Process Safety and Environmental Protection 117 (2018) 454–472

Contents lists available at ScienceDirect

Process Safety and Environmental Protection

journal homepage: www.elsevier.com/locate/psep

Development of inherent safety benefits index to analyse the impact

of inherent safety implementation
Mardhati Zainal Abidin a , Risza Rusli a,∗ , Faisal Khan b , Azmi Mohd Shariff a
a
Centre of Advanced Process Safety (CAPS), Universiti Teknologi PETRONAS, 32610 Bandar Seri Iskandar, Perak Darul Ridzuan, Malaysia
b
Faculty of Engineering and Applied Science, Memorial University of Newfoundland, 240 Prince Phillip Drive, S.J. Carew Building, EN4019, St. John’s, NL A1B
3X5, Canada

a r t i c l e i n f o a b s t r a c t

Article history: Over the years notable accidents have occurred in the chemical process industry thus creating a major
Received 20 January 2017 concern for the safeness of its operation. The disasters led to the dire consequences that claimed human
Received in revised form 17 April 2018 life and health as well as monetary losses. Inherently safer design is an effective solution to prevent acci-
Accepted 14 May 2018
dents because it postulates that the best way to reduce risk is to avoid the hazard rather than control the
Available online 17 May 2018
hazard. Although the implementation of inherently safer design can give great advantages and cost opti-
mal operation throughout process’s life cycle, slow adoption of this principle into real design practice can
Keywords:
be observed. This is partly because the inability to realise the benefits of inherent safety implementation
Inherent safety
Index-based methodology
among industrial practitioner. This paper presents a methodology for evaluating inherently safer design
Benefits alternatives and identifying the benefits using an index-based approach at preliminary design stage. The
Technical and organisational measures proposed methodology is applied to ammonia storage with the objective of preventing and minimising
Hazardous process management toxic release. The results have shown that the new methodology can identify the best inherently safer
Regulation design and the benefits where it is capable to reduce the severity of accident and the requirement to
manage hazardous process.
© 2018 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

1. Introduction
Advanced technologies brought by chemical industries have
facilitated the development of products with several benefits for
mankind. Nevertheless, the safety, health and environmental prob-
lems appearing from major accidents, such as the Flixbrough
disaster in 1974 and Bhopal disaster in 1984, have steered major
concern on chemical plant operations. The community, legislation,
and voluntary bodies are now demanding safe operations, and in
order to achieve the highest safety level, new chemical plants are
inflated with thousands of instruments, computing devices, and
safety procedures, the purpose being to reduce the potential of acci-
dent. However, record from the major accident reporting system
(eMARS) database shows that from 2006 to 2016, 225 major acci-
dents had occurred involving thirty types of industries. From this
number, eleven types of industries recorded the highest number
of accidents (Fig. 1), and the highest occurrence (fifty-four cases)
was recorded from chemical manufacturers. Similar trend can be
∗ Corresponding author.
E-mail addresses: mardhatiza@gmail.com (M. Zainal Abidin), risza@utp.edu.my
(R. Rusli), fikhan@mun.ca (F. Khan), azmish@utp.edu.my (A. Mohd Shariff).
found in offshore activities. Data from 2010 to 2012 (Figs. 2 and 3)
show that the accidents that cause fatalities and injuries continued
to occur (Barua et al., 2016). This proves that although the current
plants and operations have been equipped with additional layers
of protection that consist of passive, active, and procedural strate-
gies, the accidents could not be prevented effectively. Therefore, an
inherently safer design (ISD), which emphasises the elimination or
minimisation of hazard source, is deemed the best option.
The ISD principle was introduced by Professor Trevor Kletz in
1978 following the Flixbrough disaster. The principle proposes that
the elimination of intrinsic properties of hazards is the best way to
reduce the risk rather than the addition of external systems to con-
trol the hazards. As outlined by the Center for Chemical Process
Safety (2009), the four main ISD strategies are to minimise, sub-
stitute, moderate, and simplify. The strategies can be implemented
throughout a process life cycle according to suitability as shown
in Table 1. In order to obtain the highest benefits, an ISD must
be implemented at the earliest stage possible because this mea-
sure will not only resolve the safety problem at the lowest cost
but also require fewer efforts (Table 2). The decisions made in
the early design process will be an important factor in determin-
ing the requirement for hazard management while operating the
process, hence leading to the reduction of lifetime cost if ISD was

https://doi.org/10.1016/j.psep.2018.05.013
0957-5820/© 2018 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
 M. Zainal Abidin et al. / Process Safety and Environmental Protection 117 (2018) 454–472 455

the early stages of a project in order to consider the safety aspects.

Nomenclature This necessitates a usable tool or methodology that can provide a
swift solution to assist industrial practitioners in identifying ISD at
2DGR 2-Dimensional graphical rating the earliest design stage possible (Khan and Amyotte, 2003, 2002;
AHI Atmospheric hazard index Srinivasan and Natarajan, 2012). Numerous methodologies have
AHI Accident hazard index been developed in the last twenty years with the aims to evaluate
AHP Analytical hierarchy procedure inherent safety level at an early or preliminary design stage. The
ALARP As low as reasonably practicable following section provides an overview of the methodologies.
CIMAH Control of industrial major accident hazards
COMAH Control of major accident hazard
2. Review on the current methodologies for inherent safety
DI Damage index
evaluation
Dow F&EI Dow fire and explosion index
DOSH Department of occupational safety and health
The past decades have witnessed a remarkable progress in the
eMARS The major accident reporting system
development of methodologies to measure the inherent safety level
EHI Environmental hazard index
for alternative designs at the early or preliminary stage of process
ETA Event tree analysis
design. The following section provides an overview of the previ-
FTA Fault tree analysis
ously developed methodologies for four main categories: inherent
GERA Global environmental risk assessment
safety; inherent health; inherent environment; and inherent safety,
GHS Globally harmonised system
health and environment (SHE). Finally, the comparisons and lim-
HEN Heat exchanger networks
itations of the methodologies will be discussed in relation to the
HIRA Hazard identification and ranking analysis
aims of the study. As a side note, while some of the works reviewed
HMI Hazard management index
adopt decision making tools such as optimisation and analytical
I2SI Integrated inherent safety index
hierarchy procedure (AHP) to resolve multi-criteria SHE problems,
IBI Inherent benign-ness indicator
this section does not intend to provide a detailed overview about
IETH Inherent environmental toxicity hazard
the decision-making tools. The overview in this section will focus
I-GRAND Integrated graphical and numerical descriptive
on the reasons for the index development, their components, and
INSIDE INherent SHE in design
the indicator of ISD implementation.
IOHI Inherent occupational health index
IRA Inherent risk assessment
2.1. Inherent safety methodologies
IRDI Inherent risk of design index
ISAPEDS Inherent safety assessment technique for prelimi-
In general, the methodologies developed for inherent safety
nary design stage
assessment can be divided into three main categories: index-based
ISBI Inherent safety benefits index
methodologies, consequence-based methodologies, and risk-based
ISD Inherently safer design
methodologies. Each methodology will be elaborated in the follow-
ISI Inherent safety index
ing subsections.
KPI Key performance indicator
MAPP Major accident prevention policy 2.1.1. Index-based methodologies
NFPA National fire protection association The pioneer index-based methods such as Prototype Index of
NuDIST Numerical descriptive inherent safety technique Inherent Safety (PIIS) (Lawrence, 1996), Inherent Safety Index (ISI)
OHHI Occupational health hazard index (Heikkilä, 1999), and i-Safe Index (Palaniappan et al., 2004, 2002a,
OSHA Occupational safety and health administration 2002b) were developed to assess the hazards for safer process alter-
PCA Principle components analysis natives and their suitability to be used at the early stage of process
PFD Process flow diagram design in which limited design data are available. The score for
PI Piping and instrumentation the process alternatives is given based on a parameter value that
PIIS Prototype index of inherent safety is divided into several sub-intervals. However, these methodolo-
PPE Personal protective equipment gies suffer from over or low sensitivity issues that are contributed
PRHI Process route healthiness index by the lack of continuity in the evaluation caused by a discrete
PRI Process route index function. To overcome this problem, an index based on fuzzy logic
PSI Process stream index was developed by Gentile et al. (2003). In more recent works, a
R&D Research and development numerical method based on a logistic function called Numerical
RISI Risk-based inherent safety index Descriptive Inherent Safety Technique (NuDIST) was developed to
SAW Simple additive weighting tackle the discrete problem (Ahmad et al., 2014). This technique
SDS Safety data sheet has been extended to further assess the alternative process routes
SHE Safety, health and environment further to the root-cause of the hazards. A 2-Dimensional Graph-
STEP Stream temperature vs entalphy plot ical Rating (2DGR) was provided with the following purposes: (1)
STHI Specific terrestrial hazard index to highlight, through visualisation, the chemical and process prop-
SWHI Specific water hazard index erties that contribute significantly to the maximum hazards, (2)
SWeHI Safety weighted hazard index to group the process routes evaluated with three coloured regions
TIM Three stage ISD matrix (red, yellow, and green), and (3) to indicate the most hazardous,
UNRTDG United nations recommendations on the transport hazardous, and the least hazardous levels respectively (Ahmad
of dangerous goods et al., 2016a). The NuDIST methodology has been used to identify
WAR Waste reduction algorithm ISD for biodiesel production from the perspective of flammability
parameter (Ahmad et al., 2016b). From NuDIST and the Integrated
Graphical and Numerical Descriptive (I-GRAND) that has been pro-
chosen. However, as mentioned by Khan and Amyotte (2002), one posed, the Inherent Safety Assessment Technique for Preliminary
of the hurdles in adopting ISD is the difficulty to secure time at Design Stage (ISAPEDS) has been introduced by the same author
456 M. Zainal Abidin et al. / Process Safety and Environmental Protection 117 (2018) 454–472

Fig. 1. Industries that record the highest number of major accidents.

Table 1
Usual plant design activity and ISD review criteria in the chemical process development and design (Kidam et al., 2016)

Design phase Process design tasks ISD review criteria

Research and development (R&D)
• Reaction concept selection • ISD for reaction systems that are focused on inherent hazards of chemicals and their
• reaction chemistry reactions potential at selected process conditions.
• reaction kinetics • Proper safety investigation and laboratory test of process materials used including
• synthesis routes feedstock, intermediate impurities, by products, and services medium.
• process conditions • Identify the side effects of normal process condition and possible deviations to the
• Thermal analysis material of construction.
• calorimetric test • Detailed study on the potential of unwanted reactions and the ways to control the
• runaway reaction reactive conditions on such as by special additives and inhibitors, solvent to quench
• Examination of materials and or stop the reaction.
chemical hazards • Identify the potential internal/external process contaminants and study their
• fire and explosion impact to the reaction system.
• toxic release • Detailed analysis on the effect of the temperature, pressure, humidity, long storage
• reactivity and stability (self-life) and process deviation from the chemical properties of process materials.
• incompatibility • Detailed study on the heat generation, thermal safety and potential of runaway at
• Process scale-up normal and abnormal process conditions.
• bench and pilot scale test • Proper test and modelling of reactor system scale-up

Preliminary engineering
• Process concept selection • ISD for main process and sub-process that are focus on inherent hazards of
• mode of operations chemicals, their interactions at selected process conditions and types of unit
• process flowsheet operations.
• material balance • Designing a smaller chemical process plant to reduce the inventory of hazardous
• energy balance materials.
• Unit operation selection • Proper test and modelling of process equipment for safe design and scale-up.
• types of equipment • Select appropriate types of equipment that simple and easy to operates for
• process condition complexity and human error reduction.
• sizing • Always use and up-date the design standard timely and check the local legal
• Process modelling and simulation requirement.
• Logistics and material flow analysis • Identify the potential operational problems based on process material
• storage condition characteristics by choosing experienced design team or appoint external reliable
• transfer and handling design specialists

Basic engineering • Process equipment design
• safe operating limits
• protection material construction
• chemical and mechanical spec
• equipment layout and configuration
• feeding system
• Process system engineering
• automation and instrumentation
• protection and mitigation
• Plant-wide layout
• physical arrangement
• equipment siting and material flows
• Utilities set-up
• heating/cooling system
• heat transfer medium selection
• ISD for process and unit operation design that are focused on safer, simpler and
robust design specifications
• Improve the integrity of the process equipment by using chemically resistant and
mechanically robust construction materials.
• Design process equipment based on worst-case scenario and employ suitable
protection system where appropriate.
• Analyse chemical incompatibility and possible operational problems of the utility
system by reviewing their chemical and physical properties and process conditions.
• Choose the correct supply energy for the specific application. − Foresee potential
internal and external process contamination through process connectivity and
equipment/pipe sharing.
• Select robust, reliable, simple, and user-friendly control system.
• Use simpler control philosophy especially in emergency situations.
• Design a simpler chemical process plant to limit the process interaction during
process deviation and upset.
• Provide adequate separation between process units, storage area and other
buildings.
• Design simple and robust piping system by eliminating unwanted components,
dead-end, by-pass and standby pipelines.
 M. Zainal Abidin et al. / Process Safety and Environmental Protection 117 (2018) 454–472
Table 1 (Continued)
Design phase Process design tasks
Detailed engineering • Detailed process description on:
• normal operations
• process start up and shutdown
• automation and control
• preventive maintenance and
services
• emergency preparedness
• Safety manual
• code of practice
• safe system at work
• management system
Table 2
Costs of fixing safety problems (Kletz and Amyotte, 2010).
Stages Cost of fixing safety problem ($)
R&D stage 1
Process flowsheet stage 10
Final decision stage 100
Production stage 1000
Post incident stage 10000
(Ahmad et al., 2017). ISAPEDS introduces four improvements (1)
the parameters are assessed in relations to each other rather than
assessing these parameters as a standalone hazard factor (2) the
chemicals are assessed as a mixture instead of as individual sub-
stance (3) the evaluation will be done comprehensively on each of
the process equipment in the process flow diagram (PFD) (4) the
score of the safety hazards assessment in the whole process will
be weighted based on the chemicals composition in the mixture. A
higher score in ISAPEDS is indicated as more hazardous compared
to a lower score. Other shortcomings of the earliest index-based
methods such as PIIS, ISI, and i-Safe indices are that the interaction
among the process parameters are not considered and are treated
individually, not as a mixture. A few scholars (Leong and Shariff,
2009; Shariff et al., 2012) then developed the indices on the Pro-
cess Route Index (PRI) and Process Stream Index (PSI) based on
process parameters that can be related to an explosion accident.
The purpose was to measure the safety level of the process streams
and process options. These techniques are integrated with a pro-
cess simulator to enable the extraction of mixture data as an input
for index calculation. Salzano and Di Benedetto (2012) also pro-
posed an index to evaluate options based on explosion where the
accident scenario was considered when developing the index, thus
proposing a new index for three types of explosions: (1) uncon-
fined and partially, (2) confined gas and vapour cloud explosion,
confined gas and vapour explosion, and (3) dust explosion. Another
index-based method developed to measure inherent safety level for
process alternatives is the Integrated Inherent Safety Index (I2SI)
by Khan and Amyotte (2005, 2004), which is applicable through-
out a process life cycle. By using this tool, the benefits of ISD are
demonstrated not only by the reduction of severity but also by
the reduction of controller and safety measure requirements. In
the work of Rusli et al. (2013); the Inherent Risk of Design Index
(IRDI) was improved from the I2SI with the objective to consider
the conflict between inherently safer principles and damage index
in selecting the best ISD alternatives. In this work, the best alter-
natives with the lowest IRDI value for both factors will result in
the reduction of the technical safety measure requirement. The
lack of research on ISD for heat exchanger has motivated Chan
et al. (2014) to develop a new method for heat exchanger networks
(HEN) design that incorporates the inherent safety index during the
selection of heat exchanger matches in order to reduce the potential
hazards of the optimal HEN design. The approach uses the extended
457
ISD review criteria
• ISD for detailed process equipment designs that is resistant to process deviations,
emergency and accidents.
• Detailed study and modelling of flow related problems using simulator and
advanced metallurgy data.
• Special technical and operational consideration for layout and mechanical
performance under severe process conditions, and unwanted events (i.e.
internal/external fire or explosion).
• Select simpler, robust and user-friendly ‘buy item’. Review the conductivity of the
material of ‘buy-item’ to limit the static electricity generation and discharge.
• Choose a safer material and working method for cleaning, servicing, maintenance
and repair work
Fig. 2. Fatalities statistics in three years (Barua et al., 2016).
Fig. 3. Injuries statistics in three years (Barua et al., 2016).
graphical Stream Temperature vs Enthalpy Plot (STEP) method that
has been integrated with the ISI.
2.1.2. Consequence-based methodologies
Consequence analysis (Center for Chemical Process Safety,
1999) has been used to measure the inherent safeness of ISD
alternatives based on the severity of accident. The aimed is to
produce more reliable methodologies compared to the index-
based methodology based on mathematical models. A few scholars
(Shariff et al. 2006; Shariff and Wahab, 2013; Shariff and Zaini,
2010) proposed an integrated framework for explosion, fire, and
toxic using consequence analysis with a process simulator. Later,
Zaini et al. (2016) extended the work by proposing ISD for heat
exchanger based on consequence analysis of heat exchanger explo-
sion. In the work of Patel et al. (2010), consequence analysis was
integrated with solvent database to identify inherently safer sol-
vent. Some of the works convert the consequence in terms of
damage radii into the cost of the accident (Bernechea and Arnaldos,
2014; Bernechea and Viger, 2013; Eini et al., 2015; Medina et al.,
2009). Among the costs considered are fatalities and injuries, equip-
ment damaged, structures collapsed or damaged, lost production,
and indirect costs associated with the accident. Additionally, Khan
and Amyotte (2005) included environmental remediation costs to
quantify the impact of major accident on the environment. Other
scholars (Landucci et al., 2008; Tugnoli et al., 2012, 2009, 2007) used
458 M. Zainal Abidin et al. / Process Safety and Environmental Protection 117 (2018) 454–472
the consequence analysis as a basis in developing Key Performance
Indicator (KPI).
2.1.3. Risk-based methodologies
Risk in process safety consists of two components: severity of
accident and failure probability. In recent years, several risk-based
methodologies have been developed to evaluate inherent safety
level at the early design stage. A few works (Shariff and Leong,
2009; Shariff and Zaini, 2013) have introduced two regions of FN
curve and a two-region matrix for risk assessment at the prelimi-
nary stage of process design. Named as Inherent Risk Assessment
(IRA), the method combines the consequence analysis for severity
estimation with Event Tree Analysis (ETA) and Fault Tree Analysis
(FTA) to estimate failure probability. Different approaches can be
used for risk-based analysis. For example, cost of accidents has been
combined with probability for domino effect using the ETA analy-
sis (Bernechea and Arnaldos, 2014; Bernechea and Viger, 2013).
Similarly, Gómez et al. (2012) combined the cost of event that con-
sisted of cost of material lost, cost of repair and structure damage
with the annual frequency of event based on the literature. The
purpose was to identify the best ISD for storage tank by using a risk-
based analysis. In order to resolve the limitation of I2SI (which is
consequence-based analysis), Moradi (2010) introduced the appli-
cation of Petri-net tool for a risk-based assessment, which combines
Hazard Identification and Ranking Analysis (HIRA) methodology for
severity estimation with a stochastic probabilistic estimation for a
risk-based methodology. In the work of Rathnayaka et al. (2014),
a Risk-based Inherent Safety Index (RISI) was proposed with the
HIRA methodology for severity estimation and with the bow-tie
technique for probability estimation as the basis of index develop-
ment. According to Jha et al. (2016), the benchmarking of the worst
stream is difficult to predict by using the original PSI methodology.
Therefore, the authors proposed the benchmarking criteria using
risk-based methodology.
2.2. Inherent health methodologies
One of the pioneer indices developed to assess the inherent
health hazard for alternative process routes is the Occupational
Health Hazard Index (OHHI). Subsequently, Hassim and Edwards
(2006) proposed the Process Route Healthiness Index (PRHI) for
the same purposes. The PRHI takes into account the potential of
activities and process conditions that can cause harm to workers.
The index only uses health hazard properties from the National Fire
Protection Association (NFPA) to assess inherent material health
hazards based on their ability to cause typical occupational dis-
eases. Additionally, the PRHI takes into account all the possible
sources that can cause exposure of chemical towards workers, but
it requires extensive data, some of which are not available during
process pathways/process concept evaluation stage. Later, a new
index called Inherent Occupational Health Index (IOHI) was intro-
duced to assess alternative pathways from health point of view
(Hassim and Hurme, 2010). Two main categories were considered
in the IOHI development: the potential for harm and the potential
for exposure. Interestingly, in this work, three different approaches
were used to combine the sub-index, which are the additive type,
the average type, and the worst-case type. The results showed that
the assessment outcomes depend heavily on the type of combina-
tion approach used. The impact of the complexity can be captured
by the additive approach; the average hazard for process routes can
be identified using the average type; and the worst-case approach
can expose the worst side of a process from the aspect of inherent
healthiness.
2.3. Inherent environment methodologies
Due to the lack of methodology for assessing inherent environ-
mental friendliness for process route, Cave and Edwards (1997)
developed the Environmental Hazard Index (EHI) to assess the
impact of hazardous chemicals on aquatic and terrestrial ecosys-
tems. The EHI consists of two sub-indices: (1) Specific Water Hazard
Index (SWHI) (hazard to the aquatic ecosystem associated with
one tonne of a chemical) and (2) Specific Terrestrial Hazard Index
(STHI) (for the hazard to the terrestrial ecosystem through food
and water intake due to one tonne of the chemical). Cave and
Edward’s work was continued by Gunasekera and Edwards (2003)
who then developed the Atmospheric Hazard Index (AHI) to assess
the potential impact of airborne release from a chemical production
process. The AHI considers five categories: toxicity, photochemi-
cal smog, acid deposition, global warming, and stratospheric ozone
depletion. Both the EHI and AHI were developed to quantify the
impact of total loss of chemical containment in which the distri-
bution of chemical towards environment is estimated using level I
fugacity model (Cave and Edwards, 1997; Gunasekera and Edwards,
2003). Later, the two indices were combined as Inherent Envi-
ronmental Toxicity Hazard (IETH) to measure toxicity hazard to
the aquatic, terrestrial, and atmospheric environment due to the
catastrophic failure of chemical process (Gunasekera and Edwards,
2006). Another work developed to describe the flow and the gen-
eration of potential environmental impact is the waste reduction
algorithm (WAR) (Young et al., 2000). Among the factors considered
in the WAR are human toxicity potential by ingestion, human tox-
icity potential of exposure both dermal and inhalation, terrestrial
toxicity potential, aquatic toxicity potential, global warming poten-
tial, ozone depletion potential, photochemical oxidation potential,
and acidification potential. Similar factors were considered in the
work of Li et al. (2009). In the work of (Achoura et al., 2005), a
new index to assess environmental risk called Global Environmen-
tal Risk Assessment (GERA) was introduced. The development of
GERA took into account the two main indices: index for chemical
components present in process streams and index for unit opera-
tions. What differentiates the GERA index from other indices is that
GERA considers not only the inherent properties of chemical in a
plant (such as toxicity and flammability) but also the management
and organisational factors during assessment.
2.4. Inherent safety, health, and environment (SHE)
methodologies
Koller et al. (2000) highlights the lack of methodology that could
give a comprehensive view of the SHE aspects in a single framework
for the process development stage. Accordingly, a semi quantitative
index methodology was proposed consist of eleven factors. In this
work, the safest alternative after the consideration of technologi-
cal and organisational measures was the most favourable option.
Interestingly, Koller et al.’s work provides more flexibility for input
data because the assessor can choose the data to be used as input
according to their quality, and the data can be imported automat-
ically from an external source or flowsheeting programs (Koller
et al., 1999). The developed methodology was intended to be used
for chemical and fine chemical industries. Another work that con-
sidered the SHE components for tool development is INherent SHE
in DEsign (INSIDE Toolkit) (Mansfield et al., 2001). In INSIDE Toolkit,
the SHE components are evaluated separately and the overall per-
formance can be identified using the assistance of a multi-attribute
decision analysis technique. The developed indices are simple thus
enable rapid evaluation of process options. Another methodology
to counter the SHE problem was proposed by Srinivasan and Nhan
(2008). The method is known as Inherent Benign-ness Indicator
(IBI), which consists of fifteen factors related to SHE. In this work,
 M. Zainal Abidin et al. / Process Safety and Environmental Protection 117 (2018) 454–472
the combination of IBI with Principle Components Analysis (PCA)
techniques enable the identification of similarities and differences
between process routes as well as the identification of the main fac-
tors that lead to a route’s benign-ness. The results from the PCA are
important for route improvement, process design, and risk man-
agement. Other authors resolved the SHE multi-criteria problem
by using decision making tools. For example, Hassim and Ali (2009)
used simple additive weighting (SAW) for process route screening
at research and development stage. The index developed in this
work focuses on chemical substances data only because the pur-
pose of the index is to look at the inherent characteristics of the
process concept. The SHE components are also included in an index
form in the work by Narayanan et al. (2007) in order to identify
the most sustainable process for biodiesel production by consid-
ering their life cycle. The work of Liew et al. (2015, 2014) focuses
on developing a methodology for multi-criteria problem includ-
ing the SHE factors on the basis of the design stage. The developed
methodology was implemented to identify the best process routes
for biodiesel production.
2.5. Comparisons of developed methodologies
As discussed in the previous section, different methodologies
have been developed based on their goal and scope in order to
suit different objectives. The index methodologies are considered
good choices for a simple and quick solution. The consequence-
based methodologies, which were developed using a mathematical
model, will be a better choice for a reliable solution. For both
considerations of severity and failure potential, the risk-based
methodology will be the best option. Different methodologies will
produce different output as an indicator to identify the best ISD
alternative (Table 3). Focusing on the index methodology, simi-
lar pattern can be found in which the basis of index development
is the combination of different parameters, varies from chemical
properties related to the particular hazard in interest, process con-
dition parameters, and process equipment parameters, to list a few.
The combination will produce one final indicator that will serve
as the guidance to identify the best ISD alternative. Some of the
developed index methodologies included the parameters for tech-
nical and organisational requirement as part of the ISD assessment
steps. This inclusion will indicate the impact of ISD implementa-
tion towards the reduction of lifetime cost where a safer process
requires less technical and organisational measures for operation.
Table 4 lists previous developed methodologies with detailed infor-
mation about technical and organisational parameter considered
during their index development.
3. Gaps of the current methodologies and opportunity for
improvement
Despite the development of numerous methodologies in recent
years, there have been difficulties in adopting the ISD concept into
real design practice. In a recent study, Kidam et al. (2016) high-
lighted that one of the reasons for the inability to adopt ISD is the
low appreciation of the ISD concept, which has hastened its adop-
tion into design. Similar view is highlighted by Gupta et al. (2003),
who claimed that the real benefits of ISD are not realised by indus-
tries thus contributing to the poor response in implementing ISD
into practice. The extra costs of processing hazardous substance are
often assumed to be due to trips and alarms, compliance with regu-
lations, and workers’ training. In normal practices, these expenses
are usually estimated only as a small fraction of the total capital
and operating costs. Without good incentive, industrial practition-
ers will be more reluctant to allocate extra efforts in searching for
safer alternatives. Table 5 gives a detailed breakdown of thirty-one
459
items in order to demonstrate how ISD selection can give great
benefits in terms of economy and safety. According to Gupta et al.
(2003), by operating ISD operation, significant cost reduction can
be achieved via the items listed in Table 5; the total amount that
can be achieved is up to one-third to one-half, or even more, of the
capital and operating costs. Certainly, the items listed in Table 5
are the extra requirements for technical and organisational mea-
sures in managing a hazardous process. Details of the items listed
in Table 5 are explained as follows:
Compared to inherently safer process, operating a hazardous
process will result in additional requirements to be fulfilled by plant
owner in order to ensure safer operation. Failure to do so will cause
a hefty fine. Breach of regulations is not uncommon in the indus-
try. In 2009, The U.S. Department of Labor’s Occupational Safety
and Health Administration (OSHA), issued more than $87 million
penalties to BP Texas due to the violation of process safety man-
agement, the largest amount recorded in the OSHA history. Other
than that, risk analysis is required to be performed for operating a
hazardous process. According to Malaysia’s regulation, a detailed
quantitative risk analysis is required for safety report submission
to the regulatory body. In order to do so, the personnel need to be
trained to perform risk analysis, or the plant owner has to consult
competent persons. The service of competent persons will reduce
the number of reports that are rejected, require modification, or
require correction. On-site and off-site emergency response plants
need to be prepared, which require the involvement of civic author-
ities. The plan needs to be updated regularly and the designated
person needs to be trained. From the risk analysis, the equipment
including active (e.g., leak detectors and sensors, alarms and con-
trol equipment) and passive (e.g., dikes, catch pots) need to be
installed in the plant to limit the consequences of major accidents
on human health and environment. This requirement also covers
the installation required to handle off-spec materials and spills. In
the case of emergency, the training and mock drills are required
and need to be conducted regularly. Other than that, on-site medi-
cal facilities are required, and personal protective equipment must
be provided to the workers. Under emergency response planning,
the internal reporting systems for all near-miss incidents and acci-
dents are required and handled by trained personnel. Usually, a
safety and health committee need to be established in the company
and the meeting among committee members will be held in regular
intervals. Any expansion/modification performed in the hazardous
plant must be approved by authorities prior to the modification.
The arrangement of buildings, layout, design, and construction
must follow a standard code. According to Malaysia’s regulation,
a detailed plant layout that shows the arrangement of equipment
must be submitted to regulatory body for approval. More than that,
operating hazardous plant requires all equipment and instruments
to function properly hence require regular maintenance. Other than
the cost required to train maintenance personnel/hire maintenance
personnel, details of the maintenance need to be recorded and
the inspector from regulatory body has an authority to access the
record whenever required. Product handling, packaging, and trans-
portation need to follow the strict procedure and guidelines. The
driver must be trained and be capable of handling hazardous mate-
rials, and the off-site emergency plan needs to be prepared in case
of material release caused by accidents during transportation. To
ensure the monitoring process and controller perform as intention,
the software must be reliable and handled by trained personnel. As
the process to manage hazardous process requires the commitment
and involvement from civic authorities or inspector from regula-
tory bodies, good relations with them must be maintained. The staff
must be prepared to host their visit and prepare the equipment so
that it is ready for inspection and manage the follow up action. In
Malaysia’s regulation, there is a clause that allows the inspector to
cancel the inspection if the process/equipment is not prepared for
 460
Table 3
Comparison on ISD indication way for previous developed methodologies.

Component Methodology Authors Indicator of preferable alternatives

Changes of index value Changes of Change of Changes of Changes of Changes of

accident severity probability technical & region/category objective
organisational function
requirement

M. Zainal Abidin et al. / Process Safety and Environmental Protection 117 (2018) 454–472
Safety
√
Index- PIIS Lawrence (1996)
√
based ISI Heikkilä (1999)
√
i-Safe Index Palaniappan et al. (2004, 2002a, 2002b)
√ √
Fuzzy logic index Gentile et al. (2003)
√ √ √
I2SI Khan and Amyotte (2005, 2004)
√
PRI Leong and Shariff (2009)
√
Explosion index Salzano and Di Benedetto (2012)
√
PSI Shariff et al. (2012)
√ √ √
IRDI Rusli et al. (2013)
√ √
STEP Chan et al. (2014)
√
NuDIST Ahmad et al. (2014), Ahmad et al. (2016b)
√ √
2DGR. Ahmad et al. (2016a)
√ √
ISAPEDS Ahmad et al. (2017)
√ √
Consequence- KPI Landucci et al. (2008), Tugnoli et al. (2012, 2009, 2007)
√
based Consequence Shariff et al. (2006), Shariff and Wahab (2013), Shariff and
Risk- analysis Zaini (2010), Zaini et al. (2016)
√ √ √
based Cost of accident Bernechea and Arnaldos (2014), Bernechea and Viger
(2013), Eini et al. (2015), Medina et al. (2009)
√ √ √
IRA Shariff and Leong (2009), Shariff and Zaini (2013), Jha et al.
(2016)
√ √ √
Petri net Moradi (2010)
√ √ √ √
RISI Rathnayaka et al. (2014)
√
Health PRHI Hassim and Edwards (2006)
√ √
IOHI Hassim and Hurme (2010)
√
Environment EHI Cave and Edwards (1997)
√ √
WAR Young et al. (2000)
√
AHI Gunasekera and Edwards (2003)
√ √
GERA Achoura et al. (2005)
√
IETH Gunasekera and Edwards (2006)
√ √
Li et al. (2009)
√ √
SHE Koller et al. (2000, 1999)
√
INSIDE Mansfield et al. (2001)
√
Narayanan et al. (2007)
√
IBI Srinivasan and Nhan (2008)
√
Hassim and Ali (2009)
√ √
Liew et al. (2015, 2014)
 M. Zainal Abidin et al. / Process Safety and Environmental Protection 117 (2018) 454–472 461

Table 4 including organisations, workers, and surrounding community.

Comparisons of parameters considered for technical and organisational
Most of the methodologies in Table 4 (other than GERA and IRDI)
requirement.
focus on the impact of ISD implementation within the plant.
Methodology Authors Parameters for technical and Further, the current developed methodologies/tools are unable
organisational requirement
to reflect the ISD benefits discussed in the previous paragraph
I2SI Khan and Amyotte Controller requirement properly, which can result in the underestimating of ISD value.
(2005, 2004) Add-on safety measure requirement ii. Due to unavailability of data about technical and organisational
GERA Achoura et al. (2005) Plant preparedness index (to hazard
measures at the early stage of design, the developed index heav-
situations)
Community knowledge index (of ily depends on the subjective evaluation supported with the
hazard situations) knowledge about the process as indicated in the literature. Other
Community preparedness index than the results itself may be subject to different biases, there
Operator’s index
is a challenge in finding sufficient data for a new process or
Inspection index
Maintenance index new hazardous materials that have limitedly informed in the
Isolation index (independent unit literature. Thus, this approach will result in uncertainties dur-
operation) ing the evaluation process and inconsistency of the final results.
Failure detectability index More reliable methodologies that consider technical and organ-
Controllability index (in case of failure)
isational measures during alternatives assessment, such as Dow
Spill index (in case of failure)
Petri net Moradi (2010) Controller requirement Fire and Explosion Index (Dow F&EI) (Dow’s Fire and Explosion
Add-on safety measure requirement Index Hazard Classification Guide, 1994) and Safety Weighted
Koller et al. (2000, Technical and organisational measures Hazard Index (SWeHI) (Khan et al., 2001), are considered not
1999)
suitable to be used at the early/preliminary design stage because
IRDI Rusli et al. (2013) Controllability – basic requirement
Controllability – technical requirement
more detailed data are required.
Complexity on overall process unit and
plant
Therefore, it becomes a motivation for this work to develop a
RISI Rathnayaka et al. Hazard control requirement
(2014) Risk control requirement simple index methodology that can produce consistent and reli-
able results, which can identify the safest alternatives and reflect
the benefits of ISD implementation properly. The findings through
inspection. In such occurrence, the company needs to arrange for the detailed reviews show that the keys to resolving this problem is
another inspection date. Other than that, good relationships with by relating the parameters listed in Table 5 with regulatory require-
media and community need to be maintained. In recent years, many ments. This finding brings to the breakthrough in understanding
projects were cancelled due to poor relationship with media and the limitation of the current available methodologies where most of
community, which had led to the objection from them. The com- the regulatory frameworks on inherent safety provide no detailed
munity has the right to:(1) know the nature of the major accident mention of the incentives of using it. Due to the lack of regula-
hazards including their potential effects on human health and the tory framework, inherent safety still leaves a choice for designers
environment, and (2) summarise the details of the main types of to choose accruing to convenience. In order to resolve this problem,
major accident scenarios and the control measures to address them. an integrated inherent safety-regulatory framework is proposed in
With the increment on the awareness related to safety and health this work with the aims to show that the adoption of ISD as the
environment, operating hazardous operation requires extra efforts first element of safety can give significant benefits for the oper-
on educating the community. ation. The proposed framework was intended to be used for ISD
In comparing the list of items in Table 5 with the current evaluation during preliminary design stage in line with the design
methodology that is capable of identifying the technical and organ- practices given in Table 1.
isational measures (Table 4), two major drawbacks can be found:
4. Integrated inherent safety-regulatory framework
i. Different parameters are considered for the technical and
organisational measures for each methodology, in which the The integrated inherent safety-regulatory framework shown in
parameters considered depend heavily on the developer’s opin- Fig. 4 comprises three key steps: (1) hazard identification for the
ions. The impact of ISD may affect several interested parties, base case and ISD alternatives, (2) evaluation of Damage Index

Table 5
Items affected by safety level (Gupta et al., 2003).

No Items

1) Compliance with regulations 17) Handling spills/outdated/off-spec materials

2) Risk analysis 18) Transport, loading and unloading of hazardous materials
3) Emergency response plans 19) Software reliability
4) Fire and explosion protection 20) Accident investigation and reporting
5) Leak detectors and sensors 21) Personal protective equipment (PPE)
6) Alarms and control equipment 22) Insurance costs
7) Passive barriers 23) Decommissioning costs
8) Personnel training and mock drills 24) On-site medical facilities
9) Expansion, modification of plant/process 25) Relations with regulatory and civic authorities
10) Code-certified buildings 26) Relations with media and community
11) Sophisticated research and testing laboratory 27) A responsible care program
12) Maintenance of equipment and instruments 28) New regulations at site
13) Cost of raw material 29) Site security
14) Cost of equipment 30) Library
15) Product-related concerns 31) Litigation
16) Number of employees
462 M. Zainal Abidin et al. / Process Safety and Environmental Protection 117 (2018) 454–472
Fig. 4. Integrated inherent-safety-regulatory framework.
(DI) and Hazard Management Index (HMI) for ISD alternatives, and
(3) identification of the best option using Inherent Safety Benefits
Index (ISBI). The detailed procedure for the first step can be found
in Zainal Abidin et al. (2016b); the first step serves to identify the
hazard from the base case and to review alternative design based
on the ISD approach using Three Stage ISD Matrix (TIM Tool). The
suggested procedure is a qualitative approach that aims to help
the assessor to understand the hazard and the impact of ISD mod-
ification towards different criteria. This paper will focus on the
development of a quantitative index methodology ISBI, to simul-
taneously identify and quantify the benefits of ISD options. ISBI
compromises of two main components, DI and HMI. Details of the DI
used in this work will be explained in Section 4.1. The HMI proposed
in this work will be presented in Section 4.2. The final objective is to
identify the best ISD alternatives with the lowest accident severity
potential and the lowest technical and organisation measures to
manage hazards. The ISBI index will be proposed to comprise three
categories: (1) significant positive impact of ISD implementation,
(2) less significant positive impact of ISD implementation, and (3)
insignificant positive impact of ISD implementation. Details of the
ISBI formulation is presented in Section 4.3.
4.1. Damage Index (DI)
In this framework, the DI value is calculated using an improved
HIRA methodology as formulated in Khan et al. (2001), Khan and
Abbasi (1998), and Khan and Amyotte (2004). The HIRA methodol-
ogy proposed by Khan and Abbasi (1998) comprises the estimation
of accident severity of fire and explosion as well as toxic impact
based on the total energy and inventory for equipment, together
with penalties for process parameters that can influence the sever-
ity. Through the years, several modifications, improvements, and
developments were made to the index related to the HIRA method-
ology. Based on the Accident Hazard Index (AHI) methodology
(Khan and Abbasi, 1997), the HIRA methodology is extended to
cover not only the impact of major accident for fatalities but also
the environment, damage to property, and other support systems.
In the SWeHI, the HIRA methodology was improved by considering
the credits for safety measures. In the I2SI, the DI estimation was
used as a basis to estimate the cost of accident. The index consists
of cost of fatalities/injuries, cost of production loss, asset lost, and
environmental clean-up cost (Khan and Amyotte, 2005, 2004). HIRA
methodology was selected for DI estimation in this study because
it provides reliable results because the models and penalties were
derived from well-tried and tested mathematical models and a few
penalties are quantified using empirical models and hazard ranking
procedures. The DI estimation was included in this work with the
sole purposes of representing the indicator of insurance cost born
by the company due to the losses cause by the major accidents.
Fig. 5. HMI framework.
4.2. Hazard Management Index (HMI)
In this work, the Hazard Management Index (HMI) is used as an
indicator for the regulatory body requirements to manage the haz-
ard through technical and organisational measures. The framework
proposed for the HMI identification comprises five steps as shown
in Fig. 5. In this work, (Chemicals Act (Control Of Major Accident
Hazards Involving Dangerous Substances), 2015) has been chosen
as a platform to develop the index due to the comprehensiveness
of this regulation. The advantages using Control Of Major Accident
Hazards (COMAH 2015) as a basis of the index development are:
i Review of COMAH regulation shows that this regulation is very
comprehensive and can covers most of the items listed in Table 5.
Further, COMAH capable to address the requirement to manage
hazardous process for the process that has a potential to cause
major accidents that can harms human and environment.
ii COMAH 2015 give the flexibility to the regulated entity to deter-
mine the technical and safety measure to control hazard and
to keep risk based on as low as reasonably practicable (ALARP)
concept. It has a similar pattern with the current Malaysia’s reg-
ulation, Control of Industrial Major Accident Hazards (CIMAH
1996) since the current Malaysia’s regulation was developed
based on CIMAH UK1984
iii In the latest COMAH enforced in 2015, the class of hazard was
aligned with Globally Harmonised System (GHS). In this work,
these two regulations were linked to the United Nations Rec-
ommendations on the Transport of Dangerous Goods (UNRTDG)
requirement. The details on correlation proposed will be dis-
cussed in the next section. Therefore, the index proposed using
COMAH 2015 can covers the items related to the requirement
for transportation, packaging and labelling as shown in Table 5.
Details of each step are as follows:
Step 1
The first step for the HMI identification is to select the type of
installation. The installations, which are suggested based on the
definition given in Part 1, consisted of seven options:
 M. Zainal Abidin et al. / Process Safety and Environmental Protection 117 (2018) 454–472 463

Table 6
List of named hazardous substances (Chemicals Act (Control Of Major Accident Hazards Involving Dangerous Substances), 2015).

Hazardous substances Lower tier, QL (kg) Upper tier, QU (kg)

1-Bromo-3-chloropropane 500000 2000000

2,4 —Toluene diisocyanate 10000 100000
2,6 —Toluene diisocyanate 10000 100000
2-Methyl-3-butenenitrile 50000 2000000
3-(2-Ethylhexyloxy) propylamine 50000 200000
3-Methylpyridine 500000 2000000
4, 4 -Methylene bis (2-chloraniline) and/or salts, in powder form 0 10
Acetylene 5000 50000
Ammonium nitrate 5000000 10000000
Ammonium nitrate 1250000 5000000
Ammonium nitrate 350000 2500000
Ammonium nitrate 10000 50000
Anhydrous Ammonia 50000 200000
Arisine (arsenic trihydride) 200 1000
Arsenic pentoxide, arsenic (V) acid and/or salts 1000 2000
Arsenic trioxide, arsenious (III) acid and/or salts 0 100
Bis(2-dimethylaminoethyl) (methyl)amin 50000 200000
Boron trifluoride 5000 20000
Bromine 20000 100000
Carbonyl dichloride (phosgane) 300 750
Chlorine 10000 25000
Ethylene oxide 5000 50000
Ethyleneimine 10000 20000
Fluorine 10000 20000
Formaldehyde (concentration >=90%) 5000 50000
Hydrogen 5000 50000
Hydrogen chloride (liquefied gas) 25000 250000
Hydrogen sulphide 5000 20000
Lead alkyls 5000 50000
Liquefied flammable gases, Category 1 or 2 (including LPG) and natural gas 50000 200000
Methanol 500000 5000000
Methyl acrylate 500000 2000000
Methylisocyanate 0 150
Mixtures (*) of sodium hypochlorite classified as Aquatic Acute Category 1 [H400] containing less than 5% active 200000 500000
chlorine and not classified under any of the other hazard categories in Part 1 of Schedule 1. (*) Provided that the
mixture in the absence of sodium hypochlorite would not be classified as Aquatic Acute Category 1 [H400].
Nickel compounds in inhalable powder form: nickel monoxide, nickel dioxide,nickel sulphide, trinickel 0 1000
disulphide,dinickel trioxide
Oxygen 200000 2000000
Petroleum products and alternative fuels 2500000 25000000
(a) gasolines and naphthas,
(b) kerosenes (including jet fuels),
(c) gas oils (including diesel fuels, home heating oils and gas oil blending streams)
(d) heavy fuel oils
(e) alternative fuels serving the same purposes and with similar properties as regards flammability and
environmental hazards as the products referred to in points (a) to (d)
Phosphine (phosphorus trihydride) 200 1000
Piperidine 50000 200000
Polychlorodibenzofurans and polychlorodibenzodioxins (including TCDD), calculated in TCDD equivalent 0 1
Potassium nitrate 5000000 10000000
Potassium nitrate 1250000 5000000
Propylamine 50000 2000000
Propylene oxide 5000 50000
Sulphur dichloride 0 1000
Sulphur trioxide 15000 75000
Tert-butyl acrylate 20000 500000
Tetrahydro-3,5-dimethyl-1,3,5,-thiadiazine-2-thione (Dazomet) (see note 21)thione (Dazomet) 100000 200000
The following CARCINOGENS or the mixtures containing the following carcinogens at concentrations above 5% by 500 2000
weight:
4-Aminobiphenyl and/or its salts, Benzotrichloride, Benzidine and/or salts, Bis (chloromethyl) ether, Chloromethyl
methyl ether, 1,2-Dibromoethane, Diethyl sulphate, Dimethyl sulphate, Dimethylcarbamoyl chloride,
1,2-Dibromo-3-chloropropane, 1,2- Dimethylhydrazine, Dimethylnitrosamine, Hexamethylphosphoric triamide,
Hydrazine, 2- Naphthylamine and/or salts, 4-Nitrodiphenyl, and 1,3 Propanesultone

i All the equipment, structures, pipework, machinery, and tools vii Warehouses or similar structures
that produce dangerous substances
ii All the equipment, structures, pipework, machinery, and tools Step 2
that use dangerous substances After that, the chemical involved in the analysed installation
iii All the equipment, structures, pipework, machinery, and tools is selected from a list of 49 hazardous substances, each with its
that handle/store dangerous substances threshold value (Table 6). The list is suggested based on the infor-
iv Private railway sidings mation given in Schedule 1 (Part 2). An inventory of hazardous
v Docks and unloading quays serving the installation substances needs to be provided as an input to identify the HMI
vi Jetties value. If the analysed substance is not available from the list in
 464
Table 7
List of hazardous substance classification based on general hazard (Chemicals Act (Control Of Major Accident Hazards Involving Dangerous Substances), 2015).

Type of hazard Hazard classification Description Lower tier, Upper tier,

QL (kg) QU (kg)

Section H1 (Acute Toxic Category 1) All exposure routes 5000 20000

‘H’ H2 (Acute Toxic Category 2) All exposure routes 50000 200000

M. Zainal Abidin et al. / Process Safety and Environmental Protection 117 (2018) 454–472
H2 (Acute Toxic Category 3) Inhalation exposure route 50000 200000
H3 (STOT Specific Target Organ Toxicity) STOT Category 1 50000 200000
Section P1a(i) (Explosives) Unstable explosives 10000 50000
‘P’ P1a(ii) (Explosives) Explosives, Division 1.1, 1.2, 1.3, 1.5 or 1.6 10000 50000
P1a(iii) (Explosives) Substances or mixtures which have explosive properties according to method A.14 of Regulation 10000 50000
(EC) No. 440/2008 of 30 May 2008 laying the European Parliament and of the Council on the
Registration, down test methods pursuant to Regulation (EC) No. 1907/2006 of Evaluation,
Authorisation and Restriction of Chemicals (REACH) (see note 9) and do not belong to the hazard
classes Organic peroxides or Self-reactive substances and mixtures
P1b (Explosives) Explosives, Division 1.4 50000 200000
P2 (Flammable Gases) Flammable gases, Category 1 or 2 10000 50000
P3a (Flammable aerosols) ‘Flammable’ aerosols Category 1 or 2, containing flammable gases Category 1 or 2 or flammable 150000 500000
liquids Category 1
P3b (Flammable aerosols) ‘Flammable’ aerosols Category 1 or 2, not containing flammable gases Category 1 or 2 nor 5000000 50000000
flammable liquids category 1
P4 (Oxidising gases) Oxidising gases, Category 1 50000 200000
P5a(i) (Flammable liquids) Flammable liquids, Category 1 10000 50000
P5a(ii) (Flammable liquids) Flammable liquids Category 2 or 3 maintained at a temperature above their boiling point 10000 50000
P5a(iii) (Flammable liquids) Other liquids with a flash point ≤ 60 ◦ C, maintained at a temperature above their boiling point 10000 50000
P5b(i) (Flammable liquids) Flammable liquids Category 2 or 3 where particular processing conditions, such as high pressure 50000 200000
or high temperature, may create major accident hazards
P5b(ii) (Flammable liquids) Other liquids with a flash point ≤ 60 ◦ C where particular processing conditions, such as high 50000 200000
pressure or high temperature, may create major accident hazards (see note 12)
P5c (Flammable liquids) Flammable liquids, Categories 2 or 3 not covered by P5a and P5b 5000000 50000000
P6a(i) (Self-reactive substances and mixtures and organic peroxides) Self-reactive substances and mixtures, Type A or B 10000 50000
P6a(ii) (Self-reactive substances and mixtures and organic peroxides) Organic peroxides, Type A or B 10000 50000
P6b (Self-reactive substances and mixtures and organic peroxides) Self-reactive substances and mixtures, Type C, D, E or F or organic peroxides, Type C, D, E, or F 50000 200000
P7(i) (Pyrophoric liquids and solids) Pyrophoric liquids, Category 1 50000 200000
P7(ii) (Pyrophoric liquids and solids) Pyrophoric solids, Category 1 50000 200000
P8(i) (Oxidising liquids and solids) Oxidising Liquids, Category 1, 2 or 3 50000 200000
P8(ii) (Oxidising liquids and solids) Oxidising Solids, Category 1, 2 or 3 50000 200000
Section E1 Hazardous to the Aquatic Environment in Category Acute 1 or Chronic 1 100000 200000
‘E’ E2 Hazardous to the Aquatic Environment in Category Chronic 2 200000 500000
Section O1 Substances or mixtures which reacts violently with water 100000 500000
‘O’ O2 Substances and mixtures which in contact with water emit flammable gases, Category 1 100000 500000
O3 Substances or mixtures which contact with water liberates toxic gas 50000 200000
 M. Zainal Abidin et al. / Process Safety and Environmental Protection 117 (2018) 454–472 465

Table 8 Table 10
Criteria for HMI identification using Step 2/Step3a. Requirement from COMAH based on HMI value.

HMI Criteria HMI Requirement from COMAH

1 q < QL 1 Compliance to other relevant legislation

2 QU > q ≥ QL Health and safety legislation
3 q ≥ QU Environmental legislation
2 Compliance to other relevant legislation
Health and safety legislation
Table 9 Environmental legislation
Criteria for HMI identification using Step 3b. Notification
Information in Paragraph 8
HMI Criteria Update every five years/modification
1 L1 < 1 Preparing MAPP
2 L1 > 1 Policy on the prevention of major accident
3 U1 > 1 Information about safety management system that will be
used to put the policy into action
3 Compliance to other relevant legislation
Health and safety legislation
Table 2, general hazard classification will be used to determine the
Environmental legislation
HMI value (Step 3). Notification
Step 3a Information in Paragraph 8
In this step, the process is evaluated using the general hazard Update every five years/modification
classification as the guidance given in Schedule 1 (Part 1). There are Preparing MAPP
Policy on the prevention of major accident
four sections under this classification: Section H for health hazard, Information about safety management system that will be
Section P for physical hazard, Section E for environmental hazard, used to put the policy into action
and Section O for other hazards. (Details are shown in Table 7). The Preparing safety report
inventory of hazardous substances needs to be provided as input Demonstration
Information in Sch 3
to identify the HMI value.
Name of consultation company
Step 3b Update every 5 years/modification
In the case where Step 2 and Step 3a were performed yet no Preparing & testing an emergency plant (Sch 4)
individual dangerous substance is present in a quantity above or Internal emergency plant
equal to the relevant qualifying quantities, the indicator for HMI External emergency plant
Supply information to local authorities and public
value is determined using the aggregation formula in Eq. (1) or Eq.
Information in Sch 5 (Part 1 & 2)
(2).

L1 = q1 /QL1 + q2 /QL2 + q3 /QL3 + ... + qn /QLn (1)

U1 = q1 /QU1 + q2 /QU2 + q3 /QU3 + ... + qn /QUn (2)

Where U1 is the indicator for aggregation using upper tier value,

L1 is an indicator for aggregation using lower tier value, qn is the
inventory of nth hazardous material in the establishment in kg, QUn
is the upper tier value for nth hazardous material in the establish-
ment in kg, QLn is the lower tier value nth hazardous material in the
establishment in kg. If U1 produce the outcome more than 1, HMI
value will be determined using U1 . If calculation of U1 produce the
value less than 1, HMI will be determined using L1 .
Step 4
Based on Step 2, Step 3a, and Step 3b, the HMI value is deter-
mined according to the criteria given in Tables 8 and 9. The HMI
value indicates the requirement to manage the process that has the Fig. 6. Correlation between GHS, COMAH and UNRTDG.

potential of major accident. If the user uses Step 2, the HMI value
can be identified directly from Table 8 where q is the inventory
of hazardous material in the establishment in kg, QU is the upper
tier value for hazardous material in the establishment in kg, QL is
the lower tier value of hazardous material in the establishment in
kg. If the user uses Step 3a to identify the HMI value, the user is
required to evaluate all the hazards available (Section P, Section E,
Section H and Section O) using the criteria in Table 7, only the high-
est index obtained will be counted as shown in Eq. (3). For Step 3b,
the HMI value will be determined using U1 and L1 using the criteria
in Table 9.
HMI = IF(Named substances = "", MAX(IP , IH , IE , IO )
Where IP is the indicator for physical hazard, IH is the indicator for
health hazard, IE is the indicator for environmental hazard and IO
is the indicator for the other hazard.
Step 5
In this step, HMI can be related to the requirement to man-
age the process from the regulatory body. Unsafe alternatives with
a high HMI value indicates that more requirements need to be
fulfilled for the hazardous process operation and thus (a) requir-
ing more expertise and efforts, (b) incurring higher cost, and (c)
being time consuming. Details of the COMAH requirements based
on (Chemicals Act (Control Of Major Accident Hazards Involving
Dangerous Substances), 2015) according to the index value are
summarised in Table 10.
Although the HMI was developed using COMAH as a basis,
there is a special correlation between COMAH, GHS, and UNRTDG.
The relationship between these three regulations is summarised
in Fig. 6. Following the correlation, the requirement from GHS
(3)
and UNRTDG can be obtained concurrently as we evaluate the
alternatives for HMI value that represents the COMAH require-
ments. As can be seen, COMAH is a subset of GHS (COMAH ⊆GHS),
therefore the alternatives need to fulfil (a) the requirement from
COMAH as listed in Table 10 and (b) the GHS requirements as
well. The total requirements that need to be fulfilled by the alter-
natives will be determined according to the intersection area,
466 M. Zainal Abidin et al. / Process Safety and Environmental Protection 117 (2018) 454–472

Table 11 tionship between these regulations with ISD benefits listed in

Requirement from GHS and UNRTDG (Industry Code of Practice on Chemicals Clas-
Table 5 are given in Table 12.
sification and Hazard Communication, 2014; Recommendations on the Transport of
Dangerous Goods, 2011; Regulation (EC) No 1272/2008 on Classification, Labelling
and Packaging of Substances and Mixture, 2008). 4.3. Inherent Safety Benefits Index (ISBI)
GHS UNRTDG
In this section, the ISBI will be proposed to identify the best
Labelling requirement Consignment procedures
ISD alternative with the lowest severity of accidents and lowest
Information on a label Marking and labelling of packages
Multiple hazards and precedence of Documentation requirement from regulatory bodies to manage the hazard. ISBI
information Placarding of cargo transport units was proposed based on the relative ranking concept (Center for
Arrangements for presenting the label Chemical Process Safety, 2011; Edwards and Barron, 1994) where
elements the concept allows the assessor to compares the attributes of sev-
Special labelling arrangements
eral processes or activities in numerical grading before details
Miscellaneous
Preparation of safety data sheet (SDS) Periodically training including analysis. The concept of risk matrix was adopted in order to propose
Identification of the hazardous General awareness/familiarization the region of ISBI index. These two concepts have been adopted in
chemical and of the supplier training the ISBI development due to the simplicity factor. Both the outcome
Hazard identification Function-specific training
from Section 4.1 (DI) and Section 4.2 (HMI) will be compared to the
Composition and information of the Safety training
ingredients of the hazardous chemical base case value in order to quantify the impact of the ISD imple-
First-aid measures mentation. The changes of the DI and HMI can be calculated using
Fire-fighting measures the formula given in Eqs. (4) and (5), respectively. The changes of
Accidental release measures the DI will be presented in a range of −9 to 9, and the changes of HMI
Handling and storage
will be presented in the range of −2 to 2. A negative value indicates
Exposure controls and personal
protection that the proposed modification can give a negative impact by the
Physical and chemical properties increment of the DI or HMI value of the alternatives from the base
Stability and reactivity case. A positive value indicates a positive impact by the reduction
Toxicological information
of DI or HMI value of the alternative from the base case. A zero out-
Ecological information
Disposal information
come indicates that the modification does not give any changes in
Transportation information terms of DI or HMI value compared to the base case. By combining
Regulatory information these two components, the best ISD alternative is indicated by the
Other information ISBI index as shown in Eq. (6) with the range 0 < ISBI ≤ 18 and with
Training Security training & security plan
three potential outcomes as detailed out in Table 13. The ranking
Elements of security plan:
Specific allocation and responsibilities process will be performed according to the outcomes. In this work,
for personnel the alternatives will be ranked according to the ISBI value; the best
Records of hazardous material ISD alternative is indicated by the highest ISBI value. In the case,
transported
all the alternatives being evaluated fall into category 2 (Table 13),
Review of current operations and
vulnerability assessment
which will produce ISBI = 0, therefore the ranking process must be
Clear statements of measures including done according to HMI or DI value whichever is available.
training/policies/operating
practices/equipment and resources         
DIalt DIbc
used to reduce security risks DI = IF DIalt > DIbc , Max −9, −1 × −1 , Min 9, −1 ×1
DIbc DIalt
Effective and up to date procedures for
reporting and dealing with security
(4)
threats, breaches of security and
security incidents      
Procedures for the evaluation and HMIalt HMIbc
HMI = IF HMIalt > HMIbc , −1 × −1, −1 ×1 (5)
testing of security plants and HMIbc HMIalt
procedures for periodic review and
update of the plans ISBI = IF(AND(HMI > 0, DI > 0), HMI × DI, 0) (6)
Measures to ensure the security of
transport information contained in the Where DI is the changes of DI, HMI is the changes of HMI, DIalt
plant is the DI for ISD alternative, DIbc is the DI for base case, HMIalt is the
Measures to ensure that the HMI for ISD alternative, HMIbc is the HMI for base case.
distribution of the transport
information is limited as far as possible
Packaging 5. Results and discussion
Requirement to follow the type of
packaging based on type of hazardous This section is divided into two subsections. Section 5.1
materials
Requirement for the construction and
describes the case study used to demonstrate the proposed
testing of packaging methodology. Section 5.2 gives the explanation on DI, HMI and ISBI
results obtained using the proposed methodology. ISBI. The results
obtained using the ISBI were compared with the ISI developed by
GHS∩COMAH∩UNRTDG. Due to this special feature, the HMI can Heikkilä (1999).
show the positive impact brought by the implementation of ISD for
operation, transportation, packaging, and labelling. Table 11 sum- 5.1. Case study: ISD alternatives for ammonia storage
marises the extra requirements that need to be fulfilled when the
alternatives fall into GHS and UNRTDG according to Industry Code The objective of this section is to demonstrate how the proposed
of Practice on Chemicals Classification and Hazard Communication methodology can help to identify the best ISD for ammonia stor-
(2014), Recommendations on the Transport of Dangerous Goods age. Ammonia is a toxic material that exists in a gaseous form at
(2011), Regulation (EC) No 1272/2008 on Classification, Labelling ambient temperature. For ease and economical transportation and
and Packaging of Substances and Mixtures (2008). The details rela- storage, ammonia is liquefied either by cooling or compressing. In
 M. Zainal Abidin et al. / Process Safety and Environmental Protection 117 (2018) 454–472
Table 12
Details on the items covered by regulation.
Items Code
Compliance with regulations GHS/UNRTDG
Risk analysis COMAH Part 3, No 11, 2(b)
COMAH Part 2
COMAH Schedule 2(b),ii
COMAH Schedule 3
COMAH Schedule 2(b),iv
Emergency response plans COMAH Schedule 2(b),v
COMAH Schedule 4
COMAH Part 4
COMAH Schedule 2(b),v
COMAH Schedule 4
COMAH Part 4
Fire and explosion protection COMAH Schedule 3,No 5
COMAH Schedule 2(b),I
COMAH Schedule 2(b),v
COMAH Part 4, No 16
COMAH Part 4, No 13
COMAH Part 2,No 7,2(b)
COMAH Part 7
Leak detectors and sensors COMAH Schedule 3,No 5
Alarms and control equipment COMAH Schedule 3,No 5
COMAH Part 2, No2 (b)
COMAH Schedule 2,b(iii)
COMAH Part 2,No 7,2(b)
COMAH Part 2,No 11 (c)
Passive barriers COMAH Schedule 3,No 5
COMAH Part 7
Personnel training and mock drills COMAH Schedule 4 No1 (f)
COMAH Schedule 2 (b) (i)
COMAH Schedule 2(b) (v)
COMAH Part 4,No 15 (2)
COMAH Schedule 2(b) (i)
COMAH Schedule 2(b) (v)
COMAH Schedule 2(b) (v)
Expansion, modification of plant/process COMAH Part 2, No8, 4(b)
COMAH Part 3, No 10,1
COMAH Part 3, No 11,3(a)
COMAH Part 3,No 11,6(c)
COMAH Part 3,No 12,2
COMAH Part 8, No 25
COMAH Schedule 5
Code-certified buildings COMAH Schedule 3 No 5
COMAH Part 2,No 7,2(b)
Sophisticated research and testing laboratory GHS
Maintenance of equipment and instruments COMAH Part 2, No2 (b)
COMAH Schedule 2,b(iii)
COMAH Part 2,No 7,2(b)
COMAH Part 2,No 11 (c)
COMAH Schedule 2,b(vii)
COMAH Part 8
Product-related concerns GHS/UNRTDG
GHS/UNRTDG
Number of employees COMAH Schedule 2 (b) (i), (v), (vii)
COMAH Part 2, No 7, 2(b)
Handling spills/outdated/off-spec materials COMAH Schedule 3,No 5
Transport, loading and unloading of hazardous materials UNRTDG
Software reliability COMAH Schedule 2(b) (iii)
COMAH Schedule 2(b) (i)
Accident investigation and reporting COMAH Part 4
COMAH Part 4
COMAH Schedule 2 (b) (vi)
COMAH Schedule 2 (b) (i)
467
COMAH/GHS/UNRTDG
Requirement to comply with the regulation for packaging, labelling,
storing, transporting hazardous materials
Requirement for trained people/consultant to be hired for safety repot
Requirement for risk management system
Requirement for risk analysis and steps to minimise the risk
Requirement for MOC program
Requirement for emergency response plan including training and
updating through time
Preparation of detailed off-site emergency response plant involving
authorities including training of civic authorities staff
Requirement for equipment installed in the plant to limit the
consequences of major accidents for human health and environment
Training for personnel
Liaison and training of civic authorities staff
Requirement for security from illegal entry
Requirement for certain separation distance between plant public use
area
Requirement for equipment installed in the plant to limit the
consequences of major accidents for human health and environment
Requirement for equipment installed in the plant to limit the
consequences of major accidents for human health and environment
Requirement for maintenance
Requirement for equipment installed in the plant to limit the
consequences of major accidents for human health and environment
Requirement for certain separation distance between plant public use
area
Requirement for personnel training to handle and hazardous plant and
possible emergencies
Requirement for contract worker training
Mock drills requirement
Approval requirement from regulatory/authorities before modification
Open hearing with the community/interested parties
Requirement for equipment installed in the plant to limit the
consequences of major accidents for human health and environment
Monitoring of visitor access
Requirement for hazardous substance testing
Requirement for maintenance
Requirement for detailed record of maintenance and availability upon
request
Requirement for product filling, handling of filled bottles, storage,
labelling, and transportation
Requirement for UN/other international or bi- national conventions to
be followed and the recipient nation informed of the possible hazards
for export
Requirement of trained personnel for operation, security and
management
Requirement for equipment installed in the plant to limit the
consequences of major accidents for human health and environment
Strict rules related to the transportation construction,
Training for driver
Emergency planning for off-site transportation incident
Requirement to ensure process control software reliability
Requirement for personnel training that dealing with software and
alarm
Requirement to establish on-site and off-site emergency procedure
Requirement to maintain the liaison with civic authorities
Requirement of internal reporting system, tool, and personnel training
for investigation techniques
468 M. Zainal Abidin et al. / Process Safety and Environmental Protection 117 (2018) 454–472

Table 12 (Continued)

Items Code COMAH/GHS/UNRTDG

Personal protective equipment (PPE) COMAH Part 3, 3(a)
COMAH Part 3, 11(b)
Insurance costs Covered by DI calculation
On-site medical facilities COMAH Part 4, No 15 (2)
Relations with regulatory and civic authorities COMAH Part 6, No 22
COMAH Part 4
COMAH Part 4
Relations with media and community COMAH Part 8
COMAH Schedule 5
New regulations at site COMAH Part 2, No 7, 2(b)
COMAH Part 2, No 7, 2(d)
Site security COMAH Schedule 2 (b) (i)
COMAH Part 2, No7, 2(b)
COMAH Part 2, No7, 2(c)
COMAH Part 2, No 7, 2(d)
Library COMAH Part 2, No7, 2(d)
Litigation COMAH Part 10
Requirement to provide the PPE appropriate for the job
Insurance against business losses due to the fire, explosion, toxic release, etc.
Requirement of on-site medical facilities and personnel
Management and time allocation to manage inspector visits and follow up on
actions
Liaison with authorities for accident investigation
Liaison with civic authorities to conduct mock drill
Provision of information to the public requires a good relationship with print
and visual media to acquire good publicity. Public relations personnel need to
be hired. The company need to build good reputation by involving more with
community activities
New regulations need to be imposed to ensure safe operation such as checking
all employees, visitors, contractors, etc., entering the site for items of
safety/security concern etc.
Compliance to site security requirement, requirement for worker training to
ensure secure operations and access to the area that handled hazardous
substance will be more stringent
The provision to provide the info related to hazardous operations such as
applicable government regulations, standards, current industry practices, and
latest data on the hazards
Regulatory authorities may sue for non-compliance

Table 13
ISBI classification.

Table 14
ISD alternatives for ammonia storage.

Options Base case Alternative 1 Alternative 2 Alternative 3

Types of storage High pressure storage Atmospheric storage High pressure storage Atmospheric storage
Substance Anhydrous ammonia Anhydrous ammonia Aqueous ammonia (29%) Aqueous ammonia (19%)
Pressure (kPa) 1274.86 101.33 108.27 101.33
Temperature (C) 25.00 −33.40 25.00 25.00
Mass (kg) 40 000.00 160 000.00 95 663.00 881 118.00
No of unit 4 1 6 1

the work of Roy et al. (2011), pure ammonia was stored in a lique-
fied form in high pressure storage. The design was hazardous due
to the high concentration of toxic material and high-pressure oper-
ation. In this study, three ISD alternatives were proposed according
to the normal industrial practice where the number and capacity of
tanks were estimated to fulfil the process requirement and the lim-
itations addressed in previous studies (Chavan, 2012; Lele, 2008;
Long and Gardner, 2004; UN Industrial Development Organization,
1998; Wiley-VCH, 2003). Detailed information for alternatives is
shown in Table 14. Alternative 1 offers a hazard reduction chance
based on the moderation principle, particularly by operating the
hazardous system in less hazardous conditions. The high-pressure
tank was replaced with a cryogenic storage system, in which the
liquefied ammonia was stored using refrigerant at ambient pres-
sure. Alternative 2 also offers hazard reduction via the moderation
principle, particularly through a dilution approach, in which the
aqueous ammonia of 29% concentration was stored in a high-
pressured vessel. Alternative 3 offers hazard reduction potential
using the moderation principle, particularly by replacing the high-
pressure tank with an atmospheric storage system which stores
lower concentration of ammonia (19%) at ambient temperature. A
detailed review about the hazard imposed by the base case and the
impact of ISD modifications are discussed in Zainal Abidin et al.
(2016b).
5.2. Demonstration of integrated inherent-Safety-Regulatory
framework
5.2.1. DI identification
The DI calculation that involves acute toxicity and environmen-
tal impact is discussed in Zainal Abidin et al. (2016a), who also
discussed the influence of ISD on severity reduction via related
 M. Zainal Abidin et al. / Process Safety and Environmental Protection 117 (2018) 454–472 469

Table 15
Determination of HMI using named substances.

Approach Options Base case Alternative 1 Alternative 2 Alternative 3

Named substance Hazardous substance Anhydrous ammonia Anhydrous ammonia N.A. N.A.
Lower tier 50000 50000 N.A. N.A.
Upper tier 200000 200000 N.A. N.A.
HMI 2 2 1 1
General hazard GHS Code H331, H314, H318, H335 H331, H314, H318, H335 H302, H332, H314, H318, H335 H302, H332, H314, H318, H335
H221 H221 N.A. N.A.
H400, H411 H400, H411 H412 H412
Section H H2 H2 N.A. N.A.
Section P P2 P2 N.A. N.A.
Section E E1 E1 N.A. N.A.
HMI 2 2 1 1
Aggregation LI 3.20 3.20 N.A. N.A.
UI 0.80 0.80 N.A. N.A.
IH 2 2 1 1
LI 1.60 1.60 N.A. N.A.
UI 0.80 0.80 N.A. N.A.
IE 2 2 1 1
HMI 2 2 1 1

parameters in detail in the same paper hence will not be repeated
in this paper.
5.2.2. HMI identification
Section 5.2.2 demonstrates the use of the proposed method-
ology (Fig. 5) to identify the HMI value: (1) by using named
substances, (2) by using general hazard, (3) and by using aggre-
gation.
Table 15 summarises the results of the HMI identification using
the first approach (named substances). The HMI outcomes from
this approach indicated that only base case and Alternative 1 stored
anhydrous ammonia included in the lower tier of the COMAH regu-
lation enforcement. Alternative 2 and Alternative 3 were excluded
from the COMAH enforcement because aqueous ammonia is not
listed under the named substance category. From the results in
Table 15, Alternative 1 (single unit with 160000 kg capacity) and
base case (four units with 40000-kg capacity each) obtained the
same HMI because in accordance to the COMAH approach, the total
amount of hazardous substance that exists in the establishment
must be considered during evaluation. Thus, 160000 kg was used
for comparison with a threshold value for both alternatives.
In the second approach, general hazard was used to identify the
HMI value for the ammonia storage alternatives. In general hazard,
the HMI value was identified based on the class of hazard, where it
highly depends on the concentration of the hazardous substance.
Therefore, the first step is to verify if the concentration changes can
change the class of hazard. Based on the guidance from Regulation
(EC) No 1272/2008 on Classification, Labelling and Packaging of
Substances and Mixtures (2008), for the mixture with initial con-
centration between 25% and 100%, the permitted variation from the
initial concentration is 5%. If the variation of concentration exceeds
5%, the class of hazard need to re-evaluate. In this case, the initial
concentration of the anhydrous ammonia in the base case is 99.99%
and the ISD modification results in the same concentration with
Alternative 1, changes to 29% (Alternative 2) and 19% (Alternative
3) (Table 14). Clearly, the changes for Alternative 2 and Alternative 3
are significant enough and can change the class of hazard. After the
verification process, the class identification performed using the
GHS Code from the CF Industries (2016, 2015a, 2015b) as shown in
Table 15. Referring to Table 15, for Section H, although H331, H314
and H318 are the code for health hazard, only H331 were used for
evaluation. This decision was made because H331 was classified for
more severe effect H2, acute category 3 while H314 and H318 were
classified for H3, STOT. For section P, while the code indicates the
potential of major hazard due to the flammable gas, by consider-
ing the phase of materials, the tier value for flammable liquid was
used for HMI value identification. For section E, similar to section
H, the code with more severe effect was used as a basis of eval-
uation, in this case H411. Review on the other hazard shows that
the alternatives does not pose any hazard for section O. The HMI
value obtained following the tier value of the verified class was pre-
sented in Table 15. HMI value obtained using this approach come
to an agreement with the results obtained using the named sub-
stance approach where Base Case and Alternative 1 requires more
efforts in managing hazard compared to Alternative 2 and Alter-
native 3. According to the results obtained in Table 15, other than
toxic hazard, the Base Case and Alternative 1 also has a potential
for fire hazard. Thus, although the final HMI value has been deter-
mined according to the maximum value among IP , IE , IH and IO , the
major accident prevention policy (MAPP) or safety report prepare
in the later stage must address all type hazard that have potential to
cause the major accident. While in this work, this approach produc-
ing satisfying results, there is a potential of inaccuracy due to the
bridging principle problem as discussed by Wilday et al. (2012). To
resolve this issue, the assessor can identify the class of hazard using
another alternatives 1) determining the class of hazard using the
quantitative mixture data 2) determining the class of hazard using
the quantitative component data and formula given in Regulation
(EC) No 1272/2008 on Classification, Labelling and Packaging of
Substances and Mixtures (2008).
The identification of HMI using the third approach (aggregation
method) can only be used after the evaluation using the previ-
ous two approaches (named substances and general hazard) show
there is no individual dangerous substance was present in the
quantity above or equal to the relevant qualifying quantities in
the establishment. In order to show the function of the aggrega-
tion approach, the potential of major hazard for the anhydrous
ammonia came from acute toxicity and environmental hazard were
considered. Thus, the threshold for acute toxicity (H2) and chronic
category 1 for environment (E1) were used for aggregation calcu-
lation as outlined in Eqs. (1) and (2). The results show that the base
case and Alternative 1 were included under the lower-tier COMAH
regulation enforcement, whereas Alternative 2 and Alternative 3
were excluded from the COMAH (Table 15). The results obtained
using the aggregation approach are in agreement with the results
obtained using named substances and general hazard.
As conclusion, the findings show that the three approaches
presented in Fig. 5(using named substance, using general hazard
classification, and using aggregation formula) can be used to iden-
tify the HMI. The results obtained are in agreement with each
other indicate that with good input data, different approaches can
give consistent outcomes. In general, the HMI framework proposed
470 M. Zainal Abidin et al. / Process Safety and Environmental Protection 117 (2018) 454–472
Table 16
ISBI ranking for ammonia storage tank.
Options Base case Alternative 1 Alternative 2
DI 0 6.24 0.90
HMI 0 0 1 1
ISBI 0 0 0.90
Rank 4 3 2 1
offers more flexibility to the assessor in identifying the HMI value
according to the data available in hand. The most important, the
HMI value will serve as an indicator to measure the effectiveness
of ISD implementation in reducing the regulatory requirement to
manage hazardous process.
5.2.3. ISD identification using ISBI
The DI results from Zainal Abidin et al. (2016a) and the HMI
results (Table 15) were combined for the ISBI calculation as outlined
in the procedure given in Section 4.3. The results obtained for the
four alternatives are summarised in Table 16.
In terms of regulatory requirement, Alternative 3 and Alterna-
tive 2 are more favourable than Alternative 1 and the base case
because the former were excluded from COMAH, whereas Alterna-
tive 1 and the base case were included under the lower-tier COMAH.
According to the guidelines in Tables 10 and 11, if Alternative 1 or
base case was selected, the policy and info regarding the manage-
ment system to enable the implementation of the policy must be
provided, and the requirements for packaging, labelling, and trans-
portation must be prepared. The efforts, time, and source that are
spent in preparing all the requirements to manage the hazardous
process and the additional requirements often lead to the higher
chance of error, which might lead to a repeated tedious process or
worst, an accident. This is proven by the statistics obtained while
preparing the safety reports needed to fulfil the requirements from
upper tier COMAH in the United Kingdom; thirty-percent of the
reports were returned to the plant owner for further correction
(Britton, 2003). Similar case happened in Malaysia in which 599
errors/omissions were found from the 30 safety reports submit-
ted to the Control of Industrial Major Accident Hazards (CIMAH)
(Mustapha and Zain, 2003). The errors must be corrected and
go through an approval process from a competent person before
submission to the Department of Occupational Safety and Health
(DOSH). In addition to the tedious processes in preparing the safety
reports, the plant owner is required to inform the surrounding
community regarding the hazardous process that is being oper-
ated, as required by the COMAH/CIMAH enforcement. The public
understanding of the risk plays an important role in determining
the success of the project, therefore the communication and edu-
cation of the public need to be increased. This obligation implies
additional cost to ensure smooth communication with the commu-
nity and to win the public trust; otherwise the owner risks facing
project cancellation with considerable financial losses. Through the
years, several projects were cancelled due to public objection. A
few notable examples are the halts of $8.9-billion plant expansion
project that produced paraxylene in Ningbo (Ramzy, 2012) and the
cancellation of MIC production in Bayer CropScience at West Vir-
ginia facility after $25 million was spent to enhance operational
safety (Bayer Crop Science, 2012). These cases prove that the per-
ception of public community on the operation of hazardous process
is very important and improper handling of this issue will cause
a large investment to go down the drain. As presented in Fig. 6,
the proposed framework can identify the requirement for trans-
portation, packaging, and labelling concurrently while assessing
the COMAH requirements. The additional requirements bring to
another potential problem. One study by Ta et al. (2010) is related
to the requirements of the procedures on packaging and labelling
Table 17
ISI ranks for ammonia storage tank.
Alternative 3 Options Base case Alternative 1 Alternative 2 Alternative 3
4.76 ISI 15 13 14 8
Rank 4 2 3 1
4.76
of hazardous materials. The study involved 150 participants from
the industry, 2% of whom had never read the information on the
labels and around 10% of whom had never read the information
provided by Safety Data Sheet (SDS). Whereas the intention of SDS
and labelling itself is to give information related to the substance
being handled, due to negligence, this procedure can be the cause
of the accident. This outcome proves that hazardous processes that
require more operating procedure will have a higher chance of
accidents, therefore an ISD that is able to eliminate or reduce the
additional requirement is considered the best choice.
According to the ISBI value presented in Table 16, the most ben-
eficial ISD alternative quantified by the reduction of severity and
management requirement is Alternative 3> Alternative 2> Alter-
native 1> base case. The proposed methodology was compared
with the previous developed index, ISI by Heikkilä (1999) and the
results obtained is presented in Table 17. Small difference was
found among the ranks obtained via the ISBI and the ISI in which
the second and the third place were swapped. According to the ISBI,
Alternative 3 and Alternative 2 will be more favourable compared to
the other two alternatives because the modifications led to signifi-
cant impact at which both severity and management requirement
can be reduced. Although both Alternative 1 and base case obtained
the same value, the former will be more favourable because it was
able to reduce the severity. Alternative 3 and Alternative 1, accord-
ing to ISI, will be more favourable compared to Alternative 2 and
the base case because the ISI ranks the alternatives according to
the potential of hazard. The rank obtained does not consider the
requirement for hazard management, therefore, the results give
more weightage to the atmospheric storage in Alternative 1 com-
pared to the diluted material in Alternative 2. Compared to the
ISBI, other than the potential of accidents, the requirement for haz-
ard management was counted for the ranking process. Due to the
nature of COMAH regulation that is the basis of the HMI devel-
opment, the hazard management requirement obtained is highly
dependent on the type of hazardous materials and inventory, there-
fore the results from the ISBI favour Alternative 2 compared to
Alternative 1. In this study, both indices, the ISBI and ISI, are in
agreement that Alternative 3 with the diluted ammonia with atmo-
spheric pressure operations being the best ISD option, and the base
case that has anhydrous ammonia with high pressure operation
being the most unfavourable option.
6. Conclusion
The present work proposes an integrated inherent safety-
regulatory framework that subsequently produces a new index,
ISBI, with the aims to identify the most beneficial ISD alterna-
tive that considers both hazard reduction and the reduction of
hazard management requirements. The integration of regulatory
elements into the ISBI development led to a comprehensive and
systematic approach that can provide the holistic benefits of ISD
implementation. Further, the proposed methodology can elimi-
nate error-related biases, which are derived from the subjective
judgement in evaluating the requirements to manage a hazardous
process. This is because the data required for ISD identification
during preliminary design stage using ISBI are available. More
importantly, the ISBI can help to enlighten and increase the aware-
ness among industrial practitioners of the positive impact of ISD
implementation on their businesses, particularly in terms of safety
 M. Zainal Abidin et al. / Process Safety and Environmental Protection 117 (2018) 454–472
and hazard management that will affect the lifetime of cost. In
this paper, the implementation of ISBI for ammonia storage shows
that the tool could identify the best ISD options during preliminary
design stage effectively.
Acknowledgement
The authors would like to thank Centre of Advanced Process
Safety (CAPS), Universiti Teknologi PETRONAS for providing the
facilities and the Ministry of Higher Education Malaysia for their
sponsorship under MyBrain 15 program.
References
Achoura, M.H., Harounb, A.E., Schultb, C.J., Gasemb, K.A.M., 2005. A new method
to assess the environmental risk of a chemical process. Chem. Eng. Process. 44,
901–909.
Ahmad, S.I., Hashim, H., Hassim, M.H., 2014. Numerical Descriptive Inherent Safety
Technique (NuDIST) for inherent safety assessment in petrochemical industry.
Process Saf. Environ. Prot. 92, 379–389, http://dx.doi.org/10.1016/j.psep.2014.
03.009.
Ahmad, S.I., Hashim, H., Hassim, M.H., 2016a. A graphical method for assessing inher-
ent safety during research and development phase of process design. J. Loss Prev.
Process Ind. 42, 59–69, http://dx.doi.org/10.1016/j.jlp.2015.09.018.
Ahmad, S.I., Hashim, H., Hassim, M.H., Muis, Z.A., 2016b. Inherent safety assessment
of biodiesel production: flammability parameter. Procedia Eng. 148, 1177–1183,
http://dx.doi.org/10.1016/j.proeng.2016.06.622.
Ahmad, S.I., Hashim, H., Hassim, M.H., 2017. Inherent safety assessment technique
for preliminary design stage. Chem. Eng. Trans. 56, 1345–1350.
Barua, S., Gao, X., Mannan, M.S., 2016. Comparison of prescriptive and performance-
based regulatory regimes in the U.S.A and the U.K. J. Loss Prev. Process Ind. 44,
764–769, http://dx.doi.org/10.1016/j.jlp.2016.10.009.
Bayer Crop Science, 2012. The Use and Storage of Methyl Isocyanate (MIC) at Bayer
CropScience. National Academies Press.
Bernechea, E.J., Arnaldos, J., 2014. Optimizing the design of storage facilities through
the application of ISD and QRA. Process Saf. Environ. Prot. 92, 598–615, http://
dx.doi.org/10.1016/j.psep.2013.06.002.
Bernechea, E.J., Viger, J.A., 2013. Design optimization of hazardous substance storage
facilities to minimize project risk. Saf. Sci. 51, 49–62, http://dx.doi.org/10.1016/
j.ssci.2012.06.007.
Britton, T., 2003. Lessons learned about preparing comah safety reports. In: Sympo-
sium Series No 149. IChemE, pp. 55–67.
CF Industries, 2015a. Aqua Ammonia 29% Safety Data Sheet.
CF Industries, 2015b. Aqua Ammonia 19% Safety Data Sheet.
CF Industries, 2016. Anhydrous Ammonia Safety Data Sheet.
Cave, S.R., Edwards, D.W., 1997. Chemical process route selection based on assess-
ment of inherent environmental hazard. Comput. Chem. Eng. 21, 965–970.
Center for Chemical Process Safety, 1999. Guidelines for Consequence Analysis of
Chemical Releases, CCPS Guidelines Series. John Wiley & Sons, New York.
Center for Chemical Process Safety, 2009. Inherently safer chemical processes: a life
cycle approach. In: ACCPS Concept Book, 2nd ed. John Wiley & Sons & AIChE,
New York.
Center for Chemical Process Safety, 2011. Guidelines for Hazard Evaluation Proce-
dures, 3rd ed. John Wiley & Sons, New York.
Chan, I., Alwi, S.R.W., Hassim, M.H., Manan, Z.A., Klemeš, J.J., 2014. Heat exchanger
network design considering inherent safety. Energy Procedia 61, 2469–2473,
http://dx.doi.org/10.1016/j.egypro.2014.12.025.
Chavan, S.D., 2012. Cryogenic storage of ammonia. Chem. Ind. Dig., 76–80.
Dow’s Fire and Explosion Index Hazard Classification Guide, 7th ed, 1994. American
Institute of Chemical Engineers, New York.
Edwards, W., Barron, F.H., 1994. SMARTS and SMARTER: improved simple methods
for multiattribute utility measurement. Organ. Behav. Hum. Decis. Process. 60,
306–325.
Eini, S., Abdolhamidzadeh, B., Reniers, G., Rashtchian, D., 2015. Optimization proce-
dure to select an inherently safer design scheme. Process Saf. Environ. Prot. 93,
89–98, http://dx.doi.org/10.1016/j.psep.2014.05.002.
Gómez, G., Rodríguez, S., Cadena, J., Muñoz, F., 2012. Kletz’s legacy for developing
countries: simple systems based on inherently safer design. J. Loss Prev. Process
Ind. 25, 843–847, http://dx.doi.org/10.1016/j.jlp.2012.04.003.
Gentile, M., Rogers, W.J., Mannan, M.S., 2003. Development of a fuzzy logic-based
inherent safety index. Process Saf. Environ. Prot. 81, 444–456, http://dx.doi.org/
10.1205/095758203770866610.
Gunasekera, M.Y., Edwards, D.W., 2003. Estimating the environmental impact of
catastrophic chemical releases to the atmosphere. Trans. IChemE 81, 463–474,
http://dx.doi.org/10.1205/095758203770866638.
Gunasekera, M.Y., Edwards, D.W., 2006. Chemical process route selection based upon
the potential toxic impact on the aquatic, terrestrial and atmospheric environ-
ments. J. Loss Prev. Process Ind. 19, 60–69.
Gupta, J.P., Hendershot, D.C., Mannan, M.S., 2003. The real cost of process safety—a
clear case for inherent safety. Process Saf. Environ. Prot. 81, 406–413, http://dx.
doi.org/10.1205/095758203770866575.
471
Hassim, M.H., Ali, M.W., 2009. Screening alternative chemical routes based on inher-
ent chemical process properties data: methyl methacrylate case study. J. – Inst.
Eng. Malaysia 70, 2–10.
Hassim, M.H., Edwards, D.W., 2006. Development of a methodology for assessing
inherent occupational health hazards. Process Saf. Environ. Prot. 84, 378–390.
Hassim, M.H., Hurme, M., 2010. Inherent occupational health assessment during
process research and development stage. J. Loss Prev. Process Ind. 23, 127–138.
Heikkilä, A., 1999. Inherent Safety in Process Plant Design: An Index Based Approach.
Helsinki University of Technology.
Industry Code of Practice on Chemicals Classification and Hazard Communica-
tion, 2014, Department of Occupational Safety and Health, Ministry of Human
Resource Malaysia.
Jha, V., Pasha, M., Zaini, D., 2016. Enhanced inherent safety intervention framework.
Procedia Eng. 148, 1051–1057.
Khan, F.I., Abbasi, S.A., 1997. Accident hazard index: a multi-attribute method for
process industry hazard rating. Process Saf. Environ. Prot. 75, 217–224, http://
dx.doi.org/10.1205/095758297529093.
Khan, F.I., Abbasi, S.A., 1998. Multivariate hazard identification and ranking system.
Process Saf. Prog. 17, 157–170.
Khan, F.I., Amyotte, P.R., 2002. Inherent safety in offshore oil and gas activities: a
review of the present status and future directions. J. Loss Prev. Process Ind. 15,
279–289, http://dx.doi.org/10.1016/S0950-4230(02)00009-8.
Khan, F.I., Amyotte, P.R., 2003. How to make inherent safety practice a reality. Can.
J. Chem. Eng. 81, 2–16, http://dx.doi.org/10.1002/cjce.5450810101.
Khan, F.I., Amyotte, P.R., 2004. Integrated inherent safety index (I2SI): A tool for
inherent safety evaluation. Process Saf. Prog. 23, 136–148, http://dx.doi.org/10.
1002/prs.10015.
Khan, F.I., Amyotte, P.R., 2005. I2SI: A comprehensive quantitative tool for inherent
safety and cost evaluation. J. Loss Prev. Process Ind. 18, 310–326, http://dx.doi.
org/10.1016/j.jlp.2005.06.022.
Khan, F.I., Husain, T., Abbasi, S.A., 2001. Safety weighted hazard index (SWeHI): a
new, user-friendly tool for swift yet comprehensive hazard identification and
safety evaluation in chemical process industrie. Process Saf. Environ. Prot. 79,
65–80, http://dx.doi.org/10.1205/09575820151095157.
Kidam, K., Sahak, H.A., Hassim, M.H., Shahlan, S.S., Hurme, M., 2016. Inherently
safer design review and their timing during chemical process development and
design. J. Loss Prev. Process Ind. 42, 47–58, http://dx.doi.org/10.1016/j.jlp.2015.
09.016.
Kletz, T., Amyotte, P., 2010. Process Plants A Handbook for Inherently Safer Design,
2nd ed. Taylor & Francis Group.
Koller, G., Fischer, U., Hungerbtihler, K., 1999. Assessment of environment-: health-
and safety aspects of fine chemical processes during early design phases. Com-
put. Chem. Eng. Suppl., 63–66.
Koller, G., Fischer, U., Konrad, H., 2000. Assessing safety, health, and environmental
impact early during process development. Ind. Eng. Chem. Res. 39, 960–972.
Landucci, G., Tugnoli, A., Cozzani, V., 2008. Inherent safety key performance indi-
cators for hydrogen storage systems. J. Hazard. Mater. 159, 554–566, http://dx.
doi.org/10.1016/j.jhazmat.2008.02.080.
Lawrence, D., 1996. Quantifying Inherent Safety of Chemical Process Routes. Lough-
borough University.
Lele, G.S., 2008. Ammonia storage: selection & safety issues. Chem. Ind. Dig., 85–89.
Leong, C.T., Shariff, A.M., 2009. Process route index (PRI) to assess level of explosive-
ness for inherent safety quantification. J. Loss Prev. Process Ind. 22, 216–221,
http://dx.doi.org/10.1016/j.jlp.2008.12.008.
Li, C., Zhang, X., Zhang, S., Suzuki, K., 2009. Environmentally conscious design of
chemical processes and products: multi-optimization method. Chem. Eng. Res.
Des. 87, 233–243, http://dx.doi.org/10.1016/j.cherd.2008.07.017.
Liew, W.H., Hassim, M.H., Ng, D.K.S., 2014. Sustainability assessment for biodiesel
production via fuzzy optimisation during research and development (R&D)
stage. Clean Techn Env. Policy 16, 1431–1444, http://dx.doi.org/10.1007/
s10098-014-0763-2.
Liew, W.H., Hassim, M.H., Ng, D.K.S., Chemmangattuvalappil, N., 2015. Systematic
framework for sustainability assessment of biodiesel production: preliminary
engineering stage. Ind. Eng. Chem. Res. 54, 12615–12629.
Long, B., Gardner, B., 2004. Ammonia Storage-a Special Case. In: Guide to Storage
Tanks and Equipment, European Guide Series (REP). Wiley.
Mansfield, D., Clark, J., Malmén, Y., Schabel, J., Rogers, R., Suokas, E., Turney, R., Ellis,
G., van Steen, J., Verwoerd, M., 2001. The INSET Toolkit INherent SHE Evaluation
Tool.
Medina, H., Arnaldos, J., Casal, J., 2009. Process design optimization and risk analysis.
J. Loss Prev. Process Ind. 22, 566–573, http://dx.doi.org/10.1016/j.jlp.2009.04.
007.
Moradi, F., 2010. An Integrated Framework for Risk and Safety Evaluation and Pro-
cess Design. Murdoch University.
Mustapha, S., Zain, I.M., 2003. Safety report: maintaining standard and reviewing.
In: Symposium Series No 149. IChemE, pp. 87–101.
Narayanan, D., Zhang, Y., Mannan, M.S., 2007. Engineering for sustainable develop-
ment (ESD) in bio-diesel production. Process Saf. Environ. Prot. 85, 349–359.
Palaniappan, C., Srinivasan, R., Tan, R., 2002a. Expert system for the design of
inherently safer processes. 1. Route selection stage. Ind. Eng. Chem. Res. 41,
6698–6710, http://dx.doi.org/10.1021/ie020175c.
Palaniappan, C., Srinivasan, R., Tan, R.B., 2002b. Expert system for the design of inher-
ently safer processes. 2. Flowsheet development stage. Ind. Eng. Chem. Res. 41,
6711–6722, http://dx.doi.org/10.1021/ie0201765.
472 M. Zainal Abidin et al. / Process Safety and Environmental Protection 117 (2018) 454–472
Palaniappan, C., Srinivasan, R., Tan, R., 2004. Selection of inherently safer process
routes: a case study. Chem. Eng. Process. Process Intensif. 43, 641–647, http://
dx.doi.org/10.1016/j.cep.2002.12.001.
Patel, S.J., Ng, D., Mannan, M.S., 2010. Inherently safer design of solvent processes at
the conceptual stage: practical application for substitution. J. Loss Prev. Process
Ind. 23, 483–491, http://dx.doi.org/10.1016/j.jlp.2010.03.002.
Ramzy, A., 2012. After Protests, Chinese City Halts Chemical Plant Expan-
sion [WWW Document]. Time. URL http://world.time.com/2012/10/30/after-
protests-chinese-city-halts-chemcial-plant-expansion/.(Accessed 29 December
16).
Rathnayaka, S., Khan, F., Amyotte, P., 2014. Risk-based process plant design con-
sidering inherent safety. Saf. Sci. 70, 438–464, http://dx.doi.org/10.1016/j.ssci.
2014.06.004.
Recommendations on the transport of dangerous goods, Volume 1. ed, 2011.United
Nations.
Regulation (EC) No 1272/2008 on classification, labeling and packaging of substances
and mixtures, 2008.
Roy, P.K., Bhatt, A., Kumar, B., Kaur, S., Rajagopal, C., 2011. Consequence and risk
assessment: case study of an ammonia storage facility. Arch. Environ. Sci. 5,
25–36.
Rusli, R., Shariff, A.M., Khan, F.I., 2013. Evaluating hazard conflicts using inherently
safer design concept. Saf. Sci. 53, 61–72, http://dx.doi.org/10.1016/j.ssci.2012.
09.002.
Salzano, E., Di Benedetto, A., 2012. Inherently safe indexes for explosions in the
process industry. Rev. Ing. 37, 73–78.
Shariff, A.M., Leong, C.T., 2009. Inherent risk assessment—a new concept to evaluate
risk in preliminary design stage. Process Saf. Environ. Prot. 87, 371–376, http://
dx.doi.org/10.1016/j.psep.2009.08.004.
Shariff, A.M., Wahab, N.A., 2013. Inherent fire consequence estimation tool (IFCET)
for preliminary design of process plant. Fire Saf. J. 59, 47–54, http://dx.doi.org/
10.1016/j.firesaf.2013.03.015.
Shariff, A.M., Zaini, D., 2010. Toxic release consequence analysis tool (TORCAT) for
inherently safer design plant. J. Hazard. Mater. 182, 394–402, http://dx.doi.org/
10.1016/j.jhazmat.2010.06.046.
Shariff, A.M., Zaini, D., 2013. Inherent risk assessment methodology in preliminary
design stage: a case study for toxic release. J. Loss Prev. Process Ind. 26, 605–613,
http://dx.doi.org/10.1016/j.jlp.2012.12.003.
Shariff, A.M., Rusli, R., Leong, C.T., Radhakrishnan, V.R., Buang, A., 2006. Inher-
ent safety tool for explosion consequences study. J. Loss Prev. Process Ind. 19,
409–418, http://dx.doi.org/10.1016/j.jlp.2005.10.008.
Shariff, A.M., Leong, C.T., Zaini, D., 2012. Using process stream index (PSI) to assess
inherent safety level during preliminary design stage. Saf. Sci. 50, 1098–1103,
http://dx.doi.org/10.1016/j.ssci.2011.11.015.
Srinivasan, R., Natarajan, S., 2012. Developments in inherent safety: a review of the
progress during 2001–2011 and opportunities ahead. Process Saf. Environ. Prot.
90, 389–403, http://dx.doi.org/10.1016/j.psep.2012.06.001.
Srinivasan, R., Nhan, N.T., 2008. A statistical approach for evaluating inherent
benign-ness of chemical process routes in early design stages. Process Saf. Env-
iron. Prot. 86, 163–174.
Ta, G.C., Mokhtar, M., Mohd Mokhtar, A., Ismail, A., Abu Bin, M.F., 2010. Analysis of
the comprehensibility of chemical hazard communication tools at the industrial
workplace. Ind. Health 48, 835–844.
Tugnoli, A., Cozzani, V., Landucci, G., 2007. A consequence based approach to the
quantitative assessment of inherent safety. AIChE J. 53, 3171–3182, http://dx.
doi.org/10.1002/aic.11315.
Tugnoli, A., Landucci, G., Cozzani, V., 2009. Key performance indicators for inherent
safety: application to the hydrogen supply chain. Process Saf. Prog. 28, 156–170,
http://dx.doi.org/10.1002/prs.10303.
Tugnoli, A., Landucci, G., Salzano, E., Cozzani, V., 2012. Supporting the selection of
process and plant design options by Inherent Safety KPIs. J. Loss Prev. Process
Ind. 25, 830–842, http://dx.doi.org/10.1016/j.jlp.2012.03.008.
UN Industrial Development Organization, 1998. Transportation and Storage of
Ammonia. In: Fertilizer Manual. Springer, pp. 196–202.
Wilday, J., Fraser, S., Bailey, C., Stocks-Greaves, M., Ridgway, P., Ashcroft, S., 2012.
The forthcoming SEVESO III directive: alignment with GHS classifications and
data issues for acute toxicity. In: Hazards XXIII, SYMPOSIUM SERIES NO. 158.,
pp. 379–386.
Wiley - VCH (Ed.), 2003. Ammonia, in: Ullman n’s Encyclopedia of Industrial Chem-
istry, Ullmann’s Encyclopedia of Industrial Chemistry. John Wiley & Sons, pp.
671– 681.
Young, D., Scharp, R., Cabezas, H., 2000. The waste reduction (WAR) algorithm: envi-
ronmental impacts, energy consumption, and engineering economics. Waste
Manag. 20, 605–615.
Zainal Abidin, M., Rusli, R., Buang, A., Mohd Shariff, A., Khan, F.I., 2016a. Resolving
inherent safety conflict using quantitative and qualitative technique. J. Loss Prev.
Process Ind. 44, 95–111, http://dx.doi.org/10.1016/j.jlp.2016.08.018.
Zainal Abidin, M., Rusli, R., Mohd Shariff, A., Khan, F.I., 2016b. Three-stage ISD matrix
(TIM) tool to review the impact of inherently safer design implementation. Pro-
cess Saf. Environ. Prot. 99, 30–42, http://dx.doi.org/10.1016/j.psep.2015.10.006.
Zaini, D., Pasha, M., Kaura, S., 2016. Inherently safe heat exchanger network design
by consequence based analysis. Procedia Eng. 148, 908–915, http://dx.doi.org/
10.1016/j.proeng.2016.06.500.