Safety Science 80 (2015) 1–12

Contents lists available at ScienceDirect

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

Activity based risk assessment and safety cost estimation for residential
building construction projects
G. Emre Gurcanli, Senem Bilir ⇑, Merve Sevim
Technical University of Istanbul, Department of Civil Engineering, Division of Construction Management, Ayazaga Campus, Maslak, 34469 Istanbul, Turkey

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

Article history: Fatal construction accidents are major problem in Turkish construction industry. Although new compre-
Received 6 February 2015 hensive legislation has been inured recently; level of consciousness in the industry is unsatisfactory and
Received in revised form 19 June 2015 safety is perceived as an extra cost. Small or middle scaled residential projects have a big share in the
Accepted 1 July 2015
industry and majority of the fatal accidents usually occur on such work sites. Focus of this study is small
and middle-scaled residential construction sites. Drawings, technical specifications, bill of quantities of
25 concrete residential buildings with total areas that vary between 230 and 118,200 m2 in Istanbul were
Keywords:
examined. Site visits and information gathering were performed from site engineers. This study aims to
Risk assessment
Safety cost
give an approach for safety cost estimation for the early stages of construction bidding phase using risk
Scheduling assessment activities and construction project scheduling by focusing on construction activities.
Construction industry Additionally, a cost analysis was performed and the results were compared and analyzed with a former
Logarithmic regression project that focused on 30 concrete residential projects in Istanbul. Primavera P6 was utilized for
scheduling, and risk assessment was conducted using the techniques applicable for construction projects
(i.e., L matrix and Fine-Kinney). Finally, activity-based techniques were used for cost estimation. The
results of this study reveal that the percentage of safety cost to the total construction cost is 1.92%.
Also, to comply with the current legislation and minimize risks; 0.85 USD per man-hour should be spent
according to 2013 rates and safety cost per unit area was found as approximately 5.68 USD.
Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction playing a significant role in the economic development of Turkey.

In recent years, not only in Turkey, but also all around the
world, safety in the construction industry has become a matter
of grave concern to government bodies and private enterprises.
When safety assessments are required in construction sites, there
is often insufficient data or imprecise information available
(Gurcanli and Mungen, 2009). Additionally, different companies
tend to have different scales in various safety management
systems because of resource limitations (Tam et al., 2002). Thus,
the construction industry has a higher fatal accident rate than
other industries (Waehrer et al., 2007).
The Turkish construction industry has had double-digit growth
and success in recent years. Most of this growth has come since
2011 from projects of the Housing Development Administration
of Turkey (TOKI). After a small increase in 2012, the industry expe-
rienced 11.5% growth in terms of GNP in 2013 and its share in total
employment reached 7.4%, the highest number in the last 20 years
(TUIK, 2013). As these figures show, the construction industry is
⇑ Corresponding author. Tel.: +90 212 285 37 36; fax: +90 212 285 65 87.
E-mail address: sbilir@itu.edu.tr (S. Bilir).
However, the industry also has a reputation of being one of the
most dangerous domestic industries. Compared with other
industries, construction accidents are frequent with a diverse set
of small-scale hazard sources.
Current legislation on occupational safety in Turkey provides
for regulation and enforcement of employers to implement safety
measures as well as safety management systems. However, the
level of consciousness in the industry is unsatisfactory and safety
is perceived as an extra cost and unnecessary expenditure. More
specifically, in order to stay economically competitive, sustainable
and to reach maximum profits, many contractors only execute
basic safety measures and eliminate many important hazard
prevention-training programs during construction project
implementation (Cheng et al., 2010a).
This paper provides the results of a project conducted by the
authors, which was supported by The Scientific and
Technological Research Council of Turkey (Project Number
112M033). The aim was to combine construction project schedul-
ing, risk assessment activities and construction cost estimation to
value safety costs and its distribution throughout a project. In this
way, this research provides a better understanding of safety costs

http://dx.doi.org/10.1016/j.ssci.2015.07.002
0925-7535/Ó 2015 Elsevier Ltd. All rights reserved.
2 G.E. Gurcanli et al. / Safety Science 80 (2015) 1–12
during a building construction project. Therefore, this approach
helps in efforts to reduce loss of workers’ life and mitigation
of the safety costs arising from severe injuries, fatalities,
administrative procedures, legal obligations and litigation costs
and expenses.
2. Literature
Cost estimation models are useful in the preliminary design
stage of a construction project if they are able to estimate construc-
tion costs with minimum project information in the early stages
(Kim et al., 2004). Such estimates allow owners and planners to
evaluate project feasibility and effectively control costs in detailed
project design work. However, contractors do not have any stan-
dard for cost estimation in the European Union (EU) or in other
countries. Several sources mention various cost estimating imple-
mentations (Akarca, 1992; Akintoye, 2000; Arpaci, 1995; Cook,
1982; Cox, 1987; Curran, 1989; Ciraci et al., 1996; Feng et al.,
2015; PMBOK, 2008; Polat, 2004; Seyyar, 2000; Usta, 1994).
Some of these cost estimation methods include: Unit Method,
Superficial Method, Approximate Quantities Method, Artificial
Neural Networks, Fuzzy Logic, Genetic Algorithms, Regression
Analysis, etc. Since the accuracy of construction cost estimates
(as well as contingency for contractors) is a critical factor in the
success of the project, many cost estimation models that adopt
parametric methods have been developed. Approaches to cost esti-
mation based on statistics (Singh, 1990; Wilson, 2005) and linear
regression analysis (Kouskoulas and Koehn, 1974) were developed
in the 1970s. During the late 1980s a new approach to cost estima-
tion was introduced based on user experience and increased
research into the potential of artificial intelligence, such as with
expert systems (Kim et al., 2004).
The literature on construction project cost estimation includes
research on the costs of accidents. Costs of accidents have been a
focus point for safety experts and academicians in the field of occu-
pational safety. There are numerous studies that analyze the cost of
lost working days, insurance, delay in project duration, etc. In a
study undertaken by the Health and Safety Executive (of
England), it was shown that indirect costs could be 36 times
greater than the direct costs of an accident. In other words, the
direct costs of an accident or disease represent the tip of the ice-
berg when compared to overall costs (Cheng et al., 2010b).
Panopoulos (2003) sought optimum safety cost by estimation of
prevention cost, average accident cost and management failure
with-no-accident cost and determined whether or not safety pays
for a project. Everett and Frank (1996) reexamined the total cost of
accidents and injuries in the construction industry making neces-
sary assumptions, method changes and updating inputs. Feng
et al. (2015) calculated average direct accident costs, indirect acci-
dent costs and total accidents costs of building projects in
Singapore using bivariate correlation analysis and moderated
regression analysis.
According to Cooper and Kaplan (1992), Activity-Based Costing
(ABC) has emerged as a new approach that associates costs directly
related to business activities with manufactured products. ABC
gives more accurate information especially in complex structures
and allows one to obtain information on cost items that are
ignored in traditional methods (Rzvi and Elnathan, 1999).
Furthermore, this method is flexible for all kinds of
technology-intensive business systems and can be used success-
fully in developing countries (Alkan, 2005; Liu and Pan, 2007).
ABC provides detailed information about the cost of activities in
a particular process and helps decision-makers to develop reason-
able decisions (Khataie et al., 2011). According to the
Ríos-Manríquez et al. (2014) ABC implementation gives firms a
chance to compare their profitability, productivity and cost control
with other firms. The other objectives of this system are providing
cost reduction and management; determination of profitability of
customers, inventory valuation, and new product and service
design.
However, studies about the cost of safety measures as a part
of project costs during a construction project are very rare.
Tan (1999) compared the cost of safety measures during a project
and costs of accidents in Turkey. Aminbakhsh et al. (2013) investi-
gated the safety risk assessment during planning and budgeting of
construction projects using an analytic hierarchy process (AHP).
Chalos (1992) introduced a cost of safety model to conceptually
describe the cost-benefit analysis of accident/injury prevention.
Tappura et al. (2015) discussed management accounting for
safety-related decision-making and valuing human life in a
cost-benefit analysis. Alonso et al. (2013) studied the impact of
health and safety on the investment of construction companies
by applying a questionnaire in southern Spain. Sousa et al.
(2014) presented an Occupational Safety and Health Potential
Risk Model for estimation of the statistical costs of occupational
safety and health risk. As health and safety costs are not calculated
during the bidding period, the parties do not want to waste money
for safety expenses or reduce their contract profits.
The objective of this paper is to offer an approach for safety cost
estimation in the early stage of the construction bidding phase
using construction project scheduling data, and risk assessment
techniques applicable for construction projects, such as the L
matrix and Fine-Kinney methods (Hughes and Ferrett, 2005;
Kinney and Wiruth, 1976; Rausand, 2011) and ABC method.
Using an activity-based risk assessment and activity-based cost
estimation analysis would be consistent since the safety risks of
every activity are analyzed and costs of these risks are investigated
in this study. Furthermore, consideration of safety costs as part
of project costs and calculation of safety cost using both
activity-based risk assessment methods and ABC analysis will offer
a new approach for determination of safety expenses in the con-
struction industry. The approach provides a very practical way to
estimate the share of safety costs in total project costs for building
projects. Hence, by implementation of the approach suggested,
contractors can readily prepare and implement their safety
management plan while controlling costs and improving project
management, and most importantly save human life and protect
their workers on site.
3. Methodology
This section provides details on specific topics related with the
proposed approach and methodology. Fig. 1 shows a graphical
representation of the methodology.
3.1. Data collection
This paper attempts to estimate safety costs and its distribution
throughout a project before construction starts. For this purpose,
firstly drawings, technical specifications, bill of quantities of 25
concrete residential buildings in the region of Istanbul province
were collected, construction sites were visited and information
gathered from the site engineers about safety practices on site with
the financial support of The Scientific and Technological Research
Council of Turkey (TUBITAK). In total, the 25 concrete residential
building projects vary between 230 and 118,200 m2 in area.
Since the costs of construction projects vary due to regional
characteristics (different labor wages, different tax policies of the
municipalities, variable transportation, material cost and
additional cost due to site facilities), the study only focused on
the projects in Istanbul province.

3.1.1. Preparation of work breakdown structures
Since construction work is a project-based production type and
differs from a process-based production; depending on the struc-
ture type, different risks and different occupational accidents come
to the fore. However, although the construction industry has a
project-oriented type of production, in similar projects there are
similar activities. A work breakdown structure (WBS) divides the
project into several parts to make activities more manageable.
Each of these parts can be considered as products; planning, cost
estimates, observations and controls can be made via these parts
(Project Management Body Of Knowledge-PMBOK, 2008).
Accordingly, establishing WBS and determining the work items
precisely are beneficial and necessary efforts for effective activity
based risk assessment. Park and Kim (2012) mentioned that effec-
tive activity based risk identification is one of the most crucial
aspects for site safety management planning in efficiently accom-
plished projects.
In this study, the work breakdown structures of 25 building
projects were established and for each construction project all
work items were listed. Sources were determined and assigned
to listed work items. Fig. 2 shows the work items of column con-
crete work for Project 1. At this stage of the study, the authors
and concerned departments of the projects discussed the work
items. Ultimately, a number of adjustments and assumptions were
made about the work items of the construction projects.
3.1.2. Determination of man-hour values
A properly conducted and evaluated work program will facili-
tate successful completion of a project. An insufficient work pro-
gram may lead to an increase in time taken and costs of the
project. Companies should estimate an effective cost calculation
of required activities for successful project management. In order
to make such estimates, a proper work program is of great
R.C.C. Roof
Excavation Masonry
works works
2.1. lean 2.2. foundation 2.3. floor 2.4. beam
conc. conc. conc. conc.
G.E. Gurcanli et al. / Safety Science 80 (2015) 1–12 3
Fig. 1. Graphical representation of the methodology.
convenience for best-cost calculations (Pilcher, 1966; PMBOK,
2008).
There are many work items in construction projects and work
items have different qualifications from each other, which makes
a reliable comparison between them difficult. Therefore, in order
to evaluate productivity in the construction industry, using the
concept of man-hours allows translation of all these items into a
common metric. It is important to use man-hour values to calcu-
late the realistic costs of a project. One crucial point when using
man-hour values is that these values are durations that reflect only
the workers performing the unit work items (Kuruoglu, 2002).
Another important issue is that the man-hour values depend on
to site conditions, project restrictions and they can vary with the
nationality of the workers, etc. Hence, companies generally create
their own man-hour values to obtain more accurate results for
estimation of a budget. In this study, after arranging summary
information and work breakdown structures of all the projects,
preparation of scheduling tables, performing project schedules, a
determination of man-hour values for each activity was completed.
Project schedules were prepared using the planning tables, activity
durations and total number of workers for each activity. Primavera
P6 (http://www.oracle.com) software was utilized to prepare the
project schedules. Start and finish dates of each activity were set,
and relations between activities (finish to start or start to start
etc.) were determined.
This study utilized Kuruoglu and Bayoglu (2001) man-hour val-
ues derived from 69 different construction jobs. In their study
Kuruoglu and Bayoglu (2001) specified these values using the
man-hour values of the Turkish Ministry of Environment and
Urban Planning and averaged man-hour values obtained from sev-
eral companies. Briefly, this study comprises the man-hour values
from the Turkish Ministry of Environment and Urban Planning as
well as man-hours from real project scheduling tables of different
companies.
Construction Project
Electrical and Floor Doors & Lift Landscaping General
mechanical works Plastering Painting Grooting work
works Windows installation site works
2.5. column
conc.
2.5.1. Formwork
2.5.2. Pouring conc.
2.5.3. Re-bar bending
2.5.4. Settlement
Fig. 2. WBS of column concrete work of Project 1.
4 G.E. Gurcanli et al. / Safety Science 80 (2015) 1–12

3.2. Risk assessment

Gen. Site
Works

3.2.1. Analysis of possible hazards

According to the work break down structures and work item

x
x
lists, possible activity hazards are listed. Table 1 shows the funda-
Landscaping

mental work activities performed in a concrete residential building

project and related occupational hazards.
works

The main aim in performing hazard analysis is to determine

safety expenditures to prevent accidents by conducting risk assess-
x

x
ments and decreasing risk scores (to the level of ‘‘acceptable’’) by
inst.
Lift

x the utilization of personnel protective equipment, collective equip-

x
ment and other equipment or techniques as well as training pro-
window Inst

grams. Although many possible hazards are cited, they are the
most frequently encountered hazards according to past studies
Door &

done in Turkey (Gurcanli and Mungen, 2005; Gurcanli, 2006;

x

Gurcanli and Mungen, 2009, 2013; Mungen, 1997). Since this study
provides a point of view and establishes a new approach based on
historical data from past studies in Turkey, however, future studies
Rebar & other
metal works

should also consider other possible and contemporary hazards.

3.2.2. Assessment of risk scores

x

A risk analysis of each work item for 25 residential building

Floor

construction projects was accomplished with the help of activity

tables, a project schedule and work breakdown structures. The risk
x
x

scores of each work item were calculated and risk assessments

Painting

were completed using two different methods: L Matrix and

Fine-Kinney (Hughes and Ferrett, 2005; Kinney and Wiruth,
x

1976; Rausand, 2011).

Plastering

Risk assessments were done by the utilizing accident severity

and accident likelihood as input parameters for each construction
x

work item and sub-item. A conceptual framework and scores from

a study performed by Gurcanli and Mungen (2009) was used for
Grouting

accident likelihood and severity. This paper referenced some differ-

ent definitions and rankings of severity measures and utilized a
x

x
x

x
x

combination of these definitions (Australian and New Zealand

mech. works

Standard 4360, 1999; Management of Health and Safety at Work

Regulations (HSE), 1999; Sii et al., 2001; Tweeddale, 1997; Wang,
Elect. &

1997). In the literature, severity rankings were defined with 4 to

7 scales; while in this paper, a 5-scale ranking was chosen. After
x

a literature review (HSE, 2003; Raafat, 1995; Sii et al., 2001;

Masonry

Sii and Wang, 2002) a 5-scaled ranking was chosen for accident
likelihood. Subsequently, we obtained risk distributions for all
x

the projects. As an example, Table 2 shows the risk distribution of

Roof

Project 8. Similar risk distribution tables were prepared for all the
x

projects. Table 2 shows the ratio of risk categories to total risk scores,
total number of activities in the project, as well as number of
concr. const.
Reinforced

activities and percentages of items that have very high risk scores.
These risk analyses are generally performed in the planning
stages; hence it will be more reasonable to focus on the number
x

x
x

x
x

of very high and high-risk activities. As it can be seen from

Excavation

Table 2, the risk assessment of Project 8 gives 228 very high and
139 high-risk activities in a L Matrix Method. However, there are
450 very high and 39 high risk activities in a Fine-Kinney
x
x
x
x

x
x
x

Method risk assessment of Project 8. In order to stay on the safe

Equipment accidents (concrete pump

side, using the method that gives more very high and high-risk
Equipment accidents (roll over etc.)
(especially neighbor structures)

activities makes more sense. In Project 8, the results show that

Hazards due to machine and tool

Equipment accident (lifts, hoists,

Contact with underground lines
Hazard analysis of main job items.

the Fine-Kinney Method risk assessment was the more conserva-

Struck by flying/falling object

Struck by a moving vehicle

Building/Structure collapse

tive approach.
Manual handling hazards

usage (hummer etc.)

Fall into the excavation

Contact with electricity

Although both methods give similar results in certain aspects,

Possible hazards/jobs

the Fine-Kinney Method gave much higher levels of risk scores

Fall onto ground

for 25 projects than the L Matrix Method. In other words, the

Fall from height

Fine-Kinney Method scores reflect a safer estimate for almost all

cranes)

projects. The Fine-Kinney and L Matrix methods give the same risk
Cave-ins

etc.)

level for ‘‘cave-ins,’’ which are shown in Table 4. However, there is

Table 1

a large difference between these methods. In Fine-Kinney there

was a significant risk score difference between ‘‘explosion’’ and a
 G.E. Gurcanli et al. / Safety Science 80 (2015) 1–12 5

Table 2
Summary of risk distribution for project 8.

L Matrix method Fine-Kinney method

Very high High Average Low Very low Total Very high High Average Low Very low Total
Ratio of risk categories to total risk score (%) 23.6 34.1 37.1 5.2 0 100 74.2 13.3 12.4 0.2 0 100
Number of activity and percentages 228 139 531 450 39 531
(%42.9) (%26.2) (%84.8) (%7.3)

Table 3 and was accepted with amount of work with a unit cost and was
Percentage of very high-risk scores and percentage of very high-risk activities for 25 assigned to the each activity when put into the Primavera P6
projects.
software (2013). In other words, unit prices were considered as 1
Project L Matrix method Fine-Kinney method and risk scores of the activities were assigned as if it was an
number amount of work in the resource assignment table in Primavera
Ratio to Ratio to total Ratio to Ratio to total
total risk number of total risk number of P6. In this way, the risk of any activity, the distribution of risks
score (%) activities (%) score (%) activities (%) throughout the project and the time period with highest risk level
1 12.5 14.2 71 29.7 in the project could be seen in the scheduling tables and project
2 12.7 13.9 72.7 27.1 schedules. Utilization of the Fine-Kinney method and L Matrix
3 14.8 19 71.2 36.2
method is very advantageous for safety experts, especially in the
4 18.5 26.7 68 30
5 12.4 14.6 67.5 28.7 pre-construction phase as it allows one to focus on managerial
6 15.6 21 73.5 30 tools and decision-making about materials, equipment, construc-
7 32 72.2 78.8 80.6 tion methods, assembling of work teams, and coordination of sub-
8 23.6 42.9 74.2 84.8 contractors to reduce safety risks throughout the project.
9 6.2 6.3 63.2 14.6
10 7.2 6.9 64.8 18.5
11 19.7 23.7 73.2 37.6 3.2.3. Determination of protection measures, mitigation and
12 12.3 13.2 73.5 33.3 abatement techniques
13 20.3 36.8 53.6 47.4
14 14 20.7 75.3 48.9
Generally the safety experts on site define the risks for each
15 26.9 62.9 72.9 68.6 construction work item and determine the mitigation and abate-
16 18.4 24 79.6 48 ment measure to prevent the possible accident. It is important to
17 42.1 89.8 80.3 92.5 note that many safety measures that need to be performed should
18 16 16.5 76.9 32.2
be thought as a part of the construction job rather than as separate
19 27.2 46.8 76.2 60.3
20 16.4 22.4 69.6 39.5 safety measures and/or expenditure. For example, for excavation
21 19 26.2 81.7 45.5 work, shield, shoring, sloping and other techniques are part of
22 17.2 25.4 81.3 46.3 the job itself rather than separate safety measures. According to
23 24.3 36.6 85.3 55.6 regulations, specifications and standards, it is necessary to follow
24 23.6 35.7 84.8 55.2
25 22.5 33.7 84.2 52.8
the rules as a part of proper engineering principles. Therefore, in
this study all these types of safety measures were not calculated
Ave 19 30.1 74.1 45.7
and not included in the ‘‘cost of safety’’. Table 4 was given as an
example for both risk assessment methods (i.e., L Matrix Method
and Fine-Kinney Method). For the given example for excavation
‘‘cave-in’’ hazards. In comparison, it is impossible to specify these
(and all other job items), PPE’s (in this case helmet and safety har-
hazards using the L Matrix Method. The L Matrix Method also does
nesses) and collective protective equipment (in this case fences,
not allow the user to see differences in hazards. This is an impor-
guardrails, safety signs) are depicted in the last two columns of
tant consideration for eliminating the risks as risks are put in order
Table 4.
according to their risk scores and safety measures are taken from
the top-to-bottom in the risk scores list.
The average rate of the sum of the very high risks to total risks 3.3. Safety cost estimation
was found to be 0.19 in the L Matrix analysis; however, the same
average rate was found to be 74.1% in the Fine-Kinney analysis Cost estimating is one of the most important steps in project
(see Table 3). According to the L Matrix Method, the ratio of sum management especially during the bid and award phase. A cost
of the very high-risk activities to total number of activities was estimate establishes the baseline of project costs at different stages
found to be 30.1%; the same ratio was approximately 45.7% in of the project. For the contractor, a bid estimate is submitted to the
Fine-Kinney. If these two methods are compared, the Fine-Kinney owner either for competitive bidding or negotiation, and consists
Method identifies more activities in the region of very high risk of direct construction costs including field supervision, markup
for the 25 residential concrete construction projects. Since the to cover general overhead and profit. The direct costs of construc-
Fine-Kinney method has some difficulties in daily practice (i.e., dif- tion for bid estimates is usually derived from a combination of sub-
ficulty in accessing accident frequency values), it is generally pre- contractor quotations, quantity takeoff and required construction
ferred for use in the planning stage of construction projects in technology/procedures; but many times safety costs are not taken
order to develop an OHS management system. In contrast, the L into consideration. A cost estimate at a given stage of a project
Matrix method is easier to use for occupational safety experts development represents a prediction on the basis of available data
and also site engineers and technicians during construction pro- performed by a cost engineer or estimator; and gathering more
jects on a day-to-day basis. data provides better estimation.
After completion of the risk assessments, calculated risk scores
were used to obtain the risk distributions throughout the projects. 3.3.1. Cost of construction activities
In order to acquire risk distribution, a risk score for each work item Since the ultimate aim of the paper is to give an approximate
was calculated using both the L Matrix and Fine-Kinney methods, cost of safety to contractors before the bid and award phase or
 6 G.E. Gurcanli et al. / Safety Science 80 (2015) 1–12

before the construction phase begins, project costs including labor

and material costs, service and consultancy costs for mechanical,

Caution Signs, Rearrangement

of excavation site with cover
electrical systems as well as architectural and structural services,

Caution Signs, Safety Strips

costs for supervision and finally general expenditures for construc-
tion site facilities were calculated.
Direct and indirect labor is taken into consideration when
calculating labor costs. In order to calculate direct labor cost, it is

Caution Signs
important to determine how many hours each employee works
for a work item. Payroll information and work hours information
Ladder

plates
CPM

is evaluated together to determine direct labor costs of each work

–
–

item (Bozkurt, 2010). Therefore, daily labor payroll information

PPE

PPE

was used. In comparison, indirect labor costs were distributed to

–

–
the work items by using an ABC method as overhead.
Safety precautions to be taken in

The costs of all construction activities were calculated using the

ABC system. For these calculations, the man-hour values men-
excavation and underground

tioned in the previous section were used. The direct cost of each
work item was determined from market research. Tenders about
the projects were received from subcontractors for each work item
or cost data were used with the permission of the companies. This
study was conducted from the perspective of the main contractors
Measures

and assumes that the main contractors will take all safety
works

measures on site. Additionally, the following assumptions were

used in the calculations:
Average

 The notification published in 24 April 2013 by the Ministry of

High

High

High
level
Very

Very
high

high
Risk

Environment and Urban Planning was used for calculation of

approximate structure cost.
score

252
2400

270

120

540
200
Risk

 In absence of current prices, unit prices of the previous year

were used by increasing 5%.
Exposure

 Construction site overhead was taken as 5% of the total cost.

 For Architectural, Structural, Mechanical Design, Review and
Engineering Fee of the projects, ‘‘Project and Technical
6

6
3

Responsibility Fee Rate 2012’’ (promulgated by the Ministry of

Severity
Fine Kinney method

Environment and Urban Planning) was taken into account.

15

15
40

7
40

100

 Chamber of Electric Engineers Construction Engineering

Services and Minimum Wage Project Tender Documents
Probability

Charges for 2012 data were used when calculating the electric-
ity costs.
3

6
1

6
10

 In calculating Construction Supervision Fee, the value of

Excavation risk assessment for building construction using L Matrix and Fine-Kinney method.

‘‘Service Cost Rate’’ was taken assuming that the construction

Average

Average

Average

period is 1 year. These assumptions were taken into account

High

High
level
Very

Very
Risk

Low

in cost calculations of 25 building projects.

score
Risk

3.3.2. Cost of safety expenses

5

20
25

12

12

12

After calculating the cost of construction activities, safety costs

Severity

for personal protective equipment (PPE), collective protective mea-

sures (CPM), consultancy and training were determined. Firstly,
L Matrix method

4
5

market research was performed to estimate the safety cost for each
item. In this stage, not only were costs for PPEs taken into consid-
Likelihood

eration, but also the costs of collective protective equipment such

as fences, guardrails, stairs, roof stairs and other items identified
5

3
1

from the market survey. After first stage, PPE packages were estab-
Equipment accidents (roll over,

lished for each worker working at different jobs throughout the

struck by a moving vehicle
Building/Structure collapse

project. For example, for workers that work in excavation work,

Contact with underground
Struck by a falling object
Fall into the excavation

the package was titled as PPE-EXC (PPE for excavation worker or

(especially neighbor

PPE-Form - PPE for formwork workers) and so on. Each package

electricity etc. Lines

consists of a different combination of PPEs. In Table 5, the details

of a particular construction job, quantities, man-hours, unit price
structures)

for PPE-Package and total PPE cost for that job are provided.
Cave-ins

Job details were explained according to Ministry of

Risks

etc.)

Environment and Urban Planning Unit Prices and Definitions

Handbook. The daily work time was assumed as 9 h and the num-
Excavation

ber of working days was found by dividing total work hours by the
daily work time. In order to obtain the number of workers, work
Table 4

Job

days were divided by the number of days work will be completed.

In some small-scale construction projects, the number of workers
 G.E. Gurcanli et al. / Safety Science 80 (2015) 1–12 7

Table 5
Example of Personal Protective Equipment (PPE) cost calculation for excavation work.

Excavation Amount Unit Man Total Days Duration Number Minimum Maximum PPE Total PPE
hour working of work of number of number Package Price
hour worker worker of worker Price (USD) (USD)
Sand and gravel paving 10.08 m3 1.2 12.09 1.34 10 1 3 3 15.28 45.84
Soil Excation by Hand 330.8 m3 0.25 82.71 9.19 1 3
(Hard and Soft Soil)
Soil Excavation by Machine 28.94 m3 0.25 7.24 0.8 1 3
(Hard and Soft Soil)
Free Excavations by machine 53.75 m3 0.25 13.44 1.49 1 3
Blockage with Hearthstone 14.82 m3 6 88.9 9.88 1 3
Manually Soil Compaction 125.25 m3 1 125.25 13.92 2 3

was less than three for some specific jobs. In these cases, the num- guardrails. All these decision-making processes need hazard anal-
ber of workers was determined as three (minimum number of ysis and risk assessment scores performed by safety experts with
workers in PPE tables) according to ergonomics and feasibility of the aid of Table 4. Therefore, it is very important to collaborate
a project. PPE module costs were found by multiplying the number with safety experts before the bid and award phase to estimate
of persons working on the relevant work item by PPE module price safety expenditures for the contractors. The level of risk for each
for the same work item. Finally, the PPE costs of all work items construction activity determines the level of safety expenditure
were summed and the total PPE cost for the project was achieved. to eliminate that hazard.
In Table 6, PPE packages for different occupations are shown. After PPEs and CPMs, consultancy and training costs were calcu-
One point needs special attention because some repeating activities lated. In the study, it was assumed that a safety consultancy and
may possibly be ‘‘double-counted’’. In order to prevent this mistake, training service was provided by an expert once every two weeks
the workmanship of each work item was assigned to activities with throughout the project for a fixed 12 months project duration
a particular coding system. For unit price analysis in the last column and was calculated as 3675 USD data based on market research
of Table 6, the most expensive and high-quality equipment (in (from consultant firms). Safety Consultancy, training and expertise
other words equipment not only appropriate to required standards, cost (STCost) is the minimum price and includes auditing and
but also preferred by the safety professionals) were preferred rather training.
than the average price in the market. Finally, the safety cost of a building project can be expressed as
Hazard analyses and risk assessment scores were used to deter- follows:
mine collective protection measures and calculation of associated
SafetyCost ¼ PPECost þ CPECost þ STCost ð1Þ
costs for related construction activity. When calculating the cost
of safety measures, the overall project was examined to determine Xn
PPECost ¼ ðPPEi  NÞ ðSee Table 6 for PPEi ; n ¼ 14Þ ð2Þ
mitigation and abatement techniques, equipment and tools with i¼1
the aid of architectural and technical drawings, bill of quantities, Xn
specifications and construction schedule. It is not preferable to CPECost ¼ Ci ðSee Table 10; n ¼ 6Þ ð3Þ
divide the cost of collective protective equipment for each con- i¼1

struction job (item or sub-item) for two reasons; firstly, it is very
difficult to divide collective safety measures, and secondly, many
construction activities or many hazards, which are a result of the
nature of those jobs, can be eliminated or associated risks can be
mitigated together. For example, floor openings have risks of fall
from height and falling objects, and the installation of guardrails
is not only necessary for jobs on the slabs, but also necessary and
required when electrical and mechanical installations while work-
ers working on the floors. It is also impossible to divide the instal-
lation of fencing. For example, one way to clearly delineate the
boundaries of a construction site is by erecting a fence around
the perimeter of the area under construction. By ensuring that
work is carried out within well-defined borders, the public can
be kept safe from potentially life-threatening accidents.
Furthermore, construction site workers will have a secure area in
which to conduct their work. The installation of fences can prevent
or reduce many risks (due to excavation, heavy equipment work
and so on) and should be erected at project initiation. Therefore,
especially thru the aid of the architectural and technical drawings
and the project schedule, safety items such as safety signs, fences,
guardrails, cover sheets, safety nets, first aid, fire protection tools
and so on were determined and associated costs were calculated
for 25 building projects. Table 7 provides the Collective
Protection Measures and related techniques.
In some cases, it is necessary to prevent people from entering
the working radius of a crane thru the use of fences or in some
cases with safety tape. Floor openings can easily covered with
wooden cover, timber, etc., but sometimes it is necessary to install
STCost ¼ 12  2  1  STDaily
ðSTD daily ¼ 153:13 USD from market researchÞ ð4Þ
3.3.3. Distribution of safety costs during projects
As mentioned in the assessment of risk scores section, risk scores
calculated with the L Matrix and Fine-Kinney methods were consid-
ered as resources and these resources were assigned to each activ-
ity. Furthermore, project schedules were prepared using Primavera
P6 software and then the distributions of risks throughout the pro-
jects were obtained. In other words, the risk of each activity, the dis-
tribution of risks throughout the project and the most risky time
period in the projects can be seen in the project schedules.
Additionally, with the help of these distributions, one can create
safety measures implementation plans. While preparing the safety
measures implementation plans, ‘‘very high’’ and ‘‘high-risk’’ activ-
ities should be taken into consideration. In this manner, one can
have the safety cost of a project and organize the project budget
before the project begins by focusing on risk distributions. The dis-
tribution graphs of risk scores that were assigned to the activities
were prepared for each project. Figs. 3 and 4 are examples for risk
distributions for Project 4 estimated by the L Matrix and
Fine-Kinney Methods.
4. Results
In accordance with the performed risk assessments, safety mea-
sures were determined and the share of the safety cost to the total
8 G.E. Gurcanli et al. / Safety Science 80 (2015) 1–12

Table 6
PPE packages for different occupations.

PPE Helmet Goggle Dust mask Face Protective Reflective Safety Safety-toe High Gloves Cost
respirator Shilld clothes work vests harness protective boots (USD)
footwear
Standards CE EN CE EN CE EN CE EN CE EN CE EN CE EN CE EN CE EN
397 166 149 166 471 361 345 345 388
PPE1 PPE-EXC Exc.       15.28
Worker
PPE2 PPE-FRM Formwork       47.66
Worker
PPE3 PPE-IRW Iron Worker      47.23
PPE4 PPE-CON Concrete Worker       49.86
PPE5 PPE-ROF Roof Worker       47.66
PPE6 PPE-BRK Excavation Worker      47.23
PPE7 PPE-PPT Bricklayer       49.86
PPE8 PPE-ELC Painter, Plaster     23.55
PPE9 PPE-MEK Electrician     23.55
PPE10 PPE-FLR Mechanics, Plumber     24.34
PPE11 PPE-WEL Floor Jobs Worker       48.45
PPE12 PPE-CAR Welder       24.76
PPE13 PPE-UNS Unskilled     23.55
PPE14 PPE-DWI Door-Window ins.     23.55

Table 7
Collective protection techniques, equipment and tools.

Required measure Techniques, equipment or tools Unit/Unit Total Total

required price quantity cost
C1 Protect workers and public from the hazards of site and excavation Fencing around the site
Standard warning signs
Safety tapes
C2 Protect workers from falls into floor openings, hoist areas and slab edges Guardrails
Covering with Timbers
Safety Nets
Safety Tapes
C3 Prevent workers or public to enter the working radius of cranes, hoists etc. Fences
Safety Tapes
C4Prevent workers or public from falling objects Fences
Safety Nets
Guardrails
C5Prevention of tile breaks, fall from edge, fall from skylights Slide guards on the roof
Timbers on skylights
Roof ladders
C6 First Aid and Fire Protection First Aid Tool Boxes
Fire Protection Tools

700 35000
Total Risk Total Risk
600 30000
Total Risk Score
Total Risk Score

High Risk High Risk

500 25000
Very High Risk
400 20000 Very High Risk

300 15000

200 10000
5000
100
0
0

Project Duration
Project Duration
Fig. 4. Risk Score Distribution of Project 4 with Fine-Kinney Method.
Fig. 3. Risk Score Distribution of Project 4 with L Matrix Method.

construction cost was found. The total construction cost includes
material-labor costs, project service costs (reinforced concrete,
architectural design, mechanical design, electrical design),
building inspection service cost, site overhead and occupational
safety costs. Safety cost comprises personal protective equipment
costs, collective protective measure costs and consultancy/training
costs.
4.1. Results of cost vs. total area
Table 8 provides summarized information of all cost items cal-
culated in this study. The share of safety expenditures of total pro-
ject cost should be especially underlined, because this calculation
will be important in the bidding phase for contractors. Another
important aspect of Table 8 is that it shows the share of safety costs
 G.E. Gurcanli et al. / Safety Science 80 (2015) 1–12 9

Table 8
Cost Summary information (USD) and cost rates of 25 projects.

Project Construction Project cost Safety Total cost Project cost per Total Safety Safety cost per Safety cost per C.
number area (m2) (Except Safety cost C. Area (m2) working cost/total cost man-hour (USD) Area (USD/m2)
Cost) hour (%)
1 2618 1,171,758 21,205 1,192,963 865.8 37,792 1.78 0.56 8.11
2 9750 2,038,263 27,896 2,066,159 402.6 71,072 1.35 0.39 2.84
3 2500 381,438 11.49 392,928 298.6 11,264 2.92 1.02 4.58
4 1350 386,333 13,836 400,169 563.2 10,352 3.46 1.34 10.26
5 1260 455,286 18,776 474,062 714.9 46,184 3.96 0.41 14.89
6 11,200 2,891,402 33,899 2,925,301 496.3 95,640 1.16 0.35 3.05
7 2500 937,883 12,549 950,433 722.3 10,800 1.32 1.16 5
8 11,000 18,458,607 143,947 18,602,554 3213.20 118,002 0.77 1.22 13.11
9 3400 833,645 11,179 844,824 472.1 9056 1.32 1.24 3.26
10 3700 1,089,413 21,367 1,110,780 300.2 34,384 1.92 0.62 5.79
11 530 171,688 7728 179,416 643.2 6728 4.31 1.15 14.58
12 7330 1,997,605 31,649 2,029,254 526 51,368 1.56 0.62 4.32
13 2000 717,087 10,494 727,582 691.2 28,728 1.44 0.36 5.26
14 300 111,521 7778 119,299 755.6 9176 6.52 0.85 25.95
15 3150 729,174 14,491 743,665 448.6 19,544 1.95 0.74 4.58
16 230 107,012 6.3 113,312 936.1 15,064 5.56 0.42 27.37
17 12,000 2,699,819 84,166 2,783,985 440.8 23,728 3.02 3.55 5.26
18 500 263,385 6852 270,236 1026.90 5072 2.54 1.35 13.68
19 118,200 28,666,266 155,158 28,821,424 463.3 166,128 0.54 0.93 1.32
20 1200 239,692 8831 248,523 393.5 7664 3.55 1.15 7.37
21 2255 813,805 16,451 830,256 699.6 25,880 1.98 0.64 7.32
22 439 172,102 10,858 182,961 791.9 20,992 5.93 0.52 24.74
23 4345 1,414,536 19,962 1,434,498 627.3 22,760 1.39 0.88 4.58
24 3106 1,150,106 17,592 1,167,698 714.3 18,832 1.51 0.93 5.68
25 4042 1,476,310 19,042 1,495,352 702.9 25,952 1.27 0.73 4.74
Median 1.92 0.85 5.68

per unit of man-hour; because man-hour data are extremely
important and they are the smallest units used for the planning
of managerial activities in modern construction project manage-
ment. It is also worth mentioning that determination of safety
costs per unit area offers great advantages in bidding, budgeting,
planning and other construction management activities. The calcu-
lations indicate that the percentage of safety cost to the total con-
struction cost is 1.92%. In terms of man-hour costs, 0.85 USD safety
expenditure should be spent (with the 2013 rates). Safety cost per
unit construction area was found to be approximately 5.68 USD.
Safety cost versus construction area for 24 building projects is
shown in Fig. 4 (one project regarded as an ‘‘outlier’’ due to its total
construction area being 118,200 sqm). The most appropriate trend
can be expressed by a logarithmic curve for this data set.
Regression analysis gives the relationship shown below:
y ¼ 1:16 lnðxÞ þ 11:55 R2 ¼ 0:67 ð5Þ
y represents the percentage of safety cost in total project cost and x
is the total area of the construction project. The formula gives a very
practical way to estimate the share of safety cost in total project
cost for building projects, but deficiencies must be considered (this
is discussed in the Section 5).
Analysis of Fig. 5 shows that as the construction area gets larger,
safety costs tend to decrease. However R2 is lower than expected
(see Table 9) if we compare it with the results of the former study
shown in Fig. 6 and Table 10 in the validation section. As known
that R2 is a statistical measure for data set whether far or close
to the regression curve. R2 can be basically defined as the percent-
age of explained variation in total variation and in general, the
higher the R2, the better the model fits the data. Of course R2 does

Share of Safety Cost in Total Project Cost for Recent Research (%)

7,00

6,00 y= -1.16 ln (x) + 11.55

2
R = 0.67
5,00

4,00

3,00

2,00

1,00

0,00

0,00 2000,00 4000,00 6000,00 8000,00 10000,00 12000,00

Total Construction Area
Observed
Logarithmic

Fig. 5. Variation of Safety Cost Based on Construction Area for 24 Projects larger than 2000 sqm.
10 G.E. Gurcanli et al. / Safety Science 80 (2015) 1–12

Table 9 Table 10
Model summary of recent research for logarithmic regression. Model summary of former research with logarithmic regression.

Unstandardized Standardized T Sig. Unstandardized Standardized t Sig.

coefficients coefficients coefficients coefficients
B Std. Beta B Std. Beta
error Error
ln(x) 1.160 .175 .816 6.625 .000 ln(x) 2.497 .130 .964 19.194 .000
Constant 11.554 1.365 8.462 .000 Constant 20.321 .866 23.473 .000
Sum of df Mean square F Sig. Sum of df Mean square F Sig.
squares squares
Regression 40.618 1 40.618 43.891 .000 Regression 26.117 1 26.117 368.405 .000
Residual 20.360 22 .925 Residual 1.985 28 .071
Total 60.978 23 Total 28.102 29

accidents in the construction industry generally occur in smalls

not always point out whether a regression model is adequate; in
some cases a low R-squared value may be a good percentage for
the model. However, it can be observed from the plots and
commented intuitively that R2 here is a measure of how well our
logarithmic regression model fits the data for safety cost versus
construction area.
4.2. Validation of the results
The results are compared with an earlier study (Gurcanli et al.,
2011) where the ratio of safety cost to the total construction cost
was found to be 3.62%; man-hour cost was calculated as 0.28
USD and cost per unit area was obtained as 8.08 USD. Former
research observed and focused on relatively small construction
projects and it can be intuitively said that the share of the safety
cost decreases when total construction area increases. The most
appropriate trend can be expressed by a logarithmic curve for this
data set. Fig. 6 obviously points out that as the area of construction
gets larger, the safety costs tend to decrease.
In the former study carried out by Gurcanli et al. (2011), the R2
value can be reached as 0.93 as shown in Fig. 6 and the model sum-
mary in Table 10. A logarithmic regression model fits well for this
data set due to the scope of the projects examined. All the projects
in this data set had a total project area of less than 2000 sqm, but in
the current research the majority of the construction projects were
more than 2000 or even 4000 sqm. Briefly, the logarithmic model
works well for middle- or small-scale projects according to the
results, and especially for projects whose total construction area
is less than 10,000 sqm. Recent studies (Ceylan, 2014; Social
Insurance Institution – SII Occupational Accidents Statistics
Report, 2012; Yilmaz and Tan, 2015) show that occupational
and middle-scale projects. Therefore, this study provides a refer-
ence for contractors of small- or middle-scale construction projects
and facilitates estimation of both safety and bidding costs.
If we combine two data sets, logarithmic regression analysis
gives the relationship below:
y ¼ 1:239 lnðxÞ þ 12:053 R2 ¼ 0:72 ð6Þ
where y represents the percentage of the safety cost to total project
cost and x is the total area of the construction and p < 0.001 for all
data sets. Furthermore, when we focus on projects with a less than
10,000 sqm construction area, R2 tends to reach the value 0.80
where more than fifty percent of the standard deviation can be
explained by the model. Of course the logarithmic relationship just
gives a point of view or starting point for further analyses. However
at this stage, the formula provides a very practical way to estimate
the share of safety costs in the total project cost for a building pro-
ject, especially for projects with a less than ten-thousand sqm total
construction area.
5. Discussion
As seen in Table 8, the share of the safety expenses in the total
cost of a construction project (construction plus safety) was 1.9% at
a median. 0.85 USD per man-hour is required to proceed with the
project safely and 5.68 USD should be spent for safety expenses per
unit construction area. It should be noted that as the total area of
the project increases, the share of the safety expenses decreases.
At this point, firstly it must be stated that certain constant
expenses (fixed costs) for safety were the same for all projects
(safety training and consultancy due to legal requirements).

Share of Safety Cost in Total Project Cost for Former Research (%)
6,00
y= -2.50 ln (x) + 20.32
2
R = 0.93
5,00

4,00

3,00

2,00

200,00 400,00 600,00 800,00 1000,00 1200,00 1400,00 1600,00

Total Construction Area
Observed
Logarithmic

Fig. 6. Variation of Safety Cost Based on Construction Area for 30 Projects smaller than 2000 sqm.
 G.E. Gurcanli et al. / Safety Science 80 (2015) 1–12
However, the total number of safety expert service hours may
change only with the number of workers (according to a new reg-
ulation). Moreover, some expenses such as fencing around the site
or excavation area do not depend on the total construction area,
because the area of the construction site (for instance the fencing
cost for a ten-storey building and three-storey building erected
on a same surface area) may be same.
The authors suggest that for the future research, data from the
construction sites may be analyzed with different regression meth-
ods to model the relationship between type of construction, total
area of construction and percentage of safety expenditures. The
logarithmic relationship of course just gives us a point of view;
however, the formula provides a very practical way to estimate
the share of safety costs in total project costs for building projects.
Additionally, contractors can prepare safety plans and organization
and allocate the required budget for safety measures not only for
cost control or project management, but also in order to protect
human life and ensure the safety of their workers. Due to the char-
acteristics of the construction industry, when scope is enlarged,
many different factors will affect the project cost, and this is sim-
ilar for health and safety costs. Therefore, further research may
increase sample size and focus on multi regression models or dif-
ferent methodologies such as logistic regression, fuzzy logic
approach and expert systems.
This project provides some other results important for practi-
tioners and safety experts:
Risk assessments done with the L matrix method clearly points
out periods in the schedule where the most risky activities exist.
On the other hand, the Fine-Kinney method points out the most
risky one or two activities in a much shorter time interval.
Although the risk categories of some activities were the same,
Fine-Kinney method can give very different risk scores. In other
words, since the Fine-Kinney utilizes three parameters instead of
two and larger intervals for risk categories (for instance high-risk
means, a risk score between 200 and 400 while in the L-Matrix
method it means a risk score between 15 and 20), it is possible
to evaluate and differentiate risk scores between different activi-
ties those have same risk category. Therefore, the Fine-Kinney pro-
vides a higher time and activity risk resolution, in other words
gives estimation for the ‘‘date’’ of the most hazardous activities
during a building construction project. This fact is important for
more precise risk assessment and strategy for decision makers
when determining the most urgent measures to be taken and/or
changes of working method, materials and equipment.
A common finding throughout all the risk assessments of the 25
projects was unsurprisingly that traditional construction jobs such
as ‘‘reinforced concrete’’ work items such as formwork, pouring
concrete, re-bar bending, settlement and so on are the most haz-
ardous (and risky) jobs in building construction. The most risky
work items after ‘‘reinforced concrete’’ are excavation and work
related to electrical installations, use of electrical equipment, etc.
The risks of these work items can be easily eliminated with
advance implementation of an occupational health and safety
management plan.
The results reveal that the logarithmic regression model fits
well, especially for projects with a less than 10,000 sqm construc-
tion area, where R2 tends to reach the value 0.80, and roughly
speaking more than fifty percent of the standard deviation can be
explained by the model. The formula gives a very practical way
to estimate the share of safety cost in total project cost for building
projects, especially in small/middle-scale construction projects. As
a comparison, past statistical data of construction accidents reveal
that a majority of the fatal accidents occur in small or middle-scale
construction sites, which are very common nowadays due to the
Law on the Transformation of the Areas under Disaster Risk, No:
6306 which entered into force on May 31, 2012 (commonly known
11
as Urban Renewal Act). According to this law, many residential
buildings have begun to be demolished and new projects have
started. As mentioned above, recent studies reveal that generally
workers at small- and middle-scaled building projects are the
victims of the fatal accidents. Therefore, research as well as
guidance to give information and object oriented training
programs for contractors of small- or middle-scale construction
projects is important.
It is thought that these results will shed light on resource load-
ing and leveling purposes. It is known that resource loading
describes the amounts of individual resources an existing schedule
requires during specific time periods. Resource loading provides a
general view of the demands of a project on a construction com-
pany’s resources and it is an excellent guide for early, rough project
planning. Moreover, it is also a first step attempting to reduce
excessive demands on certain resources, regardless of the specific
technique used to reduce demands (Meredith and Mantel, 2011).
If a general project management approach is translated into health
and safety resource allocation, this study will provide information
for safety professionals as well as project managers to plan limited
resources throughout the project while acknowledging that the use
of resources on a project is generally nonlinear.
6. Conclusion
This project proposed an approach for prime contractors to esti-
mate safety costs as well as its distribution throughout a construc-
tion project. Toward this purpose, hazard analysis and risk
assessment techniques were presented. Safety risks and related
costs were assessed and calculated, and a distribution of risk and
related costs were determined to facilitate effective safety plan-
ning. Safety management is one of the most prominent elements
of construction management. Additionally, activity-based risk esti-
mation by focusing on the distribution of very high and high-risk
activities throughout a project provides necessary information for
safety budgeting and planning for contractors as well as safety
experts. Therefore, for the establishment of an occupational health
and safety management system, contract requirements, project
duration, total employment, number of employees, total construc-
tion area, number and variety of work items should be considered
in initial planning and with the proposed approach safety planning
can easily be implemented. Moreover, the results of the study may
help in resource leveling attempts to minimize period-by-period
variations in resource loading by shifting tasks within their slack
allowances.
In future research, the sample size may be increased, and data
from construction sites may be analyzed with different regression
methods to model the relationship between type of construction,
total area of construction and percentage of safety expenditures.
Briefly, it is thought that completely different project characteris-
tics like duration, number of workers and similar parameters
should also be considered and multiple regression modeling or
other techniques may be used in future studies.
Acknowledgement
This work is supported by the Scientific and Technological
_
Research Council of Turkey (TÜBITAK) under the project with no.
112M033.
References
Akarca, H., 1992. The problems and solution alternatives faced on during the
elemental cost analysis applications. MSc Thesis, Istanbul Technical University,
Institute of Graduate Studies in Science and Engineering, Istanbul.
Akintoye, A., 2000. Analysis of factors influencing project cost. Constr. Manage.
Econ., 77–79, January–February
12 G.E. Gurcanli et al. / Safety Science 80 (2015) 1–12
Alkan, A.T., 2005. Faaliyet tabanlı maliyet sistemi ve bir uygulama. J. Inst. Soc. Sci.
Selcuk Univ. 13, 39–41.
Alonso, M.L., Dávila, M.P., Gámez, M.C., Munoz, T.G., 2013. The impact of health and
safety investment on construction company costs. Saf. Sci. 60, 151–159.
Aminbakhsh, S., Gunduz, M., Sonmez, R., 2013. Safety risk assessment using analytic
hierarchy process (AHP) during planning and budgeting of construction
projects. J. Saf. Res. 46, 99–105.
Arpaci, S.S., 1995. Cost estimating with simülation approach in building
construction. MSc Thesis, Istanbul Technical University, Institute of Graduate
Studies in Science and Engineering, Istanbul.
Australian and New Zealand Standard on Risk Management AS/NZS 4360. 1999.
Bozkurt, M., 2010. A system proposal to collect and evaluate actual cost data in
construction projects. PhD Thesis, Istanbul Technical University, Institute of
Graduate Studies in Science and Engineering, Istanbul.
Ceylan, H., 2014. Türkiye’de insßaat sektöründe meydana gelen isß kazalarının analizi.
Int. J. Eng. Res. Develop. 6 (1) (in Turkish).
Chalos, P., 1992. Managing Cost in Today’s Manufacturing Environment. Prentice
Hall, Englewood Cliffs, NJ.
Cheng, C.W., Leu, S.S., Lin, C.C., Chihhao, F., 2010a. Characteristic analysis of
occupational accidents at small construction enterprises. Saf. Sci. 48 (6), 698–
707.
Cheng, M.Y., Tsai, H.C., Sudjono, E., 2010b. Conceptual cost estimates using
evolutionary fuzzy hybrid neural network for projects in construction
industry. Expert Syst. Appl. 37, 4224–4231.
Ciraci, M., Ciraci, H., Sey, Y., Tiknaztepe, M., Eski, O., Cicekyuzlu, N., Usta, H., 1996.
Konutlarda Maliyet Tahmini Için _ Bir Model. Middle East Technical University
Press, Ankara.
Cook, P.J., 1982. Estimating for the General Contractor. R.S. Means Company, Inc.,
U.S.A..
Cooper, R., Kaplan, R., 1992. Activity-based systems: measuring the cost of resource
usage. Account. Horizons 6 (3), 1–13.
Cox, D., 1987. Indirect Cost Estimating for General Contractors. MorganTown, W.V.
Curran, M.W., 1989. Range estimating. Cost Eng. 31 (3), 18–26.
Everett, J.G., Frank, P.B., 1996. Costs of accidents and injuries to the construction
industry. J. Constr. Eng. Manage. 122 (2), 158–164.
Feng, Y., Zhang, S., Wu, P., 2015. Factors influencing workplace accident costs of
building projects. Saf. Sci. 72, 97–104.
Gurcanli, G.E., 2006. A risk analysis method for occupational safety in construction
sites using fuzzy sets. PhD Thesis, Istanbul Technical University, Institute of
Graduate Studies in Science and Engineering, Istanbul.
Gurcanli, G.E., Mungen, U., 2005. Fatal traffic accidents in the Turkish construction
industry. Saf. Sci. 43, 299–322.
Gurcanli, G.E., Mungen, U., 2009. An occupational safety risk analysis method at
construction sites using fuzzy sets. Int. J. Ind. Ergon. 39 (2), 371–387.
Gurcanli, G.E., Mungen, U., 2013. Analysis of construction accidents in turkey and
responsible parties. Natl. Ind. Health 51 (6), 581–595.
Gurcanli, G.E., Korkutan, N.T., Müngen, U., 2011. An approach for estimating total
cost of occupational safety for building constructions. In: The Fourth
International Conference on Construction Engineering and Pr, Sydney, Australia.
HSE, 1999. Management Guidelines for Working Together in a Contract
Environment. Report No: 6.64/291.
HSE, Hazardous Installations Directorate, 2003. HID Safety Report Assessment
Guide: Explosives. HSE.
Hughes, P., Ferrett, E., 2005. Introduction to Health and Safety in Construction.
Elsevier.
Khataie, A.H., Bulgak, A.A., Segovia, J.J., 2011. Activity based costing and
management applied in a hybrid decision support system for order
management. Decis. Support Syst. 52, 142–156.
Kim, G.H., An, S.H., Kang, K.I., 2004. Comparison of construction cost estimating
models based on regression analysis, neural networks, and case-based
reasoning. Build. Environ. 39, 1235–1242.
Kinney, G.F., Wiruth, A.D., 1976. Practical risk analysis for safety management (No.
NWC-TP-5865). Naval Weapons Center China Lake Ca.
Kouskoulas, V., Koehn, E., 1974. 2.2 Predesign cost estimating function for buildings.
Cost Modelling. E&FN Spon, pp. 172–190.
Kuruoglu, M., 2002. Ins _ ßaat Sektöründe Bilgisayar Destekli Planlama Metot ve
Örnekleri. Çağlayan Kitabevi, Istanbul.
Kuruoglu, M., Bayoglu, F.I., 2001. Yapı üretiminde adam-saat değerlerinin
belirlenmesi üzerine bir arasßtırma ve sonuçları. In: 16th Technical Conference
of Civil Engineering, Ankara.
Liu, L.Y., Pan, F., 2007. The implementation of activity-based costing in China: an
innovation action research approach. Br. Account. Rev. 39 (3), 249–264.
Meredith, Jack R., Mantel Jr., Samuel, 2011. Project Management: A Managerial
Approach. John Wiley & Sons.
Mungen, U., 1997. Employment related accidents in the turkish construction sector
and applications of occupational safety. In: 1st South African Construction
Health and Safety Conference, Johannesburg, pp. 231–241.
Panopoulos, G.D., 2003. Economic aspects of safety in construction industry – a
research program. InterConstruct 2003, 1–29.
Park, C.S., Kim, H.J., 2012. A framework for construction safety management and
visualization system. Autom. Constr. 33, 95–103.
Pilcher, R., 1966. Principles of Construction Management for Engineers and
Managers. McGRAW Hill, London.
Project Management Bady Of Knowledge (PMBOK), 2008. Project Management
Institute, Newtown Square, Pennsylvania, USA.
Polat, D., 2004. A model to establish a database for pre-design phase-building cost
estimate in Turkey. PhD Thesis, Istanbul Technical University, Institute of
Graduate Studies in Science and Engineering, Istanbul.
Primavera P6 Software, 2013. <http://www.oracle.com/us/products/applications/
primavera/p6-professional-project-management/overview/index.html> (retrieved
04.02.13).
Raafat, H., 1995. Machinery Safety: The Risk Based Approach. Practical Guidelines
on Risk Assessment. Standards and Legislation Technical Communications
(Publishing) Ltd..
Rausand, M., 2011. Risk Assessment Theory, Methods, and Applications. Wiley, New
Jersey.
Ríos-Manríquez, M., Colomina, C.I.M., Pastor, M.L.R.V., 2014. Is the activity based
costing system a viable instrument for small and medium enterprises? The case
of Mexico. Estud. Gerenciales 30 (132), 220–232.
Rzvi, T., Elnathan, D., 1999. Activity based costing for projects. Int. J. Project Manage.
17 (1), 61–67.
Seyyar, B., 2000. Bina tasarım sürecinde bilgisayar destekli maliyet tahmin
modelleri. MSc Thesis, Istanbul Technical University, Institute of Graduate
Studies in Science and Engineering, Istanbul.
SII. Social Insurance Institution Occupational Accidents Statistics Report, 2012.
<http://www.isteguvenlik.tc/SGK2012IsKazaIstatistik.pdf> (retrieved 10.12.13,
in Turkish).
Sii, H.S., Wang, J., 2002. Safety Assessment of FPSO’s: The Process of Modeling
System Safety and Case Studies. Report of The Project – The Application of
Approximate Reasoning Methodologies to Offshore Engineering Design. EPSRC,
GR/R30624 and GR/R32413, Liverpool John Moores University, U.K.
Sii, H.S., Ruxton, T., Wang, J., 2001. A fuzzy-logic-based approach to qualitative
safety modeling for marine systems. Reliab. Eng. Syst. Saf. 73, 19–34.
Singh, S., 1990. Cost model for reinforced concrete beam and slab structures in
building. J. Constr. Eng. Manage. 116 (1), 54–679.
Sousa, V., Almeida, N.M., Dias, L.A., 2014. Risk-based management of occupational
safety and health in the construction industry – Part 1: Background knowledge.
Saf. Sci. 66, 75–86.
Tam, C.M., Tong, T.K.L., Chiu, G.C.W., Fung, I.W.H., 2002. Non-structural fuzzy
decision support system for evaluation of construction safety management
system. Int. J. Project Manage. 20, 303–313.
Tan O., 1999. Is _ß kazası olusßmadan alınacak önlemlerin maliyeti ile isß kazası
olusßtuktan sonraki harcama maliyetlerinin analizi ve karsßılasßtırılması. MSc
Thesis, Istanbul Technical University, Institute of Graduate Studies in Science
and Engineering, Istanbul.
Tappura, S., Sievänen, M., Heikkilä, J., Jussila, A., Nenonen, N., 2015. A management
accounting perspective on safety. Saf. Sci. 71, 151–159.
Turkish Statistical Institute (TUIK), 2013. Construction Labor Input Indices and
Percentage Changes Report of 2013. <http://www.tuik.gov.tr/UstMenu.
do?metod=temelist> (retrieved 15.01.2015).
Tweeddale, M., 1997. Risk Management Handbook for the Mining Industry. New
South Wales Department of Mineral Resources, Sydney.
Usta, S.H., 1994. Avan proje evresinde bina maliyetinin bina elemanlarına dayalı
olarak hesaplanması. MSc Thesis, Istanbul Technical University, Institute of
Graduate Studies in Science and Engineering, Istanbul.
Waehrer, G.M., Dong, X.S., Miller, T., Haile, E., Men, Y., 2007. Costs of occupational
injuries in construction in the united states. Accid. Anal. Prev. 39, 1258–1266.
Wang, J., 1997. A subjective methodology for safety analysis of safety requirements
specifications. IEEE Trans. Fuzzy Syst. 5 (3), 418–430.
Wilson, A.J., 2005. Experiments in Probabilistic Cost Modeling. Cost Modeling. E&FN
Spon, London.
Yilmaz, F., Tan, O., 2015. Bir insßaat sßantiyesinde isß kazalarinin neden olduğu isßgünü
kayiplarinin isßverene maliyetinin belirlenmesi. Int. J. Econ. Admin. Stud. 7 (14)
(in Turkish).