Safety Science 121 (2020) 15–29

Contents lists available at ScienceDirect

Safety Science
journal homepage: www.elsevier.com/locate/safety

A risk assessment approach for enhancing construction safety performance T

a,⁎ a b
Muizz O. Sanni-Anibire , Abubakar S. Mahmoud , Mohammad A. Hassanain ,
Babatunde A. Salamic
a
Safety Technology Program, Dammam Community College, King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia
b
Architectural Engineering Department, King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia
c
Project Management Department, King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia

A R T I C LE I N FO A B S T R A C T

Keywords: The construction industry is one of the most dangerous industries worldwide due to deadly fatalities and ac-
Construction cidents recorded yearly. Though many countries have established and implemented safety programs, the si-
Safety performance tuation does not seem to have been mitigated. This study aims at developing a risk assessment approach that can
Risk assessment be used to enhance the safety performance of construction projects. The study has employed pair-wise com-
Six sigma
parisons and weighting-by-ranking surveys to establish risk scores and weights for the various construction
accidents, and their potential causes. Data has been collected from safety professionals in 15 large construction
sites across the Eastern Province of Saudi Arabia. The study revealed that the type of accident with the highest
risk score is “falling objects”, while the most significant cause is excessive winds on the project site. The de-
veloped approach was applied on an on-going car park construction project. Results showed that slips, trips and
falls had the best safety performance. Furthermore, based on six sigma evaluation, the average project safety
performance was at 2.33-sigma which implies that 228,739 accidents may occur in every million opportunities.
The paper also provided recommendations to improve the safety performance of the case study.

1. Introduction The current mantra of the industry is to achieve zero accidents and
ultimate safety performance (Zwetsloot et al., 2013, 2017). Thus, safety
The construction industry continues to record high rates of acci- systems have been developed to prevent accidents and increase the
dents and fatalities making it one of the most hazardous industries in overall safety performance of construction projects (Wilson and Koehn,
the world (Waehrer et al., 2007; Pinto et al., 2011; Mosly, 2015). 2000; Awwad et al., 2016). Numerous research efforts have been made
Construction related injuries and accidents in the U.S. are 50% higher to study the safety performance in the construction industry globally
than those in any other industry (Huang and Hinze, 2003); in Japan, (Jannadi and Assaf, 1998; Sawacha et al., 1999; Siu et al., 2004; Tam
they are 40% of the total accidents, 50% in Ireland and 25% in the et al., 2004; Choudhry et al., 2007; Mosly, 2015; Alarcón et al., 2016;
United Kingdom (Agwu and Olele, 2014). More severe situations are Okoye, 2016; Gunduz et al., 2016, 2018; Misiurek and Misiurek, 2017;
recorded in developing countries. Occupational accident and fatality Dong et al., 2018). Despite the amount of substantial research and
rates in the Middle East have been reported to be 18.6 per 100,000 improvements that have been made through the years, accident rates in
workers, compared to 4.2 per 100,000 workers in established econo- construction are still high (Mosly, 2015). Practical tools directed to-
mies (Hämäläinen et al., 2006). Previous research has shown that the wards enhancing safety performance in the construction industry are
low level safety performance in developing countries is due to the ab- still lacking. Though, some researchers (Liu and Tsai, 2012; Patel and
sence of stringent safety and construction laws (Ofori, 2000), high rates Jha, 2016; Gunduz et al., 2016, 2018; Amiri et al., 2017) have provided
of un-skilled foreign workers and unemployment (Awwad et al., 2016). safety performance tools. However, these were either too complex to be
Furthermore, the cost of accidents ranges in billions of dollars. This is transferred to practical use, or did not provide a clear strategy for safety
aside from the loss of life, property damage, and permanent disabilities performance enhancement. What is probably lacking in the industry is a
of victims (Liu and Tsai, 2012). The lagging safety performance has led simple, comprehensive and yet effective approach. This can be achieved
some countries to invest significantly in safety standards, such as OSHA through a risk assessment approach that relies on a systematic hazard
in the US, aiming at zero-injury policies. analysis and quantification to determine the safety performance.

⁎
Corresponding author.
E-mail address: [email protected] (M.O. Sanni-Anibire).

https://doi.org/10.1016/j.ssci.2019.08.044
Received 13 September 2017; Received in revised form 19 May 2019; Accepted 31 August 2019
0925-7535/ © 2019 Elsevier Ltd. All rights reserved.
M.O. Sanni-Anibire, et al. Safety Science 121 (2020) 15–29

“Hazard analysis is the most important safety process in that, if that 3. Construction safety performance
fails, all other processes are likely to be ineffective” (Manuele, 2013).
In light of the foregoing, this study seeks to develop a risk assess- Safety performance as a concept revolves around maintaining safe
ment approach to enhance the safety performance of construction workplaces. This could be attained through organizational safety cul-
projects. The study provides significant information on the risk per- ture and workers safety behaviors, or the ability to minimize the
ception of accidents using pair-wise comparison surveys. It also pro- quantity of accidents and occupational injuries in the work sites (Al-
vides a comprehensive list of accident causation factors in construction Bsheish, 2017). It has been found in the literature that enhancing safety
sites, while providing their weights using weighting-by-ranking surveys performance can be either proactive or reactive. Proactive studies focus
administered to construction professionals across the Eastern Province on evaluating the safety climate of a region, safety culture, hazard
of Saudi Arabia. As a result, the data obtained was condensed into a risk identification and observation. Reactive studies focus on injury fre-
assessment checklist to form a risk assessment approach which also quency rates and compensation costs (Alasamri et al., 2012). Reactive
imbibes the six sigma evaluation scale developed by Motorola in 1987 studies seek to measure the safety performance based on historical data,
for quality improvement. Six sigma evaluation scale proposes a 3.4 and do not measure the status quo. Also, they highlight the problems,
defects per one million to achieve ultimate performance (Dorrington but do not proffer practical solutions. Alasamri et al. (2012) presented a
and Carla, 2007). The proposed approach has been implemented on a comparison of the safety performance of the construction industry in
case study of a car park construction project. Some limitations of the various countries. Saudi Arabia had the highest rate of major and fatal
study have been highlighted. Finally, the paper presents a discussion injuries, while the UK had the lowest rate of major and fatal injuries.
and conclusion section to summarize the themes and main contribu- Alarcón et al. (2016) studied a sample of 1180 construction firms and
tions of the study. 221 safety practices applied in these firms over a period of four years in
Chile. The study suggested that practices related to safety incentives
and rewards were the most effective, while practices related to acci-
2. Causes of accidents in construction sites dents and incidents investigation had a slightly negative impact on the
accident rate. Also, Misiurek and Misiurek (2017) presents an adoption
The main theme of construction safety is the prevention of acci- of Training Within Industry (TWI) program-derived from manu-
dents. Several studies have been carried out to provide insight into the facturing processes-in the construction industry as a means to im-
types and causes of accidents inherent in the construction industry. proving occupational safety. The study relied on the statistics of fatal
Some of these studies have described the main causes of fatalities in accidents in construction and manufacturing industries over several
construction, while others have identified the major types of con- recent years in the UK. Choudhry et al. (2007) criticized reactive studies
struction accidents. Kartam and Bouz (1998) suggested that falling from highlighting them as being poor measurement tools in assessing safety
a height was the major cause of construction injuries and fatalities in performance, due to its reliance on historical events regardless of cur-
Kuwait, while Yilmaz (2015) highlights “being hit by dashing and rent safety activities.
flitting objects”, “being hit by objects”, “falling objects” and “being More proactive studies include such concepts as design for safety
stung by something” as the major construction accidents in Istanbul- and hazard risk assessments. Gambatese et al. (2008) suggested that
Turkey. Additionally, there are studies based on more generalist ap- 42% of construction accidents were due to design, and thus re-
proaches to identifying the causes of accidents in construction sites. commended that design professionals need to play a role in addressing
Abdelhamid and Everett (2000) presented an accident root causes tra- safety in their designs. Okoye (2016) showed that safety behavior in-
cing model adopted for the construction industry. The study did not tervention factors were positively correlated with safety performance.
provide information on the various accidents applicable to the con- Raheem and Issa (2016) proposed a safety framework composed of
struction industry and their potential causes. guidelines for a reciprocal safety implementation system with aspects
Studies that sought to dig deeper into the subject matter include such as regulatory enhancements and corporate safety culture im-
Haslam et al. (2005) which studied 100 individual construction acci- provements. Wu et al. (2008) investigated the potential correlation
dents. The study identified the key factors in accidents as arising from among safety leadership scale, safety climate scale and safety perfor-
the workers, or the work team (70% of accidents), workplace issues mance scale in four universities in central Taiwan. Fung et al. (2010)
(49%), shortcomings with equipment (including PPE) (56%), problems developed a Risk Assessment Model (RAM) for assessing risk levels at
with suitability and condition of materials (27%), and deficiencies with various project stages within various work trades. Liu and Tsai (2012)
risk management (84%). Thus, the study focused more on the ergo- presented a fuzzy risk assessment method based on two-stage quality
nomic perspective. Also, Al-Tabtabai (2002) presented 30 causes of function deployment (QFD) tables to represent the relationships among
accidents in the construction industry in Kuwait based on an analysis of construction items, hazard types and hazard causes. Amiri et al. (2017)
accident reports extracted from government agencies. The causes of presented a fuzzy probabilistic rule-based expert system tested on four
accidents studied were more of a generalist nature, and did not provide major construction case studies.
insight into the hazards that may lead to the occurrence of a real ac- In addition, Jannadi and Assaf (1998) employed a standard check-
cident. Hamid et al. (2008) presented a study on the causes of accidents list to assess the safety of construction job sites in Saudi Arabia. The
in the Malaysian construction industry. The study suggested that the results of the study showed that safety levels varied between large and
main cause of construction accidents are the workers’ negligence, small projects. Likewise, Mosly (2015) investigated the safety perfor-
failure of workers to obey work procedures, work at high elevation, mance of small- to medium-sized construction projects in the private
operating equipment without safety devices, poor site management, sector of Saudi Arabia, using a simple yes/no checklist. The checklist
harsh work operation, low knowledge and skill level of workers, failure covered safety aspects under five groups: general – construction site;
to use personal protective equipment and poor workers attitude about workers’ personal protective equipment (PPE); heights and fall pro-
safety. Though the study acknowledged that there are numerous causes tection; machinery; and excavation. These risk assessment checklists
of accidents in the construction industry, it lacked a systematic cate- are based on the assumption that all hazards have an equal level of
gorization of these causes. Therefore, these studies can best be de- contribution to the occurrence of an accident, which is obviously in-
scribed as partial and not comprehensive in addressing a wide range of correct.
potential construction accidents and their causes. They also do not
provide tangible contributions in terms of practical tools and frame- 4. Research methodology
works. However, they provide recommendations that could serve as the
basis for further research and development. The methodology undertaken in this study can be summarily

16
M.O. Sanni-Anibire, et al. Safety Science 121 (2020) 15–29

categorized in four stages as follows.

Drowning 4.1. Literature review

•
An extensive review of existing literature on the causes of accidents
excavation

and safety performance in the construction industry was conducted.

accidents
Trench/

This was to provide the contextual background for the study. The au-
thors identified and summarized the main types and causes of accidents
•

•
in construction projects, and methods employed in enhancing the safety
respiratory issues

performance of construction projects.

Asphyxiation/

4.2. Questionnaire design, administration and analysis

The common types of accidents (see Table 1) and their causes (listed
•

•
in Table 4) identified were further modified and improved through
brainstorming with experts in the field. Subsequently, the information
Structural
collapse

obtained was used as the basis for the development of three ques-
tionnaire surveys: two pair-wise comparison surveys, and one
•

•
weighting-by-ranking survey. These are explained as follows.
accidents
Scaffold

4.2.1. Pair-wise comparison surveys

Pairwise comparison is a scientific method used to judge elements in
•

pairs of two, and thus, determine the preferred element. In it, subjective
toxic/hazardous

feedback from experts are converted into numerical values in the form
of weights (Luther, 2003). The common accidents presented in Table 1
Exposure to

materials

were arranged vertically and horizontally as elements of a matrix, and

assigned alphabetical denotations (see Tables 2 and 3). The elements
were compared with each other to achieve their relative weights, where
•

the comparison is limited to two elements at a time. Preference is es-

tablished in terms of impact of the accidents (Table 2) and frequency of
Struck by

moving
objects
fixed/

occurrence of the accidents (Table 3). A numerical score of 1 to 4 is

assigned to the preferred element. The scale for the relative weights
•

•
•

used include the following: (1) ‘equal preference’, (2) ‘minor pre-
Fire outbreak/

ference’, (3) ‘strong preference’, (4) ‘very strong preference’. Twelve

explosions

professionals with adequate experience in construction project man-

agement, and occupational health and safety issues were consulted in
•

carrying out the pairwise comparisons. Once all the cells in the matrix
were filled, the number of points accrued to each element was summed
Falling
object

up and inserted into the respective raw score cell of the matrix. The raw
scores were further adjusted on a scale of 1–4 based on the value of the
•

•
•

highest raw score.

height
from
Falls

4.2.2. Weighting-by-ranking survey

•

The potential causes of the accidents identified in this study are

Power tool

presented in Table 4. The weighting-by-ranking technique has been

accidents

used to derive the weights of the identified causes. These causes have
been ranked by 100 professionals across 15 large construction sites in
•

the Eastern Province of Saudi Arabia. Questionnaire surveys were de-

Heat/Chemical

signed to contain the various accidents with their corresponding causes.

Electrical/

Respondents ranked the various causes based on numerical values of 1,

2, 3, 4, …, n. The mean value of the ranks given by the professionals
burns

was estimated. To determine the assigned weight, the reciprocal values

•

•
•

of the mean values were computed (i.e. for a given criteria with rank r,
Common accidents in construction projects.

Trips and

the weight is computed as 1/r). See Table 4.

Slips,

Falls
Types of accidents

4.3. Development of the risk assessment approach

accidents

Numerous methods of risk assessment have been developed over the

Crane

years. These methods have further been improved into more hybrid and
•

unique forms tailored to specific industries and applications. These

Yilmaz (2015)
Gürcanlı et al.

Ferrett (2015)
Gunduz et al.
Haslam et al.

extant methods can be summarily classified as qualitative, quantitative,

Liu and Tsai
Bouz (1998)

Hughes and
Kartam and

Al-Tabtabai
Reference/

Fung et al.
Ling et al.
Literature

or a hybrid of both (semi-quantitative) (Liu and Tsai, 2012; Hassanain

(2002)

(2009)

(2010)

(2005)

(2012)

(2015)

(2016)

et al., 2017). The risk approach presented in this study can be described
as a semi-quantitative risk model. It utilizes qualitative data based on
Table 1

expert experience and judgment and expresses same as numerical risk

No.

10
1

7
8

ratings. The proposed risk assessment approach is illustrated in Fig. 1.

17
 M.O. Sanni-Anibire, et al.

Table 2
Pair-wise comparisons of construction accidents – Impact.
A B C D E F G H I J K L M N
Crane Slips, Electrical/ Power tool Falls Falling Fire Struck by Exposure to Scaffold Structural Asphyxiation/ Trench/ Drowning
accidents Trips and Heat/Chemical accidents from objects outbreak/ fixed/ hazardous accidents collapse respiratory issues excavation
falls burns height explosions moving materials accidents
objects

A Crane accidents 1
B Slips, Trips and A4 1
falls
C Electrical/Heat/ A4 C3 1
Chemical burns
D Power tool A4 D3 D2 1
accidents
E Falls from height A3 E3 E2 E3 1

18
F Falling objects A3 F2 C/F1 F2 E2 1
G Fire outbreak/ G3 G3 G4 G4 G4 G4 1
explosions
H Struck by fixed/ A3 B/H1 C2 H2 E2 F/H1 G4 1
moving objects
I Exposure to A3 I2 I3 D/I1 E2 I2 G4 I2 1
hazardous
materials
J Scaffold accidents A4 J2 J2 J2 E2 F2 G3 H2 J2 1
K Structural collapse K2 K3 K4 K4 K4 K4 G/K1 K4 K4 K4 1
L Asphyxiation/ A2 L2 C2 D2 E2 F2 G4 H2 I/L1 J3 K4 1
respiratory issues
M Trench/excavation A3 M2 C/M1 M2 E2 F2 G4 M2 M2 J/M1 K3 M2 1
accidents
N Drowning A3 N3 C/N1 N3 N2 N2 G3 N2 N2 J/N1 K3 N3 M/N1 1
Raw score 37 2 11 9 21 13 46 9 12 14 45 4 14 21
Adjusted total 3.22 0.17 0.96 0.78 1.83 1.13 4.00 0.78 1.04 1.22 3.91 0.35 1.22 1.83
Safety Science 121 (2020) 15–29
 M.O. Sanni-Anibire, et al.

Table 3
Pair-wise comparisons of construction accidents – Frequency.
A B C D E F G H I J K L M N
Crane Slips, Electrical/ Power tool Falls Falling Fire Struck by Exposure to Scaffold Structural Asphyxiation/ Trench/ Drowning
accidents Trips and Heat/Chemical accidents from objects outbreak/ fixed/ hazardous accidents collapse respiratory issues excavation
falls burns height explosions moving materials accidents
objects

A Crane accidents 1
B Slips, Trips and B4 1
falls
C Electrical/Heat/ C3 B3 1
Chemical burns
D Power tool D2 B2 C/D1 1
accidents
E Falls from height E2 B2 C2 D2 1

19
F Falling objects F3 B/F1 F2 F2 F3 1
G Fire outbreak/ A/G 1 B4 C3 D3 E2 F4 1
explosions
H Struck by fixed/ H2 B2 H2 D/H1 H2 F2 H3 1
moving objects
I Exposure to A/I1 B3 C/I1 I2 I2 F2 I3 H2 1
hazardous
materials
J Scaffold accidents J2 B3 C2 D2 E/J1 F3 J2 H3 I3 1
K Structural collapse A/K1 B4 C3 D3 E2 F4 G/K1 H4 I3 J2 1
L Asphyxiation/ L2 B2 C/L1 D2 L2 F3 L2 H2 I/L1 L2 L2 1
respiratory issues
M Trench/excavation M2 B3 C2 D3 E/M1 F3 M2 H3 I3 J/M1 M2 L2 1
accidents
N Drowning A/N1 B4 C3 D3 E2 F4 G/N1 H4 I4 J2 K/N1 L3 M2 1
Raw score 5 38 22 23 11 37 4 29 24 11 4 18 11 4
Adjusted total 0.53 4.00 2.32 2.42 1.16 3.89 0.42 3.05 2.53 1.16 0.42 1.89 1.16 0.42
Safety Science 121 (2020) 15–29
 Table 4
Risk assessment checklist.
Construction accidents Risk Score Causes of accidents Hazard Risk Priority Adjusted RPN Risk rating Adjusted risk rating (Adjusted
(Impact*Frequency) criticality Number (RPN) (%) RPN * Risk rating)/4 (%)
(4) High (0) No
Risk Risk
M.O. Sanni-Anibire, et al.

Crane accidents 1.69 Lack of regular maintenance 0.27 0.45 100 2 50.00
Operator error/Lack of adequate training and skill 0.25 0.43 93.7 0 0.00
Lack of regular testing/inspections 0.25 0.42 93.47 2 46.74
Incompetent personnel/Lack of specialist 0.24 0.41 90.95 0 0.00
Contact with overhead power lines 0.24 0.41 90.29 0 0.00
Overloading/Dropped loads 0.24 0.41 89.21 0 0.00
Insecure counterweights/Overturning 0.23 0.38 84.74 1 21.19
Lack of barriers to prevent collision and pedestrian 0.23 0.38 84.74 3 63.56
interference
High winds 0.23 0.38 84.55 2 42.28
Foundation failure 0.23 0.38 84.55 2 42.28
Incorrect design/Improper erection/assembling/ 0.19 0.32 71.13 1 17.78
anchorage
Average adjusted risk rating 25.80
Slips, Trips and Falls 0.70 Poor housekeeping 0.33 0.23 100 3 75.00
Poor or uneven floor/pavement surfaces 0.28 0.20 84.94 0 0.00
Physical inactivity 0.27 0.19 80.38 0 0.00
Changes in levels 0.28 0.20 84.46 1 21.12
Unsuitable floor coverings 0.28 0.20 83.52 0 0.00
Loss of balance 0.26 0.19 79.10 0 0.00
Trailing Cables 0.25 0.18 75.70 2 37.85
Poor Lighting 0.25 0.18 74.94 0 0.00

20
Footwear, wet floors 0.24 0.17 72.57 0 0.00
Average adjusted risk rating 14.89
Electrical/Heat/Chemical 2.22 Lack/Improper use of PPE 0.34 0.76 100 4 100.00
burns Exposure to fire, steam, hot objects, or hot liquids 0.29 0.64 84.35 2 42.18
Extensive exposure to the sun 0.28 0.61 80.39 4 80.39
Contact with power lines/high voltage apparatus 0.27 0.61 79.51 0 0.00
and electrical sockets
Lightning 0.27 0.60 79.29 0 0.00
Use of incorrect of damaged tools for electrical 0.27 0.60 78.44 1 19.61
work
Unsafe welding practices 0.26 0.57 75.00 0 0.00
Contact with wet cement 0.25 0.56 73.82 1 18.46
Average adjusted risk rating 31.80
Power tool accidents 1.89 Inattention/Inexperience of users 0.32 0.60 100 2 50.00
Use of tools in wet and damp conditions 0.26 0.49 81.61 1 20.40
Dysfunctional machine guards 0.26 0.49 81.61 1 20.40
Unexpected interruption while working 0.25 0.48 80.15 3 60.11
Use of tools near combustible materials/electrical 0.25 0.47 77.59 0 0.00
cables
Lack of training on proper handling of tools 0.25 0.46 77.40 2 38.70
Overloading of tools 0.24 0.46 75.90 1 18.98
Faulty design/manufacture of power tool 0.23 0.44 72.92 0 0.00
Lack of maintenance of power tool 0.13 0.24 40.33 2 20.17
Average adjusted risk rating 25.42
(continued on next page)
Safety Science 121 (2020) 15–29
 Table 4 (continued)

Construction accidents Risk Score Causes of accidents Hazard Risk Priority Adjusted RPN Risk rating Adjusted risk rating (Adjusted
(Impact*Frequency) criticality Number (RPN) (%) RPN * Risk rating)/4 (%)
(4) High (0) No
Risk Risk
M.O. Sanni-Anibire, et al.

Falls from height 2.11 Workers’ distractions/Carelessness of the worker 0.35 0.74 100 2 50.00
Improper barriers/guardrails to open edges 0.29 0.61 82.08 3 61.56
Insufficient lighting/illumination 0.29 0.60 81.38 0 0.00
Insufficient/inoperative personal protective 0.28 0.58 78.24 3 58.68
equipment (PPE), personal fall arrest systems
(PFAS)
Slippery or sloped surfaces 0.26 0.55 74.54 1 18.64
Inadequate knowledge and experience 0.25 0.54 72.08 3 54.06
Fatigue due to long hours of work 0.25 0.52 70.12 2 35.06
Weak and fragile roofs 0.24 0.50 67.46 0 0.00
Struck by moving objects at height 0.20 0.43 57.72 3 43.29
Average adjusted risk rating 35.70
Falling objects 4.40 Excessive winds 0.34 1.51 100 1 25.00
Roof/floor openings 0.32 1.41 93.57 0 0.00
Unprotected roof edges, absence of covered 0.32 1.39 92.09 3 69.07
walkways or fan scaffolds
Lack of chutes and hoists for waste material 0.32 1.39 92.09 0 0.00
Insufficient PPE (head protection for workers and 0.28 1.24 82.20 0 0.00
visitors)
Improperly fixed fixtures and fittings 0.28 1.22 80.39 0 0.00
Objects falling from vehicles 0.27 1.21 79.73 0 0.00
Loose/moving/flying objects 0.27 1.19 78.65 1 19.66
Inadequate training and awareness 0.26 1.15 76.18 2 38.09

21
Average adjusted risk rating 18.98
Fire outbreak/explosions 1.68 Lack of proper storage of combustible/flammable 0.42 0.71 100 0 0.00
substances
Electric hazards 0.40 0.66 49.07 0 0.00
Lack of fire safety precautions and procedures 0.39 0.66 93.70 1 23.43
Hot work: Welding/Cutting 0.39 0.66 92.97 1 23.24
Lack of suitable and sufficient fire-fighting 0.37 0.62 88.15 1 22.04
equipment
Faulty equipment and machinery 0.37 0.62 87.82 1 21.96
Average adjusted risk rating 15.11
Struck by fixed/moving 2.39 Vehicle/equipment overturning 0.35 0.83 100 0 0.00
objects Machinery with unguarded moving parts 0.34 0.82 99.31 0 0.00
Communication problems between vehicle drivers 0.34 0.80 96.64 1 24.16
and employees or members of the public
Insufficient/Lack of warning signs and barricades 0.33 0.80 96.32 2 48.16
Operator/Drivers’ error 0.33 0.79 95.36 2 47.68
Insufficient lighting/illumination 0.31 0.74 89.16 0 0.00
Lack of PPE/Loose clothing 0.31 0.74 88.62 2 44.31
Average adjusted risk rating 23.47
Exposure to toxic/hazardous 2.64 Lack of proper storage 0.43 1.15 100 0 0.00
materials Inadequate training and awareness 0.41 1.09 95.04 2 47.52
Lack of suitable techniques and equipment to 0.40 1.06 92.37 0 0.00
monitor working environment
Insufficient/lack of PPE 0.40 1.07 93.12 2 46.56
Lack of first aid/emergency procedures 0.38 1.01 88.12 0 0.00
Average adjusted risk rating 18.82
(continued on next page)
Safety Science 121 (2020) 15–29
 Table 4 (continued)

Construction accidents Risk Score Causes of accidents Hazard Risk Priority Adjusted RPN Risk rating Adjusted risk rating (Adjusted
(Impact*Frequency) criticality Number (RPN) (%) RPN * Risk rating)/4 (%)
(4) High (0) No
Risk Risk
M.O. Sanni-Anibire, et al.

Scaffold accidents 1.41 Defective guardrails, toes boards, braces etc. 0.33 0.46 100 1 25.00
Incorrect design/Improper erection/assembling/ 0.32 0.45 98.71 0 0.00
anchorage of scaffold
Overloading 0.32 0.45 98.71 1 24.68
Lack of regular inspections 0.32 0.44 96.85 1 24.21
High winds 0.30 0.43 93.31 0 0.00
Uneven floor/pavement surfaces 0.30 0.43 93.31 1 23.33
Lack of barriers and warnings signs 0.30 0.42 91.92 2 45.96
Lack of safe access and egress of scaffold 0.29 0.41 89.50 0 0.00
Insufficient/inoperative personal protective 0.29 0.41 88.47 0 0.00
equipment (PPE), personal fall arrest systems
(PFAS)
Average adjusted risk rating 15.91
Structural collapse 1.65 Poor construction materials 0.44 0.73 100 0 0.00
Lack of proper design 0.44 0.72 99.56 0 0.00
Incompetent personnel/supervisor 0.42 0.69 94.58 0 0.00
Overloaded structures 0.42 0.69 94.58 0 0.00
Weak foundations 0.38 0.62 85.66 0 0.00
Average adjusted risk rating 0.00
Asphyxiation/respiratory 0.66 Insufficient/lack of PPE 0.44 0.29 100 2 50.00
issues Lack of training on confined spaces 0.41 0.27 94.63 1 23.66
Lack of measurement of site conditions 0.39 0.26 89.45 3 67.09
Presence of respirable dusts (asbestos, coal, silica, 0.39 0.26 88.76 3 66.57

22
some plastics and talc).
Lack of first aid/emergency procedures 0.36 0.23 81.49 0 0.00
Average adjusted risk rating 41.46
Trench/excavation accidents 1.41 Falling loads from above a trench or excavation 0.32 0.45 100 2 50.00
site
Cave ins due to lack of sidewall shoring 0.32 0.45 99.37 0 0.00
Absence of ramp/ladder for escape 0.29 0.40 89.68 0 0.00
Toxic fumes, lack of oxygen and water 0.28 0.39 87.19 2 43.60
accumulation
Proximity of heavy machinery or structures 0.27 0.38 83.47 2 41.74
Broken gas lines/gas leaks into trench 0.26 0.37 81.51 1 20.38
Contact with underground electrical lines 0.25 0.35 77.67 2 38.84
Proximity of waste/stored materials 0.25 0.35 76.72 2 38.36
Average adjusted risk rating 29.11
Drowning 0.77 Lack of adequate supervision and security 0.55 0.42 100 N/A 0.00
Lack of required PPEs (life jackets, buoyancy aids 0.53 0.41 96.83 N/A 0.00
and grab Lines)
Lack of barriers to edges of drains and ponds 0.51 0.39 93.37 N/A 0.00
Lack of rescue equipment, lifebelts, harnesses, 0.49 0.38 89.71 N/A 0.00
safety lines, and in some cases, a rescue boat
Average adjusted risk rating 0.00
Safety Science 121 (2020) 15–29
M.O. Sanni-Anibire, et al. Safety Science 121 (2020) 15–29

Fig. 1. The risk assessment approach.

This comprises of nine steps summarily described as follows: Table 5

Six sigma evaluation scale (Dorrington and Carla, 2007).
4.3.1. Step 1. Select a safety performance base line Rating Performance (%) Defects Per Million Opportunities (DPMO)
In its application, an organization will establish its safety perfor-
1 30.85 691,500
mance baseline at any of the sigma levels as presented in Table 5. Six
2 69.15 308,500
Sigma is both a methodology and a measurement procedure with the 3 93.32 66,800
sole aim of reducing defects. The philosophy of six sigma is to achieve 4 99.38 6,200
99.99966 performance (6-sigma level), where out of a million oppor- 5 99.977 230
6 99.99966 3.4
tunities, only 3.4 chances are acceptable for error (Pyzdek and Keller,
2014). Few attempts have been made to incorporate six sigma techni-
ques into risk assessment studies. Jolin (2009) discussed how a fi-
nancial audit company employed six sigma technique to create an

23
M.O. Sanni-Anibire, et al. Safety Science 121 (2020) 15–29

optimum risk assessment process. Yazdani et al. (2018) through their 4.3.6. Step 6. Risk rating of construction project site
study suggested the incorporating of musculoskeletal disorders hazard This step involves the deployment of the checklist presented in
identification and assessment into modern tools such as six sigma to Table 4 on an on-going construction site. A risk rating of the hazards on
improve MSD prevention. However, the six sigma technique has not the project site is carried out through the identification of any potential
been thoroughly employed in risk assessment studies, and especially in causes as listed in the table. The risk ratings assigned to each cause of
the construction industry, despite its potential to enhance safety per- accident is further adjusted by multiplying with the adjusted RPN, and
formance. then divided by the highest possible risk rating (i.e. 4). This presents the
final adjusted risk ratings for all potential causes of accidents in the
construction project site.
4.3.2. Step 2. Identify potential hazards (accidents and their causes)
This entails a comprehensive list of hazards associated with the
4.3.7. Step 7. Determine safety performance
scheduled construction tasks at the project site. The result of an ex-
To determine the safety performance based on the risk assessment,
tensive literature review as discussed in Section 4.1 is presented in
the average risk rating of the accidents is computed and subtracted
Table 1. This was further developed into a unique checklist presented in
from a 100 percent. This results into a yield value which can be in-
Table 4. Pinto et al. (2011) notes that checklists are one of the most
terpreted based on the six sigma evaluation scale shown in Table 5. This
common methods used in risk assessment in the construction industry.
table shows the level of performance and its corresponding Defects per
The range of accidents and their causes would vary according to the job
Million Opportunities (DPMO). In this study, accidents are considered
task at hand, previous accident records, and the experience of the risk
to be defects, and thus the table illustrates the amount of potential
assessment team.
accidents that could occur per million opportunities based on the level
of performance achieved.
4.3.3. Step 3. Estimate accident’s risk score (RS)
In practice, risk score is one of the most common techniques used in 4.3.8. Step 8 and 9. Risk acceptance and documentation of results
estimating the level of risk. Generally, risk assessments are based on These steps comprise the decision and documentation phases of the
two dimensional risk scoring systems including risk factors of severity risk assessment process. Firstly, an assessment of the level of perfor-
of occurrence (or impact) and the probability/frequency of occurrence. mance achieved is compared with the initial safety performance base
This is denoted mathematically as follows: line established in step 1. In case the performance level falls below this
baseline, necessary corrective actions will be made to serve as feedback,
Risk Score = impact * frequency and an iterative process will continue for continuous improvement as
illustrated in Fig. 1. The safety performance base line could also be
The impact and frequency of construction accidents can be de- adjusted as necessary.
termined through subjective assessments based on experts experience
and scrutiny. Hence, in this study, the risk perception of the identified 4.4. Case study application
construction accidents have been assessed based on expert opinion
through pair-wise comparison surveys as detailed in Section 4.2.1. The A multi-story car park building under construction in the Eastern
risk factors (impact and frequency of occurrence) thus obtained were Province of Saudi Arabia was used as a case study in this research. The
used in computing the risk score. workforce consists of 100 labor and 20 consultant/contractor personnel
including quality and safety professionals. The structure is a combina-
tion of cast-in place concrete (foundations, columns, walls, beams and
4.3.4. Step 4. Estimate hazard criticality
solid slabs) and precast hollow core slabs. The car park building is
Risk assessment studies usually include two risk factors as discussed
designed to accommodate 1000 cars with a provision of 8 slots for
in Section 4.3.3, however risk assessment procedures also incorporate a
people with disabilities. The gross project cost is 14,832,599 USD.
third risk factor such as “detection of failure” in failure mode and effect
Regular construction activities include excavation and backfilling,
analysis studies, and “prevention effectiveness” in hazard studies. Based
concrete and masonry works, installation of precast elements in the
on the premise that various causes could have varying levels of con-
substructure, superstructure, and elevator shafts; mechanical works
tribution to the occurrence of an actual accident, a “hazard criticality”
that includes fire protection, plumbing, drainage, HVAC; internal and
factor is suggested in this study. The “hazard criticality” allows for the
external fixtures such as panel boards, switches, lighting and signage;
estimation of a risk priority based on the presence or absence of the
finishing works such as painting, installation of stainless steel doors and
identified causes. Section 4.2.2 describes the method of determining the
windows, road works (asphalting) and other site development works.
weights for the “hazard criticality” of various causes of accidents pre-
The site also feature major construction equipment such as a tower
sented in Table 4.
crane (60 m height, 3900 kg), concrete pump. Only mobile scaffolds
have been used in interior spaces within the building, independent tied
4.3.5. Step 5. Determine Risk Priority Number (RPN) and put-log scaffolds have not been used. The risk assessment tool
The Risk Priority Number (RPN) is a combination of three or more presented in Table 4 was deployed by a joint team of a civil/structural
risk factors. Liu and Tsai (2012) suggested that researchers have been engineer, a safety manager and the site’s construction project manager.
critical of the RPN methodology. Though, the RPN has been promoted The average risk rating by the three professionals has been adopted.
by reliable authorities in the field (Manuele, 2008; Popov et al., 2016).
Furthermore, in this study, it is instrumental to establishing the varying 5. Results
levels of influence of each hazard cause to the occurrence of an acci-
dent. The RPN value in this study was computed according to the 5.1. Risk evaluation of construction accidents in Saudi Arabia
equation below:
To determine the risk score of the identified construction accidents,
Risk Priority Number = Risk Score (impact * frequency) * hazard two pair-wise comparison matrices were established as explained in
criticality Section 4.2.1. The impact of the accidents as well as the frequency of
occurrence were established through this exercise. These values are
The RPN value this obtained was further adjusted to a range of presented in Tables 2 and 3. The results in Table 2 shows that fire
0–100, based on the largest numerical value of the RPN. outbreaks and explosions with a raw score of 46 has the most significant

24
M.O. Sanni-Anibire, et al. Safety Science 121 (2020) 15–29

G
K
4

A Medium risk High risk

3
OCCURENCE

2 N E

M
J
F
C I
1 D H

Low risk L
Medium risk B

0
0 1 2 3 4
IMPACT
Fig. 2. Impact-Occurrence risk assessment matrix.

impact/severity. Next to that are crane accidents with a raw score of 37. 5.2.3. Electrical/heat/chemical burns
The accident with the least impact/severity is slips, trips and falls with a Burns occur due to the exposure of the skin to heat sources such as
raw score of 2, followed by asphyxiation and respiratory issues with a steam, flames, hot surfaces or liquids, electrical current, chemical
raw score of 4. The results of the frequency of occurrence as presented substances, and also prolonged exposure to the sun. The results show
in Table 3 shows that slips, trips and falls is considered the most fre- that out of eight causes of electrical, heat and chemical burn accidents,
quent accident in construction sites with a raw score of 38, while fire “lack/improper use of PPE”, “exposure to fire, steam, hot objects, or hot
outbreaks/explosions, structural collapse and drowning are considered liquids”, “extensive exposure to the sun” were the three most prominent
the least occurring accidents with raw scores of 4. The adjusted total causes of the accidents. The least cause identified was “contact with wet
scores as presented in Tables 2 and 3 have been computed to a range of cement”.
0–4 based on the value of the highest raw score. The derived values are
further plotted on a risk assessment (Impact-Occurrence) matrix as 5.2.4. Power tool accidents
shown in Fig. 2. The figure shows that none of the accidents have been Power and hand tools are amongst a worker’s most valuable assets.
perceived to be in the high risk area. However, G (fire outbreak/ex- However, these tools can become very dangerous if used unsafely. Nine
plosions), K (structural collapse), C (electrical/heat/chemical burns), D causes of power tool accidents were identified. The results show that
(power tool accidents), I (exposure to hazardous materials), H (struck the three most prominent of these causes include “inattention/in-
by fixed/moving objects), F (falling objects) and B (slips, trips and falls) experience of users”, “use of tools in wet and damp conditions”, and
have been perceived to be of medium risk. “dysfunctional machine guards”. The least cause identified was “lack of
maintenance of power tools”.
5.2. Ranking of causes of construction accidents
5.2.5. Falls from heights
This section summarizes the important findings on the ranking and The results show that out of nine causes of falls from heights acci-
prioritization of causes leading to construction accidents. Table 1 pre- dents, “workers’ distractions/carelessness of the worker”, “improper
sents all fourteen accidents with their corresponding references. Ad- barriers/guardrails to open edges”, and “insufficient lighting/illumi-
ditionally, Table 4 illustrated the potential causes of the identified ac- nation” were the three foremost causes. “Struck by moving objects at
cidents arranged in descending order according to priority. height” was identified as the least cause.

5.2.1. Crane accidents 5.2.6. Falling objects

The results show that, out of eleven causes of crane accidents, “lack Falling objects entails various objects such as tools, nails, paint cans
of regular maintenance” was ranked as the most prominent cause of this and containers and other materials falling under gravity. Nine causes of
type of accident. Followed by “Operator error/Lack of adequate falling object accidents were identified. The three most common causes
training and skill” and then “Lack of regular testing/inspections”. The are “excessive winds”, “roof/floor openings”, and “unprotected roof
least cause listed is “Incorrect design/Improper erection/assembling/ edges, absence of covered walkways or fan scaffolds”. The least cause
anchorage”. was identified as “inadequate training and awareness”.

5.2.2. Slips, trips and falls 5.2.7. Fire outbreak/explosions

This study presented nine causes of slips, trips and fall accidents, Hot works on construction sites are potential sources of fire and
and identified the three most prominent causes to be “poor house- explosion risks. The risks are further exacerbated when flammable or
keeping”, “poor or uneven floor/pavement surfaces”, and “physical combustible materials are nearby. The results show that out of six
inactivity”. The least cause of this accident is identified as being due to causes of fire outbreak/explosion accidents “lack of proper storage of
“footwear, wet floors”. combustible/flammable substances”, “electric hazards”, and “lack of

25
M.O. Sanni-Anibire, et al. Safety Science 121 (2020) 15–29

fire safety precautions and procedures” were the three most prominent 5.3. Implementation of the proposed risk assessment approach
causes, while “faulty equipment and machinery” was identified as the
least cause. The risk assessment checklist presented in Table 4 was jointly de-
ployed on the car park building described in Section 4.4. The assess-
5.2.8. Struck by fixed/moving objects ment team comprised of a civil/structural engineer, a safety manager
The results show that out of six causes related to accidents due to and the site’s construction project manager. This was to determine the
struck by fixed/moving objects “vehicle/equipment overturning”, risk rating of the various potential causes of accidents listed in the
“machinery with unguarded moving parts”, and “communication pro- checklist. The procedure involved a review of project documents, and
blems between vehicle drivers and employees or members of the safety plans, discussions with the site personnel including one con-
public” were identified as the top three causes of accidents. The least sultant’s safety officer, client’s project supervisor, and contractor’s
cause of accident was identified as “lack of PPE/loose clothing”. safety manager. And finally, a walk-through of the site while observa-
tions were noted. The site is located in an arid region, away from any
water bodies, and thus observations for drowning accidents have been
5.2.9. Exposure to toxic/hazardous materials
denoted as not available (N/A).
Construction sites may harbor hazardous materials such as asbestos,
The results of the observations showed that the highest risk rating
lead, silica dust, organic solvents, sewer gases, welding fumes and
(4) included “lack/improper use of PPE”. Discussions with construction
radioactive materials. Six causes of accidents due to exposure to toxic/
workers showed that PPEs have been provided, but have been neglected
hazardous materials were identified. The top three causes are “lack of
by the workers. Thus, a problem of enforcement is inherent rather than
proper storage”, “inadequate training and awareness”, and “lack of
unavailability of the PPEs. Also, “extensive exposure to the sun” was
suitable techniques and equipment to monitor working environment”.
observed as the average temperature in summer months is recorded at
The least cause identified was “lack of first aid/emergency procedures”.
36 °C. Other highly rated risks (3) included: “poor housekeeping” which
could lead to trips and falls. “Lack of barriers to prevent collision and
5.2.10. Scaffold accidents pedestrian interference”, around the crane. Site observations showed
Scaffolds are temporary structures that are used in construction that the area around the crane has not been well guarded and workers
projects for working at heights. The results show that out of nine causes moved freely around this area. Interference around the base of the
of scaffold accidents, “defective guardrails, toes boards, braces etc.”, crane could lead to stability-related issues, or serious injuries to pe-
“incorrect design/improper erection/assembling/anchorage of scaf- destrians. “Unexpected interruption while working”, could lead to
fold”, and “overloading”, are the top most causes of accidents. power tool accidents due to lack of concentration of the operator. It was
“Insufficient/inoperative personal protective equipment (PPE), per- observed that workers interacted with tool operators without any
sonal fall arrest systems (PFAS)” was identified as the least contributing warning signs to stop or pause work. “Improper barriers/guardrails to
cause. open edges” at heights. In the erection of precast elements, the erector
has not been protected with the provision of adequate barriers around
5.2.11. Structural collapse open edges at heights. “Insufficient/inoperative personal protective
Structural collapse of buildings under construction results in sig- equipment (PPE), personal fall arrest systems (PFAS)” for works at
nificant loses in life, material and financial resources, in addition to height. Though work at heights is limited to few personnel, due to the
severe injuries to workers and its negative influence on the social image mode of construction which is primarily the use of pre-fabricated ele-
of the organizations involved. The results show that out of five causes of ments, adequate PPE has not been observed by the few who work at
structural collapse, “poor construction materials”, “lack of proper de- heights. Other issues related to working at heights include “Inadequate
sign”, and “incompetent personnel/supervisor” were the most promi- knowledge and experience”, which was observed based on interactions
nent causes. The least prominent cause was identified as “weak foun- with some of the workers. “Struck by moving objects at height”, due to
dations”. the reliance on the crane to move pre-fabricated elements into place.
“Unprotected roof edges, absence of covered walkways or fan scaffolds”
could lead to objects falling from heights. A number of workers were
5.2.12. Asphyxiation/respiratory issues
noticed not to be donning the hard hats despite its provision. “Lack of
Five causes of asphyxiation/respiratory illnesses in construction
measurement of site conditions” and “presence of respirable dusts (as-
sites were identified. The most prominent causes are “insufficient/lack
bestos, coal, silica, some plastics and talc)”, could lead to respiratory
of PPE”, “lack of training on confined spaces”, and “lack of measure-
issues. The desert nature of the site location allowed for an ample
ment of site conditions”. The least cause identified was “lack of first
amount of dust, in addition to silica present due to concrete and mortar.
aid/emergency procedures”.
Based on the risk rating provided, an adjusted risk rating has been
computed. The average risk rating for all fourteen construction acci-
5.2.13. Trench/excavation accidents dents has been presented in Table 6. The table shows that, the highest
Excavations are depressions in the earth’s surface due to the re- risk rating was due to “asphyxiation/respiration issues” with a risk
moval of earth. Eight causes of trench/excavation accidents were rating of 41.46, followed by “falls from height” with a risk rating of
identified. The top three causes include “falling loads from above a 35.7, and “electrical/heat/chemical burns” with a risk rating of 31.8.
trench or excavation site”, “cave ins due to lack of sidewall shoring”, Conversely, the highest safety performance was inversely proportional
and “absence of ramp/ladder for escape”. The least cause identified was to accidents with the least risk rating. Thus, “structural collapse”, “slips,
“proximity of waste/stored materials”. trips and falls”, “fire outbreak/explosions”, and “scaffold accidents”
have the highest safety performance with 100%, 85.11%, 84.89%,
5.2.14. Drowning 84.09%, respectively. Though these values of safety performance seem
Drowning accidents is due to respiratory impairments resulting high, on a six sigma rating scale, they score 6, 2.66, 2.65, and 2.62,
from immersion in, or under a liquid. Four causes of drowning acci- respectively. Besides “structural collapse”, all safety performance levels
dents were identified, they include, in descending order, “lack of ade- fall below 3 sigma level. At 3 sigma level, 66,800 accidents are likely to
quate supervision and security”, “lack of required PPEs (life jackets, occur per million opportunities (or 66.8 accident per thousand oppor-
buoyancy aids and grab lines)”, “lack of barriers to edges of drains and tunities). The average safety performance for the car park building is
ponds”, and “lack of rescue equipment, lifebelts, harnesses, safety lines, also computed to be 77.19%, which corresponds to 2.33 sigma level.
and in some cases, a rescue boat”. That means 228,739 accidents per million opportunities (or 228.74

26
M.O. Sanni-Anibire, et al. Safety Science 121 (2020) 15–29

Table 6
Safety performance of car-park building.
Types of accident Adjusted risk rating Safety performance Sigma rating DPMO

Crane accidents 25.80 74.20 2.21 257,743

Slips, Trips and Falls 14.89 85.11 2.66 148,978
Electrical/Heat/Chemical burns 31.80 68.20 1.98 699,160
Power tool accidents 25.42 74.58 2.22 255,326
Falls from height 35.70 64.30 1.87 741,290
Falling object 18.98 81.02 2.49 190,067
Fire outbreak/explosions 15.11 84.89 2.65 151,395
Struck by fixed/moving objects 23.47 76.53 2.31 233,573
Exposure to toxic/hazardous materials 18.82 81.18 2.49 190,067
Scaffold accidents 15.91 84.09 2.62 158,646
Structural collapse 0.00 100.00 6 3.4
Asphyxiation/respiratory issues 41.46 58.54 1.72 798,740
Trench/excavation accidents 29.11 70.89 2.07 291,581
Drowning N/A N/A N/A N/A
Average project safety performance 77.19 2.33 228,739

accidents per thousand opportunities). the global construction industry. Though Saudi Arabia has a very
buoyant construction industry, its construction market is not com-
5.3.1. Recommendations parable to other developed countries across the world. Furthermore,
In light of the potential causes of accidents identified in the risk the safety awareness and training of professionals in the region is
assessment process, countermeasures were developed to improve the still in its developing stages. This limitation however in made up for
safety performance of the car park building project. These include, in Saudi Arabia being the largest construction market in the Middle
though not limited to, the following: East region (Alasmari et al., 2012), and thus the study can be
comfortably implemented in neighboring countries. Additionally,
• Training, provision and enforcement of the use of PPE for the var- the approach presented in the study is suitable to be re-adapted to
ious construction activities on the site. any construction climate globally.
• Provision of proper PPE to prevent heat and sun burns due to pro- 2. The study has based its approach on hazard identification and
longed exposure. Monitoring of weather conditions. Work sche- analysis which is the most important safety process in improving
duling/shifts to avoid heat stress. safety performance. The authors will like to highlight that by the
• Regular housekeeping to remove all trailing cables, ropes and lines. very nature of hazard analysis, not all the causes of hazards/acci-
Also the absence of wet floors and anti-skid mats where necessary. dents can be determined nor can it be fully covered. Furthermore, a
Provision of caution signs in the right places. case study approach has been used in validating the approach. Thus,
• The area around tower crane should be properly guarded with re- this study is limited in the number of accidents, and their causes,
quired barriers to prevent pedestrian interference and risk to identified by the researchers, and also in the nature and scope of the
workers. Provision of proper tag lines to prevent dangerous spin or case study used. It is however, convenient during adaptation of the
swing of materials when raised or lowered by a crane. approach to other regions that additional accidents and/or causes of
• Power tool users should only entertain questions and comments accidents are included in the procedure and are analyzed.
without interruption.
• Proper barriers, guard rails, and safety nets should be erected where
7. Discussion and conclusion
work at heights is to be performed.
• Ensuring guard rails, toe board and warning lines are provided at
The construction industry has witnessed a high influx of research
working platforms and defective elements are replaced.
• Required PPEs and PFAS should be provided to workers working at and application to ensure that the construction process is safer and
more reliable, and ultimately reduce the high rate of fatality recorded
heights.
• Measurements of site conditions to ensure dust levels, fumes, smoke yearly. Preventing occupational injuries and accidents in construction
has often driven the industry to make changes. The efforts that have
and other particles are within threshold values. Additionally, suffi-
cient PPEs should be provided and its usage should be enforced and been made are generally described as reactive or pro-active. While re-
monitored. active techniques have been criticized as inadequate in measuring
• Continuous training on safety laws and requirements should be construction safety performance, proactive techniques have been fa-
vored by many stakeholders. Hazard identification and risk assessment
provided to all workers on site according to their job tasks.
• Enforce compliance with the municipal safety standards from top has been adjudged to be the most important safety process (Manuele,
2013). Risk assessment can both be considered as an art and science. It
management to include the last worker on the project site.
entails the ability to modify appropriate methods and express in-
6. Study limitations formation in a way that effectively communicates risks (Lyon and
Popov, 2016). Numerous risk assessment methods have been developed
This study aimed at presenting a risk assessment approach, which over the years, and have subsequently evolved into variants of the
will enhance the safety performance of the construction industry. original ideas on par with technology, communication and new dis-
However, the authors recognize the following limitations of this study coveries. Some of these efforts have focused on specific industry
as listed below: adaptation or unique applications. Other efforts have sought to create
hybrid techniques through the combination of various existing methods
1. The case study approach used in this study, where the sample of (Fung et al., 2010; Liu and Tsai, 2012; Amiri et al., 2017). These
professionals has been drawn from the Eastern Province of Saudi methods are either over-simplified, thereby not being able to accurately
Arabia, limits the potential to generalize the results of the study to estimate the inherent risks, or are too complex in nature, posing a
challenge to their adoption and implementation in the industry.

27
M.O. Sanni-Anibire, et al. Safety Science 121 (2020) 15–29

Furthermore, these tools have not provided a clear methodology that conceptual framework. Int. J. Bus. Social Res. 7 (01), 01–10.
can effectively enhance construction safety performance. The current Al-Tabtabai, H.M., 2002. Analyzing construction site accidents in Kuwait. Kuwait J. Sci.
Eng 29 (2), 213–238.
challenge is in developing a simple, yet efficient approach. Amiri, M., Ardeshir, A., Zarandi, M.H.F., 2017. Fuzzy probabilistic expert system for
Accordingly, this study provides a comprehensive list of potential occupational hazard assessment in construction. Safety Sci. 93, 16–28.
accidents in construction projects. It also seeks to establish the causes of Awwad, R., El Souki, O., Jabbour, M., 2016. Construction safety practices and challenges
in a Middle Eastern developing country. Safety Sci. 83, 1–11.
these accidents. The information gathered so far was used in developing Choudhry, R.M., Fang, D., Mohamed, S., 2007. Developing a model of construction safety
a unique risk assessment approach based on hazard identification. culture. J. Manage. Eng. 23 (4), 207–212.
Furthermore, the six sigma methodology has been used to analyze the Dong, S., Li, H., Yin, Q., 2018. Building information modeling in combination with real
time location systems and sensors for safety performance enhancement. Safety Sci.
results. This is based on the premise that the six sigma approach is a 102, 226–237.
methodology and measurement system that seeks to achieve zero de- Dorrington, M., Carla, F., 2007. Encyclopedia of Energy Engineering and Technology-3
fect/failure/accident/injury and thus, achieve the ultimate perfor- Volume Set (Print). Anwar, S., Capehart, B.L. (Eds.). CRC Press.
Fung, I.W., Tam, V.W., Lo, T.Y., Lu, L.L., 2010. Developing a risk assessment model for
mance of a system or process. The introduction of the six sigma mea-
construction safety. Int. J. Project Manage. 28 (6), 593–600.
surement scale can be viewed as an incremental innovation to already Gambatese, J.A., Behm, M., Rajendran, S., 2008. Design’s role in construction accident
existing risk assessment approaches. causality and prevention: perspectives from an expert panel. Safety Sci. 46 (4),
The developed approach applied to the case study shows that a 675–691.
Gunduz, M., Birgonul, M.T., Ozdemir, M., 2016. Fuzzy structural equation model to assess
number of accident causes received high risk ratings such as the use of construction site safety performance. J. Constr. Eng. Manage. 143 (4), 04016112.
PPEs, provisions for working at heights. High temperatures due to Gunduz, M., Birgonul, M.T., Ozdemir, M., 2018. Development of a safety performance
weather could lead to heat stress and sun burns if the proper precau- index assessment tool by using a fuzzy structural equation model for construction
sites. Autom. Constr. 85, 124–134.
tions are not put in place. Also, the site conditions—which often entails Gürcanlı, G., Baradan, S., Uzun, M., 2015. Risk perception of construction equipment
a lot of sand and silica dust—could lead to respiratory issues. operators on construction sites of Turkey. Int. J. Indust. Ergonom. 46, 59–68.
Furthermore, the study reveals that while the safety performance pre- Hämäläinen, P., Takala, J., Saarela, K.L., 2006. Global estimates of occupational acci-
dents. Safety science 44 (2), 137–156.
sented may seem to be on the high side, the six sigma measurement Hamid, A.R.A., Majid, M.Z.A., Singh, B., 2008. Causes of accidents at construction sites.
scale showed that all safety performance levels were below 3-sigma Malaysian J. Civil Eng. 20 (2), 242–259.
level, except for “structural collapse”. More specifically, the asphyxia- Haslam, R.A., Hide, S.A., Gibb, A.G.F., Gyi, D.E., Pavitt, T., Atkinson, S., Duff, A.R., 2005.
Contributing factors in construction accidents. Appl. Ergonom. 36 (4), 401–415.
tion/respiratory issues were measured at 1.72 sigma level, which im- Hassanain, M.A., Hafeez, M.A., Sanni-Anibire, M.O., 2017. A ranking system for fire
plies that for every million possibilities there will be 798,740 injuries/ safety performance of student housing facilities. Safety Sci. 92, 116–127.
accidents. Huang, X., Hinze, J., 2003. Analysis of construction worker fall accidents. J. Constr. Eng.
Manage. 129 (3), 262–271.
These results can be used by the concerned stakeholders to improve
Hughes, P., Ferrett, E., 2015. Introduction to Health and Safety in Construction: For the
the safety conditions of the site. This can be through eliminating, re- NEBOSH National Certificate in Construction Health and Safety. Routledge.
ducing, and controlling the identified risks. The approach is also in- Jannadi, M.O., Assaf, S., 1998. Safety assessments in the built environment of Saudi
strumental in the continuous enhancement of the safety performance of Arabia. Safety Sci. 29, 15–24.
Jolin, Joseph J., 2009. Risk assessments Six Sigma style: internal auditors at Textron used
the construction project until a target safety baseline set by the orga- modern process development methods to enhance the way they assess the organi-
nization is achieved, in the ranges of 3, 4, 5 or 6 sigma levels. This study zation's risks. Internal Auditor. Institute of Internal Auditors, Inc. 2009. Retrieved
presents useful insights into the risk perception of construction acci- January 27, 2018 from HighBeam Research: https://www.highbeam.com/doc/1G1-
214841443.html.
dents in the region of Saudi Arabia. It also ranks the priorities of each Kartam, K., Bouz, R., 1998. Fatalities and injuries in the Kuwaiti construction industry.
cause of accident in relation to the actual occurrence of the accident. Acc. Anal. Prevent. 30 (6), 805–814.
Finally, this study presents a unique tool that can be practically im- Ling, F.Y.Y., Liu, M., Woo, Y.C., 2009. Construction fatalities in Singapore. Int. J. Project
Manage. 27 (7), 717–726.
plemented by the construction industry to improve its safety perfor- Liu, H.T., Tsai, Y.L., 2012. A fuzzy risk assessment approach for occupational hazards in
mance. Though the data has been obtained from the regional context of the construction industry. Safety Sci. 50 (4), 1067–1078.
Saudi Arabia, it is also relevant to other countries in the Middle East Luther, C., 2003. Roof design by weighted evaluation. Interface 6–8.
Lyon, B.K., Popov, G., 2016. The art of assessing risk. Profess. Safety 61 (03), 40–51.
with a similar construction climate. Thus, it could also be adapted to
Manuele, F.A., 2008. Advanced Safety Management Focusing on Z10 and Serious Injury
other construction climates. Prevention. John Wiley & Sons.
Manuele, F.A., 2013. On the Practice of Safety. John Wiley & Sons.
Misiurek, K., Misiurek, B., 2017. Methodology of improving occupational safety in the
Acknowledgement
construction industry on the basis of the TWI program. Safety Sci. 92, 225–231.
Mosly, I., 2015. Safety performance in the construction industry of Saudi Arabia. Int. J.
The authors are thankful to King Fahd University of Petroleum and Constr. Eng. Manage. 4 (6), 238–247.
Minerals for the support and facilities that made this research possible. Ofori, G., 2000. Challenges of construction industries in developing countries: lessons
from various countries. In: 2nd International Conference on Construction in
Also, Mr. Haitham Al Marshad, Mr. Raed Khalid Jambi, and Mr. Rakan Developing Countries: Challenges Facing the Construction Industry in Developing
Saad Al Krri for helping in the data collection process. Lastly, the au- Countries, Gaborone, November 5 (24), pp. 15–17.
thors are thankful to the project management team of the case study for Okoye, P.U., 2016. Improving the safety performance of Nigeria construction workers: a
social ecological approach. Universal J. Eng. Sci. 4 (2), 22–37.
providing access and information to the car park building construction Patel, D.A., Jha, K.N., 2016. Developing a process to evaluate construction project safety
project, as well as the reviewers that helped to improve the quality and hazard index using the possibility approach in India. J. Constr. Eng. Manage. 143 (1),
value of the manuscript. 04016081.
Pinto, A., Nunes, I.L., Ribeiro, R.A., 2011. Occupational risk assessment in construction
industry–overview and reflection. Safety Sci. 49 (5), 616–624.
References Popov, G., Lyon, B.K., Hollcroft, B., 2016. Risk Assessment: A Practical Guide to Assessing
Operational Risks. John Wiley & Sons.
Pyzdek, T., Keller, P.A., 2014. The Six Sigma Handbook. McGraw-Hill Education, New
Abdelhamid, T.S., Everett, J.G., 2000. Identifying root causes of construction accidents. J.
York, pp. 25.
Constr. Eng. Manage. 126 (1), 52–60.
Raheem, A.A., Issa, R.R., 2016. Safety implementation framework for Pakistani con-
Agwu, M.O., Olele, H.E., 2014. Fatalities in the Nigerian construction industry: a case of
struction industry. Safety Sci. 82, 301–314.
poor safety culture. British J. Econ., Manage. Trade 4 (3), 431–452.
Sawacha, E., Naoum, S., Fong, D., 1999. Factors affecting safety performance on con-
Alarcón, L.F., Acuña, D., Diethelm, S., Pellicer, E., 2016. Strategies for improving safety
struction sites. Int. J. Project Manage. 17 (5), 309–315.
performance in construction firms. Acc. Anal. Prevent. 94, 107–118.
Siu, O.L., Phillips, D.R., Leung, T.W., 2004. Safety climate and safety performance among
Alasamri, H., Chrisp, M.T., Bowles, G., 2012. A framework for enhancing and improving
construction workers in Hong Kong: the role of psychological strains as mediators.
the safety culture on Saudi construction sites. In Proceedings of the 28th Annual
Acc. Anal. Prevent. 36 (3), 359–366.
ARCOM Conference. Edinburgh, UK: Association of Researchers in Construction
Tam, C.M., Zeng, S.X., Deng, Z.M., 2004. Identifying elements of poor construction safety
Management 475–485.
management in China. Safety Sci. 42 (7), 569–586.
Al-Bsheish, M.A., 2017. Enhancing safety performance by recognizing the role of per-
Waehrer, G.M., Dong, X.S., Miller, T., Haile, E., Men, Y., 2007. Costs of occupational
ceived management commitment to safety in the Jordanian healthcare industry:
injuries in construction in the United States. Acc. Anal. Prevent. 39 (6), 1258–1266.

28
M.O. Sanni-Anibire, et al. Safety Science 121 (2020) 15–29

Wilson, J.M., Koehn, E.E., 2000. Safety management: problems encountered and re- Yilmaz, F., 2015. Monitoring and analysis of construction site accidents by using acci-
commended solutions. J. Constr. Eng. Manage. 126 (1), 77–79. dents analysis management system in Turkey. J. Sustain. Develop. 8 (2), 57.
Wu, T.C., Chen, C.H., Li, C.C., 2008. A correlation among safety leadership, safety climate Zwetsloot, G.I., Aaltonen, M., Wybo, J.L., Saari, J., Kines, P., De Beeck, R.O., 2013. The
and safety performance. J. Loss Prevent. Process Indust. 21 (3), 307–318. case for research into the zero accident vision. Safety Sci. 58, 41–48.
Yazdani, A., Hilbrecht, M., Imbeau, D., Bigelow, P., Neumann, W.P., Pagell, M., Wells, R., Zwetsloot, G.I., Kines, P., Wybo, J.L., Ruotsala, R., Drupsteen, L., Bezemer, R.A., 2017.
2018. Integration of musculoskeletal disorders prevention into management systems: Zero accident vision based strategies in organisations: innovative perspectives. Safety
a qualitative study of key informants’ perspectives. Safety Sci. 104, 110–118. Sci. 91, 260–268.

29