Austroads - AP-R445-13 - Standardised Bridge Barrier Design PDF

AP-R445-13
AUSTROADS RESEARCH REPORT
Standardised Bridge Barrier Design

Published October 2013
© Austroads Ltd 2013
This work is copyright. Apart from any use as permitted under the Copyright Act 1968,
no part may be reproduced by any process without the prior written permission of Austroads.
ISBN 978-1-925037-26-5
Austroads Project No. TT1601
Austroads Publication No. AP-R445-13
Project Manager
Nigel Powers, VicRoads
Prepared by
Dr Hanson Ngo, Rudolph Kotze and Dr Neal Lake
ARRB Group
Acknowledgements
The project team would like to acknowledge the valuable contributions
from the Working Group, Australian and New Zealand road jurisdictions
and consultants, and the Austroads Bridge Technology Task Force.
Published by Austroads Ltd

Level 9, Robell House
287 Elizabeth Street
Sydney NSW 2000 Australia
Phone: +61 2 9264 7088
Fax: +61 2 9264 1657
Email: austroads@austroads.com.au
www.austroads.com.au
Austroads believes this publication to be correct at the time of printing and does not accept
responsibility for any consequences arising from the use of information herein. Readers should
rely on their own skill and judgement to apply information to particular issues.
Sydney 2013
About Austroads
Austroads’ purpose is to:
 promote improved Australian and New Zealand transport outcomes
 provide expert technical input to national policy development on road and road transport
issues
 promote improved practice and capability by road agencies.
 promote consistency in road and road agency operations.
Austroads membership comprises the six state and two territory road transport and traffic
authorities, the Commonwealth Department of Infrastructure and Regional Development, the
Australian Local Government Association, and NZ Transport Agency. Austroads is governed by a
Board consisting of the chief executive officer (or an alternative senior executive officer) of each of
its eleven member organisations:
 Roads and Maritime Services New South Wales
 Roads Corporation Victoria
 Department of Transport and Main Roads Queensland
 Main Roads Western Australia
 Department of Planning, Transport and Infrastructure South Australia
 Department of Infrastructure, Energy and Resources Tasmania
 Department of Transport Northern Territory
 Department of Territory and Municipal Services Australian Capital Territory
 Commonwealth Department of Infrastructure and Regional Development
 Australian Local Government Association
 New Zealand Transport Agency.
The success of Austroads is derived from the collaboration of member organisations and others in
the road industry. It aims to be the Australasian leader in providing high quality information, advice
and fostering research in the road transport sector.
Standardised Bridge Barrier Designs
SUMMARY
Consistency in the design of traffic barriers for bridges has been identified as a major issue in
current practice throughout Australia and New Zealand. The need for standard solutions was
recognised by road jurisdictions, consultants and contractors during the Austroads 6th Bridge
Conference Workshop on Bridge Barriers, held in 2006. The main areas of concern are the
determination of appropriate barrier performance levels, structural design criteria, the lack of
standard barrier design details, guidance on retrofitting existing bridge barriers, bridge approach
barriers and overpass bridge support protection.
The aim of this project is to develop a set of guidelines to assist bridge designers, contractors and
jurisdictions to provide consistency and cost savings in the selection and development of relevant
bridge barriers throughout Australia and New Zealand. This report builds on, and provides input to,
the Australian Bridge Design Standard AS 5100 (2007), currently under review.
This report includes the following guidelines and key features:

 Selection of bridge barrier performance levels, introducing site condition categories and
barrier specification standards to facilitate the selection of barrier performance levels based
on the provisions of AS 5100.1 (2004).
 Bridge barrier design procedure, providing detailed guidance with practical examples for the
structural design of popular types of bridge barriers adopting the AASHTO Load and
Resistance Factor Design (LRFD) method based on yield line theory.
 Development of standardised bridge barrier designs, providing a set of standardised concept
designs which incorporate most road jurisdictions’ policy and future directions on bridge
barriers. This set of concept designs has been endorsed by members of Austroads Bridge
Technology Task Force (BTTF).
 Retrofitting of existing bridge barriers, providing a methodology for the selection of
appropriate barrier performance levels, evaluation of existing bridge structures and design of
barrier retrofit solutions.
 Bridge approach barriers, providing design procedures and methodologies for determination
of barrier performance levels, length of need and consideration of various topics such as
barrier layout, transition, and end treatment examples.
 Bridge support protection barriers, providing a methodology for determining the level of
protection, selection of barrier performance level and barrier layout design including
examples.
The report forms part of the current review of AS 5100 (2007) and identifies key areas of future
work to be undertaken after AS 5100 has been published. These areas include the detail design
and specification of agreed standard barrier types and a testing program to verify actual in-service
performance.
Austroads 2013
— i—
CONTENTS
1 INTRODUCTION ................................................................................................................... 1
1.1 Background ........................................................................................................................... 1
1.2 Aims ...................................................................................................................................... 1
1.2.1 Overall Aims ................................................................................................................. 1
1.2.2 2010–11 Assess Current Practice ................................................................................. 2
1.2.3 2011–12 Develop Guidelines and Identify Research ..................................................... 3
1.2.4 2012–13 Finalise Guidelines and Recommend Research ............................................. 4
1.3 Scope .................................................................................................................................... 5
1.4 Outline ................................................................................................................................... 5
2 DISCUSSIONS ON THE DEVELOPMENT OF THE GUIDELINES ....................................... 7

2.1 Jurisdictions’ Current Practice ................................................................................................ 7
2.1.1 Common Bridge Barrier Systems .................................................................................. 7
2.1.2 Design Guidelines......................................................................................................... 7
2.1.3 Issues and Recommendations ...................................................................................... 8
2.2 Current Design Standard AS 5100 ......................................................................................... 9
2.3 Computer Simulation ........................................................................................................... 10
2.4 Safe System Approach ........................................................................................................ 11
2.4.1 Safe System ............................................................................................................... 11
2.4.2 Contribution of Bridge Barrier Design to the Safe System ........................................... 11
2.5 Development of the Guidelines ............................................................................................ 12
2.5.1 Guidelines for Selection of Bridge Barrier Performance Levels (Section 3) ................. 12
2.5.2 Bridge Barrier Design Procedure (Section 4) .............................................................. 15
2.5.3 Standardised Bridge Barrier Designs (Section 5) ........................................................ 16
2.5.4 Guidelines for Bridge Barrier Retrofitting (Section 6) ................................................... 17
2.5.5 Guidelines for Bridge Approach Barriers (Section 7) ................................................... 18
2.5.6 Guidelines for Bridge Support Protection Barriers (Section 8) ..................................... 18
3 SELECTION OF BRIDGE BARRIER PERFORMANCE LEVELS ....................................... 19

3.1 General Procedure .............................................................................................................. 19
3.1.1 Bridge Barrier Performance Levels ............................................................................. 19
3.1.2 Site Condition Categories ........................................................................................... 20
3.2 Criteria for Site Condition CAT1 ........................................................................................... 21
3.5 CAT1 Method....................................................................................................................... 23
3.6 CAT2 Method – Chart-based Procedure .............................................................................. 23
3.6.1 Determine AADT and Portion of Commercial Vehicles................................................ 23
3.6.2 Determine Adjusted AADT .......................................................................................... 24
3.6.3 Determine Barrier Performance Level ......................................................................... 27
3.6.4 Example 1................................................................................................................... 29
3.7 CAT3 Method – Specification Standards and B/C Analysis .................................................. 30
3.7.1 Specification Standards for Medium Performance Level ............................................. 30
3.7.2 Specification Standards for High Performance Level .................................................. 31
3.7.3 Specification Standards for Special Performance Level .............................................. 31
3.7.4 Benefit-cost Analysis .................................................................................................. 32
3.7.5 Example 2................................................................................................................... 41
Austroads 2013
— ii —
4 BRIDGE BARRIER DESIGN PROCEDURE........................................................................ 49

4.1 Barrier Performance Specifications ...................................................................................... 49
4.2 Structural Design Criteria ..................................................................................................... 49
4.3 Selection of Bridge Barrier Types......................................................................................... 53
4.3.1 Concrete Barriers........................................................................................................ 53
4.3.2 Post-and-rail Barriers .................................................................................................. 55
4.3.3 Combined Concrete Parapet and Metal Rails ............................................................. 55
4.3.4 Acceptance Criteria for Non-standard Bridge Traffic Barriers ...................................... 56
4.4 Concrete Parapets ............................................................................................................... 56
4.4.1 Barrier Geometry ........................................................................................................ 56
4.4.2 Design Process........................................................................................................... 56
4.4.3 Design of Bridge Cantilevered Deck Slab ................................................................... 61
4.5 Post-and-rail Barriers ........................................................................................................... 63
4.5.2 Design Process........................................................................................................... 65
4.6 Concrete Parapet and Metal Rail ......................................................................................... 70
4.6.2 Design Process........................................................................................................... 70
4.7 Design Examples ................................................................................................................. 73
4.7.1 Design Example 1: Concrete Parapet – Medium Performance Level .......................... 73
4.7.2 Design Example 2: Post-and-rail Barrier – Regular Performance Level ...................... 75
4.7.3 Design Example 3: Combined Concrete Parapet and Metal Rail – Medium
Performance Level ................................................................................................... 77
5 STANDARDISED BRIDGE BARRIER DESIGNS ................................................................ 85

5.1 Introduction .......................................................................................................................... 85
5.2 Bridge Barrier Designs Currently in Use in Jurisdictions ...................................................... 85
5.2.1 Low Performance Level Barriers ................................................................................. 86
5.2.2 Regular Performance Level Barriers ........................................................................... 89
5.2.3 Medium Performance Level Barriers ........................................................................... 94
5.2.4 High Performance Level Barriers ................................................................................ 97
5.2.5 Special Performance Level Barriers ............................................................................ 99
5.3 Specific Design Requirements from Jurisdictions ............................................................... 100
5.4 Selection of Bridge Barrier Concept Designs for Standardisation ....................................... 102
6 RETROFITTING OF EXISTING BRIDGE BARRIERS ....................................................... 109

6.1 Introduction ........................................................................................................................ 109
6.2 Reasons for Retrofitting Existing Bridge Barriers ............................................................... 109
6.3 Barrier Retrofitting Design Methodology............................................................................. 110
6.3.1 Design Procedure ..................................................................................................... 110
6.3.2 Determine Ideal Barrier Performance Level .............................................................. 110
6.3.3 Evaluate the Existing Bridge Structure ...................................................................... 110
6.3.4 Rationalisation of Barrier Performance Level ............................................................ 112
6.3.5 Design Methodology ................................................................................................. 112
6.4 Determination of Appropriate Barrier Performance Level ................................................... 113
6.4.1 Site-specific Risk Assessment .................................................................................. 113
6.4.2 Benefit-cost Analysis ................................................................................................ 113
6.4.3 Additional Considerations ......................................................................................... 114
Austroads 2013
— iii —
6.5 Barrier Retrofit Design Procedure ...................................................................................... 115

6.5.1 Characteristics of the Bridge Location....................................................................... 115
6.5.2 Design for Upgrading Existing Barriers ..................................................................... 115
6.5.3 Design for Replacement of Existing Barrier............................................................... 120
6.5.4 Retrofitting Existing Structure.................................................................................... 123
6.5.5 Design of Connections .............................................................................................. 124
6.5.6 Retrofitting of Bridge Approach Barriers .................................................................... 124
6.6 Bridge Barrier Upgrade Exceptions .................................................................................... 124
6.7 Special Considerations for Historic Bridges ....................................................................... 125
7 BRIDGE APPROACH BARRIERS .................................................................................... 126

7.1 Introduction ........................................................................................................................ 126
7.2 Current Design and Specification Requirements ................................................................ 126
7.2.1 Terminology .............................................................................................................. 126
7.2.2 Current AS 5100 Requirements ................................................................................ 127
7.2.3 Additional Requirements from Jurisdictions’ Guidelines ............................................ 128
7.3 Design Procedure .............................................................................................................. 130
7.4 Determination of Length of Need ....................................................................................... 130
7.4.1 Run-out Length Method ............................................................................................ 131
7.4.2 Angle of Departure Method ....................................................................................... 136
7.5 Selection of Barrier Performance Level .............................................................................. 141
7.5.1 Methodology ............................................................................................................. 141
7.5.2 Performance Level and Associated Length Required to Contain Errant Vehicles...... 142
7.5.3 Considerations to Preventing Errant Vehicles Entering High Risk Areas behind
Approach Barriers .................................................................................................. 142
7.5.4 Considerations Relevant to Minimum Lengths of Different Performance Level
Bridge Approach Barriers ....................................................................................... 143
7.5.5 Methodology for Determining Performance Levels and Associated Length of
Median Bridge Approach Barrier Systems .............................................................. 144
7.6 Typical Layouts .................................................................................................................. 144
7.7 Transition Between Different Performance Level Barriers .................................................. 148
7.7.1 General..................................................................................................................... 148
7.7.2 Design Criteria – Physically Connected Barriers ....................................................... 148
7.7.3 Typical Interfaces Between Barrier Types ................................................................. 149
7.8 End Treatments ................................................................................................................. 157
7.9 Gore Area .......................................................................................................................... 160
7.10 Freeway and Highway Ramps and Side Roads ................................................................. 160
7.11 Bridge Approach Off-structure Barrier Foundation ............................................................. 160
7.12 Examples of Approach Barrier ........................................................................................... 160
7.12.1 Example 1: Bridge Approach Barrier Layout Design ............................................... 160
8 BRIDGE SUPPORT PROTECTION BARRIERS ............................................................... 165

8.1 Introduction ........................................................................................................................ 165
8.2 Current Design and Specification Requirements ................................................................ 165
8.2.1 Current AS 5100 Requirements ................................................................................ 165
8.2.2 Additional Requirements from Jurisdictions’ Guidelines ............................................ 166
8.3 Design Procedure .............................................................................................................. 166
8.4 Determination of Level of Protection .................................................................................. 166
8.4.1 Levels of Protection .................................................................................................. 166
8.4.2 Risk Assessment ...................................................................................................... 168
8.4.3 Selection of Performance Level ................................................................................ 169
Austroads 2013
— iv —
8.5 Barrier Layout and Design ................................................................................................. 170

8.5.1 Length of Need ......................................................................................................... 170
8.5.2 Dynamic Deflection ................................................................................................... 170
8.5.3 Vehicle Roll Allowance.............................................................................................. 171
8.5.4 Typical Barrier Layout for Bridge Supports Adjacent to Undivided Roadways ........... 172
8.5.5 Typical Barrier Layout for Bridge Supports Located in Median Area ......................... 173
8.5.6 Height Considerations............................................................................................... 174
8.5.7 Foundations for Rigid Barriers .................................................................................. 175
8.6 Bridge Support Protection from Railway Traffic .................................................................. 175
8.7 Typical Design Drawings ................................................................................................... 176
8.8 Examples ........................................................................................................................... 180
8.8.1 Example 1: Protection Barrier for a Bridge Pier Adjacent to an Undivided
Two-way Road ....................................................................................................... 180
8.8.2 Example 2: Protection Barrier for a Bridge Pier Located in a Median ........................ 184
9 PROPOSED FUTURE RESEARCH .................................................................................. 188

9.1 Development of Detailed Standard Barrier Designs ........................................................... 188
9.2 Computer Simulation ......................................................................................................... 189
9.3 Safe System Approach ...................................................................................................... 190
9.4 Barrier Test Program ......................................................................................................... 190
9.5 Update to Austroads Guide to Bridge Technology ............................................................. 190
REFERENCES ........................................................................................................................... 191
APPENDIX A RECOMMENDATIONS OF 2006 WORKSHOP ................................................. 194
APPENDIX B MEETING OF EXPERT GROUP 2010............................................................... 197
APPENDIX C COMMENTS FROM CONSULTANTS AND CONTRACTORS .......................... 203
APPENDIX D EXPERT CONSULTATION ............................................................................... 207
APPENDIX E BTTF CONSULTATION .................................................................................... 219
Austroads 2013
— v—
TABLES
Table 1.1: Project working phases .............................................................................................. 1
Table 2.1: Jurisdictions’ technical design guidelines for bridge barriers ....................................... 7
Table 2.2: Proposed standard barrier types............................................................................... 16
Table 3.1: Bridge barrier performance levels ............................................................................. 20
Table 3.2: Road type factor ....................................................................................................... 25
Table 3.3: Example of summary table for crash cost related to severity index ........................... 40
Table 3.4: Example 2 – road data ............................................................................................. 42
Table 3.5: Example 2 – lane data .............................................................................................. 42
Table 3.6: Example 2 – traffic data ............................................................................................ 42
Table 3.7: Example 2 – base encroachment frequency ............................................................. 44
Table 3.8: Example 2 – distribution of encroachment angles ..................................................... 44
Table 3.9: Example 2 – accident cost ........................................................................................ 46
Table 3.10: Example 2 – costs of the two barrier alternatives ...................................................... 47
Table 4.1: Traffic barrier design loads, contact lengths and effective heights ............................ 50
Table 5.1: Agreed standard designs for Low performance level barriers ................................. 102
Table 5.2: Agreed standard designs for Regular performance level barriers ........................... 102
Table 5.3: Agreed standard designs for Medium performance level barriers ........................... 104
Table 5.4: Agreed standard designs for High performance level barriers ................................ 106
Table 5.5: Agreed standard designs for Special performance level barriers ............................ 107
Table 7.1: Run-out lengths for barrier design .......................................................................... 131
Table 7.2: Flare rate ................................................................................................................ 132
Table 7.3: Shy line offset values.............................................................................................. 132
Table 7.4: Clear zone distances from edge of through travelled way ....................................... 135
Table 7.5: Curve adjustment factors ........................................................................................ 136
Table 7.6: Angles of departure from the road .......................................................................... 137
Table 7.7: VicRoads typical layouts of bridge approach barriers ............................................. 144
Table 7.8: Typical barrier transitions used in TMR................................................................... 157
Table 8.1: Indicative deflection for concept/feasibility design ................................................... 170
Table 8.2: Vehicle roll allowance ............................................................................................. 172
Table 8.3: Applicable height of rigid concrete barriers ............................................................. 175
Table 9.1: Changes to bridge barrier design loads .................................................................. 189
FIGURES
Figure 3.1: Barrier performance level flowchart .......................................................................... 22
Figure 3.2: Chart-based procedure............................................................................................. 24
Figure 3.3: Grade factor GD ....................................................................................................... 25
Figure 3.4: Curvature factor CU.................................................................................................. 26
Figure 3.5: Under-bridge condition factor US.............................................................................. 26
Figure 3.6: Threshold limits 60 km/h ........................................................................................... 27
Figure 3.7: Threshold limits 80 km/h ........................................................................................... 27
Figure 3.8: Threshold limits 100 km/h ......................................................................................... 28
Figure 3.9: Threshold limits 110 km/h ......................................................................................... 28
Figure 3.10: Example with threshold limits 110 km/h .................................................................... 30
Figure 3.11: Simple process for risk assessment ......................................................................... 33
Figure 3.12: Flowchart of risk assessment process ...................................................................... 38
Figure 3.13: Curvature adjustment factor CU ............................................................................... 39
Figure 3.14: Probability encroachment curve................................................................................ 43
Figure 4.1: Bridge barrier component layout ............................................................................... 49
Figure 4.2: Barrier design forces ................................................................................................ 50
Austroads 2013
— vi —
Figure 4.3: Barrier design diagram ............................................................................................. 51

Figure 4.4: Typical types of concrete barriers ............................................................................. 54
Figure 4.5: Failure mechanism for impact within wall segment ................................................... 57
Figure 4.6: Failure mechanism for impact near end wall segment .............................................. 59
Figure 4.7: Calculation of Mb, Mw and Mc .................................................................................... 59
Figure 4.8: Possible failure mechanisms for F-shape barriers and the like ................................. 60
Figure 4.9: Design sections in the overhang region .................................................................... 62
Figure 4.10: Post-and-rail barriers ................................................................................................ 64
Figure 4.11: Possible failure modes for post-and-rail barriers ....................................................... 66
Figure 4.12: Contribution of yield line resistance of rails ............................................................... 67
Figure 4.13: Effective length of cantilever for carrying concentrated post loads ............................ 68
Figure 4.14: Concrete parapet and metal rail – impact at mid-span of rail .................................... 70
Figure 4.15: Concrete parapet and metal rail – impact at post...................................................... 71
Figure 4.16: Barrier design example 1 .......................................................................................... 73
Figure 4.17: Barrier design example 2 .......................................................................................... 75
Figure 4.18: Barrier design example 3 – barrier cross-section ...................................................... 77
Figure 4.19: Barrier design example 3 – parapet .......................................................................... 78
Figure 6.1: Barrier retrofitting design procedure ....................................................................... 111
Figure 6.2: Standard timber handrail for timber bridges ............................................................ 116
Figure 6.3: Timber handrail for concrete bridges ...................................................................... 117
Figure 6.4: Pipe handrail .......................................................................................................... 117
Figure 6.5: Reinforced concrete handrail type 1 – details at concrete post ............................... 118
Figure 6.6: Reinforced concrete handrail type 1 – intermediate post details ............................. 119
Figure 6.7: Reinforced concrete handrail type 2 ....................................................................... 119
Figure 6.8: Proprietary bridge barrier retrofitting system ........................................................... 120
Figure 6.9: Casterton-Edenhope Rd over Glenelg River ........................................................... 120
Figure 6.10: Replacement option for concrete bridge decks, narrow kerbs................................. 121
Figure 6.11: Replacement option for concrete bridge decks, intermediate kerb widths ............... 121
Figure 6.12: Replacement option for concrete bridge decks, wide kerbs .................................... 122
Figure 6.13: Replacement option for concrete bridge decks, deck or footway wider
than 650 mm .......................................................................................................... 122
Figure 6.14: Replacement option for concrete bridge decks, for Regular performance
level ........................................................................................................................ 123
Figure 7.1: Terminology used for determining performance levels and extents ........................ 129
Figure 7.2: Barrier LON – single direction ................................................................................. 130
Figure 7.3: Barrier LON – two directions................................................................................... 131
Figure 7.4: Run-out length method of determining LON ........................................................... 133
Figure 7.5: Length of need on outside curve using run-out length method................................ 134
Figure 7.6: Length of need on inside curve using run-out length method .................................. 134
Figure 7.7: Angle of departure method of determining LON...................................................... 137
Figure 7.8: Angle of departure method on curves where leading angle meets the rear
of hazard ................................................................................................................ 140
Figure 7.9: Angle of departure method where leading angle does not meet the rear of
hazard .................................................................................................................... 141
Figure 7.10: Bridge approach barriers for Medium and High performance levels – left
verge on steep terrain............................................................................................. 145
Figure 7.11: Bridge approach barriers for Medium and High performance levels – left
verge on flat terrain................................................................................................. 146
Figure 7.12: Typical layouts of guard fence on bridge approach ................................................. 147
Figure 7.13: Interface details of type F to W-beam (meets TL-3) ................................................ 150
Figure 7.14: Interface details of type F to thrie-beam (meets TL-3) ............................................ 151
Figure 7.15: W – thrie transition section ..................................................................................... 152
Figure 7.16: Structure connector – thrie-beam ........................................................................... 152
Austroads 2013
— vii —
Figure 7.17: Interface details of type F to thrie-beam.................................................................. 153

Figure 7.18: Layout for WRSB match with existing structures .................................................... 154
Figure 7.19: Transition from Regular performance level bridge barrier to roadside
barrier ..................................................................................................................... 154
Figure 7.20: Transition from High performance level bridge barrier to roadside barrier............... 155
Figure 7.21: Transition from Medium performance level bridge barrier to roadside
barrier ..................................................................................................................... 156
Figure 7.22: Termination details of Medium performance level steel bridge barriers................... 158
Figure 7.23: Termination details of Regular performance level steel bridge barriers ................... 159
Figure 7.24: Example: determine the layout of bridge approach barrier ...................................... 161
Figure 7.25: Example: using run-out length method ................................................................... 163
Figure 7.26: Example: using angle of departure method ............................................................ 164
Figure 8.1: Vehicle roll allowance ............................................................................................. 171
Figure 8.2: Typical barrier layouts for bridge supports adjacent to undivided roadways ............ 172
Figure 8.3: Examples of road safety barriers for use in narrow medians................................... 173
Figure 8.4: An example of a barrier layout for shielding a bridge pier located in a
median ................................................................................................................... 173
Figure 8.5: Two-stage protection layout.................................................................................... 174
Figure 8.6: Example: VicRoads typical bridge support protection system ................................. 177
Figure 8.7: Typical collision protection system for a median pier with inclined faces
parallel to the trafficked lanes ................................................................................. 178
Figure 8.8: Typical collision protection system for a median pier with vertical faces
parallel to the trafficked lanes ................................................................................. 178
Figure 8.9: Example 1: Bridge pier protection barrier................................................................ 180
Figure 8.10: Example 1: Options for bridge pier protection barrier layout using run-out
length method......................................................................................................... 182
Figure 8.11: Example 1: Options for bridge pier protection barrier layout using angle of
departure method ................................................................................................... 183
Figure 8.12: Example 2: Site plan............................................................................................... 184
Figure 8.13: Example 2: Layout of rigid barrier for approaching traffic using run-out
length method......................................................................................................... 186
Figure 8.14: Example 2: Layout of rigid barrier for approaching traffic using angle of
departure method ................................................................................................... 187
Austroads 2013
— viii —
1 INTRODUCTION
1.1 Background
During the Austroads 6th Bridge Conference Workshop on Bridge Barriers in 2006, a number of
issues regarding current bridge barrier design practice were raised. Refer to Appendix A for details.
 It was recognised by conference delegates that the need for individual consultants and
contractors to develop new details for many projects caused inefficiency and uncertainty.
There was recognition that although road jurisdictions may install barriers with the same
nominal performance level, current practice resulted in barriers that have a range of details
and actual performance levels.
 Consequently there was strong support from consultants and contractors attending the
workshop for Austroads and individual road jurisdictions to rationalise the design and
construction of bridge (and roadside) barriers by developing guidelines. Such guidelines
should specify performance requirements and standard barrier systems, and provide more
guidance on bridge approach treatments.
 It was identified that there is a need to address the current practice of individual designers
undertaking risk assessments for determining required performance levels and selecting or
designing appropriate barrier systems and foundations for individual sites. Further, there was
an urgent need for road authorities to take a lead role in specifying and standardising
requirements for common bridge and bridge approach situations on the road network.
 In particular, there was a need to achieve more consistency in the provision of bridge
approach barriers and their foundations. There was also a need to develop recommended
systems for providing pier protection for road over road, road over rail and rail over road
overpasses. An important requirement to identify barrier systems that have been tested by
various organisations and are considered acceptable to Austroads members was noted. In
some instances, there may also be some barrier systems that individual road authorities wish
to nominate as being satisfactory for their needs.
This project aims to develop a set of guidelines (building on the requirements of AS 5100 (2007)
and the recommendations of the Austroads 6th Bridge Conference Workshop on Bridge Barriers)
that will assist bridge designers, contractors and asset owners and provide consistency and cost
savings in the selection and development of relevant bridge barriers, their installation and
maintenance for bridge sites throughout Australia.
1.2 Aims
1.2.1 Overall Aims
This is a three-year project, with three working phases as shown in Table 1.1:
Table 1.1: Project working phases

Period Working phase
2010–11 Assess current practice
2011–12 Develop guidelines and identify research
2012–13 Finalise guidelines and recommend research
Austroads 2013
— 1—
The overall aims of the project include:

 development of standard practice for standard bridge barrier systems
— risk assessment
— testing
— design
— installation
— maintenance
 incorporation of the current practice into a set of barrier design guidelines
 development of standard practice for retrofitting of existing barriers
 recommendations for amendments to the Austroads Guide to Bridge Technology (as
appropriate).
1.2.2 2010–11 Assess Current Practice

Task (a): Review existing material
Task (a) involved an assessment of current provisions and procedures for risk assessment for
barriers associated with bridges, and the collation and comparison of current bridge barrier
practices. It provided input to task (b).
Method (a)
 Review provisions of AS 5100 for risk assessment and determination of barrier performance
level requirements, lengths of need and end treatments.
 Review common procedures for undertaking risk assessments for a representative range of
typical bridging situations.
 Review current practices and barrier systems that are being designed and constructed in
accordance with the requirements of AS 5100 by the various road authorities, consultants
and contractors.
 Review any existing road authority guidelines or other guidelines for bridge and bridge
approach barriers.
2010–11 Output and milestones

The output for 2010–11 was a progress report entitled Toward Standardised Bridge Barrier
Designs – Stage 1 which summarised the various reviews and consultations, and identified issues
to be resolved. It was proposed that this report would be an internal report.
A number of milestones identified in the 2010–11 program was covered by interim reports to the
project manager and advisory group and consolidated into the progress report (i.e. output),
namely:
 review of AS 5100
 review of risk assessment procedures
 review of current practice for bridge barriers.
Relevant aspects of the review of AS 5100 were provided as input to a future revision of that
standard.
Austroads 2013
— 2—
It should be noted that a crash data analysis was included in the original contract note. This task
involved the determination of the numbers and types of crashes occurring on bridges and major
culverts and the approaches to these structures. The objective of the task was to identify what
contribution the crash data may make to the risk assessment procedures used by road authorities.
However, after conducting a comprehensive search and working group discussions, it was
identified that crash data is not usually available at the location of a new bridge on a new road. It is
different from a ‘black spot’ on an existing road where crashes may occur and be recorded. Crash
data relating to each bridge barrier type is rare and not practical to obtain. Therefore, it was agreed
that this task should be removed from the project scope.
1.2.3 2011–12 Develop Guidelines and Identify Research

Task (b): Develop guidelines and identify further research activities
Task (b) focused on the development of a set of guidelines that will assist bridge designers,
contractors and asset owners and provide consistency and cost savings in the selection and
development of relevant bridge barriers, their installation and maintenance for bridge sites
throughout Australia. The guidelines should take account of the recommendations of the Austroads
6th Bridge Conference Workshop on Bridge Barriers, namely that they should include:
 new bridges and retrofitting to existing bridges
 risk assessment guidance and examples
 performance level selection guidance, procedures and examples
 barrier design guidance and examples
 review of computer simulation
 Austroads approved tested barrier systems
 Austroads other accepted bridge and approach barrier systems.
Method (b)
The progress report for 2010–11 was used as the basis for workshops of the expert group and
other experts to:
 resolve issues relating to differences in practice between jurisdictions
 develop an agreed structure and general content for the guidelines
 investigate and confirm the role of computer simulation in the design of barrier systems
 facilitate the preparation and review of draft versions of the guidelines
 facilitate agreement to guidelines that best meet the needs of member authorities
 identify and scope further research required to fill gaps in the guidelines or resolve issues
relating to different practices in jurisdictions. If appropriate, this may include
— physical dynamic and static testing of barrier systems including pier protection systems
— numerical simulation of vehicle – barrier impacts.
Austroads 2013
— 3—
2011–12 Output and milestones

The output for 2011–12 was:
 a draft set of guidelines
 a report that
— summarises the findings of the review of existing material and data analysis, discusses
guidelines and standards that reflect the common requirements of road authorities and
identifies specific requirements of individual authorities that reflect local requirements
— discusses the process involved in developing the guidelines
— includes details of acceptable tested barrier systems, and other standard drawings or
diagrams of recommended details and systems
— discusses the implementation of the recommendations of Austroads 6th Bridge
Conference Workshop on Bridge Barriers
— identifies issues that require research to complete the guidelines or enhance future
versions of the guidelines
— provides recommendations.
This report was entitled Standardised Bridge Barrier Designs – Stage 1.
1.2.4 2012–13 Finalise Guidelines and Recommend Research

Task (c): Implement research – finalise guidelines
Task (c) involves the undertaking of research that is able to be accommodated within the available
project budget and the finalisation of the guidelines for bridge barriers. It also includes:
 recommended amendments to the Austroads Guide to Bridge Technology (as appropriate)
 preparation of a PowerPoint™ presentation to disseminate the project findings.
2012–13 Outputs and milestones

The outputs for 2012–13 include:
 a set of guidelines for bridge barriers
 a report that
— describes and recommends research that the Bridge Technology Task Force (BTTF)
considers necessary in relation to the further development of bridge barrier design
guidelines
— includes recommendations for amendments to the Austroads Guide to Bridge
Technology (as appropriate)
 a PowerPoint™ presentation to disseminate findings.
The final report entitled Standardised Bridge Barrier Designs consists of:
 a set of guidelines for bridge barriers
 descriptions and recommendations of research that the BTTF considers necessary in relation
to the further development of bridge barrier design guidelines
 recommendations for amendments to the Austroads Guide to Bridge Technology (as
appropriate).
Austroads 2013
— 4—
The set of guidelines includes the following sections:

 risk assessment and selection of bridge barrier performance levels
 bridge barrier design
 retrofitting existing bridge barriers
 bridge approach barriers
 bridge support protection barriers
 standardised bridge barrier systems.
The final report incorporates all information generated during the three years of the project, and
supersedes all previous interim and draft reports.
It should be noted that, due to the fact that changes in the barrier design loads have been
proposed in the on-going revision of the Australian Standard Bridge Design: Design Loads
AS 5100.2 (2004), the BTTF has decided that a set of standardised bridge barrier systems will be
developed after the revised AS 5100.2 has been published. The same approach is applied to the
update of the Austroads Guide to Bridge Technology. As a result, this project will only provide a set
of concept barrier designs with agreed shapes for different categories.
1.3 Scope
This report covers road bridge barriers, bridge approach barriers, barriers for pier protection for
road over road and road over rail overpasses. Noise barriers, pedestrian barriers, bicycle path
barriers or barriers on railway or pedestrian bridges are out of the scope for this report.
1.4 Outline
The report includes nine sections including the Introduction (Section 1). The content of other
sections is summarised below.
Section 2 summarises findings from a review of AS 5100 (2007) and current practices in bridge
barrier design in various jurisdictions and industry in Australia and New Zealand. A brief review of
computer simulation used in safety barrier design and the Safe System approach is also
presented. This section also discusses the process involved in the development of the guidelines
included in this project.
Section 3 covers the guideline for risk assessment and selection of the bridge barrier performance
level in which a procedure is developed to facilitate the selection process based on the provisions
of AS 5100.1 (2004). Site condition categories and specification standards for barrier performance
levels are introduced for ease of implementation. A benefit-cost (B/C) analysis procedure for
justification of the selection of the barrier performance level is also presented. Examples are
provided for illustration purposes.
Section 4 provides a suggested procedure and detailed guidance for the structural design of
popular types of bridge barriers. The outcome of this section can be used as a tool to assess the
structural adequacy of existing bridge barrier designs, and assist in the development of
standardised bridge barrier design systems. Detailed numerical examples are provided to illustrate
the design procedure.
Section 5 presents a set of standardised concept designs which incorporate most road
jurisdictions’ policy and future directions on bridge barriers. This set of concept designs has been
endorsed by members of the BTTF.
Austroads 2013
— 5—
Section 6 provides a guideline for retrofitting of existing bridge barriers. Methodology for the
selection of appropriate barrier performance level, evaluation of existing bridge structure and
barrier retrofitting design is included. Barrier retrofitting examples from various road jurisdictions
are also provided.
Guidelines for designing bridge approach barriers are presented in Section 7. This section consists
of a design procedure, methodologies for determination of the barrier performance level and length
of need, and considerations on various topics such as barrier layout, transition, and end treatment.
An example is also provided.
Guidelines for the design of bridge support protection barriers are provided in Section 8. This
guideline presents the methodology for determining the level of protection, selection of barrier
performance level and barrier layout design. Examples are provided to illustrate the design
process.
Discussions are presented in Section 9 regarding the future research directions, with identified
issues that require further research to complete the guidelines as well as to enhance future
versions of the guidelines.
Appendices are provided to include the implementation of the recommendations of the Austroads
6th Bridge Conference Workshop on Bridge Barriers (Appendix A), the consultation with the
working group (Appendix B and Appendix D), industry (Appendix C) and the BTTF (Appendix E).
Austroads 2013
— 6—
2 DISCUSSIONS ON THE DEVELOPMENT OF THE

GUIDELINES
This section summarises the results of a review of current practice throughout road jurisdictions in
Australia and New Zealand on bridge barriers and presents the discussions related to the process
involved in the development of the guidelines included in this project.
2.1 Jurisdictions’ Current Practice

2.1.1 Common Bridge Barrier Systems
A collection of bridge barrier designs currently in use in various jurisdictions throughout Australia
and New Zealand have been collected and reviewed. This collection includes:
 standard barrier designs issued by road jurisdictions
 barrier designs used in actual projects.
Refer to Section 5.2 for a list of common barrier designs which have been used in Australia and
New Zealand.
It should be noted that not all road jurisdictions have issued standard bridge barrier designs. Some
road authorities rely on consultants/contractors to provide the designs on a project basis.
Significant variations have been observed among these designs in barrier type, typical
cross-section, and structural details.
Although these designs have been classified by their performance level, there is no proven
evidence of their compliance to the current standard (AS 5100.1–2004) such as structural capacity
verification and crash test results.
2.1.2 Design Guidelines

Table 2.1 lists the additional key technical design guidelines for bridge barriers currently used in
jurisdictions.
Table 2.1: Jurisdictions’ technical design guidelines for bridge barriers

Jurisdiction Technical document
RMS Road design guide: section 6:safety barriers for roads and bridges (Roads and Traffic Authority 1996)
BTD2008/07 Design of bridge supports for collision load from road traffic (Roads and Traffic Authority 2008b)
BTD2007/08 Design of replacement traffic barriers on existing bridges (Roads and Traffic Authority 2009)
Standard bridge drawings for barriers and railings (Roads and Maritime Services 2012)
VicRoads BTN2001/007 Selection requirements for bridge traffic barriers (VicRoads 2001)
BTN2005/001 Improving existing bridge barriers (VicRoads 2005)
BTN2009/001 Protection of bridge supports (VicRoads 2009a)
BTN2009/002 Guidelines for bridge approach barriers (VicRoads 2009b)
RDN06/07 The performance safety barrier treatments at bridge approaches (VicRoads 2010)
Standard drawings for guard fences and barriers (VicRoads 2012)
TMR Road planning and design manual: interim advice: chapter 8: safety barriers and roadside furniture (Department of Main
Roads 2009)
QR MCE-SR-007 Design and selection criteria for road-rail interface barriers (Department of Main Roads and Queensland
Rail 2010)
Structural drafting standards: volume 3: chapter 19: bridge barriers (Department of Transport and Main Roads 2011)
Design criteria for bridges and other structures (Department of Transport and Main Roads 2012a)
Standard drawings roads manual: part 18: bridges (Department of Transport and Main Roads 2012b)
Austroads 2013
— 7—
Jurisdiction Technical document

MRWA Structures engineering design manual (Main Roads Western Australia 2009)
MRWA supplement to Austroads Guide to Road Design - Part 6 (Main Roads Western Australia 2012a)
Standard contract drawings: traffic barriers and balustrading for structures (Main Roads Western Australia 2012b)
DPTI Guide to the Selection of Safety Barriers - GD100 (2002)
Specification: Part 244 Concrete Safety Barrier (2011)
Standard drawings for bridge barriers (Department of Planning, Transport and Infrastructure 2012)
DIER Road safety barriers: design guide (Department of Infrastructure, Energy and Resources 2007)
DOT Not obtained
Roads ACT DS07 Bridges and related structures
NZTA Bridge manual: section B: bridge side protection (Transit New Zealand 2004)
The following are derived from the review:

 The selection of bridge barrier performance levels is based on the AS 5100.1 (2004)
risk-based procedure; however, this procedure applies to ‘normal’ sites only. For sites with
high risks, only descriptive advice is available. As a result, the selection of the barrier
performance levels for these sites is very subjective and may lead to variations in similar
situations.
 In the majority of the cases, the road jurisdictions specify the required performance level of
barriers. In some other cases consultants carry out risk analyses to determine the
performance levels. Most of the jurisdiction’s guidelines just restate the AS 5100.1 (2004)
provisions.
 AS 5100 (2007) does not explicitly provide a detailed structural analysis procedure for bridge
barriers. As a result, current practice uses American Association of State Highway and
Transportation Officials (AASHTO) design procedure (see AASHTO 2012) based on yield
line analysis for concrete parapets and inelastic analysis for post-and-beam barriers.
 No detailed guidelines for retrofitting existing bridge barriers, bridge support protection, and
bridge approach barriers have been available.
2.1.3 Issues and Recommendations

The following issues are summarised from project TT1601 Notes of Meeting on 12 October 2010:
 Department of Planning, Transport and Infrastructure SA (DPTI) has invested significant
effort into the development of barrier types over the past two years.
 Barrier costs can be significant (DPTI).
 Pier protection may depend on financial constraints provided by the project manager (DPTI).
No guidelines exist for the protection barrier and the distance from travel lane to the barrier.
 Barriers associated with reinforced soil walls should be included in AS 5100 and bridge guide
(RMS).
 AS 5100 requires a barrier to be provided if piers cannot withstand load, but no distance is
mentioned (RMS).
 Pier protection is an area requiring improved guidance (RMS).
 Guidelines for retrofitting of current bridges are urgently required since requests are received
from regions to retrofit damaged barriers and cognisance must be taken of the strength of the
existing structures (TMR).
 Computer models regarding the progressive collapse of barriers should be investigated
(TMR).
Austroads 2013
— 8—
 The same level of protection should be applied to road and bridge barriers (TMR).
 It is difficult for practitioners to work over state boundaries.
Recommendations from the project group meeting at VicRoads on 6 December 2010:

 Evaluate the risk assessment process in AS 5100 and how jurisdictions apply their risk
assessments (i.e. do they use AS 5100.1–2004 for this or another process?). An assessment
of how crash data analysis might influence this process is required.
 Investigate the relationship between road barriers and bridge barriers, with specific reference
to their physical transition (refer to AS 5100–2004 and AS/NZS 3845–1999).
 Investigate the issues for future consideration including high performance vehicles/changing
fleet through PBS, etc.
 As an objective of standardisation, ensure that design requirements cannot be misinterpreted
by design consultants.
2.2 Current Design Standard AS 5100

A number of points for discussion have been identified as a result of a review of AS 5100 (2007),
including:
 It is specified that the bridge barrier performance levels be determined by a relevant
authority. Such a relevant authority should be specified.
 Should there be a clause on the acceptance of barrier testing by computer simulation for
future use?
 What is an appropriate number of performance levels? Should the two special performance
levels presented in Appendix B, AS 5100.1 (2004) as informative be included in the main text
of AS 5100.1 to take into account the high incidence of heavy vehicles on the current
Australian road network?
 It is recommended that heavier utility cars (2.4 t/70 km/h/20°), local buses (9 t/75 km/h/15°),
metro buses (15 t/100 km/h/15°) and inter-city buses (22 t/90 km/h/15°) be included in
vehicle crash testing criteria.
 A detailed and quantitative site-specific risk assessment procedure should be developed,
particularly for the medium to high-risk sites.
 Benefit-cost (B/C) analysis criteria and procedure should be developed.
 Should a detailed structure analysis procedure (which is based on the AASHTO LRFD
method) be included for bridge barriers?
 For bridge approach barriers, detailed descriptions on lengths of transition segments, length
of need, typical design drawings, interfaces between barrier types and acceptable types of
end treatments should be developed. Good references include VicRoads BTN 2009/002
(VicRoads 2009b) and Department of Infrastructure, Energy and Resources (2007).
 For retrofitting of barriers on existing bridges, reference should be made to the RMS’s
BTD 2007/008 (Roads and Traffic Authority 2009) for a site-specific risk assessment
procedure.
 For bridge support protection, a procedure for risk assessment and determination of
protection level should be developed. Typical design drawings on profile and plan for typical
situations should be developed. Refer to VicRoads BTN 2009/001 (VicRoads 2009a).
Interface barriers between road and rail should be considered as specified in the
specification MCE-SR-007 (Department of Main Roads and Queensland Rail 2010).
Austroads 2013
— 9—
2.3 Computer Simulation

Computer simulation has been increasingly used to study the crashworthiness of vehicles as well
as the behaviour of highway structures during collisions. The simulation models allow researchers
to examine a variety of impact scenarios, test barrier configurations and test conditions that
typically cannot be done by full-scale crash tests due to cost constraints. The simulation results
can be used to assess deficiencies of the testing models and make adjustments to existing
roadside safety features. Therefore the design of roadside safety hardware can also be optimised
based on the simulation results.
LS-DYNA, ANSYS and ABAQUS are the typical programs utilising finite element method (FEM)
used to develop simulation models.
In the USA, computer simulation has been used as a tool to examine a range of the barrier
performance characteristics and provide the basis for improving the performance of the barrier
before the actual crash testing (Williams et al. 2007). In another case, the Engineering Simulation
System Italy (ENSIS n.d.) has used computer simulation to verify a bridge barrier design that has
some details different from that of a crash-tested barrier. This process includes two steps:
 Validate the FEM model by comparing the FEM model results to the test results. Calibration
of simulation results with test results is carried out using an iterative process.
 After the validation is done, the simulation is used to generate test parameters which are
then compared to the test results to show the conformity of the design.
A number of comparisons have been made between computer simulation models and full-scale
crash tests to validate the reliability of the computer models. Pernetti et al. (2007) successfully
carried out simulation tests for articulated trucks. They reported that the results obtained from the
simulation models agreed well with that of full-scale crash tests. As a result, this model was
validated to be used instead of full-scale crash tests.
Various existing LS-DYNA compatible models are available for free download on established
websites, such as that of the FHWA/NHTSA Crash Analysis Center and National Crash Analysis
Center. The computer models include every element involved in a crash: the vehicles, dummies
with human biomechanical components, and roadside hardware.
The rapid improvement in computer technology plays an important part in the development of the
crash simulation models. Not long ago, even simple simulations – such as those of small models
representing just a 100 millisecond event – required a day or more on a vector supercomputer to
complete. Today, using low-cost parallel computing, NCAC, for example, can perform large-scale
car-to-car simulations – complete with occupants and airbags – in just a few hours. The powerful,
integrated workstations enable researchers to view and manipulate simulation data immediately;
such a task was virtually impossible just a few years ago.
Although there have been increased validations of computer models of barrier crash tests,
computer simulation has not been accepted as an alternative for the full-scale crash tests, which is
desirable for validation purposes.
In Australia, since full-scale crash tests are unlikely to be conducted to develop new bridge barrier
designs, computer simulation can potentially be used to produce assessments to corroborate the
acceptance criteria specified in AS 5100.1 (Section 4.3.4) and especially to validate the barrier
retrofitting designs.
Austroads 2013
— 10 —
2.4 Safe System Approach

2.4.1 Safe System
The Safe System as specified by the Australian Transport Council (2011) is a road safety approach
which acknowledges that people will continue to make mistakes and that roads, vehicles and
speeds should be designed to reduce the risk of crashes and to protect people in the event of a
crash.
This is an inclusive approach that caters for all groups using the road system, including drivers,
motorcyclists, passengers, pedestrians, bicyclists, and commercial and heavy vehicle drivers.
There are several guiding principles to this approach:

 People make mistakes: humans will continue to make mistakes, and the transport system
must accommodate these. The transport system should not result in death or serious injury
as a consequence of errors on the roads.
 Human physical frailty: there are known physical limits to the amount of force bodies can
take before they are injured.
 A ‘forgiving’ road transport system: a Safe System helps to ensure that the forces in
collisions do not exceed the limits of human tolerance. Speeds must be managed so that
humans are not exposed to impact forces beyond their physical tolerance. System designers
and operators need to take into account the limits of the human body in designing and
maintaining roads, vehicles and speeds.
Four cornerstones and the strategic intent are described in Australian Transport Council (2011) as
follows:
1 Safe roads: roads and roadsides designed and maintained to reduce the risk of crashes
occurring and to lessen the severity of injury if a crash does occur. Safe roads prevent
unintended use through design and encourage safe behaviour by users.
2 Safe speeds: speed limits complementing the road environment to manage crash impact
forces to within human tolerance, and all road users complying with the speed limits.
3 Safe vehicles: vehicles which not only lessen the likelihood of a crash and protect occupants,
but also simplify the driving task and protect vulnerable users. Increasingly this will involve
vehicles that communicate with roads and other vehicles, while automating protective
systems when crash risk is elevated.
4 Safe people: encourage safe, consistent and compliant behaviour through well-informed and
educated road users. Licensing, education, road rules, enforcement and sanctions are all
part of the Safe System.
2.4.2 Contribution of Bridge Barrier Design to the Safe System

The bridge barrier system should contribute to the safe roads cornerstone. The Safe System
approach requires, in part (Australian Transport Council 2007):
 Designing, constructing and maintaining a road system (roads, vehicles and operating
requirements) so that forces on the human body generated in crashes are generally less
than those resulting in fatal or debilitating injury.
 Improving roads and roadsides to reduce the risk of crashes and minimise harm: measures
for higher-speed roads including dividing traffic, designing ‘forgiving’ roadsides, and providing
clear driver guidance. In areas with large numbers of vulnerable road users or substantial
collision risk, speed management supplemented by road and roadside treatments is a key
strategy for limiting crashes.
Austroads 2013
— 11 —
 Managing speeds, taking into account the risks on different parts of the road system.
Beside vehicle and speed-related countermeasures, road-related countermeasures consistent with

Safe System principles from jurisdictions around the world include:
 flexible wire barriers along the centre and sides of roads, to prevent run-off-road and head-on
crashes
 flexible wire barriers to separate vulnerable road user groups (pedestrians, cyclists) from
vehicles
 roundabouts, to reduce possible impact speeds and to alter crash types, and in so doing to
simplify the decisions drivers have to make.
2.5 Development of the Guidelines

The outcome of this project includes a set of guidelines for the following six topics:
 Selection of Bridge Barrier Performance Levels (Section 3)
 Bridge Barrier Design Procedure (Section 4)
 Standardised Bridge Barrier Designs (Section 5)
 Retrofitting of Existing Bridge Barriers (Section 6)
 Bridge Approach Barriers (Section 7)
 Bridge Support Protection Barriers (Section 8).
The guidelines are developed based on the following sources of information:

 provisions of AS 5100.1 and AS 5100.2 (2004)
 consultation from working group experts
 available examples from various actual projects.
The processes involved in developing these guidelines are described in the following sections.
2.5.1 Guidelines for Selection of Bridge Barrier Performance Levels (Section 3)

Number of bridge barrier performance levels
See Section 3.1.1 for a discussion of the number of bridge barrier performance levels.
Criteria for no barrier

Current criteria for no barrier specified in AS 5100.1 are as follows:
Traffic barriers may be omitted where all of the following conditions apply:
1 The bridge is less than 1.5 m above the ground.
2 Traffic volumes are less than 150 vehicles per day.
3 The radius of curvature of the bridge is such that the road approaches have a sight distance
greater than the stopping distance.
4 The width between kerbs is not less than 6.5 m for a two-lane bridge or 4.0 m for a
single-lane bridge.
5 The edge of the bridge is at least 1.0 m from the edge of the traffic lanes.
6 The location is without anticipated pedestrian traffic.
Austroads 2013
— 12 —
7 Any water beneath the bridge is normally less than 1.2 m deep.
8 The provision of barriers prevents the passage of debris or the barriers would be frequently
damaged by heavy debris or both.
Comments from the working group showed that:

 From points (d) and (e), 1.0 m edge distance means that the width of the traffic lane is only
2.0 m for single-lane bridges and 4.5 m for two-lane bridges. Nominally, traffic lanes are
3.0 m minimum. Hence, the code is not consistent.
 Point (h) is not relevant in the context of all other items: If all items except the last one
conform for the no barrier situation, then the last item does not make sense. The last item
needs to be in a separate paragraph.
These criteria refer back to the criteria for Level 4 barriers in the superseded 1992 Austroads
Bridge Design Code. They imply that the no barrier case is only considered for sites with flooding
debris problems and have favourable site conditions.
Consider the original points (d) and (e) and assume that the kerb has a minimum width of 0.25 m
and the lane width is 3.2 m. If condition (e) applies then the width between kerbs will not be less
than 2 x 3.2 + 2 x (1 – 0.25) = 7.9 m for a two-lane bridge or 3.2 + 2 x (1 – 0.25) = 4.7 m for a
single-lane bridge. In addition, the original condition (e) is covered in the revised condition (d). It
should be noted that kerbs are required for the no barrier case.
The following criteria for no barrier are suggested:
Traffic barriers may be omitted where the provision of barriers prevents the passage of debris or
the barriers would be frequently damaged by heavy debris or both, and all the following criteria
apply:
1 The bridge or culvert is less than 1.5 m above the ground.
3 Bridges with an essentially straight alignment (e.g. with a radius of horizontal curvature of
greater than 1500 m) and the road approaches have a sight distance greater than the
stopping distance.
single-lane bridge.
Introduction of site categories CAT1, CAT2 and CAT3

The introduction of a site condition category is in addition to the current requirements of
AS 5100.1 (2004) but has been included in the methodology with the acknowledgement that there
are three different methodologies used in this design standard. The introduction of a site condition
category results is a formal method to select the appropriate methodology and give guidance on
when a particular methodology should be selected. Effectively the site categories are equivalent to
the following terminologies:
 CAT1: subjective criteria
 CAT2: site risk cost criteria algorithm
 CAT3: empirical and B/C consideration based.
Austroads 2013
— 13 —
For a specific site, it is easy to determine the site category based on its initial site condition. The
subsequent steps are obvious, based on the methodology associated with the selected site
categorisation.
Specification standards
The criteria set out in this guide provide a framework for the input from jurisdictions. The criteria
need to be as specific as possible to provide effective guidance, but generic enough to allow the
road authority discretion as appropriate.
Typical case scenarios are developed based on experiences from jurisdictions. An appropriate
standard of barrier is nominated for each scenario. Such a system of case scenario and associated
barrier performance level would mean less reliance on traffic parameters (e.g. AADT number) and
B/C analysis.
Examples of typical case scenarios for critical locations may include (Austroads 2010a):
 a heavy vehicle falling from a bridge or embankment onto a passenger railway line or onto a
major road, causing multiple fatalities in many vehicles plus significant societal disruption
 a heavy vehicle impacting a water or steam pipeline, electrical transmission tower, or the like
where major disruption to supply, or other consequential damage, such as fire or
contamination could occur
 a heavy vehicle impacting the supports of a structure, such as a bridge or pedestrian
overpass, causing the structure to collapse onto the road. This would be catastrophic if there
were people on the structure or if the collapse impacted vehicles or pedestrians on the road
and/or caused long-term transport disruption
 a high-occupancy vehicle, such as a bus, falling into deep water from a bridge or
embankment or over a drop of sufficient height, with consequential fatalities to the
passengers in the vehicle
 a heavy vehicle leaving the road at a curve and impacting a community, commercial or
residential building (e.g. a school or playground located beside the road). This would be
catastrophic if there were many people in or around the building
 roads passing over the railway main control room.
More case scenarios need to be obtained from jurisdictions to complete this system.
Benefit-cost analysis
It was advised by the working group that a B/C analysis should be avoided, or conducted by
experts who then provide a limited number of standards within the code for the user to choose
from, or needs to be well defined, simple and quick to undertake.
In this report, it is proposed that a B/C analysis is optional, which is based on an incremental B/C
approach. Once it is identified that a B/C analysis is needed, a comparison between all
performance levels can be conducted.
By introducing the two limits for the performance level specifications, ‘minimum’ and ‘preferable’,
the adoption of preferable limits subject to an incremental B/C analysis will at least incentivise B/C
analysis and potentially penalise those who do not do the B/C analysis.
Austroads 2013
— 14 —
The following points have been identified for discussion:

 jurisdictions’ agreement on the use of site categories and number of barrier performance
levels is required
 criteria for site conditions and specification standards for CAT3 method
— jurisdictions’ review and agreement on the proposed numbers
— a supplementary of additional criteria.
Refer to Appendix E.1 for the outcome of the discussion.
2.5.2 Bridge Barrier Design Procedure (Section 4)

Current AS 5100 (2007) does not provide specific provisions for structural design of bridge
barriers. Current practice shows that consultants/contractors are using the AASHTO LRFD
methodology, which is based on the yield line theory.
According to AS 5100.2 Supplementary, design forces specified for High and Special performance
level barriers are taken respectively as approximately 50% and 100% greater than those specified
for Medium performance level barriers to provide clearly recognisable steps in containment
capacity over those defined for lower performance levels. These may not correctly reflect the
current heavy vehicle data. Therefore design forces should be revised.
The design procedures were developed for the three most popular bridge barrier types: concrete
parapet, steel rail-and-post, and combined barriers.
The methodology for the analysis of the bridge deck slab under the barrier impact loads is also
provided, however, the bridge deck slab should be designed based on specific requirements of the
specific bridge structure rather than being standardised.
Excel spread sheets have been developed based on these procedures and can be used in the
development of standard barrier designs.
It is suggested that the forces (moment, shear, normal force) occurring in the bridge deck slab due
to design impact loads be determined for each bridge barrier type and performance level. These
forces will be used to verify the strength of the bridge deck slab.
Regarding the capacity reduction factor, φ:

 In AASHTO LRFD, impacts of vehicles colliding with barriers are considered in an extreme
event limit state; therefore the load factor and the capacity reduction factor is 1.0.
 In AS 5100 (2007), ultimate limit states cover the extreme limit state. The capacity reduction
factor shall be used, e.g. φ = 0.8 for concrete and φ = 0.9 for steel (in bending). The load
factor for designing bridge barriers is 1.0.
 Current practice in Australia uses the AASHTO method in combination with AS 5100 (2007)
φ factors in designing bridge barriers. As a result, the barrier designed in Australia is
overdesigned (10% to 20% stronger than the US design). This also increases the capacity
requirement of the bridge deck slab.
 A capacity reduction factor φ = 1.0 was proposed.
Austroads 2013
— 15 —
2.5.3 Standardised Bridge Barrier Designs (Section 5)

This section discusses the development of standardised bridge barrier design systems. The
process involves the following steps:
 collect standard bridge barrier designs specified by various jurisdictions and bridge barrier
design examples used in recent actual projects
 select the most suitable designs for each performance level or application. The selection is
based on the popularity of the design and the review from working group
 carry out structural verification of the above selected designs and recommend the
modification to the detailed design if necessary.
It should be noted that this section will not try to invent new designs or reproduce detailed
drawings. The outcomes of this section are a set of barrier design types with detailed descriptions
and associated design information. Jurisdiction may develop detailed design drawings based on
this outcome.
Although the selected bridge barrier designs may be accepted as per acceptance criteria specified
in AS 5100.1 (2004), the conformity of these designs to the NCHRP Report 350 or MASH
(AASHTO 2009) is questionable and will only be corroborated by physical crash tests.
Based on a decision of the BTTF, the detailed design of selected barrier types would be
undertaken only once the revision of AS 5100 has been completed and published.
The points noted in Table 2.2 and below have been identified for discussion.
Table 2.2: Proposed standard barrier types

Barrier type/performance level Low Regular Medium High Special
Concrete barrier N/A RL1, RL2 ML1, ML2 HL1, HL2 SL1, SL2, SL3
Rail-and-post barrier LL1, LL2 RL3, RL4 ML3 N/A N/A
Combination barrier N/A RL5, RL6 ML4, ML5, ML6 HL3, HL4 SL4
Note: Refer to Section 5 for definitions of barrier types.
 Concrete barriers: single-slope and F-shape are preferred. This type of barrier should be
developed for Regular, Medium, High, and Special performance levels.
 Rail-and-post barriers: should be developed only for Low and Regular performance levels.
Different barrier configurations may include: one rail, two rails and three rails with steel posts.
 Combination barriers: should be developed only for Regular, Medium and High performance
levels. Different barrier configurations may include: concrete parapet of different heights
combined with one rail or two rail barriers with steel posts.
 For the Special performance level, concrete barrier and combination barrier should be
developed.
 For a concrete parapet, two typical shapes are preferred, F-shape and single slope. Different
performance level barriers may have the same shape; however, the barrier height, thickness
and reinforcement details may be different.
 Barrier designs with unique architectural forms should be treated on a project-specific basis.
Austroads 2013
— 16 —
2.5.4 Guidelines for Bridge Barrier Retrofitting (Section 6)

Retrofitting design methodology
The methodology generally involves determining the required performance level for the site,
evaluating the structural capacity of the existing bridge, and determining the most feasible
retrofitting option. The selected barrier retrofitting option will largely depend on the required
performance level and the available capacity of the existing structure. Consideration is given to a
lower than required performance level barrier with some superstructure strengthening options.
Reasons for retrofitting existing bridge barriers

Generally, retrofitting of bridge barriers will be considered when the bridge is to be
widened/upgraded, or when there have been excessive on-going vehicle collision costs and safety
risks at the bridge site and the existing barrier is substandard.
Evaluating existing bridge structure

It is essential that the existing bridge structure be evaluated to make sure that the bridge can
accommodate the design loads associated with the new barrier.
In some circumstances where the capacity of an existing bridge structure is not sufficient then the
strengthening solutions should be determined.
Determination of appropriate barrier performance level

If the structural capacity of the existing bridge is not sufficient to accommodate new loads
associated with the required barrier performance level, site-specific risk assessment and B/C
analysis should be carried out to determine the optimal option. The outcome of this process may
be one of the following:
 retrofitting the existing barriers to the required performance level combined with structural
strengthening of the existing bridge
 retrofitting the existing barriers to a lower performance level than the required performance
level combined with structural strengthening of existing bridge and accepting some level of
risk remaining at the site
 if no retrofitting can be done to the existing barrier, some safety measures should be carried
out at the site to reduce the risk.
Retrofitting design procedure

The retrofitting design may include upgrading or replacing existing barriers.
For the replacing option, the design of the new barrier can be taken from the standard barrier
design systems for the selected performance level. Detailed design will include anchorage details
and structural strengthening if necessary.
The upgrading design is based on the actual retrofitting projects. The acceptance of the upgraded
barrier, however, will only be based on the theoretical structural verification and practical
performance.

 It is unlikely that a set of standard designs for retrofitting existing bridge barriers will be
developed, as the designs are usually developed based on a project-specific basis.
Austroads 2013
— 17 —
 The barrier retrofitting examples presented in Section 6.5 are not structurally verified to the
associated performance level.
 The methodology to assess the conformity of the structurally upgraded barrier to the current
standard should be developed.
 Bridge barrier retrofitting examples from various jurisdictions should be provided. Examples
should show the initial barrier, old performance level, new performance level, ideal
performance level, justifications for design processes etc.
2.5.5 Guidelines for Bridge Approach Barriers (Section 7)

 There are two methods for the determination of the length of need, run-out length and angle
of departure. It is suggested that both methods be used and the worst case scenario be
applied.
 Future work: guidelines for designing the off-structure foundation (on the bridge approach)
should be developed, which include traffic barriers along the edges of high reinforced soil
walls.
2.5.6 Guidelines for Bridge Support Protection Barriers (Section 8)

 For collisions from road traffic, AS 5100.2 (2004) requires that the bridge supports be
designed for a minimum equivalent load of 2000 kN applied at an angle of 10° from the
direction of the road centre line and at 1.2 m above the ground level, if they are not located
behind appropriate protective traffic barriers.
 It is suggested that the recent update of AASHTO LRFD Bridge design specifications
(AASHTO 2012) be incorporated, which specifies that the equivalent static load is 2670 kN
acting in a direction of zero to 15° with the edge of the pavement in a horizontal plane, at a
distance of 1.5 m above ground.
 If a rigid barrier is provided, the barrier should be designed to the associated design loads of
the barrier performance level. The barrier should not be designed to the equivalent static
loads specified to act on the bridge support.
 Criteria for the determination of the performance level of bridge support protection barriers in
Section 8.4.3 should be reviewed and agreed by jurisdictions.
 Queensland Rail technical guideline Collision protection of supporting elements adjacent to
railways (Queensland Rail 2010) is suggested for collision protection of supporting elements
adjacent to railways.
Austroads 2013
— 18 —
3 SELECTION OF BRIDGE BARRIER PERFORMANCE

LEVELS
Current practice shows that most Australian road authorities determine the required bridge barrier
performance levels based on a risk assessment procedure specified in the Bridge Design Standard
(AS 5100.1–2004) while the New Zealand Transport Agency (NZTA) uses a similar procedure
specified in its Bridge manual (Transit New Zealand 2004). In this procedure, the charts are
provided for normal site conditions; however, for site-specific, medium to high risk situations, only
descriptive advice is available. As a consequence, several jurisdictions have developed their own
guidelines which lead to variations in undertaking barrier risk assessments and hence may result in
different barrier performance levels for similar situations.
This guideline aims at providing a generic procedure for undertaking risk assessment for
determining bridge barrier performance levels, which can be used consistently throughout
jurisdictions. It should be noted that this section only focuses on the selection of the performance
level instead of the selection of the bridge barrier type or design. After the barrier performance
level is selected, it will be used as the basis for determining the bridge barrier performance
specifications, type and details.
This guideline includes:

 a methodology and flowchart for selection of the barrier performance level
 a process to determine site condition categories
 a chart-based procedure
 minimum specifications for higher performance levels based on risk assessment and B/C
analysis
 several examples.
3.1 General Procedure

The general procedure is to determine a site condition category which then determines the
methodology to be used to select a site-specific barrier performance level.
3.1.1 Bridge Barrier Performance Levels

AS 5100.1 (2004) specifies Low, Regular, Medium and Special performance levels with a
possibility of including a ‘no barrier’ case and Special level. The Special level was divided into
three sub-levels (36 t tanker, 44 t articulated van, and as specified by the road authority). This
system relates to the NCHRP Report 350 test level system (Ross et al. 1993), in which TL-1 and
TL-3 are not used while a TL-6 is added.
Jurisdictions in general follow this specification, however they introduced some departures: NZTA
uses no barrier, TL-3, TL-4, TL-5, TL-6 and Special level; VicRoads specifies no barrier, Low,
Regular, Medium, High level and Special level; RMS uses no barrier, Low, Regular, Medium,
Special I, Special II and Special III levels and two additional levels including Basic and
Intermediate for barrier retrofitting purposes.
It is suggested in this guide that five performance levels (Low, Regular, Medium, High, and
Special) be used together with the no barrier case. The Special III performance level should be
omitted. This approach ensures AS 5100 conformity and also follows most jurisdictions’ best
practice.
Austroads 2013
— 19 —
Table 3.1 presents the general containment requirements and the equivalent test levels of NCHRP
Report 350 (Ross et al. 1993) for these performance level barriers (Main Roads Western Australia
2009 and VicRoads 2001).
Table 3.1: Bridge barrier performance levels

Barrier NCHRP 350 Containment properties Test vehicle
performance test level
level
Provided for the containment of light vehicles. Generally to be Required to contain a level of impact
used for short, low-level structures on rural roads and areas typical of a 2.0 tonne utility vehicle (light
Low TL-2
where a very small number of mixed heavy vehicles are vehicle) at 70 km/h and an impact angle
expected and for a low speed environment. of 25°.
Provided for the containment of cars, heavy utilities and light Required to contain a level of impact
trucks. Generally to be used on high speed main roads and typical of an 8.0 tonne truck at 80 km/h
Regular TL-4
highways with a mixture of heavy vehicles. This barrier is and an impact angle of 15°.
applicable and appropriate to the majority of bridge sites.
Provided for the containment of most buses and medium mass Required to contain a level of impact
vehicles. Generally to be used for high speed freeways, arterial typical of a 36.0 tonne articulated van at
Medium TL-5
main roads and major highways with a high volume of mixed 80 km/h and an impact angle of 15°.
heavy vehicles and site-specific risk situations.
Provided for high risk situations and the containment of heavy Required to contain a level of impact
high centre-of-gravity vehicles. Generally to be used on routes typical of a 36.0 tonne tanker-type vehicle
High TL-6 with a higher volume of mixed heavy vehicles and maximum (high centre-of-gravity and rigid vehicle)
tolerable speeds such as freeways with variable cross slopes at 80 km/h and an impact angle of 15°.
and reduced radius of curvature.
Provided for site-specific, unusual conditions and critical Required to contain a level of impact
locations where penetration or vaulting by very high typical of a 44.0 tonne articulated van at
Special Above TL-6 centre-of-gravity and/or heavy vehicles under varying impact 100 km/h and an impact angle of 15°.
conditions must be prevented. There is no barrier rated higher than TL-6
in NCHRP 350.
3.1.2 Site Condition Categories

The introduction of a site condition category is in addition to the current requirements of
AS 5100.1 (2004) but has been included in the methodology with the acknowledgement that there
are three different methodologies used in the current standard. The introduction of a site condition
category results in a formal method to select the appropriate methodology and give guidance on
when a particular methodology should be selected.
The site conditions are classified into three categories as follows:

 CAT1: for sites which satisfy all criteria for the no barrier case in Section 3.2.
 CAT2: for sites with low to medium risk levels which satisfy the criteria specified in
Section 3.3.
 CAT3: for sites with medium to high risk levels and/or that have special conditions as
specified in Section 3.4.
Site condition category CAT1 is determined by conformance to a set of prescribed, ‘deemed to

comply’ requirements listed in Section 3.2. If all requirements are met, no barrier can be specified.
All requirements must be met otherwise a higher category must be considered. It needs to be
emphasised that where possible, a barrier should be considered and that this guide is specifying
minimum requirements. Special situations may exist where no barrier can be justified due to issues
with flood integration/strategy. In these situations the jurisdiction must do a comprehensive risk
study including a B/C analysis.
Austroads 2013
— 20 —
Site condition category CAT3 is selected on the basis of the site-specific characteristics described
in Section 3.4. In this case, a barrier performance level is selected on the basis of minimum
performance requirements and a B/C analysis.
Site condition category CAT2 is the default category if CAT1 and CAT3 are not relevant. In this
case, the chart-based procedure can be used. In addition, all risks listed in the site condition CAT3
must not be relevant to select this site condition category.
A flowchart for the selection of barrier performance levels is shown in Figure 3.1.
For the purpose of this section, high risk situations include:

1 bridges over major roadways with an annual average daily traffic (AADT) of 10 000 vpd per
lane, or over roads with an AADT of 40 000 vehicles or more per day
2 bridges over high frequency passenger rail lines with a peak intensity of more than six trains
in an hour in each direction or goods lines carrying noxious, flammable or large volumes of
freight, or critical railway infrastructure as specified by the relevant rail authority
3 bridges over high occupancy land use such as houses, factories, areas for congregating, etc.
3.2 Criteria for Site Condition CAT1

Traffic barriers may be omitted where the provision of barriers prevents the passage of debris or
the barriers would be frequently damaged by heavy debris or both, and all the following criteria for
site condition CAT1 apply:
1 The bridge or culvert is less than 1.5 m above the ground.
3 Bridges with an essentially straight alignment (e.g. with a radius of horizontal curvature of
greater than 1500 m) and the road approaches have a sight distance greater than the
stopping distance.
single-lane bridge.
Austroads 2013
— 21 —
Figure 3.1: Barrier performance level flowchart

A site condition CAT2 applies when any of the criteria specified for CAT1 and CAT3 does not hold.
For this site condition category, a Low, Regular or Medium performance level barrier may be
selected by undertaking a chart-based risk assessment procedure described in Section 3.6.

A site shall be considered to be in a CAT3 site condition where any of the following criteria apply:
1 site-specific, unusual conditions and critical locations as specified by the relevant road
jurisdiction
2 locations where it is essential that penetration or vaulting by vehicles specified by the
jurisdiction under impact conditions needs to be prevented
3 bridges on roads with special classes of heavy vehicle such as high mass, high
centre-of-gravity vehicles
Austroads 2013
— 22 —
4 roads with a commercial traffic level of greater than or equal to 3000 commercial vehicles per
day per carriageway and are in a high risk situation
5 roads with a volume of buses greater than or equal to 150 buses per day per carriageway in
a high risk situation or in any of the following situations
(a) bridges more than 10 m high
(b) bridge over water more than 3 m deep (normal flow)
(c) bridges on horizontal curves with a radius of 600 m or less.
3.5 CAT1 Method

No barrier shall be provided if the site conditions meet all the criteria listed in Section 3.2 for site
condition CAT1.
3.6 CAT2 Method – Chart-based Procedure

This procedure is used to select a barrier performance level between Low, Regular and Medium for
a CAT2 site condition using site-specific traffic data and road environment details such as road
grade, horizontal curvature and under-structure conditions.
The detailed steps of the chart-based procedure are straightforward as described in Appendix B of
AS 5100.1 (2004) and presented in Figure 3.2.
Detailed steps are described below.
3.6.1 Determine AADT and Portion of Commercial Vehicles

For the chart-based method, it is specified in AS 5100.1 (2004) that the AADT is taken as the total
estimated average annual daily traffic for the first year after construction on all lanes of the bridge.
The projected traffic growth assumed is 2% per annum for 30 years. For sites with a design speed
of 80 km/h or greater and a construction year AADT greater than 10 000 vehicles per day per lane
(vpdpl), the construction year AADT may be limited to 10 000 vpdpl, in order to account for the
effect of traffic congestion on traffic speeds.
It should be noted that for growth rates other than 2% per annum, the construction year AADT for
use in this procedure can be adjusted by dividing the 30-year after construction AADT by
(1 + 2%)30 = 1.81. The error using this estimation is acceptable and within the assumption of this
methodology (Transit New Zealand 2004).
The portion of commercial vehicles is usually obtained from the predicted traffic data derived from
the project’s environmental impact statement (EIS).
Austroads 2013
— 23 —
3.6.2 Determine Adjusted AADT

The adjusted AADT is calculated by the following formula:
Figure 3.2: Chart-based procedure
Adjusted AADT = AADT x RT x GD x CU x US 1
where
RT = road type adjustment factor, is derived from Table 3.2

(Table B1, AS 5100.1–2004)
GD = road grade adjustment factor, is taken from Figure 3.3

(Figure B2, AS 5100.1–2004)
CU = road curvature adjustment factor, is taken from Figure 3.4

(Figure B3, AS 5100.1–2004)
US = adjustment factor for deck height and under-structure conditions, is taken from
Figure 3.5 (Amendment No. 1, AS 5100.1–2010).
Austroads 2013
— 24 —
Table 3.2: Road type factor

Road type AADT Number of lanes* Adjustment factor RT
One way** Based on one-way traffic 1 or more 2.0
Two-way divided Based on two-way traffic 2 or more 1.0
Two-way undivided Based on two-way traffic 1 to 4 1.5
Two-way undivided Based on two-way traffic 5 or more 1.0
* The number of lanes is the total number of lanes on the bridge.
** Includes one-way roads, freeway ramps and bridges on separated carriageways of freeways and highways.
Source: AS 5100.1 (2004).
Figure 3.3: Grade factor GD
Austroads 2013
— 25 —
Figure 3.4: Curvature factor CU
Figure 3.5: Under-bridge condition factor US
Austroads 2013
— 26 —
3.6.3 Determine Barrier Performance Level

The barrier performance level will be determined from Figure 3.6 to Figure 3.9 (AS 5100.1–2004),
based on design speed, adjusted AADT, portion of commercial vehicles and rail offset.
Figure 3.6: Threshold limits 60 km/h
Austroads 2013
— 27 —
Austroads 2013
— 28 —
3.6.4 Example 1
A bridge site in a low-risk situation has the following traffic data and road environment details:
 AADT 12 000 vpd at the census year 2008 (two-way traffic)
 construction year 2012
 growth rate 3% per annum
 portion of commercial vehicle 17% with no high mass (vehicle class ≥ class 7) and high
centre-of-gravity vehicles
 design speed 110 km/h
 two-way undivided road
 number of lanes 2
 lane width 3.5 m
 offset from lane to barrier 1.2 m
 alignment 700 m radius horizontal curve
 road grade in the direction of traffic –2.5%
 the bridge deck height 10 m
 under-bridge condition: low occupancy land use.
Determine site condition category

The AADT with adjustment of the growth rate 3% in four years from 2008 to 2012 is calculated as:
AADT = 12 000 x (1 + 0.03)4 = 13 506 vpd
Thus, the volume of commercial vehicles at construction year is 17% x 13 506 = 2296 vpd.
The following were derived from the traffic data and environmental details:
 There are no high mass and high centre-of-gravity vehicles or special conditions that require
a Special performance level.
 The volume of commercial vehicles is less than 3000 vpd, which does not require a High
performance level to be considered.
 The site is not a medium to high risk situation, which does not require a B/C analysis.
From this analysis, it is obvious that the site condition does not belong to either the CAT3 or CAT1
site conditions, thus a CAT2 site condition is selected. As a result, a chart-based risk assessment
procedure is used to select a performance level.
Chart-based procedure
The following calculations are performed:
 Road type factor: for a two-way undivided road with two traffic lanes, the adjustment factor
for road type is RT = 1.5 (Table 3.2).
 Road grade factor: according to Figure 3.3, based on a downgrade of –2.5%, the road grade
factor is GD = 1.13.
 Curvature factor: with a radius of curvature of 700 m, a curvature factor CU = 1.25 is derived
from Figure 3.4.
 Adjustment factor for deck height and under-bridge condition: a factor of US = 1.2 is derived
from Figure 3.5 with low occupancy land use and deck height of 10 m.
Austroads 2013
— 29 —
 Adjusted AADT (Equation 1):

Adjusted AADT = 13 506 x 1.5 x 1.13 x 1.25 x 1.2 = 34 339 vpd.
With the design speed of 110 km/h, Figure 3.9 is used to determine the barrier performance level.
In Figure 3.10, for an adjusted AADT of 34 339 vpd and 17% commercial vehicles, the
corresponding point belongs to the Medium level region formed by the 1.2 m barrier offset lines,
thus the selected barrier level is a Medium performance level.
34.3
17
Figure 3.10: Example with threshold limits 110 km/h
3.7 CAT3 Method – Specification Standards and B/C Analysis

There are two standards for each performance level, ‘minimum’ and ‘preferable’. While the
minimum standard requires that a barrier associated with that performance level must be provided,
the preferable standard requires a B/C justification to investigate the possibility of using a lower
performance level barrier.
The following are deemed to be minimum specifications for the selection of a barrier performance
level within a CAT3 site condition. If all conditions are not met, the next highest level must be
considered.
3.7.1 Specification Standards for Medium Performance Level

A Medium performance level barrier may be provided at specific locations where agreed by the
relevant road jurisdiction for the containment of buses and medium mass vehicles on high speed
carriageways, major carriageways, and urban roads with a medium to high level of mixed heavy
vehicles, and site-specific risk situations.
Minimum standard
Medium performance level barriers shall be provided where the site belongs to a CAT3 site
condition and does not satisfy the criteria for a High performance level barrier.
Preferable standard
There are no preferable standards for Medium performance levels.
Austroads 2013
— 30 —
3.7.2 Specification Standards for High Performance Level

Where there is a high probability of loss of life or serious injuries due to a vehicle penetrating the
barrier, a High performance level barrier shall be provided at specific locations agreed to by the
relevant road jurisdiction.
Minimum standard
High performance level barriers shall be provided where the following condition applies:
1 Bridges with heavy, high centre-of-gravity vehicles on high speed freeways, major highways
and urban arterial roads with a medium to high volume of commercial vehicles (between 15%
and 30% AADT) in a high risk situation.
Preferable standard
High performance level barriers shall be provided, subject to an appropriate B/C justification, for
bridges with heavy, high centre-of-gravity vehicles on high speed freeways, major highways and
urban arterial roads with a medium to high volume of commercial vehicles (between 15% and
30% AADT) not in a high risk situation or roads with a high volume of commercial vehicles (greater
than or equal to 30% AADT) in any of the following conditions:
1 bridges more than 10 metres high (deck to ground)
2 bridges over water greater than 3 metres deep (average water level)
3 bridges on horizontal curves with a radius of 600 metres or less.
3.7.3 Specification Standards for Special Performance Level

Special performance level, non-penetrable barriers shall be provided at specific locations agreed
by the relevant road jurisdiction where vaulting by high mass and high centre-of-gravity vehicles
must be prevented.
Minimum standard
Special performance level barriers shall be provided where any of the following conditions applies:
1 roads pass over the railway main control room
2 bridges with heavy, high centre-of-gravity vehicles on high speed freeways, major highways
and urban arterial roads with a high volume of commercial vehicles (greater than or equal to
30% AADT) in a high risk situation.
Preferred standard
Special performance level barriers shall be provided, subject to an appropriate B/C justification, for
bridges with heavy, high centre-of-gravity vehicles on high speed freeways, major highways and
urban arterial roads with a high volume of commercial vehicles (greater than or equal to
30% AADT) not in a high risk situation.
In addition to the above minimum and preferable standards, the following particular hazards where
the consequences of a heavy vehicle running off the road would be catastrophic may trigger the
use of a Special performance level barrier (Austroads 2010a):
 a heavy vehicle falling from a bridge or embankment onto a passenger railway line or onto a
major road, causing multiple fatalities in many vehicles plus significant societal disruption
 a heavy vehicle impacting a water or steam pipeline, electrical transmission tower, or the like
where major disruption to supply, or other consequential damage, such as fire or
contamination could occur
Austroads 2013
— 31 —
 a heavy vehicle impacting the supports of a structure, such as a bridge or pedestrian

overpass, causing the structure to collapse onto the road. This would be catastrophic if there
were people on the structure or if the collapse impacted vehicles or pedestrians on the road
and/or caused long-term transport disruption
 a high occupancy vehicle, such as a bus, falling into deep water from a bridge or
embankment or over a drop of sufficient height, resulting in casualties to the passengers in
the vehicle
 a heavy vehicle leaving the road on a curve and impacting a community, commercial or
residential building (e.g. a school or playground located beside the road). This would be
catastrophic if there were significant numbers of people in or around the building.
3.7.4 Benefit-cost Analysis

In addition to the minimum specifications, a B/C analysis should be considered to determine the
most economic performance level. The performance level is selected from the highest level
determined by minimum specifications and B/C analysis.
A quantitative evaluation includes an assessment of the risk associated with hazards and
computation of an annual cost of crashes. The same method can be used to analyse the risk and
determine annual crash costs associated with treatment options. This information can then be used
together with installation, construction and maintenance costs to undertake B/C analysis.
B/C analysis is a method that estimates the benefits derived from a specific course of action
compared to the costs of implementing that action. If the estimated benefit of a safer design option
exceeds the cost of constructing and maintaining that design over a period of time, the safer
design should be implemented subject to project funding constraints.
The primary benefit obtained from selecting one design over another is the expected reduction in
future crash costs. These typically included property damage costs and personal injury costs. In
some cases, the total number of crashes may be reduced by a given treatment, such as providing
a significantly wider roadside recovery area than previously existed. In other instances, the safety
treatment may not reduce the total number of crashes but may reduce their severity
(e.g. installation of a median barrier).
A B/C analysis should consider the period of time (project life) over which each alternative
treatment provides a benefit. Since different treatments can have different project lives, both
benefits and costs must be annualised so that direct comparisons between alternative treatments
can be made. To reduce total costs to annualised costs, discount rates must be considered. An
annualised B/C ratio (BCR) thus compares the expected savings to society (through reduced crash
costs) to the costs (construction and maintenance) incurred to provide a specific treatment.
Available quantitative assessment methods
There are six methods that may be used to undertake a quantitative assessment of the risk
associated with a hazard. Jurisdictions in Australia and New Zealand may use one or several of
these methods depending on the complexity of the hazards. The methods available are:
 simple manual method
 detailed quantitative manual method
 Queensland Department of Main Roads (now Queensland Department of Transport and
Main Roads – TMR) Roadside Impact Severity Calculator (RISC – Department of Main
Roads 2005)
 AASHTO Roadside safety analysis program (RSAP – AASHTO 2006)
Austroads 2013
— 32 —
 RTA 1 method (Roads and Traffic Authority 2008a)

 ARRB Road safety risk manager
(http://www.arrb.com.au/Safe-Systems/Tools-for-improving-road-safety.aspx).
These methods are briefly described as follows.
1 Simple manual method.
This method involves (Figure 3.11):
— an assessment of the likelihood that a vehicle will crash into an object or feature and
the severity of such crashes
— an assessment of the number of crashes that would occur with a treatment and the
severity of those crashes
— a comparison of the number and severity of crashes that would occur at an untreated
hazard with the number and severity of crashes if a treatment was provided (e.g. road
safety barrier).
Source: Based on Department of Main Roads (2005).
Figure 3.11: Simple process for risk assessment
This method is usually used for hazard identification in a risk assessment process.
2 Detailed quantitative manual method
The detailed manual method is based on a risk assessment and economic analysis. The
process involves:
— from crash data for an existing situation, determine the annual crash cost for the
hazard
— determine options for treatment of the hazard
— estimate costs associated with each treatment option including crash costs,
construction costs, annual maintenance costs and operating costs where applicable
— undertake a B/C analysis of the whole-of-life costs for the existing situation (i.e. the
untreated hazard) and of all treatment options identified for evaluation and
consideration
1
Roads and Traffic Authority (RTA) NSW became Roads and Maritime Services (RMS) NSW during the course of this
project.
Austroads 2013
— 33 —
— use the results of the B/C analysis in conjunction with any qualitative analysis to
establish which treatment was recommended.
For bridge barriers, treatment options may include barriers of different performance levels.
The assessment will compare the BCRs when providing different performance level barriers
for the bridge site, based on the specific site conditions.
3 TMR RISC
RISC is a program developed by the TMR based on the AASHTO software ROADSIDE and
is used to perform quantitative evaluation of hazardous roadside objects (Department of
Main Roads 2005).
RISC requires the user to model roadside objects and potential treatments for these objects
using an array of numerical parameters. Once this is done the relative benefits and costs for
different treatments are automatically calculated using an algorithm based on the AASHTO
Roadside Design Guide (AASHTO 1996). The most cost-effective treatment for each hazard
can be determined and the decision-making process can then continue to the next step. The
program operates through a series of windows and menus.
The modelling can be used to determine the possible BCRs achievable by comparing the
treatment options available. For example, a comparison can be made between leaving an
end-on culvert as it is, installing bar grates, redesigning the end wall to reduce its severity,
and the installation of a road safety barrier.
It should be noted that TMR does not use this method specifically for bridge barriers,
however, the principles of the B/C analysis method used in this program may be utilised.
4 AASHTO RSAP
RSAP software is described in Appendix A of the Roadside Design Guide (AASHTO 2006)
and represents one approach to using this guide. RSAP provides an improved
computer-based cost-effectiveness analysis procedure for use in:
— assessing alternative roadside safety treatments at both point locations and for
sections of roadway
— developing warrants and guidelines including those which consider performance levels
of safety features.
RSAP provides a simple and structured means for data entry and four separate reports
summarising the analysis results and input data, namely:
— BCR – presents the incremental B/C ratios associated with the alternatives in a tabular
format for all combinations of alternatives.
— Alternative cost – presents the predicted crash frequencies, and the annual installation,
maintenance, and repair costs associated with each of the alternatives in a tabular
format.
— Feature cost – presents the predicted impact frequencies, average severity, and crash
costs associated with individual features of each alternative in a tabular format.
— Input data – presents the input data for each alternative in a summary form.
Austroads 2013
— 34 —
5 RMS method
This method calculates the risk associated with a hazard and compares it to an intervention
benchmark for the particular type of road. The process includes the following steps (Roads
and Traffic Authority 2008a):
— Determine the crash consequences based on crash severity indices.
— Determine the combined risk of injury and fatality based on the probability of injury and
fatality.
— The combined risk is then compared to a benchmark for the type of road to determine
the required action. If the combined risk is below the benchmark, consideration is given
to the consequences of leaving the hazard as it is, and if it is greater than or equal to
the benchmark treatment options are considered.
6 ARRB RSRM
RSRM has been developed to provide road safety professionals with a tool to proactively
assess road safety hazards and treatments for the purpose of prioritising actions. The tool
adopts a risk management approach, with the ultimate aim of maximising the risk reduction
on the road network for a given budget. RSRM provides a method to evaluate treatments
and assist designers in making optimal investment decisions. It enables relative risks to be
examined for different treatments at a site, including those associated with proposals to
provide road safety barriers.
RSRM is specifically focussed on the prioritisation of appropriate treatments. It is not a
replacement for sound engineering judgement. The method calculates a risk reduction to
cost ratio (discounted risk reduction/discounted costs) and uses this as the basis for
prioritisation of treatments.
Proposed B/C analysis method

The quantitative evaluation processes used by various road jurisdictions and computer programs
are similar but the detail of some parameters may vary to suit the requirements of the jurisdiction.
The methods involve a hazard risk assessment and a typical process should include:
 Step 1 – calculation of the frequency of errant vehicle crashes
 Step 2 – determination of the severity of the crashes with the hazard/s
 Step 3 – comparison of the results for a hazard with those for a treatment.
The information presented below makes use of the principles of the evaluation procedure used by
the RISC software (Department of Main Roads 2005) to describe the methodology for calculating
BCRs.
Austroads 2013
— 35 —
The BCR is defined as the net present benefit (NPB) divided by the net present cost (NPC) as
(Equation 2):
NPB 2
BCR =
NPC
where
defined as the total value of benefits due to crash reduction over a defined
period based on an economic discount rate, calculated as:
NPB = NPB = (factor) x B
where (factor) is a discounting factor, for different values of rate and period;
B is the value of annual benefits (i.e. annual reduction in road crash cost)
defined as the cost of implementation (discounted if not undertaken in the

NPC =
first year).
The following factors are considered in determining a BCR:

 cost savings in crashes prevented or reduced severity (reduction in road crash cost)
 cost of implementing treatment
 cost of maintaining treatment
 cost of repairing treatment if hit
 length of analysis period
 discount rate.
The BCR will be checked to compare the cost of barrier installation for a given performance level
against the probable crash cost for the design life. This allows assessment of different barrier
performance levels and use of a higher crash cost to reflect the consequences to third parties.
A BCR > 1 is considered the criterion for installing roadside barriers with a higher performance
level. It should be noted that TMR uses the criterion of a BCR greater than 1.5 for rural roads and
2.5 for urban roads. VicRoads uses a BCR of 4.0 or higher.
Figure 3.12 presents a flowchart for determining the crash cost in a risk assessment process
based on the procedure used by the RISC software (Department of Main Roads 2005).
The steps in the flowchart are described in detail below:

1 Determine road environment variables
These variables define the roadway characteristics and are used to determine the base
encroachment frequency (the number of expected encroachments per km per year). The
road environment variables include:
— road type: divided, undivided and one-way
— number of lanes (on each carriageway)
— width of lanes (marked lanes)
— design speed = 85th percentile speed or the posted speed limit +10 km/h if the data is
not available.
Austroads 2013
— 36 —
The following roadside object attributes, in combination with vehicle speed and road
curvature, which define the probability of impact with the object need also be identified:
— horizontal offset of the object from the edge of the travelled way. This should be
calculated for each lane as the distance from the edge of each lane to the inner face of
the barrier
— object length: is the length of bridge barrier
— object width. For bridge barriers as a roadside object, the width of barrier is not taken
into account as the vehicle impact to the barrier’s back and ends is unlikely to occur
— determine traffic data.
The traffic volume is obtained from traffic survey counts and/or available reports. The
following information is derived from the traffic data:
— AADT: total estimated average annual daily traffic. AADT is taken from the project’s
EIS (environmental impact statement). A traffic growth rate, design life and maximum
lane capacity should be provided to calculate the average AADT over the design life
— growth rate p.a. (for example, Department of Main Roads (2005) uses 4%,
AS 5100.1–2004 uses 2%)
— design life
— maximum lane capacity
— the percentage of commercial vehicles.
The traffic volume is divided into the number of carriageways. For example, on a two-lane
two-way road, the traffic volume would remain unchanged (i.e. it is a single carriageway),
whereas for a four-lane divided facility, the volume is divided by two for a 50%/50% split
unless another split of traffic is evident (Department of Main Roads 2005). It should be noted
that this AADT is different from the AADT used in the chart-based procedure, which is the
AADT at the construction year.
2 Determine horizontal and vertical curvature factors

These factors modify the encroachment frequency, due to the increased probability of a
vehicle leaving the road on horizontal curves or grades. Note that the grade adjustment
factor in Department of Main Roads (2005) is the same as that of AS 5100.1–2004
(Figure 3.3); however the curvature factor is different (Figure 3.13).
Austroads 2013
— 37 —
Source: Based on Department of Main Roads (2005).
Figure 3.12: Flowchart of risk assessment process
Austroads 2013
— 38 —
Source: Department of Main Roads (2005).
Figure 3.13: Curvature adjustment factor CU
3 Calculate encroachment frequency

This is an estimation of the number of vehicles that will leave the roadway per km per year.
Not all vehicles that leave the roadway will necessarily collide with a roadside object.
Variables such as the object’s size and offset from the edge of the travelled way and vehicle
speed influence the likelihood of impact with the object. The encroachment frequency is
calculated as (Equation 3):
EF = BER x AADT x CU x GD 3
where
EF = encroachment frequency (encroachment/year/km)
BER = base encroachment rate (0.00030 encroachments/km/year/vpd)
CU = curvature adjustment factor
GD = grade adjustment factor.
The base encroachment rate of 0.00030 is taken from AASHTO (2006). This rate should be
adjusted when actual data at a specific location is available, or modified based on
engineering adjustment for non-typical conditions.
Austroads 2013
— 39 —
4 Determine severity index SI for roadside object

The SI defines the severity of the outcome of an impact with a particular roadside feature.
According to Austroads (2010a), the severity indices are based on average crash costs when
a vehicle impacts the hazard. The SI can be derived from AASHTO (1996) for traffic barriers;
however, these values which are valid for occupants of light vehicles, are not suitable for
motorcyclists and are probably not appropriate for trucks. If the situation has more than one
factor, the case with the highest SI should be selected. According to Department of Main
Roads (2005), however, a separate SI for each impact zone of the hazard should be applied.
5 Determine crash costs

The estimated crash costs per year can be calculated as:
Annual crash costs per year ($) = (impacts per year) x (severity index crash cost per impact).
Based on the calculated SI, the severity index crash cost per impact is derived from a table
issued by the relevant jurisdiction. In this table the costs related to different SI ranging from 0
to 10 are provided. The percentage values and the associated costs of the crash outcomes
(property damage, minor injury, moderate injury, hospitalisation, and fatal) may all vary
between jurisdictions. Table 3.3 (Austroads 2010a) shows an example using the 2007 costs
in Queensland.
Table 3.3: Example of summary table for crash cost related to severity index
Severity Property Minor injury Moderate Hospitalisation Fatal Cost
index (SI) damage injury ($)
0 0 0 0 0 0 0
0.5 100 0 0 0 0 7 534
1 90.4 7.3 2.3 0 0 8 526
2 71 22 7 0 0 10 531
3 43 34 21 1 1 39 801
4 30 30 32 5 3 104 121
5 15 22 45 10 8 237 550
6 7 16 39 20 18 502 132
7 2 10 28 30 30 808 931
8 0 4 19 27 50 1 218 507
9 0 0 7 18 75 1 704 580
10 0 0 0 0 100 2 144 096
Note: The crash outcome descriptions and their unit costs are based on information published in the Queensland Department of Transport and Main Roads ‘RISC
Crash Costs Update’; 23 June 2008 (cost calculation date June 2007). The SI crash cost per impact shown in the right column is the sum of the cost for each crash
outcome (determined by multiplying the percentage in the column by the unit cost for the relevant outcome).
Source: Austroads (2010a).
These crash cost are based on Road Crash Costs in Australia (Bureau of Transport
Economics 2000). Current values as determined by the Australian Bureau of Statistics
should be used consistently throughout the jurisdictions.
Comparing different performance levels, the primary benefit obtained from selecting one
design over another is the expected reduction in the future crash costs.
Austroads 2013
— 40 —
6 Determine barrier cost

The barrier costs include the cost of barrier (per m), barrier maintenance cost (per m per
year) and barrier repair cost (per penetration per year). While the cost of barrier is the
present value, the other two costs are assumed to be equally distributed over the design life
(DL). Their associated present values should be calculated.
7 Determine BCR
The BCR between two bridge barrier alternatives is calculated by (Equation 4):
BCR = (NB1 – NB2) / (NC2 – NC1) 4
where
NB1 = net present values of benefit of the barrier alternative 1

(e.g. Medium performance level barrier)
NB2 = net present values of benefit of the barrier alternatives 2

(e.g. High performance level barrier)
NC1 = net present value of cost of the barrier alternative 1
NC2 = net present value of cost of the barrier alternative 2.
3.7.5 Example 2
In this example, the bridge site has a site condition CAT3 and according to the minimum standard,
a High performance level can be specified. However, a B/C analysis according to the preferred
standard shows that a Medium performance level can be used.
The same set of bridge site data in Example 1 in Section 3.6.4 is used except that the commercial
vehicles at the bridge site consist of heavy, high centre-of-gravity vehicles.
Determine site condition category

With the presence of heavy, high centre-of-gravity vehicles, this site has a CAT3 site condition
category. As a result, CAT3 methodology will be used to determine the required performance level
for this site.
Determine performance level based on specification standards

The site condition of this site satisfies the criteria for a ‘preferable’ standard High performance level
based the following information:
 heavy, high centre-of-gravity vehicles are present on the road
 the volume of commercial vehicles is 17%, which is between 15% and 30% of AADT.
Austroads 2013
— 41 —
Benefit-cost analysis
A B/C analysis is required to assess if a performance level lower than the High performance level
could be provided. This analysis will compare the BCR of Medium and High performance level
barriers. A B/C analysis based on the method proposed is carried out. The detailed calculations
are presented below:
 Road environment variables and roadside object attributes (Table 3.4) and lane data
(Table 3.5)
Table 3.4: Example 2 – road data
Road environment variables and roadside Abbreviation Unit Value
object attributes
Design speed V km/h 110
Design life DL Years 20
Road type (divided, undivided or one-way) N/A N/A Undivided
Road grade in the direction of traffic N/A N/A –2.5%
Number of lanes NL N/A 2
Length of bridge barrier BL m 100
Table 3.5: Example 2 – lane data

Lane Abbreviation Unit Lane 1 Lane 2
Lane width W m 3.5 3.5
Offset from lane to object Offset m 1.2 4.7
 Traffic data (Table 3.6)

Table 3.6: Example 2 – traffic data
Traffic data Abbreviation Unit Value Formula
AADT at census year AADTC vpd 12 000
Number of years to the construction year n1 Year 4
Growth rate of traffic volume g % pa 3.0%
Base AADT (survey data at year '0') AADT0 vpd 13 506 AADT0 = AADTC (1+g)n1
AADT at year 'n' AADTn vpd AADTn = AADT0 (1+g)n
Maximum lane capacity AADTlane.max vpd 10 000
Average AADT over the design life AADT vpd 17 538 = total AADT/DL
Portion of commercial vehicles of AADT CV % 17%
The AADT at year ‘n’ after construction is calculated as AADTn = AADT0(1 + g)n. At year 'x'
when AADT is greater than NL*AADTlane.max, the AADT of subsequent years will be
NL*AADTlane.max until the design life DL is reached. With the above data, the maximum lane
capacity is not reached for the whole design life. Thus the average AADT over the design life
is AADT = 17 538 vpd.
Austroads 2013
— 42 —
The traffic growth adjustment factor TG is used to average the traffic volume over the life of
the project and is calculated as follows (Equation 5):
∑ (1 + g/ 100 )
DL
1
TG = n
5
n =1 DL
Commercial vehicles (CV) will have more severe crash consequences, thus the crash cost
due to CV impact should be calculated. The portion of CV in AADT is taken from survey data
and is assumed to be unchanged over the design life of the barrier.
 Grade and curvature factors

The grade factor, GD, and curvature factor, CU, are derived from Figure 3.13 and Figure 3.4,
respectively. As the bridge is located on a straight road section with a 4% downgrade, these
factors are GD = 1.13, and CU = 1.25.
 Encroachment frequency
Basic encroachment rate can be taken as (AASHTO 2006) (Equation 6):
BER = 0.00030 encroachments/km/AADT/year 6
The probability of encroachment, Probenc_lane is derived from Figure 3.14 based on the design
speed and the lane’s edge offset. It should be noted that values of the probability of
encroachment are also available in tabulated form. The probability of encroachment should
be calculated for each lane as the edge offset of each lane varies. With the design speed
= 110 km/h, lane 1 has an edge offset = 1.2 m, thus Probenc_lane = 82%, lane 2 has an edge
offset = 4.7 m, thus Probenc_lane = 46%.
Source: Department of Main Roads (2005).
Figure 3.14: Probability encroachment curve
Austroads 2013
— 43 —
The base encroachment frequency for each lane is calculated using Equation 7:
BEFlane = BER x Probenc_lane x AADTlane 7
and has following values for each lane (Table 3.7):
Table 3.7: Example 2 – base encroachment frequency

Lane Abbreviation Unit Lane 1 Lane 2
Lane average AADT AADTlane vpdpl 8769 8769
Probability of encroachment from lane Probenc_lane % 82% 46%
Base encroachment frequency per lane BEFlane encr/km/yr/lane 2.16 1.21
The total base encroachment frequency is the sum of the base encroachment frequency for each
lane: BEF = 2.16 + 1.21 = 3.37.
Therefore the encroachment frequency is calculated as (Equation 8):
EF = BEF x CU x GD 8
Thus, EF = 4.76.
For each barrier test level, the portion of CV which results in penetration if impacting with barriers
varies. An assessment should be performed based on the site-specific conditions to determine this
parameter for each performance level barrier. For this example, the assessment is presented
below.
For Medium and High performance level barriers, there is no data available for the penetration
rates. They should be determined based on an assessment of the CV characteristics based on the
features of the traffic flow at the site.
It is assumed that a Medium performance level barrier will prevent all non-tanker trucks up to 36 t
from penetrating the barrier. Class 8 and 9 vehicles (Austroads 2006) can have masses in excess
of 36 t, and up to 44 t.
The critical encroachment angle for the barrier design is 15° (test condition for Medium
performance level). The distribution of encroachment angles obtained by Hutchinson and Kennedy
(quoted in Lynam & Kennedy 2005) is shown in Table 3.8.
Table 3.8: Example 2 – distribution of encroachment angles

Angle (degrees) 5 10 15 20 30
Cumulative per cent 25 60 75 85 95
The following assumptions are made to determine the probability of barrier penetration of heavy
vehicles:
 50% of CVs can potentially exceed 36 t mass (i.e. class 8 or higher vehicle).
 75% of impacts are at an angle of less than or equal to 15°. Therefore 25% of impacts will
potentially exceed the Medium performance level loading.
 20% of CVs that will impact at the barrier at an angle of greater than 15° will have a mass
greater than 36 t.
Austroads 2013
— 44 —
 50% of CVs heavier than 36 t will actually penetrate when impacting with barrier at an angle
of greater than 15°.
Based on these assumptions, the probability of penetration is therefore 0.5 x 0.2 x 0.25 x 0.5 =
1.25%. Due to the assumptions in determining this figure, a value of 1% is proposed.
For High performance level barrier loading, it is assumed that only 0.5% of all commercial vehicles
will penetrate the barrier.
The penetration rates for heavy vehicles adopted for this example are as follows:
 Estimated portion of CV causing penetration for Medium performance level barriers (needs
High performance level barriers), CVMedium = 1%.
 Estimated portion of CV causing penetration for High performance level barriers (needs
Special performance level barriers), CVHigh = 0.5%.
The probability of reaching an object after penetration is 1, since for a bridge it is certain that after
penetration of the barrier, the vehicle will reach an object. As a result, the following can be derived:
 Number of penetrations of Medium performance level barrier per year is:
CVPN = CV*EF*CVMedium*BL/1000*Probpene = 0.00081.
 Number of penetrations of High performance level barrier per year is:
CVPN = CV*EF*CVHigh*BL/1000*Probpene = 0.00040.
 Crash cost.
The crash cost is calculated for each barrier performance level, including CV penetration cost
(truck driver and third party) and barrier impact cost. A discount factor of 6% is used to annualise
the cost to calculate the net present cost.
In general, the severity index crash cost per impact is taken from the relevant jurisdiction. For this
example, the severity index is derived from AASHTO (1996). For crashes that cause barrier
penetration, the vehicles are extremely likely to fall off the bridge, thus an SI = 10 was selected.
The penetration cost for a truck impacting with an object under the bridge should be determined
based on a site-specific assessment, taking into account the under-bridge condition – water,
passenger rail crossing, freight rail crossing, and road crossing.
With a severity index for penetration of CV (to the driver) of SI = 10, the unit penetration cost to the
driver of a CV that causes penetration, Cost_drv, is taken as A$2 485 492 (2008 cost). For this
example, a crash cost of A$30 000 000 is also assumed for a crash involving freight trains
(cost_fr).
Austroads 2013
— 45 —
The penetration cost over the design life of the barrier is calculated for each barrier performance
level by (Equation 9):
Penecost = (Cost_drv + Cost_fr) x CVPN x DL 9
where
Cost_drv = unit penetration cost to the driver of the commercial vehicle that causes the
penetration
Cost_fr = unit penetration cost for the truck impacting with an object under the bridge
number of penetrations of commercial vehicles on the barrier performance

CVPN =
level under consideration.
For a Medium performance level barrier,

Penecost = (2 485 492 + 30 000 000) x 0.0008086 x 20 = A$525 352
For a High performance level barrier,

Penecost = (2 485 492 + 30 000 000) x 0.000404 x 20 = A$262 676
For comparison purposes, it is likely that the costs due to vehicles impacting with the barrier are
the same for all barrier types, thus this cost will not be considered in this example. As a result, only
penetration cost related to trucks impacting with objects under the bridge will be taken into
account.
Assuming that this cost is equally distributed over the design life of the barrier DL, the equivalent
present value is calculated as PV_penecost = PV(6%, 20 year, Penecost/DL).
It should be noted that the present value PV of amount A invested equally in n year with a discount
rate g (%) is calculated as (Equation 10):
PV = A (1 – (1 + g)-n) / g 10
Values of Penecost and PV_penecost for all performance level barriers are as in Table 3.9.
Table 3.9: Example 2 – accident cost

Alternative barrier performance level Unit Medium High
Penetration cost A$ 525 352 262 676
Average penetration cost/year A$ 26 268 13 134
Total accident net present cost A$ 301 287 150 644
The penetration cost (which can be avoided by the provision of a barrier) is the net benefit in the
B/C analysis (NB).
Austroads 2013
— 46 —
 Barrier cost
The barrier costs include the construction cost of barrier (per m), barrier maintenance cost
(per m per year) and barrier repair cost (per penetration per year). These values are
assumed as shown in Table 3.10 for each barrier type. While the cost of barrier is the
present value, the other two costs are assumed to be equally distributed over the design life
DL and their associated present values should be calculated, as follows:
— Present value of maintenance cost (A$/m) (Equation 11):
PV_mtnce = PV(6%, 20 year, maintenance cost per year) 11
— Repair cost per year (A$/year) (Equation 12):
Repair cost = number of crashes x unit repair cost 12
— Present value of repair cost (A$) (Equation 13):
PV_repair = PV(6%, 20 year, repair cost per year) 13
— The total net present cost of barrier including barrier cost, maintenance cost and repair
cost over the design life is calculated as (Equation 14):
Total barrier net present cost = (PV_barrier + PV_mtnce) x BL + NPVrepair 14
where
PV_barrier = unit penetration cost to the driver of the commercial vehicle

that causes the penetration
BL = total length of the bridge barrier
The cost of barrier is the net cost in the B/C analysis (NC) and is calculated in Table 3.10.
Table 3.10: Example 2 – costs of the two barrier alternatives

Alternative barrier performance level Unit Medium High
Construction cost of barrier A$/m 600 1 100
Maintenance cost per year A$/m/year 10 10
Present value of maintenance cost A$/m 115 115
Unit repair cost A$/crash/year 2 000 2 000
Repair cost per year A$/year 2 1
Present value of repair cost A$ 19 9
NP cost of barrier A$ 60 180 110 090
Austroads 2013
— 47 —
 Benefit-cost ratio
The BCR of a High performance level barrier over a Medium performance level barrier is
calculated as in Equation 4:
BCR = ($301 287 – $150 644) / ($110 090 – $60 180) = 3.02
A BCR of 3.02 is a good basis for a High performance level bridge barrier to be provided for
this site. This confirms the selection based on the criteria of minimum specification standard
for a High performance level.
It is acknowledged that while the methodology presented refers to and uses the principles of
the RISC method, TMR does not use RISC for bridge barriers. As far as possible, generic
data has been used, however where applicable, jurisdictions need to introduce
bridge-specific data.
Austroads 2013
— 48 —
4 BRIDGE BARRIER DESIGN PROCEDURE

Figure 4.1 shows a general layout of bridge barrier, bridge approach barrier, transition and end
treatment.
Source: AASHTO (2006).
Figure 4.1: Bridge barrier component layout
This section provides suggested procedures and detailed guidance for the structural design of
bridge barriers located in the area A in Figure 4.1 where the bridge abutments consist of
spill-through earth batters. For bridges comprising vertical wall or near vertical wall abutments,
such as reinforced soil walls, and where such walls extend parallel to the direction of traffic, area A
in Figure 4.1 shall extend to the approach ends of the walls. Where earth embankments are
retained by vertical wall or near vertical wall structures, such structures shall be considered no
different to a bridge structure when designing traffic barriers, and traffic barriers on such structures
shall be designed as for traffic barriers in area A in Figure 4.1.
4.1 Barrier Performance Specifications

Table 3.1 shows the link between the barrier performance levels with the containment
requirements and the required design loads specified in the equivalent NCHRP 350 test levels.
4.2 Structural Design Criteria

The ultimate design loads and load distribution lengths for all performance levels as defined by
AS 5100.1 (2004) are given in Table 4.1 and illustrated in Figure 4.2.
Two design checks shall be performed to verify the structural adequacy of the barrier, including:
 Structural adequacy: the barrier total transverse resistance, R*, shall be greater than or equal
to the transverse loads Ft specified in Table 4.1.
 Barrier height: the effective height Y* of the barrier shall be greater than or equal to the
minimum effective height, He specified in Table 4.1.
Austroads 2013
— 49 —
Figure 4.2: Barrier design forces
Table 4.1: Traffic barrier design loads, contact lengths and effective heights
Barrier performance level Unit Low Regular Medium High Special
Ultimate transverse outward load (Ft) kN 125 250 500 750 1 000
Ultimate longitudinal or transverse inward load (FL) kN 40 80 170 250 330
Ultimate vertical downward load (Fv) kN 20 80 350 350 450
Vehicle contact length for transverse loads (Lt) and
mm 1 100 1 100 2 400 2 400 2 500
longitudinal loads (LL)
Vertical contact length for vertical loads (Lv) mm 5 500 5 500 12 000 12 000 15 000
Minimum effective height He mm 500 800 1 100 1 400 1 400
Source: Adopted from AS 5100.1 (2004).
The following should be noted:

 The data given in Table 4.1 is based on a lateral combined barrier/vehicle deformation of
300 mm for the Low and Regular performance levels, and 500 mm for Medium, High and
Special performance levels.
 Load combinations to be considered include: (a) transverse and longitudinal loads applied
simultaneously; and (b) vertical loads only.
 The loads are applied uniformly over the relevant specified contact lengths.
 All loads shall be applied to the longitudinal barrier elements. The distribution of the
longitudinal loads to posts shall be consistent with the continuity of rail elements. Distribution
of transverse loads shall be consistent with the assumed failure mechanism of the barrier
system.
The values in Table 4.1 were derived based on information from AASHTO (2012) and AS 5100.2
Supplement 1 (2007).
Austroads 2013
— 50 —
Contact lengths
The lengths for Lt, LL and Lv relate to the length of significant contact between the vehicle and
barrier that has been observed in films of crash tests. The length of 1100 mm for the Regular
performance level (TL-4) is the rear-axle tyre diameter of the truck. The length of 2400 m for the
Medium performance level (TL-5) is the length of the prime mover or trailer rear tandem axles (two
1050 mm diameter tyres, plus 300 mm between them).
For High performance level barriers, the contact lengths are taken from the AASHTO (2012) test
level TL-6.
The contact lengths of Special performance level barriers correspond approximately to the
AASHTO (1989) performance level PL-4.
Design forces
Fv is the weight of the vehicle lying on top of the bridge barrier, and is distributed over the length of
the vehicle in contact with the rail, Lv.
Design forces specified for High and Special performance level barriers are taken respectively as
approximately 50% and 100% greater than those specified for Medium performance level barriers
to provide clearly recognisable steps in containment capacity over those defined for lower
performance levels.
Minimum effective heights

The required minimum effective heights of Low, Regular and Medium performance level barriers
are based upon theoretical assessments supported by overseas testing and in-service
performance, to prevent the relevant vehicles given in Table 10.4 of AS 5100.1 (2004) from
overturning these barriers under the specified impact criteria.
Theoretically, the effective height of the vehicle rollover force, He is taken as (Figure 4.3):
Figure 4.3: Barrier design diagram
Austroads 2013
— 51 —
WB 15
He =G −
2Ft
where
G = height of vehicle centre-of-gravity above bridge deck (mm)
W = weight of vehicle corresponding to the required test level (N)
B = out-to-out wheel spacing on an axle (mm)
Ft =
transverse force specified in Table 4.1 assumed to be acting at top of a
concrete wall (N)
The minimum effective height of 500 mm of Low performance level barriers is based upon the tyre
height of the test vehicles and other light vehicles.
The effective height of 800 mm of Regular performance level barriers is based on the theoretical
calculation (Equation 15) and on satisfactory performance in preventing light test vehicles and the
8 tonne rigid test truck rolling over such barriers, under the impact criteria specified in the crash
test.
The minimum height, He of 1100 m for Medium performance level (TL-5) barriers is based on the
height used for successfully crash-tested concrete barrier engaging only the tyres of the truck. For
post-and-rail metal bridge railings, it may be prudent to increase the height by 300 mm so as to
engage the bed of the truck.
The minimum height, He, shown in Table 4.1, for High performance level (TL-6) barriers is the
height required to engage the side of the tank as determined by crash test.
For High performance level barriers, the effective height of 1400 mm is aimed at containing the tray
of heavy high centre-of-gravity van type semi-trailers. This is to prevent them from overturning.
In order to prevent a Special performance level heavy tanker type semi-trailer from overturning, a
greater effective height, in the order of 1700 mm to 2000 mm may be required for a combination
parapet to prevent overturning of such a vehicle. Hence, in the case of a very high-risk situation, an
individual B/C assessment is still required.
In the case of other Special performance level barriers, the appropriate effective height shall be
determined and specified by the relevant jurisdiction.
It should be noted that the procedures in Appendix A in AASHTO (2012) are not applicable to
traffic railings mounted on rigid structures, such as retaining walls or spread footings, when the
cracking pattern is expected to extend to the supporting components.
It should also be noted that the ‘semi-rigid’ barrier designs used commonly in New Zealand (for
Low and Regular performance levels) are designed based on the US tested systems. Refer to
Transit New Zealand (2004) for the design criteria.
Austroads 2013
— 52 —
4.3 Selection of Bridge Barrier Types

There are numerous bridge barrier designs for each performance level among the jurisdictions.
These designs have been developed based on overseas crash-tested barriers with minor
modifications. Based on the available materials obtained from various jurisdictions, there are three
types of bridge barriers. They include:
 concrete barriers
 post-and-rail barriers
 combined concrete parapet and metal rails.
4.3.1 Concrete Barriers

Typical cross-sections of bridge concrete barrier include:
 vertical wall
 single slope
 F-shape and New Jersey.
The properties of these types of barriers are discussed below.
Vertical wall barrier

Vertical wall concrete barriers (Figure 4.4a) do not lift the vehicle and hence do not have the
energy management feature of the F-shape, single slope or New Jersey barriers. Vertical wall
concrete barriers can be an effective alternative to the wider safety-shape barriers and can
preserve available median width at narrow locations such as in front of bridge piers. Vehicle
damage in crashes with a vertical wall is greater than with safety-shaped barriers, but injuries are
comparable and preservation of shoulder width is a safety benefit (AASHTO 1996). In a crash with
a vertical wall all four wheels remain on the ground and this minimises the potential for vehicles to
roll over. The trajectory of cars after they crash into vertical walls is also uncertain because wheel
damage may occur as the axle makes contact with the barrier.
For these reasons vertical wall barriers are not preferred. However, they may be suitable in some
situations on urban roads where road width is highly constrained or on low speed roads where the
appearance of the road environment is important; for example, where it is desired to construct or
face the wall using natural stone or some other material. An advantage of the vertical wall is that its
profile is not affected by re-sheeting or resurfacing of the pavement.
Austroads 2013
— 53 —
Figure 4.4: Typical types of concrete barriers
Single slope barrier

Single slope barriers (Figure 4.4b) have a more consistent performance than vertical wall barriers.
Single slope barriers can also facilitate pavement resurfacing without the profile being adversely
affected.
Two types of single slope barriers have been developed. The Texas barrier has a wall slope of
10.8° and the Californian has a wall slope of 9.1°. These were developed by the Texas
Transportation Institute and the California Department of Transport respectively.
Crash tests have indicated that the performance of the Texas single slope barrier is comparable to
New Jersey barrier and that the performance of the Californian single slope barrier is comparable
to that of the F-shape barrier. In approving the use of single slope barriers, the US Federal
Highway Administration (FHWA) suggested that the Californian single slope barrier is an
improvement over both the standard New Jersey concrete barrier shape and the Texas single
slope barrier because of the reduced vehicular climb seen upon impact and the less severe
post-crash vehicular trajectories observed in crash test videos.
It appears that the Californian constant slope barrier is the preferred concrete barrier shape.
Austroads 2013
— 54 —
New Jersey barrier

The New Jersey barrier (Figure 4.4c) has been used throughout Australia and has provided
satisfactory service. For common shallow-angle impacts the shape is intended to minimise sheet
metal body damage by allowing the vehicle tyres to ride up on the lower sloped face. Energy is
dissipated by lifting and lowering of the vehicle, compression of the vehicle suspension and
deformation of the body of the vehicle. While New Jersey barriers remain in service, the shape was
modified to the F-shape barrier to address the likelihood of vehicle rollover when impacted by small
to medium vehicles. Consequently, jurisdictions generally no longer install New Jersey barriers.
For higher impact angles the New Jersey shape results in a staged response by an impacting
vehicle, namely:
 The vehicle bumper impacts the upper sloped face and slides upwards, lifting the vehicle.
 As the vehicle becomes more parallel with the barrier, the wheel contacts the lower sloped
face causing additional lift through compression of the front suspension.
 The lifting reduces friction between the tyres and the paved surface and this facilitates
banking and redirection of the vehicle.
Excessive lifting of the vehicle may cause it to yaw, pitch or roll during contact with the barrier, and
to roll over when the tyres make contact with the road again. As wheel side-rubbing forces can
provide additional lift, exposed aggregate and other rough surfaces should be avoided.
The 75 mm high vertical face at the base of the New Jersey barrier is intended to provide an
allowance for future pavement overlays. Apart from increasing the extent to which a vehicle is
lifted, this vertical face plays no significant role in the performance of the barrier.
F-shape barrier
The F-shape barrier (Figure 4.4d) has a similar profile to the New Jersey barrier, the main
difference being that the height of the lower sloped surface is smaller. The major performance
difference is that the lower slope of the F-shape profile significantly reduces the lifting of an
impacting vehicle, resulting in a reduced tendency for vehicles to roll, particularly small cars.
Refer to Section 5 for typical concrete barriers used in Australia.
4.3.2 Post-and-rail Barriers

Post-and-rail barriers have been used mainly for Low and Regular performance levels. The
components of this barrier type include:
 Posts: steel hollow box or channel sections, connected to the bridge deck by bolts.
 Rails: steel hollow box or W-beam or thrie-beam sections bolted to the posts. There may be
one, two or three-rail configurations.
Refer to Section 5.2 for typical post-and-rail barriers used in Australia.
4.3.3 Combined Concrete Parapet and Metal Rails

This type of barrier has been used for Regular, Medium and High performance levels. The barrier
consists of a post-and-rail system on top of a concrete parapet.
The heights of the parapet and post are selected based on the requirement of minimum effective
height of the barrier.
Refer to Section 5.2 for typical combined concrete parapet and metal rail barriers used in Australia.
Austroads 2013
— 55 —
This type of barrier has been used widely, for example, in NSW and Victoria for Regular and
Medium performance levels.
4.3.4 Acceptance Criteria for Non-standard Bridge Traffic Barriers

The current version of AS 5100.1 (2004) specifies that only barriers which have been
demonstrated to the relevant jurisdiction to have acceptable vehicle/barrier interaction performance
shall be used. This performance shall be demonstrated by either:
 full-scale crash tests
 methods based on crash-tested barriers used elsewhere
 methods based on barriers that can be geometrically and structurally evaluated as equivalent
to a crash-tested system
 for a barrier that has the same details as those of an existing barrier, a method based on an
evaluation of the existing barriers performance, to the approval of the relevant jurisdiction.
As it is unlikely that physical crash tests have been conducted for all bridge barriers in Australia,
the barrier designs that are geometrically equivalent to crash-tested systems developed in the USA
should be selected.
In addition to fulfilling the above requirements, the alternative barrier shall be designed in
accordance with the design methods presented in this document and to the appropriate
performance specification presented in Section 4.2.
4.4 Concrete Parapets

4.4.1 Barrier Geometry
The concrete barrier profiles shown in Figure 4.4 shall be used as the basis for parapet, and
combined parapet and horizontal rail traffic barriers. For these profiles, the height of the barrier
shall be determined by the required barrier performance level.
4.4.2 Design Process

The method used in this section has been adopted from AASHTO (2012), which is based on yield
line failure analysis.
It is assumed in this analysis that the yield line failure pattern occurs within the parapet only and
does not extend into the deck. This means that the deck must have sufficient resistance to force
the yield line failure pattern to remain within the parapet. If the failure pattern extends into the deck,
the equations for resistance for the parapet are not valid.
The analysis is also based on the assumption that sufficient longitudinal length of parapet exists to
result in the yield line failure pattern shown in Figure 4.5 or Figure 4.6. For short lengths of parapet,
a single yield line may form along the junction of the parapet and deck. Such a failure pattern is
permissible, and the resistance of the parapet should be computed using an appropriate analysis.
This analysis is based on the assumption that the negative and positive wall resisting moments are
equal and that the negative and positive beam resisting moments are equal.
The measurement of system resistance of a concrete barrier is R* = Rw in this case, which is

compared to the loads in Table 4.1 to determine structural adequacy. The flexural resistances, Mb,
Mw and Mc, (Figure 4.5), are related to the system resistance Rw through the yield line analysis. Rw
is the ‘nominal resistance’ because it is compared to the ‘nominal load’ given in Table 4.1.
Austroads 2013
— 56 —
The procedure includes the following steps:

 Step 1: Calculate the concrete capacities of the various components of the barrier, which
includes Mb – flexural capacity of the beam at the top of the wall about the beam’s vertical
axis (if the beam exists), Mw – flexural capacity of the wall about the vertical axis, and Mc –
flexural capacity of the wall about an axis parallel to the bridge’s longitudinal axis.
 Step 2: Calculate the critical length of the yield line failure pattern Lc.
 Step 3: Calculate the barrier’s resistance Rw.

Step 4: Check the adequacy of the barrier.
Two impact cases may occur. Firstly, impact within the wall segment (Figure 4.5) and secondly,
impact near the end of the wall segment (Figure 4.6).
Figure 4.5: Failure mechanism for impact within wall segment
Austroads 2013
— 57 —
Impact within wall segment

The barrier resistance Rw and the critical wall length Lc are calculated as (Equation 16 and
Equation 17):
 2  M L2  16
Rw =    8M b + 8Mw + c c 
 2Lc − Lt  H 
 L  8H (M b + Mw )
2 17
Lt
Lc = +  t +
2 2 Mc
where
H = height of wall (mm)
Lc = critical wall length over which the yield line mechanism occurs (mm)
Lt = longitudinal length of distribution of impact force Ft specified in Table 4.1 (mm)
Rw = total transverse resistance of the barrier (N)
additional flexural resistance of beam about the vertical axis in addition to M w ,

Mb =
if any, at top of wall (Nmm)
Mw = flexural resistance of the wall about the vertical axis (Nmm/mm)
flexural resistance of cantilevered wall about an axis parallel to the longitudinal

Mc =
axis of the bridge, per metre strip (Nmm/mm)
Impacts near end wall segment

In this case, the barrier resistance R and the critical wall length Lc are calculated as (Equation 18
and Equation 19):
 2  M L2  18
Rw =    M b + Mw + c c 
 2Lc − Lt  H 
L L  (M + Mw )
2
Lc = t +  t  + H b
19
2 2 Mc
Austroads 2013
— 58 —
Figure 4.6: Failure mechanism for impact near end wall segment
The flexural capacity of the barrier components, Mb, Mc, and Mw can be calculated as shown in
Figure 4.7 using Equation 20 (based on AS 5100.5 (2004) for a rectangular concrete section with
tensile steel).
Figure 4.7: Calculation of Mb, Mw and Mc
Austroads 2013
— 59 —
 Af  20
M = φMu = φAs fy d 1 − 0.6 s y 
 b d f 'c 
where
φ capacity reduction factor, taken from AS 5100.5 (2004) for concrete
=
members
As = total cross-section of reinforced bar in the beam (mm2)
fy = minimum yield strength of steel reinforcement (N/mm2)
d = effective height of the beam section (mm)

b = effective width of the beam section (mm)
f 'c = minimum yield strength of concrete
The following should be noted:

 For use in the above equations, Mc and Mw should not vary significantly over the height of the
wall. For other cases, a rigorous yield line analysis should be used.
 Where the width of the concrete wall varies along the height, Mc should be taken as the
average of its values along the height of the wall.
 If the parapet length between joints is less than Lc then Lc must be set to the actual length.
 The yield line analysis shown in Figure 4.5 and Figure 4.6 includes only the ultimate flexural
capacity of the concrete component. Stirrups or ties should be provided to resist the shear
and/or diagonal tension forces.
 Reinforcing steel for concrete barriers shall have sufficient embedment length to develop the
yield strength.
 The ultimate flexural resistance of the bridge deck or slab should be determined recognising
that the deck is also resisting a tensile force, caused by the components of the impact forces,
Ft.
 For F-shape barriers and the like, two possible failure mechanisms should be taken into
account, as shown in Figure 4.8.
Figure 4.8: Possible failure mechanisms for F-shape barriers and the like
Austroads 2013
— 60 —
4.4.3 Design of Bridge Cantilevered Deck Slab

According to AS 5100.2 (2004), the loads transmitted to the bridge deck shall be determined from
the results of load testing and ultimate strength analysis of the barrier system using the loads given
in Table 4.1.
Sections on the deck slab that should be checked include (Figure 4.9):
 Section A-A at inside face of parapet
 Section B-B at the far edge of the overhang
 Section C-C at the design section in the first interior span.
A load factor of 1.1 shall apply to the design of deck cantilevers for the effects of barrier loads. This
is to ensure that the bending resistance of the deck slab is larger than the bending resistance of
the parapet, thus ensuring failure occurs first in the barrier.
The barrier impact loads and traffic loads on the deck need not be applied simultaneously when
designing the deck. For barrier impacts, the following two design cases shall be considered:
 Design case 1: permanent loads + the transverse and longitudinal forces specified in
Table 4.1.
 Design case 2: permanent loads + the vertical forces specified in Table 4.1.
For design case 1, the deck slab shall be designed to resist a bending moment of 1.1 x Mc acting
coincident with the axial tensile force 1.1 x T. The axial tensile force T is calculated using
Equation 21:
Austroads 2013
— 61 —
Figure 4.9: Design sections in the overhang region
Rw 21
T =
(Lc + 2H )
where
Rw = parapet resistance (N)
Lc = critical length of yield line failure pattern (mm)
H = height of wall (mm)
T = tensile force per unit of deck length (N/mm)
Austroads 2013
— 62 —
In the above equation, the tensile strength of the deck slab is calculated based on the actual
resistance of the parapet Rw, which ensures that the deck overhang region is designed to have a
resistance larger than the actual resistance of the concrete parapet. For tensile forces, a 45° angle
for load distribution is assumed, thus the length of the deck slab carrying the tensile load T is
Lc + 2H.
At the inside face of the parapet, the collision forces are distributed over a distance Lc for the
moment and Lc + 2H for axial force. It is reasonable to assume that the distribution length will
increase as the distance from the section to the parapet increases.
The value of the distribution angle is not stated in the specifications and is determined using
engineering judgment. Distribution angles between 30° and 45° have been used to calculate the
increase in the distribution length, which is acceptable. This distribution angle can be used for both
sections B-B and C-C in Figure 4.9. An angle of 30° is preferable as it is more conservative.
For design case 2, the design of the deck slab shall be based on the cantilevered portion of the
deck. However, for concrete parapets, the case of vertical collision never controls.
It should be noted that if the deck overhang capacity is less than that specified, the yield line failure
mechanism for the parapet may not develop as shown in Figure 4.5 and Equation 16, and
Equation 17 will not be correct.
The crash testing program is oriented toward survival, not necessarily the identification of the
ultimate strength of the barrier system. This could produce a barrier system that is significantly
overdesigned, leading to the possibility that the deck overhang is also overdesigned.
4.5 Post-and-rail Barriers

For traffic barriers with post and horizontal rail components, the following geometric requirements
shall be applied (Figure 4.10).
The total depth of longitudinal barrier components, including longitudinal rails and any kerb or
parapet section above the reference surface, in contact with the vehicle, ΣA, shall not be less than
25% of the height of the barrier (H). The height of the barrier (H) is measured from the reference
surface to the upper surface of the top vehicle contact rail.
The clear vertical opening (cb) below the lowest rail shall be not greater than 380 mm. The clear
vertical opening between rails (c) or between the top of a concrete barrier and a horizontal rail (A1,
A2…) shall be not greater than 380 mm. The vertical clear opening between rails is the distance
between square or rectangular shapes and the extreme traffic-side face of curved or circular
shapes.
Unless approved otherwise by the relevant jurisdiction, posts shall be set back from the traffic face
of the traffic rails by a minimum of:
 100 mm for ΣA/H ≥ 0.5
 200 mm for ΣA/H ≤ 0.3.
Linear interpolation shall be used where ΣA/H is between 0.3 and 0.5.
Austroads 2013
— 63 —
Figure 4.10: Post-and-rail barriers
The post setback, S, is defined as the horizontal distance between the traffic faces of the railing
and the traffic faces of plane-faced posts or the centre of circular posts.
The traffic faces of all traffic rails shall be within 25 mm of a vertical plane through the face of the
rail closest to the traffic. Rails further back than 25 mm or centred lower than 380 mm above the
reference surface shall not be considered as traffic rails for the purpose of resisting the design load
specified in AS 5100.2 (2004).
The two- and three-rail barrier systems used by VicRoads and MRWA (Section 5.2) are suggested
for Low and Regular performance levels.
Austroads 2013
— 64 —

It is suggested that an inelastic analysis be used for design of a post-and-rail barrier under failure
conditions. The post on each end of the plastic mechanism must be able to resist the rail or beam
shear. One, two, or more barrier spans may be involved in a failure mode. Figure 4.11 presents
several possible failure modes, including single-span, two-span, and three-span failure modes.

 Step 1: Calculate the ultimate capacities of the posts and rails.
 Step 2: Calculate the combined barrier resistance.
 Step 3: Check the adequacy of the barrier.
Depending on the location of the failure, there may be two cases, including failures that involve and
do not involve the end post of a segment.
The critical nominal resistance, R* = Rr in this case, when the failure does not involve the end post
of a segment, shall be taken as the lowest value determined from Equation 22 and Equation 23 for
various numbers of railing spans, N, as follows:
 For failure modes involving an odd number of barrier spans, N:
16M p + (N − 1)(N + 1)PpL 22

Rr =
2 NL − Lt
 For failure modes involving an even number of barrier spans, N:
16M p + N 2PpL 23
Rr =
2NL − Lt
where
L = post spacing or single-span (mm)
Mp inelastic or yield line resistance of all of the rails contributing to a

=
plastic hinge (Nmm)
Pp shear resistance of a single post which corresponds to the inelastic

=
resistance of the post and is located Y* above the deck (N)
Rr = total transverse resistance of the rails and post (N)
Austroads 2013
— 65 —
Figure 4.11: Possible failure modes for post-and-rail barriers
The inelastic or yield line resistance of steel post or rail, Mp can be calculated by Equation 24:
M p = φ fy S 24
where
φ = capacity reduction factor, φ = 0.9 for steel material
S = plastic section modulus of the rail’s cross-section
fy = yield strength of the rail material
It should be noted that the contribution of each rail to Mp is proportional to the height of the rail
(distance from the rail to the base of the post). For example, in Figure 4.12, rail 2 contributes 100%
Mp2 and rail 1 contributes (h1/h2) x Mp1.
Austroads 2013
— 66 —
Figure 4.12: Contribution of yield line resistance of rails
The shear resistance of the post, Pp is calculated by the following formula, based on the post’s
inelastic resistance Mp-post (Equation 25):
M p − post 25
Pp =
Y*
where
Y* = effective height of the barrier
 For impacts at the end of rail segments that cause the end post to fail, the critical rail nominal
resistance, R r , shall be calculated as follows for any number of barrier spans, N
(Equation 26):
26
∑i)
N
2M p + 2PpL(
Rr = i =1
2NL − Lt
The total transverse resistance of a post-and-rail barrier R*, shall be greater than or equal to the
transverse loads specified in Table 4.1.
The effective height of the barrier, Y*, is calculated as (Equation 27):
27
∑
n
RiYi
Y* =
i =1 Rr
where
n = number of rails
Ri = proportion of the total ultimate resistance taken by the ith rail (N)
Yi = distance from the bridge deck or reference surface to the ith rail (mm)
For multiple rail systems, each of the rails may contribute to the yield mechanism shown
schematically in Figure 4.11, depending on the rotation corresponding to its vertical position.
Austroads 2013
— 67 —
It should be noted that, if the total resistance, Rr , of a post-and-rail barrier system with multiple rail
elements is significantly greater than the applied load, Ft, then the resistance for the lower rail
element(s) used in calculations may be reduced. The reduced value of Rr will result in an increase
in the computed value of Y*. The reduced notional total rail resistance and its effective height must
satisfy the requirements in Section 4.2.

Post-and-rail systems, such as a metal system with wide flange or tubular posts, impose large
concentrated forces and moments on the deck at the point where the post is attached to the deck
(Figure 4.13).
Two design cases as described in Section 4.4.3 shall be considered.
Figure 4.13: Effective length of cantilever for carrying concentrated post loads
Austroads 2013
— 68 —
 For design case 1, the deck shall be designed to resist a moment 1.1 x Md in N.mm/mm, and
a tensile force 1.1 x T, in N/mm, calculated as follows (Equation 28 and Equation 29):
M p−post 28
Md =
(Wb + D )
Pp 29
T =
(Wb + D )
where
M p − post = flexural resistance of barrier post (N)
Pp = shear corresponding to M p − post (N)
Wb = width of base plate of the barrier post (mm)
distance from the outer edge of the base plate to the innermost of
D =
bolts (mm)
 For design case 2, the shear force, Pv and cantilevered moment, Md are taken as
(Equation 30 and Equation 31):
Pv X 30
Md =
b
Fv L 31
Pv =
Lv
in which b = 2X + Wb ≤ L
where
vertical force of vehicle laying on top of rail after impact forces Ft and
Fv =
FL are over (N)
L = post spacing (mm)
Lv = longitudinal distribution of vertical force Fv on top of barrier (mm)
distance from the outside edge of the post base plate to the section
X =
under investigation, as specified in Figure 4.13 (mm)
It should be noted that the yield strength of anchor bolts for steel barriers shall be fully developed
by bond, hooks, attachment to embedded plates, or any combination thereof.
Austroads 2013
— 69 —
4.6 Concrete Parapet and Metal Rail

The geometric requirements described in Section 4.5.1 shall also apply to traffic barriers with
concrete parapet and metal rails (Figure 4.14).
Standard barrier designs used by RMS and VicRoads are suggested. Refer to Section 5.2 for
details.

The resistance of each component of a combination parapet and rail system shall be determined
as specified in Section 4.4 and Section 4.5. The flexural strength of the rail, Rr shall be determined
over one span, and over two spans. The resistance of the post on top of the wall Pp, shall be
determined including the resistance of the anchor bolts or post.
The barrier resistance of the combination parapet and rail shall be taken as the lesser of the
resistances determined for two failure modes, including impacts at the mid-span of the rail
(Figure 4.14) and impacts at the post (Figure 4.15).
Figure 4.14: Concrete parapet and metal rail – impact at mid-span of rail
Austroads 2013
— 70 —
Figure 4.15: Concrete parapet and metal rail – impact at post

 Step 1: Calculate the ultimate capacities of the various components, including concrete
parapet and rail-and-post.
 Step 2: Calculate the resistance of the combination parapet and rail.
 Step 3: Check the adequacy of the barrier.
Impact at mid-span of the metal rail

Where the vehicle impact is at mid-span of the metal rail, the flexural resistance of the rail, Rr, and
the maximum strength of the concrete wall, Rw, shall be added together to determine the combined
resultant strength, R* and the effective height Y*, taken as (Equation 32 and Equation 33):
R * = Rr + Rw 32
Rr H r + Rw H w 33
Y* =
R*
where
Rr = ultimate capacity of the rail over one span (N), calculated by Equation 22
Rw = ultimate capacity of the wall (N), calculated by Equation 16

Y* = effective height of the barrier (mm)
Hw = height of wall (mm)
Hr = effective height of rails (mm)
Austroads 2013
— 71 —
Impact at a post
Where the vehicle impact is at a post, the post takes load until the plastic moment of the post is
reached. Beyond this point the remaining load has to be transferred by the deflection of post and
rails to the adjacent post.
The wall resistance R’w is the reduced resistance of the wall due to the load the post is taking. This
is because in the end, the post load has to be transferred to the concrete.
The maximum resultant strength R*, located at a height Y*, shall be taken as the sum of the post
capacity, Pp, the rail strength, R’r, and a reduced wall strength, R’w as follows (Equation 34,
Equation 35 and Equation 36):
R * = Pp + Rr' + Rw' 34
Pp H r + R r' H r + Rw' H w 35
Y* = *
R
in which Rw H w − Pp H r 36
Rw' =
Hw
where
Pp = ultimate transverse resistance of post (N)
Rr' = ultimate transverse resistance of rail over two spans (N)
Rw = ultimate transverse resistance of walls (N)
Rw' = capacity of wall, reduced to resist post load (N)
It should be noted that AASHTO (2012) defines Pp as the resistance of the post located Y* above
the deck; however, it should be located at (Hr – Hw) above the top of the parapet, and is calculated
as (Equation 37):
Mp 37
Pp =
H r − Hw
It should also be recognised that a maximum effective height, Y*, equal to the centroid rail height,
Hr, could be obtained, but at reduced resultant strength, R*, equal to the post capacity, Pp, and rail
capacity, R’r, only.
The total transverse resistance of a combination concrete wall and metal rail barrier R*, shall be
greater than or equal to the transverse containment loads specified in Table 4.1.
The analysis herein does not consider impacts near open joints in the concrete wall or parapet.
The metal rail will help distribute load across such joints. Improved rail resistance will be obtained if
the use of expansion joints is minimised.
For impact near the end of barrier segments, the resistance may be calculated as the sum of the
wall resistance, calculated using Equation 18, and the metal rail resistance over one span,
calculated using Equation 26.
Austroads 2013
— 72 —

The procedure described in Section 4.4.3 shall be applied.
4.7 Design Examples

4.7.1 Design Example 1: Concrete Parapet – Medium Performance Level
Referring to Table 4.1, for the design forces associated with a Medium performance level barrier:
 ultimate lateral containment force Ft = 500 kN
 vehicle contact length Lt = 2400 mm
 minimum effective height He = 1100 mm.
For a concrete parapet to satisfy the minimum effective height requirement, the parapet must have
a minimum height of 1100 mm.
Assume that a parapet-type barrier design shown in Figure 4.16 is considered. The yield strength
of steel and concrete materials is fy = 400 MPa and f’c = 32 MPa, respectively.
Figure 4.16: Barrier design example 1
Two impact cases need to be considered:

 impact within wall segment
 impact near end wall segment.
Step 1: Calculate the concrete capacities of the various components on the parapet
 Beam at top of wall:
In this example, the beam at the top of the parapet contributes to the capacity of the parapet.
The moment capacity of the beam Mb is calculated about the vertical beam axis Y-Y using
Equation 20 as follows:
 400 × 400 
M b = 0.8 × 400 × 400 × 200 × 1 − 0.6 ×  = 23.5 × 106 N.mm
 180 × 200 × 32 
Austroads 2013
— 73 —
 Flexural capacity of wall:

The wall flexural capacity Mw is calculated about the vertical axis of the wall Y-Y. It should be
noted that the portion of the top beam is not included in the calculation of Mw. For a slope
parapet with variable thicknesses along the height, Mw is approximately calculated by
dividing the parapet wall into a number of segments with constant thickness. In this example,
it was assumed that the flexural capacity of the wall was constant along its height.
For a unit height of the wall section (Equation 20):

 550 × 400 
Mw 0 = 0.8 × 550 × 400 × 150 × 1 − 0.6 ×  = 25.7 × 103 N.mm/mm
 1000 × 150 × 32 
For the total wall section: Mw = Mw0 x Hw, where Hw is the height of the wall excluding the top
beam portion.
Mw = 25.7 x 103 x (1100 – 180) = 23.6 x 106 N.mm
 Flexural resistance of cantilevered wall:

The flexural resistance of the cantilevered wall is the wall flexural capacity about the
longitudinal axis X-X per metre strip (Figure 4.16), using Equation 20:
 1000 × 400 
Mc = 0.8 × 1000 × 400 × 250 × 1 − 0.6 ×  = 77.6 × 103 N.mm/mm
 1000 × 250 × 32 
It should be noted here that As is taken on a strip of one metre.
Step 2: Calculate the critical length of yield line failure pattern, Lc:
 For impact within a wall segment (Equation 17):
 2400  8 × 1100(23.5 × 106 + 23.6 × 106 )

2
2400
Lc = +   + = 3804 mm
2  2  77.6 × 103
 For impact at end of wall segment (Equation 19):
 2400  1100(23.5 × 106 + 23.6 × 106 )

2
2400
Lc = +   + = 2651 mm
2  2  77.6 × 103
Step 3: Calculate the wall resistance, Rw:

 For impacts within a wall segment (Equation 16):
 2  77.6 × 103 × 3804 2 
Rw =   8 × 23.5 × 106 + 8 × 23.6 × 106 +  = 537 kN
 2 × 3804 − 2400  1100 
 For impacts at end of wall segment (Equation 18):

 2  77.6 × 103 × 26512 
Rw =   23.5 × 106 + 23.6 × 106 +  = 374 kN
 2 × 2651 − 2400  1100 
Austroads 2013
— 74 —
Step 4: Check the structural adequacy of the barrier

From the above calculations, the critical capacity of the barrier is 374 kN for impacts at the end of
wall segment. This value is less than the applied load of Ft = 500 kN, thus the barrier fails.
Results from a numerical analysis show that the requirement of Rw ≥ Ft was satisfied when the
vertical reinforcements are φ 16 mm at 150 mm spacings instead of φ 16 mm at 200 mm spacings.
Thus As = 1407 mm2/m. The calculations are presented below.
 For impacts at end of wall segment:

 1407 × 400 
Mc = 0.8 × 1407 × 400 × 250 × 1 − 0.6 ×  = 108 × 103 N.mm/mm
 1000 × 250 × 32 
 2400  1100(23.5 × 106 + 23.6 × 106 )

2
2400
Lc = +   + = 2586 mm
2  2  108 × 103
 2  108 × 103 × 2586 2 

Rw =   23.5 × 106 + 23.6 × 106 +  = 508 kN > Ft OK
 2 × 2586 − 2400  1100 
The effective height, Y*, in the case of a concrete parapet, is equal to the height of the wall:
Y* = 1100 mm ≥ H c OK
The barrier is now adequately designed.
4.7.2 Design Example 2: Post-and-rail Barrier – Regular Performance Level

Referring to Table 4.1, for the design forces associated with a Regular performance level barrier:
Assume that a post-and-rail barrier design shown in Figure 4.17 is considered.
Figure 4.17: Barrier design example 2
Austroads 2013
— 75 —
Step 1: Calculate the ultimate capacities of the post and the rail
Referring to standard section property tables:
150 x 100 x 6 mm RHS has a plastic modulus S = 134 x103 mm3
200 x 200 x 6 mm SHS has a plastic modulus S = 327 x103 mm3
For one rail (Equation 24),

M p = 0.9 × 350 × 134 × 103 = 42.2 × 106 N.mm
For the post (Equation 24),
M p − post = φ fy S
M p − post = 0.9 × 350 × 327 × 10 3 = 103.0 × 10 6 N.mm

The effective height of the barrier, Y* is calculated by Equation 27, with the assumption that only
the two top rails contribute to resisting the design load, in order to maximise the effective height of
the barrier (Figure 4.17):
42.2x10 6 (650 + 950 )
Y* = = 800 mm
2x42.2x106
The shear resistance of the post, Pp, is the shear force on a single post which corresponds to Mp of
the post and is located Y* above the deck and is calculated using Equation 25:
103.0 × 10 6
Pp = = 128.8 × 103 N
800
Step 2: Calculate the barrier resistance

The critical barrier nominal resistance shall be taken as the least value of the following, with an
assumption that the bottom rail does not contribute in resisting the design load:
 For failure modes involving an odd number of railing spans, N (Equation 22):
16 × 2 × 42.2 × 10 6 + 0
For N = 1: R * = = 276 kN
2 × 1× 3000 − 1100
16 × 2 × 42.2 × 106 + (3 − 1)(3 + 1) × 128.8 × 103 × 3000

For N = 3: R * = = 263 kN
2 × 3 × 3000 − 1100
16 × 2 × 42.2 × 106 + (5 − 1)(5 + 1) × 128.8 × 103 × 3000

For N = 5: R * = = 368 kN
2 × 5 × 3000 − 1100
 For failure modes involving an even number of railing spans, N (Equation 23):
16 × 2 × 42.2 × 106 + 22 × 128.8 × 103 × 3000
For N = 2: R * = = 266 kN
2 × 2 × 3000 − 1100
16 × 2 × 42.2 × 10 6 + 42 × 128.8 × 103 × 3000

For N = 4: R * = = 329 kN
2 × 4 × 3000 − 1100
16 × 2 × 42.2 × 106 + 62 × 128.8 × 103 × 3000

For N = 6: R * = = 437 kN
2 × 6 × 3000 − 1100
Austroads 2013
— 76 —
Step 3: Check the structural adequacy of the barrier

From the above cases, it is clear that the capacity of the barrier is controlled by the failure mode
involving three railing spans (N = 3) with a capacity of 263 kN. This meets the capacity requirement
(R* ≥ Ft = 250 kN).
Because the top two rails have equal capacity, the computed effective height can now be found
using Equation 27 as:
131.5 × 103 × 950 + 131.5 × 103 × 650
Y* = = 800 mm ≥ He OK
263
The barrier is now adequately designed.
4.7.3 Design Example 3: Combined Concrete Parapet and Metal Rail – Medium
Performance Level
Referring to Table 4.1, for the design forces associated with a Medium level barrier:
The barrier must have sufficient strength to resist the ultimate lateral containment force, 500 kN at
the required minimum effective height of 1100 mm.
Assume that a combined concrete parapet and metal rail barrier design shown in Figure 4.18 is
considered.
Figure 4.18: Barrier design example 3 – barrier cross-section
Austroads 2013
— 77 —
Step 1: Calculate the ultimate capacities of the various components

 Concrete parapet
For this example, the following material properties are used:
Yield strength of reinforcement fy = 400 MPa
Yield strength of concrete f 'c = 24 MPa
Strength reduction factor for concrete φc = 0.8
There is no concrete beam at the top of the wall, thus Mb = 0.
As the thickness of the concrete parapet varies along its height, it is assumed that the parapet is
divided into three portions as shown in Figure 4.19. Two possible failure mechanisms shown in
Figure 4.8 are considered.
Figure 4.19: Barrier design example 3 – parapet
The flexural resistance of cantilevered wall, Mc is calculated as the average of the values of Mc at
sections 1, 2 and 3 as follows:
Section 1: At top of wall Name Value Unit
Height of cross-section 300 mm
Cover thickness of vertical rebars 50 mm
Diameter of vertical rebars 16 mm
Spacing of vertical rebars b 200 mm
Cross-section area of 1 vertical rebar As 201 mm2
Height of top segment H1 590 mm
Effective height of cross-section d 242 mm
As fy
a= a 19.7 mm
0.85f'c b
Moment capacity of cantilevered wall at section 1:

φc As fy a
Mc = �d- � Mcs1 74 681 Nmm/mm
b 2
Austroads 2013
— 78 —
Section 2: At bottom of top slope Name Value Unit

As fy
a= a 19.7 mm
0.85f'c b
φc A fy a Mcs2 74 681 Nmm/mm

s
Mc = �d- �
b 2
Section 3: At bottom of wall Name Value Unit

Height of bottom segment H2 260 mm
As fy
a= a 30.8 mm
0.85f'c b
φc A fy a
Mc = s
�d- � Mcs3 205 887 Nmm/mm
b 2
Assuming failure mechanism 1, Mc is the average of the three sections:
Mcs1 +Mcs2 Mcs2 +Mcs3 1

Mc = � H1 + H2 � � � Mc1 94 747 Nmm/mm
2 2 H1 +H2
Assuming failure mechanism 2, Mc is the average of the top two sections:

Mcs1 +Mcs2
Mc = � � Mc2 74 681 Nmm/mm
2
The flexural capacity of the wall, Mw is calculated separately for positive bending and negative
bending for each portion. The total flexural capacity of the wall is the sum of the average capacities
of the number of wall portions in association with each failure mechanism.
Austroads 2013
— 79 —
Top portion Name Value Unit

Width at top section 300 mm
Width at bottom section 320 mm
Height of portion Hp 200 mm
Distance from outer fibre to center of tension rebar 76 mm
Diameter of longitudinal rebar 12 mm
Cross-section area of 1 rebar 113 mm2
Number of longitudinal rebars in positive moment 2
Number of longitudinal rebars in negative moment 2
Cross-section area of rebar
As 226 mm2
(same for both positive and negative)
Average effective height of wall section d 234.2 mm
As fy
a= a 22.2 mm
0.85f'c Hp
Moment capacity of wall Mw of top portion Mpositive = Mnegative

φc As fy a Mw 16 147 152 Nmm
Mpositive =Mnegative = �d- � Hp
Hp 2
Centre portion Name Value Unit

Distance from outer fibre to centre of rebar for positive
76 mm
moment
Cross-section area of rebars in positive moment As 226 mm2
Average effective height of wall section in positive moment d 244.2 mm
As fy
a= a 11.4 mm
0.85f'c Hp
Moment capacity of wall Mw in positive bending

φc A fy a Mpositive 17 261 972 Nmm
s
Mpositive = �d- � Hp
Hp 2
Distance from outer fibre to centre of rebars for negative
76 mm
moment
Austroads 2013
— 80 —
Cross-section area of rebars in negative moment As 226 mm2

As fy a 11.4 mm
a=
0.85f'c Hp
Moment capacity of wall Mw in negative bending
φc As fy a Mnegative 17 261 972 Nmm

Mnegative = �d- � Hp
Hp 2
Average moment capacity of middle portion =

Mave 17 261 972 Nmm
average (Mpositive, Mnegative)
Bottom portion Name Value Unit

Distance from outer fibre to center of rebar for positive
76 mm
moment
Cross-section area of rebars in positive moment As 226 mm2
As fy a 17.1 mm
a=
0.85f'c Hp
Moment capacity of wall Mw in positive bending

φc As fy a
Mpositive = �d- � Hp Mpositive 23 015 455 Nmm
Hp 2
Distance from outer fibre to centre of rebars for negative

76 mm
moment
Cross-section area of rebars in negative moment As 226 mm2
As fy
a= a 17.06 mm
0.85f'c Hp
Moment capacity of wall Mw in negative bending

φc As fy a Mnegative 23 015 455 Nmm
Mnegative = �d- � Hp
Hp 2
Austroads 2013
— 81 —
Average moment capacity of bottom portion = average

Mave 23 015 455 Nmm
(Mpositive, Mnegative)
For collision within a segment, the average of the positive and negative values of the wall’s flexible
capacity is used. This is acceptable because the yield line mechanism for this case will have some
positive moment hinges and some negative moment hinges.
For collision near expansion joints, the moment resistance that corresponds to a moment causing
tension along the inside face of the parapet, i.e. the positive moment, is used. This is required
because the only yield line to form is caused by a moment causing tension along the inside face.
The parapet combined capacity, Rw, is the smaller value of two mechanisms, calculated as follows.
Collision within a segment:
Failure mechanism 1 (all three portions) Mw 56 424 580 Nmm
Length of critical yield line (Equation 17) Lc 3 543 mm
Total transverse resistance of barrier (Equation 16) Rw 789 856 N
Failure mechanism 2 (two top portions) Mw 33 409 124 Nmm
Height of failure mechanism 2 H1 590 mm
Therefore, Rw = 780 kN for collision within a segment.
Collision near an end segment:

Failure mechanism 1
Mw is the sum of Mpositive in three portions Mw 56 424 580 Nmm
Failure mechanism 2
Height of failure mechanism 2 H1 590 mm
Mw is the sum of Mpositive in two top portions Mw 33 409 124 Nmm
Therefore, Rw = 578 kN for collision near an end segment.
 Metal rail-and-post
Referring to standard section property tables:
150 x 150 x 9 mm RHS has a plastic modulus, S = 248 x103 mm3
The rail and post have the same section, and the ultimate capacity is (Equation 24):
M p = 0.9 × 350 × 248 × 10 3 = 78.1× 10 6 N.mm
Austroads 2013
— 82 —
The resistance of the post, Pp is calculated using Equation 37 as:

78.1× 10 6
Pp = = 208 x 10 3 N
1225 − 850
For impact involving one railing span, N = 1 (Equation 22):
16 × 78.1× 10 6 + 0
Rr = = 347 200 N
2 × 1× 3000 − 2400
For impact involving two railing spans, N = 2 (Equation 23):
16 × 78.1× 10 6 + 2 2 × 208 × 10 3 × 3000
R'r = = 390 600 N
2 × 2 × 3000 − 2400
Step 2: Calculate the resistance of the combination parapet and rail

 For impact at mid-span of the metal rail (one span failure):
Resistance of the parapet Rw 780 868 N
Resistance of the metal rail over one span Rr 347 200 N
Resistance of combined barrier (Equation 32) R* 1 128 068 N
Check the resistance of barrier to transverse design load:

R* ≥ Ft? YES 500 000 N
Effective height of barrier (Equation 33) Y* 965 mm
Check the effective height of barrier:

Y* ≥ He? NO 1 100 mm
In this case the effective height of the barrier, Y*, does not meet the requirement of Y* ≥ He.
However, because the total barrier resistance is 1128 kN which is significantly greater than
the applied load of Ft = 500 kN, Y* can be increased by reducing the contribution of the
concrete parapet Rw with a corresponding reduction in the total lateral resistance, R*.
Try using 20% of Rw; the barrier’s total resistance and effective height are re-calculated as
follows:
Revised resistance of the metal rail over one span Rw 156 174 N
Revised resistance of combined barrier (Equation 32) R* 503 374 N

R* ≥ Ft? YES 500 000 N
Effective height of barrier (Equation 33) Y* 1 109 mm

Y* ≥ He? YES 1 100 mm
Austroads 2013
— 83 —
 For impact at a post (two span failure):

Resistance of the post Pp 208 320 N
Resistance of the metal rail over two spans R'r 390 600 N
Resistance of the parapet, reduced to resist the post load
R'w 480 642 N
(Equation 36)
Resistance of combined barrier (Equation 34) R* 1 079 562 N

R* ≥ Ft? YES 500 000 N

Y* ≥ He? NO 1 100 mm
Again, the effective height of the barrier, Y*, does not meet the requirement of Y* ≥ He.
However, because the barrier’s total resistance is significantly greater than the applied load
of Ft = 500 kN, Y* can be increased by reducing the contribution of the concrete parapet Rw
with a corresponding reduction in the total lateral resistance, R*.
Try using 75% of Rw; the barrier’s total resistance and effective height are re-calculated as
follows:
Revised resistance of the metal rail over one span Rw 585 651 N
Resistance of the parapet, reduced to resist the post load
R'w 285 425 N
(Equation 36)
Resistance of combined barrier (Equation 34) R* 884 345 N

R* ≥ Ft? YES 500 000 N

Y* ≥ He? YES 1 100 mm
From the above calculations, the barrier is adequately designed.
Austroads 2013
— 84 —
5 STANDARDISED BRIDGE BARRIER DESIGNS

5.1 Introduction
A collection of bridge barrier designs currently in use in various jurisdictions throughout Australia
and New Zealand has been collected and reviewed. This collection includes:
 standard barrier designs issued by road jurisdictions
 barrier designs used in actual projects.
It should be noted that not all road jurisdictions have issued standard designs for bridge barriers.
Some road jurisdictions rely on consultants/contractors to provide the designs on a
project-for-project basis.
Significant variations have been observed in these designs in barrier type, typical cross-section,
and structural details.
Although these designs have been classified by their performance level, there is no proven
evidence of conformity covering features such as structural capacity verification and crash test
results.
Based on the performance specifications, design requirements specified in AS 5100 (2007), and
the popularity of these designs, several barrier designs have been selected and standardised for
each performance level and material type. The selected barrier designs can be checked for
structural adequacy by the methodology presented in Section 4.
While the design guidelines for bridge deck slabs under barrier loads are provided, the actual
design of the bridge deck slab should be undertaken by the bridge design engineer.
5.2 Bridge Barrier Designs Currently in Use in Jurisdictions

The following tables list available bridge barrier designs obtained from various jurisdictions and
consultants/contractors. It should be noted that while this list was prepared based on available
information provided and with the intention to cover as far as possible all types and performance
levels of barriers, it is not necessary to address all available barrier designs used in Australia and
New Zealand. The list of current barrier types used for particular performance levels forms the
basis for selection of agreed barrier solutions to be used for developing standardised designs
(Section 5.3).
It should be noted that full details of the particular barrier type contained in these tables can be
obtained from the relevant jurisdiction.
Austroads 2013
— 85 —
5.2.1 Low Performance Level Barriers
Barrier type Indicative barrier profile# Barrier description

VRL1 Typical two rail low level containment barrier over bridge.
VicRoads Post system: galvanised steel 150 U.C. 37.2 x 1300 mm,
Low level steel 800 mm height above the finished surface; guardrail: steel
barrier over RHS; post bolted to deck overhang using post brackets.
bridges
VRL2 Typical single guardrail fence low level barrier over culvert.
VicRoads Double nested guardrail, 725 mm height above the finished
Low level steel level.
barrier over Bolted to bridge concrete parapet.
culverts
VRL3 Single guard fence low level barrier with pedestrian fence.
VicRoads Same as the low level barriers over culverts, except
Low level additional wire fencing is added.
special steel
barrier
Austroads 2013
— 86 —

WAL1 This barrier is used for maintaining existing bridge barriers
MRWA only.
Heritage Rail, Two rails plus top rail, post system; guardrail: steel RHS,
0530 - two rail top rail: steel SHS, post: steel SHS; rail spacing: 375 mm;
plus top rail post spacing: 2500 mm typical.
Expansion joints, transition, connection with roadside
barrier are provided.
Post embedded in concrete footings at approach.
Refer to MRWA standard drawing 0530–1344 and
associated drawings for details.

MRWA only.
Weak post (two Two rail and three rail barrier with post system; rail: steel
and three rail) RHS, post: steel SHS; rail spacing: 375 mm; post spacing:
barrier 3000 mm typical.
Expansion joints, transition, and end post are provided.
Posts are embedded in predefined holes on the parapet.

MRWA only.
Side mounted Thrie-beam rail and steel post; height of barrier 820 mm.
thrie-beam Posts are embedded in concrete footings at the bridge
barrier approach and post-connected to the parapet by bolt
connections.
Austroads 2013
— 87 —

SAL1 This barrier design has its independent foundations. It can
SA DPTI be installed in situations where the structure is too weak to
Single span take barrier loads. The system is based on tested Ohio
type of TL3 DoT barriers.
barriers
NZTAL1 Semi-rigid barrier system; post height 655 mm from base,

NZTA Low 2000 mm typical spacing.
level,
TL3 thrie-beam
barrier
# Details of the particular barrier type may be obtained from the relevant jurisdiction
Austroads 2013
— 88 —
5.2.2 Regular Performance Level Barriers

RMS Combined concrete parapet and two metal rail-and-post
RAO, RAI barrier; designed for both outer and inner traffic barriers
(RAO and RAI, respectively); maximum post spacing is
2700 mm; rails are RHS cross-section; M20 high strength
steel bolts grade 8.8; height of parapet 410 mm, height of
steel barrier 590 mm, total height 1000 mm.
The outer face of the concrete parapet must be vertical for
inner traffic barrier type RAI.
No provision for cyclists.
Termination: expansion joint, transition and end post
provided.

RBO, RBI barrier; designed for both outer and inner traffic barriers
(RBO and RBI, respectively); maximum post spacing is
2700 mm; the rail is RHS cross-section; M20 high strength
inner traffic barrier type RBI.
Provision for cyclists.
provided.
RMS Concrete parapet; designed for both outer and inner traffic
RCO, RCI barriers (RCO and RCI, respectively); total height 820 mm.
inner traffic barrier type RCI.
CHS grab rail for cyclists can be installed on top of these
barriers.
Austroads 2013
— 89 —

RMS Concrete parapet; designed for combined traffic/pedestrian
RD applications; total height 1300 mm.
VRR1 Typical three rail regular level barrier over bridges,

VicRoads 1200 mm height steel posts, bolted to 600 mm diameter
Three rail steel concrete column or retaining wall.
regular level Three RHS guardrails spacing 350–450 mm.
barrier
VRR2 Parapet without rail barrier on structure. Minimum 800 mm

VicRoads height above the finished level.
Concrete No reinforcement and anchorage details are available.
parapet
Austroads 2013
— 90 —

VRR3 Steel one rail-and-post, 500 mm height combined with
VicRoads 500 mm height concrete parapet.
Combined steel Steel SHS post and rail.
one rail-and- No reinforcement and anchorage details are available.
post and
concrete
parapet Regular
level barrier
VRR4 Same as the three rail steel regular level barrier above,
VicRoads except additional pedestrian rails are included.
Three rail steel
Regular level
special barrier
VRR5 Two rail steel barrier system bolted to top of parapet

VicRoads 800 mm height SHS posts.
Two rail steel Two RHS guardrails spacing at 350 mm.
Regular level
barrier
Austroads 2013
— 91 —

WAR1 Two rail bridge barrier system with steel posts; rails: steel
MRWA RHS or SHS, post: steel W section; post spacing: 2500 mm
Two rail type typical; barrier height: 755 mm.
(Regular Expansion joints, transition barriers and connection to
performance roadside barriers are provided.
level barrier) Post connected by embedded bolts on bridge concrete
parapet and driven to ground at approach.
Refer to MRWA standard drawing 0430–0760 and related
drawings for details.
WAR2 Three rail bridge barrier system with steel posts; rails: steel
Three rail type typical; barrier height: 1115 mm.
WAR3 Four rail bridge barrier system with steel posts; rails: steel
Four rail type typical; barrier height: 1345 mm.
Austroads 2013
— 92 —

WAR4 Thrie-beam rail and steel post bolted on concrete parapets;
MRWA total height of barrier 800 mm; post is 580 mm high.
thrie-beam Posts are embedded in concrete footings at the bridge
(Delaware approach and post-connected to the parapet by bolt
Regular connections. Refer to MRWA standard drawing 0530–1101
performance and related drawings for details.
level)
TMRR1 Precast concrete panel, 1100 mm height and 110 mm

TMR Precast thickness. Combined precast and cast-in situ parts form a
concrete panel single slope F-type barrier, 220 mm thickness at the top of
barrier.
Panel length is 2400 mm typical. Compressible filler and
backing rods are used for panel connections.
NZTAR1 Semi-rigid barrier system; post height 1078 mm from base,

NZTA Regular 2000 mm typical spacing.
level,
TL4 thrie-beam
with modified
blockout barrier
Austroads 2013
— 93 —
5.2.3 Medium Performance Level Barriers

MAO, MAI barrier; designed for both outer and inner traffic barriers
(MAO and MAI, respectively); maximum post spacing is
2700 mm; rails are RHS cross-section; M20 high strength
inner traffic barrier type MAI.
provided.
RMS Combined concrete parapet and one metal rail-and-post

MBO, MBI barrier; designed for both outer and inner traffic barriers
(MBO and MBI, respectively); maximum post spacing is
2700 mm; the rail is RHS cross-section; M20 high strength
inner traffic barrier type MBI.
provided.
RMS Concrete parapet; designed for both outer and inner traffic
MCO, MCI barrier; total height 1300 mm.
inner traffic barrier type MCI.
Austroads 2013
— 94 —

VRM1 Two rail steel barrier system bolted to top of parapet
VicRoads 800 mm height SHS posts.
Two rail steel Two RHS guardrails spacing at 350 mm.
Medium level Parapet: same as VRM2.
barrier
VRM2 Steel two rail-and-post, 800 mm height combined with

Combined steel Steel SHS post and rails.
two rail-and- Rail: SHS 150x150x9; post: SHS 200x200x9.
post and No reinforcement and anchorage details are available.
concrete
parapet
Medium level
barrier
VRM3 Steel rail-and-post, 450 mm height combined with 800 mm

VicRoads height concrete parapet.
Combined steel Steel SHS post and rail.
one rail-and- No reinforcement and anchorage details are available.
post and
concrete
parapet
Medium level
barrier
Austroads 2013
— 95 —

MRWA Similar to the RMS's MAO barrier.
MA Combined concrete parapet and two metal rail-and-post
Combination barrier; designed for both outer and inner traffic barrier;
Medium level maximum post spacing is 2700 mm; rails are RHS cross-
barrier section; M20 high strength steel bolts grade 8.8; height of
parapet 650 mm, height of steel barrier 650 mm, total
height 1300 mm.
inner traffic barrier type MAI.
provided.
TMRM1 Precast concrete panel, 1100 mm height and 110 mm

TMR Precast thickness. Combined precast and cast-in situ parts form a
concrete panel single slope F-type barrier, 220 mm thickness at the top of
barrier.
Panel length is 2400 mm typical. Compressible filler and
backing rods are used for panel connections.
Same as Regular performance level barriers.
SAM1 Three rail system; post spacing typically 3000 mm

SA DPTI maximum.
Post-and-rail
barrier
Austroads 2013
— 96 —

SAM2 1400 mm high, single slope combination barrier.
SA DPTI
Single slope
combination
barrier
NZTAM1 This is Texas HT80 barrier design.

NZTA
Combination
Medium level
barrier
5.2.4 High Performance Level Barriers

VRH1 Steel two rail-and-post, 1025 mm height combined with
two No reinforcement and anchorage details are available.
rail-and-post
and concrete
parapet High
level barrier
Austroads 2013
— 97 —

VRH2 Steel two rail-and-post, 725 mm height combined with
two No reinforcement and anchorage details are available.
rail-and-post
and concrete
parapet High
level barrier
VRH3 One steel rail-and-post, 450 mm height combined with

one No reinforcement and anchorage details are available.
rail-and-post
and concrete
parapet High
level barrier
SA DPTI 1500 mm high single slope combination barrier.

Combination,
Single slope
High level
barrier
Austroads 2013
— 98 —
5.2.5 Special Performance Level Barriers

TMRS1 Precast concrete panel, 1400 mm height and varying
TMR Concrete thickness to create a unique architectural form, 125 mm thick
F-type barrier on average. Combined precast and cast-in situ parts form a
single slope barrier, 220 mm thickness at the top of barrier.
Panel length is 2400 mm typical. Compressible connection
types are used. Precast section over wing-walls tapers from
1400 to 1100 high.

TMR Special thickness to create a unique architectural form, 125 mm thick
1.4 m on average. Combined precast and cast-in situ parts form a
single slope barrier, 220 mm thickness at the top of barrier.
1400 to 1100 high.

TMR Special thickness to create a unique architectural form, 125 mm thick
1.6 m on average. Combined precast and cast-in situ parts form a
single slope F-type barrier, 220 mm thickness at the top of
barrier.
1400 to 1100 high.
Austroads 2013
— 99 —

SA DPTI This is based on High level HL4 barrier (1500 mm high single
Single slope slope combination barrier), with higher design loads.
combination
barrier
Special level
5.3 Specific Design Requirements from Jurisdictions

A number of jurisdictions specify more specific requirements for the selection of bridge barrier
types for specific site situations and associated design requirements.
TMR’s Design criteria for bridges and other structures (Department of Transport and Main Roads
2012a) requires that:
 Steel (posts and rails) bridge traffic barrier shall be used on streams, creeks and rivers with a
flood immunity of less than ARI 2000 years to reduce afflux, if this is a design constraint.
 Concrete traffic barriers rather than steel barriers shall be used on overpass bridges to
prevent debris falling off the bridge onto the traffic below.
 Bridges on major roads and those on small horizontal radius curves may also be required to
have concrete barriers.
 For a design speed equal to or in excess of 80 km/h and a radius less than 400 m, bridges
shall have 1500 mm minimum height bridge barriers on the outside of the curve. For a design
speed less than 80 km/h and a radius less than 400 m, consideration shall be given to a
bridge barrier higher than 1100 mm for concrete parapets and/or 1000 mm for steel traffic
barrier to prevent overturning on the barrier and loss of side friction on the pavement. Special
consideration shall be given where the speed environment transitions from high speed to a
lower speed (for example, off ramps). In such cases the barrier shall be designed for the
higher speed.
In addition, overpass bridges crossing roads or railways have concrete barriers rather than steel
bridge traffic rails. This is to help prevent debris falling off the bridge onto traffic below. Bridges on
major roads and those on small horizontal radius curves may also be required to have concrete
barriers (Department of Transport and Main Roads 2011).
Austroads 2013
— 100 —
RMS’s direction is that traffic barriers with two metal rails and a short height concrete parapet are
likely to remain a standard traffic barrier type in NSW for the following reasons:
 Many bridges carry cyclists and the height of traffic barriers on such bridges in order to
contain cyclists must be at least 1300 mm in accordance with Clause 12.1(d) in AS 5100.1
(2004). If a short height concrete parapet with two metal rails is provided, it will reduce the
overall weight of the traffic barrier and consequently the dead load on the edge bridge
girders, which are often the most heavily loaded girders.
 It is often an urban design requirement to have short height concrete parapets in order to
maximise a commuter’s view from the bridge deck outwards. This is particularly true for long
bridges.
 It is also often an urban design requirement to have short height concrete parapets to
increase the slenderness and improve the aesthetics of a bridge when viewed in elevation.
MRWA specifies that (Main Roads Western Australia 2009) rigid concrete barriers must be
adequately anchored to the bridge deck by reinforcement and the deck must be able to sustain the
impact force without distress. Open rail-and-post barrier systems require continuity of the
longitudinal rails to be maintained across any connection.
MRWA’s directions for use of rigid barriers include (Main Roads Western Australia 2012a):
 The New Jersey barrier profile is no longer the preferred profile for use.
 Vertical face barriers are generally not favoured.
 F-shape profile is currently the best technology available for a concrete barrier.
 The Californian single slope barrier is an acceptable alternative that may be preferred where
road jurisdictions wish to provide for future surface overlays that will not affect the barrier
profile or require its replacement.
New Jersey concrete barriers have been used on major structures that carry high volume and high
speed traffic in high risk locations and in locations where the barrier deflections have to be
minimised, e.g. as a median barrier (Main Roads Western Australia 2009).
The post-and-rail barriers (two or three rails) are typically used along the edge of the bridge for
both aesthetic purposes and to maintain a view from the bridge.
Single slope and F-shape concrete barriers are used widely in South Australia for Regular and
Medium performance levels.
Austroads 2013
— 101 —
5.4 Selection of Bridge Barrier Concept Designs for Standardisation

Based on the current most popular bridge barrier designs and specific requirements of
jurisdictions, the designs listed in the following tables are selected to be used as concept designs
for standardisation. This list of concept designs was the result of discussions between the
members of the working group and was finalised at the BTTF meeting in December 2012.
Table 5.1: Agreed standard designs for Low performance level barriers
Original Agreed barrier Indicative barrier profile# Barrier description
barrier type name
VRL1 LL1 Typical two rail low level containment barrier over
VicRoads bridge. Post system: galvanised steel 150 U.C.
Low level steel 37.2 x 1300 mm, 800 mm height above the
barrier finished surface; guardrail: steel RHS.
Post bolted to deck overhang using post brackets.
WAL3 LL2 This barrier is used for maintaining existing bridge

MRWA barriers only.
Low level, Thrie-beam rail and steel post; height of barrier
Side mounted 820 mm.
thrie-beam Posts are embedded in concrete footings at the
barrier bridge approach and post-connected to the
parapet by bolt connections.
Notes:
 Proposed barrier names have been changed to update the agreed barrier types.
 TMR does not use Low performance level barrier on its bridges.
Table 5.2: Agreed standard designs for Regular performance level barriers
barrier type name
RMS RL1 Concrete parapet; designed for both outer and
RCO, RCI inner traffic barriers (RCO and RCI, respectively);
Regular level, total height 820 mm.
The outer face of the concrete parapet must be
F-shape vertical for inner traffic barrier type RCI.
concrete barrier CHS grab rail for cyclists can be installed on top of
these barriers.
Austroads 2013
— 102 —

barrier type name
RMS RL2 Concrete parapet; designed for combined
RD Regular traffic/pedestrian applications; total height
level, 1300 mm.
Concrete
barrier
VRR1 RL3 Typical three rail regular level barrier over bridges.
VicRoads 1200 mm height steel posts, bolted to 600 mm
Regular level, diameter concrete column or retaining wall.
Three RHS guardrails spacing 350–450 mm.
Three rail steel
barrier
NZTA Regular RL4 Semi-rigid barrier system; post height 1078 mm

level, from base, 2000 mm typical spacing.
TL4 thrie-beam
with modified
blockout barrier
RMS RL5 Combined concrete parapet and two metal rail-

RAO, RAI and-post barrier; designed for both outer and inner
Regular level, traffic barriers (RAO and RAI, respectively);
maximum post spacing is 2700 mm; rails are RHS
Combination cross-section; M20 high strength steel bolts grade
barrier 8.8; height of parapet 410 mm, height of steel
barrier 590 mm, total height 1000 mm.
vertical for inner traffic barrier type RAI.
No provision for cyclists.
Termination: expansion joint, transition and end
post provided.
Austroads 2013
— 103 —

barrier type name
RMS RL6 Combined concrete parapet and two metal rail-
RBO, RBI and-post barrier; designed for both outer and inner
Regular level, traffic barriers (RBO and RBI, respectively);
Combination maximum post spacing is 2700 mm; the rail is
barrier RHS cross-section; M20 high strength steel bolts
grade 8.8; height of parapet 650 mm, height of
vertical for inner traffic barrier type RBI.
post provided.
Notes:
 TMR does not use combination barriers on its bridges.
Table 5.3: Agreed standard designs for Medium performance level barriers
barrier type name
RMS ML1 Concrete parapet; designed for both outer and
MCO, MCI inner traffic barrier; total height 1300 mm.
Medium level, The outer face of the concrete parapet must be
vertical for inner traffic barrier type MCI.
F-shape Provision for cyclists.
concrete barrier
TMRM1 ML2 Precast concrete panel, 1100 mm height and

TMR Medium 110 mm thickness. Combined precast and cast-in
level, situ parts form a single slope F-type barrier,
Single slope, 220 mm thickness at the top of barrier.
Precast Panel length is 2400 mm typical. Compressible
concrete panel filler and backing rods are used for panel
connections.
Austroads 2013
— 104 —

barrier type name
DPTI Medium ML3 Three rail system; post spacing typically 3000 mm
level, Post and maximum.
rail barrier This design was proposed by SA during the
meeting 18 December 2012 and should be
reviewed before adopting.
RMS ML4 Combined concrete parapet and two metal

MAO, MAI rail-and-post barrier; designed for both outer and
inner traffic barriers (MAO and MAI, respectively);
maximum post spacing is 2700 mm; rails are RHS
cross-section; M20 high strength steel bolts grade
8.8; height of parapet 650 mm, height of steel
barrier 650 mm, total height 1300 mm.
vertical for inner traffic barrier type MAI.
post provided.
RMS ML5 Combined concrete parapet and one metal

MBO, MBI rail-and-post barrier; designed for both outer and
inner traffic barriers (MBO and MBI, respectively);
maximum post spacing is 2700 mm; the rail is
RHS cross-section; M20 high strength steel bolts
grade 8.8; height of parapet 820 mm, height of
vertical for inner traffic barrier type MBI.
post provided.
Austroads 2013
— 105 —

barrier type name
SA Medium ML6 1400 mm high, single slope combination barrier.
level, Single This design was proposed by SA during the
slope meeting 18 December 2012 and should be
combination reviewed before adopting.
barrier
Notes:
 NZTA uses Texas HT80 barrier.
Table 5.4: Agreed standard designs for High performance level barriers
barrier type name
High level, HL1 F-shape, to be developed based on Medium level F-shape, 1400 mm height.
F-shape concrete barrier ML1.
Concrete
barrier
High level, HL2 Single slope, to be developed based on Medium Single slope, 1400 mm height.
Single slope level concrete barrier ML2.
Concrete
barrier
VRH2 HL3 Steel two rail-and-post, 725 mm height combined

VicRoads with 800 mm height concrete parapet.
High level, Steel SHS post and rails.
Combination No reinforcement and anchorage details are
barrier available.
Consideration should be given to changing the
details of the concrete part to match with MR4,
MR5.
Austroads 2013
— 106 —

barrier type name
SA High level, HL4 1500 mm high single slope combination barrier.
Single slope This design was proposed by SA during the
combination meeting 18 December 2012 and should be
barrier reviewed before adopting.
Notes:
 Consideration should be given to changing the details of the concrete part to match with MR4, MR5.
Table 5.5: Agreed standard designs for Special performance level barriers
Original barrier Agreed barrier Indicative barrier profile# Barrier description
type name
TMRS1 SL1 Precast concrete panel, 1400 mm height and
TMR Special varying thickness to create a unique architectural
level form, 125 mm thick in average. Combined precast
Concrete Single and cast-in situ parts form a single slope barrier,
slope barrier 220 mm thickness at the top of barrier.
Panel length is 2400 mm typical. Compressible
connection types are used. Precast section over
wing-walls tapers from 1400 to 1100 high.
Austroads 2013
— 107 —
Original barrier Agreed barrier Indicative barrier profile# Barrier description

type name
TMR Special varying thickness to create a unique architectural
level form, 125 mm thick in average. Combined precast
Concrete Single and cast-in situ parts form a single slope barrier,
slope barrier 220 mm thickness at the top of barrier.

TMR Special 1.6 varying thickness to create a unique architectural
m form, 125 mm thick on average. Combined precast
and cast-in situ parts form a single slope F-type
barrier, 220 mm thickness at the top of barrier.
SA Special level, SL4 This is based on High level HL4 barrier (1500 mm
Single slope high single slope combination barrier), with higher
combination design loads.
barrier This design was proposed by SA during the
meeting 18 December 2012 and should be
reviewed before adopting.
# Details of the particular barrier type may be obtained from the relevant jurisdiction.
Notes:
 Barrier designs for Special performance level are provided to use as examples only.
 More examples should be developed for Special performance level combination barrier and F-type concrete barrier.
It has been agreed in the BTTF meeting in December 2012 that the detailed design of standard
barriers identified in this section would be developed after the revision of the AS 5100 is
completed, incorporating the changes in the design loads and required minimum barrier heights.
Austroads 2013
— 108 —
6 RETROFITTING OF EXISTING BRIDGE BARRIERS

6.1 Introduction
This section provides suggestions and guidance on the retrofitting of traffic barriers on existing
bridges, taking into account the design loads and barrier performance levels specified in
AS 5100.1 (2004).
This section covers the following:

 barrier retrofitting design procedure
 determination of appropriate barrier performance level
 barrier retrofit design
 special considerations for design exceptions and historical bridges.
It should be noted that it has been agreed in the BTTF meeting in December 2012 that a standard
set of designs for retrofitting existing bridges barriers is not possible due to the designs being
site-specific and highly variable.
6.2 Reasons for Retrofitting Existing Bridge Barriers

Since the release of AS 5100 in April 2004, the design of replacement traffic barriers on existing
bridges has been affected by the generally higher barrier design loads specified in this design
standard. In NSW, for example, most bridges built before 2005 have been designed using loads
significantly less than those specified in AS 5100.2 (2004) e.g. for the Regular performance level.
This situation is further complicated when the required barrier performance level is higher than the
Regular performance level (Roads and Traffic Authority 2009).
The following situations are generally considered for barrier retrofitting:

 where on-going vehicle collision costs and safety risks are excessive and a retrofit will
reduce the collision frequency, severity of damage or cost of repair
 where the bridge is being widened or rehabilitated (Texas Department of Transportation
2006), the barrier shall be brought up to current standards
 where improvements are made to any bridge or its approach roadway, the bridge barriers
shall be evaluated for conformance to current standards.
Other reasons that may be considered include:

 the need for separating pedestrian or cycle traffic from vehicle traffic for safety reasons due
to an increase in the traffic volume
 handrail systems that do not have sufficient strength to comply with the AS 5100.2 (2004)
load requirements
 kerbs that do not have sufficient strength to comply with the AS 5100 (2004) requirements
 kerbs that do not have sufficient height for vehicle re-direction
 kerbs that have a ‘flat’ face slope allowing easy vehicle mounting.
Bridge widening is typically considered in the following situations:

 widening an inadequate roadway width on an existing bridge
 adding lanes to a highway segment to increase the traffic carrying capacity
 adding an auxiliary lane across the bridge.
Austroads 2013
— 109 —
6.3 Barrier Retrofitting Design Methodology

6.3.1 Design Procedure
The barrier retrofitting design procedure will involve the following four-staged process:
1 Determine the ideal barrier performance level for the site using the procedure in Section 3.
2 Evaluate the existing bridge structure to determine the strength available and the capability
of the structure in accommodating the design loads associated with the ideal barrier
performance level without superstructure strengthening using Section 6.3.3.
3 If the ideal barrier performance level is higher than what can be accommodated by the
existing structure then determine the optimum achievable performance considering
superstructure strengthening options.
4 Design the barrier retrofitting which may be a new barrier design or a barrier upgrading
design and may include the retrofitting of the existing structure.
The barrier retrofitting design procedure in Figure 6.1 is suggested. Detailed steps are provided in
the following sections.
6.3.2 Determine Ideal Barrier Performance Level

The ideal barrier performance level for the site is the performance level required by the site-specific
condition and traffic data as for a new bridge construction. The selection of this barrier
performance level is not constrained by the structural condition of the existing bridge.
A site-specific risk analysis shall be carried out to determine the ideal barrier performance level for
the bridge site as per Section 3.
If the selected barrier performance level does not satisfy AS 5100, the actual level shall be noted in
the drawing’s title block, as a percentage of AS 5100.1 performance level, for example, ‘50% of
AS 5100.1 Low performance level’ (Department of Transport and Main Roads 2011).
6.3.3 Evaluate the Existing Bridge Structure

Evaluation requirements
Once the ideal barrier performance level has been selected the bridge needs to be evaluated to
determine if this barrier performance level is achievable. This consists of a review and calculations
to determine:
 the logistics of demolition of the old barrier and constructing of the ideal barrier on the
existing structure
 if the existing structure is adequate to carry the higher loads associated with the ideal barrier;
this needs to include an assessment of the
— transverse deck capacity
— girder capacity
— clearance for placement of the new barrier
— other relevant requirements.
Austroads 2013
— 110 —
Details of the slab and kerb reinforcement of the existing bridge shall be reviewed to determine if
the slab edge is capable of being retrofitted with an adequate new barrier. The following
information should be acquired to perform the deck evaluation:
 as-built plan of the whole structure including detailed design drawings of bridge’s
superstructure components
 a plot locating existing defects on the slab such as delaminations, spalls and cracks
 measurement of the depth of cover on the top mat of reinforcing steel on a grid pattern
 deck concrete compressive strength assessed through destructive testing of deck core
samples.
Figure 6.1: Barrier retrofitting design procedure
Evaluation methodology
It is assumed that there is a section of deck slab cantilever that needs to transfer the barrier
collision loads to the exterior beam. AS 5100 (2007) limits deck analysis methods to elastic plate
bending and the empirical method for traffic load rating and posting. However, due to the extreme
nature of barrier collisions and the low probability of occurrence, a simplified yield line method
based on AASHTO (2012) has been adopted (Section 4). Inherent in the use of yield line theory is
the assumption that the historic mild steel reinforcement used for bridge construction has sufficient
ductility to allow the yield line structural response to develop. Refer to Section 4.5 for
post-and-railing systems and Section 4.4 for barriers with a concrete parapet.
Austroads 2013
— 111 —
In addition, AS 5100.2 (2004) requires that the bridge barriers and supporting elements be
designed as a progressive strength system in which barriers and then their connections fail prior to
the failure of the bridge deck slab. Therefore a load factor of 1.1 should be used when evaluating
the deck slab cantilever section under the above impact load combinations.
It should be noted that if the bridge can be successfully retrofitted to the ideal barrier performance
level then a B/C analysis can be carried out to determine the economic feasibility of the upgrade.
However a Safe System approach should be employed where possible rather than decisions made
on economics alone (Section 2.4).
Load combinations
While barrier collision loads are known for a predefined barrier performance level, the specific
bridge loading at the time of the collision is less certain. The barrier impact loads and traffic loads
on the deck need not be applied simultaneously when designing the deck. For the purpose of
assessing the capacity of the existing deck due to barrier impacts, the following two load
combinations shall be considered:
 Case 1: permanent loads + the transverse and longitudinal impact forces.
 Case 2: permanent loads + the vertical impact forces.
The impact forces are taken from the barrier design loads for the performance level under
consideration (Section 4.2).
6.3.4 Rationalisation of Barrier Performance Level

From the results obtained by an evaluation presented in Section 6.3.3, in many situations it may
not be possible to upgrade an existing bridge to the ideal barrier performance level. Reasons for
this may include:
 insufficient capacity of the existing deck
 logistics of the demolition and/or the retrofit process
 incompatibility of the ideal barrier and the existing bridge
 high cost of retrofit (whole-of-life-cycle cost)
 remaining life of the bridge.
Where it is not viable to strengthen to the ideal level, a lower performance level may be justified by
using site-specific risk evaluation and B/C analysis. Section 6.4 provides a suggested process to
determine if a lower performance level is appropriate for the particular site. A Safe System
approach should also be taken into consideration.
6.3.5 Design Methodology

The following design principles are suggested:
 Yield line failure analysis method adopted from AASHTO (2012) will be used.
 The retrofitted barrier system should allow for progressive collapse, i.e. the failure of the
barrier system should occur in the following order: barrier, anchorage then deck slab.
 The retrofit design should match the structural components of the existing structure, including
splice locations, as far as practical.
Austroads 2013
— 112 —
6.4 Determination of Appropriate Barrier Performance Level

Where it is not viable to install the ideal performance level barrier for the site, as determined in
Section 6.3.3, the procedure described in the following sections is suggested to determine the
most suitable barrier retrofit design solution.
6.4.1 Site-specific Risk Assessment

The following typical hazards of existing bridge barriers should be assessed to provide additional
information for the determination of the most suitable barrier retrofitting design solution (Roads and
Traffic Authority 2009):
 unprotected rigid barrier (e.g. barrier ending on an end post that does not have any guardrail
attached on the approach side)
 elements that may spear vehicles
 discontinuities in horizontal rails
 protruding kerbs
 inadequate barrier height
 barrier too brittle/fragile (e.g. lightly reinforced concrete rails, timber barriers, etc.)
 barrier too flexible (e.g. chain mesh)
 inadequate guardrail connection
 inadequate guardrail anchorage
 approach barrier too flexible
 insufficient stiffness transition
 lack of containment.
6.4.2 Benefit-cost Analysis

A B/C analysis can be carried out to select the most cost-effective option for the site.
The following options may be considered:

1 do nothing
2 strengthen the existing bridge to accommodate new barrier loads
3 reduce the required barrier performance level
4 reduce site hazards including speed reduction
5 build a new bridge.
It is possible to combine options (b) and (c) or (d) above.
The following should be taken into consideration (Roads and Traffic Authority 2009):
 If the bridge cannot be economically strengthened to install a traffic barrier to the barrier
performance level specified in Section 3 then the highest barrier performance level
achievable within the economic constraints applicable to the site shall be adopted.
 If one of the barrier performance levels specified in Section 3 can be provided, then, as a
minimum, the hazards identified in Section 6.4.1 or as otherwise determined shall be
minimised.
Austroads 2013
— 113 —
 Where the proposed barrier performance level is less than that required from the risk
assessment, then alternative amelioration strategies shall be considered including reduced
speed limits or advance warning signs.
 Alternatives to upgrading barriers on large culverts include widening the culvert and
continuing the flexible barrier system on the approaches across the culvert.
 The resulting barrier design solution shall take into account any adverse impact on
vulnerable road users such as cyclists or pedestrians.
 Barrier designs must not introduce or allow the continued existence of the barrier hazards
identified.
 The information gathering, risk assessment, design, decision-making and approval as well as
as-built drawings shall be recorded.
6.4.3 Additional Considerations

The following additional situations should be taken into account when selecting a barrier retrofitting
solution (Department of Transport and Main Roads 2012a).
Retrofitting of bridge balustrades

Balustrades are normally not installed on the roadside of a footpath/bikeway on new works. A
traffic barrier plus safety rail should be used in these applications. However, historically,
balustrades may have been installed between the road and footpath to stop pedestrians and
cyclists falling off the footpath onto the road where there is a large vertical drop to the road or
where the footpath is narrow. Vehicle impact of balustrades poses a hazard to road users and
pedestrians.
On existing bridges with balustrades next to the road, the balustrade may be allowed to remain in
place if the speed limit is 70 km/h or less and commercial vehicles are less than 5% of the traffic
volume. In all other situations, balustrades shall be removed and replaced by conforming barriers.
Retrofitting barriers on widened bridges

The barrier on the widened side must conform to current design requirements. The barrier on the
unwidened side would have been designed to the design criteria current at the time of design. The
following action shall be undertaken:
 For bridges with a repeated accident history of vehicle impacts or vehicles overturning, the
barriers shall be replaced with barriers conforming to AS 5100 (2007).
 Where AS 5100.1 (2004) would currently require a performance level higher than Regular,
the bridge must be modified to provide a higher level of protection. Exemptions based on
impractical or uneconomic criteria shall be accepted by the relevant road jurisdiction.
 Existing barriers with little strength (concrete post and water pipe rails and concrete posts
with balustrade) should as a minimum be replaced by a steel bridge rail utilising the full
existing structural capacity of the deck, taking into consideration the requirement that the
minimum strength for replacement rails is 50% of the Low performance level.
 Barriers on bridges that do not conform to any performance level specified in
AS 5100.1 (2004) shall be replaced with conforming barriers. Exemptions based on
impractical or uneconomic criteria shall be approved by the relevant road jurisdiction.
 For bridges not satisfying the above conditions, the barriers on the bridges may remain in
place.
Austroads 2013
— 114 —
6.5 Barrier Retrofit Design Procedure

As part of the planning for a bridge that will be widened or rehabilitated, the following aspects of
the project shall be considered in the selection of a retrofit barrier:
 characteristics of the bridge location
 design for upgrading or replacement of existing barrier
 design for retrofitting existing structure if required
 design of connections
 details of approach barriers.
The details of potential retrofit designs, such as the following, and their effect on selection of a
retrofit barrier should be reviewed carefully:
 placement or spacing of new anchor bolts or dowels
 reinforcement anchorage
 approach guard fence post positioning
 shoulder width required by the new barrier.
6.5.1 Characteristics of the Bridge Location

Evaluate details of the location, such as the following, and consider their effect on selection of a
retrofit barrier:
 bridge structure’s height above lower terrain or waterway
 approach roadway width, alignment, and grade
 position of adjacent streets and their average daily traffic
 bridge design speed, average daily traffic, and percentage of heavy vehicles
 accident history on the bridge.
The effect of a full-strength retrofit on the shoulder width of the bridge should be evaluated to
ensure that a reduction in effective shoulder width or in sight distances at adjacent intersections
will not increase the accident rate. The following should also be considered:
 bridge width, alignment, and grade
 type, aesthetics, and strength of existing barrier
 bridge length and its potential for posting speed limits.
6.5.2 Design for Upgrading Existing Barriers

Typical treatments for various types of existing deficient barriers as provided in VicRoads (2005)
are adopted in this section as examples for the design for upgrading existing bridge barriers.
Additional modifications shall be introduced to existing bridge barriers where applicable to fulfil the
new performance levels. The existing deficient barriers include standard timber handrails for timber
and concrete bridges, pipe handrails, and reinforced concrete handrails. Due to the limited height
of the existing barriers, the upgraded performance levels can be of a Low or a Regular level at
best.
Austroads 2013
— 115 —
Standard timber handrail for timber bridges

The option for timber structures is essentially a guardrail acting as a tensile rubbing rail, based on
the strong beam – weak post principle. The new guardrail shown in Figure 6.2 replaces the existing
timber rails, thereby reducing snagging and spearing aspects of the existing railing, while providing
tensile continuity for redirecting light vehicles.
This arrangement incorporates additional timber posts midway between the existing posts, thereby
strengthening the barrier and improving its re-directional capacity. Two extra-large washers are
required for each bolt through timber, to minimise the potential for punching through the timber.
The capacity of the guardrail can be marginally improved, by replacing the existing undersized
timber posts (152 mm x 102 mm) with larger 175 mm x 125 mm hardwood sections, and increasing
the size of connecting bolts.
The performance level for this arrangement due to its low height is Low. However, with the full
replacement of the timber posts it is considered that NCHRP TL-3 light vehicles under 2 tonnes,
impacting at standard crash conditions (namely 100 km/h and 25° incidence angle) will be
redirected by this barrier.
Source: VicRoads (2005).
Figure 6.2: Standard timber handrail for timber bridges
Timber handrail for concrete bridges

For this handrail treatment the guardrail is bolted to steel block-outs, which in turn are bolted to
new steel posts (Figure 6.3). The steel posts are fixed to the kerbs by grouting bolts in position. At
the existing timber post positions, the new parallel flange channel (PFC) posts are fixed to the
existing bolts through square hollow section (SHS) packers. Anchorage capacity is improved by
incorporating 76 mm x 76 mm x 6.3 mm steel packers, welded to the PFC posts.
The performance level for this arrangement due to its low height does not match the Regular
performance level but will redirect TL-3 light vehicles under two tonnes. The barrier can be
effectively turned into a Low performance level by using the following members:
 Sections consist of 178 mm x 76 mm x 6 mm Grade 300 standard cold formed steel channel.
 Posts can be at 1.25 m maximum spacing with a single guardrail.
Austroads 2013
— 116 —
Figure 6.3: Timber handrail for concrete bridges
Pipe handrail
The proposal shown in Figure 6.4 essentially provides a rubbing guardrail, to avoid the potential for
snagging of the existing system and improve vehicle redirection. The guardrail is further
strengthened by fixing it to separate posts, at 1.0 to 1.25 metre centres.
It is possible to upgrade to the Regular performance level by adopting a marginally modified

arrangement of that shown in Figure 6.10 or replacing the existing barrier.
Figure 6.4: Pipe handrail
Reinforced concrete handrail type 1

The option shown in Figure 6.5 is basically a continuous rubbing guardrail, fixed to the concrete
posts through SHS steel block-outs. The access hole in the steel SHS is positioned on the
departure side of the traffic. This hole is required for tightening the guardrail nut.
Austroads 2013
— 117 —
The single guardrail is weak in bending and may be strengthened by doubling up the guardrail, or
providing additional intermediate posts (Figure 6.6). The barrier can also be replaced with details
shown in Figure 6.11 to fulfil the Low performance level.
Figure 6.5: Reinforced concrete handrail type 1 – details at concrete post
Reinforced concrete handrail type 2

This proposal is similar to the Figure 6.5 rubbing guardrail treatment, but allows for wider (180 mm)
PFC block-outs, which in turn are bolted to the concrete posts through hardwood packers, as
shown in Figure 6.7. The system has been previously installed without packers, which are
considered optional.
This system allows for partial removal of the precast kerb sections, with a resultant increase in
trafficable deck width of approximately 160 mm.
Low performance level alternatives to this arrangement would be to leave the kerb in position and
replace the barrier with the barrier shown in Figure 6.12 (with reduced width block-outs), or to
replace both the barrier and kerbs by the barrier shown in Figure 6.13. It is also possible to
upgrade to the Regular performance level by adopting the arrangement shown in Figure 6.14 or
replacing the existing barrier.
Austroads 2013
— 118 —
Figure 6.6: Reinforced concrete handrail type 1 – intermediate post details
Figure 6.7: Reinforced concrete handrail type 2
Proprietary bridge barrier retrofitting system

Where removal and replacement of the concrete barrier is not considered a viable option and spot
repair may have a questionable life-span, some proprietary bridge barrier retrofitting systems may
be used to retrofit the existing barrier. One example is the Transpo's Visi-Barrier™ precast polymer
concrete panels shown in Figure 6.8.
Austroads 2013
— 119 —
Source: http://www.transpo.com/Visi-Barrier.html, accessed 27 March 2012.
Figure 6.8: Proprietary bridge barrier retrofitting system
6.5.3 Design for Replacement of Existing Barrier

A barrier replacement option shall be used where the existing barrier is not structurally or
economically effective to carry the new design loads.
The following barrier replacement options for timber bridges and various configurations of concrete
bridge decks are presented as examples (VicRoads 2005).
Timber bridges
For timber bridges, timber look-alike barriers can be fabricated from steel hollow sections to meet a
Low performance level (see Figure 6.9 for an example).
Figure 6.9: Casterton-Edenhope Rd over Glenelg River
Austroads 2013
— 120 —
Concrete bridges
The barrier replacement solutions shown in Figure 6.10 to Figure 6.13 can be utilised for concrete
bridges with narrow, intermediate, wide and wider kerbs, respectively, to comply with the Low
performance level.
Figure 6.10: Replacement option for concrete bridge decks, narrow kerbs
Figure 6.11: Replacement option for concrete bridge decks, intermediate kerb widths
Austroads 2013
— 121 —
Figure 6.12: Replacement option for concrete bridge decks, wide kerbs
Figure 6.13: Replacement option for concrete bridge decks, deck or footway wider than 650 mm
Austroads 2013
— 122 —
Figure 6.14 presents a barrier replacement option for concrete bridges to comply with the Regular
performance level.
Figure 6.14: Replacement option for concrete bridge decks, for Regular performance level
6.5.4 Retrofitting Existing Structure

Retrofitting of the existing structure, if required, should be based on the results from a detailed
structural evaluation as presented in Section 6.3.3 for the selected barrier performance level.
Possible bridge components needing to be retrofitted when retrofitting the barrier include
cantilevered deck slab and external longitudinal girders.
Strengthening options for the deck slab may include:

 externally bonded fibre reinforced polymer (FRP) lamination
 post-tensioning
 additional cross-section (cross beam, slab thickness)
 deck slab replacement.
An appropriate strengthening option should be selected based on site-specific structural

conditions.
Austroads 2013
— 123 —
6.5.5 Design of Connections

Upgrading existing barriers
Bolt connections are popularly used for the connection between new attachments with the
components of the existing barriers or with the deck slab.
Barrier replacement
The connection of the new barrier with the existing slab shall be designed so that it is capable of
carrying the impact loads. Typical types of connection include:
 Bolted connection: the bolts can be embedded in the existing concrete slab or can be an
expansion type. This connection type can be used for steel post-and-rail barrier types.
 Cast-in-place connection: the steel reinforcement of the new barrier is welded to the existing
steel bars of the slab. This connection type is used for concrete parapet barrier types.
The barrier anchorage should be designed using a progressive strength system approach.
6.5.6 Retrofitting of Bridge Approach Barriers

The bridge approach barrier will be considered for upgrade in two situations:
 where a bridge barrier is retrofitted and the roadside barrier is of a flexible type or lower
performance level
 where a roadway guard fence is upgraded but the existing bridge barrier will remain in place,
customised bridge barrier to guard fence transitions may be required.
Refer to Section 7 for guidance on bridge approach barrier design.
6.6 Bridge Barrier Upgrade Exceptions

Rarely but occasionally, an upgrade to a barrier on an existing structure could degrade rather than
improve bridge safety. A possible design exception should be considered when planning a
bridge-widening or rehabilitation project in which raising the barrier to meet current standards will
reduce overall safety. Factors to consider include the following:
 design speed of traffic in the bridge location
 resistance to impact of the existing barrier
 whether the bridge ends are intersections protected by stop signs or traffic lights
 whether the geometry is straight into, along, and out of the bridge
 whether traffic on the bridge is one-way or two-way
 accident history on the bridge, including damage to and repairs of the barrier
 risk of falling over the side of the bridge
 whether the bridge has a lower roadway
 whether a barrier upgrade will further narrow an already narrow lane width.
Austroads 2013
— 124 —
6.7 Special Considerations for Historic Bridges

Federal law protects historically significant bridges, and any rehabilitation or improvement projects
on them require special attention. Bridges that are designated historic and that are listed or eligible
to be listed in the National Heritage List introduce a special design challenge because the
appearance of the bridge may be protected even though the historic barrier may not meet current
standards.
The original barrier on a historic bridge is unlikely to meet current design requirements, particularly
the structural capacity and barrier height requirements.
Options for upgrading the barrier on historic bridges usually include the following:
 introduce an approved traffic barrier alongside the existing barrier, leaving the existing barrier
in place. This is sometimes appropriate when a pedestrian walkway exists on or is planned
for the bridge
 replace the existing barrier with an acceptable approved barrier
 remove the current barrier and incorporate it into a new acceptable barrier. This may be
appropriate in rare instances where an existing barrier is especially decorative
 design a special barrier to match the appearance of the existing barrier.
Austroads 2013
— 125 —
7 BRIDGE APPROACH BARRIERS

7.1 Introduction
In an incident of a vehicle crashing into a roadside barrier on a bridge approach, severe
decelerations and the consequent high risk of injury can occur if the transition from a flexible
approach barrier to a rigid bridge barrier is not properly detailed and transitioned in strength and
stiffness. Such incidents represent a potentially major risk to the occupants of these vehicles.
In addition, the road section in the proximity of the bridge may have the same risks as the bridge
section itself, particularly in the instances where the area beneath the bridge is a major roadway or
railway, highly developed area, deep waterway or similar feature. Such situations may represent a
high risk to third-party persons and property if, for example, an errant heavy vehicle or high
occupancy vehicle, such as a bus, impacts the bridge approach area.
AS 5100.1 (2004) requires that a transition barrier be provided on the approach to all bridge traffic
barriers. A number of requirements have been specified for the design of bridge approach barriers,
however, detailed guidelines for the determination of the following factors need to be provided:
 the required length of need (LON) of bridge approach barriers
 the length of each performance level provision within the LON
 the specification of the transition between different performance level barriers
 the specification of end treatments.
The objective of the guidelines contained in this document is to specify bridge approach barriers
that will results in a reduction in the severity of crashes on bridge approaches and to achieve
rational uniformity of standards. The methodology used in the guidelines is primary based on the
AS 5100 provisions for bridge approach barriers. The guidelines provide a process for the selection
of multiple performance level bridge approach barrier systems and considerations relevant to the
determination of appropriate extents for each performance level. They cover barriers on
approaches to bridges on major freeways and divided highways where the use of rigid concrete
bridge barriers is most prevalent. They also cover barriers on approaches to bridges on less
heavily trafficked roads where bridge barriers commonly comprise lower performance levels and
more flexible steel post-and-rail systems.
7.2 Current Design and Specification Requirements

7.2.1 Terminology
The following terminologies are used in this guideline and are taken from Austroads (2010b):
 Clear zone (CZ) is the area adjacent to the traffic lane that should be kept free from features
that would be potentially hazardous to errant vehicles. The clear zone is a compromise
between the recovery area for every errant vehicle, the cost of providing that area and the
probability of an errant vehicle encountering a hazard. The clear zone should be kept free of
non-frangible hazards where economically and environmentally possible. Alternatively,
hazards within the clear zone should be treated to make them safe or be shielded by a safety
barrier.
 Gore area is a triangular area where two roads either meet or split.
Austroads 2013
— 126 —
 Length of need is the length of a road safety barrier system, excluding leading or trailing
terminals, needed to prevent an errant vehicle from impacting a hazard, and represents the
length over which a road safety barrier will re-direct an impacting errant vehicle. It is the
distance between the leading and trailing points of need – effectively the length of the road
safety barrier less terminals.
 Point of need is the start or end of the LON, defining the length over which an errant vehicle
is redirected by the road safety barrier and would otherwise strike the hazard if a road safety
barrier was not provided.
 Right of way (ROW) width is the amount of land reserved or used for transport purposes.
The methodology described in the guidelines may be used for determining bridge approach
performance levels and extents for the following cases:
 twin bridges on divided carriageways, where there is a median between the bridges
 a single bridge on a divided carriageway, where there is continuous median barrier along the
roadway on both sides of the bridge
 a single bridge on an undivided carriageway, where there is no median.
7.2.2 Current AS 5100 Requirements

AS 5100.1 (2004) specifies that an appropriate transition of compatible stiffness shall be provided
for bridge approach barriers.
Transition barriers are specified for bridge approaches with provisions for strength, stiffness,
exposed surfaces, and end treatments. Strength and stiffness shall vary to provide a transition
between any flexible roadside barrier and the rigid or semi-rigid bridge traffic barrier.
A smooth face and tensile continuity shall be maintained throughout. Exposed rail ends, posts and
sharp changes in the geometry of the barrier components, kerbs, and the like, shall be avoided or
transitioned out with a maximum taper of 1 in 10 for the barrier components, and a maximum taper
of 1 in 20 for kerb discontinuities (Clause 10.6.3, AS 5100.1–2004).
It is specified in AS 5100.1 (2004) that the selection of a barrier performance level for bridge
approaches and departures be based on the same procedure as for the bridge barriers. The length
of each performance level provision shall take into account local factors including, but not limited
to, the following:
 the distance and clearance to the ROW boundary as it may affect the risk to occupants of
errant vehicles and third parties
 the distance to hazards, including rigid objects and steep descents, as it may affect the risk
to occupants of errant vehicles and third parties
 the risk associated with the use of the crossing underneath the bridge and the proximity of
that crossing
 the risk associated with the existence of service roads or parallel walkways and the like.
There is, however, no detailed guidance for determining the LON of bridge approach barriers and
the length of each performance level barrier on the bridge approach.
Austroads 2013
— 127 —
7.2.3 Additional Requirements from Jurisdictions’ Guidelines

VicRoads
BTN 2009/002 (VicRoads 2009b) in conjunction with RDN 06–07 (VicRoads 2010) provides
guidelines for bridge approach barriers. These guidelines expand the provisions of AS 5100.1
(2004) for the design of bridge approach barrier systems for bridges on arterial road networks, on
major freeways and divided highways, and on less heavily trafficked roads. The following
requirements are specified in addition to the AS 5100.1 (2004) requirements:
 For a High performance level bridge barrier, a series of bridge approach barrier systems
transitioning progressively from a roadside barrier to Regular, Medium and then High
performance level approach barrier systems shall be provided.
 In the case of a Medium performance level bridge barrier, the transition from the roadside
barrier shall include Regular and Medium performance level approach barrier systems.
 For a Regular performance level bridge barrier, the transition may simply include a local
stiffening and transition section joining the bridge barrier and the roadside barrier.
 In the case of rigid barriers, the extended length shall be sufficient to contain errant vehicles
at the representative 15° divergence angle from the nearest traffic lane in order to provide a
minimum protected width of clear zone at the bridge. In the case of non-rigid barriers, the
length shall be determined by using a nominal 3° divergence angle (Figure 7.1).
Road and Maritime Services, NSW

RMS provides standard drawings for two types of transition barriers as below (RMS’s standard
drawings B505 and B506):
 W-beam/thrie-beam connection to rigid barrier (of Regular or Medium performance levels) or
bridge end post using thrie-beam and reduced post spacing
 thrie-beam connection to rigid barrier or bridge end post and reduced post spacing.
It should be noted that RMS’s Supplement to Austroads guide to road design – Part 6: Roadside
design, safety and barriers (Roads and Maritime Services 2011) specifies the RMS’s departure
from the Austroads guide (Austroads 2010a) by using the guidelines presented in Roads and
Traffic Authority (1996).
Department of Transport and Main Roads, QLD

For bridge approach barriers, the requirements specified in AS 5100.1 (2004) are stated in more
detail. These include (Department of Transport and Main Roads 2012a):
 There is no snagging hazard to stop the errant vehicle moving between the two types of
barriers.
 The entire length of the road barrier, transition and bridge barrier must be smooth and
continuous to support an errant vehicle’s passage along the carriageway.
 The stiffness of the barrier must ensure pocketing does not occur. Bridge transition barriers
are detailed in TMR standard drawings (Department of Transport and Main Roads 2012b).
 The maximum slope on a transition in height shall be no steeper than 1:10.
 The approach rail shall not direct an errant vehicle into a hazard on the bridge, e.g. kerb.
Similarly the exit shall not direct a vehicle onto a hazard. The layout must also address
crossover crashes. The approach barrier shall also stop an errant vehicle from travelling
down embankments, into creeks or onto a road beneath the bridge.
Austroads 2013
— 128 —
Note: RDG Road design guidelines; CZ Clear Zone; LON Length of Need; --/--/-- right of way boundary.
Source: VicRoads (2009b).
Figure 7.1: Terminology used for determining performance levels and extents
For the transition section, the following criteria shall apply (Department of Main Roads 2005):
 The connection point of the two systems should have adequate strength to ensure the
connection will not fail on impact by pulling out.
 The possibility of snagging an errant vehicle should be minimised.
 The bridge railing end behind the approach transition should be tapered.
 The length of the transition should be appropriate to minimise any significant changes in
deflection. It is suggested to be 10 – 12 times the difference in the lateral deflection of the
two barrier systems.
 A high degree of continuity is required for the change in stiffness from the less rigid barrier to
the more rigid barrier.
For road/rail interface barriers, TMR technical specifications MCE-SR-007 (Department of Main
Roads and Queensland Rail 2010) shall be applied.
Various types of end treatments are specified in Department of Main Roads (2005), including
MELT, SKT, FLEAT, Brakemaster, QuadGuard, React, 350, Sand Filled Barrels, TRACC, TAU II,
QuadTrend, and thrie-beam Bullnose. The selection of an end treatment depends on speed limit
and barrier application.
Austroads 2013
— 129 —
New Zealand Transport Agency

Transit New Zealand (2004) specifies a length of approach transition of not less than 31.8 m
unless justified by a detailed assessment of the LON.
Other jurisdictions’ guidelines

Other jurisdictions’ guidelines such as Main Roads Western Australia (2012a), and Department of
Infrastructure, Energy and Resources (2007) follow AS 5100.1 (2004) design requirements for
bridge approach barriers with supplement requirements specified in AS/NZS 3845 (1999) and
AASHTO Roadside design guide (AASHTO 2011) for the transition sections. A number of standard
drawings of the transition between typical types of barriers and barrier end treatments are also
provided in these documents.
7.3 Design Procedure

The following design steps are suggested:
 determine the LON for the bridge approach barrier
 select appropriate performance levels for different section along the required LON
 determine transitions between different barrier types
 select end treatments
 determine structural solutions.
These steps are presented in detail in the following sections.
7.4 Determination of Length of Need

The LON of a roadside barrier to protect vehicles from a roadside hazard can be determined from
the likely trajectory of a vehicle that leaves the road in the vicinity of the roadside hazard.
Figure 7.2 and Figure 7.3 illustrate the LON of a barrier that is used to protect a roadside hazard
from impact from a single direction and two directions of travel, respectively.
Figure 7.2: Barrier LON – single direction
Austroads 2013
— 130 —
Figure 7.3: Barrier LON – two directions
There are two geometric methods used to determine the LON, including the ‘run-out length’ and
‘angle of departure’ method. It is indicated in Austroads (2010a) that, compared to the angle of
departure method, the run-out length method is likely to result in a longer LON for large widths
requiring shielding and a shorter LON for narrow widths. Both methods are acceptable for use in
Australia; however, it is agreed by most jurisdictions that the angle of departure method be applied
(see Appendix E).
7.4.1 Run-out Length Method

The run-out length (LR) is shown in Table 7.1 and is the length of clear run-out area that should be
made available as a passageway for deceleration between the start of the barrier and a
non-bypassable hazard. It is the theoretical distance needed for a vehicle that has left the roadway
to come to a stop. It is measured from the upstream extent of the obstruction along the roadway to
the point at which a vehicle is assumed to leave the roadway.
Table 7.1: Run-out lengths for barrier design

Design speed (km/h) Run-out length LR (m) for AADT range
> 10 000 5 000–10 000 1 000–5 000 < 1 000
110 110 101 88 76
100 91 76 64 61
80 70 58 49 46
60 49 40 34 30
50 34 27 24 21
Straight section of road

For installation where the barrier is flared (Equation 38):
b
LA + L – L2
X= a 1
b LA 38
+
a LR
Austroads 2013
— 131 —
For parallel installations that have no flare (Equation 39):

LA – L2
X= 39
LA
LR
The lateral offset, Y, from the edge of the running lane to the beginning of the LON may be
calculated from (Equation 40):
LA
Y = LA – X 40
LR
where
X = the required LON in advance of the area of concern (hazard)
Y = lateral distance from edge of traffic lane to point of need
L2 = run-out length (Table 7.1)
b
= flare rate (Table 7.2)
a
LA = lateral extent of the area of concern
L1 = tangent length of the barrier upstream from the area of concern
LR = barrier’s lateral distance from the edge of the traffic lane.
Table 7.2: Flare rate

Design speed (km/h) For barrier within the shy For rigid barrier outside For non-rigid barrier
line offset the shy line offset outside the shy line offset
110 30:1 20:1 15:1
100 26:1 18:1 14:1
90 24:1 16:1 12:1
80 21:1 14:1 11:1
70 18:1 12:1 10:1
60 16:1 10:1 8:1
50 13:1 8:1 7:1
Note: Shy line is the distance from the edge of the traffic lane beyond which a roadside object will not be perceived as an obstacle and result in motorists changing
their behaviour. The shy line offset values are shown in Table 7.3.
Table 7.3: Shy line offset values

Design speed (km/h) Shy line offset (m)
50 1.1
60 1.4
70 1.7
80 2.0
90 2.2
100 2.4
110 2.8
See Figure 7.4 for the illustration of these parameters.
Austroads 2013
— 132 —
It should be noted that flared barriers should be used where practical, e.g. where the batter slope is
relatively flat so that the barrier can be located a lateral distance from the road shoulder.
Figure 7.4: Run-out length method of determining LON
Curved sections of road

The LON formula is applicable only to straight sections of road. For barrier designs on the outside
of horizontal curves, it is assumed that a vehicle's exit path from the road will follow a tangential
run-out path if the area outside the roadway is flat and traversable. Therefore, rather than using the
theoretical LR distance to determine the barrier LON, a line from the outside edge of the hazard (or
the clear zone for a continuous non-traversable feature) to a tangent point on the curve should be
used to determine the appropriate LON for the barrier (Figure 7.5).
The barrier length then becomes a function of the distance it is located from the edge of the driving
lane and can most readily be obtained graphically by scaling (AASHTO 2011). Depending on the
radius of the curve, a flare may not be required on the barrier but a properly designed and
installed, crashworthy end treatment will be required.
Austroads 2013
— 133 —
Figure 7.5: Length of need on outside curve using run-out length method
For barrier designs on the inside of curves, the LON is based on the length of run-out (LR)
projected from the edge of the traffic lane to the rear of the hazard (Figure 7.6). This is based on
the premise that a vehicle leaving the road in advance or at the departure point will be able to stop
before reaching the hazard or pass to the rear of it. The various possible vehicle trajectories
beyond this departure point will be shielded from the hazard.
Figure 7.6: Length of need on inside curve using run-out length method
The following points should be noted for determining the LON for bridge approach barriers using
the run-out length method:
 The lateral extent of the area of concern, LA, used in the run-out length method is taken as
the clear zone width for a bridge approach case, since the area underneath the bridge
represents a continuous non-traversable feature. As a result, this method is applied to a point
in the roadside a lateral distance from the bridge end post equal to the clear zone
(Austroads 2010a).
 The clear zone distances from the edge of the through travelled way can be derived from
Table 7.4 based on design speed, design AADT and batter slope.
 The barrier’s lateral distance from the edge of the running lane, L2, is normally the shoulder
width, since the bridge approach barrier is aligned with the bridge barrier.
 The tangent length of the barrier upstream from the area of concern (from the bridge barrier’s
end post in the case of the bridge approach), L1 (Figure 7.4), is chosen by the designer on
case-by-case basis. For the situation where a semi-rigid railing is connected to a rigid barrier,
it is suggested (AASHTO 2011) that this length needs to be at least as long as the transition
Austroads 2013
— 134 —
section. This measure reduces the possibility of pocketing at the transition and increases the
likelihood of smooth redirection if the barrier is struck immediately adjacent to the rigid
barrier.
 There is typically a section of off-structure barrier which has the same performance level as
that of the bridge barrier. L1 can be taken as the length of this section.
 For single-direction carriageways, the trailing point of need is at the end point of the bridge
barrier (e.g. end of the abutment’s wing wall); therefore the bridge approach barrier on the
bridge departure end may include transition section(s) connecting to the roadside barrier or
to a terminal section.
 For two-direction carriageways, the lengths of approach barriers required on both bridge
approaches are the same if the sign-posted speed limits are the same for both directions.
Table 7.4: Clear zone distances from edge of through travelled way
Design speed Design ADT Clear zone width (m) Steeper than
(km/h) Fill batter Cut batter 3:1(3)
6:1 to flat 4:1 to < 6:1 Steeper 6:1 to flat 4:1 to < 6:1 4:1 to 3:1
than 4:1(2)
≤ 60 < 750 3.0 3.0 (2) 3.0 3.0 3.0 (3)
750-1500 3.5 4.5 (2) 3.5 3.5 3.5 (3)
1501-6000 4.5 5.0 (2) 4.5 4.5 4.5 (3)
> 6000 5.0 5.5 (2) 5.0 5.0 5.0 (3)
70–80 < 750 3.5 4.5 (2) 3.5 3.0 3.0 (3)
750-1500 5.0 6.0 (2) 5.0 4.5 3.5 (3)
1501-6000 5.5 8.0 (2) 5.5 5.0 4.5 (3)
> 6000 6.5 8.5 (2) 6.5 6.0 5.0 (3)
90 < 750 4.5 5.5 (2) 3.5 3.5 3.0 (3)
750-1500 5.5 7.5 (2) 5.5 5.0 3.5 (3)
1501–6000 6.5 9.0 (2) 6.5 5.5 5.0 (3)
> 6000 7.5 10.0(1) (2) 7.5 6.5 5.5 (3)
100 < 750 5.5 7.5 (2) 5.0 4.5 3.5 (3)
750–1500 7.5 10.0(1) (2) 6.5 5.5 4.5 (3)
1501–6000 9.0 12.0(1) (2) 8.0 6.5 5.5 (3)
> 6000 10.0(1) 13.5(1) (2) 8.5 8.0 6.5 (3)
110 < 750 6.0 8.0 (2) 5.0 5.0 3.5 (3)
750–1500 8.0 11.0(1) (2) 6.5 6.0 5.0 (3)
1501–6000 10.0(1) 13.0(1) (2) 8.5 7.5 6.0 (3)
> 6000 10.5(1) 14.0(1) (2) 9.0 9.0 7.5 (3)
1 Where a site-specific investigation indicates a high probability of continuing crashes, or such occurrences are indicated by crash history, the designer may
provide clear zone distances greater than the clear zone shown in Table 7.4. A jurisdiction may limit clear zones to 9 m for practicality and to provide a consistent
roadway template if previous experience with similar projects or designs indicates satisfactory performance.
2 For fill batters steeper than 4:1 the batter width shall be treated as non-recoverable and not be considered as part of the clear zone. If a clear zone is to be
provided then:
 Providing that the embankment is not considered hazardous then the clear zone can be provided by the recoverable area at the top and bottom of the
embankment. If this summation is equal to or greater than the required clear zone for the appropriate slopes of these areas then the clear zone is satisfied.
 If the embankment is hazardous, then unless the embankment is offset a distance equal to the clear zone for the appropriate slope from the edge of the
travelled way to the embankment it is within the clear zone.
3 No clear zone widths are provided for cut batters steeper than 3:1. Therefore unless an appropriate clear zone is provided prior to the cut batter it shall be treated
as being within the clear zone. The cut batter and any objects contained on it shall be assessed.
Notes:
 The design ADT in Table 7.4 is the average daily traffic volume in both directions and in all lanes, other than for divided roads where it is the total traffic in all
lanes in one direction. In selecting the traffic to be used for the assessment of the clear zone a 20 year timeframe and allowance for growth over this period shall
be considered.
 Where the road is curved the values in Table 7.4 should be adjusted by the curve adjustment factors in Table 7.5.
Source: Adapted from AASHTO (2011).
Austroads 2013
— 135 —
Table 7.5: Curve adjustment factors

Radius Design speed (km/h)
(m)
60 70 80 90 100 110
900 1.1 1.1 1.1 1.2 1.2 1.2
700 1.1 1.1 1.2 1.2 1.2 1.3
600 1.1 1.2 1.2 1.2 1.3 1.4
500 1.1 1.2 1.2 1.3 1.3 1.4
450 1.2 1.2 1.3 1.3 1.4 1.5
400 1.2 1.2 1.3 1.3 1.4 –
350 1.2 1.2 1.3 1.4 1.5 –
300 1.2 1.3 1.4 1.5 1.5 –
250 1.3 1.3 1.4 1.5 – –
200 1.3 1.4 1.5 – – –
150 1.4 1.5 – – – –
100 1.5 – – – – –
7.4.2 Angle of Departure Method

In the angle of departure method, the length of barrier required is determined from angles at which
vehicles are assumed to leave the road.
Straight sections of road

The angle of departure of vehicles leaving the road varies over a range of values. In this method
vehicle trajectories are plotted based on angles at which most vehicles are likely to depart from the
traffic lane, in order to establish the barrier LON. This method is illustrated in Figure 7.7. The angle
of departure can be derived from Table 7.6 based on the posted speed limit.
Austroads 2013
— 136 —
Source: Roads and Traffic Authority (1996).
Figure 7.7: Angle of departure method of determining LON
Table 7.6: Angles of departure from the road

Signposted speed limit 15th percentile angle (1:X) used as 85th percentile angle (1:X) used as
(km/h) leading angle ‘a’ (Figure 7.7) leading angle ‘b’ (Figure 7.7)
60/70 5.7° (1:10) 22° (1:2.5)
80/90 3.8° (1:15) 22° (1:2.5)
100/110 2.9° (1:20) 22° (1:2.5)
Austroads 2013
— 137 —
In Figure 7.7 (a), the points of need for a two-direction carriageway can be established by the
following steps:
1 Identify the first possible point of contact with the hazard for direction 1 (the lane adjacent to
the barrier) (Point A).
2 Using the impact angles from Table 7.6, project a line at the leading impact angle until it
intersects the offset line of the barrier (Point B).
3 Record this as the leading point of need.
4 Identify the first possible point of contact with the hazard for direction 2 (the opposite lane)
(Point C).
5 Using the impact angles from Table 7.6, project a line at the leading impact angle until it
intersects the offset line of the barrier (Point D). Record this as the trailing point of need.
From these points establish the LON of the barrier installation.
It should be noted that, for this case, the posted speed limits may be different in the two directions.
In Figure 7.7 (b), the method to establish points of need for a single-direction carriageway includes:
1 Identify the first possible point of contact with the hazard (Point A).
2 Using the impact angles from Table 7.6 (as a leading angle), project a line at the leading
impact angle until it intersects the offset line of the barrier (Point B).
3 Record this as the leading point of need.
4 Identify the last possible point of contact with the hazard (Point C).
5 Using the impact angles from Table 7.6 (as a trailing angle), project a line at the trailing
impact angle until it intersects the offset line of the barrier (Point D). Record this as the
trailing point of need.
If the hazard is located on a median and is in the area of interest (i.e. clear zone distance) for the
opposing direction of traffic, repeat the process for the opposing direction of traffic.
Curved sections of road

When determining the leading point of need for a road safety barrier, the angle of departure of an
errant vehicle should be taken from the outer edge of the travel lane in all cases. Working back
from the obstacle will give the same result if the lane/road alignment is straight, but when the
alignment is curved, the leading and trailing angles of departure should be determined from a
tangent on the outside of the edge of the travel lane.
For a curve, the leading angle of departure from Table 7.6 (2.9° for speeds equal to or greater than
100 km/h) is taken of a tangent to determine where the initial point of need lies when this angle
meets with the back of a hazard that is located within the clear zone. The trailing angle of
departure at 22° is then taken from a tangent in front of the hazard to determine the final point of
need for a one-way road. Figure 7.8 (a) to (d) illustrates the situations for hazards on the outside
and inside of a curve, and for two-way and one-way carriageways.
In determining the LON for a road safety barrier, there is a range of angles of departure that are
considered between the leading angle of 2.9° (at 100 km/h) and the trailing angle of 22° (for all
speeds). These are general limits and when applied in cases where the leading angle from
Table 7.6 does not meet with the hazard, a departure angle that is somewhere between the
leading and trailing limits must be considered.
Austroads 2013
— 138 —
On the inside of a horizontal curve, a slightly different procedure is required if the leading angle of
departure does not meet with the back of the hazard (i.e. the line passes through or in front of the
hazard), and as a consequence the initial point of need for the road safety barrier does not relate to
the rear of the hazard. However, the leading and trailing angles cover a range and an angle within
these limits can be used as a leading angle for establishing the initial point of need. Therefore, in
these situations a chord to the curve should be drawn across the back of the hazard, square to the
centre of the curve. This process is illustrated in Figure 7.9 (a) and (b) for two-way and one-way
carriageways. The chord should be extended to intersect with the edge of travel lanes at point A
and B. Point A is where the leading angle of departure begins for traffic in the lane adjacent to the
hazard, and B is the corresponding point for the opposing traffic. The leading angle of departure is
the angle between the chord and the tangent to the curve at A. It can be calculated and will be
somewhere in the range of 2.9° to 22° for a speed limit of 100 km/h or greater.
The following points should be noted for determining the LON for bridge approach barriers using
the angle of departure method:
 The first possible point of contact with the hazard can be taken as the point in the roadside a
lateral distance from the bridge end post equal to the clear zone (Austroads 2010a).
 From Table 7.6, the angle of departure used to determine the trailing point of need is 22° for
all ranges of posted speed limits.
 For single-direction carriageways, the trailing point of need is at the end point of the bridge
barrier (e.g. end of the abutment’s wing wall), therefore the bridge approach barrier on the
bridge departure end may include transition section(s) connecting to the roadside barrier or
to a terminal section.
 For carriageways with traffic in two directions, the lengths of approach barrier required on
both bridge approaches are the same if the sign-posted speed limits are the same for both
directions.
Austroads 2013
— 139 —
Figure 7.8: Angle of departure method on curves where leading angle meets the rear of hazard
Austroads 2013
— 140 —
Figure 7.9: Angle of departure method where leading angle does not meet the rear of hazard
7.5 Selection of Barrier Performance Level

7.5.1 Methodology
The method developed in Section 3 is utilised to determine the appropriate bridge barrier
performance level at successive cross-sections along the bridge approach, taking into account the
AS 5100.1 (2004) requirements as summarised in Section 7.2.2.
For bridge sites with CAT1 site conditions, no barrier on the bridge or bridge approach is required.
For bridge sites with CAT2 site conditions, since the maximum possible performance level of the
bridge barrier is the Medium performance level, only a transition section between the bridge barrier
and the roadside barrier is required. The minimum length of roadside barrier is determined based
on the LON.
For bridge sites with CAT3 site conditions, the following principles are suggested:
 The performance level of the first section on the approach adjacent to the bridge is the same
as that of the barrier on the structure (bridge barrier). The length of this segment shall be
determined based on the consideration given in Section 7.5.2.
Austroads 2013
— 141 —
 Subsequently, the LON determined in Section 7.4 is divided into successive cross-sections
along the road centreline, from the bridge deck end or approach slab end (if provided). The
distance between the sections is determined based on considerations given in Section 7.5.3
and Section 7.5.4.
 The bridge approach barriers are considered off-structure barriers. Therefore the barrier
foundation should be taken into consideration.
It should be noted that, in existing urban areas or when city side streets and/or footpaths prevent
installation of the approach barrier system, consideration should be given to (Colosimo 2011):
 extending the bridge rail and or approach barrier around the intersection corner in such a
manner as to prevent encroachment of a vehicle onto any highway/terrain system below the
bridge
 providing traffic-control measures such as speed-restriction or intersection signs
 providing run-off recovery areas
 if the approach barrier is connected to a roadside barrier, it should be continuous with the
bridge approach barrier system and only a transition from a flexible (or semi-rigid) to a rigid
barrier system is required.
The assessments proposed for bridge sites with CAT3 site conditions are presented in the
following sections.
7.5.2 Performance Level and Associated Length Required to Contain Errant Vehicles
Section 7.2.2 provides the basis for undertaking a risk-based assessment, in conjunction with B/C
considerations, to determine the risk and consequences of a range of vehicles penetrating or
vaulting the bridge and bridge approach barriers at different cross-sections, and thus determining
the appropriate performance level at each cross-section.
These assessments shall consider the relative probabilities and risks of different errant vehicles,
relevant to the site, penetrating or vaulting:
1 the bridge barrier and landing on the area beneath the bridge, which may be a major road or
railway, a highly developed area or deep water or other high risk environment
2 the bridge approach barrier immediately before the bridge and also potentially ending up
within the area beneath the bridge
3 the bridge approach barrier and rolling over a vertical retaining wall or steep descent, or
impacting a rigid object or encroaching on the ROW boundary or similar, as local factors
given in Section 7.2.2
4 the bridge approach barrier at other locations.
7.5.3 Considerations to Preventing Errant Vehicles Entering High Risk Areas behind
Approach Barriers
These assessments shall also consider the relative probabilities and risks of different errant
commercial vehicles, leaving the trafficked roadway at the divergence angle associated with the
flare rate from the left-hand traffic lane, bypassing the end of the rigid bridge approach system and
travelling a sufficient distance to encroach on the:
1 area beyond the clear zone
2 area beyond the ROW boundary
Austroads 2013
— 142 —
3 continuation of the area beneath the bridge, particularly when this is a major roadway or
railway, high-use land area, deep water or other high-risk environment.
Consideration shall be given to the probable stopping distance of the errant vehicle.
Site-specific consideration must also be given to the possibility and the consequences of the driver
of a heavy vehicle falling asleep or becoming incapacitated and failing to brake after leaving the
roadway.
Consideration may also be given to constructing a secondary protection system to safeguard

against an errant vehicle entering the high-risk area under the bridge.
If there are additional factors which necessitate a long length of rigid barrier, it may be appropriate
to curtail the length of the rigid roadside barrier and to incorporate a secondary barrier in front of
the ROW boundary.
The total length of bridge approach barrier shall be determined by using either the run-out length
method (Section 7.4.1) or the angle of departure method (Section 7.4.2) in order to provide a
protected width of CZ as a minimum at the bridge. An additional length of barrier may be required
on flat or near flat terrain due to the greater probability and consequences of a vehicle travelling
across the terrain and behind the barrier towards the hazard.
7.5.4 Considerations Relevant to Minimum Lengths of Different Performance Level

Bridge Approach Barriers
In addition to the methodology specified above for determining the minimum allowable
performance level for bridge approach barriers at successive cross-sections along the bridge
approach and departure, there are practical minimum lengths of each performance level that need
to be considered.
In the case of sites requiring a High performance level bridge barrier, where there is normally a
very high risk to third parties if a heavy vehicle penetrates or vaults the barrier and enters the area
beneath bridge, it is essential that a site-specific risk assessment be made of the minimum length
of High performance level rigid or semi-rigid barrier to be provided immediately prior to the bridge
to provide a safe transition to the bridge including:
1 maintaining any errant heavy vehicle and its freight upright and redirecting it safely onto and
over the bridge
2 preventing any errant heavy vehicle from penetrating or vaulting the barrier at a cross-section
sufficiently close to the bridge that it would potentially have sufficient remaining momentum
for part or all of the vehicle or its heavy freight to continue into the area beneath the bridge.
The above minimum length of High performance level bridge approach barrier shall generally be
preceded by a length of Medium performance level bridge approach barrier or the High
performance level barrier may be extended over the length of warrant for the High and Medium
performance level barriers and transition immediately to a Regular performance level barrier
(Figure 7.20).
Given the above considerations, the minimum length of a High performance level barrier prior to a
Medium performance level bridge barrier shall be determined by site-specific risk assessment as
per Section 3.
Austroads 2013
— 143 —
It should be noted that typical minimum length of barrier needs to relate to the typical
manufacturing length of precast barriers to facilitate easy replacement. Minimum lengths also need
to consider the typical length required for a gradual change of stiffness in the transition.
Similar site-specific considerations shall be given to the approaches to bridges requiring Medium
performance level bridge barriers (Figure 7.21).
In the case of a bridge with semi-rigid Regular performance level bridge barrier, a non-rigid bridge
approach barrier may be transitioned over a relatively short length using an appropriate local
increase in strength, stiffness and height.
7.5.5 Methodology for Determining Performance Levels and Associated Length of

Median Bridge Approach Barrier Systems
Median width greater than the clear zone distance
If the median width behind the extended rigid approach barrier is more than the clear zone
distance (CZ), the rigid barrier length shall be sufficient to contain errant vehicles at the
representative 15° divergence angle from the nearest traffic lane, to protect as a minimum the CZ
width between the bridges.
The total barrier length including the non-rigid barrier shall be the LON determined from the
nearest traffic lane to protect a width equal to CZ between the bridges.
Median width less than the clear zone distance

If the median width behind the extended rigid bridge barrier is less than the CZ, the same
requirement as for the case when the median width is greater than the clear zone distance shall
apply, with the additional requirement that the extended rigid barrier shall be double-sided from its
approach end along the length exposed by a 15° divergence angle from the nearest point of the
opposite carriageway at the end of the bridge barrier.
7.6 Typical Layouts

The typical layouts of bridge approach barriers in Table 7.7 are suggested for various situations.
Table 7.7: VicRoads typical layouts of bridge approach barriers
Drawing Description Features

VicRoads Left verge on steep terrain for Medium and Non-flared extended bridge barrier includes a min. 80 m segment and a
SD 4521 High performance level barriers (Figure 7.10). guard fence transition segment. Actual length depends on height difference,
extent of steep terrain and road curvature.
VicRoads Left verge on flat terrain for Medium and High Single sided barrier is extended from bridge barriers where trucks shall be
SD 4531A performance level barriers (Figure 7.11). contained. The bridge approach barrier includes a min. 20 m segment flared
at 1:20, and a min. 30 m segment flared at 1:30 installed independently with
the WRSB.
VicRoads Guard fence layouts for bridge approach Layouts for three typical situations are included:
SD 4071 (Figure 7.12).  full shoulder on structure
 30 m long guard fence narrow shoulder on approach
 narrow shoulder on structure.
Austroads 2013
— 144 —
Source: VicRoads standard drawing SD 4521.
Figure 7.10: Bridge approach barriers for Medium and High performance levels – left verge on steep terrain
Austroads 2013
— 145 —
Figure 7.11: Bridge approach barriers for Medium and High performance levels – left verge on flat terrain
Austroads 2013
— 146 —
Figure 7.12: Typical layouts of guard fence on bridge approach
Austroads 2013
— 147 —
7.7 Transition Between Different Performance Level Barriers

7.7.1 General
Transitions are used to provide a safe interface whenever it is necessary to change from one type
of barrier to another.
There are three broad types of barrier: flexible, semi-rigid and rigid barriers. In general they have
the following features:
 flexible barriers comprise of tensioned wire ropes
 semi-rigid barriers comprise of horizontal steel beams (e.g. thrie-beam or W-beam)
 rigid barriers comprise of concrete components.
A satisfactory interface may be achieved by:

 providing a structurally designed and tested physical connection between the systems. The
connections are facilitated through transition sections of barrier that are designed to provide
gradually increasing lateral stiffness and hence continuity of protection for vehicles that
impact the barrier in the vicinity of the interface. Transitions can be used only between
semi-rigid systems (i.e. steel to steel) or between semi-rigid and rigid systems (i.e. steel to
concrete)
 overlapping the barriers by commencing the more rigid system behind the less rigid system.
The purpose of a transition section is to produce a gradual increase in stiffening between the
barrier systems so that vehicular pocketing, snagging or penetrations are prevented at any position
along the transition. The overlapping of the barriers achieves a similar outcome by providing
adequate lateral separation between them.
In practice transitions are achieved by:

 increasing the rigidity of a W-beam system by
— decreasing the post spacing
— nesting one rail behind another
— using another steel section behind the W-beam
 using a heavier rail for the transition (e.g. thrie-beam).
Overlapping different types of barrier is only possible where adequate space is available to
accommodate deflections. This may be used for any systems but is the only way of achieving a
transition from wire rope barrier to a more rigid barrier.
Specially designed barrier sections or connections are used for situations where W-beam is to be
connected to thrie-beam, or where either of these semi-rigid barriers is to be connected to a rigid
barrier (such as on the approach to bridges that have rigid barriers).
7.7.2 Design Criteria – Physically Connected Barriers

Several criteria are important when designing a transition section or connection. It is suggested
that the principles below be followed where any semi-rigid barrier system is connected to a rigid
barrier (AASHTO 2011).
 The connection point of the two systems must be as strong as the approach barrier to ensure
that the connection will not fail on impact by pulling out. The use of a cast-in-place anchor or
through-bolt connections is recommended.
Austroads 2013
— 148 —
 The transition must be designed to minimise the likelihood of snagging an errant vehicle,
including one from the opposing lane on a two-way facility.
 When providing a transition section to a bridge railing end it is recommended to taper the
bridge railing end behind the approach transition to prevent pocketing on vehicle impact.
 The transition should be long enough to ensure that changes in deflection do not occur over
a short distance. The change in stiffness from the less rigid barrier to the more rigid barrier,
over the transition length, should increase with a high degree of continuity. This may be
achieved by reducing the post spacing, strengthening the rail element or a combination of
these techniques.
 As with longitudinal barriers, kerb and slope features must be addressed. The slope between
the edge of the road and the barrier should not be steeper than 10:1.
 Drainage features such as kerbs, kerb inlets, raised inlets or open drains should not be
constructed in front of barriers or a transition area, as they may initiate vehicle instability and
adversely affect the crashworthiness of the barrier or transition.
7.7.3 Typical Interfaces Between Barrier Types

General
AS/NZS 3845 (1999) provides detailed illustrations of transitions between semi-rigid and rigid
barriers. These transitions have been tested or are deemed to be acceptable with respect to
NCHRP Report 350 (Ross et al. 1993). The transitions are achieved through stiffening of the steel
safety barrier by the use of special sections and connectors, reduced post spacing and nesting (i.e.
two sections of rail, one inside the other) of the beams.
Wire rope safety barriers (WRSB) are not designed to be connected to semi-rigid or rigid safety
barriers or bridge ends. However, WRSB may be transitioned to more rigid barriers provided that
the WRSB overlaps the more rigid barrier by an adequate longitudinal distance and the lateral
separation is sufficient to accommodate the maximum likely deflection of the WRSB. Such
arrangements should enable the two systems to work independently while providing continuous
shielding of hazards.
The Federal Highway Administration website provides details of crash-tested transitions

(http://safety.fhwa.dot.gov/roadway_dept/policy_guide/road_hardware/barriers/techadvs/archive/t5
04026/, viewed 12 July 2012).
Where a concrete safety F-shape barrier needs to be transitioned to a vertical or flared back
vertical wall, the transition must not be abrupt.
AS/NZS 3845 (1999) requires that where different road safety barrier systems adjoin, the interface
details shall be designed to achieve a gradual transition in the horizontal stiffness and height. The
difference in their horizontal stiffness shall not exceed 40% over a minimum length of 2.5 m.
Various typical transitions are provided in the following sections.
W-beam to thrie-beam
The transition is achieved through the use of a product that bolts to the W-beam at one end and to
the thrie-beam at the other end. This transition is 2 m between post centres and is illustrated in
Figure 7.13.
Austroads 2013
— 149 —
Source: AS/NZS 3845 (1999).
Figure 7.13: Interface details of type F to W-beam (meets TL-3)
W-beam to concrete
W-beams are connected to a concrete barrier either through the use of a thrie-beam transition
(Figure 7.13) or by connecting the W-beam directly to the concrete using an acceptable direct
transition (for example, see Figure 7.14 for long end post). Both treatments provide a structurally
sound connection and a smooth and stiffened transition to prevent snagging and pocketing of
impacting vehicles.
Austroads 2013
— 150 —
Figure 7.14: Interface details of type F to thrie-beam (meets TL-3)
Austroads 2013
— 151 —
The thrie-beam transition involves:

 the use of a prefabricated product to connect the W-beam to the thrie-beam (Figure 7.15)
 the post spacing being reduced from the standard spacing (2 m) to 1 m for five spaces and
then to 0.5 m for the two spaces prior to the concrete barrier
 nesting of the thrie-beam over the last 4 m prior to the concrete barrier
 the use of a structure connector to bolt the thrie-beam into a recess in the concrete barrier
(Figure 7.16).
Figure 7.15: W – thrie transition section
Figure 7.16: Structure connector – thrie-beam
Austroads 2013
— 152 —
The W-beam transition (directly to concrete barrier) involves the:

 W-beam being recessed into the concrete barrier to provide a flush barrier face at the
connection
 transition being strengthened by the post spacing being reduced progressively from the
standard spacing (2 m) to 1 m and then 0.5 m over the last 10 m of the beam
 transition being further strengthened by nesting of the W-beam over the last 5 m
 concrete barrier being flared away from the W-beam, the latter being stiffened by circular
hollow sections bearing on the face of the concrete at the rear of the beam (Figure 7.14).
Thrie-beam to concrete
The transition between the thrie-beam barrier and concrete barrier is achieved through the use of a
structure connector, as shown in Figure 7.16, which enables the thrie-beam to be bolted into a
recess in the concrete barrier. Details of the transition are shown in Figure 7.17. The thrie-beam is
stiffened in the same manner as the W-beam to concrete transition.
Figure 7.17: Interface details of type F to thrie-beam
Austroads 2013
— 153 —
Wire rope safety barrier to semi-rigid barrier

These transitions involve the wire rope safety barrier overlapping the W-beam or thrie-beam barrier
by a nominal longitudinal distance based on site conditions. Where space is available the barriers
can be separated laterally so that they operate independently (Figure 7.18).
Figure 7.18: Layout for WRSB match with existing structures
Rigid barriers to roadside barriers

The transition from a high containment bridge barrier to a roadside barrier should be gradually
established. Typically, for rigid barriers, the transition from one performance level should be to its
nearest lower performance level. Typical transitions for rigid barriers include:
 transition from Regular performance level bridge barrier to roadside barrier (Figure 7.19)
 transition from High performance level bridge barrier to roadside barrier (Figure 7.20)
 transition from Medium performance level bridge barrier to roadside barrier (Figure 7.21).
Figure 7.19: Transition from Regular performance level bridge barrier to roadside barrier
Austroads 2013
— 154 —
Figure 7.20: Transition from High performance level bridge barrier to roadside barrier
Austroads 2013
— 155 —
Figure 7.21: Transition from Medium performance level bridge barrier to roadside barrier
Examples of various types of barrier transition used in Queensland are listed in Table 7.8.
Austroads 2013
— 156 —
Table 7.8: Typical barrier transitions used in TMR

Drawing Description Features
TMR Steel beam guardrail – Installation on bridge  W-beam connection to thrie-beam
SDRM 1475 and barrier approaches  W-beam/thrie-beam connection to bridge rail
 thrie-beam connection to single slope concrete barrier or concrete bridge
parapet
 W-beam connection to F-shape concrete barrier or concrete bridge
parapet
TMR Concrete terminal with the thrie-beam Single slope concrete parapet of 1000 mm height above the finished surface
SDRM 1486 guardrail connection using a 3 m concrete barrier terminal
TMR Steel beam guardrail W-beam direct connection for concrete end posts
SDRM 1493
TMR Steel beam guardrail Thrie-beam connection for concrete end posts using a transition rail
SDRM 1494
TMR Steel bridge traffic rail – end post-W-beam Two rail-and-post barrier
SDRM 1509 connection
TMR Steel bridge traffic rail – end post-thrie-beam Two rail-and-post barrier
SDRM 1510 connection
7.8 End Treatments

As specified in AS 5100.1 (2004), ends of approach barriers shall be crashworthy and shall be
protected by a suitable transitioned traffic barrier and/or proprietary impact attenuation device. The
barrier termination shall be designed to ensure acceptable performance on impact at the required
performance level for the vehicle(s) under consideration. Any end treatments are in addition to the
LON.
The ends of a rigid barrier that may be connected to and/or splay away from a non-rigid approach
system shall be suitably terminated. End blocks, for example, shall be detailed to ensure that the
occupants of vehicles and others nearby are protected during an end-on impact.
Where rigid bridge barriers are transitioned to a flexible roadside barrier on the bridge approach,
the bridge approach barriers will use the same end treatments of the appropriate roadside barriers.
Refer to Types of safety barrier terminals in Appendix J of Austroads (2010a) for the types and
detailed requirements of barrier terminal treatments.
Where bridge barriers end at the bridge approach (do not connect to roadside barriers), the bridge
High performance level barriers should be transitioned to Regular or Medium performance level
barriers before connecting to a barrier termination. Typical termination details of Medium and
Regular performance level barriers are shown in Figure 7.22 and Figure 7.23, respectively. In
these configurations, a 780 mm long, 820 mm high concrete segment is used as the connector and
a flare of 1:20 is used at the transition segment.
Austroads 2013
— 157 —
Source: RMS standard drawings B505 and B506 (2009).
Figure 7.22: Termination details of Medium performance level steel bridge barriers
Austroads 2013
— 158 —
Source: RMS standard drawing B502 (2009).
Figure 7.23: Termination details of Regular performance level steel bridge barriers
Austroads 2013
— 159 —
7.9 Gore Area

For sites where the required total length of the approach barrier extends to the gore area
attenuator, the rigid barrier component shall be extended to the attenuator. For other sites the
extent and type of approach treatment should be based on B/C considerations.
7.10 Freeway and Highway Ramps and Side Roads

Barriers in gore areas near to the bridge that include gaps for access or side roads shall be
designed to safely contain errant vehicles at all probable impact angles that might result from an
errant vehicle leaving the nearest traffic lane at a flare rate determined from Table 7.2.
If the bridge approach barrier is connected to a side road barrier, the side road barrier must be
continuous with the bridge approach barrier system. A transition from a flexible to a rigid barrier
system at the junction of the two systems is then the only requirement.
Appropriate barrier systems, normally the rigid bridge barriers, shall be extended down freeway or
major highway entrance and exit ramps a suitable distance to protect against errant vehicles
penetrating the freeway or highway trafficked lanes.
7.11 Bridge Approach Off-structure Barrier Foundation

The loads transmitted to the barrier base/foundation shall be determined from an ultimate limit
states analysis of the barrier system using the loads specified in AS 5100.2 (2004).
In the case of foundations for rigid barrier systems, the capacity of the foundation at the base of the
barrier shall be not less than the barrier capacity in order to prevent failure of the foundations.
In accordance with the requirements of AS 5100.2 (2004), the bridge approach barrier,
connections and supporting system shall be designed as a progressive strength system to limit the
damage to the supporting system and potential reconstruction requirements that may arise from a
collision.
7.12 Examples of Approach Barrier

7.12.1 Example 1: Bridge Approach Barrier Layout Design
Determine the layout of a bridge approach barrier on a divided rural highway as shown in
Figure 7.24.
The road at the bridge approach section has the following data:
 AADT: 15 000 vpd
 design speed: 110 km/h
 alignment: essentially straight with flat gradient
 bridge plan: twin bridge with spill-through abutments and descent to stream; 3.0 m shoulder
on left, 1.0 m shoulder on right; bridge barrier ends immediately behind the shoulder; batter
slope at left side 4:1 and at median 6:1; the batter is 3.5 m high and 14.0 m wide with a clear
run-out area of greater than 3.0 m in width beyond the toe.
Austroads 2013
— 160 —
Figure 7.24: Example: determine the layout of bridge approach barrier
For a bridge approach, in most cases, the task is to shield vehicles from leaving the road close to
the bridge barrier’s end posts and either crashing into the end post or plummeting into the stream
or the road underneath the bridge.
Run-out length method

 For the left side of the carriageway
Clear zone distance is determined from Table 7.4 based on the design speed, design AADT
and the slope of the fill/cut batter. For a design speed of 110 km/h, design AADT of 15 000
vpd, and 4:1 fill batter slope, the clear zone width is 14.0 m.
It should be noted that for this case, on the left side of the carriageway it is impracticable to
flatten the batter slope 4:1 to provide a flared barrier layout. As a result, a tangent barrier
system and associated terminal treatment is the appropriate installation.
The run-out method is applied to a point in the roadside a lateral distance from the end post
equal to the clear zone (i.e. 14.0 m). As there is no flare and the rail is coincident with the
rear of the shoulder, the length (X) from the end of the bridge barrier to the point of need is
computed by Equation 39, in which:
— LA is the lateral distance from the end post equal to the clear zone width, LA = 14.0 m.
— L2 is the barrier’s lateral distance from the edge of the running lane, L2 = 3.0 m
(Figure 7.24).
— LR is the run-out length, LR = 110 m, derived from Table 7.1 based on the design speed
of 110 km/h and an AADT of greater than 10 000 vpd.
Therefore, X = 86.4 m.
A crashworthy transition should be provided to the bridge end post (Section 7.7) and its
length will be included in the 86.4 m length. A crashworthy leading end treatment should also
be provided. The point of need is assumed to be about 4 m from the leading end of the
barrier (varies depending on the product used) and consequently the overall length of barrier
required is 86.4 m + 4 m = 90.4 m.
Austroads 2013
— 161 —
 For the median side at the bridge

Table 7.4 gives a clear zone width of 10.5 m for a design speed of 110 km/h, design AADT of
15 000 vpd, and 6:1 fill batter slope.
A flared barrier alignment could be used as the 6:1 median slope can be flattened to 10:1 in
the area between the carriageway and the barrier. The median slope behind the barrier may
have to be steepened to satisfy drainage requirements. Equation 38 is used to determine the
LON of the barrier.
From Table 7.3, a shy line offset of 2.8 m is applicable for a design speed of 110 km/h in this
site. Therefore, with a 1.0 m shoulder, the barrier is within the shy line offset, and Table 7.2
gives an applicable flare rate of 30:1. Thus b/a = 1/30. Other parameters are taken as
follows.
— LA is the lateral distance from the end post equal to the clear zone width, LA = 10.5 m
(Table 7.4).
— L1 is the tangent length of the barrier upstream from the area of concern, L1 can be
taken as the length of transition L1 = 10.0 m (Section 7.7).
— L2 is the barrier’s lateral distance from the edge of the running lane, L2 = 1.0 m
(Figure 7.24).
— LR is the run-out length, LR = 110 m, derived from Table 7.1 based on the design speed
of 110 km/h and an AADT of greater than 6000 vpd.
A MELT terminal (non-proprietary product) can be accommodated in the median to provide a

gating terminal with a traversable area 18 m long by 6 m wide behind the barrier, measured
from the point of need. The point of need for the MELT is two post spacings from the leading
end. Hence the overall length of road safety barrier is 76.4 + 4 = 80.4 m.
Adoption of a higher flare angle (1 in 15 when the barrier is outside the shy line) would
reduce the required barrier length by about 15 m (62.7 m). However, increasing the flare
angle has operational disadvantages in that the angle of impact and severity of crashes
increase and there is an increased likelihood that a vehicle will be redirected back into or
across the carriageway following an impact.
The barrier arrangements determined above are illustrated in Figure 7.25.
Austroads 2013
— 162 —
Figure 7.25: Example: using run-out length method
Angle of departure method

In the angle of departure method (Figure 7.7) a vehicle travelling at 100 to 110 km/h is assumed to
leave the road at a leading angle of 2.9° (Table 7.6). This equates to a rate of lateral shift of 1 in
20. This angle is applied from the edge of the travel lane to the rear extremity of the hazard for the
adjacent traffic, and from the centre of the road to the clear zone for opposing traffic. Lengths
relating to the application of the angle of departure method are shown in Figure 7.26.
Consider the left side of the carriageway. For a design speed of 110 km/h, design AADT of
15 000 vpd, and 4:1 fill batter slope, the clear zone width is 14.0 m (Table 7.4).
Because of the batter characteristics a non-flared safety barrier is necessary. Assuming that the
road safety barrier is coincident with the back of the shoulder, it can be seen that:
LON = (14.0 – 3.0) x 20 = 220 m
Length of road safety barrier = 220 + 4 (terminal) = 224 m
Consider the median side of the carriageway. Because the median shoulder is 1.0 m wide the road
safety barrier is within the shy line and a flare rate of 1 in 30 applies (refer to Table 7.2).
The distance between points of need can be determined either graphically or by applying geometry
to the triangle enclosed by the vehicle trajectory, the back of the shoulder and the barrier (refer to
Figure 7.26).
The distance from the end of the structure to the point where the vehicle trajectory crosses the rear
of the shoulder is:
(10.5 – 1.0) x 20 = 9.5 x 20 = 190 m
Assuming that the transition between the road barrier and the bridge barrier is 10 m, the length
(along the shoulder) of the base of the triangle is 190 m minus the transition length = 180 m.
Austroads 2013
— 163 —
The vehicle trajectory rate of lateral shift and the barrier flare rate can be used to compute the
length of barrier along the side of the triangle:
30 / (30 + 20) x 180 = 108 m
Therefore the LON of the barrier is = 108 m + transition length = 118 m and the overall length of
road safety barrier = 108 + 4 (MELT) = 122 m.
Figure 7.26: Example: using angle of departure method
It should be noted that, from Figure 7.25 and Figure 7.26, the LON using the angle of departure
method is greater than the LON calculated using the run-out length method.
Austroads 2013
— 164 —
8 BRIDGE SUPPORT PROTECTION BARRIERS

8.1 Introduction
The supports (abutments and piers) of overpass bridges are required to be structurally protected
from vehicle collisions. They are considered roadside hazards, and therefore need to have safety
treatments to protect road users.
It is specified in Section 11 of AS 5100.1 (2004) that the relevant jurisdictions need to make an
assessment of the risk of a vehicle impacting bridge supports, bridge superstructure and bridge
elements above the bridge deck, and determine the level of protection and the associated
appropriate performance levels of the barrier systems. A detailed procedure for risk assessment
and determination of the protection level, however, is not available. This section provides a design
guideline for bridge support protection barriers considering the following:
 design requirements
 determination of level of protection
 barrier layout and design.
8.2 Current Design and Specification Requirements

8.2.1 Current AS 5100 Requirements
AS 5100.1 (2004) specifies that:
 For road traffic collisions, the relevant authority shall determine the minimum ‘no protection’
clearance of a pier/column from the roadway.
 For pedestrian bridges, piers/columns shall be either located at a safe distance or protected
from collision.
 For railway traffic collisions, bridges over railways shall have a clear span between
abutments. Otherwise an alternative load path or deflection walls or heavy construction
columns/piers shall be provided.
Section 10 of AS 5100.2 (2004) specifies the design collision loads and design requirements for
protection of bridge supports from collisions of roadway and railway traffic. For collisions from road
traffic, bridge supports shall be designed for a minimum equivalent static load of 2000 kN applied
at an angle of 10° from the direction of the road centre-line and at 1.2 m above the ground level, if
they are not located behind appropriate protective traffic barriers.
It is suggested that the recent update of AASHTO (2012) be incorporated, which specifies that the
equivalent static load is 2670 kN acting in a direction of zero to 15° with the edge of the pavement
in a horizontal plane, at a distance of 1.5 m above ground.
For collision from railway traffic, AS 5100.2 (2004) specifies that alternative load paths be provided
to prevent the superstructure of the overpass bridge from collapsing. Otherwise, the supporting
elements located within 10 m of the centre-line of the railway track shall be designed to resist a
minimum collision load of 3000 kN parallel to the rails and 1500 kN normal to the rails. Where
supporting elements are located between 10 m to 20 m from the centre-line of the railway track,
the level of protection shall be determined based on a risk analysis. The piers and columns shall
be designed to resist a minimum collision load of 1500 kN at any angle in the horizontal plane.
Austroads 2013
— 165 —
It should be noted that, for individual column shafts, the load should be considered as a point load.
For wall piers, the load may be considered to be a point load or may be distributed over an area
deemed suitable for the size of the structure and the anticipated impacting vehicle, but not greater
than 1500 mm wide by 600 mm high. These dimensions were determined by considering the size
of a truck frame (AASHTO 2012).
8.2.2 Additional Requirements from Jurisdictions’ Guidelines

All jurisdictions follow the AS 5100 (2007) requirements for the design of bridge support protection
barriers. Several jurisdictions such as RMS and VicRoads provide clarifications and more detailed
guidelines.
Roads and Traffic Authority (2008b) specifies that:

 No protection is required if the clearance of the piers or columns from road traffic exceeds
the required minimum distance specified in Roads and Traffic Authority (1996).
 Road safety barriers (flexible and semi-rigid barriers) may be provided to protect occupants
of errant vehicles in the vicinity of bridge supports in accordance with the following
documents: Roads and Traffic Authority (1996) and AS/NZS 3845 (1999).
 The road safety barriers should be disregarded when assessing the need for protection of
the bridge supports from collision, as the errant vehicles may penetrate these barriers,
especially at high impact angles.
It was stated in Roads and Traffic Authority (1996) that a rigid barrier may be undesirable if the
distance from the outside of the travelled lane to the barrier is greater than 4 m because of the
possible high impact angles. Where the clear zone is traversable, it is preferred that all barriers be
as far from the carriageway as possible. However, where a rigid barrier is proposed, a check is
required to ensure that high impact angles are unlikely.
VicRoads (2009a) provides clarifications and more detailed guidance for the design of bridge
support protection barriers.
8.3 Design Procedure

The following design steps are suggested:
 Assess the site conditions to determine the level of protection, including the selection of an
appropriate barrier performance level.
 Determine the barrier layout, including the barrier LON and lateral offset of the barrier.
 Determine structural solutions, including barrier type, height consideration and foundations.
These design steps are presented in more detail in the following sections.
8.4 Determination of Level of Protection

8.4.1 Levels of Protection
When assessing the need for protection of bridge supports for collision from road traffic, risk
assessment and determination of protective measures carried out in accordance with Clauses 11.1
and 11.2 of AS 5100.1 (2004) may result in one of the levels of protection discussed below.
Austroads 2013
— 166 —
No protection
Bridge supports located at a distance greater than the clear zone width, measured from the
nearest face of the bridge supports to the edge of the traffic lane as determined from Table 7.4
(taking into consideration batter slope, design speed and design ADT) do not need to be designed
for collision from road traffic if adequate sight distance has been provided. New bridge designs
should take this into consideration.
Road safety barriers

Flexible or semi-rigid barriers may be provided to protect occupants of errant vehicles in the vicinity
of bridge supports where the bridge supports are located within the clear zone determined from
Table 7.4 and there is sufficient space to accommodate the barrier’s dynamic deflection
(Section 8.5.2) and vehicle roll allowance (Section 8.5.3). These barriers shall be designed in
accordance with Austroads (2010a) and AS/NZS 3845 (1999).
Under these circumstances, bridge supports shall be designed for the collision load specified in
Section 8.2.1. The collision load may be assumed to vary linearly from the full load when the bridge
supports are located at a lateral distance equal to the clearance required for the barrier to deflect
from the edge of the nearest travel lanes, to zero load when the bridge supports are located a
distance equal to the clear zone width from the edge of the nearest travel lanes.
Rigid barriers
Where rigid barriers are provided, the barriers shall be structurally and geometrically designed in
accordance with AS 5100 (2007) for an appropriate performance level determined in Section 8.4.3.
The barriers shall be crashworthy, ground-mounted and structurally independent. The bridge
supports do not need to be designed for collision from road traffic.
The following should be noted when considering the adoption of a high containment rigid barrier
(Austroads 2010a):
 Road safety barriers suitable for all trucks are rarely used and are expensive.
 A heavy vehicle will not be contained by a normal road safety barrier.
 A car may be extensively damaged by impact with a barrier designed for trucks.
Except for barriers associated with bridges and other situations where the consequences of
vehicles leaving the road are extreme, road safety barriers are not normally designed to contain
van or tanker type semi-trailers (of Medium and High performance levels). This design limitation
has been practised primarily because of the relatively low volumes of these vehicles on many
roads and the high cost of providing barriers to contain them. The increased severity of passenger
car crashes into high-containment barriers is also an important consideration.
A risk assessment as presented in Section 8.4.2 may be carried out to determine if rigid barriers
should be provided.
Where a risk assessment indicates that the run-off-road risk associated with heavy vehicles is
particularly high, a Medium or High performance level may be considered. If available, local
information on truck encroachment frequency should also be considered.
It should be noted that at the BTTF meeting in December 2012, an agreement was reached by all
jurisdictions on the principle that the bridge piers should be designed for collision from road traffic,
regardless of the presence of roadside barriers.
Austroads 2013
— 167 —
8.4.2 Risk Assessment

A site-specific risk assessment needs to be undertaken in conjunction with B/C considerations to
determine the level of protection for bridge supports for the construction of a new roadway, a
widening or a road safety upgrade of an existing roadway. See Section 3 for guidance on how to
undertake a site-specific risk assessment and a B/C analysis.
The site-specific risk assessment shall take into consideration all relevant parameters including,
but not limited to the:
 class of road
 design and operational speed of the road
 total traffic volume
 percentage of commercial vehicles
 type and proportion of heavy vehicles
 road crash records (if existing) for this and similar roadways
 potential outcomes of a vehicle impact with a bridge support considering vehicle occupants
and third party persons and property
 potential risks associated with the introduction of bridge support protection system(s)
 road geometry
 differences in levels between divided carriageways, surface levels and drainage
requirements in the central median area
 distance from trafficked lanes to the face of the bridge supports
 presence of a sealed shoulder or emergency stopping lanes between the trafficked lanes and
the bridge supports
 strength and robustness of the bridge supports
 geometry of the bridge supports, with particular attention given to tapered columns that might
snag high vehicles.
Specifically, the assessment should determine the need for a barrier and type of barrier by
considering:
 if a barrier is required to protect the bridge supports, which may involve a B/C consideration
 what barrier performance level/barrier type is required to protect the site, based on traffic
volume, proportion/type of heavy vehicle, and site conditions
 if the horizontal clearance is sufficient for the selected type of barrier taking into account the
likely deflection of the barrier system.
Specific attention shall be given to the following situations:

 an existing divided carriageway with a single continuous concrete median barrier that is
being widened on the outside
 an existing divided carriageway that is being widened into the median with bridge supports
within this median
 a new divided carriageway with bridge supports located within the median
 situations where bridge supports are located close to the outside edges of trafficked lanes
 situations where bridge supports are located close to entry and exit ramps from freeways and
major highways.
Austroads 2013
— 168 —
To aid the risk assessment process, a formula provided in AASHTO (2012) is suggested to be
used to evaluate the annual frequency of impact from heavy vehicles, with the approval of the
relevant jurisdiction, as follows (Equation 41):
AFHBP = 2(AADT) (PHBP)365 41
where
AADT = the number of trucks per day in one direction
the annual probability for a bridge pier to be hit by a heavy vehicle;

PBHP = 3.457 x 10-9 for undivided roadways in tangent and horizontally
PHBP = curved sections; PBHP = 1.090 x 10-9 for divided roadways in tangent
sections; PBHP = 2.184 x 10-9 for divided roadways in horizontally curved
sections
Design for vehicular collision force is not required if AFHBP is less than 0.0001 for critical or
essential bridges or 0.001 for typical bridges.
8.4.3 Selection of Performance Level

Where road safety barriers are provided, the barrier performance level and type shall be selected
in accordance with Austroads (2010a) and AS/NZS 3845 (1999), based on the specific
requirements of the site such as the design vehicle and speed environment.
Where a rigid barrier is required to either protect road users or bridge supports, the type and
performance level of the barrier must be determined based on the traffic data and the site
conditions. The following scenarios are suggested:
 Roads with low volumes of commercial vehicles (less than 15% AADT or 1500 commercial
vehicles per day per carriageway): Regular performance level barrier should be provided.
 Roads with medium to high volumes of commercial vehicles (from 15% AADT to 30% AADT
or up to 3000 commercial vehicles per day per carriageway): Medium performance level
barrier should be provided.
 Roads with high volumes of commercial vehicles (greater than or equal to 30% AADT or
3000 commercial vehicles per day per carriageway); roads with special site-specific
conditions as specified by relevant road jurisdiction: High performance level barrier should be
provided.
 Site-specific, unusual conditions and critical locations as specified by the relevant road
jurisdiction: Special performance level barrier should be provided.
It should be noted that the Regular performance level is the minimum performance level to be
selected where rigid barriers are required.
Austroads 2013
— 169 —
8.5 Barrier Layout and Design

Where a barrier is required to either protect road users or bridge supports, the layout of the barrier
should be determined based on site condition and vehicle characteristics. Factors that need to be
determined include:
 barrier length of need (LON), to determine the extent of the barrier to prevent an errant
vehicle from crashing into the bridge support
 clearance for barrier dynamic deflection, to determine the lateral position of the barrier to
ensure sufficient clearance for the barrier to deflect in a vehicle crash
 vehicle roll allowance, to ensure that no part of the errant vehicle impacts the bridge support.
The dynamic deflection and vehicle roll allowance are used to determine the transverse location of
the barrier.
In addition, it is recommended that the face of guardrail or other devices be at least 600 mm from
the face of the bridge support and at least 600 mm outside the normal shoulder line
(AASHTO 2012).
8.5.1 Length of Need

The estimate of the LON can be calculated using either the run-out length method or the angle of
departure method to determine the extent of the protection barrier. Refer to Section 7.4 for details
of how to calculate LON.
8.5.2 Dynamic Deflection

When a vehicle strikes a road safety barrier the dynamic deflection of a barrier varies according to
the characteristics of the impacting vehicle, impact speed, angle of impact and the characteristics
of the barrier system. Sufficient dynamic clearance should be provided between the face of a
barrier and the bridge support to accommodate the appropriate dynamic deflection.
The deflection information given in Table 8.1 is only suitable for concept design. More detailed
deflection data for use in detailed design should be obtained from the relevant road jurisdiction
guidelines or specific product information. If the envelope for deflection is too small to
accommodate the dynamic deflection of a flexible barrier, then a semi-rigid or rigid barrier must be
used.
Table 8.1: Indicative deflection for concept/feasibility design

Barrier type Indicative deflection caused by 2000 kg vehicle,
at 100 km/h impacting at 25° (m)
Work zone barrier Refer to relevant road authority
Wire rope road safety barrier 1.7–2.2 (depending on product type)
W-beam road safety barrier 1.4
Thrie-beam road safety barrier 0.9
Type F concrete barrier (permanent system attached
0.0
to the pavement)
Notes:
 Refer to relevant road authority for product specific deflections for use in detailed design.
 The deflections quoted in the table are based on the crash test results which typically involve short sections of barrier (e.g. 100 m).
 The deflection of wire rope barrier systems will depend on post spacing, barrier length and road curvature.
 Where the barrier type is known designers should check specific product information to determine design deflections.
Austroads 2013
— 170 —
8.5.3 Vehicle Roll Allowance

All barrier systems deflect under impact by errant vehicles except for rigid concrete barriers. The
behaviour of barriers under impact varies depending on the type of system. The posts in semi-rigid
systems act to reduce deflection but may induce more vehicle roll whereas the weaker posts in
flexible systems do not restrain the cables under impact and vehicle roll is likely to be less.
The clearance required between the face of a road safety barrier and a bridge support includes the
dynamic deflection of the road safety barrier system and a vehicle roll allowance and is described
with the term ‘working width’ as shown in Figure 8.1.
The vehicle roll allowance values shown in Table 8.2 are for a height of 4.3 m above the pavement.
These values are based on the vehicle dynamics of a 4.3 m high van type rigid or articulated truck.
The vehicle roll allowance values may be interpolated where the hazard is less than 4.3 m high but
caution should be applied in their use.
Source: Adapted from Roads and Traffic Authority (2008a).
Figure 8.1: Vehicle roll allowance
Austroads 2013
— 171 —
Table 8.2: Vehicle roll allowance

Vehicle roll allowance at 4.3 m height above pavement (m)
Design Crossfall (%)
speed –7 –6 –5 –4 –3 –2 –1 Flat +1 +2 +3 +4 +5 +6 +7
(km/h)
40 0.70 0.65 0.60 0.55 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50
50 0.75 0.70 0.65 0.60 0.55 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50
60 0.80 0.75 0.70 0.65 0.60 0.55 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50
70 0.85 0.80 0.75 0.70 0.65 0.60 0.55 0.55 0.55 0.50 0.50 0.50 0.50 0.50 0.50
80 0.90 0.85 0.80 0.75 0.70 0.65 0.60 0.60 0.60 0.55 0.50 0.50 0.50 0.50 0.50
90 1.00 0.95 0.90 0.85 0.80 0.75 0.70 0.70 0.70 0.65 0.60 0.55 0.50 0.50 0.50
100 1.10 1.05 1.00 0.95 0.90 0.85 0.80 0.80 0.80 0.75 0.70 0.65 0.60 0.55 0.50
110 1.20 1.15 1.10 1.05 1.00 0.95 0.90 0.90 0.90 0.85 0.80 0.75 0.70 0.65 0.60
Notes:
Interpolate vehicle roll slope line for objects that are less than 4.3 m high.
The vehicle roll allowance will depend on the height and type of barrier. The values in this table apply to a 710 mm high (approximately) W-beam barrier and rigid
barriers.
Source: Roads and Traffic Authority (2008a).
8.5.4 Typical Barrier Layout for Bridge Supports Adjacent to Undivided Roadways
For bridge supports located adjacent to undivided roadways, typical layouts are shown in
Figure 8.2. Non-flared and flared barriers may be selected based on the batter slope and available
clearance. Single-sided barrier cross-sections can be used.
Figure 8.2: Typical barrier layouts for bridge supports adjacent to undivided roadways
Austroads 2013
— 172 —
8.5.5 Typical Barrier Layout for Bridge Supports Located in Median Area
Single-stage barrier system
For bridge piers located within a narrow median, F-shape or single slope rigid barriers are often
chosen, but back-to-back semi-rigid or flexible barriers may be used where the dynamic deflection
can be contained within the median width so that an unacceptable risk does not eventuate for
opposing traffic. Typical profiles of median barriers for use in narrow medians are shown in
Figure 8.3. Appropriate end treatments must be used to suit each type of barrier and situation (i.e.
width available and behaviour of end treatment on vehicle impact).
Figure 8.3: Examples of road safety barriers for use in narrow medians
Figure 8.4 shows how a bridge pier can be shielded in a narrow median. Where the median is wide
enough to accommodate the barrier’s deflection, a flexible or semi-rigid barrier is preferred.
Figure 8.4: An example of a barrier layout for shielding a bridge pier located in a median
Two-stage barrier system

For bridge supports located in a wider median, a two-stage protection barrier system may be
considered (Figure 8.5).
Austroads 2013
— 173 —
Source: Roads and Traffic Authority (2008a).
Figure 8.5: Two-stage protection layout
The following issues must be considered when designing two-stage protection (Austroads 2010a):
 Design requirements for working width, LON and minimum length of barrier must be met.
The traffic face of the rigid barrier must be offset from the face of the bridge pier to
accommodate working width.
 The structural design of the bridge pier may require a gap between the rigid barrier and the
bridge pier, or the rigid barrier may be an integral part of the bridge pier, if the pier has been
designed for high-mass, high-speed impact loadings.
 The rigid barrier must be designed to ensure it does not overturn when impacted by errant
large mass vehicles.
 It is preferable that the rigid barrier be aligned parallel to the carriageway, rather than flared.
This will facilitate extension of the rigid barrier if additional lanes are added in the future.
 In some situations the end of the rigid barrier may require a crash attenuator (refer to
individual road authority requirements).
 Wherever possible, the offset between the flexible barrier and the rigid barrier should be
more than the working width of the flexible barrier.
 A cross slope of 10% or flatter should be provided between the flexible and the rigid barrier.
8.5.6 Height Considerations

The height of the road safety barriers shall be in accordance with Austroads (2010a) and
AS/NZS 3845 (1999).
For rigid concrete barriers, the height of the barrier shall be in accordance with the performance
level selected for the site as presented in Section 8.4.3. Applicable barrier heights and designs
based on standard barrier designs presented in Section 5 are summarised in Table 8.3.
Austroads 2013
— 174 —
Table 8.3: Applicable height of rigid concrete barriers
Barrier performance level Height above finishing level (mm) Applicable barrier design
Regular 810–820 RMS concrete parapet RCO, RCI
DPTI single slope concrete barrier SAR1
DPTI F-shape concrete barrier SAR2
Medium 1070–1100 RMS concrete parapet MCO, MCI
DPTI single slope concrete barrier SAM1
DPTI F-shape concrete barrier SAM2
High 1400–1500 Single sided F-shape concrete barrier
Double sided F-shape concrete barrier
Special 1500–2000 Single sided F-shape concrete barrier
Double sided F-shape concrete barrier
The height of the bridge support protection barrier associated with the barrier performance level
shown in Table 8.3 may be modified to safely redirect a heavy commercial vehicle and minimise
the lateral rotation of a high centre-of-gravity vehicle and potential for snagging of the bridge
supports.
The determination of appropriate height and clearance of the pier protection system from the
bridge supports shall be based on the vehicle roll allowance (Figure 8.1 and Table 8.2).
The height of the protection barrier should be gradually reduced away from the bridge supports to
reduce the size of the potential impact area at each end of the system. However, the slope on the
top surface of the barrier should minimise the probability of the barrier acting as a ramp and
launching a heavy vehicle. This slope should be of the order of 1 in 10. The height and width of the
barrier at each end should be minimised or shaped to reduce the severity of a direct impact with
the end of the barrier.
For applications where the operational speed of the road is 80 km/h or less, site-specific risk
assessment might indicate that a reduced height of about 1100 mm might be adequate to re-direct
heavy vehicles, subject to considerations about the probability of bridge support snagging and the
robustness of these supports.
8.5.7 Foundations for Rigid Barriers

Where rigid barriers are provided, appropriate foundations, in the form of bored piles or similar,
shall be provided to resist the barrier design loads referred to above in accordance with the
requirements of AS 5100 (2007).
8.6 Bridge Support Protection from Railway Traffic

In addition to the requirements set out in AS 5100.1 (2004), the Supplementary 1 of AS 5100.1
(2006) provides the following considerations for the design of bridges and other structures over or
adjacent to railways:
 Bridge abutments should be designed with wing walls close to parallel to the track and flush
with the abutment, to act to deflect a derailed train past the abutment. Wing walls
perpendicular to rail tracks are most undesirable.
 In general, unprotected piers near railway tracks should be avoided. Adequate alternative
load paths should be designed to permit a derailed train to demolish supports without
collapse of the superstructure onto the train.
Austroads 2013
— 175 —
 Collisions of railway traffic with heavy support structures may lead to loss of life to train
occupants or third parties due to carriages becoming airborne, telescoping and jack knifing.
 The designer should consider whether the emergency authorities should be advised on
recommended traffic control on the bridge once a support is damaged or removed, or if an
electronic system to detect large displacements of the supports coupled with traffic control
devices should be installed.
 The provision of alternative load paths should increase the structures’ ability to resist
damage from explosions.
It should be noted that the piers and columns of heavy construction and deflection walls that are
designed for the minimum collision load given in AS 5100.2 (2004), can only be expected to deflect
a train when the collision is a glancing blow, not a head-on collision of a locomotive. In a head-on
collision, the pier, column or wall may fail, resulting in support failure, or the train may telescope,
concertina or become partly airborne, in which case the support and/or superstructure may be
impacted and fail.
Continuous walls are required in order to deflect a derailed train and prevent the demolition of
bridge piers or structure supports.
Deflection walls significantly improve the performance of supports of heavy construction in a

derailment. They perform more effectively the more they can deflect in a derailment before loading
the support.
Where piers, columns or walls are located between 7.5 m and 20.0 m from a designated main line,
a risk analysis in consultation with the relevant authority should determine the level of protection
required, but not be less than the provisions of AS 5100.2 (2004).
The wall geometry should be selected so that a derailed locomotive is maintained in an upright
position if possible.
In some circumstances, it may be impractical to design for the loads specified in AS 5100.2 (2004)
and the relevant authority may specify an alternative load condition.
Reference should be made to Queensland Rail (2010) for a comprehensive guidance on collision
protection of supporting elements adjacent to railways.
8.7 Typical Design Drawings

This section presents concept details of typical systems currently being used by VicRoads for the
collision protection of existing and new bridge supports located in the central median of freeways.
Similar details can be used on major highways and arterial roads with divided carriageways.
Plan view and side elevation for a typical median pier collision protection system are shown in
Figure 8.6. The typical cross-sections are shown in Figure 8.6 and Figure 8.7, respectively, for an
inclined face bridge support and a vertical face bridge support.
Austroads 2013
— 176 —
Source: VicRoads (2009a).
Figure 8.6: Example: VicRoads typical bridge support protection system
Austroads 2013
— 177 —
Figure 8.7: Typical collision protection system for a median pier with inclined faces parallel to the trafficked lanes
Figure 8.8: Typical collision protection system for a median pier with vertical faces parallel to the trafficked lanes
Austroads 2013
— 178 —
This is a two-stage barrier system which consists of flexible (WRSB) and rigid (concrete) barriers.
The main component of the bridge support protection system, shown in Figure 8.6, is the
reinforced concrete coffer dam type section. This system is intended to avoid medium to heavy
vehicles directly imparting collision loads to the bridge support system.
This type of system is intended to provide protection to an existing, non-robust bridge support
system. It shall be designed to act independently of the bridge support system, by relying on its
self-weight and independent foundations. A clearance cavity shall be provided around the bridge
supports to minimise the probability of collision load transferring to them.
In the case of new bridge supports or existing robust supports that satisfy AS 5100.1 (2004)
requirements, consideration may be given to designing a similar concrete pier protection system
that acts integrally with the bridge supports.
A secondary flexible (wire rope or similar) system is provided in front of the concrete collision
protection system to safely contain and redirect light to medium-mass vehicles without any contact
with the rigid concrete barrier.
The combination of flexible barrier and tapered rigid concrete system is also intended to safely
redirect an errant medium to heavy vehicle impacting at a low angle of incidence. It is also
intended to partially contain and redirect heavier vehicles, particularly buses to minimise the
severity of any collision with the rigid system and avoid head-on impact with the end of the system.
The above features are aimed at minimising the risk of serious injury to vehicle occupants.
The ends of the concrete barrier system potentially represent a hazard to multi-passenger,
medium-mass vehicles such as buses. Such vehicles may have sufficient momentum to deflect the
flexible barrier system more than the minimum 2.5 m offset specified in Figure 8.6. Site-specific
consideration shall be given to the provision of additional end protection, such as a crash cushion
impact system.
As can be seen from Figure 8.6, the approach traffic face of the rigid barrier is flared at a 1 in 10
slope away from the traffic from a minimum offset of 1 m at the bridge supports until a minimum
offset of 2.5 m to the flexible (WRSB) system is achieved at the approach end.
The trafficked face of the rigid barrier runs parallel with the trafficked lanes from a distance of
approximately 1.5 m prior to the bridge supports for the distance required to achieve the
appropriate offset from the flexible barrier system on the other carriageway.
If the road alignment is curved, the rigid barrier may have to be flared at a rate of 1 in 10 on both
sides to maintain the minimum clearance to the flexible barrier system. This will result in a longer
length of concrete barrier.
The height of the rigid barrier system varies by a slope of 1:10. The height of the barrier at the
bridge supports can be 1500 mm for a High performance level barrier and can be 1100 mm for a
Medium performance level barrier.
An appropriate foundation for the rigid barrier shall be provided to resist the design loads
associated with the barrier performance level as presented in Section 4.
Austroads 2013
— 179 —
8.8 Examples
8.8.1 Example 1: Protection Barrier for a Bridge Pier Adjacent to an Undivided Two-way
Road
Determine the layout of a road safety barrier to shield a bridge pier (6 m long x 2 m wide) adjacent
to a rural two-lane, two-way road as illustrated in Figure 8.9.
Figure 8.9: Example 1: Bridge pier protection barrier
The road has the following characteristics:

 AADT = 14 000
 commercial vehicles = 2%
 design speed = 100 km/h
 plan view: as per Figure 8.9; 2.5 m shoulder; pier 4 m from the edge of the traffic lane; fill
batter slope 6:1.
Determine level of protection
Clear zone distance is determined from Table 7.4 based on the design speed, design AADT and
the slope of the fill/cut batter. For a design speed of 100 km/h, a design AADT of 14 000 vpd, and
6:1 fill batter slope, the clear zone width is 10.0 m. In this case, the pier is located within the clear
zone width; therefore the pier should be protected from a road traffic collision by either road safety
barriers or rigid barriers.
As there is only 1.5 m from the rear edge of the shoulder to the face of the pier there is insufficient
space to accommodate the relatively large deflections of a flexible road safety barrier (refer to
Table 8.1 which provides indicative deflections for planning purposes, and bearing in mind that a
WRSB will be located a small distance behind the edge of the shoulder).
Referring to Table 8.1, a W-beam (i.e. single rail with 2.0 m post spacing) may require a deflection
distance of up to 1.4 m, measured between the face of the barrier in its initial position and its final
position after impact. As a W-beam system has a system width of about 0.44 m, its dynamic
deflection cannot be accommodated within the 1.5 m. If a lesser dynamic deflection was assumed
(e.g. 1.0 m from AS/NZS 3845–1999) a total distance of 1.44 m would be required and could be
accommodated. However, this does not allow for a vehicle to roll should a high van impact the
barrier (Figure 8.1 and Table 8.2) and adding a vehicle roll allowance of 0.8 m results in a total
working width of 1.44 + 0.8 = 2.24 m. A similar process shows that a short section of thrie-beam
Austroads 2013
— 180 —
(deflection 0.6 m to 0.9 m) resulting in a minimum working width of 0.6 + 0.44 + 0.8 = 1.84 which is
also greater than 1.50 m.
It is clear that a system stiffness greater than that of a standard W-beam or thrie-beam is required.
Options that could be considered include the use of a rigid concrete barrier or a thrie-beam that
has reduced post spacing, or perhaps crash cushions. As the percentage of commercial vehicles is
relatively low (2%) a Regular performance level barrier should be sufficient.
Length of need using run-out length method

Consider first the lane nearest the hazard. The LON for traffic in the lane nearest to the hazard on
the approach to the hazard can be calculated using Equation 38 where:
LA = is the lateral distance from the edge of traffic lane to the rear of the pier, LA =
6.0 m.
L1 = is the tangent length of the barrier upstream from the area of concern (i.e.
from the leading face of the pier)? In this case, L1 can be taken as the length
of one rail L1 = 4.0 m.
L2 = is the barrier’s lateral distance from the edge of the running lane? Assume
that the barrier is located at a 0.3 m lateral distance from the edge of the
shoulder, then L2 = 2.5 m + 0.3 m = 2.8 m.
LR = is the run-out length, LR = 91 m, derived from Table 7.1 based on the design
speed of 100 km/h and an AADT of greater than 10 000 vpd.
b = is the barrier flare rate. A flared barrier alignment could be used as the 6:1
a batter slope can be flattened to 10:1 in the area between the carriageway
and the barrier. From Table 7.3, a shy line offset of 2.4 m is applicable for a
design speed of 100 km/h in this site. Therefore, with a road safety barrier
offset of 2.8 m from traffic lane, the barrier is outside the shy line offset, and
Table 7.2 gives an applicable flare rate of 18:1 for rigid barriers. Thus
b/a = 1/18.
Note that this is the distance from the leading face of the pier to the point of need for traffic
approaching in the lane nearest the hazard, and that the length computed is for a flared end
(i.e. Line A in Figure 8.10).
The end of the concrete barrier should be shielded using a suitable leading terminal (i.e. crash
cushion) which will increase the overall length.
Consider the second lane (furthest from the bridge pier). The rear of the pier is offset from the
opposing traffic lane by 9.5 m (6.0 + 3.5) and is therefore 0.5 m within the clear zone for the
opposing direction. In this case:
 LA is the lateral distance from the edge of the running lane to the rear of the pier, LA = 9.5 m.
 L1 = 4.0 m.
 L2 is the barrier’s lateral distance from the edge of the running lane,
L2 = 3.5 m + 2.5 m + 0.3 m = 6.3 m.
 LR is the run-out length, LR = 91 m.
Austroads 2013
— 181 —
 With a road safety barrier offset of 6.3 m from the opposing traffic lane, the barrier is outside
the shy line offset (2.4 m), and Table 7.2 gives an applicable flare rate of 18:1 for rigid
barriers. Thus b/a = 1/18.
Note that this is the distance from the trailing face of the pier to the point of need for traffic
approaching in the lane furthest from the hazard, and that the length computed is for a flared end
(i.e. Line A in Figure 8.10). A suitable trailing terminal (i.e. crash cushion) which will increase the
overall length should also be provided.
The overall length of road safety barrier required for traffic approaching from both directions
therefore comprises the sum of the length required for the traffic in the lane adjacent to the hazard
(leading), the length required for traffic in the opposing lane (trailing), plus the length of pier parallel
to the road (i.e. 6.0 m).
Hence the overall length of need of the barrier is 28.2 + 21.4 + 6.0 = 55.6 m, which is rounded to
56 m.
The road safety barrier layout determined above is illustrated as the Line A installation (flared
barrier) in Figure 8.10. A Line B installation (parallel barrier) is also illustrated showing that a longer
road safety barrier would be required with a tangential alignment (i.e. no flared ends). The required
LON, X, can be computed for both the leading and trailing sides using Equation 38 as follows:
 for leading point of need, X = (6.0 – 2.8) / (6.0 / 91) = 48.5 m
 for trailing point of need, X = [(9.5 – 6.3)] / (9.5 / 91) = 30.7 m.
The overall length of need is given by the length on the leading side plus the length of hazard plus
the length on the trailing side is 48.5 + 6.0 + 30.7 = 85.2 m, which is rounded to 85 m.
In summary, a barrier that allows for flaring of both ends would be 56 m long between points of
need and a barrier that is not flared at the ends would be 85 m long between points of need. The
choice of the barrier alignment would often be governed by embankments and other geometric
features.
Source: Adapted from Austroads (2010a).
Figure 8.10: Example 1: Options for bridge pier protection barrier layout using run-out length method
Austroads 2013
— 182 —
Length of need using the angle of departure method

Using the angle of departure method (Section 7.4.2) a vehicle travelling at 100 km/h is assumed to
leave the road at a leading angle of 2.9° (Table 7.6). This equates to a rate of lateral shift of 20:1.
This angle is applied from the edge of the travel lane to the rear extremity of the bridge pier for the
adjacent traffic, and from the centre of the road to the clear zone for opposing traffic.
It is assumed that the road safety barrier in this example is located 2.8 m from the edge of the
nearest travel lane.
Figure 8.11 shows the general layout options for the barrier. For a straight alignment the lengths
can be determined either graphically or by using simple algebra that relates to the geometry of
triangles.
As the barrier is to be located 2.8 m from the edge of the travel lane it is situated 6.0 – 2.8 = 3.2 m
from the rear of the hazard at both the leading side and trailing side (i.e. for opposing traffic). The
dimensions resulting from the angle of departure method are shown in Figure 8.11.
For Line B, the key factor is the rate of divergence (20:1) of the vehicle travel path from the edge of
traffic lane, and as the barrier is parallel to the rear of the shoulder, the length of barrier between
points of need is given by:
L = 20 x 3.2 + 6 + 20 x 3.2 = 64 + 6 + 64 = 134 m.
For Line A the length of barrier between the points of need can be determined either graphically or
by applying geometry to the triangle enclosed by the vehicle trajectory, Line B and Line A
(Figure 8.11).
Bridge pier
134 m
Source: Adapted from Austroads (2010a).
Figure 8.11: Example 1: Options for bridge pier protection barrier layout using angle of departure method
Austroads 2013
— 183 —
For the leading length of need, the base of the triangle is 64 – 4 = 60 m long. As the vehicle
trajectory is 1 in 20 and the flare of the barrier is 1 in 18 the length of the side of the triangle along
the road safety barrier can be determined from equations for the height of the triangle as follows:
h = c / 20 = d / 18 and as c = 60 – d the equation becomes:
(60 – d) / 20 = d/18 which can be solved to determine that d = 28.5 m.
The length of the barrier can then be solved but at such a small slope will approximate d. The
length of barrier required from the leading point of need to the hazard is therefore 28.5 m plus the
nominal 4 m long section of straight barrier (i.e. 32.5 m). The same length will apply to the trailing
side of the hazard (Figure 8.11).
Therefore, the length of barrier between points of need = 32.5 x 2 + 6 = 71 m.
It should be noted that, in both flared and non-flared barrier installations, the LON using the angle
of departure method is greater than the LON calculated by using the run-out length method,
respectively.
8.8.2 Example 2: Protection Barrier for a Bridge Pier Located in a Median

Determine the layout of a barrier to shield the end of a bridge barrier on a divided rural highway as
illustrated in Figure 8.12. The road at the bridge approach section has the following data:
 AADT: 30 000 vpd
 commercial vehicles = 20%
 design speed: 100 km/h
 plan: a concrete bridge pier located in the median area of a two-way divided road; 3.0 m
shoulder on left, 1 m shoulder on right; batter slope in the median 6:1.
Figure 8.12: Example 2: Site plan
Austroads 2013
— 184 —
Determine level of protection

The same assessment as in Example 1 (Section 8.8.1) was carried out for this example. The clear
zone distance is determined based on the design speed, design AADT and the slope of the fill/cut
batter. From Table 7.4, for a design speed of 100 km/h, design AADT of 30 000 vpd, and 6:1 batter
slope, the clear zone width is 10.0 m. Therefore, the pier is located within the clear zone (8.0 m
from the traffic lane). There is a high probability that it will be hit by errant vehicles and the
consequences will be extremely severe. As a result, the pier should be protected from traffic
collision.
As the percentage of commercial vehicles is 20%, the number of commercial vehicle is 6000 vpd.
From Section 8.4.3, a High performance level barrier that caters for heavy vehicles is warranted.
Length of need using run-out length method

Consider first the lane nearest the bridge pier. A flared barrier alignment could be used as the 6:1
batter slope can be flattened to 10:1 in the area between the carriageway and the barrier. The LON
for traffic in the lane nearest to the bridge pier on the approach to the pier can be calculated using
Equation 38 where:
LA = is the lateral distance from the edge of traffic lane to the rear of the pier
(13.2 m), or to the clear zone, whichever is smaller. In this case, CZ =
10.0 m, therefore LA = 10.0 m.
L1 = is the tangent length of the barrier upstream from the area of concern (i.e.
from the leading face of the pier)? In this case, L1 can be taken as the length
of one precast barrier segment L1 = 3.0 m.
L2 = is the barrier’s lateral distance from the edge of the running lane? Assume
that the barrier is located 0.6 m from the rear edge of the shoulder, then L2 =
1.6 m.
LR = is the run-out length, LR = 91 m, derived from Table 7.1 based on the design
speed of 100 km/h and an AADT greater than 10 000 vpd.
b = is the barrier flare rate. A flared barrier alignment could be used as the 6:1
a batter slope can be flattened to 10:1 in the area between the carriageway
and the barrier. From Table 7.3, a shy line offset of 2.4 m is applicable for a
design speed of 100 km/h in this site. Therefore, with a road safety barrier
offset of 1.6 m from traffic lane, the barrier is within the shy line offset, and
Table 7.2 gives an applicable flare rate of 26:1 for rigid barriers. Thus
b/a = 1/26.
The end of the concrete barrier should be shielded using a suitable leading terminal (i.e. crash
cushion). The layout of the barrier for approaching traffic is illustrated in Figure 8.13.
Austroads 2013
— 185 —
Figure 8.13: Example 2: Layout of rigid barrier for approaching traffic using run-out length method
Similarly, the barrier layout for the opposing traffic is determined based on the clear zone width
since the distance from the edge of the traffic lane (22.0 m – 8.0 m = 14.0 m) is greater than CZ
(10.0 m). The same LON of 57.4 m was obtained when the barrier is located at the rear edge of the
traffic lane.
It should be noted that the lateral location of the rigid barrier should not be greater than 4.0 m from
the edge of traffic lane because of the likelihood of higher impact angles and resultant higher
severity of impact. If the rigid barrier is located at a lateral distance of 4.0 m from the edge of traffic
lane, i.e. L2 = 4 m, then b/a = 18:1, and the barrier LON is 37.3 m. However, consideration should
be given to the distance from the barrier to the face of the pier to ensure a vehicle roll allowance.
Under this circumstance, a second layer of WRSB may be installed near the edge of traffic lanes in
both directions to reduce the risk of errant vehicles colliding with the rigid barriers. As the clearance
from the edge of traffic to the edge of the concrete barrier is 4 m, there is sufficient space to
accommodate the relatively large deflection of this barrier system.
Length of need using angle of departure method

Using the angle of departure method (Section 7.4.2) a vehicle travelling at 100 km/h is assumed to
leave the road at a leading angle of 2.9° (Table 7.6). This equates to a rate of lateral shift of 20:1.
This angle is applied from the edge of the travel lane to the rear extremity of the bridge pier for the
adjacent traffic. As the rear extremity of the bridge pier is outside the clear zone, the clear zone
width CZ = 10.0 m is used instead, as shown in Figure 8.14.
It is assumed that the road safety barrier in this example is located 1.6 m from the edge of the
nearest travel lane. With the thickness of the barrier of approximately 0.6 m, the rear face of the
barrier is at 8.0 – 1.6 – 0.6 = 5.8 m from the rear face of the pier. The dimensions resulting from
the angle of departure method are shown in Figure 8.14.
The length of barrier between the points of need can be determined either graphically or by
applying geometry to the triangle enclosed by the vehicle trajectory and barrier flare rate.
The base of the triangle is (10 – 1.6) x 20 – 3 = 165 m long. As the vehicle trajectory is 1 in 20 and
the flare of the barrier is 1 in 26 the length of the side of the triangle along the road safety barrier
can be determined from equations for the height of the triangle as follows:
Austroads 2013
— 186 —
h = c / 20 = d/26 and as c = 165 – d the equation becomes:

(165 – d) / 20 = d/26 which can be solved to determine that d = 93 m.
The length of barrier required from the leading point of need to the hazard is therefore 93 m plus
the nominal 3 m long section of straight barrier (i.e. 96 m).
Figure 8.14: Example 2: Layout of rigid barrier for approaching traffic using angle of departure method
Similar to Section 8.8.1, the LON using the angle of departure method is greater than the LON
calculated by the run-out length method.
Austroads 2013
— 187 —
9 PROPOSED FUTURE RESEARCH

This section presents identified issues that require further research to complete the guidelines or
enhance future versions of the guidelines.
9.1 Development of Detailed Standard Barrier Designs

The BTTF members, in the meeting in December 2012, have agreed that the detailed standard
barrier designs would be developed after the revision of the AS 5100 is completed, so that possible
changes in the design loads and required minimum barrier heights as a result of MASH will be
incorporated.
The Manual for Assessing Safety Hardware (MASH) (AASHTO 2009) has replaced NCHRP Report
350, effective December 2010 in the USA.
Jurisdictions in the USA specify that highway hardware accepted prior to the adoption of MASH –
using criteria contained in NCHRP Report 350 – may remain in place and may continue to be
manufactured and installed. In addition, highway safety hardware accepted using NCHRP Report
350 criteria is not required to be retested using MASH criteria. However, new highway safety
hardware not previously evaluated must utilise MASH for testing and evaluation.
Changes were made in test matrices, test installation, test vehicles, evaluation criteria, test
documentation, and performance evaluation.
Changes in test matrices

 The small car impact angle is increased from 20° to 25° to match the impact angle used with
light truck testing.
 The impact speed for the single-unit truck test is increased from 80 km/h to 90 km/h to better
distinguish the TL-4 test from TL-3.
 The impact angle for length-of-need testing of terminals and crash cushions is increased
from 20° to 25° to match that for longitudinal barriers.
Changes in test vehicles

The size and weight of test vehicles have been increased to reflect the increase in vehicle fleet
size:
 The 820C test vehicle is replaced by the 1100C.
 The 2000P test vehicle is replaced by the 2270P.
 The single unit truck mass is increased from 8000 kg to 10 000 kg.
 The light truck test vehicle must have a minimum centre-of-gravity height of 28 inches
(710 mm).
Decisions still have to be made regarding testing of standardised bridge barriers in Australia.
However, it is recommended that the changes outlined above be incorporated in the design loading
for structural verification of barriers.
Changes of bridge barrier design loads in the revised AS 5100.2 (2004)

As a result of the replacement of NCHRP Report 350 by MASH, changes to the bridge barrier
design loads have been proposed in the future revision of AS 5100.2 (Table 9.1). The numbers in
parentheses are the values specified by the current AS 5100.2 (2004).
Austroads 2013
— 188 —
Table 9.1: Changes to bridge barrier design loads

Barrier performance level Unit Low Regular Medium High Special
Design transverse impact load Ft kN 150 (125) 300 (250) 600 (500) 750 1000
Design longitudinal impact load FL kN 50 (40) 100 (80) 200 (170) 250 330
Design vertical impact load Fv kN 22 (20) 100 (80) 300 (350) 350 450
Transverse length over which impact is applied Lt and LL mm 1.1 1.2 (1.1) 2.4 2.4 2.5
Vertical length over which impact is applied Lv mm 5.5 6.0 (5.5) 12 12 15
Height of impact load above asphalt He mm 0.6 (0.5) 0.9 (0.8) 1.2 (1.1) 1.4 1.4
If the above changes are accepted, there may be significant impacts on the current bridge barrier
designs, especially for Regular and Medium performance level barriers. The heights of most of the
current Regular performance level barrier designs are in the range of 800 mm to 820 mm, while
this range is 1100 mm to 1300 mm for the Medium performance level barrier designs. This needs
to be taken into account in the development of the standardised barrier designs.
It should be noted, however, that in the latest version of AASHTO LRFD Bridge Design
Specifications (2012), although MASH crash test criteria have been incorporated, there are no
changes to the design forces (Table A13.2–1).
The detailed standard barrier designs should include the following elements:
 cross-section
 module length
 reinforcement details
 anchorage details
 connection between modules.
9.2 Computer Simulation

Refer to Section 2.3 for a discussion of the current practice of computer simulation. The following
points should be further investigated for computer simulation of bridge barrier crash tests:
 assessment of the reliability of the computer simulation results
 acceptance of computer simulation instead of full-scale crash tests
 barrier acceptance criteria based on computer simulation results
 verification/calibration of test results of barrier designs with computer simulation.
Austroads 2013
— 189 —
9.3 Safe System Approach

Refer to Section 2.4 for an overview of the Safe System approach. Incorporating the principles of
the Safe System approach into the design of bridge barriers consists of:
 compliance with the current Australian Building Code
 a holistic approach to design of the whole bridge barrier system: bridge barrier, bridge
approach barrier, transition, and end treatment
 the interaction of the barrier with the bridge superstructure and the bridge approach:
on-structure barrier anchorages and off-structure barrier foundations
 special provisions for pedestrians and cyclists.
9.4 Barrier Test Program

Current practice shows that while compliance of a bridge barrier design to AS 5100 is assessed on
a theoretical basis, historical performance or engineering judgment, the need for a physical testing
program of bridge barriers is apparent due to the following issues:
 The Safe System approach requires the bridge barrier to be crashworthy.
 Theoretical assessment can only assess the structural adequacy of the barrier but cannot
fully assess the barrier’s performance.
 Methods based on overseas crash-tested barriers may not be applicable to Australian and
New Zealand conditions and vehicle fleets.
The aim is to develop a nationally consistent set of physically crash-tested bridge barrier designs
conforming to AS 5100. It is suggested that a barrier test program be developed in consultation
with jurisdictions and the BTTF, taking into consideration the agreed standardised barrier designs.
9.5 Update to Austroads Guide to Bridge Technology

It has been identified that the outcomes of this project should be incorporated into the Austroads
Guide to Bridge Technology (AGBT-09). Related sections in the AGBT-09 include:
 Part 3 – Typical Superstructures, Substructures and Components (Section 16)
 Part 6 – Bridge Construction (Section 20).
It was agreed at the BTTF meeting in December 2012 (Appendix E) that the AGBT-09 would be
updated after the revision of AS 5100.2 is completed, so that the effects of the changes in the
barrier design loads and required minimum heights are taken into account, as well as incorporating
final standardised bridge barrier details.
Austroads 2013
— 190 —
REFERENCES
AASHTO 1989, Guide specification for bridge railings, American Association of State Highway and
Transportation Officials, Washington DC, USA.
nd
AASHTO 1996, Roadside design guide, 2 edn, American Association of State Highway and Transportation
Officials, Washington DC, USA.
rd
AASHTO 2009, Manual for assessing safety hardware (MASH), American Association of State Highway and
rd
th
AASHTO 2012, Bridge design specifications, 6 edn, American Association of State Highway and
Australian Transport Council 2007, National road safety action plan: 2007 and 2008, ATC, Canberra, ACT.
Australian Transport Council 2011, National road safety strategy 2011-2020, ATC, Canberra, ACT.
Austroads 2006, Automatic vehicle classification by vehicle length, AP-T60/06, Austroads, Sydney, NSW.
Austroads 2010a, Guide to road design: part 6: roadside design, safety and barriers, AGRD06/10,
Austroads, Sydney, NSW.
th
Austroads 2010b, Austroads glossary of terms, 4 edn, AP-C87/10, Austroads, Sydney, NSW.
Bureau of Transport Economics 2000, Road crash costs in Australia, report 102, Bureau of Transport
Economics, Canberra, ACT.
th
Colosimo, V 2011, ‘Guide for bridge approach barriers’, Australian small bridges conference, 4 , 2011,
Melbourne, Victoria, Australia, Hallmark Conferences and Events, Hampton, Vic, 21 pp.
Department of Infrastructure, Energy and Resources 2007, Road safety barriers: design guide, DIER,
Hobart, Tas.
Department of Main Roads 2005, Road planning and design manual, chapter 8: safety barriers and roadside
furniture, QDMR, Brisbane, Qld, viewed 13 September 2011, <http://www.tmr.qld.gov.au/Business-
industry/Technical-standards-publications/Road-planning-and-design-manual.aspx>.
Department of Main Roads 2009, Road planning and design manual: interim advice: October 2009: chapter
8: safety barriers and roadside furniture, QDMR, Brisbane, Qld.
Department of Main Roads and Queensland Rail 2010, Design and selection criteria for road-rail interface
barriers, QR MCE-SR-007 revision A, QDMR and Queensland Rail, Brisbane, Qld.
Department of Planning, Transport and Infrastructure 2012, Standard drawings: concrete barriers and steel
beam barriers, DPTI, Adelaide, SA, viewed 02 May 2012,
<http://www.dpti.sa.gov.au/standards/structures>.
Department of Transport and Main Roads 2011, Structural drafting standards: volume 3: chapter 19: bridge
barriers, TMR, Brisbane, Qld.
Austroads 2013
— 191 —
Department of Transport and Main Roads 2012a, Design criteria for bridges and other structures, TMR,
Brisbane, Qld.
Department of Transport and Main Roads 2012b, Standard drawings roads manual: part 18: bridges, TMR,
Brisbane, Qld.
ENSIS n.d., Design and verification of barrier for bridge, ENSIS, Venice, Italy, viewed 20 June 2012,
<http://www.ensis-ve.com/docs/01%20Barriere%20da%20Ponte_EN.pdf>.
Lynam, DA & Kennedy, JV 2005, The travel of errant vehicles after leaving the carriageway, report no. PPR
298, TRL Ltd, Crowthorne, Berkshire, UK, viewed 15 August 2011,
<http://www.dft.gov.uk/ha/standards/tech_info/files/Errant_vehicle_review_PPR_2.pdf>.
Main Roads Western Australia 2009, Structures engineering design manual, document no. 3912/03, MRWA,
East Perth, WA.
Main Roads Western Australia 2012a, MRWA supplement to Austroads Guide to Road Design - Part 6,
document no. D11#38472, version 1C, MRWA, East Perth, WA.
Main Roads Western Australia 2012b, Standard contract drawings: traffic barriers and balustrading for
structures, MRWA, East Perth, WA, viewed 28 March 2013,
<https://www.mainroads.wa.gov.au/BuildingRoads/StandardsTechnical/MainRoadsDrawings/Pages/St
andard_Contract_Drawings.aspx>.
Pernetti, M, Scalera, S & Cibllis, G 2007, ‘Development of validated finite element model of an articulated
th
truck suitable to simulate collisions against road safety barriers’, 6 European LS-DYNA users’
conference, Gothenburg, Sweden.
Queensland Rail 2010, Collision protection of supporting elements adjacent to railways, MCE-SR-012,
revision B, Queensland Rail, Brisbane, Qld.
Roads and Maritime Services 2011, Supplementary to Austroads Guide to Road Design – Part 6: roadside
design, safety and barriers, RMS, Sydney, NSW.
Roads and Maritime Services 2012, Standard bridge drawings for barriers and railings, RMS, Sydney, NSW,
viewed 03 May 2012,
<http://www.rta.nsw.gov.au/doingbusinesswithus/designdocuments/bridge_drawings.html>.
Roads and Traffic Authority 1996, Road design guide: section 6: safety barriers for roads and bridges, RTA,
Sydney, NSW.
Roads and Traffic Authority 2008a, ‘Road design guide: section 6: design for errant vehicles: hazard
mitigation and safety barrier design’, unpublished draft, RTA, Sydney, NSW.
Roads and Traffic Authority 2008b, Design of bridge supports for collision load from road traffic, bridge
technical direction BTD 2008/07, RMS, Sydney, NSW.
Roads and Traffic Authority 2009, Design of replacement traffic barriers on existing bridges, bridge technical
direction BTD 2007/08, revision 1, RMS, Sydney, NSW.
Ross, HE, Sicking, DL, Zimmer, RA & Michie, JD 1993, Recommended procedures for the safety
performance evaluation of highway features, NCHRP report 350, Transportation Research Board,
Washington DC, USA.
Texas Department of Transportation 2006, Bridge railing manual, TxDOT, Austin, Texas, USA.
Austroads 2013
— 192 —
Transit New Zealand 2004, Bridge manual: section B: bridge side protection, Transit New Zealand,
Wellington, New Zealand.
VicRoads 2001, Selection requirements for bridge traffic barriers, bridge technical note BTN 2001/007,
VicRoads, Kew, Vic.
VicRoads 2005, Improving existing bridge barriers, bridge technical note BTN 2005/001, VicRoads, Kew, Vic.
VicRoads 2009a, Protection of bridge supports, bridge technical note BTN2009-001, v1.0, VicRoads, Kew,
Vic.
VicRoads 2009b, Guidelines for bridge approach barriers, bridge technical note BTN 2009/002, VicRoads,
Kew, Vic.
VicRoads 2010, The performance safety barrier treatments at bridge approaches, road design note RDN 06-
07, VicRoads, Kew, Vic.
VicRoads 2012, Standard drawings for guard fences and barriers, VicRoads, Kew, Vic, viewed 2 May 2012,
<http://www.vicroads.vic.gov.au/Home/Moreinfoandservices/RoadManagementAndDesign/DesignStan
dardsManualsNotes/RoadDesign/GuidelinesStandardDrawingsForRoadworks.htm#guard>.
Williams, WF, Abu-Odeh, A, Alberson, DC, Menges, WL & Haug, RR 2007, Bridge rail analysis and crash
testing, report no. FHWA PA-2007-019-BRACT, Texas Transportation Institute, College Station,
Texas, USA.
Standards Australia
AS 5100 Set-2007, Bridge design set.
AS 5100.1-2004, Bridge design: scope and general principles.
AS 5100.2-2004, Bridge design: design loads.
AS 5100.5-2004, Bridge design: concrete.
AS 5100.1-2006, Supplementary 1.
AS 5100.2-2007, Supplement 1.
AS 5100.1-2010, Amendment 1.
AS/NZS 3845-1999, Road safety barrier systems.
Austroads 2013
— 193 —
APPENDIX A RECOMMENDATIONS OF 2006 WORKSHOP

A.1 Background
The publication of the new AS 5100 bridge design Code saw the introduction of substantial
changes to the provisions for the design of bridge barriers. The extent of the changes created a
degree of concern amongst the bridge engineering profession and the Austroads Bridge
Technology Group decided that a public workshop that could provide information on the new
standards as well as collate the issues concerning the design engineers would be valuable. The
workshop was held in conjunction with the Austroads 6th Bridge Conference Workshop on Bridge
Barriers.
The Bridge Barrier Workshop was held on 12 September, 2006 at the University Club on the
grounds of the University of Western Australia. The workshop was chaired by Dr. John Fenwick,
Executive Director Technical Services of Queensland Main Roads and was very well attended with
a total of 73 participants including invited guest speakers.
A.2 Guest Speakers

The first part of the workshop consisted of four presentations by invited guest speakers to provide
background information and reference points for the general discussion that followed. The four
presenters and their topics were as follows:
 Geoff Boully, (Senior Bridge Engineer VicRoads, AS 5100 code committee) provided
information on the background and reasons for the development of the new Code and
provisions for bridge barriers.
 Raphael Grzebieta (Principal Forensic Engineer, DVExperts International) provided
information on the assessment of barriers, crash testing and dynamic simulation.
 Jan Karpinski (Road Design Engineer, Main Roads WA) gave a presentation on the interface
between road and bridge barriers, and transition lengths.
 Geoff Bowman (Technical Director Maunsell) provided a perspective from the designer’s
viewpoint.
A.3 Introduction
The attendees mostly appeared to fall into one of three general categories:
 Designers who have not used the Code barrier provisions or have no knowledge of the Code
requirements and had attended the workshop in the hope that they might learn what was
required.
 Code (barrier provision) users seeking affirmation that they have applied the code provisions
correctly.
 Code users who had some concerns or reservations about certain provisions and were
seeking a deeper understanding of the barrier provisions in the Code.
The workshop was an information sharing session with the further intention of collating together
issues and opportunities for improving the Code and general practice in the future. The
proceedings from the General Discussion were reviewed to capture the main points.
Austroads 2013
— 194 —
A.4 General Discussion Session

It was reiterated by the Chair that one of the outcomes of the forum was to provide an opportunity
to air these concerns and resolve these issues where possible. It was pointed out that Australian
society was changing and in following an American trend, it was becoming more litigious. The
holistic approach adopted in the Code for bridge barriers was a more responsible approach to
barrier design.
It was clear from the nature of the questions that there are a number of areas where greater
education on the intent and requirements of the Code is required and where clearer guidance
would assist greatly in reducing the confusion and application of the Code requirements.
The discussion session highlighted a number of areas of concern over the new Code for bridge
barrier design and a high degree on interest in understanding the requirements and being able to
address the requirements in the correct manner.
The areas of concern may be broadly categorised as follows:
 clarification of the Code requirements or intent
 additional background information to promote better understanding
 advice on sources of supplementary supporting information
 supply of methodologies and examples to illustrate standard practice for implementing the
Code requirements
 preparation and supply of standard designs.
In addition there were a number of suggestions and ideas on improvements made which have
been captured with the Appendix A.7 – Summary of Issues.
Attendees expressed the view that many designers felt frustrated by the apparent ‘black box’
approach to the determination of the performance level for the barrier based as it is on a
non-transparent methodology (in Appendix B of AS 5100.1). Further, due to the requirement for the
use of only crash-tested barriers, there is a lack of options to select from and little realistic
opportunity to design or develop an alternative.
AS 5100 also requires that bridge barriers are appropriately transitioned to the approach or
roadside barriers. The comments received during the workshop indicated that there was a
perceived need for the Code and asset owners to supply more guidance and/or more information
in this area.
A.5 Summary
The large number of attendees is an indication of the interest in this topic amongst the bridge
design engineers. The participants included many leading bridge design practitioners from around
Australia.
It is considered by the conveners that there was a clear and broad consensus arising from the
workshop to request Austroads and the road authorities to pursue a number of improvement
actions. These are:
 Develop and publish a set of Austroads bridge barriers guidelines or a ‘user guide’. The
guidelines would cover a wide range of information including such aspects as:
— new bridges and retrofitting to existing bridges
— risk assessment guidance and examples
— performance level selection guidance, procedures and examples
— barrier design guidance and examples
Austroads 2013
— 195 —
— computer simulation guidelines and examples

— Austroads approved tested barrier systems
— Austroads other accepted bridge and approach barrier systems
 Develop and publish a suite of national standard barrier systems.
In addition there was broad support from consultants and contractors that the asset owner should,
in the preparation of design briefs, consider specifying:
 Performance requirements for the barrier and provide more guidance on bridge approach
treatments.
A.6 Recommendation
That the Bridge Technical Group prepares a set of development and research proposals for
Austroads Council to consider, which addresses the concerns and issues identified in the
workshop.
A.7 Bridge Barrier Issues Raised in the Workshop

Category Issues of concern
More Information Request for standard bridge barriers designs and guidelines on road authority websites.
Protection of bridge piers – not well covered in AS 5100, how to determine the required performance level and at
what off-set.
Are there any crash-tested W-beam barriers for the Low PL category?
What is AS 5100 based on – US LRFD.
Discrepancy in barrier design forces – AS 5100 forces to be adopted.
Supporting information Design forces published in AS 5100 based on average force over 25 ms not the peak force.
Guidance on the extent of modifications to crash-tested barriers allowed – legal issues.
Additional guidance What is an acceptable level of risk i.e. acceptance criteria in determining the omission of a barrier or the selection
of one.
When can a no-barrier be used to ‘protect’ a bridge pier.
Guidance on upgrading barriers when bridge upgrade is carried out on one side only.
Guidance on shielding the ‘elephant trap’, i.e. vehicle runoff between two parallel bridges.
Request for guidance and examples on how to design a barrier.
Request for guidance and examples on how to design a transition barrier – appropriate length.
Request for guidance and examples on how to design an approach barrier – appropriate length.
Request for guidance and examples on how to design a retrofit/upgrade barrier.
Guidance or principles for ranking outcomes in decision making e.g. priority for protecting innocent 3rd parties.
Request for a risk assessment procedure with examples including B/C analysis methodologies.
Future improvements Suggestion that a barrier test program be developed and costed with the expectation of the road authorities to fund
the program.
Combination steel and concrete barriers are based on crash-tested systems from the USA.
Road authorities should develop a common design for barriers, that is, a barrier standard for various design types
e.g. Super T standard.
Request that road authorities as the client must specify the required barrier performance level and thereby identify
the risk.
Review of AS 5100 to consider including a 750 kN design force.
Initiate an in-service accident data management measurement – via traffic cameras.
Interchange road and bridge personnel on roadside and bridge barrier committees.
Austroads 2013
— 196 —
APPENDIX B MEETING OF EXPERT GROUP 2010

A meeting between the project’s working group members was held at Hilton – Melbourne Airport
on 12 October 2010. The following notes of the meeting were made by Gary Veith and Andrew
Zivanovik (ARRB).
B.1 Present
Nigel Powers (VicRoads), Vince Colosimo (VicRoads), Joe Canceri (RTA), Adam Pyrzakowski
(DTEI 2), Gary Veith (ARRB), Andrew Zivanovic (ARRB). And on tele-conference: Kazik Rassalski
(TMR) and Nishantha Palliyaguru (TMR).
B.2 Welcome and Introduction

Nigel explained the background to the project and its importance, a lot of work has been going on
over the years and hence Austroads have funded the project to bring it together. Standardisation
should be good for consultants and contractors who work across jurisdictions. Nigel noted that
agreeing on a standard may be difficult as times but asked everyone to remain open minded to
options, suggestions and outcomes.
Using a PowerPoint™ presentation Gary covered the requirements of the proposal and contract
note including the proposed program and outputs. The importance of stakeholder views,
expectations and active support for the project in providing information, comments, and resolution
of standard designs was emphasised.
Some initial discussion took place and comment was provided:

 All authorities use AS 5100 and are moving toward developing guides around that.
 Not everyone follows AS 5100; therefore it may require a supplement/amendment?
 Suggestion of teleconference every 2–3 months and meet in person to resolve key issues.
 SA has invested significant effort into the development of barrier types over the past two
years.
 Barrier costs can be significant (SA).
 Barriers associated with reinforced soil walls should be included in AS 5100 and bridge guide
(RTA).
ACTION: Nigel to follow up with Austroads Bridge Technical Review Panel on whether to include
this as part of the project scope.
 Pier protection may depend on how much the project manager is prepared to pay (SA).
There is no guide for the protection barrier and no distance from travel lane to the barrier.
 AS 5100 requires pier to withstand load or a barrier to be provided – distance not mentioned
(RTA).
 Pier protection is an area requiring improved guidance (RTA).
 A most important area is retrofitting current bridges – requests are received from regions to
retrofit damaged barriers and cognisance must be given to strength of existing
superstructures (TMR).
— There will be some differences in detail but we should have general standardisation.
2
Department for Transport, Energy and Infrastructure, South Australia (DTEI) became Department of Planning,
Transport and Infrastructure, South Australia (DPTI) during the course of this project.
Austroads 2013
— 197 —
— Research should investigate computer models regarding the progressive collapse of

barriers.
 There will be a number of full-scale tests done in USA.
 Victoria has used simulation to test ‘low level’ barriers.
 Good articles are available on TL-4 testing (SA).
 Cannot cater for tankers in Australia.
 Same level of protection should be applied to road and bridge barriers (i.e. apply same test
vehicles) (TMR).
 Pier protection – should design for impact loads – comes into the road arena (TMR).
 RTA insists that the road barrier details be shown on the bridge drawings.
 Difficult for practitioners to work over state boundaries.
 If we cannot get full agreement, work from common principles; tackle grey areas first, then
look at existing practices.
 Models should be verified against actual tests (US).
B.3 Role of the Group

Nigel explained the roles.
Nigel
 Project Manager, and is ultimately responsible for its delivery.
 Liaise with Bridge Technology Group.
 Report to Program Manager.
 Henry Luczak will have role of assistant Project Manager.
Gary
 Project Leader.
 Collate and process information.
 Organise and run future meetings.
 Coordinate and deliver project outputs including reporting.
Expert group
 Provide information and comment throughout project.
 Provide the technical input required on moving toward standardisation.
 Resolve issues and assist the development of standard guidelines through workshops and
other means/communications.
B.4 Jurisdictional Expectations

VicRoads
 Focus on the roadside aspect – not so much the structural design of them but what is used
and where it is placed. Transitions mentioned.
 Design in VicRoads – VR only does a small percentage of barrier design – most is done by
consultants and this has led to a variety of barrier designs.
Austroads 2013
— 198 —
 VR does not use standard drawings rather, base new designs on previous similar designs.
Yield theory and hinging? Consultants tend to over-design often specifying twice the required
thickness.
 Practice adopted at some point where a load applied and minimum distribution assumed to
adjacent barrier panels leading to heavier, expensive designs. More recently design is based
on US design where load spread over a number of panels – barrier still performs but is lighter
(e.g. 4 mm rather than 5 mm thick tube) and yields a bit more. Rotation is available to the
designer – results in barriers not much different to the 60’s and 70’s standard where barrier
fails over a number of panels.
 Reference made to a 1997 standard to design a multiple performance barrier. A lot of
technical notes are available (Vince and Dennis Eastwood) e.g. size of members, new one
regarding bridge approaches, performance level guidelines.
 Standard spec sections are used for contracts - how to design barriers. There is new US
system of testing. VR test requirements based on US and Europe (mainly UK).
RTA
 Need to have a clear distinction as to where we use bridge type barriers compared to road
traffic barriers.
 Include areas where we currently do not have good guidance e.g. pier protection, soil walls,
Transition area – bridge-road hybrid.
 Re AS 5100 – In NSW when looking to composite barrier of concrete with steel top rail the
member sizes are still bigger – probably because solid state design is used rather than a
more plastic approach. Different design of composite than in Victoria.
 RTA has a set of standard drawings.
 Also provide 1300 high barriers for cyclists (VicRoads also does this).
 Contracting out – cite standard drawings and Technical Directions. Where a clear Technical
Direction is not available clauses are provided in the scope of works.
 Standardisation of precast sections of concrete barrier e.g. outside is a shell and the front is
cast-in-situ. Style where space is provided behind the barrier to allow movement under a
transverse load, and hence no forces are distributed to the deck. Attention to detail of starter
bars is important to avoid them clashing with the shell. Comment by VR – compaction of
in-situ concrete can be an issue.
 Retrofitting of new barriers to existing bridges – concern about legal situation. It should be no
issue as long as the design etc. is well investigated and documented – RTA has a Technical
Directive on the subject.
DTEI
 Have an installation of barriers guide, generic drawings and contract drawings as well.
 Systems – had retrofitting system for a long time with heavy?. Good performance after 20
years – based on technical articles from the USA.
 TL-3, TL-4 and TL-5 systems for new bridges. TL-5s are all 1300 high. Have a three rail RHS
system and a composite design.
 Now moving to single slope concrete (with steel rail on top?
 Selection – Try to use AS 5100. Generally they are all ‘regular’ performance level.
 In terms of expectations – always hoping for new ideas; however DTEI has developed a lot
of designs/information over last few years so well situated. Hence it may be problematic to
have to change – i.e. introduce new approach; training etc.
Austroads 2013
— 199 —
TMR
 Looking for improved ways of doing things.
 Categorisation of bridges e.g. New bridge over rail involves QR; all other bridges are regular
performance level for traffic and speed as per AS 5100.
 Retrofit - standard regular barrier RHS 150 mm wide and 6 mm wall thickness – Work on the
assumption that collapse is progressive and there is to be no damage to the superstructure.
 Standard drawings are available; some under review e.g. transitions on bridge approaches.
 Have a team in bridge design for an alliance project. Provide a lot of
information/requirements to be given to contractor such as scope of works technical
documents and contract documents.
 DTEI provides barrier information as part of the safety aspects of contracts – design posts to
fail at a particular point (consistent with AS 5100).
B.5 2010–11 Work

B.5.1 Consultation with Consultants and Contractors
 Consultation to be by letter/email.
 Make it clear that we are not going to get them actively involved.
 Ask them about their expectations – what they want to get out of it – problems/issues.
 Ask them if they have any barriers/treatments that they would like to submit i.e. a future
standard.
 How to contact them? RTA, DTEI and TMR all have a list of panel consultants/contractors
and can recommend selected ones re barriers.
ACTION: Gary to send out letter notifying relevant consultants and contractors or project.
B.5.2 AS 5100
 Vince Colosimo explained that AS 5100 followed a US change and left out TL-3 altogether –
replaced TL-3 with TL-2, TL-4 and TL-5. Reason for no TL-3–TL-4 has the same strength but
deals with both cars and 4WD vehicles as well.
 People are now asking why they can’t use TL-3 – practice is that authority includes a clause
to allow TL-3 where a road authority believes it is needed.
 Regarding risk assessment – Determine cost difference – work out likelihood – compute a
cost /benefit analysis. Refer to Chapter 6 of RTA Road Design Guide.
 Need to request deficiencies in AS 5100 and suggested amendments from jurisdictions.
 Whilst contract note includes only a review of AS 5100 Part 1, meeting was of the view that
Part 2 should be included.
 The outcome of this review will feed into the AS 5100 group to aid a review of AS 5100 that
is soon to commence.
B.5.3 Risk Assessment

 AS 5100 has been in use since 2004 – some complaints but all SRAs use it.
 Regarding retrofitting barriers to existing bridges or reinforcing existing barriers – tend to give
them a label i.e. ‘low’, ‘deficient low’, ‘regular’ with some deficiency – move away from
AS 5100.
Austroads 2013
— 200 —
 The procedure in AS 5100 results in a ‘regular’ performance level in about 80% of cases.
SRAs can provide higher level on particular routes – if a ‘medium’ level is desired/required it
can be provided. Use a B/C approach as per Chapter 6 of RTA Road Design Guide.
 The procedure in AS 5100.1 has only been questioned in relation to its application on bridge
approaches (Vince).
B.5.4 Crash Study

 The involvement of Raphael Grzebieta in crash investigation was noted.
 It is likely that the number of crashes into bridge barriers is relatively low given their length in
relation to the length of road on the network.
 General feeling was that crash information from SRA databases (i.e. based on DCAs or
police report forms) would not yield any worthwhile information. Had been done before by
VicRoads and the RTA, and nothing useful came out of it. Information on fatal crashes from
police reports for coronial enquiries could be helpful but access to that information is
generally not available.
 Mentioned that risk assessment for bridges is based on an encroachment rate derived from
the number of crashes that generally occur on roadsides – 0.0003 typically used, RTA
determined 0.00027.
 US study found that bridges have three times the encroachment rate as general roadsides.
 AS 5100 based on US data.
 It was suggested that road safety departments have books of crashes which could be
assessed in conjunction with barrier type etc. It may be worthwhile trying to obtain pictures of
accidents in particular those showing the resultant damage to the bridge barrier. However,
without information on details of crash vehicle, angle of impact, speed of impact etc. photos
are of limited use.
 Intent of project was to get a full picture of the crashes into bridge barriers. Need to talk to
Road safety Departments regarding information and a way forward.
B.6 2011–12 Work

B.6.1 Simulation Software
 Noted that Raphael Grzebieta uses a US model called Dyna.
 Models are getting accurate to the point where they reflect test results – however, it is a time
consuming process that may take 2–3 years to model one vehicle and then one also has to
model the barrier.
 MUARC and overseas was mentioned but generally too expensive to justify.
 Nigel mentioned that Swinburne University was developing ‘Virtual Networks’ that was
promising for the future.
 Vince suggested that he had heard that simulation can be accurate and cheap. Cost of
simulation vs testing is relevant.
 Guidelines should focus on models available and where and when to use modelling. The
intention was always to use simulation where the barrier design was a little beyond what was
required and the use of full-scale test could not be justified.
ACTION: Nigel to follow up with Austroads Bridge Technical Review Panel to clarify the project
scope in relation to testing and simulation.
Austroads 2013
— 201 —
B.7 Some General Comment

 Comment was made on installation and maintenance issues – the need to enquire as to the
principles/design practices are undertaken in design to minimise any difficulties in installation
and maintenance activities – i.e. cater for the ease of construction and maintenance.
 MASH was mentioned – reason why VicRoads barriers a bit higher – new test vehicle with
higher centre-of-gravity.
 Intent is to strive for the future rather than justify a barrier based on current stock. New
barriers need to conform by testing or by analysis to AS 5100.
 There is a need to reduce the number of barrier options. It is anticipated that we may initially
have more than 20 barrier types but may/should not be difficult to reduce to less than half
that number.
ACTION: All group members consider possible acceptance criteria or how we could rationally
decide on standard barrier designs if the need arose and discuss at next meeting.
 Pedestrian barriers/screens should not really form part of this review.
 As an indication of improvements to tested barriers, a few are available (refer to Vince’s
papers).
 Acceptance Criteria? RTA - mentioned that they rely on AS 5100.1, clause 10.4 (d), a
method based on evaluation of existing barriers.
B.8 Conclusion
A couple of workshops will be required to clarify information and resolve standardisation issues. A
first workshop is required early in 2011 to enable jurisdictions to present their barrier designs,
basis, use, advantagesrco, disadvantages etc. in order to and promote discussion of issues.
Gary to request road authorities to provide the comment and information required for the project
and allow adequate time for response. Road authorities should put forward barriers that they are
comfortable with for consideration in standardisation discussions.
ACTION: ARRB to put together minutes and circulate to all group members.
ACTION: Gary to send out letter to all road authorities requesting information relative to the project
including consultant and contractor contacts, barrier design standards, processes, procedures,
protocols etc.
Austroads 2013
— 202 —
APPENDIX C COMMENTS FROM CONSULTANTS AND

CONTRACTORS
Requests have been sent to some main consultants/contractors to gather their comments and
information on current practice in design and construction of bridge barriers. The request covered
the following areas:
 bridge barrier design procedure
 determination of bridge barriers’ performance level
 risk assessment procedure
 current key issues in design of bridge barriers
 comments on the current version of AS 5100 (2007) for design of bridge barriers.
Several responses have been received. The comments are briefly summarised below.
Worley Parsons
An email from Min Khong, Chong on 1 November 2010 (communication with Robert Scanlon
MRWA) notes that:
 For crash testing requirements, AS 5100.1–2004 refers to TRB-NCHRP Report 350 or ‘other
appropriate’ standards as determined by the authority. While Table 10.4 of AS 5100.1–2004
shows the ‘crash test vehicles and criteria’ for different performance level barriers with
reference to the TRB-NCHRP Report 350 test levels, it is not clear how this came about
or how to relate to other ‘appropriate’ standards, considering that AS 5100.1–2004 classifies
the performance levels according to traffic volume, nature of crossing (i.e. over highways,
etc.), site-specific, unusual conditions, etc. (Clause 10.5, AS 5100.1–2004).
 The classification of performance levels in AS 5100.1–2004 could be confusing and even
arbitrary to some extent. BS 6779–2:1991 and the more recent Euro version
BS EN 1317–2:1998 classify containment levels primarily with respect to types of vehicles
(vehicle mass), and vehicle speeds. In unusual sites (high risk to loss of lives), the minimum
height of containment could be specified. The project group might wish to review the
classifications since they refer to the NCHRP Report 350 for testing requirements.
 It would help to have the testing requirements, standard limiting geometric dimensions of
tested barriers, and other necessary requirements to be clearly specified either within the
code or appended rather than referring to external documents.
 It might be useful to indicate the minimum equivalent static nominal load to which the bridge
barrier shall be designed, to enable checks, if necessary, to be carried out on its capacity
and the load effects on the foundation.
 The Euro codes are setting a common platform for standard practices across Europe with
local annexes to suit the individual countries. The Austroads project group might like to
compare notes. It would help to come to some common agreement.
Austroads 2013
— 203 —
Parsons Brinkerhoff
In his email on 17 February 2011, Richard Lipianin provides the following comments on AS 5100
for bridge barriers:
 The determination of an appropriate level of containment at a bridge should be determined
by a road safety expert as part of the need for barriers for the whole road system rather than
leaving the choice of containment level at a bridge to structural engineers or bridge design
specialists within road authorities.
 There should be a link between AS 5100 and AS 3845 for the design of bridge barriers.
 For bridge approach barriers, some authorities, notably VicRoads, require barriers
approaching bridges to be designed to cater for the same containment level as those on the
associated bridge. This seems to be a reasonable requirement considering that an errant
vehicle leaving a carriageway just before a bridge is likely to end up in a similar predicament
as it would have, had it breached a bridge barrier (i.e. on a railway, freeway, etc.). It is
therefore appropriate that warrants for such barriers be addressed along with those for other
road barriers, in AS 3845 as noted above.
 From a structural perspective, the biggest challenge in designing bridge approach barriers is
achieving the required containment capacity at the start (i.e. the end remote from the bridge).
There is usually a transition at this point from a non-rigid barrier (guardrail or similar) to a
rigid concrete structure that is intended to cater for the full containment load. As this point is
at the limit of the length of barrier intended to cater for a higher risk situation, it is reasonable
to assume the level risk is lower than at the bridge end of this barrier. Accordingly, the
transition in the containment level between that provided by the non-rigid barrier and the full
containment level would also seem to be a reasonable approach. This transition could be
defined either by specifying a lower containment level over the first (say) 6 m, or specifying
that the required containment must be achieved within (say) 6 m and that the barrier
preceding that point is built to the same design as the rest of it. The intention is to avoid
building an unnecessarily strong foundation at the remote end of the barrier while not
materially compromising its capacity.
 In terms of testing, AS 5100.1 Clause 10.4 requires that all barriers must be proven,
essentially by testing, to also comply with TRB-NCHRP 350. This sets criteria which include
deceleration limits on vehicle occupants. Current practice is that barriers are designed to
equivalent static loads, without testing and with no regard to impact on occupants. This
should be addressed.
 The use of precast concrete components means that connections have to be made between
the barrier sections and the deck and, for high containment level barriers, between the
barrier sections. With the relatively thin decks that are generally cast over precast beams or
panels there is considerable doubt that the so called ‘kerb bars’ are adequately anchored in
the deck, even though they are deemed to be so under some interpretations of AS 5100.
Likewise the detail used to transfer the load from the reinforcement in the barrier sections to
the ‘kerb bars’ is not one that would be generally acceptable in reinforced concrete
structures. Therefore the only way to verify that the required containment levels are actually
being achieved is to test typical barriers. Testing individual barrier sections under static loads
will provide information on the adequacy of structural components to resist nominated
containment loads, but only testing barrier systems for containment of the design vehicles
will verify the adequacy of the barrier.
 In terms of standardisation, one of the bugbears of designers and contractors is the need to
design barriers for every project and have the design verified independently. It would be of
great benefit to both designers, contractors and asset owners if standard details were
developed, ideally after testing, for each barrier type (concrete parapet, post and rail or a
combination of the two) for each containment level.
Austroads 2013
— 204 —
 Failure mechanism in post and rail barriers is another issue. There is an inconsistency
between Clause 3.8 of AS 3845 and best practice for the design of rigid barriers with steel
rails. This clause requires that anchors for the posts be the first point of failure. A failure of
this type is invariably a punching shear failure at the top of the concrete barrier which means
that all load resistance at the post associated with the anchor failure is lost. Conversely
failure of the post in bending does not reduce its capacity and is a means to absorb some of
the energy of an impact.
 Retrofitting existing bridges: many existing bridge decks are not strong enough to withstand
loads imparted to them from barriers designed to contemporary standards. Retrofitting old
bridge decks using stronger barriers is therefore likely to lead to deck failure under barrier
impact loads. To overcome this, barriers can be custom designed to match the available
deck strength. One option is the use of semi-flexible barriers with fixing strength limited to the
deck capacity and made continuous to contain a vehicle on the bridge (i.e. similar functioning
to wire rope barriers and/or guardrails). Systems of this type (e.g. SISTEMA) have been
tested and used overseas. This option should be included in the range of suitable barriers.
Retrofitting should consider the whole system, including approaches and transition to the
bridge barrier.
ARUP
David Morris, in his email on 22 March 2011, provides some information on current practice in
Queensland and Victoria as summarised below:
 Typically the road authority specifies the performance level of the barrier. Where (as an
alternative example) the road authority is not the client, the barrier containment levels may
be determined by the designer based on the method in AS 5100.
 The selection of bridge barrier performance level generally follows AS 5100 procedure, which
is based on descriptive assessments of the site-specific condition. Onsite risk assessments
and B/C analyses may also be carried out to verify the selection.
 The load specified in AS 5100 are significantly higher than historical impact loads which were
in the order of 70 kN.
 There is limited data available when designing for very low speeds (i.e. less than 50 kph)
such that for car parks or ambulance ramps, which can lead to over-designed barriers.
 AS 5100 is not clear about extending existing barriers and whether the required barrier
length is needed beyond the last expansion joint or if additional sections can be added on as
required.
SMEC
In his email on 14 March 2011, Vishnu Balakrishnan comments that:
 The code (AS 5100) should have a separate section to cater for off-structure barriers.
 Poor guidance in AS 5100 – refers to AASHTO but there is no clear direction provided in
AASHTO for combined steel rail-concrete barriers. AS 5100 should clearly spell out what it
needs rather than refer us to external codes.
 No guidance on termination of barriers – currently no barrier is expected to comply at the
ends of the expansion joint.
 The VicRoads detail of tying in barriers should be reviewed. The detail achieved on-site is
not satisfactory while it still meets code requirements. This detail should be replaced.
 No guidance for single slope barriers with a combined steel rail on top as to where the steel
rail is to be located.
Austroads 2013
— 205 —
GHD
Kabir Patoary (email on 21 February 2011) provides following comments:
 AS 5100.1 cl 10.2 and cl 10.5 specifies that the performance level shall be determined by the
relevant authority. It is however not clear who is that relevant authority? Is it the client,
contractor or consultant or road authority? AS 5100.1 cl 10.5.3 to 10.5.5 provides guidelines
for barrier selection based on terms of ‘low traffic volume, general traffics, major roadways,
high occupancy land use etc.’ The definition of these terms is not clear in the code. The code
also does not provide any clear guidelines about the length beyond which the required
performance level is no longer applicable.
 It will be very good step to rationalise the procedure for undertaking risk assessment across
the industry for a typical bridge situation.
 Current practice is that in majority of the cases, the road authority specifies the required
performance level of barrier and their extent. The consultant designs the barrier to achieve
the specified performance level. In some instances, the consultant does the risk assessment
in consultation with the client following AS 5100 guideline to select the appropriate
performance level.
 Recommend to have common guidelines across the country for bridge and approach
barriers.
Austroads 2013
— 206 —
APPENDIX D EXPERT CONSULTATION

This appendix presents the comments obtained from the working group and the responses from the project delivery team.
Comments Responses
From: Nishantha.S.Palliyaguru@tmr.qld.gov.au [mailto:Nishantha.S.Palliyaguru@tmr.qld.gov.au] Point (d): Assuming kerb is of 0.25 m minimum width and lane width is 3.2 m. If condition (e) applies then
Sent: Friday, 14 October 2011 3:32 PM the width between kerbs will not be less than 7.9 m (2 x 3.2 + 2 x (1 – 0.25)) for a two-lane bridge or 4.7 m
(3.2 + 2 x (1 – 0.25) for a single-lane bridge. Condition (e) will be covered in (d). It should be noted that
Criteria for site condition CAT1 kerbs are required for no barrier case.
Point (a) The bridge or culvert is less than 1.5 m above the ground. Clause 9.4: 0.6 m is the required minimum edge clearance from the edge of the traffic lane to the face of the
safety barrier. However, to be eligible for no barrier, it needs to be 1000 mm min.
Having a bridge barrier or traffic barrier in culvert may not be feasible due to crown of the culvert not having
adequate capacity. However, a guardrail option on culvert is possible. Point (h): Should refer back to Cl. 1.5.7 Level 4 Barriers in 1992 Austroads Bridge Design Code, where the
condition (h) is the first criteria. It implies that no barrier case is only considered for sites with flooding debris
Point (d) The width between kerbs is not less than 6.5 m for a two-lane bridge or 4.0 m for a single-lane
problem AND have favourable site conditions.
bridge, and Point (e) The edge of the bridge is at least 1.0 m from the edge of traffic lanes.
1.0 m edge distance means 2 m traffic lane for single-lane and 4.5 m for two-lane bridges. Nominally traffic
lanes are 3+m. Hence, the code is not logical. Suggest to revise the criteria for site condition CAT1 as follows:
What we have suggested is 600 mm edge clearance as per Table 9.5 and suggested 3.2 m lane. This Traffic barriers may be omitted where the provision of barriers prevents the passage of debris or the barriers
means 4.4 m for one lane and 7.6 m for two lanes. This generally agrees with Clause 9.4, 4.2 m to 4.5 m for would be frequently damaged by heavy debris or both, and all the following criteria for site condition CAT1
one lane. apply:
(a) The bridge or culvert is less than 1.5 m above the ground.
Hence suggested updates are: (b) Traffic volumes are less than 150 vehicles per day.
 The width between kerbs is not less than 7.6 m for a two-lane bridge or 4.4 m for a single-lane bridge. (c) Bridges with an essentially straight alignment (e.g. with a radius of horizontal curvature of greater
 Edge of the bridge is at least 0.6 m from the edge of traffic lanes. than 1500 m) and the road approaches have a sight distance greater than the stopping distance.
(d) The width between kerbs is not less than 7.9 m for a two-lane bridge or 4.7 m for a single-lane bridge.
Point (h) The provision of barriers prevents the passage of debris or the barriers would be frequently (e) The location is without anticipated pedestrian traffic.
damaged by heavy debris or both. (f) Any water beneath the bridge is normally less than 1.2 m deep.
This item is not relevant in the context of all other items: If all items except the last one conform for no
barrier situation, then last item does not make sense. The last item needs to be in a separate paragraph.
Austroads 2013
— 207 —
Comments Responses
From: SCANLON Robert (Jock) (SES) [mailto:robertjock.scanlon@mainroads.wa.gov.au]
Sent: Sunday, 16 October 2011 2:13 PM
I haven’t been able to read this closely prior to going on leave, so in terms of the ‘highlighted’ areas for SRA
input on numbers that will have to come later. However without wishing to undervalue the work done, I did
have some general thoughts from my first very quick look.
In the choice of any standard there is an assessment of risk made at some point in time. Mostly this has
been performed as a once-off and then a subjective appraisal is made to determine and set an acceptable
standard.
One of my question marks over the current code is its requirement and in my view over prescriptive reliance Noted: Generic risk assessment undertaken by experts to provide a limited number of standards within the
on risk assessments, BCA, traffic numbers, grades and etc. etc. for every site rather than having the generic code.
risk assessment undertaken by experts who then provide a limited number of standards within the code for
the user to choose from.
It is a difficult area but I would have liked to see this guideline tackle various typical case scenarios and Noted: Typical case scenarios should be developed.
nominate the necessary or appropriate standard of barrier which could be applied across a fairly broad By introducing the minimum specifications, we are attempting to reduce the reliance on BCA. This analysis
spectrum. This would have meant less reliance on AADT numbers etc., and BCAs, rather than more. A low will be optional and carried out on a case-by-case basis where required.
or certainly a regular barrier will contain light vehicles irrespective of AADTs, so anything higher is really
about containing heavy vehicles. A BCA is useful tool in considering the prioritising for retro fitting of existing
systems but I don’t see it should be necessary for new works.
From: John Reynolds [mailto:John.Reynolds@nzta.govt.nz]
Sent: Monday, 17 October 2011 9:15 AM
I am a bridge designer with an interest in barrier design. I am not a safety expert. I have provided the
document to two of NZTA's safety experts (James Hughes and Julian Chisnall) for additional comments.
We think there are three principal issues to address:
1. The compatibility of the Bridge Barrier Design principles with current and future highway and approach
barrier design principles.
2. The principles of barrier selection.
3. The specific criteria used for barrier selection.
Austroads 2013
— 208 —
Comments Responses
Compatibility of bridge and highway design principles
In NZ we now design highway barriers using ‘Safe System’ principles. It is our understanding that Austroads We would like to have references and examples of implementing the ‘Safe System’ approach in highway
has initiated a project under the Highway Safety Technology Review Panel (?) to investigate the barrier design.
implementation of the ‘Safe System’ principles for barrier design.
It is our view that the Bridge Barrier project should pre-empt the changes to ensure that the ‘Safe System’ We are looking into it, but it is not currently in the scope.
approach is incorporated.
We presume that the outcome from Project TT1601 will be reviewed by the Highway Safety Technology This is not currently in the scope.
Review Panel.
Principles of barrier selection

The key principles seem to be;
 different safety risk exposures demand different barrier performance levels
 the barrier selection method should require as little time input as necessary, consistent with the level of
risk
 the current principles of AS 5100 have largely been followed
 we agree with this principle.
A few questions/comments
1. Have you confirmed that the principles are consistent with international best practice? If Austroads is to An international literature review is out of scope of this project; however, changes of MASH over NCHRP
adopt this methodology, then it would be prudent to confirm that it aligns with international best practice. 350 and use of AASHTO design method have been considered.
2. We consider that this needs to align with ‘Safe System’ principles. Agree, it is out of scope, but we are looking into it.
3. The use of bridge performance level terminology ‘Low, Regular, Medium, High and Special’ in NZ has Current work considers separating the terminologies of low-regular-medium-special performance levels and
been replaced with ‘TL-3, TL-4, TL-5, TL-6, and Special’. I note the example uses TL-5 and TL-6. Part associates them with the performance specifications rather than with the actual barrier test level TLs.
of the reason for NZ adopting the TL system, is apparently the connotation associated with ‘Low
Performance’. It would be useful at least to have a table showing the equivalence of TL terminology to
the stated terminology.
Austroads 2013
— 209 —
Comments Responses
4. Is it mandatory that the B/C Analysis be undertaken for CAT3 sites? My experience with bridge B/C analysis should be optional. Minimum specification level can be used instead.
designers is that they have a passion for designing steel and concrete beams etc., but that they
consider barriers are an ‘off the shelf’ item with little discretion for departure from standard solutions. I
am sure designers would prefer not to have to look at B/C's – its not their area of interest or expertise. It
will make the design process more convoluted and we are not convinced of the validity of the outcome.
So we would prefer to see a methodology without a B/C component.
An alternative approach could be to set lower thresholds for high and special barriers using the specification We agree with this approach.
level method, and then allow the performance level to be lowered using the B/C method. This would then
provide a potential incentive to do the B/C method, rather than the current disincentive. Perhaps the criteria
for the specification method could have 2 limits - a ‘minimum’ and ‘preferable’ limit, with adoption of We like this concept. We need to get agreement on the ‘minimum’ criteria before moving to ‘preferable’
‘minimum’ limits subject to an incremental B/C analysis. That will at least incentivise B/C analysis and limits.
potentially penalise those who don't do the B/C analysis.
The B/C approach will need to be well defined, simple and quick to undertake.
Specific criteria used for barrier selection
We have not addressed this in any detail at this stage. We would rather do that once some of the bigger
issues have been reviewed and agreed.
1. CAT1 Criteria
No comment at this stage
2. CAT3 Criteria Agree. Need to find out specific situations where a Special PL is needed.
 Criteria (a) and (b) will need to be expanded to include more specific criteria. We think this needs to be
as specific as possible to provide effective guidance, but generic enough to allow Road Controlling
authority discretion as appropriate. Perhaps it could include ‘minimum’ and ‘preferable’ criteria.
 B/C Analysis method adopts an ‘incremental B/C’ approach. Why not compare ‘Medium’, ‘High’ and Current document does reflect this, but much opposition was put up from getting to a ‘Medium’ answer via
‘Special’ options? If the incremental B/C is less than 1 (say), then the lower performance level can be the two different methodologies.
adopted. Should there even be a B/C approach? More work needed once the approach agreed.
 If there is to be a B/C Analysis, it will have to be very simple with standard charts etc. Perhaps even Standard cost and benefit charts require extensive inputs from jurisdictions. It is unlikely to be achievable in
standard cost and benefit charts. this project scope.
Austroads 2013
— 210 —
Comments Responses
From: CANCERI Joseph P [mailto:Joseph_CANCERI@rta.nsw.gov.au]
Sent: Wednesday, 19 October 2011 4:14 PM
As you can see from my comments, I believe that traffic barriers along the edges of high reinforced soil This is a good point; however, it is out of scope of the project. Maybe additional work based on the outcome
walls (RSW) should be included in this project. of this project.
During the last meeting we had in Melbourne, I put forward the case that the vertical drop from a high RSW
was no different to a vertical drop from a bridge and as such, barriers along the edges of high RSW should
be treated and designed similarly to barriers on bridges. This is particularly true for bridges which have been
designed with RSW abutments.
Also, in NSW there are no design guidelines that I am aware for the design of barriers on high RSW and
this project provides an opportunity to provide such guidelines.
From: Vince.Colosimo@roads.vic.gov.au [mailto:Vince.Colosimo@roads.vic.gov.au]
Sent: Wednesday, 21 December 2011 2:58 PM
Summary of information provided previously and from yesterday's webex meeting are in part extracted from
the following research papers etc.:
1 Bridge Barriers – Towards National Standards Austroads NZ Conference 2009
2 Guide for Bridge Approach Barriers – Small Bridges Conference 2011
3 Bridge Barriers – Implementing the AS 5100 Bridge Code provisions. Austroads 2004. We have also considered the Paper in 2nd ARRB conference 2006.
The categorisation is no more than a superimposed category number on the following performance levels:
1) No barrier Performance level = Category 1 (Subjective criteria) The categorisation is used for ease of implementation of the code provisions. There is no new methodology
2) Performance Level Low, Regular & Medium (TL-2, 4 & 5) = Category 2 (Site risk cost criteria algorithm) invented, it largely based on the current AS 5100 code provisions (and it should be) to make it easier for
3) Special performance levels T44 High & 36 t tanker TL-5 = Category 3 (Empirical and Benefit - cost understanding.
consideration based).
The categories are site risk factor based but do not make the code any easier to read or work with on a There are three methodologies, the structure just repeats this.
practical basis i.e. they simply add an additional layer of terminology required.
Most of the rest is based on the existing description and coverage in the current standard AS 5100 with We recognised and addressed the contribution of these papers. The goals of this work are to develop
additional considerations into the MASH 2009 test loading update as it may affect the standard criteria. comprehensive methodology and standardised systems.
Note that most of the above has already been considered in previous recent research papers.
Austroads 2013
— 211 —
Comments Responses
From: John Reynolds [mailto:John.Reynolds@nzta.govt.nz]
Sent: Thursday, 5 January 2012 1:42 PM
I have provided comments on the spread sheet. Hopefully it is helpful, but I haven't spent much time on it.
Give me a call if you need more info or explanation. Overall I concur with the approach being taken.
I will be able to provide drawings but these will follow later.
Criteria for site condition CAT1

Criterion d) The width between kerbs is not less than 8.4 m for a two-lane bridge or 5.2 m for a single-lane Please see the response to Nishantha Palliyaguru (TMR).
bridge.
Not currently a NZ required. But agree in principle. Does this imply kerbs are required? Noted
Criterion e) The edge of the bridge is at least 1.0 m from the edge of traffic lanes. Please see the response to Nishantha Palliyaguru (TMR).
Agree but probably not relevant given d) above.
Criterion h) The provision of barriers prevents the passage of debris or the barriers would be frequently Possibly it needs to be covered by individual jurisdiction.
damaged by heavy debris or both.
Doesn't seem sensible to require that this criterion MUST be met. All other criteria are safety based. Or is it
intended that only where barriers could be damaged or inhibit flows that a no barrier option be adopted?
Needs discussion.
Criterion i) NZ has a separate criterion as well – Or, when the edge of the bridge or culvert is greater than Agree
9 m from the edge of the carriageway, or when the cross-sectional area of a culvert is less than 3.5 m2
and with ends flush with the embankment batter.
Criteria for site condition CAT3

Criterion a) Site-specific, unusual conditions and critical locations. Need to get jurisdictions’ agreement on specific volumes. Can you please provide your estimate?
Agree in principle. But needs to be more specific. Perhaps refer to the clauses below. Or refer to Road
Controlling Authority discretion.
Criterion (c) Bridges on roads with special classes of heavy vehicle such as high mass (HML), high
centre-of-gravity vehicles, B-double, double- and triple‑road trains (vehicle class 9 or higher).
Agree. But need to be more specific e.g. volumes?
Criterion (e) Roads with 15 or more buses per day per carriageway, and any of the situations (d)-1 to (d)-3, (d)-1 to (d)-3 represent unfavourable site conditions where the risk of fatality is high when a crash occurs.
plus any of the following site conditions apply: Together with the comments by Ross Pritchard (page 10), it is likely that for routes with buses, the barrier
 bridges more than 10 m high performance level can be ‘Medium’ as the highest regardless of the bus volume or site conditions. ‘Regular’
 bridge over water more than 3 m deep (normal flow) performance level barrier may be considered where the volume of buses is less than a limit, say 15 buses a
 bridges on horizontal curves with a radius of 600 m or less. day?
I can't understand why (d)-1 to (d)-3 is relevant for buses. 15 buses per day seems very low. In NZ we don't In that case, conditions (d)-1 to (d)-3 can be removed.
differentiate buses, and we include items (e)-1 to (e)-3 under (d) as ‘Or any of the following conditions apply’
Austroads 2013
— 212 —
Comments Responses
Minimum specs for Medium performance level

(a) Class 10 vehicles or higher are not present on route. These are minimum criteria for when Medium PL is needed. The site must be in CAT3 site condition before
This doesn't seem to set sensible minimum criteria for Medium performance level. The majority of sites looking at these criteria.
would meet these criteria. Agree that it’s a good idea to provide criteria but need to rethink the criteria. As To make it clear, one criterion is added:
stated, the criteria are for when Medium PL is not needed. Shouldn't the criteria be for when Medium PL is ‘(a) the site belongs to a CAT3 site condition.’
needed?
Minimum specs for High performance level
Need to revise and include the specific volumes of vehicle types. Can you provide your estimate?
(b) Class 10 vehicles or higher are not present on route.
Seems superfluous
Minimum specs for Special performance level
Need to find out scenarios for site-specific, unusual conditions and critical locations from all jurisdictions.
(a) Site-specific, unusual conditions and critical locations
Not specific. Perhaps allow Road Controlling Authority discretion.
(c) Bridges on roads with special classes of heavy vehicle such as high mass (HML), high centre-of-gravity Need to discuss further with all jurisdictions on what types of vehicle automatically require a Special
vehicles, B-double, double- and triple‑road trains (vehicle class 10 or higher). performance level.
Too generic. Need volumes specified.
Other information
Have you considered the link between the TL levels and performance level? We need information on the Noted
basis of this.
The NZTA Bridge Manual provides a Table of Performance Level and Test Level (NCHRP Report 350 Test
Level) equivalence
Are there any different loads used compared to AS 5100 table?

We have never carried out crash testing. For design we use table 4.2 loads for rigid barriers. We use other Noted
loads for semi-rigid barrier design, but I would like another table to be included for this case.
Are there any specific performance criteria or site characteristics used in the selection of a barrier over and Noted
above AS 5100? None that I am aware of.
Are there any special design methodologies used?

None that I am aware of Noted
Austroads 2013
— 213 —
Comments Responses
From: ross.w.pritchard@tmr.qld.gov.au [mailto:ross.w.pritchard@tmr.qld.gov.au]
Sent: Monday, 19 December 2011 6:13 PM
CAT1 - TMR agrees with suggestion
Special barriers – rail.

TMR has a special barrier agreement with Queensland Rail. There are barriers in excess of AS 5100. Noted: TL-5 barriers with 1.5 m height; TL-6 barriers with 1.5 and 2.0 m height and 1000 kN transverse
design load. 2.0 m height is used for fire protection.
CAT3: We have a special level barrier where our road passes over the Queensland Rail main control room. Noted. We need more such scenarios from other jurisdictions.
It was decided that we must never penetrate that building.
Re buses. On our busways, all buses are wheel chair friendly and have a low level floor. Our standard Noted: no special barrier is required regardless of bus volume.
barrier is single slope 1100 mm high. Hence, I do not believe a special barrier is necessary.
However, for long distance coaches and a high centre-of-gravity, I would agree a special class barrier is Noted: special barrier may be considered for high centre-of-gravity vehicles.
necessary where there are a high number of buses. Similarly, for some rural local buses with high
centre-of-gravity, special barriers may be necessary.
From: Pyrzakowski, Adam (DPTI) [mailto:Adam.Pyrzakowski@sa.gov.au]
Sent: Friday, 23 December 2011 9:05 AM
I had a very quick look at your Stage 1 Document and noted that the South Australian DTEI (or as it called Noted
now DPTI) barriers are not listed in your Table 4.4. The barriers are:
 Low level:
W-beam and thrie-beam backed with two RHS sections, 710 mm high.
 Regular:
double thrie beam or thicker ’bridge’ thrie beam and UC posts, 820 mm high
two RHS rail and UC posts, 820 mm high
two RHS rail + hand rail and UC posts, nom 1000 mm height.
 Median
combination barrier, concrete plinth and one or two rails, minimum 1300 mm high
three RHS rails and UC posts 1300 mm high.
I noted in your Section 4.4.3 (based on AASHTO LRFD) a typo – the equation for tension force should Agreed, it has been updated.
be T=Rw/(Lc+2H).
It is not clear either in AASHTO LRFD or in your Document what width of deck should be used for cantilever Will clarify in the document.
bending design in Design option 1 – in the past I have used the same distribution as for tension force T but 45° distribution should be used, however, conservatively the width of deck can be taken as the same
it should be clarified. Further, it is also not clear if the above distribution applies only at the cantilever just distribution as for tension force. The same is for the width of deck at the root of the cantilever.
behind the barrier and what width should be used at the root of the cantilever.
Austroads 2013
— 214 —
Comments Responses
I have also noted that you propose a load factor of 1.1 for design option 1. What is the justification for 1.1 is used for design of the cantilever deck slab. This factor is used to ensure that the failure occurs in the
this considering that in AASHTO (and AS 5100) load factor for Ft (<= Rw) is 1.0 and furthermore in AASHTO barrier, not in the deck slab.
material reduction factors are 1.0 (Extreme event). When Rw is >> Ft the design is already over-
conservative.
In cl 4.8.2 a design example of post and 3 rail barrier is shown. However, in your calculation only the two AASHTO LFRD CL A13.2, commentary:
top rails out of three are considered. A justification is required. If the total resistance, R*, of a post-and-beam railing system with multiple rail elements is significantly
In the past I have noted that for example TL-5 barrier design with, say, the top rail at 1.07 m which meets greater than the applied load, Ft, then the resistance, R*, for the lower rail element(s) used in calculations
AASHTO minimum rail requirement (refer to AASHTO tab A 13.2–1) cannot meet the Y*>=He (min) may be reduced.
condition as there is always force distribution in more than one rail. A discussion regarding this is The reduced value of R will result in an increase in the computed value of Y*. The reduced notional total rail
required as in example in cl 4.8.3 resistance and its effective height must satisfy R* > Ft and Y* > He.
If the design load located at a height H falls between rail elements, it should be distributed proportionally to
In cl 4.8.3 I find your method of reducing Rw by 84% to 0.26 Rw to obtain required Y* not well explained. A rail elements above and below such that Y* > He.
better rationale should be provided for the benefit of designer. Please note that this barrier o/a height is in This will be clarified in the report.
excess of the min height of 1070 mm specified by AASHTO for TL-5 which could offer a justification.
Notwithstanding the above comments I can see great benefit of having a good barrier design manual based Noted.
on AASHTO which would, in my opinion, contribute much to barrier standardisation. The manual
should provide information beyond what is already readily available in AASHTO LRFD. I would suggest that
going to the source and contacting the AASHTO LRDF authors directly as well as practising engineers in
US could be the way forward.
In order to conduct a risk analysis we would need credible crash data at bridge sites. This was in scope
In your spread sheet in CAT3 criteria for Special barrier selection I have chosen 3000 CV/cway/day (15% of
earlier in the project however based on information that was available it was agreed to drop this from the
20000 AADT) and 150 buses/cway/day which are above recorded traffic counts on one of our main
scope of the project.
highways. It is difficult to guess if the risk-based on this numbers is acceptable and I would prefer if a
meaningful risk analysis is conducted by ARRB.
Other information
Have you considered the link between the TL levels and performance level? Noted.
AASHTO TL and AS 5100 Performance Levels as in AS 5100 are used interchangeably.
Are there any different loads used compared to AS 5100 table? No Noted.
Are there any specific performance criteria or site characteristics used in the selection of a barrier over and
above AS 5100?
DPTI has recently adopted 1500 mm high, combination barrier over the high frequency passenger railway In the USA, barriers with height > 2 m have been used for vehicles with high centre-of-gravity.
lines design to 750 kN, TL-6 loading. However in AASHTO TL-6 level has minimum height He= 2290 mm – This will be addressed in the performance specifications.
rather impractical in my opinion. The recommended height for Special barrier should be addressed in the
Document.
Are there any special design methodologies used?
DPTI follows AS 5100, however in case of our lower level TL-4 barrier (thrie-beam+ UCpost ) we use an Noted.
arbitrary maximum limit of adjusted AADT.
Austroads 2013
— 215 —
Comments Responses
From: Neal Lake
Barrier performance specifications Will be attempting to link performance level to barrier selection through performance criteria.
Need a discussion linking:
 Barrier performance level and actual vehicles likely to use a road at that performance level Should refer back to AASHTO LRFD Section 13, CL 13.7.2 Test level selection criteria:
 How the above point relates to the TL level, i.e. determine if TL-2 is appropriate for Low level barrier TL-2 – taken to be generally acceptable for work zones and most local and
collector roads with favourable site conditions as well as where a small number of heavy vehicles is
performance level by determining if the vehicles and speeds listed above are suitable and relevant.
expected and posted speeds are reduced;
E.g. is it reasonable to expect that any road is likely to have a 44 t articulated van on it. So if the road is
100 km/h then only TL-6 would be deemed suitable? TL-3 – taken to be generally acceptable for a wide range of high-speed arterial
It is also very important to link the above with the actual loads used in the performance specification. Can highways with very low mixtures of heavy vehicles and with favourable site conditions;
these be prorated?
TL-4 – taken to be generally acceptable for the majority of applications on
Need to discuss with the committee to determine if any work has been done in these areas.
high speed highways, freeways, expressways, and Interstate highways with a mixture of trucks and heavy
vehicles;
TL-5 – taken to be generally acceptable for the same applications as TL-4 and
where large trucks make up a significant portion of the average daily traffic or when unfavourable site
conditions justify a higher level of rail resistance.
(TL-5 provides for a van-type tractor-trailer that will satisfy design requirements where TL-4 railings are
deemed to be inadequate due to the high number of this type of vehicle anticipated, or due to unfavourable
site. Conditions where rollover or penetration beyond the railing could result in severe consequences);
TL-6 – taken to be generally acceptable for applications where tanker-type trucks or similar high centre-of-
gravity vehicles are anticipated, particularly along with unfavourable site conditions.
(TL-6 provides for a tanker-type truck that will satisfy design requirements where this type vehicle with a
higher centre-of-gravity has shown a history of rollover or penetration, or unfavourable site conditions may
indicate the need for this level of rail resistance);
Unfavourable site conditions include but are not limited to reduced radius of curvature, steep downgrades
on curvature, variable cross slopes, and adverse weather conditions.
AASHTO LRFD, Low speed: < 70 km/h; high speed: > 70 km/h.
Table 4.1: Traffic barrier design loads and contact lengths

How do the values of forces, contact lengths, minimum heights in this table link to TL barriers. Need to Could not find source document/formulae from AASHTO/NCHRP 350 for Ft and FL. Other values are
discuss if the magnitude is appropriate for the types of vehicles that will use the given performance level explained in AASHTO LRFD 2010.
route. How to derive values for B-doubles, Road trains?
Austroads 2013
— 216 —
Comments Responses
From: CANCERI Joseph P [mailto:Joseph.CANCERI@rms.nsw.gov.au]
Sent: Wednesday, 15 February 2012 4:05 PM See the below table.
ID Section/para no. Comment Responses
(don’t use page nums) (may include proposed new text and reason for change)
1 Section 1.3 Scope The scope should also include traffic barriers along the edges of reinforced soil wall (RSW) abutments and along the edges of high It is an important point, but out of scope of this
RSW. The drop from a near-vertical RSW is essentially the same as the drop from the edge of a bridge. Quite often off-loading and project. Suggest additional work based on the
on-loading vehicular ramps are located adjacent to and at the bottom of RSW. outcome of this project.
2 Section 3 It is stated that the risk assessment procedure described in the charts contained in AS 5100.1–2004 are for ‘regular site conditions’. Noted.
1st para. I am not sure if this statement is correct. It can be argued that the adjustment factor ‘US’ derived from Figure B4, distinguishes
different site conditions or under structure conditions.
3 Section 3.1 to On first reading the description of the ‘Site Conditions CAT1, CAT2 and CAT3’, it seems to unnecessarily introduce the same or Noted. This is exactly what we are attempting
Section 3.4 very similar information that already exists in Cl 10.5 in AS 5100.1–2004 (Performance Levels) but in a different format. However, to achieve.
and on reflection, classifying site conditions first and then determining the appropriate performance levels as described in Sections
3.1 to 3.4 and the Flowchart in Figure 3.1 of the Draft, is a more comprehensive and systematic approach.
Section 3.4(c) Do not agree that bridges carrying high mass limit vehicles (HML), B-doubles, Double & Triple- road trains (vehicles Class 9 or
higher) necessarily constitutes a CAT3 Site Condition. A CAT3 Site Condition could be justified if a bridge carries a large volume of
these heavier vehicles. However, if an agreed minimum number of these heavier vehicles per day per carriageway are specified
together with other conditions such as those specified for buses in Condition (e) of Section 3.4 then a CAT3 Site Condition can be
justified.
Section 3.4 The three classifications for Site Conditions (CAT1, CAT2 & CAT3), as well as the draft document in general, seem to only focus on
bridges with edge traffic barriers. Some bridges on highly trafficked major roadways which carry two-way traffic and heavy vehicles
travelling at very high speeds have median barriers with narrow shoulders. Should an errant heavy vehicle crash through or vault
over the median barrier on to on-coming traffic the results would be disastrous. The provision for median traffic barriers also needs
to be discussed and incorporated in the document.
4 Section 3.6.1 – Medium It is noted that medium performance level barriers may be selected ‘when the bridge site is on a road with medium to high volumes
performance of buses’. The reference to medium to high volume of buses needs to be qualified by a number of buses per day per carriageway.
Also, it is noted that medium performance level barriers may be selected if the conditions specified apply whilst in the paragraphs
that follow, high & special performance level barriers shall be selected if the respective conditions that have been specified apply.
Since the term ‘may’ is non-mandatory, should it be changed to ‘shall’ for the selection of the medium performance level barriers?
5 Section 3.6.1 – High ‘A high performance level barrier shall be provided at specific locations where ‘any or all of’ the following conditions apply:’ Need to get jurisdictions’ agreement on a
performance Need to clarify if any or all of the conditions apply. minimum number of these vehicles which
1st sentence requires special PL barriers.
6 Section 3.6.1 – Special Do not agree that bridges carrying high mass limit vehicles (HML), high centre-of-gravity vehicles, B-doubles, Double & Triple- road Agreed.
performance trains (vehicles Class 10 or higher) necessarily have to have a Special Performance Level barriers unless a minimum number of
these heavier vehicles per day per carriageway is specified and agree to. Refer also to the comment made above on Section 3.4.
Also, we need to define better what are high centre-of-gravity vehicles e.g. buses etc.
Austroads 2013
— 217 —
ID Section/para no. Comment Responses

(don’t use page nums) (may include proposed new text and reason for change)
7 Section 4.6 This section only depicts a high concrete parapet with a single metal rail. In NSW shorter height concrete parapets with two metal Median barriers should be taken into
Concrete Parapet and rails are the most commonly used traffic barriers. consideration, however more in B/C analysis.
Metal Rail There are three reasons why traffic barriers with two metal rails and a short height concrete parapet have been adopted as a Selection procedure and design process are
standard barrier type in NSW: the same with edge traffic barriers.
 Firstly, many bridges carry cyclists and the height of traffic barriers on such bridges in order to contain cyclists must be at least
1300 mm in accordance with Clause 12.1(d) in AS 5100.1–2004. If a short height concrete parapet with two metal rails is
provided, it will reduce the overall weight of the traffic barrier and consequently the dead load on the edge bridge girders -
which are often the most heavily loaded girders.
 Secondly, it is often an urban design requirement to have short height concrete parapets in order to maximise commuter’s view
from the bridge deck outwards. This is particularly true for long bridge.
 Thirdly, it is also often an urban design requirement to have short height concrete parapets to increase the slenderness and
improve the aesthetics of a bridge when viewed in elevation.
For the above reasons, traffic barriers with two metal rails and a short height concrete parapet are likely to remain a standard traffic
barrier type in NSW and it is suggested that such a barrier type be included in the future Austroads’ ‘Standardise Bridge Barrier
Design’ the draft of which is the subject of this review.
Austroads 2013
— 218 —
APPENDIX E BTTF CONSULTATION

After the first draft of the project report was sent to the working group for comment, a BTTF
meeting was organised in Austroads National Office in December 2012 to discuss the outcomes of
the project and obtain an agreement to a set of nationally consistent barriers to take into the future.
The BTF discussed a number of points for each proposed guideline. The discussions are
summarised as follows.
E.1 Section 3 – Selection of Bridge Barrier Performance Levels

E.1.1 Changes of Barrier Design Loads and Heights as a Result of MASH
As part of the revision of AS 5100.2 (2004), changes to the impact forces and the required
minimum height for bridge barriers have been proposed by Vince Colosimo (VicRoads) as a result
of the introduction of MASH (AASHTO 2009) in the USA. Although the controlling strength test
vehicles do not change for all performance levels except for Low and Regular performance levels,
the design forces and the required minimum height for the Medium performance level barrier was
proposed to be increased.
In addition, there is no change in the design loads for bridge barriers in the latest version of
AASHTO LRFD (2012 version) as a result of the introduction of MASH.
The following is noted from a review by Joseph Canceri (RMS) of Vince Colisimo’s design
calculations for his proposed increases in both impact forces and minimum height for bridge
barriers:
 It is noted in Vince’s calculations that the proposed changes to both impact forces and
impact heights for the Low and Regular performance level barriers have been deduced by
comparing the test results in NCHRP 350 with the MASH test levels. However, it is noted that
for Test Level 4 (Regular performance level barriers), the increase in the mass of the test
vehicle (8 t to 10 t) has been accounted for in the design calculations but the increase in the
speed of the test vehicle (80 km/h to 90 km/h) has not.
 Also, it is noted in Vince’s calculations that the proposed increased impact forces and height
of application of these forces for the Medium and High performance level barriers have not
been deduced by comparing the test results in NCHRP 350 with the MASH test results since
the test vehicle for these higher performance level barriers has not changed. Vince’s
proposed increases for these traffic barriers have been deduced by comparing the test
results in NCHRP 350 with either the European (CEN1317) test results or previous test
results contained in various AASHTO publications.
 Even though Vince’s proposed impact forces and impact heights have not been consistently
deduced from the MASH test results, the proposed forces and heights increase consistently
with the increase in the performance level of the barriers. The actual calculations associated
with the respective increased impact forces and heights seem to be correct.
 Since the masses of heavy vehicles in Australia are larger than those in the USA, I
understand Vince’s rationale in increasing traffic barrier impact forces and heights. It is really
up to the committee to decide if Vince’s proposed increased impact forces and height of
application of impact forces for traffic barriers should be adopted for the subject Austroads
project or not.
 Resolution: this issue is put on hold. This project proceeds with current design loads and
minimum height for bridge barriers as specified by AS 5100.2 (2004).
Austroads 2013
— 219 —
E.1.2 The use of Site Categories and Number of Barrier Performance Levels
 RMS agreed in principle with five level of barrier s but RMS only used Regular and Medium
barrier. Refer to RMS comments for site categories by Joe Canceri in Appendix C.
 VicRoads agrees with the proposed levels but acknowledges that the test vehicle definition is
incorrect and will need to be updated based on the future of AS 5100. VicRoads agrees with
the site categories.
 TMR has not seen any evidence of how the ‘design load’ relate to NCHRP350 test level.
Based on submission of crash test of road barrier submitted as a bridge barrier, TMR has
specifically banned using crash tests instead of design method. B/C is not supported.
 DIER agrees with site categories. Should there be a speed limit to CAT1, say less than
60 km/h? It would appear precarious for a heavy vehicle travelling at 100 km/h to traverse
across the bridge without any barrier.
 NZTA: Table 3.1 should be updated to reference MASH test levels as well as NCHRP 350.
The traffic volume criteria for ‘high risk category’ are identical to the NZ Bridge Manual
Appendix B (Transit New Zealand 2004); the number of trains, relating to railway crossings,
is not a condition in Transit New Zealand (2004). Use of a B/C analysis is not a Safe System
way to determine the test level.
 It was agreed by all participants that five performance levels be used for bridge barriers.
 Resolution: agree on five levels.
E.1.3 Criteria for Site Conditions and Specification Standards for CAT3 Method
 VicRoads does not agree to the criteria in the current form. VicRoads will agree with the
criteria if the following changes are made:
— Revise point (c) by removing reference to B-double and double- and triple-road trains.
— Revise point (e) by removing 30% AADT (will very rarely happen) and (3) as this is not
considered suitable on its own and requires other factors to be influential.
 TMR owns busways which satisfy this clause (Section 3.4). If buses are suburban commuter
buses with low centre-of-gravity, Medium performance level is not warranted. Instead
Regular performance level is recommended. Agree with Medium performance level barriers
for interstate buses with high centre-of-gravity and have one such situation.
 DPTI: Current AS 5100.1 (2004) selection process works well. Perhaps some expansion is
needed on how depth of water relates to high, medium and low risk under-bridge condition in
Fig B4.
 All other jurisdictions agreed with the proposal.
 Resolution: Agree with proposal. Jurisdictions to consider and agree on site category criteria,
i.e. number or other considerations.
Austroads 2013
— 220 —
E.2 Section 4 – Bridge Barrier Design Procedure

E.2.1 Capacity Reduction Factor, φ
 In AASHTO LRFD, impacts of vehicles colliding with barriers are considered in an extreme
event limit state; therefore the load factor and the capacity reduction factor are both equal to
1.0.
 In AS 5100 (2007), ultimate limit states cover the extreme limit state. The capacity reduction
factor shall be used, e.g. φ = 0.8 for concrete and φ = 0.9 for steel (in bending). The load
factor for designing bridge barriers is 1.0.
 Current practice in Australia uses AASHTO method in combination with AS 5100 (2007)
φ-factors in designing bridge barriers. As a result, the barrier design in Australia is
overdesigned (10% to 20% stronger than the US design). This also increases the capacity
requirements of the bridge deck slab.
E.2.2 Jurisdictions’ Comments and Resolution

 RMS: AS 5100 Part 2 is under revision. RMS is agreed in principle in line with revised
AS 5100.
 VicRoads does not agree with the proposed change as insufficient evidence is provided to
prove the need. VicRoads proposes to stay with current factors for consistency with the rest
of AS 5100.
 TMR: Adopt AASHTO.
 DPTI: For reasons of consistency, AS 5100 reduction factors would be preferred to be used.
Note that in AASHTO extreme event reduction factor 1.0 does not apply to bolts (and is
0.80). Discussion needs also to include design of cantilever deck supporting the barrier.
AS 5100.2, cl.11.3 recommends load factor 1.1 for barrier load and does not include traffic
load in the design. In AASHTO, 1.0 barrier load factor is used but 0.5 of traffic load acts
simultaneously with the barrier load.
 DIER: Propose to be consistent with AS 5100.
 NZTA: The current AS 5100 barrier design approach (Load factor = 1.0 and φ = 0.8 or 0.9)
seems appropriate and is consistent with NZ practice. It is also consistent with the NZ
approach to seismic design (being an ultimate limit state load). Use of φ = 1.0 clearly
changes the probability of demand exceeding capacity. To really answer this question, we
need to understand the derivation of the load. Usually φ = 1.0 is associated with ‘max
credible load effect’ rather than ‘ultimate design load’.
 Resolution: Agree that if design is based on displacement method then φ = 1.0.
E.3 Section 5 – Standardised Bridge Barrier Designs

E.3.1 Directions in Selecting Existing Bridge Barrier Designs for Standardisation
 Details of barrier designs will be completed when detailed designs of selected standardised
barriers are agreed.
 There are three types of bridge barriers: concrete, rail-and-post, and combination (concrete
parapet and rail-and-post barrier).
 It is proposed that a maximum of two standard designs (primary and secondary) be
developed for each barrier type for each performance level, if applicable.
Austroads 2013
— 221 —
 Concrete barriers: single slope and F-shape are preferred. This type of barrier should be
developed for Regular, Medium, High, and Special performance levels.
 Rail-and-post barriers: should be developed only for Regular, Medium and High performance
levels. Different barrier configurations may include: one rail, two rails and three rails with
steel posts.
 Combination barriers: should be developed only for Regular, Medium and High performance
levels. Different barrier configurations may include: concrete parapet of different heights
combined with one rail or two rail barriers with steel posts.
 For Special performance level, only concrete barrier should be developed.
 For concrete parapet, two typical shapes are preferred, F-shape and single slope. Different
performance level barriers may have the same shape; however the barrier height, thickness
and reinforcement details may be different.
 Barrier designs with unique architectural forms should be treated on a project-specific basis.

Low performance level barriers
 Rail-and-post barriers will be used for standardisation. Two standard designs LL1 and LL2
based on VicRoads Low level steel barrier and MRWA thrie-beam barrier, respectively
(Table 5.1) will be developed.
 No concrete or combination barriers will be developed for Low performance level.
Notes:
 RMS and TMR use Regular performance level as a minimum thus they do not use Low
performance level barriers.
 NZTA proposes to use W-beam barrier.
Regular performance level barriers

Two standard designs will be developed for each barrier type including concrete barrier, steel
barrier, and combination barrier as shown in Table 5.2. These designs are based on the following
existing designs:
 RMS Regular level, F-shape concrete barrier RCO, RCI
 RMS Regular level, single slope concrete barrier RD
 VicRoads Regular level, three rail steel barrier
 NZTA Regular level, TL4 thrie-beam with modified blockout barrier
 RMS Regular level, combination barrier RAO, RAI
 RMS Regular level, combination barrier RBO, RBI.
Medium performance level barriers

The barrier designs shown in Table 5.3 are selected for standardisation. These designs are based
on the following existing designs:
 RMS Medium level, F-shape concrete barrier MCO, MCI
 TMR Medium level, single slope, precast concrete barrier
 DPTI Medium level, post and rail barrier
Austroads 2013
— 222 —
 RMS Medium level, combination barrier MAO, MAI

 RMS Medium level, combination barrier MBO, MBI
 SA Medium level, single slope combination barrier.
Notes:
 NZTA uses Texas HT80 barrier.
High performance level barriers

Two standard designs will be developed for each of the concrete barrier and combination barrier
types for the High performance level (Table 5.4). These designs include:
 F-shape concrete barrier, based on Medium performance level barrier ML1 in Table 5.3
 Single slope concrete barrier, based on Medium performance level barrier ML2 in Table 5.3
 VicRoads High level, combination barrier
 SA High level, single slope combination barrier.
Notes:
 Consideration should be given to changing the details of the concrete part to match with
MR4, MR5 (Table 5.3).
Special performance level barriers

The barrier designs shown in Table 5.5 are selected for standardisation. These designs are based
on the following existing designs:
 TMR Special level, concrete single slope barrier, SL1
 SA Special level, single slope, combination barrier.
E.4 Section 6 – Retrofitting of Existing Bridge Barriers

E.4.1 Points for Discussion
 It is unlikely that a set of standard designs for retrofitting existing bridge barriers will be
developed, as the designs are usually developed based on a project-specific basis.
 The barrier retrofitting examples presented in Section 6.5 are not structurally verified to the
associated performance level.
 The methodology to assess the conformity of the structurally upgraded barrier to the current
standard should be developed.
 Bridge barrier retrofitting examples from various jurisdictions are needed. Examples should
show initial barrier, old performance level, new performance level, ideal performance level,
justifications of design process etc.
Austroads 2013
— 223 —

RMS
 Designs usually developed based on project-specific risk assessment.
 Risk-based assessment for bridge specific conditions needs to be developed. However, this
task will not be straightforward.
 Example for retrofitting bridge barrier will vary from bridge to bridge and application or
adoption of such a design will not be straightforward for other bridges.
VicRoads
 VicRoads agrees that a standard set of designs for retrofitting existing bridge barriers is not
possible due to the designs being site-specific and highly variable.
 VicRoads agrees that a methodology should be developed to assist designers with items to
consider.
 VicRoads disagrees with providing examples as there will be little benefit for the
time-consuming task due to the highly variable and site-specific nature of the designs.
TMR
 TMR recommends Austroads use TMR criteria for retrofit.
DPTI
 Details of traffic barrier retrofitting should remain in the hands of individual jurisdictions
although general examples could be given.
Resolution
 Standard designs are not possible for retrofitting. Provide procedure in code.
 Method for assessing the conformity of the structurally upgraded barrier to the current
standard should be developed. Need to discuss – possibly out of scope.
 Do not provide examples.
E.5 Section 7 – Bridge Approach Barriers

 There are two methods for the determination of the length of need, run-out length and angle
of departure. It is recommended that both methods be used and the worst case scenario be
applied.
 Guidelines for designing off-structure foundation (on bridge approach) should be developed,
which include traffic barriers along the edges of high reinforced soil walls.

VicRoads
 VicRoads agrees with the use of the angle of departure method. VicRoads disagrees with the
run-out length method in its current form as it can require excessively long barriers where
there is a high batter and significant skew. VicRoads is currently working through this
methodology to refine.
 VicRoads agrees that future work is required for off-structure barrier foundations.
Austroads 2013
— 224 —
DPTI
 Traffic barrier along the edges of high reinforced soil walls is possibly outside of the scope of
this project. Only a form of general guide rather than detailed recommendations should be
developed.
Resolution
 Use angle of departure method.
 Variation to be agreed with ANO on the guideline for traffic barrier along the edges of high
reinforced soil walls.
E.6 Section 8 – Bridge Support Protection Barriers

 For collisions from road traffic, current AS 5100.2 (2004) requires that the bridge supports be
designed for a minimum equivalent load of 2000 kN applied at an angle of 10° from the
direction of the road centre line and at 1.2 m above the ground level, if they are not located
behind appropriate protective traffic barriers.
 It is recommended that the recent update of AASHTO LRFD Bridge design specifications
(AASHTO 2012) be incorporated, which specifies that the equivalent static load is 2670 kN
acting in a direction of zero to 15° with the edge of the pavement in a horizontal plane, at a
distance of 1.5 m above ground.
 If a rigid barrier is provided, the barrier should be designed to the associated design loads of
the barrier performance level. The barrier should not be designed to the equivalent static
loads specified to act on the bridge support.
 Criteria for the determination of barrier performance level in Section 8.4.3 should be
reviewed and agreed by jurisdictions.
 Queensland Rail technical guideline Collision protection of supporting elements adjacent to
railways (Queensland Rail 2010) is recommended for collision protection of supporting
elements adjacent to railways.

VicRoads
 VicRoads agrees with the load increase but believes this should also apply if a barrier is
protecting as there have been cases where vehicles have mounted the protection barrier and
still collided with the bridge.
 VicRoads agrees with protection barriers being designed to the associated design loads of
the barrier performance level.
 VicRoads disagrees with the approach presented in Section 8.4.3. As a minimum the barrier
should be designed to a load greater than the bridge support collision load.
 VicRoads is not familiar with the technical guideline and it seems that all states have different
rail authority requirements. Adopting such a guideline may not be possible without involving
other rail authorities.
Austroads 2013
— 225 —
TMR
 The design forces are taken as per current draft of AS 5100 Part 2.
 Maximum performance level of protection barrier is Medium. Need to discuss.
DPTI
 It is recommended that wherever possible, such bridge supports be designed for full collision
load even if protected by (appropriate) traffic barriers.
 Barrier function should primarily be to provide a measure of safety for road users not the
structure.
NZTA
 We need to highlight the concern that the use of a safety barrier system in front of the
sub-structure will not guarantee the protection of the sub-structure. For example, a TL4
barrier will only protect against an 8 tonne vehicle at a certain impact angle. The design of
the sub-structure must therefore always allow for a residual impact load whose magnitude is
dependent on the size of a design vehicle.
Resolution
 Adopt AASHTO for the minimum equivalent load – but round up to 2700 kN.
 Principle is that piers will be designed.
 Cannot include Queensland Rail technical guideline due to state and operator differences.
Austroads 2013
— 226 —
INFORMATION RETRIEVAL
Austroads, 2013, Standardised Bridge Barrier Designs, Sydney, A4, pp.234.

AP-R445-13.
Keywords:
bridge barrier, railing, barrier performance level, site condition category,

concrete barrier, steel barrier, combination barrier, post-and-rail barrier,
concrete parapet, risk assessment, benefit-cost analysis, bridge approach
barrier, barrier retrofitting, standard barrier design, Safe System, barrier design
procedure, bridge support protection, protection barrier, barrier selection,
barrier performance specification, barrier type, length of need, rigid barrier,
semi-rigid barrier.
Abstract:
Consistency in the design of traffic barriers for bridges has been identified as a
major issue in current practice throughout Australia and New Zealand. The
need for standard solutions was recognised by road jurisdictions, consultants
and contractors during the Austroads 6th Bridge Conference Workshop on
Bridge Barriers, held in 2006. The main areas of concern are the determination
of appropriate barrier performance levels, structural design criteria, the lack of
standard barrier design details, guidance on retrofitting existing bridge barriers,
bridge approach barriers and overpass bridge support protection.
This report addresses these issues and provides a set of design guidelines for
the identified areas. These guidelines will assist bridge designers, contractors
and jurisdictions to provide consistency and cost savings in the selection and
development of relevant bridge barriers throughout Australia and New Zealand.
Existing jurisdictional technical specifications, standards and guidelines, as well
as outcomes from extensive expert discussions have been incorporated in this
report. Key areas of future work to be undertaken after the on-going revision of
AS 5100 has been published are also included.

Austroads - AP-R445-13 - Standardised Bridge Barrier Design PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Austroads - AP-R445-13 - Standardised Bridge Barrier Design PDF

Uploaded by

Copyright:

Available Formats

AP-R445-13

AUSTROADS RESEARCH REPORT

Standardised Bridge Barrier Design

Published October 2013

© Austroads Ltd 2013

Austroads Project No. TT1601

Austroads Publication No. AP-R445-13

Published by Austroads Ltd

This report includes the following guidelines and key features:

2 DISCUSSIONS ON THE DEVELOPMENT OF THE GUIDELINES ....................................... 7

3 SELECTION OF BRIDGE BARRIER PERFORMANCE LEVELS ....................................... 19

4 BRIDGE BARRIER DESIGN PROCEDURE........................................................................ 49

5 STANDARDISED BRIDGE BARRIER DESIGNS ................................................................ 85

6 RETROFITTING OF EXISTING BRIDGE BARRIERS ....................................................... 109

6.5 Barrier Retrofit Design Procedure ...................................................................................... 115

7 BRIDGE APPROACH BARRIERS .................................................................................... 126

8 BRIDGE SUPPORT PROTECTION BARRIERS ............................................................... 165

8.5 Barrier Layout and Design ................................................................................................. 170

9 PROPOSED FUTURE RESEARCH .................................................................................. 188

REFERENCES ........................................................................................................................... 191

APPENDIX A RECOMMENDATIONS OF 2006 WORKSHOP ................................................. 194

APPENDIX B MEETING OF EXPERT GROUP 2010............................................................... 197

APPENDIX C COMMENTS FROM CONSULTANTS AND CONTRACTORS .......................... 203

APPENDIX D EXPERT CONSULTATION ............................................................................... 207

APPENDIX E BTTF CONSULTATION .................................................................................... 219

Figure 4.3: Barrier design diagram ............................................................................................. 51

Figure 7.17: Interface details of type F to thrie-beam.................................................................. 153

Table 1.1: Project working phases

The overall aims of the project include:

1.2.2 2010–11 Assess Current Practice

2010–11 Output and milestones

1.2.3 2011–12 Develop Guidelines and Identify Research

2011–12 Output and milestones

This report was entitled Standardised Bridge Barrier Designs – Stage 1.

1.2.4 2012–13 Finalise Guidelines and Recommend Research

2012–13 Outputs and milestones

The set of guidelines includes the following sections:

2 DISCUSSIONS ON THE DEVELOPMENT OF THE

2.1 Jurisdictions’ Current Practice

2.1.2 Design Guidelines

Table 2.1: Jurisdictions’ technical design guidelines for bridge barriers

Jurisdiction Technical document

The following are derived from the review:

2.1.3 Issues and Recommendations

Recommendations from the project group meeting at VicRoads on 6 December 2010:

2.2 Current Design Standard AS 5100

2.3 Computer Simulation

2.4 Safe System Approach

There are several guiding principles to this approach:

2.4.2 Contribution of Bridge Barrier Design to the Safe System

Beside vehicle and speed-related countermeasures, road-related countermeasures consistent with

2.5 Development of the Guidelines

The guidelines are developed based on the following sources of information:

2.5.1 Guidelines for Selection of Bridge Barrier Performance Levels (Section 3)

Criteria for no barrier

Comments from the working group showed that:

The following criteria for no barrier are suggested:

Introduction of site categories CAT1, CAT2 and CAT3

The following points have been identified for discussion:

Refer to Appendix E.1 for the outcome of the discussion.

2.5.2 Bridge Barrier Design Procedure (Section 4)

The following points have been identified for discussion:

Regarding the capacity reduction factor, φ: