised.
This is the number of hours per year (8760), less the time allowed for crew rest, time lost
loading, unloading, refuelling, finding cargo, repairs, etc.
! HRD - the numbers of hours driven
This is the hours that the vehicle is operated. It can be calculated from the annual
kilometreage divided by the average annual speed.
! HWK - the number of hours worked
This is similar to the hours driven (HRD), except it includes the time spent loading,
unloading and refuelling.
The HDM-III adjusted utilisation model was based on the hours drove. HDM-4 is based on
the hours-worked approach.
Using a standard working week, a vehicle is typically available for approximately 1800 hours
per year. However, since there are substantial periods of time when the vehicle is not in use,
for example due to loading/unloading, the driving time would often be less than 50 per cent of
this value. In the Brazil study, for example, the vehicles were available 839 - 2414 h, but only
driven 652 - 1863 h (Watanatada et al., 1987c). Trucks and buses had the highest utilisations;
utilities the lowest.
As discussed in Bennett (1995), it is important that the value adopted for hours driven be
consistent with the annual utilisation and the average speed. If not, the predicted costs could
be distorted. To this end it is recommended that in the absence of more detailed data, the
hours driven be calculated using the following equation:
AKM
HRD = …(6.4)
S0
where:
To calculate hours worked it is necessary to have sufficient data to identify the times spent
“undertaking the essential tasks of making a complete round trip, in normal circumstances.
Time spent idle, where the crew is eating, sleeping or otherwise resting should not be
included. Time in repair should also, in general, be excluded because it is not part of a regular
trip. Driving, loading and unloading should be included together with refuelling” (Hine,
1996).
One may also include the administration time that a driver had to spend finding loads or the
time that must be spent waiting with the vehicle to move it up the queue, if the driver can’t do
anything else. It should be appreciated that in some circumstances not all loaded time will be
working time, for example if the driver stops the vehicle to sleep while it is loaded.
Working time is established by conducting small surveys covering vehicle activities over
several days. The work of Hine (1996) in Pakistan gives valuable insight into issues arising
from these surveys and special considerations that must be made in approaching the issue.
This can be expected to differ considerably between countries, and even regions within
the same country. VDESIR represents the maximum speed of travel adopted by the
driver of a vehicle when no other physical constraints, such as gradient, curvature,
roughness or congestion, govern the travel speed. The value of VDESIR is influenced by
factors such as speed limits and enforcement, road safety, cultural and behavioural
attitudes.
! β - the ‘draw down’
Indicates how far from the constraining speeds the predicted speed will be. β is the
Weibull Shape Parameter. As described in Watanatada et al. (1987c), it is functionally
related to the dispersion of the underlying distribution of the constraining speeds.
This assumes that the physical performance of the vehicle has been properly calibrated by
identifying valid representative values of other important vehicle characteristics, namely:
! Vehicle mass
! Used driving power
! Braking power
For calibration purposes it is important to understand the role of β. The HDM speed model
predicts that speeds are the probabilistic minimum of five constraining speeds based on:
! Power
! Braking
! Curvature
! Roughness
! Desired speed
As shown in Figure 6.3 (Bennett and Greenwood, 1999), when β approaches zero, the mean
speed for a given road section would be equal to the minimum of the five constraining speeds.
The greater the value for β, the further away the predicted mean speed will be from the
constraining speeds. Thus, if the desired speed is 100 km/h, with a value of β = 0, the
predicted speed would be 100 km/h. However, if a non-zero value for β is used, the predicted
speed would be less than 100 km/h.
In calibrating the model, Watanatada et al. (1987c) found value of β=0.24-0.31 for Brazil. A
similar calibration from India found β=0.59-0.68. Watanatada et al. (1987c) considered that
the higher values for β reflected the more congested traffic conditions in India. It was
recommended that “For environments, such as India, which have congested rural roads and
low level of traffic discipline, somewhat higher values of the β parameter (than Brazil) may
be used”. They note that the quantification of β would require a major field study, that is, a
Level 2 calibration, so it is discussed later in Section 6.4.1.
30
β = 0.01
25
β = 0.2
β = 0.4
20
Speed in m/s
15 β = 0.6
10
0
-10 -8 -6 -4 -2 0 2 4 6 8 10
Gradient in per cent
section and then these values averaged1. As described above, it may also be necessary to
adjust the value for β to ensure a reasonable value for VDESIR.
Diesel vehicles
where:
This equation is from Watanatada et al. (1987c) and so applies to older technology vehicles
(that is, pre-1985). Bennett (1994) indicated that for modern vehicles the factor should be
0.75. His data showed that larger vehicles tended to use more power and the following
alternative equation was developed relating power to mass:
Petrol vehicles
The modern technology petrol equation is an update of the old technology equation from
Watanatada et al. (1987c) using data from Bennett (1994). For use in HDM-III the values will
need to be divided by the factor of 0.736 to convert to MPH.
The SAE rated engine power is generally available from motor publications and
manufacturers. A weighted-average should be calculated based on the frequency of different
vehicle types, and thus engine powers, in the vehicle fleet.
In applying these equations it should be noted that there is evidence that there is a
relationship between the gradient and power usage, with vehicles having higher power usage
on higher grades (Bennett, 1994). It may therefore be necessary to increase the driving power
from those predicted using the above equations for vehicles operating in hilly conditions.
1
If there are marked differences in the values of VDESIR it may be appropriate to eliminate the greatest
outliers - as well as reviewing the estimated speeds.
2
Engine power conversions: 1 BHP = 0.746 kW; 1 MPH = 0.736 kW; 1 BHP = 0.987 MPH
10.00 R 20 14 (G)
Load Range
Ply Rating
Rim Diameter in Inches
Radial Construction
Nominal Section Width in Inches
Ply Rating
Radial Construction
Aspect Ratio
The ply rating and load range usually only apply to heavy vehicle tyres so the standard
typology reduces to:
xxx/yyRzz …(6.7)
where:
For light vehicles the aspect ratio of 82 is often omitted so they are specified, for example, as
175/R13 instead of 175/82R13. The value of 82 can be assumed for heavy vehicles when
omitted.
1
The aspect ratio is the tyre’s section height, which is the distance from the bead to the centre of tread, to the
section width. An aspect ratio of 65 means that the tyre’s section height is 65% of the tyre’s section width.
2
A less common typology uses the nominal width in inches and the aspect ratio separated by the slash, for
example, 14/80R20. Because of the much lower magnitude of the nominal width, it is readily apparent
when this case arises that the data should be converted to mm.
The tyre type can be established from manufacturer’s specifications since motorists tend to
purchase the same replacement tyres as new vehicles. Discussing with tyre retailers can check
this information. Alternatively, a small study can be conducted which records the tyres on a
sample of vehicles, for example in a parking lot or at a truck stop.
For Level 1 calibration the wheel diameter can be estimated from the tyre typology using the
following equation (Greenwood, 1997):
yy
DIAM = 25.4 zz + 2 xxx …(6.8)
100
The number of wheels can be estimated from the manufacturer’s specifications or from a
small study.
Tyre Casing
Grooves
Minimum Legal
Tread Depth
Maximum Depth
of Grooves
Length of Lower Trapezoid
Base (LL)
LL + LU
AREA = (DE − DEmin ) ARUB …(6.9)
2
where:
The volume of wearable rubber is calculated from the tread area as:
PI DIAM AREA
VOL = …(6.10)
1,000,000
where:
To calculate the tyre volume it is therefore necessary to predict the tyre diameter, the tread
depth, the tread width and the area of rubber versus grooves.
The tread width as a function of the nominal section width is predicted as (Bennett, 1998):
LL = a0 xxx - a1 …(6.11)
where:
It was found that the coefficients for Equation 6.11 above varied by vehicle class so separate
equations were developed for light and heavy vehicles. The coefficients are given in Table
6.6.
The tread width LL corresponds to the lower length of the trapezoid base in Figure 6.5. For
the upper width (UL) Bennett (1998) recommended an increase of 10 mm for heavy vehicles;
6 mm for light vehicles and 4 mm for motorcycles. Table 6.6 gives these recommended values
by vehicle class.
Table 6.6
Wearable rubber model parameters
Tread depths are available from manufacturer’s specifications, or can be easily measured
from a sample of tyres, or the values given in Table 6.6 can be used. This table also contains
the values for ARUB: the tread area as a decimal (Bennett, 1998). The latter can be easily
measured by recording the groove widths and frequency of grooves on a tyre.
The volume of rubber is calculated as follows:
1 Establish the tyre size for the representative vehicle (for example, 175SR13).
2 Calculate the tyre diameter from the tyre typology using Equation 6.8 above.
3 Calculate the tread width LL from the tyre typology using Equation 6.11 above.
4 Establish the value for UL relative to LL from Table 6.6 (for example, for passenger cars
UL = LL + 6).
5 Establish the value for ARUB from Table 6.6 (for example, for passenger cars ARUB =
0.85).
6 Establish the tread depth DE from a survey or from Table 6.6.
7 Establish the legal minimum tread depth DEmin or, if there is no enforcement, assume a
value of 0 mm.
8 Substitute the values for ARUB, DE, DEmin, LL and LU into Equation 6.9 above to
establish the tyre area.
9 Substitute the tyre area and diameter into Equation 6.10 above to establish the volume of
wearable rubber.
With the constant life method, the depreciation is calculated as the replacement value less
residual value1 divided by the vehicle life. When less than 50 per cent of the trips are for
private use, this life is adjusted by the number of hours worked. The calibration of this
method is therefore achieved using the calibrated values for annual utilisation, service life
and hours worked.
For the optimal life calibration, the user needs to establish the average service life of the
vehicle at an average roughness. Using the HDM Tools software (See Appendix K ), the
service life as a function of roughness is predicted. A regression equation is then fitted to
these data and used in HDM-4. Details of how this is done are given in Appendix K .
1
The residual value is assumed to be 1% of the replacement vehicle price.
For a Level 1 calibration it can be assumed that the default HDM values are adequate,
although if the analyses include steep gradients it may be desirable to calculate new values.
This can be done using the following equations:
where:
The HDM-III equation is from Watanatada et al. (1987c); the HDM-4 is from an analysis of
the default parameter values from NDLI (1995a). For use in HDM-III the values will need to
be divided by the factor of 0.736 to convert from kW to MPH.
Having established the mean operating speed, one then uses this in conjunction with HDM-4
to determine the appropriate value for VDESIR. HDM-4 is run iteratively using the same
road section parameters as prevailed in the speed survey varying VDESIR until the predicted
and observed speeds agree.
It should be recognised that the desired speed is a function of bendiness. This effect was
observed in several studies when vehicles travelling on roads with severe alignments had
lower desired speeds (McLean, 1991). NDLI (1995a) proposed a negative exponential model
should be used to reduce the desired speed as a function of bendiness within HDM-4.
However, on testing it was found that this model gave unreasonable predictions at extreme
levels of bendiness so it was not included in HDM-4.
If HDM will be applied at different levels of bendiness, it would be prudent to establish
separate values of VDESIR to apply at these different levels.
Width
The HDM-4 desired speed-width model is illustrated in Figure 6.6. It is based on the work of
Hoban et al. (1994) and Yuli (1996). The underlying assumption is that there is critical width
below which speeds will be unaffected by width (CW1). Between this minimum speed
(VDESMIN) and the desired speed on two-lane highways, there is a linear increase in the
speed. On roads wider than two-lanes there is a continued increase in speed, but at a much
lower rate.
The minimum desired speed, in HDM-4, is assumed to be 75 per cent that of the two-lane
highway desired speed. Conducting free speed studies on narrow and two-lane roads can
check this value. It is important that the roads studied have similar levels of roughness and, if
possible, be on flat, tangent sections. By holding all other factors constant the differences in
speeds will be due to width. A similar approach is used when roads are wider than two lanes
to obtain the speed increase slope.
VDESIR' in m/s
a3
1m
a2
1m
a1
a0
Roughness
The speed-roughness model is calibrated through the maximum rectified velocity.
Data for calibrating the maximum rectified velocity are best obtained from a before and after
study on a pavement selected for maintenance which will reduce the roughness. If there are
no changes to the width or alignment, any increases in speed after the maintenance is
attributable to the roughness reduction. Section 7.5 describes special considerations in
undertaking such a study, particularly the need for control sections to check the roughness
meter calibration.
Table 6.7 shows the change in speeds before and after overlays from such a study in India
(NDLI, 1997). In all cases there was an increase in speed accompanying the decrease in
roughness.
Calibration of the model is done by running HDM-4 at each of the roughnesses in the before
and after study and recording the predicted speeds. The other speed model parameters should
reflect the conditions of the speed survey sites.
The speed-roughness slope (km/h/IRI) should be calculated and compared to that from the
studies. The ratio of the means is used to modify the maximum average rectified velocity used
as input to HDM-4. Since the speed-roughness model is non-linear, the process should be
repeated at least two times to ensure that consistent results are obtained.
Table 6.7
Data on speed-roughness study
(km/h/IRI)
Before After Change Before After Change By site Mean
1 1 61.1 65.4 4.3 5.7 4.0 1.7 2.5
Curve speed
In order to calibrate the curve speed model it is necessary to undertake a series of
measurements on roads with different radius of curvature. HTC (1999) and Bennett (1994)
describe in detail how to undertake such studies. Data are collected on speeds at the
midpoints of curves over a range of curve radii and analysed to develop a non-linear model
relating curve speed to the radius of curvature. Figure 6.7 is an example of the data for
passenger cars from Thailand (HTC, 1999).
90
80
70
60
Curve Speed in km/h
50
40
30
20
10
0
0 100 200 300 400 500 600
Curve Radius in m
VCURVE = a0 × R a1 …(6.14)
where:
HTC (1999) discusses in detail the analytical requirements for developing such a model.
Capacity
As described earlier with Figure 6.2, HDM requires the nominal and ultimate capacities, and
a Level 2 calibration undertakes field trials to establish these capacities.
As shown in Figure 6.8, there are several different methods available for estimating capacity
(Minderhoud et al., 1997). These authors make an important observation: “Attempts to
determine the capacity of a road by existing methods will generally result in a capacity value
estimate, but the validity of this value is hard to investigate because of the lack of a reference
capacity value, which is supposed to be absolutely valid. A clear, reliable method does not
appear to be available at this time.”
Capacity
Estimation
Observed
Observed
Observed Observed Volumes, Guidelines
Volumes and Simulation Models
Headways Volumes Densities and (eg HCM)
Speeds
Speeds
Irrespective of what method is used, it is necessary to sample over sufficient time intervals to
obtain a reliable estimate. One normally measures the flow in 15-minute periods and
converts this to an hourly flow. In most situations capacity is a difficult condition to reliably
encounter so one must often extrapolate data collected at lower flows to estimate the capacity.
The capacity estimates must consider the range of factors influencing capacity. In Section
6.3.2 a series of default capacity reduction factors were given. The capacity analyses should
consider these to establish local values.
Width is a particularly important factor. The HDM approach results in a step-function of
capacity versus width. However, as shown in Figure 6.9, other studies have treated capacity
as a continuous function of width. This has the advantage of avoiding discontinuities that can
arise at the boundaries of step functions. Figure 6.9 also shows the different capacities found
in different countries, and highlights the importance of local calibration.
3500
3000
HDM-95
HDM-95 Wide Two-Lane
Two-Lane
2500
Australia
(McLean, 1989)
Capacity in pcse/h
2000
1500
HDM-95
Narrow Two-Lane
1000
500
HDM-95
Single Lane
0
0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0
Road WIdth in m
From a practical perspective, it is important to focus on the pavement widths that are most
relevant. Many applications consider incremental widening in the range of 6-8 m and in this
band minor width changes can have a marked impact on capacity. If pavements are less than
6 m wide, they often deal with a major widening of several metres that will have a substantial
increase in capacity, analogous to the step functions in HDM. Conversely, small changes in
width below 6 m have little impact on capacity. One should therefore establish realistic
capacity estimates that will reflect the operational impacts expected.
In the context of HDM speed modelling, both the nominal and ultimate capacities are of
interest. As described in HTC (1999), the ultimate capacity prevails over short time periods so
is best represented by the maximum flow observed on a road. The nominal capacity is the
maximum flow rate that can reasonably be expected to be maintained so it should be based on
a longer period. In Thailand HTC (1999) used 5 minute intervals for the ultimate capacity and
10-15 minute intervals for the nominal capacity.
Given the limited data available for most Level 2 calibrations, there are two techniques that
should be considered for estimating the capacity:
! Headways
! Observed flows
Headways are the time difference, in seconds, between successive vehicles (usually measured
from rear bumper to rear bumper although if using axle detectors from front axle to front
axle). The capacity of the road is defined as:
3600
Qult = …(6.15)
hc
where:
The method is based on the theory that at high flows there are still two populations of
vehicles:
! Constrained (followers)
! Unconstrained (leaders)
The distribution of following headways is expected to be the same as for constrained drivers
in any stationary traffic stream.
Although relatively easy to collect, it is considered that this method tends to overestimate
capacity due to the assumption that the distribution of constrained drivers can be compared at
capacity with that below capacity.
With observed flows, data are collected on roads that will reach capacity at some point during
the study period. The road capacity is taken to be the maximum flow, or the mean of several
very high flows, observed during the analysis period. This is illustrated in Figure 6.10.
Flow Rate
Maximumum Flow
Averaging Interval
Cycle [hour]
(15 min)
Figure 6.11 is an example of speed-flow study results from Thailand (HTC, 1999). This figure
shows the capacity of 2121 PCSE/h. This was based on 5-minute peak flows measured using
data loggers.
The disadvantage of this approach is that the roads may not always reach capacity. However,
if the sites are carefully selected this situation can be minimised. One way of ensuring high
flows is to plan the data collection around special events, such as holidays when traffic flows
can be expected to be high.
120
Speed in km/h 80
60
40
20
0
0 500 1000 1500 2000 2500 3000
Flow in pcse/h
Ultimate capacity
As noted by HTC (1999): “The speed at ultimate capacity is notoriously difficult to estimate.
This is because the flows become unstable in that region when the demand exceeds capacity
there is a breakdown in the flow that leads to markedly lower speeds.” Measurements of
speeds at flows such as shown in Figure 6.11 should be used to determine an appropriate
speed.
1
There are a variety of methods for assessing the validity of a regression equation. The easiest is to ensure
that the 't' statistics of the regression coefficients are significant at 95% confidence. HTC (1999) found that
this yielded the best results.
Table 6.8
PCSE values by vehicle class
Recommended values
Notes:
1 Basic data from Hoban et al. (1994)
2 Time headway calculated from space headway using 72 km/h speed
3 Truck and Trailer average length increased to 15 from 11 given in Hoban et al. (1994)
based on difference between total space and space headway
1
The PCSE is analogous to truck equivalencies calculated using the headway method. This defines the
equivalency of a truck as, the ratio of the average headway for trucks in the stream (in s), to the average
headway for cars in the stream (in s). See McLean (1989) for a detailed discussion of equivalency factors
and their determination.
The PCSE values in Table 6.8 were calculated by firstly establishing the total space occupied
by passenger cars (default = 36.0 m). The Basic PCSE values for the other vehicle classes
were then determined from the ratio of their total space to the passenger car space. This basic
value was then subjectively adjusted for width effects to obtain the recommended values in
the final three columns.
Local calibration of the PCSE values can be done by adjusting the basic assumptions in Table
6.8 to reflect local vehicles. The easiest data to obtain is that on vehicle lengths. Field studies
could focus on establishing the average headways for vehicles thereby giving the total space.
Axle detectors are quite useful in this type of work since the time between the first axles of
successive vehicles divided by the velocity (in m/s) gives the total space occupied.
Overview
The HDM-III and HDM-4 fuel consumption models are based on mechanistic principles.
HDM-III was developed using a regression-based approach while HDM-4 is more of a pure
mechanistic formulation that more effectively reflects the different components affecting fuel
consumption. To calibrate the fuel consumption model parameters it is either necessary to
have a series of raw fuel consumption measurements or, alternatively, existing fuel
consumption models, such as those developed using multiple linear regression techniques.
Each of these will be discussed individually. The discussion will focus on the HDM-4 fuel
model, with some reference to HDM-III.
where:
The fuel model calibration therefore focuses on the parameters α and ξ. It is assumed that the
models used to predict the total power have been correctly calibrated elsewhere. The
equations used to predict the total power are as follows:
Ftr = Fa + Fg + Fr …(6.18)
Fa = R1 v 2 …(6.19)
Fg = R2 GR …(6.20)
Fr = R3 + R4 v 2 …(6.21)
The variables R1 to R4 are simplifications of the variables in the actual HDM-4 equations:
R2 = M g …(6.23)
v
Ptr = Ftr …(6.27)
1000
Ptr =
[(R3 + R2 GR ) v + (R1 + R4 ) v 3 ] …(6.28)
1000
where:
! ξb)
Base engine efficiency (ξ
! Change in engine efficiency at higher power
Ptr
Ptot = + Pengaccs …(6.29)
edt
6 Divide the fuel by the total power (Ptot). This gives the power-to-fuel conversion factor
ξ for that power.
7 Undertake a regression analysis and fit the model parameters ξb and ehp in the model:
ξ = ξ b1 + ehp
Ptot
…(6.30)
Pmax
where:
8 If the analysis does not allow for the simultaneous estimation of ξb and ehp, assume the
HDM-4 default relationship for ehp.
Generating data
1 Use the existing models to generate a series of fuel consumption predictions covering the
full range of operating conditions likely to be encountered, for example: speeds,
roughness, gradients.
2 Make an assessment of the range of conditions typically encountered and weight the data
set accordingly. For example:
70% operate on flat terrain between 50 and 80 km/h;
20% on rolling terrain between 40 and 70 km/h;
10% in mountainous terrain between 20 and 50 km/h.
3 Create a data set that contains data reflecting the assessments from (2). This will ensure
that the results have the appropriate weightings.
4 From this data set, follow the instructions given above for calibration with raw fuel data.
Model coefficients
As described in NDLI (1997), a commonly used model form that has been fitted to fuel
consumption data is:
where:
It can be shown that this model is compatible with the HDM-4 model. Ignoring the fall term,
the following are the terms that correspond between the two models:
Regression HDM-4
A0 + A3 IRI (ξ/edt) R3
A1 3600 ξ Pengaccs
A4 (ξ/edt) R2
The terms A0 and A3 are thus mainly related to rolling resistance; A1 to the fuel required to
operate the engine; A2 to aerodynamic resistance, and A4 to the gradient resistance.
Calibration is done as follows:
1 Calculate the SFC for a range of conditions similar to those described above under
Generating Data.
2 The idle fuel consumption α is given by A1/3600, or ξ Pengaccs.
3 Subtract A1/S from the total SFC to obtain the fuel consumption due to forces opposing
motion. Convert this to the IFC in ml/s by dividing SFC by the speed.
4 Calculate the total tractive power using the HDM equations with the vehicle
characteristics and divide by the driveline efficiency.
5 Calculate ξ by dividing IFC by the tractive power, corrected for driveline efficiency.
In applying this method in India, NDLI (1997) found some inconsistencies between the results
from Indian equations and the default values in HDM-4. For some vehicles the values were
outside of the range of values considered viable, while for others they agreed well. This
method should therefore be applied prudently.
Overview
Acceleration noise is used in HDM-4 to predict the effects of speed variations due to
congestion or road conditions on vehicle operating costs. It is a relatively simple parameter to
measure and can be done through a small controlled experiment. Once the appropriate
equipment has been obtained, the work should only take a few days.
Figure 6.12 is an example of the acceleration profile of a vehicle on a motorway (NDLI,
1995a). It can be observed that the accelerations before entering the motorway were much
greater than once on the motorway. If a vehicle has the same start and end speeds over a
section, the mean acceleration will be zero. HDM-4 therefore uses the standard deviation of
acceleration, called the acceleration noise to represent the magnitude of accelerations; the
higher the acceleration noise, the greater the magnitude of the accelerations.
3
Entering Motorway Once on Motorway
1
Acceleration in m/s/s
-1
-2
-3
0 50 100 150 200 250 300 350 400 450 500
Time in s
! Driver (σadr)
! Roughness (σairi)
where:
HDM-4 uses a sigmoidal function for the traffic interactions and a linear function for the
others. The acceleration noises are combined to obtain the total acceleration noise for a given
road section.
A Level 2 calibration of any of the model consists of making measurements at two extremes
and assuming that the existing model applies. Level 3 calibration sees the model formulations
requantified.
It should be noted that it is not necessary to calibrate all components of the model. If one is
looking only at the effects of capacity improvements, the traffic interaction model should be
calibrated and possibly the NMT/side-friction model. If roughness improvements are of
interest then only the roughness model needs to be calibrated.
model. Further information is available through the HTC Infrastructure Management web site
given in the About This Manual section of this document.
Irrespective of which component(s) of the acceleration noise model are being investigated,
the basic approach is the same:
! Outfit a vehicle with the measurement equipment. The vehicle should be as
representative of the vehicle class as possible. Ideally, the measurements should be made
with several vehicles and several drivers, but NDLI (1995a) found that the results were
broadly similar so this is not critical.
! A test route should be established which would allow for the two critical points to be
monitored:
❏ low impact
❏ high impact
For example, if roughness is of interest the measurements should be made on a low and
high roughness road, for traffic impacts on non-congested and highly congested roads. It
is vital that all the other road attributes (for example, width, alignment, traffic
composition) be similar. That way, any differences in acceleration noise can be
attributed to the parameter of interest.
! It is important that the route be long enough for the process to stabilise. Accelerations
are assumed to be random and if too short a route is used the process will not stabilise.
Figure 6.13 shows the effect of sampling interval on data based on an analysis of the
NDLI (1995a) measurements (Greenwood, 1998). On the basis of this it is recommended
that a minimum sample of 5 minutes be collected in a single run for each attribute of
interest.
1.0
0.9
0.8
A cceleration Noise Results (m/s²)
0.7
0.6
0.5
Max
0.4
Min
Mean
0.3
0.2
0.1
0.0
0 30 60 90 120 150 180 21 0 240 270 300 330 360 390 420 450 480 510 540
The measurements should be made on as good a standard road as is available for the type of
road of interest. HDM allows data to be input for different road classes so measurements can
be made, for example, on two-lane and multi-lane highways.
If no other measurements are made (for example, alignment) it is assumed that the natural
noise is entirely due to the driver: σan = σadr.
The value for σatmax is used in conjunction with a sigmoidal relationship to predict the
traffic noise at intermediate volume-to-capacity ratios.
to the rear axle and another immediately above on the cargo tray. By mounting the axle and
body accelerometers in close proximity to each other, the degree to which roughness induced
axle vibrations were transmitted to the vehicle body could be determined.
The vehicles were driven at different speeds over sites with different roughnesses and the
accelerations were recorded. Figure 6.14 is an example of the body accelerations versus
roughness for medium trucks1. It can be seen that the body accelerations increase with
increasing roughness and this can be used to calibrate the roughness effects in the HDM parts
consumption model. It will also be noted that at lower roughnesses the accelerations do not
have much of an impact. This can be used to confirm the roughness level below which there
is no impact on parts consumption.
3.0
mcbdv.grf
MCV - Body Acceleration
2.8 Legend Title
50km/h
2.6 70km/h
Fit: 50km/h
2.4
RMS Acceleration (m/s )
Fit: 70km/h
2
2.2
2.0
1.8
1.6
1.4
1.2
1.0
One finding of this research which is not considered in HDM is the speed-roughness
dependence. The results clearly show that the higher the speed, the greater the impact of
roughness on body accelerations.
1
The roughness units in this figure are NAASRA counts/km. A conversion to IRI is 1 IRI = 26.4 NAASRA
counts/km.
Crawl speeds
On a long, steep upgrade vehicles will travel at a constant crawl speed where all the forces
are in balance. Since there is no acceleration, the observed speed can be substituted into the
following equation to calculate the used power:
where:
Speed differences
When the gradients are insufficient for vehicles to reach their crawl speeds the speed
reduction between two points on the grade can be used to establish the driving power. Called
the speed difference, or spatial method, this uses the following equation to establish the used
power:
__
Pu v 1 − v o 0.5 RHO CD CDmult AF + R3 SL 3 R4 SL
2 2
M g HC
= + + + …(6.36)
M' 2T M' T M' T M' T
where:
It should be noted that this method gives the average power used over the section, as opposed
to the crawl speed method which gives the maximum used power.
Bennett (1994) compared these two methods and concluded that both will give similar results,
with the spatial method giving the best results when measurements are made as far along the
grade as possible. When using the spatial method there is almost a linear relationship between
the predicted power used and the distance measured. Thus, any errors in measured distances
will lead to a corresponding over or under estimation of the used power.
Used power will likely be proportional to the gradient, with vehicles using more power on
steeper gradients (Bennett, 1994). However, HDM only allows for a single value to be input
so either the average or a value based on the most typical gradient should be adopted.
6.4.10 Utilisation
For a Level 2 calibration a field survey is required. This entails stopping a sample of vehicles
and recording their odometer readings or, for commercial vehicles, recording data from the
driver’s logbook . In some countries the data are available from fleet management companies.
The data should be grouped by age with the average annual kilometreage of all vehicles of a
given age being calculated. If a survival curve is available, the average lifetime kilometreage
can be calculated by multiplying the average annual kilometreage for each year by the
survival ratio. The total of these products represents the average lifetime kilometreage.
Appendix I illustrates this technique using data from Bennett (1985). The average annual
kilometreage is then calculated as the average lifetime kilometreage divided by the average
service life.
If a survival curve is not available, a reasonable estimate of the average annual kilometreage
will be obtained by calculating the average kilometreage of all the survey data. If the sample
is unbiased, this should yield a similar value to that from the survival curve technique.
Bennett (1985) used this approach and it yielded an average of 10,800 km/year that compared
with 9,850 km/year using the survival curve approach with the same data. The differences are
probably largely due to the biasing of the Bennett (1985) data towards newer vehicles that had
higher average utilisations.
OILCAP
OILCONT = …(6.37)
DISTCHG
where:
Oil loss due to operation is more difficult to measure and, as it likely has limited impacts on
the total costs, should not usually be calibrated. If calibration is desired it should be done as
follows for a sample of vehicles:
1 Change the oil and replace with a known quantity of oil.
2 Operate the vehicles for at least 5-10,000 km, keeping exact records of the amount of
fuel used and, if necessary, additional oil added.
3 At the end of the period drain the oil completely and measure the amount remaining. The
difference between this volume and the initial volume is the oil lost due to operation.
Dividing this by the total fuel consumption gives the operation loss rate.
Overview
Compared with a Level 2 calibration, a Level 3 calibration of acceleration noise can address:
1.04
σat = σatmax (a0 + a1 VCR)
…(6.38)
1+ e
The effort should focus on calibrating the parameters a0 and a1 and then developing a new
model form around the calibrated parameters. This will entail taking measurements of σat at
different levels of VCR and then fitting a curve to the data. Because of the difficulties in
estimating VCR, it is recommended that surveys be conducted at VCR levels of 1, 0.5 and
0.3. This will provide an indication of the trend in the data. Additional measurements can be
then be made to supplement the data as necessary.
6.5.2 Speed β
The quantification of the speed β requires a major field study. A series of speed data are
required covering the full range of conditions under which the constraining speed models will
be applied.
This data set should then be analysed using the same statistical methods as those described in
Watanatada et al. (1987c). The analysis should consider the same issues, such as logarithmic
transformations, and requantify the model from first principles.
7 RDWE calibration
Table 7.1
Ranking of impacts of road deterioration factors
Note:
1 Criteria 1 = Amount of cracking
2 = Rut depth
3 = Roughness
4 = EIRR for patching
5 = EIRR for reseal
6 = EIRR for overlay
Impact sensitivity A = S-I
B = S-II
C = S-III
D = S-IV
The roughness-environment factor is clearly the most important, due to the wider range of its
values, followed by the cracking initiation and progression factors. The general roughness
progression factor has low priority, despite its moderate sensitivity, because its range is small
based on many inter-country validation studies. As shown in the table, for convenience, these
adjustments can be grouped into just two classes:
! High impact
! Low impact
The following sections describe calibrating these factors for the three levels of calibration.
In HDM-4, the flexible pavement deterioration and works effects models have been
generalised to allow more specific adaptations to local materials, and adding a few improved
relationships. The changes include an increase in the number of calibration factors available
for adjusting the bituminous RDWE model from 6 to 20.
Consequently, while this chapter discusses mainly the determination of calibration
adjustments for the HDM-III RDWE model, there are only relatively brief and incomplete
references to specific HDM-4 calibration factors. Cross-references are given to HDM-4
model descriptions published in Volume 4 of this document series, Analytical Framework and
Model Descriptions.
In most cases, the guidelines provided may be directly transposed to apply to the HDM-4
model, although such approach requires a good knowledge of the specific relationships. The
HDM-4 aspects of these guidelines will be published in a future update to this Guide.
∆R te = K ge 0.023 R t …(7.1)
where:
In HDM-4, the calibration factor Kgm is equivalent to the HDM-III factor Kge. The following
discussion uses Kge.
Method 1
For a Level 1 Calibration, the values are established based on the general environmental
conditions and the road construction, drainage standard. This is done as follows:
Step 1 Identify the environment applicable to the immediate vicinity of the road
project in terms of the classifications provided in Table 7.2.
Step 2 Select the appropriate value of m from Table 7.3 according to the
environmental classification.
Step 3 Determine the effective m-value, meff, by multiplying m by a factor km
according to the standard of road construction and drainage Table 7.4, as
follows:
m eff = m k m …(7.2)
Table 7.2
Classification of road environment
Table 7.3
Recommended values of environmental coefficient, m
Table 7.4
Modifying factor of environmental coefficient for road construction and
drainage effects
Swelling soil subgrade without remedial treatment 1.3 - 2.0 1.2 - 1.6
Method 2
The values for Kge estimated using this approach could be enhanced with limited field data, as
follows:
A small number (two to five) of strong old pavements should be selected in one environment
for which accurate roughness, traffic loading and age data are already available, and for
which:
! Pavement structural adequacy: PSA > 1; and,
! Pavement surfacing age > 10 years.
The observed estimates of m are then calculated for the selected pavements (following the
method in Section 7.3.1), averaged and compared with the values selected from the tables. If
the values are similar, no changes should be made to the originally chosen values. If the
differences are consistently different, such as mostly higher or lower, a partial correction may
be made (say about one half of the difference in the means). But care should be taken not to
attach too much significance to these few results until a more thorough application of a Level
2 calibration can be made.
where:
The predictive relationships for each surfacing type take account of the interactive fatigue
effects of pavement strength and traffic loading, and the durability effects of ageing. What the
relationships could not do without requiring significantly more complex data entry was to
define how satisfactory were the material design, manufacture and construction quality, or
define the oxidising power of the environment, except in average terms. The calibration
adjustment therefore compensates for these in specific situations. Cracking behaviour of
bituminous materials tends to be variable, even under fairly identical situations, so it can be
dangerous to make general adjustments based on only one or two cases.
The Level 1 calibration attempts to estimate the durability properties by considering the
actual or approximate behaviour of lightly-trafficked or strong pavements. It assumes that the
predicted structural effects are correct. Thus, it is necessary to estimate the quality of each
bituminous surfacing type and likely adjustment required.
Careful binder temperature control, adequate binder content, low air voids
in asphalt mixtures (or good compaction), high standard of mixing plant
and compaction equipment, use of medium or soft bitumens (for example,
80/100 penetration or higher).
! Fair (FC)
Step 4 Select adjustment factor based on binder quality, oxidising climate and
construction quality from Table 7.5.
Table 7.5
Level 1 adjustment factor for cracking initiation
1
K cp = …(7.5)
K ci
Progression at the early stages is a strong positive function of relative compaction, which is
the average ratio of layer density to reference compaction standards. After the first 1-3 years,
the rate of progression slows down and only accelerates in the presence of cracking and rain.
The adjustment factor is a direct multiplier of the predicted increase so a higher factor
accelerates rut depth progression.
If the predicted rut depth is low when the pavement is young (< 5 years) but the observed rut
depth is much higher, then the adjustments should first be made through decreasing the value
for the relative compaction.
The HDM-III prediction model is a good representation of the effects of structural
deformation and densification through the pavement and subgrade. For badly cracked
pavements in wet climates, the model may underestimate the rate of progression. The model
does not represent the:
! Plastic deformation - which can occur in thick bituminous surfacings under high
temperatures and heavy loading, nor
! Abrasion - that occurs under studded tyres used in some freezing climates
In HDM-4, the rut depth progression model has been decomposed to separately represent the
four components of rutting:
! Initial densification
! Structural deformation
! Plastic deformation
! Surface wear due to studded tyres
Each component may be independently calibrated using separate K adjustment factors (see
Chapter C2, Section 10 of Analytical Framework and Model Descriptions).
The procedures given in this edition of this Guide relate primarily to calibrating the
prediction of rut depths in HDM-III. Further guidance on calibrating the HDM-4 rut depth
component models will be given in a future edition of this Guide. In the meantime, the
procedures given in this Guide for calibrating rut depth progression will assist in planning an
HDM-4 rut depth model calibration study.
In most cases for a Level 1 calibration of HDM-III or HDM-4, adjustment is not considered
necessary. Economic results in the model are barely sensitive to rut depth and the small
benefits derive from reduction of the rut depth variation that is associated with rut depth.
However, if one of the exceptional situations mentioned above is significant and dominant,
the user should consider some adjustment, as follows:
1 High temperature - plastic flow deformation in asphalt
Effects are visible as pock-marked surface wear in a rut without side ridges.
❏ All pavements:
Krp = 1.2 – 2
3 Cracked pavement - clayey or silty base materials, and high to moderate rainfall:
❏ All pavements:
Krp = 1.5(1.1 – 2)
As the available adjustment is approximate for traffic volume-related distress (cases (1) and
(2)), a wide degree of latitude is provided and iterative adjustments may be needed until the
results meet a test of reasonableness (if the user has some data for verification). If calibration
of this mode of distress is important to the user, a Level 2 or 3 calibration is advisable.
influence on roughness inside the model, so the user is advised to be careful not to exaggerate
the effects.
The adjustment factor directly multiplies the incremental area of potholing of the analysis
year. Adjustments may vary widely, for example, Kph = 0.3 - 3.0.
Field sampling
About five pavement segments are selected in each of 2-4 climatic zones as follows:
1 Select 2-4 climatic zones representative of the study area in which each climate zone is
typical of one (or, at most, two adjacent zones) of the moisture-temperature zones
defined in Table 7.2, for example, semi-arid/subtropical, or humid-perhumid/tropical.
2 Select about 5 (between 3 and 10) pavement segments in each zone as follows:
(a) Segment
Select segments about 5 km long (minimum 3 km, maximum 10 km), which may
comprise non-contiguous sections of a minimum 1 km length on the same road and
reasonably homogeneous in the basic data (type 1) in Field measurements below.
(b) Pavement
Select pavement (rehabilitation) age preferably > 10 year (minimum 6 year.)
Note: AGE3 is the age since the last overlay, reconstruction or construction
whichever is the most recent, but excluding surface treatments.
Field measurements
For HDM-4 roughness model relationships, refer to Chapter C2, Section 11 of Analytical
Framework and Model Descriptions.
Pavement age (AGE3), cumulative traffic loading (NE ESA), roughness (m/km IRI),
pavement structural number (SNP), pavement type, and drainage environment type.
2 If pavement surface distress exceeds average PSD 2.0 - then also collect:
Evaluation
The parameter m is estimated directly from the summary model for roughness progression
(Paterson and Attoh-Okine, 1992), assuming the rest of the summary predictive model is
correct, as follows:
1 Estimate original roughness, RI0 (m/km IRI) for each pavement type within either the
climate zone or the study zone, whichever is considered most practicable. This is best
estimated from pavements of the given type less than 3 year old with mean PSD < 1.0,
but also may be made from other estimates (this estimate does not need to be restricted to
the same pavements as selected for the m-calibration).
2 Estimate m from the summary model rewritten as follows depending on which data are
available, and using a spreadsheet or the worksheet provided in HDM Tools:
m=
{ ln[RI ] − ln[RI
t 0 + 263 NE(1 + SNP )
-5
]} …(7.6)
AGE3
m=
{ln[1.02 RI - (0.143 RDS + 0.0068 ACRX + 0.056 APAT )] − ln[RI
t 0 + 263 NE(1 + SNP )
-5
]}
AGE3
…(7.7)
If the observed values of m differ from the standard values (Table 7.3) by more than half of
the difference between m-values for adjacent environmental zones, the assumptions and work
of the calibration study should be carefully reviewed to identify possible errors before
adopting the observed values.
Field sampling
First, stratify the sample by:
1 Surfacing type and subtype (material type); and
This stratification needs to be made realistically into the minimum number of surfacing-
climate groups that is considered necessary and practical to be applied (typically 1-4). Note
that original surfacings need to be classed as separate groups from second and subsequent
surfacings (for example, overlays and reseals).
Second, in each surfacing-climate group, identify a minimum of 15 pavement sections (of
300 m lane length) with Low surface distress (less than 5 per cent area of cracking), selected
from a wide range of annual traffic loading (ESA/lane/year) representing the range found on
the network. The requirement for low surface distress is ideal (since it relates to the initiation
state). It should not include sections with nil cracking unless:
The surfacing age is Medium (6-15 yrs) or Old (> 15 yrs) and they comprise less than 20 per
cent of the sample; and sections of Medium distress (5-30 per cent cracking) may be included
where necessary, but with adjustment to the observed surfacing age (for example, reduced by
one year for each 10 per cent of cracked area).
For a high precision calibration (S-I), to be applied when cracking is considered the primary
intervention parameter or when higher precision is generally preferred, the size of the sample
should be increased to a minimum of 30 pavement sections in each surfacing-climate group.
Field measurements
On each identified cracking calibration pavement section, the following data should be
determined and recorded:
1 Surfacing age (years)
3 Percentage area of wide cracking (more than 3 mm width or spalled narrow cracks)
Evaluation
For cracking model relationships, refer to Chapter C2, Section 5.1 of Analytical Framework
and Model Descriptions.
For each pavement section data record, the predicted all cracking initiation age should be
calculated by the model relationship using a spreadsheet or the HDM Tools software (see
Appendix K). The calibration adjustment factors and prediction errors should be determined
separately for each surfacing-climate group from the mean predicted (mean PTCI) and
observed initiation (mean OTCI) ages as follows:
mean OTCI
K ci = …(7.8)
mean PTCI
{ [
RMSE = SQRT mean (OTCI j - PTCI j )
2
]
j=1,n
} …(7.9)
1
K cp = …(7.10)
K ci
Alternatively, the following explicit field sampling based approach may be adopted:
Field sampling
First, stratify the sample by surfacing type. In most cases, further stratification by surfacing
material subgroup and climate type (as for cracking initiation) is unlikely to add reliable
differences, but may be done optionally.
Second, in each surfacing group, identify a minimum of 15 pavement sections (of 300 m lane
length) with either medium surface distress (5-30 per cent area of cracking), or high surface
distress (>30 per cent area of cracking).
Field measurements
On each identified cracking calibration pavement section, the following data should be
determined and recorded:
1 Surfacing age (years)
3 Percentage area of wide cracking (more than 3 mm width or spalled narrow cracks)
Evaluation
For cracking model relationships, refer to Chapter C2, Section 5.1 of Analytical Framework
and Model Descriptions.
For each pavement section data record, determine:
1 The predicted cracking initiation age, calculated by the model relationship using a
spreadsheet or the HDM Tools software and adjusted by the calibration adjustment
factors determined under Section 7.3.2
2 The estimated age since cracking initiation, calculated by subtracting the predicted
cracking initiation age from the observed surfacing age
3 Fit a sigmoidal curve to the observed cracking area versus the estimated age since
initiation data, and determine the estimated age at 30 per cent cracking area (ET30) by
interpolation or extrapolation
4 Calculate the predicted age at 30 per cent cracking area (PT30) using the HDM-4
equation with coefficients appropriate for the pavement and surface type
5 Calculate the adjustment factor from the mean values of ET30 and PT30 across all
calibration sections separately for each surface type.
mean PT30
K cp = …(7.11)
mean ET30
Field sampling
Identify a minimum of:
! 20 pavement sections - (200 m lane-length) if the flexible pavements are
predominantly of the thin surfacing type (less than 50 mm bituminous material), or
! 30 sections - (about half thin and half thick) if thick asphalt pavements are common
(more than 50 mm thickness of asphaltic layers).
At least 50 per cent of the sample should have medium to high mean rut depth (greater than 6
mm) and the range should include the highest level of rutting prevailing on the road network.
In high standard road networks this sampling requirement may be difficult to meet, in which
case the calibration of rut depth prediction is unlikely to be important.
Field measurements
On each identified section, the following data should be measured or determined:
1 Mean and standard deviation of rut depth values - measured in each wheelpath at 10
m intervals under a 2 m straight-edge (or computed from automated non-contact survey
data for a 2 m straight-edge simulation); and
2 The IQL-2 explanatory parameters - required by the HDM-4 predictive relationships,
for example, SNP, YE4, COMP, ACX, HS etc.
Evaluation
For rut depth model relationships, refer to Chapter C2, Section 10 of Analytical Framework
and Model Descriptions.
For each calibration section j, compute the predicted mean rut depth (PRDMj) and the
predicted standard deviation of rut depth. Calculate the adjustment factor for mean rut depth
progression, by geometric means or from log values (LORDMj and LPRDMj) as follows:
Alternatively, determine Krp and the prediction error by linear regression of LORDMj v.
LPRDMj.
where:
EORItn is the regression estimate of observed roughness RI at time n, and tn is
the last applicable time and t1 is the first applicable survey time; or
[AVG(ORI3, ORI4 ) - AVG(ORI1, ORI2) ] (OT4 - OT1)
∆ORI t = …(7.14)
[AVG(OT3, OT4 ) - AVG(OT1, OT2 )]
where:
ORI4 is the observed roughness RI at the t=4th (or last) observation, and
OT4 is the corresponding time in years of the t=4 observation.
SUM(ORI1 : ORIn )
MORI = …(7.15)
n
where:
the notation is similar to the above for RI.
The incremental and mean values of the other parameters would be determined
similarly.
Step 4 Calculate the residual errors by differencing the observed and predicted values
of incremental roughness for each calibration section j, for example:
RESRIj = ∆(PRIj t - ∆ORIj t , ) …(7.17)
Determine correlation and slope (b) without intercept between RESRI and
MORI. If the correlation and the determination of ‘b’ are significant, then
determine the adjustment factor, Kgp, as follows:
K gp = 1 + b …(7.18)
Field sampling
The sampling should include about 15 pavements in each stratum as follows:
! Samples should be stratified by:
❏ surface type and material type; and
❏ problem and non-problem areas, defined as described above, where problem areas
have been assigned a non-zero CQ value, or where separate adjustment factors are
to be determined for each area.
! In each group, select a minimum of 15 pavement calibration sections, of 300 m lane
length, either with:
❏ low positive incidence of ravelling (0 < area < 10 %), or
❏ nil incidence of ravelling if the surfacing age is six years or older (provided that this
group does not exceed 20 per cent of the total sample).
Field measurements
On each calibration section, the following data should be measured or recorded, in
accordance with the HDM-4 relationships:
! Surface type
! Material type
! Surfacing age
! Annual vehicle axle passes (YAX)
Evaluation
For ravelling initiation model relationships, refer to Chapter C2, Section 6.1 of Analytical
Framework and Model Descriptions.
For each calibration section, the observed time to ravelling initiation (OTRVj) for the groups
should be set as shown below:
! Low area of ravelling
Should be set equal to 10 per cent less than the surfacing age, that is, 0.9 AGES.
! Nil ravelling (OTRV)
Should be set to 20 per cent more than the surfacing age, that is, 1.20 AGESj.
The predicted time to ravelling (PTRVj) should be determined for each section by applying
the appropriate HDM-4 relationship and coefficients, using a spreadsheet or the worksheet in
HDM Tools. The calibration factor for ravelling initiation is determined for each strata group
(g) as follows:
mean[OTRVj(g)]
K vi (g) = …(7.19)
mean[PTRVPj(g)]
HDM-4 also provides for a separate calibration factor for ravelling progression Kvp. It may be
derived in a similar way to the initiation factor, or else set to equal the inverse of the initiation
factor.
7.2.7, potholing progression is highly variable and generally difficult to predict. A local
calibration is particularly valuable when potholing is highly prevalent and severe in
substantial portions of the road network, because potholing progression is dependent on the
sensitivity of the base material to disintegration and moisture and is a function of construction
and maintenance practices. Otherwise, if potholing is rare or quickly repaired by routine
maintenance, a detailed calibration of potholing progression is not worthwhile and may be
omitted.
In HDM-4, the conceptual approach to modelling of initiation and progression of potholes has
been expanded from that of HDM-III, and allows for the separate internal adjustment of
contributions to potholing from structural cracking, ravelling and enlargement of potholes
(refer to Chapter C2, Section 7 of Analytical Framework and Model Descriptions).
The following procedure relates only to calibrating the prediction of potholes in HDM-III.
Adjustment of any of the HDM-4 coefficients should not be attempted in a Level 2
calibration.
Field sampling
Select about 10 pavement sections of 100 lane-m area with incidence of potholing ranging
from 10 to 500 pothole units (10 litre volume, about 0.1 m2 area), that is, from 1 m2 or 0.3 per
cent, up to about 50 m2 or 15 per cent of carriageway area.
Field measurements
On each calibration section, measure or estimate the IQL-2 parameters used in the HDM-III
predictive relationship, that is, pavement surface thickness, annual average heavy axles,
surfacing age (AGES), area of open potholes and estimation of number of pothole units.
Evaluation
For each calibration section, estimate the time for initiation of cracking (PTCI), the time for
initiation of potholing (PTPI), and the time for progression of potholing to X units (PTPX) up
to 500 potholing units. Compute the observed and predicted potholing times as follows:
Determine the potholing adjustment factor, either by linear regression of OTPXj against
PTPXj, or as follows:
mean(OTPXj)
K ph =
mean(PTPXj)
! Factorial design
The main factorial matrix should comprise the primary explanatory variables and cover a
range of values as close as possible to the range over which they will be used. The
HDM-III and HDM-4 relationships, and the sensitivity classes assigned in Chapter 4, are
the starting point to identify which parameters are primary and which secondary. When
important interactions are to be determined, especially between parameters that may
exhibit collinearity, there must be sufficient range for the cross-effects to be estimated.
But to estimate the effect statistically it is necessary to have data with high, normal and
low traffic loading on a given SNP, and strong, normal and weak pavements under a
given loading YE4. It is not necessary to fill all cells of a factorial for all primary
parameters; a partial factorial can be used to reduce the number of sections to be found.
A summary of factorial designs for road deterioration studies is given in Paterson (1987)
and GEIPOT (1982), and design principles can be found in relevant textbooks.
! Cross-sectional and longitudinal analyses
For discrete events such as the initiation of cracking, ravelling or potholing, studies
which are limited to only pavements showing the initial stages of distress can risk being
biased if the set excludes pavements that out-perform the norm or that have premature
failures. The application of maximum likelihood statistical estimation (MLE) enabled
the HDM-III analyses to determine the concurrent effects of ageing and fatigue, and the
MLE techniques have become readily accessible now in currently available advanced
statistical software.
! Analysis of residuals
Analysis of the residual errors of a statistical model should always be used to determine
whether they correlate with any of the possible explanatory variables and to reveal any
bias trends that may remain in the model. This can lead to the testing of further model
formulations and improving their explanatory power.
! Model form
The formulation of predictive models was derived in the first instance from:
❏ The received knowledge of mechanistic and empirical research,
❏ Selecting and clustering explanatory parameters on the basis of a mechanistic model
where this was available, and
❏ Adding other parameters in the forms in which they were expected to interact, for
example, in multiplicative, additive, or power functions.
Advanced statistical methods were used to test the validity of the relationships, derive
the observed coefficients and determine the real significance of individual parameters or
parameter clusters. Thus they represent structured empirical models, combining
mechanistic and logical theory with empirical estimation. The resulting model forms are
therefore sound starting points for further research. It is strongly advised that Level 3
adaptation always begin with a statistical re-estimation of the coefficients of the HDM
relationship directly from the local data.
1 The sample sizes should be doubled or tripled to a minimum of 30 per surfacing type-
material type combination; and
2 The sections selected for modelling of distress initiation should be monitored over a 4-6
year period to derive distress progression data on the same sections, so that the direct
interactions between initiation and progression behaviour can be identified and
characterised.
Two sets of models should be evaluated:
1 The structured model defined for HDM (that is, the HDM-4 road deterioration model for
bituminous pavements described in Chapter C2 of Analytical Framework and Model
Descriptions), specifying the same model parameters and estimating the coefficients
from the local data; and
2 Variations on the structured model, including for example the mechanistic (strain-based)
relationships tested in (Paterson, 1987 - chapters 5 and 6) and others.
Some tips from previous research:
1 Structural and traffic loading variables must always be clustered interactively and not
additively.
5
IRI After Overlay (m/km)
4 HDM-III
Thailand
Indonesia
Sweden
3
0
0 2 4 6 8 10 12
IRI Before Overlay (m/km)
7.5.1 Level 1
A Level 1 calibration applies a constant value for the roughness after an overlay that is
applied to all pavements receiving a particular treatment. The default values in HDM-4 may
be used. However, since this is an S-II or S-I sensitivity, it is valuable to use some simple
field measurement. A number of pavements which have recently been overlaid and for which
the current roughness is available should be identified. On the basis of their roughness levels
an average roughness after overlay is established. If different overlay thicknesses are used the
sample can be stratified by overlay thickness and different values determined as a function of
thickness.
7.5.2 Level 2
A Level 2 calibration conducts a small study into the effects of overlay on roughness and
develops a local relationship for the effects of overlay on roughness. A series of sections
scheduled for overlays should be selected. There should be at least five sections, with a
minimum length of 500 m each. The roughness should be measured on each section. If a
manual method is used (for example, Dipstick, Walking Profilometer) only a single run is
required in each wheelpath; if using a vehicle a minimum of five runs is recommended.
In addition to measuring the roughnesses of the sections selected for maintenance,
measurements should also be made on untreated sections on the same road near the
maintenance sections. These provide control sections: since no maintenance is scheduled for
them, and the time between surveys is short, it can be expected that there will be little, if any
change in the roughness. They can therefore be used to correct the data for any roughness
measurement instrument drift over time.
The before and after roughness data should be analysed to establish a general relationship
between these roughnesses. If different overlay thicknesses are available the relationship
should take this into account.
If possible, speed measurements should be made at the same time, as this will allow for the
effects of roughness on speed to be established as well.
9 References
Arup (1992). Maintenance Strategy for Songea-Makambako Road Tanzania 1993-1997.
Report to ODA, Arup Aikins, London.
Bennett, C.R. (1985). A Highway Economic Evaluation Model for New Zealand. School of
Engineering Report 368, Department of Civil Engineering, University of Auckland,
Auckland, New Zealand.
Bennett, C.R. (1989). The Depreciation of Motor Vehicles in New Zealand. Road Research
Unit Occasional Paper, Transit New Zealand, Wellington, New Zealand.
Bennett, C.R. (1994). A Speed Prediction Model for New Zealand. School of Engineering
Report 368, Department of Civil Engineering, University of Auckland.
Bennett, C.R. (1995). Modelling Depreciation, Interest, Utilisation, Crew and Overhead
Costs in HDM-4. Report to the International Study of Highway Development and
Management Tools, University of Birmingham.
Bennett, C.R. (1998). Modelling Road User Effects in HDM-4. International Study of
Highway Development and Management Tools, University of Birmingham.
Bennett (1999). Generating RUE Elasticities for the HDM-4 Calibration Reference Manual.
Internal Report, HTC Infrastructure Management Ltd., Auckland.
Bennett, C.R. and Dunn, R.C.M. (1992). Error Considerations in Speed Surveys. Proc. ARRB
Conference.
Bennett, C.R. and Greenwood, I.D. (1999). Modelling Road User and Environmental Effects
in HDM-4. Final Report Asian Development Bank Technical Assistance Project
5819-REG. To be published as Volume 7, the Highway Development and
Management Series, the World Road Association (PIARC) and the World Bank,
Paris.
Biggs, D. C. (1987). Estimating the Fuel Consumption of Light to Heavy Vehicles. Australian
Road Research Board Internal Report AIR 454-1, Nunawading, Australia.
Button, K. and Ngoe, N. (1991). Vehicle Ownership and Use Forecasting in Low Income
Countries. TRL Contractors Report 278. Transport Research Laboratory,
Crowthorne.
CES (1989). Vehicle Fleet Modernisation and Road Use Charges Study. Report to
MOST.Consulting Engineering Services (India) Private Limited, New Delhi, India.
CES (1991). Development of Long Term Plan for Expressways in India. Report to the
Ministry of Surface Transport, New Delhi.
Cenek, P. D. and Patrick, J. E. (1991). Prediction of Road Roughness Prediction. Central
Laboratories Report 91-29301. Mworks Consultancy Services, New Zealand.
CESTRIN (1994). Vehicle Operating Costs: Derivation of Vehicle Fleet Characteristics and
Unit Costs. Report to Centre for Road Engineering Studies, Bucharest.
Chamala, R. (1993). Economic Maintenance Intervention Standards. Report to the
Pavements and Asset Strategy Branch, Queensland Transport, Brisbane.
Chesher, A., and R. Harrison (1987). Vehicle Operating Costs: Evidence from Developing
Countries. World Bank Publications, Washington, D.C.
CID. (1993) Projet Routier Complement du Programme oes Routes A Elargir, Rapport de
Presentation. Direction des Routes et de la Circulation Routiere. Ministere des
Travaux Publics de la Formation Profressionnelle et de la Formation de Cadres.
Royaume du Maroc.
Cox, B.E. (1983). The Value of Travel Time. RRU Occasional Paper. Transit New Zealand,
Wellington.
Curtayne, P.C., Visser, A.T., du Plessis, H.W. and Harrison, R. (1987). Calibrating The
Relationships Between Operating Costs of Buses and Road Roughness on Low-
Volume Roads. Transportation Research Record 1106, Transportation Research
Board, Washington, D.C.
Daniels, C. (1974). Vehicle Operating Costs in Transport Studies. EIU Technical Series No.
1, The Economist Intelligence Unit, London.
du Plessis, H.W. and Meadows, J.F. (1990). A Pilot Study to Determine the Operating Costs
of Passenger Cars as Affected by Road Roughness. In Report PR 88/070/3, Roads
and Transport Technology, CSIR, Pretoria, South Africa.
du Plessis, H.W., Visser, A.T. and Harrison, R. (1991). The Effects of Road Condition on the
Operating Costs of Buses. In Report PR 88/070/3, Roads and Transport Technology,
CSIR, Pretoria, South Africa.
du Plessis, H.W. and Schutte, I.G. (1991). Road Roughness Effects on Vehicle Operating
Costs: Southern Africa Relations for use in Economic Analyses and in Road
Management Systems. Report PR 88/070/3, Roads and Transport Technology, CSIR,
Pretoria, South Africa.
Estudio (1993). Estudio de Implantacion de los Modelos HDM-III y EBM para la
Preparacion de Programas de Mantenimiento y Mejora de la red de Carreteras de la
Generalitat de Catalunya. Report to Department de Politica Territorial I Obres
Publiques Direccio General de Carreteres, Generalitat de Catalunya. Estudio HC,
SA., Barcelona.
Findlayson, A.M. and du Plessis, H.W. (1991). Operating Costs of Medium to Heavy Trucks
as Affected by Road Roughness. In Report PR 88/070/3, Roads and Transport
Technology, CSIR, Pretoria, South Africa.
Fonseca, V. M. R., Queroz, C. and dos Santos A. B. (1991). Economic Analysis of Guinea-
Bissau Roads Program Using HDM-III and EBM Models. 5th International
Conference on Low Volume Roads, North Carolina State University, Raleigh, North
Carolina, USA.
Gaete, R., Echeverria, G., Videla, C. and Gutierrez, P. (1991). Sistema de Autocalibracion de
los Modelos de Deterioro del HDM-III. Proc. Sexto Congreso Ibero - Latino-
Americano del Asfalto, Santiago, Chile.
GEIPOT (1982). Research on the Interrelationships Between Costs of Highway Construction,
Maintenance and Utilization. Final Report on the Brazil-UNDP Highway Research
Project. (12 Volumes). Brazilia, Brazil.
GITEC (1992). Estudio de Factibilidad Para El Programa De Mantenimiento Vial. Report
to the Ministry of Communications, Republic of Guatemala.
Greenwood, I.D. (1997). Calculating Tyre Diameter. Memo to the ISOHDM Study,
University of Birmingham.
Kampsax - Beca (1990) Provincial Roads and Bridges Investment Programme, Phase II,
Extract Ch 4: Vehicle Operating costs. Final Report. The Independent State of
Papua New Guinea, Department of Transport, Asian Development Bank, New
Guinea.
Kannemeyer, L. and Visser, A.T. (1994). Calibration of HDM-III Performance Models for
Use in the Pavement of South African National Roads. Paper offered for Presentation
at the January 1995 Annual Meeting of the Transportation Research Board,
Washington, D.C.
Kinhill (1998). Tamil Nadu Project Co-ordinating Consultancy Project. Correspondence with
authors. Madras.
Lea Associates (1988). Saskatchewan Highways and Transportation Pavement Management
Information System: PMIS Technical Manual Appendix I - Vehicle Operating Costs.
Lea Associates Ltd., Vancouver, B.C., Canada.
Liautaud, G. and Archondo-Callao, R. (1994). Brazil Federal Highway Network- A Study to
Define Appropriate Policies, Standards, Budget Allocations and Rehabilitation
Programs and Priorities for the Next Five Years 1995-2000.
Louis Berger International (1990). The Federal Trunk Roads Study - Final Report. Report to
the Federal Ministry of Works and Housing, Nigeria.
McLean, J.R. (1989). Two-Lane Highway Traffic Operations: Theory and Practice. Gordon
and Breach Science Publishers, Melbourne.
McLean, J.R. (1991). Adapting the HDM-III Vehicle Speed Prediction Models for Australian
Rural Highways. Working Document WD TE 91/014, Australian Road Research
Board, Nunawading.
Minderhoud, M.M., Botma, H., and Bovy, P.H.L. (1997). Assessment of Roadway Capacity
Estimation Methods. Transportation Research Record 1572, pp. 59-67.
Transportation Research Board, Washington, D.C.
Mrawira, D., Haas, R. and Paterson, W.D.O. (1998). Streamlining the World Bank’s HDM-III
Model for Network Level Application. Proc. 4th International Conference on
Managing Pavements, Durban.
MRCU (1992). Vehicle Operating Costs in Nepal - Working Paper. Maintenance and
Rehabilitation Co-ordination Unit, Ministry of Works and Transport Department of
Roads, Katmandu.
MVA Consultancy, Institute of Transport Studies (Leeds), Transport Studies Unit (Oxford)
(1987). The Value of Travel Time Savings. Policy Journals, Newbury, Berkshire, UK
NDLI (1991). Calibration of Vehicle Operating Costs to Thailand. Report to the Department
of Highways, N.D. Lea International Ltd., Vancouver, B.C., Canada.
NDLI (1992). Thailand Road Maintenance Project: A.D.B. T.A. 1106-THA Final Report.
N.D. Lea International Ltd., Vancouver, B.C., Canada.
NDLI (1993). Nepal Road Maintenance Programme: Final Feasibility Report. N.D. Lea
International Ltd., Vancouver, B.C. Canada.
NDLI (1994a). Road Network Rationalisation Study - Draft Final Report. Report to Public
Works Division, Barbados, N.D. Lea International, Vancouver.
NDLI (1994b). Comprehensive Highway Maintenance Study - Phase I Final Report. Report
to Ministry of Works and Transport, Trinidad and Tobago, N.D. Lea International,
Vancouver.
NDLI (1994c). Four States Pavement Management Project - Draft Interim Report. Report to
Ministry of Surface Transport (MOST), New Delhi, India. N.D. Lea International,
Vancouver, Canada.
NDLI (1995a). Modelling Road User Effects in HDM-4. Report to the Asian Development
Bank, N.D. Lea International Ltd., Vancouver, B.C., Canada.
NDLI (1995b). Modelling Road Deterioration and Maintenance Effects in HDM-4. Report to
the Asian Development Bank, N.D. Lea International Ltd., Vancouver, B.C., Canada.
NDLI (1997). Gujarat Project Co-ordinating Consultancy: Calibration of Road User Effects.
N.D. Lea International, Vancouver, B.C., Canada.
Opus (1999). Review of VOC Pavement Roughness Relationships Contained in Transfund’s
Project Evaluation Manual. Central Laboratories Report 529277.00. Opus
International Consultants Ltd., Wellington.
Paterson, W.D.O. (1986). Data Collection for HDM-III. Internal Report. Transportation
Department, The World Bank.
Paterson, W.D.O. (1987). Road Deterioration and Maintenance Effects: Models for Planning
and Management. World Bank Publications, Washington, D.C.
Paterson, W.D.O. and Scullion, T. (1990). Information Systems for Road Management: Draft
Guidelines on System Design and Data Issues. World Bank Technical Paper INU 77,
Infrastructure and Urban Development Department, The World Bank, Washington,
D.C.
Riley, M.J. (1998). Correspondence with authors. Riley Partnership, Launceston, UK
Riley, M.J., Petersen, H.H. and Miller, L.H. (1987). Optimizing Pavement Rehabilitation and
Maintenance Policies for Rural Roads. Asia-Pacific Conference on Roads, Highways
and Bridges, Jakarta, Indonesia.
RITES (1992). Feasibility Study for Rehabilitation and/or Upgrading of Quelimane-
Namacurra Road. Report to DNEP, Maputo, Mozambique. RITES, New Delhi, India.
RITES (1994). Gujarat Pavement Management System. New Delhi.
RPT (1990). Transport Policy and Planning Project UGA/90/018 Vehicle Operating Costs
Report. Report to Ministry of Works, Transport & Communications, Republic of
Uganda, Rendel Palmer & Tritton, London.
RPT (1993). Internal Report National Highway PMS Project. Rendel Palmer & Tritton,
London.
Sammour, A. (1992). Feasibility Study Report: Swaileh-Queen Alia International Airport
Road Using Highway Design and Maintenance Standards Model, HDM-III Model.
Report to Ministry of Planning, The Hashemite Kingdom of Jordan.
Sayers, M.W., Gillespie, T.D., and Paterson. W.D.O. (1986). Guidelines for Conducting and
Calibrating Road Roughness Measurements. World Bank Technical Paper Number
46, World Bank Publications, Washington, D.C.
Watanatada, T., Harral, C.G., Paterson, W.D.O., Dhareshwar, A.M., Bhandari, A., and K.
Tsunokawa (1987b). The Highway Design and Maintenance Standards Model:
Volume 2 - User's Manual for the HDM-III Model. World Bank Publications,
Washington, D.C.
Watanatada, T., Dhareshwar, A.M., and P.R.S. Rezende-Lima (1987c). Vehicle Speeds and
Operating Costs: Models for Road Planning and Management. World Bank
Publications, Washington, D.C.
Winfrey, R. (1969). Economic Analysis for Highways. International Textbook Co., Scranton,
Pennsylvania.
World Bank (1995). Hashemite Kingdom of Jordan Third Transport Project Staff Appraisal
Report. The World Bank, Washington, D.C.
Wyley, W. J., Petts, R. C. and Brooks, J. (1986). Evaluation of the World Bank’s Highway
Design and Maintenance Standards Model on United Kingdom Highways. Contractor
Report 27, Transport and Road Research Laboratory, Crowthorne, Berkshire, U.K.
Yuli, P. (1996). Development of Speed and Fuel Consumption Models for Chinese Vehicles.
Ph.D. Dissertation, Department of Civil Engineering, University of Birmingham.
Zaniewski, J.P., Butler, B.C., Cunningham, G., Elkins, G.E., Paggi, M.S., and Machemal, R.
(1982). Vehicle Operating Costs, Fuel Consumption, and Pavement Type and
Conditions Factors. FHWA Report FHWA-PL-82-001. Federal Highways
Administration, Washington, D.C.
Zukang, Y., Zhongren, W. and Xinghua, L. (1992). Calibration of Vehicle Operating Cost
Model for Pavement Management System. Selected Papers of Tongji University,
1992. Tongji University Press, China.
Table A.1
HDM data and model parameters and calibration levels
1
Sub-model Description Units Discussed Calibration level where local values
4
in section or should be established in place of defaults
appendix Level 1 Level 2 Level 3
Aerodynamic drag coefficient 6.3.10 •
…Continued
1
Sub-model Description Units Discussed Calibration level where local values
4
in section or should be established in place of defaults
appendix Level 1 Level 2 Level 3
Optimal life depreciation parameters 6.3.9 •
…Continued
1
Sub-model Description Units Discussed Calibration level where local values
4
in section or should be established in place of defaults
appendix Level 1 Level 2 Level 3
Acceleration noise -traffic noise – a0 6.4.5 •
…Continued
1
Sub-model Description Units Discussed Calibration level where local values
4
in section or should be established in place of defaults
appendix Level 1 Level 2 Level 3
Speed roughness - maximum rectified mm/s 6.4.1 •
velocity
Drivetrain efficiency •
…Continued
1
Sub-model Description Units Discussed Calibration level where local values
4
in section or should be established in place of defaults
appendix Level 1 Level 2 Level 3
Power due to engine drag % •
Rolling resistance - a0 •
Rolling resistance - a1 •
Rolling resistance - a2 •
Speed curve - a0 •
Speed curve - a1 •
Tyre stiffness - a0 •
Tyre stiffness - a1 •
Tyre stiffness - a2 •
…Continued
1
Sub-model Description Units Discussed Calibration level where local values
4
in section or should be established in place of defaults
appendix Level 1 Level 2 Level 3
Altitude m •
…Continued
1
Sub-model Description Units Discussed Calibration level where local values
4
in section or should be established in place of defaults
appendix Level 1 Level 2 Level 3
Roughness progression 7.2.5 •
Superelevation % Appendix D •
…Continued
1
Sub-model Description Units Discussed Calibration level where local values
4
in section or should be established in place of defaults
appendix Level 1 Level 2 Level 3
Ravelling initiation 7.2.6 • •
…Continued
1
Sub-model Description Units Discussed Calibration level where local values
4
in section or should be established in place of defaults
appendix Level 1 Level 2 Level 3
Cost of travel time cost/h Appendix F •
Table B.1
Vehicle tare weights
Note: Operating weight in tonnes. It is defined as the sum of the tare weight plus load.
Table B.3
Projected frontal area
Table B.4
Aerodynamic drag coefficient
Table B.5
Elasticity of vehicle utilisation
NDLI (1994c) Barbados 0.60 0.80 0.85 0.80 0.85 0.85 0.85 0.85 0.75
SWK (1993a) Botswana 0.10 0.15 0.20 0.35 0.40 0.50 0.35 0.40
IBRD (1990) Burundi 0.00 0.80 0.00 0.85 0.85 0.85 0.80
SWK (1993b) Ethiopia 0.10 0.20 0.20 0.20 0.30 0.40 0.20 0.30
JBP (1990) India 0.60 0.60 0.90 0.80 0.90 0.90 0.90 0.90
INDEC (1988) Indonesia 0.50 0.75 0.45 0.45 0.45 0.65 0.65 Sumatra
TSPC (1992) Sri Lanka 0.50 0.50 0.50 0.50 0.50 0.50 0.60 0.60
NDLI (1991) Thailand 0.10 0.17 0.21 0.36 0.15 0.17 0.23
Table B.6
Number of hours driven per year
Table B.7
Vehicle service life
Table B.8
Annual utilisation
Table B.9
Value for fuel consumption model adjustment factor - α1
Table B.10
HDM-III fuel consumption model adjustment factor - α2
Table B.11
Used driving power
Table B.12
Values for braking power
Table B.13
Parts consumption model parameter - C0SP
Table B.14
HDM-III parts consumption model parameter - CSPQI
Table B.15
Labour hours model parameter - C0LH
Table B.16
Labour hours model parameter - CLHPC
Table B.17
Tyre consumption model parameter - C0TC
Table B.18
Tyre consumption model parameter - CTCTE (x 10-3)
Table B.19
Volume of wearable rubber
Botswana Existing ST + Base, Crushed N 0.67 1.00 0.67 1.00 0.82-1.72 0.51 1.00 1.3-1.6 10-80 117-2438 NA 1786-3266 NA NA SWK (1993a)
Pavement Rock/ Stab. Grav/ mm/km BI
Natural Gravel
Brazil Existing Road Network Data Y 1.00 1.00 1.00 1.00 1.00 1.00 1.00 NA NA NA NA NA NA NA Liautaud and
Pavement Archondo-Callao
(1994)
Chile New Pavement ST30 + Gran Base240 N 0.50 NA 1.00 1.95 0.70 NA NA 13 Low Med NA NA NA NA Gaete et al. (1991)
Base 290
Overlay On Dense Surf 35-45 + N 0.67 0.67 NA NA 0.80-1.06 0.43 3.00 NA 3-10 Medium Dry NA NA NA
Asphalt Open Asph Base40-
Pavement 60+ Old Surf 60+
Gran155-170
Overlay On Dense Asph Surf 55- N 0.58- NA NA NA 0.74-0.99 0.70 2.20 NA 4-13 Medium Med NA NA NA
Asphalt 70+ Binder 70-100 + 0.67
Pavement Dense Asph
Base120-130 + Old
Surf130 + Gran Base
130
…Continued
New Pavement Dense Asphalt Surf40 N 0.83 NA NA 0.51 NA 1.74 NA NA NA Medium Heavy NA NA NA
+ Dense Asph Base95
+ Gran. Base 320
Guinea Bissau Existing Road Network Data Y 1.00 1.00 1.00 1.00 1.00 1.00 1.00 All All NA NA NA NA NA Fonseca et al.
Pavement (1991)
India Reseal & Asph. Surfacing N 1.00 1.25 0.70 1.25 1.25 1.00 1.00 NA NA NA Low, Medium, NA Patch + Routine NA CES (1989)
Overlaid +Granular Base High
1.00 1.50 0.60 1.50 1.50 1.50 1.15
Reseal & Asph. Surfacing N 1.00 1.50 0.60 1.50 1.50 1.15 1.15 NA NA NA NA NA NA NA JBP (1990)
Overlaid +Granular Base
Reseal & Asph. Surfacing N 1.50 1.50 1.00 1.50 1.50 1.00 1.50 2 5 1-8 msa/year Medium 2500 mm/km Patch + Routine 15 RITES (1994)
Overlaid +Granular Base BI
Reseal & Asph. Surfacing Y 1.00 1.00 1.00 1.00 1.00 0.70 1.00 2.0 , NA 2 msa/year Medium 2500 mm/km Patch + Routine 20 NDLI (1994a)
Overlaid +Granular Base 3.0 BI
Jordan Existing AC Over Granular N 2.50 1.00 1.50 3.00 1.00 1.00 0.50 NA 13 NA 0.025 m/month NA NA 12 Sammour (1992)
Base
New AC Over Granular N 0.50 1.00 2.00 1.00 1.00 2.00 1.00 NA 15 2.5msa/year NA NA NA
Base
Malaysia Reseal & Existing & New N 0.70- 1.20-1.40 1.00 1.00 1.00 1.00-1.30 1.00 All All NA NA NA Routine+ NA JKR (1991)
Overlaid Pavements 0.80
Patch+ Reseal &
Overlay
Morocco Existing Road Network Data Y 1.00 1.00 1.00 1.00 1.00 1.00 1.00 NA NA NA NA NA NA NA CID (1993)
Pavement
…Continued
Nepal Existing ST and AC over Y 1.00 1.00 1.00 1.00 1.00 1.00 1.00 SNC= NA Low & Medium/ Heavy 10m/km Routine + Patch NA NDLI (1993)
Pavement Granular Base 3 Medium
IRI
New Zealand Existing Overlays & Reseals NA NA NA NA NA 0.76 1.00 2-4 NA 500-6700, NA NA NA 7 years Cenek and Patrick
Pavement (1991)
0.01-0.09
msa/lane/ year
Nigeria Existing Road Network Data Y 1.00 1.00 1.00 1.00 1.00 1.00 1.00 NA NA NA NA NA NA NA Louis Berger
Pavement (1990)
Pakistan Existing Road Network Data Y 1.00 1.00 1.00 1.00 1.00 1.00 1.00 NA NA NA NA NA NA NA Riley et al. (1987)
Pavement
Peru Existing Road Network Data Y 1.00 1.00 1.00 1.00 1.00 1.00 1.00 NA NA NA NA NA NA NA Liautaud and
Pavement Archondo-Callao
(1994)
South Africa New Pavement Surface Treatment N 1.00- 0.10-0.30 NA NA 1.50-1.75 0.392(1) 0.8-1.2(4) NA NA NA NA NA Reseal/ Overlay 20 Kannemeyer and
1.50 (avg. @8y Visser (1994)
021) (avg. 1.57) 0.607(2) 0.6-1.4(5) (1975-1995)
(avg.
1.21) 0.886(3) 0.8-1.2(6)
Existing Overlays & Reseals N 0.40- 0.30- NA NA 1.00 0.392(1) 0.8-1.2(4) NA NA NA NA NA Reseal/ Overlay 20
Pavement 0.80 @8y
0.70 0.607(2) 0.6-1.4(5) (1975-1995)
(avg.
0.63) (avg. 0.886(3) 0.8-1.2(6)
0.21)
Spain Existing AC and ST over Y 1.00 1.00 1.00 1.00 1.00 1.00 1.00 3.1-4.7 NA NA 0.045 m/ 2253-3352 NA 15 years Estudio (1993)
Pavement Granular Base mm/km BI
month
…Continued
Tunisia Existing Road Network Data Y 1.00 1.00 1.00 1.00 1.00 1.00 1.00 NA NA NA NA NA NA NA Houcine (1982)
Pavement
UK Existing AC60+Old Surf 50+ 1.00 1.00 NA NA NA NA NA 80 5535-7057 0.03 m/month 1500 mm/km Patch + Routine 10 Wyley et al. (1986)
Pavement Cement Stab. Base BI
120 (1974-1984)
Reseal on AC SD10+Old Surf 90 + NA NA NA NA 3.00 NA NA 2.3 80 5535-7057 0.03 m/month 1500 mm/km Cold Mix 20 @ 10
Gran Base150 BI 5Y + Patch +
Routine (1974-1984)
Surface ST30 +Old Surf. NA NA NA NA 2.70 NA NA 1.5 8 2892-3659 0.03 m/month 1500 mm/km AC30@ 15/ 7Y 10
Treatment 20+Granular Base 205 BI +Patch +Routine
(1974-1984)
D.1 Introduction
The road and pavement data required for HDM consists of general road data, pavement
characteristics, pavement condition and the maintenance alternatives. Each of these data are
discussed in the following sections after a description of how one classifies pavements.
For HDM-4 the pavement classification system is much more comprehensive to reflect the
wider range of pavement types modelled.
Table D.1 lists the HDM-4 bituminous pavements classification system and Table D.2 defines the
pavement codes (ISOHDM, 1997). In addition to these there are concrete, block and unsealed
pavements in HDM-4.
Table D.1
HDM-4 bituminous pavement classification system
PMA AB AB AMAB
AM RAC CS
SB AMSB
CM LS
PA TNA
xx
CAPE CS
GB STGB
DBSD GM
SBSD AB AB STAB
ST SL CS
SB STSB
PM LS
TNA
AP STAP
FDA
Source: ISOHDM (1997)
Note:
1 AM and ST surfacings on concrete pavements, that is, AMCP & STCP, are modelled in
HDM-4 as concrete pavement types in the rigid pavements sub-model.
Table D.2
Definitions of HDM-4 pavement codes
SL Slurry Seal
PM Penetration Macadam
xx User Defined
The available HDM-4 models are based on different factors. Many are based on surface and
base type, while some are based on surface material. Accordingly, the modelling is initially
done in terms of surface material and base type, even though base materials can be specified.
HDM-4 has default coefficients for the bituminous pavement types given in Table D.3.
Table D.3
Generic HDM-4 bituminous pavement types
Without adequate maintenance records the only way of classifying a pavement into one of the
post-maintenance categories is by conducting field surveys and gathering data on a sample of
pavements.
! Superelevation (m/m)
These data are usually straightforward to obtain. The first two data are available from
topographic maps, highway information sheets, or route surveys. For the carriageway and
shoulder widths a sample of data should be used to estimate average or median values. These
are often a function of traffic volume so it may be useful to stratify the data as a function of
volume. The superelevation can either be supplied, or it can be calculated in HDM-III as a
function of curvature using the following equation developed from the Brazil data
(Watanatada et al., 1987a):
e = a0 C …(D.1)
where:
Another value termed the effective number of lanes is used in HDM-III. This is used to
reflect the practice of traffic on narrow roads to travel towards the middle of the pavement,
thereby increasing pavement deterioration. Watanatada et al. (1987a) give the following
recommended values:
It should be noted that in places where there is poor lane discipline where vehicles do not
tend to drive in set wheelpaths, such as India, it may be appropriate to increase the number of
effective lanes above these recommended values due to the reduced levels of effective traffic
loading.
Watanatada et al. (1987a) provide similar equations for calculating the deflection from the
SNC. Paterson (1987) who used an orthogonal regression to ensure that the equations give
consistent predictions when going to/from either attribute developed these equations.
Appendix J describes orthogonal regression.
NDLI (1995b) describe several techniques for estimating SNC from FWD deflection and
these have been adopted for HDM-4.
If the existing pavements are strong, SNC of 5.0 or greater, there is not a great need to
accurately specify this parameter. This is because there are limited differences in the
predictions of the HDM-III pavement deterioration model with strong pavements. Thus, it is
only when there are weak pavements that the SNC is a critical parameter.
The pavement layer information is more difficult. When there are set standards it is often
possible to assume a standard surface thickness based on an estimated average age and the
number of years between typical maintenance treatments.
D.5.1 Roughness
The pavement roughness is a very important parameter. It is not only a measure of the
condition of the pavement but it also has a major impact on the RUE.
When measuring sections of road, roughness is usually measured using a response-type road
roughness measuring system (RTRRMS) or a non-contact profilometer (laser/accelerometer).
For short sections or studies manual methods such as a rod-and-level survey, dipstick or
walking profilometer may be used.
RTRRMS consist of an instrument mounted on the floor of a vehicle that is connected to the
rear axle. The instrument records the displacement of the vehicle body relative to the axle and
usually expresses it in terms of a vertical displacement per unit distance travelled, such as
mm/km. The TRRL Bump Integrator is probably the most widely used RTRRMS.
Profilometers usually employ sensors to measure the elevation of the vehicle to the pavement.
Accelerometers are double integrated to get the movement of the vehicle through space. The
difference between these values gives the road profile elevation. These data are processed to
obtain the roughness.
Since the response of a vehicle to road roughness will depend upon the characteristics of the
suspension system, tyres and other factors, it is necessary to calibrate the vehicle against a
standard reference roughness. The International Roughness Index (IRI) was developed by the
World Bank (Sayers et al., 1986) and is based on a complex simulation of a vehicle
suspension. HDM-III used QI but since then IRI has become the international standard and all
roughnesses should be expressed in terms of IRI m/km. The conversion between QI and IRI is
1 QI = 13 IRI m/km.
In order to calculate the IRI it is necessary to obtain an accurate profile of the pavement
surface on a number of test sections. From this profile, one calculates the IRI. The vehicle is
then operated over these sections and a regression equation is developed between the
roughness meter readings and the reference roughness. This constitutes roughness meter
calibration.
Once the vehicle has been calibrated, it is necessary to measure the roughness of the
pavements. There are several points that should be considered in undertaking these
measurements:
! Sampling interval
It is best to record the roughness at the shortest practical interval - if possible, every 100
m. When longer lengths such as 1 km are used, short sections of poor pavements will be
disguised by other sections in better condition.
! Measurement speed
The measurements should be made at a single standard speed. Where data loggers are
available and roughnesses are low, this can be upwards of 80 km/hr. If different speeds
are to be used, it is necessary to have calibration equations for each speed.
! Vehicle load
The vehicle should be calibrated and operated at standard loads. In addition, the tyre
pressures should also be standard.
D.5.3 Rutting
To be included in a subsequent edition of this document
D.5.4 Potholes
To be included in a subsequent edition of this document
Table D.4
Example of maintenance treatments
Fixed sections are defined at a regular interval, for example between km stones. Dynamic
sections are defined on the basis of the data attributes.
When characterising the condition of a road, each measurement has its own appropriate
sampling interval:
! Roughness - may be measured at intervals of, say, 100 m,
! Deflection - at intervals of 50 m,
When using fixed links it is necessary to aggregate the data over the arbitrary link length,
even if the interval is not the best for the data items being measured.
Dynamic links avoid this by creating links based on the condition (or any other parameter
such as traffic) of the pavement. This leads to a much more realistic representation of the
network since the treatments one applies will be based on the pavement condition. The
process by which these links are created is termed automatic or dynamic sectioning.
Dynamic sectioning is achieved by defining the allowable variation in an attribute, for
example roughness + 1.0 IRI m/km; width + 0.5 m. The condition data are then analysed and
when the condition deviates from these variations a new section is created. This is illustrated
in Figure D.1 for a hypothetical road section.
Roughness Measurements
Roughness Sections
Rut Measurements
Rut Sections
Analysis Sections
It is not unusual for fixed section systems to lead to inappropriate treatments due to the
averaging of characteristics over the section disguising short, poor sections. This is illustrated
in Figure D.1 where the fixed sections would result in fundamentally different treatments
being applied to the dynamic analysis sections.
One can establish homogeneous sections based on a range of attributes. Attributes commonly
used to establish homogeneous sections include:
! Roughness
! Strength
! Width
! Traffic volume
! Environment
The sectioning process can be visualised as driving along a road. Whenever an attribute
changed by a certain amount a new section would be created. For example, a:
! Roughness - change of more than x IRI m/km from the mean roughness of the section to
that point
! Strength - change of more than y from the mean strength of the section to that point
! Width - change of more than z m from the mean width of the section to that point
Having established the homogeneous sections it is necessary to prepare a single data file for
use in the HDM modelling. This is done using dynamic segmentation and is illustrated in
Figure D.2. Depending upon the data attribute, one will take the average, weighted average,
maximum, minimum, or dominant feature.
500 m 300 m
Section 1 Section 2
Dynamic Segmentation
800 m
Analysis Section
E.1 Introduction
Traffic volume is a critical input data item for all HDM analyses. Since the total transport
costs are dominated by road user costs which are proportional to the traffic volume. In
addition, the commercial vehicle volumes influence the rate of pavement deterioration. This
appendix describes traffic volumes, growth rates and other associated issues.
n
q= …(E.1)
t
where:
Annual traffic figures are expressed in veh/day. There are two important measures, depending
on how the data are sampled:
! Average Annual Daily Traffic (AADT)
! Passenger Car Units (PCU)
Average Annual Daily Traffic (AADT) is the average number of vehicles on a road over the
year:
Annual Traffic
AADT = …(E.2)
365
This can only be calculated at locations where there is continuous traffic counting.
Since it is more common to conduct short-term traffic counts, most data are termed Average
Daily Traffic (ADT). This is dependent upon the length of the traffic count and it is important
to specify what it refers to; for example:
! 5 - day ADT - is used for urban streets
The base traffic data supplied to HDM must be an AADT. ADT are converted to AADT
using correction factors. These are often based on the time of year when the traffic is counted.
If, for example, the count is made in winter or during the monsoon season the factor usually
increases the ADT to convert it to an AADT. Conversely, summer or dry season counts are
often reduced to make them an AADT. Table E.1 shows seasonal correction factors from
India (NDLI, 1997). The data in this table indicate, for example, that an ADT from a count
made at Location 1 in August would be multiplied by 1.13 to convert it to an AADT.
Hourly traffic flows are expressed in veh/h. This is the most common measure to use in traffic
evaluations. HDM uses hourly flows in its congestion analyses.
The traffic stream is comprised of a number of different vehicle types. It is therefore common
to convert these to more homogeneous measures.
Table E.1
Example of seasonal correction factors from India
Passenger Car Units (PCU) is a measure that converts all vehicles to equivalent passenger
cars. These are most commonly applied on rural road appraisals where the larger vehicles
have a significant negative impact on traffic flow. The values for PCU conversions vary
between countries and depend upon factors such as the vehicle size and power to weight ratio.
Recently, the measure Bicycle Space Equivalent (BSE) has been proposed as a measure for
use in countries such as China and India where there are high levels of non-motorised traffic
(Yuli, 1996).
In HDM, traffic is expressed in terms of Passenger Car Space Equivalents (PCSE). The PCSE
differs from the PCU in that it is based on the area occupied by the vehicle1. A discussion of
PCSE and how they are calculated was given in Section 6.4.3.
It is important to consider the traffic composition as well as the total volume since different
vehicle classes have different operating costs and impacts on pavements. The traffic stream is
modelled in HDM using representative vehicles and so the default HDM representative
vehicles should be used as a guide when considering the traffic survey classes. Issues such as
1
PCSE are primarily based on length. However, as described in Section 6.4.3, Hoban et al. (1994)
recommend increased values of PCSE for narrower roads to reflect the greater effects of the vehicles on the
total road space under these conditions.
the likely operating cost differences between vehicle classes, occupancies, loading levels, etc.
all should be assessed. As a minimum, the following are the recommended vehicle classes for
HDM analyses:
! Non-motorised traffic
! Motorcycles
! Passenger cars
! Light trucks
! Medium/heavy trucks
! Mini-buses
! Heavy buses
20
18
16 USA - Seasonal
14
Hourly Flow as % AADT
12
10
USA - Commuter
6 India
USA - Intercity
4
New Zealand
2
0
0 1000 2000 3000 4000 5000 6000 7000 8000 9000
Cumulative Number of Hours per Year
For HDM it is necessary to establish the number of hours per year the traffic is at different
flow levels. It is common to adopt at least five levels. These should consist of a low and high
flow level, as well as an appropriate number of intermediate levels. When there are non-
motorised traffic (NMT), it is necessary to account for these in the flow bands. Analysing the
NMT traffic associated with the same flow bands as the motorised traffic does this.
Table E.2 is an example of how the hourly distribution is established (NDLI, 1997). Short-
term traffic counts were expanded to cover a full year (8760 h/year) and broken down into
five flow bands. After reviewing the distribution five flow-bands were selected and the
number of hours in each band was established along with the mean flow.
Table E.2
Example of hourly distributions from India
365 AADT =
∑ HRYR HRVOL
i =1
i I …(E.3)
where:
The values for HRVOLi should be adjusted to ensure this check is made.
closure or a local catastrophe. If such drops are not recorded elsewhere, they should probably
be ignored.
30000
T raffic Count Data from Surat-Dhulia Road, Gujarat, India
25000
20000
7 Day ADT in PCU .
15000
10000
5000
0
April '88
April '89
April '90
April '91
April '92
April '93
April '94
April '95
April '96
May '87
Oct '87
Oct '88
Oct '89
Oct '90
Oct '91
Oct '92
Oct '93
Oct '94
Oct '95
Figure E.2 Example of variation in traffic over time
The two usual ways of expressing growth are as a geometric or as an arithmetic growth rate.
These are calculated as:
! Geometric
(YEAR −1)
AADTyear = AADTbase 1+
GROWTH
…(E.4)
100
! Arithmetic
GROWTH
AADTyear = AADTbase 1+ (YEAR − 1) …(E.5)
100
where:
Using historical traffic trends is the least accurate method for predicting growth rates. This is
because many developing countries are experiencing rapid motorisation and economic growth
which will alter the historical trends.
Calculating growth from economic trends is often a more sound method than historical data
since there are strong correlations between traffic volume and economic activity, something
that a straight historical model will not capture. The underlying approach is to collect data on
traffic growth rates and then to do a statistical analysis that fits a model to the data. The
variables often used are income, population, fuel use, industrial production, agricultural
production, or any other measure of the economy. These lead to elasticities which express the
change in traffic as a function of the economic measures. Typical equations would be:
where:
The use of fuel is preferred by some since it gives a good overall view of the historical
growth in traffic. By comparing the changes in fuel use with changes in GDP one calculates
the effect of GDP changes on fuel use. Coupling this with the forecast GDP gives the total
forecast traffic growth. This method has advantages in that it implicitly considers the
population growth.
As an example of the elasticities, in India CES (1991) gave values for the GNP elasticity of:
! 1.75 and 1.0 for freight vehicles
! 1.0 and 0.5 for passenger vehicles
in the periods 1994-2005 and after 2005 respectively. Values of 1.43 and 2.71 were used for
the population elasticity for freight and passengers respectively. In Nepal (NDLI, 1993)
adopted a value of 1.5 for the elasticity of fuel. The use of declining elasticities with time is
common as this reduces the impacts of future uncertainties on the predictions.
One important feature of traffic growth is that it is different depending upon the level of a
country’s development. Developed countries, with high levels of vehicle densities have
markedly different car ownership growth rates than developing countries that are building up
their densities. The model that best reflects this is a sigmoidal (‘S’ shaped) model. This is
discussed in Button and Ngoe (1991) who present a generalised model with data covering a
number of developing countries to use with the model.
WARNING
When forecasting traffic, always consider whether or not the predictions are
reasonable. It is easy to adopt what seems to be a relatively low growth rate,
such as 5 per cent, but this may mean that after a period of time your facility is
at capacity. For this reason it is often prudent to adopt several different growth
rates based on short, medium, and long-term considerations, with the rates
declining in future years.
There are four principal categories of future traffic which are forecast:
! Normal traffic
The future traffic that can be expected assuming the current trends (for example,
historical patterns) remains steady.
! Diverted traffic
The traffic that can be expected to divert to the road because of the improvement.
! Generated traffic
Traffic that would not have existed but is expected because of reduced travel times or
diversion from other modes.
! Induced traffic
Traffic expected because of the new development created by better access (for example,
building a road into a new area will open it up for development).
x+y
q= …(E.8)
t
where:
This expression assumes that the flows in each direction are balanced. To apply this equation
one first converts the flows to veh/h for by-hour-of-day. Using data on the variation of flow
by-hour-of-day, the one-hour flow is converted to an ADT. This is then factored up to an
AADT using the procedure described earlier.
In the Tamil Nadu Project Co-ordinating Consultancy Study, Riley (1998) adjusted the hourly
flow from the moving survey using factors based on 7-day counts that were done using
manual methods. Figure E.3 shows the excellent agreement between 7-day count data
factored from the moving survey and those using data from the actual 7-day count stations.
8,000
R2 = 0.97
6,000
5,000
4,000
3,000
2,000
1,000
0
0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000
Figure E.3 Comparison of moving car and 7-day survey counts from India
When a wheel crosses the rubber tube and air impulse is sent along the tube which
activates an air switch. They are very common and relatively reliable over several days
and simple to operate. Their disadvantages are that the tube in time wears and breaks and
they are unsuitable where there is heavy breaking or turning. They are mainly confined
to short-term count stations.
! Inductive loops
detecting the vehicle. It is almost universally used for permanent count stations as it is
very reliable, relatively inexpensive and largely unaffected by the weather. It is used at
almost all traffic signals.
! Positive contact
The detector is activated by the weight of the vehicle causing two thin metal strips to
touch and so to complete an electrical circuit. Provides an excellent signal but fairly
short life so limited to special counting and research.
! Video
Consists of a two-core cable that creates a charge when it comes into contact with a
wheel.
❏ Piezo-electric film
D
v1 = …(E.9)
t 21 − t 11
D
v2 = …(E.10)
t 22 − t 12
SPACING 1 = v 1 (t 12 - t 11 ) …(E.11)
SPACING 2 = v 2 (t 22 - t 21 ) …(E.12)
where:
The values for v1 and v2 represent the velocity of axle 1 and the velocity of axle 2 (in m/s).
The spacings are the distances between axle 1 and axle 2 (in m), based on these velocities.
These values are usually very similar, with the differences due to timing errors in the
detectors. It is common to average the values or else to adopt only one.
It must be appreciated that there are errors associated with the measurement of all automated
traffic counters and even when installed perfectly there will still be errors due to the rounding
off of the timing data. For example, if the recorder is accurate to a millisecond, the time will
fall within the range of 0.001 + 0.0005 secs. While this may seem to be quite small, if the
detectors are placed close together (D in the above equations), it can result in a sizeable error
when measuring speeds. As described in Bennett and Dunn (1992), on the basis of this it is
recommended that detectors be spaced at 5 m intervals.
The axle spacing data allows a vehicle to be classified into a number of different classes. It
needs to be appreciated that all automatic classifiers follow certain logic in assigning the
vehicles to specific classes. These are usually based on the common vehicle types in the
country of development so may not be appropriate to every country. Unfortunately, suppliers
of traffic counters are often reticent to modify their systems for local conditions so the output
often needs to be scrutinised carefully.
There are a number of caveats to the use of automated counters, the:
! Count sites should not be located in places where they are likely to be interfered with
(for example, near schools).
! Road at the site should be straight and level.
! Detector must be at right angles to the traffic flow; and if two detectors are used they
must be parallel.
! Detector must be firmly fixed to the road surface.
! Detector or connections must not cross the footpath.
! Counter must be securely locked to a permanent object such as a telephone pole.
If only a single detector is used the counter will only record the total number of axles and the
counter assumes that two axles equals one vehicle. This overestimates the total volume and a
correction factor must be applied, calculated from manual classified counts as follows:
Number of Vehicles
Correction Factor = …(E.13)
0.5 Number of Axles on Vehicles
A correction factor is calculated for a specific site, but various sites can be averaged to
produce factors for an area, or different classifications of roads.
vehicles one needs to weigh to achieve suitable confidence in the data from the WIM scale.
Slavik (1998) addresses this issue and shows that the sample size is much smaller for
estimating the average load as opposed to the number of equivalent standard axles. Figure E.4
gives the 90% confidence intervals for both of these based on a sample of 218 axles (Slavik,
1998).
The Slavik (1998) approach takes into consideration the properties of the WIM equipment
used, the condition of the road surface, the composition and loading of truck traffic which
makes the procedure site specific. The raw axle loads are multiplied by a correction factor to
convert them to adjusted axle loads:
a i = k ri …(E.14)
where:
80
70
ESA
60
90% confidence inteval, % of mean
50
40
30
20
Load
10
0
50 100 150 200
Number of Axles Used in Calibration
This suppresses the systematic error but affects the distribution of axle loads. The
distribution is therefore corrected by converting the adjusted axle loads into corrected axle
loads using the equation:
(
c i = a i + f ai − a i ) …(E.15)
where:
f is a correction factor
The values for k and f are established from the on-site calibration data using the following
equations. Their derivation is described in Slavik (1998).
si − σ si si + σ si
n n
k = 0.5 + …(E.16)
ai ai
f = 1 − Ve i / Va i …(E.17)
where:
F.1 Introduction
The HDM road model operates by predicting the amount of resources consumed and
multiplying them by the unit costs. It is therefore necessary to supply unit cost data for RUE
and works effects (WE). These costs can be supplied as financial and/or economic costs,
which are defined below.
The RUE unit costs can be established from various sources that are listed in Table F.1. In
establishing these unit costs, they should reflect the cost over the life of the project, normally
10 - 20 years. For most items the current prices can be used as the basis since inflation can be
expected to influence the various components similarly so that they maintain the same cost in
relation to one another. An exception to this is often fuel and lubricants that are heavily
influenced by world supply and demand. As described later, it is therefore better to estimate
these based on the long-term average future oil prices1.
Table F.1
Sources for unit cost data
For works effects, unit costs are best established from recent tender prices or contracts.
1
For example, Hoff & Overgaard (1994) indicate that in early 1994 the prevailing price for a barrel of oil
(158.6 l) was $USD 15 whereas the long-term average price was expected to be $USD 20.
2
HDM-III also included foreign exchange costs.
encountered in using shadow pricing. While shadow pricing is desirable, for most studies it is
sufficient to quantify the economic costs, as the market costs net of taxes.
To illustrate how this is done, consider Figure F.1. This shows the economic and financial
cost components of petrol and a medium truck from Thailand. Petrol had a series of taxes and
levies added to the cost of production and supply. The taxes with the truck had two
components, the:
1 Import duty on the CKD chassis
Since these were levied at different stages of production one cannot just deduct them directly
from the financial cost.
Petrol Truck
6.2% Dealer
6.3% Sales Tax
8.9% Body
Tax
Tax
Tax 56.2%
17.9% Assembly
Financial Cost
Retailing 6.7%
Transport 5.6%
Economic Cost
Refining 4.5%
51.8%
CKD Chassis
Shipping 4.5%
Oil 22.5%
Figure F.1 Breakdown of 1992 Thailand petrol and medium truck prices
When deducing the taxes from the market cost, it is important to consider not only the direct
but also the indirect taxes. To illustrate this, consider the cost of petrol from Thailand shown
in Figure F.1.
The indirect tax component of this price amounted to 56.2 per cent. The cost of production
and supply 43.8 per cent, however, it is necessary to deduct the direct tax component from
this remaining cost. This would entail assessing profit margins and the direct taxes on these
profits. The foreign exchange component would consist of the refinery cost less production
costs and margins along with a component of distribution costs. For simplicity, most analyses
would ignore the direct taxes and just use the total indirect taxes.
Hoff & Overgaard (1992) as well as Hoff & Overgaard (1994), Kampsax (1992), NDLI
(1997), Kinhill (1998) used an alternative approach to that outlined above. Instead of starting
with the financial cost of fuel and deducting the taxes, this method begins with the cost of a
barrel of oil and increases the costs to reflect the refining, transport and marketing costs. It
has the advantage that it will identify any hidden taxes that may be missed using the above
approach. These hidden taxes may arise in situations such as Nepal where the Indian
government taxes fuel exports to Nepal and these taxes are later refunded as a credit to the
Nepal government. Since these taxes are external to Nepal several projects missed removing
them when calculating the economic cost of fuel. An example of this approach from Hoff &
Overgaard (1994) is given in Table F.2.
Table F.2
Example of calculating economic cost of fuel
Another argument in favour of using the price of a barrel of oil for calculating fuel prices is
that oil exporting countries often price fuel below the prevailing world price for oil. As noted
by Hoff & Overgaard (1994), changes in fuel consumption as a result of road projects affect
the quantity of fuel available for export and the potential export as opposed to the market
price is therefore the appropriate value of fuel to use.
1
It should be noted that in HDM-III this would lead to an underestimation of the interest costs.
from financial to economic before calculating the weighted-average. If there are uniform
taxes for all vehicles within a class, the weighted-average financial cost can be directly
converted to the weighted-average economic cost with a single factor.
In many countries vehicles are completely rebuilt at a certain kilometreage. This sees major
components replaced or refurbished. If the cost of rebuilding is included in the new vehicle
price the depreciation costs will be higher over the life of the vehicle. It is better to have two
different vehicles, one for vehicles that have not been rebuilt and a second for those that have
been rebuilt. This will reflect the different parts costs while not biasing the depreciation costs.
In HDM-III there was no mechanism to include a residual value for the vehicle in the
depreciation cost. These can be quite high, particularly in countries where old vehicles are
assigned to different activities, such as port work. To circumvent this problem, Transroute
(1992) assumed that vehicles had a 30 per cent residual value. The residual value was
discounted at the assumed accounting rate of interest and deducted from the replacement
vehicle price. This was done to reflect the “cascading down of both types of vehicles in their
later years to short distance, low annual mileage, low maintenance, long life work”. Since this
procedure served to reduce the maintenance and repair costs, the latter were adjusted through
the calibration parameters. HDM-4 provides for a residual value in the analysis.
1 Establish the cost of inner tubes per 1000 km (ITC) for an average road.
2 Run HDM using average conditions and establish the tyre cost for each representative
vehicle class in cost/1000 km (TYRE).
3 Multiply the average tyre cost by 100 ITC/TYRE to obtain the adjusted tyre cost. This
adjusted tyre cost should be used in all further modelling.
The issue of radial versus bias tyres; high quality, long-life versus lower-quality, shorter life
tyres is accounted for by calculating a weighted-average cost. The number of different tyre
sizes sold for each vehicle within a representative vehicle class should be weighted by the
average cost of the tyres to obtain a weighted-average cost. If there are significant differences
in prices between countries of origin, the averaging should also reflect this. Contacting tyre
retailers and asking them for an assessment of the percentages of sales of different tyre types
can readily collect these data. For example, in Nepal NDLI (1993) recorded the tyre types on
a sample of vehicles stopping at a customs checkpoint.
The HDM tyre model also requires the cost of retreads as a percentage of the new tyre price.
This is used in the calculations to establish an average tyre cost for vehicles that use a mix of
retreads and new tyres. There are two considerations here, the:
1 Average value
The average value is calculated in the same manner as for a regular tyre. This value then
needs to be reduced based on the probability of a retreaded tyre being purchased since HDM
assumes an equal likelihood of a retread or a new tyre being on a vehicle. This is done as
follows:
1 Interview tyre retailers, other industry sources, truck drivers, etc. and establish the
percentage of tyres in use, which are retreads (PCTRET).
2 Express the average retread cost as a percentage of the new tyre cost (RETPCT).
3 Calculate the retread cost as:
RREC = PCTRET RETPCT
1
This relates to the question of to the appropriateness of treating a large number of small time savings as
equal to a single large time saving of the same magnitude. There are several arguments in favour of this, but
the most common is that since small time savings cannot be perceived, they should not be valued lower
than large time savings - hence the non-linearity of the value of time savings. Symonds (1997) reviews the
issue and concludes “… there are no compelling theoretical or empirical reasons to adopt non-linear
assumptions on values of time savings, and there are strong practical reasons in favour of a constant
[value]”. They note that the exception to this is leisure time where there are different levels of disutility but
that for practical reasons linear values should be used.
While some studies have simply ignored passenger time, this is not wise as it can bias the
results in favour of projects with high transport costs. Conversely, an unrealistically high
value for passenger time can bias projects in favour of those with large speed increases.
In quantifying the value it should be recognised that there are two sectors of the economy: the
formal and the informal sectors. The formal sector is comprised of wage earners. The
informal sector is those without salaries - such as many of the rural population in developing
countries.
For simplicity, travel is usually separated into work and leisure or non-work travel. Work
travel is comprised of those on business activities. Leisure, comprised of other activities (for
example, social, visiting family, personal business, going to school, etc.) is often a significant
component of travel in developing countries.
Savings in time when journeys are related to work clearly have a value; if less time is spent
travelling more time in the working day can be used for economically productive purposes.
Another way of looking at this is the employer pays the employee an hour’s wages for no
return. The employer would be willing to pay equal to an hour’s wages to reduce travel time
by one hour. It can be argued that due to overheads and social charges the employer would
be prepared to pay even more, but the common practice in developing countries is to equate
the value of work time to the earnings rate of the traveller. In developed countries, where
there are often large social costs, this gross wage is increased by the employer’s on-costs.
The use of wage rates is complicated by the fact that official statistics on wages will probably
underestimate the earnings of travellers. Wage statistics do not usually cover the earnings of
the highest paid workers, and the wages of those travelling during working time may be
higher than the average. Often, there are also regional variations in wages that make it
impractical to adopt a national average. For example, in Thailand the 1995 average wage was
Baht 9900/month in Bangkok versus 4500 in the north-eastern region of the country.
Those in the informal sector or travelling in leisure time are not considered to be productive
in the same way as those travelling in work time. Ultimately, the value of non-working time
should reflect Government policy. If the policy is to maximise GDP, ignoring leisure time
preferences and increasing the welfare of passengers, then a zero value should be placed on
non-work time. It must be recognised that assigning a zero value to the time for those in the
informal sector will serve to bias the results in favour of those who contribute to the modern,
cash economy.
There is evidence that the leisure time savings are valued, particularly since these travellers
still prefer their trips to be faster than slower and are often willing to pay more for this to
happen. How much a person is prepared to pay for a quicker trip is based upon their income
and wealth. It is therefore common practice to assume a value of personal time related to the
individual’s income. Various percentages have been assumed in different studies, usually in
the range 20-50 per cent, but 20-25 per cent seems to be the most common.
Many who travel in personal time do not earn any income and so using this approach would
have no value for their time. In affluent societies this would not be true, but it is argued by
some that in some countries a zero value of time is appropriate.
In these instances the mean income is used to calculate the value of time. The alternative
approach is to calculate the value of time based only on those working and then to apply the
value to all travellers. This will yield a higher value of time than using the mean income of
travellers.
To summarise, there are three sets of passenger time values to be considered:
1 Employed, travelling in work time,
There is evidence that travel time values are higher for traffic travelling under congested as
opposed to free-flow conditions. MVA et al. (1987) suggested a factor of 1.4 for the situations
they studied, and an even higher factor under more congested conditions.
When establishing the value of time, particularly for truck and bus operators, it is important
to include extra income that may be obtained above the base salary, for example:
! Daily allowances - to cover food and rest,
! Backhaul of goods by trucks - where the operator instead of the owner keeps the
income.
These can be significant, for example in India NDLI (1997) estimated that the total salaries
for the truck driver and helper were Rs 4500 but that there was a further monthly income of
Rs 3000 from collecting passengers and Rs 3000 from daily allowances. About 20-25 per cent
of drivers managed to backhaul goods, earning a further Rs 7000 per month.
It is also common to differentiate between modes of travel. This, for example, sees different
passenger time costs for passengers in private transport and those in buses or other public
transport.
Table F.3 gives an example of calculating passenger time and crew costs, using basic data
from India NDLI (1997) along with some assumed additional values. The notes at the bottom
of the table detail how the individual values were calculated.
Notes:
(3) = (2) x 0.20. Assumed 20% for value of work time.
For cars which had less than one crew, (15) = (14) x Rs 3,000
(19) is calculated from the extra revenue (here Rs 7,000/month) times the percentage of backhauls
(here, approximately 20 per cent).
If goods spend less time in transit the amount of goods held in inventory may be able to be
reduced. Some goods will not benefit from the time savings, for example, if they arrive before
the business opens and thus cannot be unloaded. Thus, only a portion of the goods in transit
should be included in the calculations. The value will depend upon the nature of operations in
the country, or even in the area of the country, but values of 50-75 per cent are common.
The value of cargo time is calculated using the opportunity cost and the following equation:
where:
1
Some analysts adopt the approach of calculating an average hourly rate based on a percentage of
skilled/semi-skilled/unskilled time spent working on a vehicle. This method is generally not appropriate
since in many countries the practice is to have more than one person working on the vehicle at a time. Thus,
the costs should be accumulated rather than weighted.
G.1 Introduction
s
E= …(H.1)
n
where:
The true mean of the population is aid to lie between x + E. Thus, the larger the sample size
the smaller the value of E and the closer the sample mean will be to the population, thereby
limiting the bias.
The necessary sample size for a given level of accuracy and confidence is given by:
s
e=K …(H.2)
n
where:
Rewriting the above equation gives the following equation for determining the required
sample sizes:
K 2 s2
n= …(H.3)
e2
Hamilton (1990) gives the values in Table H.1 for 90 and 95 per cent confidence intervals
using the t Distribution (n < 30). Table H.2 gives the values of K for the Normal Distribution
(n > 30).
In order to use the above equations for determining the required sample size, it is necessary to
know the standard deviation of the sample. Since this is not available until after the study has
been completed, it appears impossible to estimate the required sample sizes before the study.
However, the distributions of many data items are relatively stable between sites and between
countries and have a consistent value for the coefficient of variation. This is defined as:
σ
COV = …(H.4)
µ
where:
µ is the mean
Table H.1
Critical ‘t’ values (n < 30)
Table H.2
Critical K values (n > 30)
Confidence level K
68.3 1.00
90.0 1.65
95.0 1.96
95.5 2.00
99.0 2.58
99.7 3.00
1
For those years preceding the sampling factor the survival ratio is set to 1.0.
Table I.1
Example of establishing survival curve and lifetime utilisation
Notes:
Sampling factor:
where:
The analysis is done in such a way that the error term is minimised.
There are instances where the dependent and independent variables are interchangeable, for
example as in the HDM equations to predict modified structural number from the Benkelman
Beam deflection and vice versa. This gives rise to two sets of equations:
The issue here is one of orthogonal regression where instead of minimising the sum of
squares in the x or y plane the minimisation needs to be done in the orthogonal plane. To
illustrate the importance of this, consider Table J.1. Columns 1 and 2 contain sample data.
These data are plotted in Figure J.1 along with the linear regression equations.
Table J.1
Orthogonal regression illustrative data
9.6 2.5
9.5
9.4 2.0
Y = -0.5561X + 9.9987
2 X = -1.6568Y + 16.695
9.3
R = 0.92
R2 = 0.92
9.2 1.5
9.1
Y
9.0 1.0
8.9
8.8 0.5
8.7
8.6 0.0
0.0 0.5 1.0 1.5 2.0 2.5 8.6 8.7 8.8 8.9 9.0 9.1 9.2 9.3 9.4 9.5 9.6
X Y
Column 3 in Table J.1 shows the value of Y predicted using Equation J.2 above along with
the original X data from Column 1. Using these values as input data for Equation J.3 above
gives the value of X. As shown in Column 5, there is a difference between these predicted
values of X and the original X values. This difference would not exist had an orthogonal
regression been done.
Orthogonal regressions require sophisticated statistical techniques. However, when the
residuals are similar in each direction the following simple solution may suffice.
y=ax +b …(J.6)
x=cy +d …(J.7)
so:
(x − d)
y=
c
a+ 1
c
( ) (
b−d
c
)
y= x+ …(J.8)
2 2
and:
1 b−d ( )
x= y− c
( ) ( )
…(J.9)
a+
a+ c
1 1
c
2
Using this approach with Equations J.4 and J.5 above results in the following equations:
Column 6 in Table J.1 shows the predicted values of y using Equation J.4 above in
conjunction with the data from Column 1. Substituting these values into Equation J.5 above
gives the predicted values of x (Column 7). These values are the same as the original data in
Column 1.
K.1 Introduction
HDM Tools is a set of software applications designed to assist in the calibration of HDM-4.
The software is available from the Internet at:
! http://www.opus.co.nz
! http://www.htc.co.nz
The HDM Tools applications have been designed to run completely independently of the
HDM-4 software, although the data can be used to update the HDM-4 data files.
The applications fall into three groups:
! RUE
! RDWE
! Analysis Tools
These applications are run from a single launch button and output data that can be imported
to the HDM Vehicle Fleet.
If the default HDM-4 vehicle parameters are used it is not necessary to run the HDM Tools
applications. However, if the default values are changed then the guidelines in Table K.1
should be used to establish if the HDM Tools applications need to be re-run.
Table K.1
Criteria for re-running RUE HDM tools
OPTIMAL For all HDM applications using the optimal life for • • •
capital costs
1
Comments and suggestions on the software can be sent to: ian.greenwood@opus.co.nz.
The second option enables for comparison of simulated versus observed fuel consumption
data such that the input vehicle parameters may be adjusted to yield the best match. Here, the
user defines a speed cycle and the software predicts the fuel consumption associated with
travelling that cycle. Comparing this to known fuel consumption enables key parameters to be
calibrated.
To run ACCFUEL the user supplies the number of vehicles to simulate and the minimum
length of road that each vehicle should travel over. It is recommended that a minimum of 250
vehicles travelling 5-10 km be simulated. For each mean initial speed, vehicles are simulated
travelling with acceleration noises from 0.05 to 1.00 in 0.05 m/s2 increments. This results in a
matrix as illustrated in Figure K.3 and Figure K.4 for fuel and tyres respectively. The fuel
data are also presented as Table K.2.
1.2
1.0
0.8
dFUEL 0.6
0.4
0.90
0.2
0.75
0.60
Acceleration Noise
0.45 in m/s/s
0.0
0.30
10
20
30
0.15
40
50
60
70
0.00
80
90
Speed in km/h
100
While the values show consistent trends, the fuel results show the inherent variations that
arise with Monte Carlo simulations. The tyre results do not have these irregularities due to the
way in which the tyre consumption is calculated (Bennett and Greenwood, 1999). The
following points will be noted in the figures:
! The highest additional fuel consumption arises in the area of the minimum fuel
consumption. This is most noticeable with heavier vehicles.
! There is a significant increase in congestion effects with heavier vehicles over light
vehicles for the same speed and acceleration noise. This is because of their higher mass
and, thus, inertial resistance.
! There is only a very limited effect of congestion on motorcycles - speed has a much more
significant impact. This is because motorcycles have such a low mass that the inertial
effects on fuel are minimal.
9.0
8.0
7.0
6.0
5.0
dTYRE
4.0
3.0
2.0
0.90
0.75
1.0
0.60
0.45 Acceleration Noise
0.0 in m/s/s
0.30
10
20
30
0.15
40
50
60
70
0.00
80
90
Speed in km/h
100
When generating a matrix there are two output files per vehicle, one for fuel consumption and
one for tyre wear. Each file has the format of vehicle speed down the rows and acceleration
noise across the columns. The name of the files is dfuel.xx and dtyre.xx, where xx is the
vehicle number within the vehicle database.
If you run the Combine files option from the File menu, then the program will combine all
the various files into two large files ready for inclusion into HDM-4. These files are called
dfuel.out and dtyre.out.
To simulate a known speed profile, the user must supply a text file with the vehicle number to
simulate followed by rows containing the time (seconds), velocity (m/s), speed (km/h),
acceleration (m/s/s), distance (m). An example of this is:
This table is then used in the analysis and then a single output file is created with the same
name as the input file but the extension .OUT. This contains all the input data plus two extra
columns, the first being the instantaneous fuel consumption (ml/s) and the second the
cumulative fuel consumption (ml).
Table K.2
Additional fuel due to accelerations – medium truck
100
LIFEKMPCT = MIN100, a1
…(K.1)
1 + EXP(a0 RI )
where:
Figure K.5 shows the input screen for the program. The program uses the following data from
the VEHICLES data file:
! Replacement value of the vehicle
To run the program it is necessary to supply a base roughness level. This is the roughness at
which the lifetime utilisation applies. For example, if the average roughness on the network is
6 IRI m/km and the average lifetime utilisation is 100,000 km you would supply the value of
6. However, if you estimated that on smooth roads the lifetime utilisation would be 200,000
km and this value is entered in the data file, you would enter 3 IRI m/km for the roughness.
The user selects the vehicles to analyse and the optimal life at different roughnesses is
generated. Varying the parts consumption coefficient kp based on the user-supplied values
and this roughness level does this. Once it has the value of kp that matches the optimal life
information, it then calculates the life at a roughness of the 3 IRI m/km (which if the base
roughness is 3 equals the lifetime utilisation). It then goes on to calculate the percentage of
life at various roughness levels relative to the life at a roughness of 3 IRI m/km (that is, not
the supplied lifetime utilisation).
Table K.3 is an example of the program output.
Table K.3
Optimal life analysis output
1 3.5 89.6
1 4.0 83.3
1 4.5 77.9
1 5.0 73.3
1 5.5 69.6
1 6.0 65.8
1 6.5 62.9
1 7.0 60.0
1 7.5 57.5
1 8.0 55.4
1 8.5 53.3
1 9.0 51.3
1 9.5 49.6
1 10.0 47.9
The data from OPTIMAL need to be analysed to quantify the coefficients a0 and a1. This is
done with the non-linear regression program supplied with HDM Tools through the script file
illustrated in Figure K.6.
Figure K.6 Script file for fitting HDM-4 OL equation to raw data
RPM = a0 + a1 SP + a2 SP 2 + a3 SP 3 …(K.2)
where:
Figure K.7 shows the input screen for the program. The user defines the gear ratio data,
which are readily available from manufacturer’s specifications, along with certain other key
attributes. Bennett and Greenwood (1999) discuss these values in detail. Selecting calibrate
the program then applies the simulation logic described in Bennett and Greenwood (1999) to
generate a table of the mean engine speed and mean effective mass ratio as a function of road
speed.
1
NDLI (1995a) proposed a different formulation based on three zones but as shown in Bennett and
Greenwood (1999), the above polynomial formulation gives more than adequate results.
1
The weight of the product at a wind angle of 0 and 180 degrees was halved to account for the full 360
degrees of possible wind angles.
Table K.4
Engine speed simulation results
Road speed Engine speed Effective Road speed Engine speed Effective mass
(km/h) (RPM) mass ratio (km/h) (RPM) ratio
20 1693 1.719 75 2347 1.114
70 2230 1.117
It is recommended that ψc be assumed to be a constant value of 30o for all vehicle classes.
The vehicle and wind speeds will vary depending on applications. As shown in Table K.5,
the value for h varies between vehicle classes.
Table K.5
Aerodynamic resistance default parameters
1
Vehicle number Type h CDmult CD AF
2
(m )
1 Motorcycle 0.4 1.12 0.70 0.8
Notes:
1 Calculated assuming wind speed of 14.4 km/h and vehicle speed of 75 km/h
! A range of values for the work zone capacity, work zone length, work zone speed and
AADT.
! An hourly flow distribution over the 24-hour period.
For each AADT level and the hourly flow distribution, the program generates the arrivals at
the work zone. These arrivals will apply irrespective of the work zone capacity or length.
For each combination of work zone capacity, work zone length and work zone speed the
program will calculate the average delay and queue size, in two directions.
The WORKZONE simulation model is capable of analysing the impact of road closures on
road users. The user inputs the configuration of the lane closure from any of the following:
! Lane closure - typically motorway situations
The start-up screen for the software is illustrated in Figure K.9. The first tab is used to
execute the analysis while the other tabs are used for data.
Figure K.9 Screenshot of Start-up screen to the Work Zone simulation model
Figure K.10 shows the traffic flow data. Instead of defining the number of individual vehicles
of each class to simulate, the program works in terms of an aggregate number of vehicles.
Accordingly, it is necessary to provide the average PCSE for each vehicle. This should be
based on the PCSE values for individual vehicles, weighted based on their frequency in the
traffic stream.
The user must define speed-flow data, which is based on the HDM speed-flow model
presented in Chapter 6. The hourly flow profiles give the variation in flow over the course of
the day. Both of these data are used to create a unique traffic ID that consists of a volume,
speed-flow profile, hourly flow profile, and a capacity for the site.
As shown in Figure K.11, the user must define the work zone closure type, the length of the
work zone, the maximum speed through the work zone, the work zone capacity, and the
number of lanes in each direction. If the capacity will vary over a day (for example if the road
will be closed for periods of time) this can also be specified.
Once all the input data have been defined the program is used to calculate the delays. The
output from the program is in a format suitable for performing a regression analysis to
determine various model coefficients for input to the HDM-4 road user effects model or to
apply as an exogenous cost.
K.5 Analysis
The non-linear regression program is NLREG. This is a shareware program and you are
expected to register it with the author should you decide to use the program. Registration
details are provided with the software.
volume five
A short executive summary describing the HDM-4 system. It is intended to be used by all readers new to HDM-4,
particularly high-level management within a road organisation.
ISBN: 2-84060-059-5
PIARC, Registered in France These HDM-4 products have been produced by the International Study of Highway
Development and Management Tools (ISOHDM), sponsored by The World Bank, the
Registered Office Asian Development Bank, the Department for International Development (UK), the
The World Road Association (PIARC) Swedish National Road Administration, and other sponsors. HDM-4 is jointly published
La Grande Arche by The World Road Association (PIARC), Paris, and The World Bank, Washington, D.C.
Paroi Nord, Niveau 8
92055 La Defénse Cedex PIARC endeavours to ensure that the information in this document is correct and fairly
France stated but does not accept liability for any error or omission.
Tel: +33 1 41 02 05 84 The development of PIARC products and services is continuous and published
Fax: +33 1 49 00 02 02 information may not be up to date. It is important to check the current position with
Email: piarc.hdm4@attglobal.net PIARC. This document is not part of a contract or licence save insofar as may be
Web: http://hdm4.piarc.org expressly agreed.