Safety risk assessment using analytic hierarchy process (AHP) during planning and budgeting of construction projects

Safety risk assessment using analytic hierarchy process (AHP) during
planning and budgeting of construction projects
Saman Aminbakhsh, Murat Gunduz ⁎, Rifat Sonmez
Department of Civil Engineering, Middle East Technical University, 06800 Ankara, Turkey
a b s t r a c ta r t i c l e i n f o
Article history:
Received 5 October 2012
Received in revised form 1 May 2013
Accepted 22 May 2013
Available online 6 June 2013
Keywords:
Occupational health and safety
Analytic hierarchy process
Risk assessment
Finance
Prioritization
Introduction: The inherent and unique risks on construction projects quite often present key challenges to
contractors. Health and safety risks are among the most significant risks in construction projects since the
construction industry is characterized by a relatively high injury and death rate compared to other industries.
In construction project management, safety risk assessment is an important step toward identifying potential
hazards and evaluating the risks associated with the hazards. Adequate prioritization of safety risks during
risk assessment is crucial for planning, budgeting, and management of safety related risks. Method: In this
paper, a safety risk assessment framework is presented based on the theory of cost of safety (COS) model
and the analytic hierarchy process (AHP). The main contribution of the proposed framework is that it
presents a robust method for prioritization of safety risks in construction projects to create a rational budget
and to set realistic goals without compromising safety. The impact to the industry: The framework provides a
decision tool for the decision makers to determine the adequate accident/injury prevention investments
while considering the funding limits. The proposed safety risk framework is illustrated using a real-life con-
struction project and the advantages and limitations of the framework are discussed.
© 2013 National Safety Council and Elsevier Ltd. All rights reserved.
1. Introduction
Risk management has been studied extensively in construction
project management in recent years due to its practical importance. In
today's world, where changes rapidly take place with underlying immi-
nent risks, the prerequisite for survival is to have profound knowledge
of the environment and to be capable of making flawless decisions.
The construction industry is recognized to be highly prone to risks
and is characterized to be very complex, dynamic, and unique where
uncertainties arise from various sources. Along with being highly
risky, construction projects engage firm financing to bear the direct
and the indirect costs. These costs include charges associated with var-
ious aspects of construction processes, such as managements of safety
and risk.
Numerous methods have been proposed to assist contractors and
project managers in selection and management of projects. Applica-
tion of these methods enables project managers to avoid potential
problems. Occupational health and safety problems (e.g., falling of
materials or people from heights, stepping on objects, injuries by
hand tools, explosions, electrical accidents) have been one of the
major challenges in the construction industry. Within this context,
safety risk assessment has become a very important topic in the
construction project management in recent years.
A novel framework is proposed in this paper to facilitate the safety
risk assessment process. The proposed method is intended to aid
decision makers during evaluation of the impacts of different safety
risk items. In the framework, the safety risk items are prioritized by
the experts by means of the analytic hierarchy process (AHP). This
framework assists project managers in planning, budgeting, and man-
agement of safety related risks. The proposed method is illustrated
using a case project.
The remainder of the paper is organized as follows: Section 2 is
devoted to the literature review and cost of safety (COS) model, and
background on AHP is presented in Section 3. The proposed frame-
work is described in Section 4 and a case study is presented in
Section 5. Finally, concluding remarks are made in Section 6.
2. Literature review
Numerous techniques have been proposed for risk analysis and as-
sessment ranging from simple classical methods to fuzzy approaches.
Howard and Matheson (1981) used influence diagram method for
risk analysis. Sensitivity analysis technique was adopted by Norris
(1992) to forecast the effect of variation of a single independent var-
iable on the dependent variable. Monte Carlo simulation (PMBoK,
2004) was also suggested to assess concurrent change in multiple
independent variables. Decision analysis bearing decision matrices
and decision trees (PMBoK, 2004) along with multicriteria decision
making techniques such as the simple multiattribute rating technique
(Von Winterfeldt & Edwards, 1986) and the analytic hierarchy
Journal of Safety Research 46 (2013) 99–105
⁎ Corresponding author.
E-mail addresses: [e-mail] (S. Aminbakhsh), [e-mail]
(M. Gunduz), [e-mail] (R. Sonmez).
0022-4375/$ – see front matter © 2013 National Safety Council and Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.jsr.2013.05.003
Contents lists available at SciVerse ScienceDirect
Journal of Safety Research
j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / j s r

process (AHP; Saaty, 1990) were used to facilitate making decisions
under risky or uncertain situations.
Occupational health and safety (OHS) risk is defined as the signif-
icance of a hazard in terms of the probability and severity of an injury/
illness that varies from one industry to another. The term ‘risk’ was de-
fined by OHSAS as a combination of the probability and the conse-
quences of the occurrence of a specified dangerous event (OHSAS
Project Group, 2007). Woodruff (2005) considered risk as the chance
that someone or something that is valuated will be adversely affected
by the hazard. The construction industry is characterized by relatively
high fatal and non-fatal injuries compared to other industries. Pinto,
Nunes, and Ribeiro (2011) indicated that occupational injuries and
illnesses not only affect safety and health, but also impact economics
because of high costs related with work injuries. When considered
within the context of construction projects, the injuries account for
7.9–15% of the cost of non-residential projects (Everett & Frank,
1996). Hence, proper implementation of safety risk assessment tech-
niques is important to promote project success.
Historical data were used in numerous studies to develop OHS risk
models. Gürcanli and Müngen (2009) and Pinto, Nunes, and Ribeiro
(2010) proposed qualitative models to conduct health and safety risk
assessment based on fuzzy logic approach. Gürcanli and Müngen
(2009) dealt with uncertain and insufficient data through adoption of
the subjective judgments of experts, and using the existing safety
level of a construction project.
A substantial part of safety literature focused on identifying and
describing various methods for improving occupational safety on
site. Hallowell (2008) developed and validated a formal method to
evaluate construction safety risk. Hallowell's (2008) method assumed
that every construction activity is associated with specific safety risks
and that each safety program is able to mitigate a portion of these
risks. Fung, Tam, Lo, and Lu (2010) attempted to improve the system-
atic risk assessment approach by introducing procedures to check the
reliability of the decisions made. The severities of the accidents were
decided using three parameters: man-days lost, fracture or amputa-
tion, and compensation. In measurement of safety risks, Mitropoulos
and Namboodiri (2011) introduced an observational method provid-
ing an objective assessment of an activity's task demand based on
observable risk factors and production variables. Mitropoulos and
Namboodiri (2011) also reflected on the difficulties to perform activ-
ities safely. Hallowell and Gambatese (2009) performed a study to
facilitate determination of the relative effectiveness of safety program
elements through quantification of their individual ability to mitigate
construction safety and health risks using Delphi process. Krallis and
Csontos (2007) examined the internal and external factors that
shape individuals' perception and explored the relevance between
risk perception and safe behavior and suggested possible initiatives
for the organizations. Although the majority of the mentioned risk
management techniques can be applied to any kind of risk assess-
ment process, very few of these studies focused on quantitative
assessment of safety risks in the construction industry.
2.1. Cost of safety (COS) model
Costs of construction injuries can have a substantial impact on the
financial success of construction organizations and may increase the
overall construction costs up to 15% (Everett & Frank, 1996). Hence,
investing in accident/injury prevention is important not only for
OHS management but also for decreasing costs of construction pro-
jects. However, there is a point where additional investment yields
diminishing returns and the return on investment becomes negative;
thus, it is crucial for construction organizations to objectively evalu-
ate the cost–benefit of investments in accident/injury prevention
through a robust process.
The cost of safety (COS) model was introduced by Chalos (1992)
to conceptually describe the cost–benefit analysis of accident/injury
prevention. The COS model is illustrated in Fig. 1. According to the
COS model, there is a theoretical equilibrium point at which the
total costs of prevention and detection are equal to the total costs of
injuries, and this point reflects the optimum investment. COS model
also supports the presumption that some level of safety risk must
be considered as acceptable to maintain an organization's financial
stability. Manuele and Main (2002) support this assumption and
state that there exists some level of inherent risk in most of the
work processes and that the costs of mitigating such risk can be over-
whelming. Subsequently, in practice, beyond a certain level of very
high safety, the goal of zero accidents imposes significant investments
in the accident/injury appraisal and prevention; that is, appraisal and
prevention costs must be substantially increased to either achieve
zero accidents or to get close to zero accidents. Consequently, the
COS model aims to provide a structure for the managers to analyze
costs, prepare budgets, and to set realistic goals.
3. The analytic hierarchy process (AHP)
Analytic hierarchy process (AHP) is a structured multi-attribute
decision method (Saaty, 1990). The main advantage of AHP is its
capability to check and reduce the inconsistency of expert judgments.
While reducing bias in the decision making process, this method pro-
vides group decision making through consensus using the geometric
mean of the individual judgments. AHP derives scales of values from
pairwise comparisons in conjunction with ratings and is suitable for
multi-objective, multi-criterion, and multi-actor decisions with any
number of alternatives. AHP involves assessing scales rather than
measures; hence, it is capable of modeling situations that lack mea-
sures (e.g., modeling risk and uncertainty). AHP is comprised of
three main principles: decomposition of the structure, comparison
of judgments, and hierarchical composition (or synthesis) of priori-
ties. Decomposing a decision problem into its constituent parts facil-
itates building hierarchies of criteria to determine the importance of
each criterion.
AHP was used for OHS initially by Freivalds (1987) and Henderson
and Dutta (1992). Padma and Balasubramanie (2009) used AHP to
develop a decision aid system in order to rank risk factors associated
with the occurrence of musculoskeletal problems in the shoulder and
neck. AHP was also adopted by Zhang, Zhan, and Tan (2009) to com-
pare risk factors associated with human error and with the causes of
accidents in the maritime transport sector. Kim, Lee, Park, and Lee
(2010) proposed a safety risk assessment methodology considering
the risk influence factors of construction sites using expert surveys
and the AHP. Badri, Nadeau, and Gbodossou (2012) proposed a proce-
dure for evaluation of the OHS risks based upon the multi-criteria
analysis techniques (e.g., AHP) and expert judgment.Safety BudgetLevel of SafetyPrevention CostsInjury CostsOptimal EquilibriumPoint
Fig. 1. COS model.
100 S. Aminbakhsh et al. / Journal of Safety Research 46 (2013) 99–105

3.1. The theoretical background of AHP
In AHP, the decision problem is usually divided into a hierarchy of
sub-problems, each of which can be analyzed independently. The
elements of the hierarchy can relate to any aspect of the decision
problem. Once the hierarchy is built, a numerical scale is assigned to
each pair of n alternatives (Ai, Aj) by the experts (see Table 1). Numer-
ical scales are attributed by making pairwise comparisons among the
alternatives with respect to their impact on an element placed in a su-
perior level in the hierarchy. The term aijk expresses the individual
preference of expert k regarding alternative A i compared to alterna-
tive A j.
Once the overall expert judgments are created and computed
using the geometrical mean (1), they are inserted into the compari-
son matrix D (2):
aij ¼
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
aij1Xaij2X...Xa ijn
n
q
ð1Þ
D ¼
a11 a12 ... a1n
a21 a22 ... a2n
... ... ... ...
an1 a n2 ... a nn
2
6
6
4
3
7
7
5: ð2Þ
Matrix D is a comparison matrix and has the following properties:
aij > 0; a ij ¼ 1
aji
; ∀ i where j ¼ 1; 2; ...; n: ð3Þ
Matrix D is considered as consistent when its elements meet con-
dition (4) while satisfying condition (3):
aij:a jk ¼ aik; ∀k where i; j ¼ 1; 2; ...; n: ð4Þ
The ordering of alternatives is taken as a result of the approxima-
tion of comparison matrix D using matrix P:
P ¼
p11 p12 ... p1n
p21 p22 ... p2n
... ... ... ...
pn1 pn2 ... pnn
2
6
6
4
3
7
7
5: ð5Þ
The elements of matrix P are consistent judgments presented in
the form of weight ratios among alternatives:
pij ¼ pi
pj
where i; j ¼ 1; 2; ...; n: ð6Þ
pi signifies the weights of the alternatives of the order vector p:
p ¼ p1; p2; ...; pnð ÞT : ð7Þ
The standardized order vector after the arithmetic normalization
is obtained as follows:
p ¼ p
1; p
2; ...; p
n
 T ð8Þ
where:
p
i ¼ pi
∑n
i¼0p i
: ð9Þ
Saaty (1990) proposed using the maximum eigenvalue method to
determine the judgment matrices as:
D:p ¼ λmax p ð10Þ
where; λmax is the maximum eigenvalue of matrix D. For a reliable
comparison, it is important to note that the inconsistency of the com-
parison matrix D must be less than 10%, that is, the number of times
condition (4) is not met must be below 10%. According to Saaty
(1990), the consistency of judgments can also be evaluated using
the Eq. (11):
Consistency ratio ¼ CR ¼ CI
RC ð11Þ
and,
Consistency index ¼ CI ¼ λmax−n
n−1 : ð12Þ
RC (random consistency index) can be acquired from Table 2.
Since the column(s) of any 1 × 1 or 2 × 2 comparison matrices are
dependent, RC is assumed to be 0. This means division by zero in
Eq. (11) and causes CR to tend toward infinity; that is, matrices of
sizes 1 and 2 are always consistent.
4. AHP in construction safety risk management
Project safety risk assessment is a fundamental component of the
project management since construction projects are prone to diverse
occupational health and safety problems such as falling of materials
or people from heights, electrical accidents, and so forth. However,
only a few studies have been carried out using the AHP method
along with the safety risk assessment techniques. In previous studies,
the most prevalent topic was the use of AHP throughout the decision
making process for evaluation of the alternatives (viz.: projects, con-
tractors, etc.). Badri et al. (2012) focused on assessment of magnitude
of the OHS risks for the sake of risk ranking; but, there was no explicit
consideration regarding AHP's application from the financial perspec-
tive. On the other hand, the cost of occupational injuries may account
for 15% of overall costs and adequate investments in accident/injury
prevention can affect the competitiveness of an organization. Al-
though investing in safety is critical to safety success, there is a
Table 1
AHP scale for combinations.
Numerical scale Definition Verbal explanation
1 Equal significance of the two elements Two elements contribute equally to the property
3 Low significance of one element compared to another Experience and personal assessments favor one element slightly over another
5 Strong significance of one element compared to another Experience and personal assessments favor one element strongly over another
7 Confirmed dominance of one element over another One element is strongly favored and its dominance is borne out in practice
9 Absolute dominance of one element over another The evidence favoring one element over another appears irrefutable
2, 4, 6, and 8 Intermediate values between two neighboring levels The assessment falls between two levels
Reciprocals (1/x) A value attributed when activity i is compared to activity j
becomes the reciprocal when j is compared to i
Table 2
Random consistency (RC) index [n = size of the reciprocal matrix].
n 1 2 3 4 5 6 7 8 9 10
RC 0 0 0.58 0.9 1.12 1.24 1.32 1.41 1.45 1.49
101S. Aminbakhsh et al. / Journal of Safety Research 46 (2013) 99–105