Risk perception of construction equipment operators on construction sites of Turkey

International Journal of Industrial Ergonomics 46 (2015) 59e68

Contents lists available at ScienceDirect

International Journal of Industrial Ergonomics journal homepage: www.elsevier.com/locate/ergon

Risk perception of construction equipment operators on construction sites of Turkey G.E. Gürcanlı a, S. Baradan b, *, M. Uzun a a b

Technical University of Istanbul, Department of Civil Engineering, Division of Construction Management, Ayazaga Campus, Maslak, 34469 Istanbul, Turkey _ Ege University, Department of Civil Engineering, Division of Construction Management, Bornova, Izmir, Turkey

a r t i c l e i n f o

a b s t r a c t

Article history: Received 28 March 2014 Received in revised form 10 November 2014 Accepted 14 December 2014 Available online 23 January 2015

Turkey has been an attractive country for construction industry in the last decade. Many large-scale construction projects, which have been realized by both international and local construction firms, helped the economy and provided employment opportunities for many. At the same time, many construction workers have been losing their lives on construction sites, which involve the usage of heavy equipment on a daily basis. Past research studies suggest that employee participation and their perception of safety risks could be valuable for determining and eliminating hazards on construction site. Therefore, this study aimed to determine and evaluate the risk perception of construction equipment operators in Turkey. The study is mainly based on a questionnaire survey performed in 51 construction projects that involved 198 heavy equipment operators. A statistical analysis was first performed on the results of the survey to observe the frequency distribution of parameters, such as safety and health training, using flagger, experience, type of equipment, working conditions and other project related data. Then, statistical methods such as, t-test, ANOVA analysis, KruskalleWallis one way analysis of variance, and ManneWhitney U test were performed to seek statistically meaningful differences in risk perception of operator groups with different attributes. Results revealed the importance of safety and health training as well as working with an assistant, such as a flagger. It was observed that operators who took safety and health training and operators who worked with flaggers perceived risk differently than others. It was also found that the project type influences the risk perception of equipment operators due to diversity of construction equipment activities performed, as well as number of incidents occurred in those projects. Relevance to industry: The authors expect this research to lead to discussion and further research on risk assessment for construction industry. The risk assessment findings of this study, in particular, could help the safety professionals detect possible unforeseen risks and design safety and health plans for construction sites that require usage of heavy equipment on a daily basis. Heavy equipment manufacturers could also devise a similar research that involves operators’ risk perception to design more ergonomic and safe equipment. © 2014 Elsevier B.V. All rights reserved.

Keywords: Construction equipment Heavy equipment operator Risk perception Occupational safety and health

1. Introduction Utilization of heavy equipment on construction sites is on the rise for the last three decades in Turkey due to unique and complex construction projects that feature creative, ergonomic and effective design, and the ever-increasing demand for residential projects. Heavy equipment unquestionably provides efficiency and speed in

* Corresponding author. E-mail address: [email protected] (S. Baradan). http://dx.doi.org/10.1016/j.ergon.2014.12.004 0169-8141/© 2014 Elsevier B.V. All rights reserved.

construction projects, but at the same time it creates a hazardous work environment for all workers who are directly or indirectly involved in heavy equipment operation. Workers that are directly involved mainly consists of: operators who are specially trained to drive and operate the vehicle, and co-operators (flaggers, signal persons and spotters) who direct traffic through a construction site and help to backup vehicles using gestures, signs or flags. Workers that are indirectly involved are usually on-foot construction workers who are engaged in other construction activities in the same construction site. Variety of fatal hazards exist on heavy

60

G.E. Gürcanlı et al. / International Journal of Industrial Ergonomics 46 (2015) 59e68

Table 1 Comparison of fatal accidents and fatal incidents rates between Turkey, EU-15, U.S.A. and P.R of China, 2000e2010. Fatal construction accidents and share (%) in total, fatal incidence rate Years

Turkeya

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

379 341 319 274 263 290 397 359 297 156 475

(32.3) (33.8) (36.6) (33.8) (31.3) (27.1) (24.9) (34.4) (34.3) (13.3) (33.1)

EU-15b 49.8 50.0 44.7 39.9 35.0 31.1 33.5 28.8 24.0 12.7 33.2

1285 1223 1198 1248 1119 1065 1151 1019 831 774 693

(24.3) (24.6) (24.8) (26.7) (25.4) (21.8) (23.4) (28.2) (26.2) (26.1) (22.6)

U.S.A.c 11.3 10.6 10.4 10.6 9.6 8.8 9.0 7.5 N/Ae 6.0 5.5

1154 1225 1121 1131 1234 1186 1239 1204 969 834 774

(19.5) (20.8) (20.3) (20.4) (21.4) (20.8) (21.2) (20.9) (19.1) (18.3) (16.5)

P.R of Chinad 16.8 17.4 16.0 16.1 16.9 15.6 15.6 15.3 13.0 13.4 13.4

3778 4056 4538 4522 4274 4202 4157 4121 4055 4017 3945

(32.3) (32.3) (30.4) (30.1) (30.5) (30.1) (29.8) (29.7) (29.4) (29.1) (28.4)

50.8 55.3 60.0 58.5 55.0 49.2 45.7 42.9 41.8 40.5 39.2

a

Social Insurance Institution General Directory and Statistical Institution of Turkey Statistics. http://epp.eurostat.ec.europa.eu/portal/page/portal/statistics/search_database; http://laborsta.ilo.org/, In 2008, EU statistical classification changed and Norway was excluded from the database. Therefore, the figures in the last 3 rows do not contain Norway data. c http://www.bls.gov/iif/oshwc/cfoi/worker_memorial_data.htm#fatal_injuries.xls, http://www.bls.gov/iif/oshcfoiarchive.htm#rates (calculated as the number of new cases of injury (fatal) during the calendar year divided by the number of workers in the reference group during the year, multiplied by 100,000. US figures were converted). d http://laborsta.ilo.org/, http://www.stats.gov.cn/english. e Data for Greece was not available due to the change in classification system in 2008. b

construction sites that harbor such workers, mainly physical hazards such as struck by vehicle, struck by objects, rollovers and caught in/between. In Turkey, not only heavy equipment but also every loadingunloading operation must be performed by the aid of a flagger or spotter as mandated by safety and health regulations for both general industry (article 379, 1974) and construction works (article 17, 1974). After the arrival of new “Safety and Health Law, 2013” in Turkey, all the regulations were updated. Health and Safety in Construction Works Regulation (Article 52 and 54) obliges a trained flagger in equipment operations, excavation works and material hauling, handling and moving operations. Health and Safety Symbols and Signals Regulation (Article 2.1e2.6) describes principal duties and responsibilities, on-site tasks and garments of flaggers. Even though it is clear that working with a flagger is required by the legislation in Turkey, using flaggers is not a common practice in Turkish construction sites yet due to the regulations being rather new. Statistics also reveal that heavy equipment fatalities rank fourth in Turkish construction industry and 7.3% of the workers who lost their lives on construction sites are equipment operators (4%) and co-operators (3.3%) (Gurcanli et al., 2008). According to U.S. Bureau of Labor Statistics, construction equipment operators' deaths comprised 4 and 6 percent of fatalities in construction in the years of 2008 and 2009 respectively, which are very similar to the figures in Turkey. If Turkey and European Union (EU) countries are compared, Turkey's fatal incidence rate in the construction industry is almost 4 to 6 times of EU's. Table 1 shows the grim safety and health record of Turkey since 2000, when compared with EU-151 countries, United States of America and Public Republic of China. Implementing a comprehensive safety and health program or management system is usually considered as the key to systematically decreasing injuries and deaths in such hazardous work sites. Participation of workers in determining and mitigating occupational risks is recognized as an invaluable component in such safety management systems. Particularly, work activities requiring team work and communication, as established between heavy equipment operators and co-operators (flaggers and spotters), would

1 For consistency purposes only the 15 countries who were part of the European Union since 2000 were included in the data. Countries who are a member of EU since 2007 were excluded.

benefit from a safety approach that embraces workers' participation and their perception of risk. Research topics, such as workers participation and risk perception have been areas of interest for several safety researchers in the past. According to Vredenburgh (2002), Worker participation (or employee involvement) is a behavioral-oriented technique that involves individuals or groups in the upward communication flow and decision-making process within the organization. The amount of participation can range from no participation, where the supervisor makes all decisions, to full participation, where everyone connected with, or affected by the decision is involved. Dedobbeleer and Beland (1991) state that management's safety concerns and actions should be highly publicized among the workers and that “workers' involvement” can include participation in the development of safety programs, conduct of safety audits, and identification of solutions. Meridian Research Report (1994) highlights some of the examples of meaningful employee participation as participating in the development of safety programs and in workplace inspections, having a membership on joint labor/ management committees, and actively getting involved in accident and “near-miss” investigations. Hallowell (2008) states that risk perceptions must be carefully solicited in a standardized fashion to quantify and compare among risk tolerances (i.e. an individual's subjective assessment of acceptable risk). Risk perception is defined as the subjective judgment that one makes about the frequency and severity of particular risks. Typically, these values are obtained by questioning individuals about specific risk scenarios and aggregating the data. According to Starr (1969), some individuals may rate the same environment differently but the statistical aggregation of risk perceptions represents actual circumstances when the major sources of bias have been minimized. Hallowell's article presents a good literature research on risk studies and risk perception in construction industry and he cites his doctorate dissertation, which presents that construction workers are capable of identifying and rating occupational safety and health risks with a reasonable level of accuracy and he puts forward an objective method of quantifying safety risk using frequency estimates defined as incidents per 200,000 worker-hours (w-h) and severity based on impact of the accident to the worker (Hallowell, 2010). In Zohar's study (2000), there are specific findings about the measure of employees' perceptions on the relative importance of safe conduct in their occupational behavior. According to Zohar, their occupational behavior

G.E. Gürcanlı et al. / International Journal of Industrial Ergonomics 46 (2015) 59e68

61

Table 2 The ways of occurrence of fatal traffic and equipment accidents on construction sites (Gurcanli et al., 2008). The way of occurrences

Number of incidences for traffic accidents

%

Number of incidences for equipment cases

%

1. Crushed/run-over by highway vehicle/construction equipment 2. Crushed/run-over by train 3. Crushed/run-over by maneuvering vehicle/equipment 4. Crushed/run-over by vehicles entering the site 5. Motor vehicle/construction equipment fall-over 6. Fall from moving vehicle/equipment while getting on/off 7. Fall from moving vehicle/equipment 8. Motor vehicle traffic accident involving collision with other vehicle 9. Caught between the equipment elements 10. Material falls (rocks etc.) on the equipment 11. Electric shock by contacting the high voltage electric lines 12. Unknown way of occurrence or other Total cases

46 32 21 11 17 16 15 5 e e e 5 168

27.4 19.0 12.5 6.5 10.1 9.5 8.9 3.0 e e e 3.0

40 e e e 71 4 11 e 38 7 35 e 206

19.4 e e e 34.5 1.9 5.3 e 18.4 3.4 17.0 e

reflects the safety climate in a given company and serves as a proxy for site safety in general. Aneziris et al. (2008) focus on risk assessment for crane activities and suggest a logical model for quantifying occupational risk in case of collapsing or overturning cranes, falling loads or falling objects struck by cranes, using quantification of bowties to obtain primarily probabilities of occurrences. State-of-the-art research on heavy equipment safety showed that only a few studies in the past dealt with analysis of construction accidents that heavy equipment operators were exposed to. Uzun (2012), in his Master of Science thesis, focuses on the risks associated with equipment usage and Bhide (2006) gives broad information on 513 accidents which OSHA has recorded over the years 1984e2000. Gurcanli et al. (2008) analyzed traffic and heavy equipment accidents in the Turkish construction industry. Findings of this study are displayed in Table 2, which shows the frequency distribution of incidents, categorized by their ways of occurrence. Moreover, the same research reveals the distribution of trades for construction equipment accidents on construction sites, where equipment operators (27.3%) and co-operators (22.7%) constitute more than half of recorded cases. Even though, Chamber of Mechanical Engineers and private training institutions have been offering courses for flaggers in Turkey, majority of the flaggers do not take special training and in most cases unskilled workers (in addition to their usual tasks) are deployed to assist the operators on construction sites. Unskilled so-called flaggers create new risks and even increase existing risks, while heavy construction equipment activities are being performed. Therefore, it is no surprise that operators, co-operators, and unskilled workers around the working radius of the construction equipment are under the risk of equipment related activities. In another study, McCann (2006) focuses on excavation industry solely and examined 256 accidents. Causes of equipment and truck related deaths were found as follows: struck by vehicle (30%), struck by objects (24%), rollovers (23%), caught in/between (12%) and others (11%). McCann's (2006) study also shows the occupations of workers involved in heavy equipment- and truck-related deaths in excavation work, 1992e2002 in USA and according to that study in 43% of the total fatal cases, equipment operators lost their lives. The past research showed that determining equipment related hazards and assessment of associated risks is very important in preventing heavy equipment related incidents. Moreover, employee participation should be an integral part of risk assessment process and later on influence the development of mitigation/ abatement strategies on sites. Therefore, the risk perception of the construction workers that are exposed to heavy equipment risks,

such as operators, flaggers and spotters, should be a hinge point for further analyses and research in heavy equipment and construction safety. For the purpose of expanding this research topic, this paper aims to determine risk perception of construction equipment operators in Turkey. 2. Scope and method This paper summarizes a study that involved a survey conducted on 198 heavy equipment operators, who were engaged in 51 construction projects. The survey used in this study was based on a two-part questionnaire designed by the authors. The survey was performed by 6 graduate students who were taking the “Construction Equipment Management” graduate course taught by one of the authors at the time. Before conducting the survey, a 3-hours crash course was given to the students to discuss the questions and the conceptual framework of the risk assessment and its “severity” and “likelihood” components. The respondents were construction equipment operators working in cities of Istanbul and Bursa, two of the major cities in Turkey, where grand-scale construction projects are conceived. The operators were informed on the risk and severity concepts associated with construction equipment activities before asking the questions. Face to face interview method was preferred, because specifically questions about severity and likelihood in the questionnaire were required to be explained to the operators by surveyors. The first part of the questionnaire consists of multiple-choice and yes/no questions that aims to gather information about the operators (experience, license, type of equipment being used, work experience with flaggers and way of communicating with them), their working environment (type of construction project, phase of the project, number of employees working in the construction site, daily average working hours, and periodic control frequency of the heavy equipment), and safety background and experience (existence of safety training and their opinion on the necessity of training, type and result of accident if any). The second part consists of rating the risk perception of the operators. Since the definition of risk implies two components: severity and likelihood; heavy equipment operators were asked to give scores for both likelihood and severity of incidents that they could be exposed to on site. The accident severity (or consequence) definitions, in particular, vary based on the concept of the work practices or industry, as seen in Table 3, which lists the consequence categories from different literature sources. Therefore, a combination of different definitions and rankings of severity measures was considered for this research and a ranking scale with 5 levels was chosen eventually.

62

G.E. Gürcanlı et al. / International Journal of Industrial Ergonomics 46 (2015) 59e68

Table 3 Consequence Categories and levels of severity. Ranking AS/NZS 4360

Nakahara and Yokota (2011) (begins with 1e6)

0 1

Minor Moderate

Insignificant- No injuries

Serious

Tweeddale (1997).

Fatality or permanent disability Serious lost time injury of illness

HSE (1999)

No injury/illness Slight injury/illness

2

Minor- First aid treatment

Major injury/illness.

3

Moderate-medical Severe treatment required

Moderate lost time Major injury/illness injury or illness

4

Major-Extrinsic injuries

Critical

Minor lost time injury or illness

5

CatastrophicDeath

Unsurvivable

No lost time

Single fatality/ permanent total disability or unfitness for work Multiple fatalities

DOE/AL 5481.1 B

Wang (1997).

Sii et al. (2001); Sii and Wang (2002).

Catastrophic- May cause deaths, or loss of the facility operations. Critical- May cause severe injury, or severe occupational illness …

Catastrophic System loss and-or death Critical Major system damage and/or severe injury Marginal Minor system damage and/or minor injury Negligible Less than minor system damage and/or less than a minor injury

Negligible No injury (1)

Marginal- May cause minor injury, or occupational illness Negligible- Will not result in a significant injury or occ. illness

Minor Single or minor injury (2,3)

Moderate Multiple injuries (4,5,6) Severe Single fatality or multiple severe injuries (7,8) Catastrophic Large number of deaths (9,10)

related incidents that are most frequently reported, such as crushed/run-over by heavy equipment, heavy equipment fall-over, caught between the equipment elements, electrocution due to contact with high voltage electric lines, crushed under heavy load and so on. To facilitate gathering information, a simple matrix, that shows typical construction heavy equipment activities in column and accident types in row, was utilized. For the construction equipment activities, a table proposed by Greenberg et al. (2006) was used, but the activities were reduced to ten most frequent jobs observed on Turkish construction sites. Respondents were asked to rate each construction activity for every accident scenario; first for severity and then for likelihood component of risk. Table 5 depicts a sample likelihood assessment of an operator. With the aid of this table (and a similar one for severity) all likelihood and severity assessments of operators were recorded and analyzed. Table 5 was also an efficient and practical tool for collecting information (i.e. assessment of the operators) on site. Moreover, just by multiplying accident severity and likelihood scores, it is simple to observe risk perception (i.e. likelihood x

It is also crucial to define accident likelihood for construction works. Different from the other industrial branches, like automotive or textile industry, construction industry does not reveal itself as a recurrence of same kinds of processes. Every construction work and on-site conditions have their unique characteristics. Hence, the implementation of the probability theory falls short for explaining the probability of an event. When historical data are insufficient or unavailable, the probability of each risk on potential failure paths cannot be evaluated by using simple probability theories (Gurcanli and Mungen, 2009). At this point, incorporation of past data and subjective judgement may be a tool to overcome these situations; and thus, definition of “likelihood” becomes more important. Definitions of accident likelihood from various literature sources can be seen in Table 4. Similar to severity portion of the survey, a 1e5 Likert scale was selected; where 1 is very low and 5 is highly frequent. After being informed on severity concepts and scoring system, operators were asked to give severity and frequency scores on a 1e5 Likert scale as described above, for construction equipment Table 4 Definition of accident likelihood from different resources (Gurcanli and Mungen, 2009). Definition

Occupational accident is unlikely but may possible during project lifetime under special circumstances Likely to happen once during project lifetime

Terms, ranking and corresponding frequencies (per year) for accident likelihood HSE (2003)

Raafat (1995)

Sii and Wang (2002).

Sii et al. (2001)

1 Incredible <10 7 2 Remote 10 5 > Fa > 10

1 Almost impossible <10 6 2 Very very low 10 5

1, 2, 3 Very Low <10 6 4 Low 0.25  10 5 5 Reasonably Low 0.25  10 4 6 Average 10 3 7 Reasonably Freq. 0.25  10 2 8.9 Frequent 0.125  10 1 9.10 Highly Frequent >0.25  10 1

1 Very Low <10 8 2,3 Düs¸ük 10 6e10 7 4, 5 Reasonably Low 10 4e10 5 6, 7 Average 10 2e10 3

7

Between low and average

Occasional accident

Will probably occur in most circumstances

Repeated accidents

Expected to occur (very likely to occur) in most circumstances in the project time a

F: frequency.

3 Unlikely 10 3 > F > 10 4 Occasional 10 1 > F > 10 5 Likely 10 > F > 10 1 6 Frequent F > 10

5

3

3 Very Unlikely 10 4 4 Unlikely 10 3 5 Likely 10 2 6 Frequent 10 1

8.9 Frequent 1e10 1 9.510 Highly Frequent >1

G.E. Gürcanlı et al. / International Journal of Industrial Ergonomics 46 (2015) 59e68

63

Table 5 Sample Accident likelihood table of an operator from survey. Construction Activities

Excavation above grade (j1) Excavating below grade (j2) Hauling soil or load (j3) Lifting, loading (soil or material) (j4) Demolition Works (j5) Soil windrowing, soil material spreading, finish grading, grubbing (j6) Soil compaction (j7) Culvert/trench box or utility pipe placement into the channel (j8) Spare part replacement (j9) Repair and maintenance (j10) Driving or manoeuvring equipment (j11) a

Accident types Crushed/run-over by heavy equipment (a1)

Heavy equipment fall-over (a2)

Caught between the equipment elements (a3)

Crushed under Electrocution (i.e contact with heavy load (a5) electric lines) (a4)

Traffic accident on site (a6)

Fall from Crushed, equipment jammed/ (a8) pinched in/ between objects (a7)

Material falls Other (a10) on the equipment (a9)

3a 4 2 2

5 4 2 3

3 2 1 2

4 1 1 5

2 2 3 4

3 4 2 2

1 4 2 3

3 2 3 4

4 5 5 5

2 2 3 4

5 1

4 1

1 1

3 1

5 1

5 1

4 1

1 1

3 1

2 1

1 1

1 3

1 1

1 1

1 1

1 1

1 3

1 1

1 1

1 1

4 4

1 1

5 5

2 2

1 1

4 4

5 5

5 5

1 1

1 1

1

1

1

1

1

1

1

4

1

1

1-Very Low, 2-Low, 3-Average, 4-Frequent, 5-Highly Frequent.

severity) of each operator or compare different groups of operators, such as operators with or without training. All the data collected from the survey were analyzed using “Statistical Package for Social Sciences Version 13.0 (SPSS 13.0)” software. Various statistical methods namely, IndependentSamples T-test, ANOVA analysis, KruskalleWallis one way analysis of variance, and ManneWhitney U test were used in different parts of the analyses. Independent-Samples T test was applied to find differences if any, between the respondents whether they took safety and health training, worked with flaggers, had accident experience or possess operator license (answers were “yes” or “no”). Since the Independent-Samples T Test procedure compares means for two groups of cases, the subjects were in advance randomly assigned to two groups, to see whether any difference in their perception of severity or likelihood exists or not. For each variable, sample size, mean, standard deviation, and standard error of the mean were calculated. Then, independent two sample t-test with equalvariances was applied to understand the statistically meaningful differences between groups in terms of mean and standard error, according to the confidence interval (p < .05 chosen). Equalvariance T-test was chosen, because the observations were independent, random samples from normal distributions with the same population variance. ANOVA analysis was initially performed for categories of “number of employees in construction”, “experience of the operator (less than one year data was excluded because only 3 operators were in the group)” and “daily average working hours of the operator (less than 8 h was excluded)”. These analyses did not generate any significant outcome. On the other hand, the sample sizes of groups for “type of project” in which operators work were less than 30 and ANOVA could not be applied. Instead, KruskaleWallis one way analysis of variance test was performed. Since there was only one operator working on a dam construction site, this operator was excluded from the analysis, which reduced the analyzed data to 189. Here it should be noted that, KruskalleWallis one way analysis of variance is a non-parametric method for testing whether samples originate from the same distribution. Moreover, it is used for comparing more than two independent samples, and

thus it can be regarded as the non-parametric equivalent of ANOVA test. When the KruskaleWallis test leads to significant results, then at least one of the samples differ from the other samples. Nevertheless, KruskaleWallis does not identify how many and where the differences occur. It is essentially an extension of the ManneWhitney U test to 3 or more groups. Since, the ManneWhitney test would help analyze the specific sample pairs for significant differences; this statistical test was also applied for specific sample pairs. 3. Results and discussions This section summarizes the results of the conducted survey in three parts. The first part discusses the results based on non-risk related questions and provides a general idea about the profile of workers and the nature of their work. The second and third part addresses the findings from severity and likelihood portions of the survey respectively. It should also be noted that eight of the questionnaires were not taken into consideration due to lack of sufficient information about severity and likelihood. Therefore, the results below represent 190 construction equipment operators. 3.1. General statistical information Table 6 reveals the results of the first part of the survey and gives the distribution of the respondents according to data categories in the questionnaire, such as project in which they are working (type of construction, number of employees, phase of the project), their experience and level of s&h training, working atmosphere, equipment used and so on. The information gathered here was used to create clusters for further statistical analyses. The respondents of the survey were mostly (65.8%) engaged in wide-scaled projects that employs more than 50 workers as seen in Table 6. Majority of the respondents were involved in highway and bridge construction projects (30%) and more than half of the projects were during excavation and ground level works phase when the survey was conducted. Crane and excavator were the most frequently used heavy equipment among operators with 24.7% and 18.4% respectively. 41.1% of the workers had more than 10 years

64

G.E. Gürcanlı et al. / International Journal of Industrial Ergonomics 46 (2015) 59e68

Table 6 Equipment operator survey results.

Table 6 (continued ) Frequency Frequency

Type of construction Highway and bridge 57 Othera 57 Residential building 24 Infrastructural works 23 Shopping Center 11 High-rise buildings 10 More than one 7 Dam 1 Number of employees in the construction site 50e99 64 More than 200 61 less than 50 38 99e200 27 Phase of the project Excavation and ground level works 105 Rough construction (structural) works 66 Finishing/trim works 15 Mobilization 4 The type of heavy equipment respondent uses Crane 47 Excavator 35 Other 15 Cylinder 14 Finisher 13 Back Hoe Loader 13 Grader 11 Loader 10 More than one equipment 10 Truck 9 Concrete Mixer 5 Forklift 4 Dozer 3 Mixer 1 Experience of the Operator More than 10 years 78 1e5 years 57 5e10 years 52 Less than one year 3 Operator Driver License Yes 179 No 11 Total 190 Safety and Health (s&h) training taken? Yes 160 No 30 Total 190 S&h training provider (institution/person etc.) From my craftsman 86 Private Training Institution 58 I did not take any training 24 Ministry of Education 22 Total 190 Opinion about the s&h training on heavy equipment Absolutely necessary 119 Necessary 55 No makes difference 7 Unnecessary 5 No idea 4 Total 190 Currently work with flagger? Yes 132 No 58 Way of Communication with flagger Walkie-talkie 53 No flagger 43 Verbal 34 Verbal and Hand-Arm 27 Hand and arm signs 18 Changes for each situationb 15 Total 190 Occupational accident experience No 148 Yes 42

Percent 30.0 30.0 12.6 12.1 5.8 5.3 3.7 0.5 33.7 32.1 20.0 14.2 55.3 34.7 7.9 2.1 24.7 18.4 7.9 7.4 6.8 6.8 5.8 5.3 5.3 4.7 2.6 2.1 1.6 0.5 41.1 30.0 27.4 1.6 94.2 5.8 100.0 84.2 15.8 100.0 45.3 30.5 12.6 11.6 100.0 62.6 28.9 3.7 2.6 2.1 100.0 69.5 30.5 27.9 22.6 17.9 14.2 9.5 7.9 100.0 77.9 22.1

Type of the accident operators experienced No accident 149 Technical failure 13 Traffic accident on site 9 Material fall from height 8 More than one factor 5 Collision with material 4 Scratch with bouncing material 2 Result of the Accident No accident 149 Non-serious injury 33 Severe injury 5 Fatal 3 Frequency of the periodical control of the heavy equipment 3 months 140 6 months 35 1 year 14 More than one year 1 Daily average working hours of the operator 8e10 h 90 10e12 h 56 More than 12 h 37 Less than 8 h 7 Total Number of Correspondents

190

Percent 78.4 6.8 4.7 4.2 2.6 2.1 1.1 78.4 17.4 2.6 1.6 73.7 18.4 7.4 0.5 47.4 29.5 19.5 3.7 100.0

a

Cement production plants, bituminous asphalt production plants, quarries, small scaled municipal public works other than infrastructural projects. b 15 of the respondents answered that they do not work with flagger and for the next question they answered as “communication with flagger changes for each situation”. This means that they were not working with flaggers at those workplaces where survey was performed.

experience 94.2% of the operators had the appropriate driver license as expected. Still, 11 of the operators did not possess the required license. According to respondents, periodical control of the heavy equipment was mostly (73.7%) being performed every 3 months. It was also observed that working conditions of the operators were quite strenuous; 96.7% of the operators worked more than 8 h/day. Furthermore, 1 of the 5 respondents claimed that they've been working more than 12 h a day. The survey also contained questions about safety and health. The responses were encouraging. Majority of the respondents had taken training (84.2%) and only about 10% of them thought that it is not necessary. Table 6 also shows that 132 (69.5%) of the operators were working with a flagger at that time, while only 53 of them used walkie talkies and the rest preferred verbal communication. Interestingly, nearly 80% of the operators did not experience any construction accidents and only 8 of the 41 accidents had severe results (5 injury, 3 fatality) as displayed in Table 6.

3.2. Determination of severity perception As mentioned above, the second part of the survey deals with severity and likelihood perception of the operators on a 1-5 Likert scale. Table 7 shows the average scores for severity of accidents rated by 190 respondents, where “crushed/run over by equipment” is found to be the most severe accident with a rating of 4.36, followed by “crushed under heavy load” and “electrocution”. On the other hand, categories like “fall from equipment” and “Crushed, jammed or pinched in or between objects” is perceived to be less severe by the respondents. Results of independent-samples T test to determine differences between operators who took safety training and operators who did not take safety training, and operators who work with a flagger and operators who do not work with a flagger, are shown in Tables 8 and 9 respectively.

G.E. Gürcanlı et al. / International Journal of Industrial Ergonomics 46 (2015) 59e68 Table 7 Average scores for accident severity (1 for negligible, 5 for very severe).

(a1) (a2) (a3) (a4) (a5) (a6) (a7) (a8) (a9) (a10)

Type of Accident

Mean

Std. Deviation

Crushed/run-over by heavy equipment Heavy equipment fall-over Caught between the equipment elements Electrocution (i.e contact with high voltage electric lines) Crushed under heavy load Traffic accident on site Crushed, jammed or pinched in or between objects Fall from equipment Material falls (rocks etc.) on the equipment Other

4.36 3.99 3.83 4.06

1.08 1.15 1.24 1.23

4.15 3.46 3.27

1.02 1.25 1.14

3.12 3.75 2.23

1.32 1.17 1.49

65

Table 9 Independent Samples t Test for severity perception between groups of operators whether they work with a flagger or not. Levene's test t-test for equality of means for equality of variances F

According to the results displayed in Table 8, the severity scores given by 160 operators who took safety and health training are greater than those of 30 untrained operators. Even though, the mean differences are 0.39 and 0.43 for crushed/run-over and traffic accident on site respectively, independent samples t test analysis gives two tailed p values greater than 0.05 for these two accident types. Additionally, caught between the equipment elements' p value is 0.52 and slightly above the confidence level. However, if these values were compared with the other analyses, two tailed p values are much more statistically significant and appropriate for future research. For other job items, the differences are statistically significant. Based on the results displayed in Table 9, it is found that the severity scores given by 132 operators who worked with a flagger are greater than those of 58 operators that did not work with flaggers or any other assistant personnel. For three accident types, the differences could not pass the t-test but are much greater when compared with other analyses. Of course, if the p-value is 0.05, then it means that there is a 5% chance of observing a difference as large as observed in the survey between the sample means, even if the population means are identical. However, it does not in any way imply that there is a 95% chance that the differences observed is due to real differences between populations and a 5% chance that the difference could be due to luck. The results of the KruskaleWallis one way analysis of variance for different types of construction projects are given in Table 10 (The descriptions of values a1 through a10 can be obtained from Table 5). The test results show that there is significant difference between at least one group and any other groups with a significance level less than 0.01 for all accident types in the survey. Initially, it is observed that the highest scores were given by the heavy equipment operators who are working at residential building and large shopping center projects. However, it must be proven

Crushed/run-over by heavy equipment (a1) Heavy equipment fall-over (a2) Caught between the equipment elements (a3) Electrocution (i.e contact with high voltage electric lines) (a4) Crushed under heavy load (a5) Traffic accident on site (a6) Crushed, jammed or pinched in or between objects (a7) Fall from equipment (a8) Material falls (rocks etc.) on the equipment (a9) Other (a10)

Sig.

t

df

Sig. Mean (2-Tailed) difference

21.633 0.000 3.244 188 0.001

0.54

21.046 0.000 3.488 188 0.001

0.61

8.185

0.005 2.467 187 0.015

0.48

0.179

0.673 0.983 188 0.327

0.19

4.960

0.027 2.276 188 0.024

0.36

0.674 1.854

0.413 2.016 188 0.045 0.175 1.460 188 0.146

0.39 0.26

0.582 0.770

0.447 3.434 188 0.001 0.381 1.573 188 0.117

0.70 0.29

15,971 0.000 2.040 188 0.043

0.47

that whether they gave more than the others and therefore pairwise comparisons should be performed by ManneWhitney U test for operators from different construction projects. The ManneWhitney U test is the most popular of the two-independentsamples tests. It is equivalent to the Wilcoxon rank sum test and the KruskaleWallis test for two groups. ManneWhitney tests that two sampled populations are equivalent in location. The observations from both groups are combined and ranked, with the average rank assigned in the case of ties. The number of ties should be small relative to the total number of observations. If the populations are identical in location, the ranks should be randomly mixed between the two samples. The number of times a score from group 1 precedes a score from group 2 and the number of times a score from group 2 precedes a score from group 1 are calculated. The ManneWhitney U statistic is the smaller of these two numbers (SPSS 13.0 Help Files). Mann Whitney U test reveals that there is statistically significant difference for the severity scores: a. between the operators from residential building sites and other projects (asphalt power plant, municipality downtown works, concrete production plant and so on) in all types of accidents,

Table 8 Independent Samples t Test for severity perception between groups of trained and untrained operators.

Crushed/run-over by heavy equipment (a1) Heavy equipment fall-over (a2) Caught between the equipment elements (a3) Electrocution (i.e contact with high voltage electric lines) (a4) Crushed under heavy load (a5) Traffic accident on site (a6) Crushed, jammed or pinched in or between objects (a7) Fall from equipment (a8) Material falls (rocks etc.) on the equipment (a9) Other (a10)

Levene's test for equality of variances

t-test for equality of means

F

Sig.

t

df

Sig. (2-Tailed)

Mean difference

0.691 2.572 0.051 0.201 1.388 0.138 2.284 0.610 0.902 55.417

0.407 0.110 0.822 0.654 0.240 0.711 0.132 0.436 0.343 0.000

1.799 2.430 1.957 2.444 3.081 1.744 2.115 2.857 2.869 3.403

188 188 187 188 188 188 188 188 188 188

0.074 0.016 0.052 0.015 0.002 0.083 0.036 0.005 0.005 0.001

0.39 0.55 0.49 0.59 0.61 0.43 0.48 0.74 0.66 0.98

66

G.E. Gürcanlı et al. / International Journal of Industrial Ergonomics 46 (2015) 59e68

Table 10 KruskalleWallis One Way Analysis of Variance for different types of construction projects for accident severity. Type of accident (a1 for crushed or run over equipment, and so on)

Chi-Square df Asymp. Sig.

(a1)

(a2)

(a3)

(a4)

(a5)

(a6)

(a7)

(a8)

(a9)

(a10)

31.269 6 0.000

17.981 6 0.006

35.609 6 0.000

18.319 6 0.005

23.908 6 0.001

23.888 6 0.001

22.501 6 0.001

40.193 6 0.000

25.642 6 0.000

39,770 6 0.000

Table 11 Average scores for accident likelihood (1 for very low, 5 for highly frequent). Typical heavy equipment Activities on construction sites

Accident types a1

a2

a3

a4

a5

a6

a7

a8

a9

Excavation above grade (j1) Excavating below grade (j2) Hauling soil or load (j3) Lifting, loading (soil or material) (j4) Demolition Works (j5) Soil windrowing, soil material spreading, finish grading, grubbing (j6) Soil compaction (j7) Culvert/trench box or utility pipe placement into the channel (j8) Spare part replacement (j9) Repair and maintenance (j10) Driving or maneuvering equipment (j11)

2.4 2.2 2.1 2.4

2.2 2.5 2.3 2.4

2.1 2.0 2.1 2.3

2.3 2.1 1.9 2.0

2.6 2.4 2.3 2.5

2.1 2.1 2.4 2.3

2.5 2.4 2.4 2.5

2.1 2.1 2.0 2.1

2.7 2.7 2.3 2.6

2.5 2.2

2.5 2.0

2.3 2.0

2.3 1.8

2.7 2.0

2.1 2.0

2.8 2.0

2.1 1.8

2.9 1.9

2.1 2.3

1.8 2.3

2.0 2.2

1.8 2.0

1.8 2.4

1.9 2.0

2.0 2.2

1.9 1.9

1.8 2.2

1.9 1.9 2.3

1.8 1.8 2.1

2.2 2.1 1.9

1.9 1.9 1.9

2.0 2.0 2.0

1.7 1.7 2.3

1.9 1.8 2.0

1.9 1.9 1.9

1.8 1.8 1.9

b. between the operators from shopping centers and other projects in all types of accidents, c. between the operators from sub-structural works and shopping center projects in 9 of 10 accident types, d. between the operators from highway projects and other projects in 8 of 10 accident types, and e. in 7 of 10 accidents there is statistically significant difference between the operators from highway and shopping center projects, in 6 of 10 from high storey and shopping center projects, in 5 of 10 from sub-structural works and residential and so on.

flagger or not in 9 of 99 for p less than 0.05. Similar to the analyses for severity scores, the KruskalleWallis one way analysis of variance, unlike ANOVA analysis, pointed out differences between the groups of operators from different projects. Table 12 shows the two tailed p values for KruskalleWallis analysis, and analysis reveals that there is significant difference between “at least” in one of the groups and any other groups with a significance level less than 0.05 (many of them were less than 0.01) while performing specific activities (significant difference exists in 89 of 99 likelihood values). Respondents answered in questionnaire that “while excavating below grade, the likelihood of equipment fall over (a2) is frequent (4)” for instance. Here again, since KruskalleWallis showed that statistically significant differences exist between groups, it is needed to perform ManneWhitney U test for the pairwise comparison of the groups of operators from different construction projects. Because, it is important to find which group(s) differ(s) from the others to advance the analysis one step forward to evaluate characteristics of the construction equipment operators' perception. Mann Whitney U test results for likeliness evaluation scores reveal that there is statistically significant difference: a. between the operators from residential building sites and other projects (asphalt plants, municipality downtown works, concrete production plant and so on) in 79 of 99, b. between the operators from substructure works and other projects in 64 of 99, c. between the operators from shopping mall and other projects in 61 of 99, d. between the operators from highway projects and other projects in 61 of 99 and operators from sub-structural works and shopping mall projects in 54 of 99 items and so on.

3.3. Determination of likelihood perception and survey results

4. Conclusions

Depiction of likelihood scores is given in the form of a matrix due to practical reasons. In Table 11, it is quite easy to follow the average likelihood scores for each equipment activity (The descriptions of values a1 through a9 can be obtained from Table 5). For example, the score 2.1 marked bold in Table 11 corresponds to the likelihood of accident for electrocution while excavating above grade. Basically, likelihood is the other component of the risk concept. Therefore, likelihood scores given by the operators should be analyzed as well as severity scores to evaluate the “risk perception”. Same procedure as severity was followed for likelihood scores. However, at this stage comparisons between the 99 likelihood scores in Table 11 and the operators had to be performed according to different classifications. The first step was again to apply independent-samples T test to find differences for likelihood scores between the respondents based on whether they took s&h training, worked with a flagger, had accident experience or possess operator license. Analysis revealed that there are statistically significant differences between the trained and untrained operators in 27 of 99 likelihood scores and between the operators who worked with a

Heavy equipment operators are exposed to variety of hazards on construction sites and experience many different kinds of workrelated incidents that result in injury or death. They also observe the consequences of those events for years. This undesirable experience, on the other hand, is invaluable and may increase their awareness of the risks on construction sites. This study stems from the assumption that employee involvement and their perception of risk is crucial in eliminating work-related hazards and investigates the factors that affect the risk perception of construction equipment operators. Previous studies that investigate risk perception mostly focused on assessing risk perception by asking respondents to name or identify certain occupational risks (Cezar-Vaz et al., 2012). This study takes this research one step further and asks likelihood and severity assessments of the respondents to predetermined occupational risks (by the aid of past accident data). The aim of this approach was to ensure visualization of the risks for each phase of construction by considering severity and likelihood perception of construction equipment operators. Since majority of the safety professionals in Turkey have been performing risk assessment by

G.E. Gürcanlı et al. / International Journal of Industrial Ergonomics 46 (2015) 59e68

67

Table 12 p values for KruskalleWallis analysis according to type of construction. Typical heavy equipment Activities on construction sites

Excavating below grade Excavating above grade Haul of soil or load Loading and storage (soil or material) Demolition works Finish grading, grubbing Soil compaction Culvert/trench box or utitily pipe placement Spare part replacement (especially large parts) Equipment repair and maintenance Movement and maneuvering on site

Accident types a1

a2

a3

a4

a5

a6

a7

a8

a9

0.0001 0.0009 0.0001 0.0001 0.2079 0.0006 0.0117 0.0467 0.0063 0.0001 0.0009

0.0466 0.0161 0.0007 0.0016 0.2798 0.0001 0.0001 0.0003 0.0231 0.0466 0.0161

0.0017 0.0001 0.0090 0.0099 0.0064 0.0155 0.1124 0.0199 0.0008 0.0017 0.0001

0.1927 0.2544 0.0003 0.0000 0.5282 0.0229 0.0263 0.0000 0.4978 0.1927 0.2544

0.0015 0.0013 0.0025 0.0051 0.3550 0.0008 0.0001 0.0013 0.0917 0.0015 0.0013

0.0001 0.0029 0.0000 0.0012 0.0125 0.0019 0.0001 0.0002 0.0000 0.0001 0.0029

0.0017 0.0005 0.0000 0.0028 0.0002 0.0180 0.0000 0.0007 0.0006 0.0017 0.0005

0.0148 0.0000 0.0003 0.0062 0.0003 0.0001 0.0002 0.0002 0.0184 0.0148 0.0000

0.0000 0.0001 0.0026 0.0102 0.0000 0.0002 0.0006 0.0006 0.0012 0.0000 0.0001

using 5  5 or 3  3 matrix method on construction sites, it becomes important to compare the perception of operators and assessment of safety professionals to efficiently implement mitigation and abatement techniques. Results of this study revealed that safety and health training and working with a flagger had a positive effect on risk perception of workers. It was observed that operators who took safety and health training gave higher scores for severity (or consequence) of accidents compared to operators without training. Since, training is an integral part of a successful s&h management system, more emphasis should be put on the characteristics and risk perception of workers while designing a job-specific s&h training program. A similar case was observed for operators who worked with flaggers; they had higher severity scores compared to operators who do not work with flaggers or any other assistant personnel. This finding points out the importance of having assistant personnel (flaggers or spotters) when performing heavy equipment activities. The assistant personnel (flaggers) are mainly focused on occupational risks during the activity and inform or warn the operator and consequently increase the risk perception level of operator. The combined positive effect of “safety and health training” and “working with a flagger” was also observed during the interviews with the operators. Most of the operators insisted in the interviews that without the help of flaggers, their movements/manoeuvers would be slower and accordingly their work would be less efficient. Moreover, flaggers' absence would lead them to neglect safety rules and work faster to accomplish the job on time. The reason of the operators' awareness was directly a result of their job-specific training with emphasis on heavy communication with flaggers, who are required to perform duties more than flagging such as, site-preparation, grade-checking, and surveying. They also stated that they remembered the training modules when they interact with them. Therefore, it can be presumed that cognitive development occurred for operators who had worked with flaggers and/or received safety and health training. Results of this study pointed out that type of construction project has an effect on operators' risk perception, which was reflected on both severity and likelihood scores. Results of Mann Whitney U tests for both severity and likeliness also revealed that heavy equipment operators who are engaged in larger scale construction projects (such as, shopping malls and housing) tend to develop higher risk perception compared to operators who only have experience in smaller construction projects (such as, asphalt plants, small public works, and concrete production plants). Since these building projects have variety of construction activities, it can be assumed that the operators observe and experience wide range of hazards that ultimately influence their risk perception.

Consequently, it is possible to comment that the project type and working environment may change the “severity” perception of the operators. Especially residential building and shopping center projects deserve paying attention for Turkey, since investments of new contractor firms have focused mostly on such projects in developing cities of Turkey. A great number of residential building and shopping centers surrounding them have been constructed for the past ten years all around the country. These development projects incorporate almost all types of construction activities. This diversity (or plurality) also brings safety risks along with it. A past research by Gurcanli (2009) points out that 46.1% of fatalities in construction occurred on residential and commercial building construction projects in Turkey. Nevertheless, being a part of such large scale projects, having exposed to numerous hazards and risks, and the public awareness on worksite safety due to increasing number of occupational injuries and fatalities might lead to an increase in severity perception of operators. It should be noted that criteria provided herein address the consequences of accidents to humans on site and worker/operator. Environmental or financial consequences of the accidents are not cited here, but in future research studies, it could be easily handled. Companies that manufacture heavy equipment could benefit from this study and devise a similar research that involves operators' risk perception to design more ergonomic and safe equipment, such as devices for visualizing or sensing the presence of humans or obstacles that are in the path of travel. Since, operators perceive accident types such as “Crushed/run-over by heavy equipment”, “crushed under heavy load”, and “electrocution” more severe than others as determined in this study, the emphasis should be on eliminating factors that cause aforementioned accident types. The findings of this study, risk assessment results of operators in particular, could especially help the safety professionals detect possible unforeseen risks and design safety and health plans for construction sites that require usage of heavy equipment on a daily basis. Without having any knowledge about risk perception of construction equipment operators, safety training programs may not attain their goals towards a construction site free of hazards and risks.

References Aneziris, O.N., Papazoglou, I.A., Mud, M.L., Damen, M., Kuiper, J., Baksteen, H., Ale, B.J., Bellamy, L.J., Hale, A.R., Bloemhoff, A., Post, J., Ohg, J., 2008. Towards risk assessment for crane activities. Saf. Sci. 46, 872e884. AS/NZS 4360, 1999. Australian and New Zealand Standard on Risk Management. Bhide, A., 2006. Analysis of Accidents and Injuries of Construction Equipment Operators. Master of Science thesis submitted to the Division of Graduate Studies and Research of the University of Cincinnati. Department of Civil and Environmental Engineering of the College Of Engineering. Cezar-Vaz, M.R., Rocha, L.P., Bonow, C.A., da Silva, M.R.S., Vaz, J.C., Cardoso, L.S., 2012. Risk perception and occupational accidents: a study of gas station workers in southern Brazil. Int. J. Environ. Res. Public Health 9 (7), 2362e2377.

68

G.E. Gürcanlı et al. / International Journal of Industrial Ergonomics 46 (2015) 59e68

Dedobbeleer, N., Beland, F., 1991. A safety climate measure for construction sites. J. Saf. Res. 22, 97e103. Greenberg, D., Popescu, C.M., Ryan, R.C., 2006. Construction Equipment Management for Engineers, Estimators, and Owners. Broken Sound Parkway NW CRC Press, Taylor & Francis Group. Gürcanli, G.E., Müngen, U., Akad, M., 2008. Construction equipment and motor vehicle related injuries on the construction sites in Turkey. Ind. Health 46, 375e388. Gürcanli, G.E., 2009. Who is at fault? Third party and child injuries at construction sites in Turkey. Saf. Sci. 47 (3), 364e373. Gürcanli, G.E., Müngen, U., 2009. An occupational safety risk analysis method at construction sites using fuzzy sets. Int. J. Ind. Ergon. 39 (2), 371e387. Hallowell, M., 2008. A Formal Model of Construction Safety and Health Risk Management. Doctoral dissertation. Oregon State University, Corvallis, OR. Hallowell, M., 2010. Safety risk perception in construction companies in the Pacific Northwest of the USA. Constr. Manag. Econ. 28, 403e413. HSE (Health and Safety Executive), September, 1999. Management Guidelines for Working Together in a Contract Environment. Report No: 6.64/291. HSE, December, 2003. Hazardous Installations Directorate (HID) Safety Report Assessment Guide: Explosives. Date of Publication 1st. McCann, M., 2006. Heavy equipment and truck-related deaths on excavation work sites. J. Saf. Res. 37, 511e517. Meridian Research Inc, 1994. Worker Protection Programs in Construction. Final Report, Prepared for:. Office of Program Evaluation Occupational Safety and Health Administration 200 Constitution Avenue, NW Washington, DC 20210. Nakahara, S., Yokota, J., 2011. Revision of the international classification of diseases to include standardized descriptions of multiple injuries and injury severity. Bull. World Health Organ. 89 (3), 238e240.

Raafat, H., 1995. Machinery Safety: The Risk Based Approach, Practical Guidelines on Risk Assessment. Standards and Legislation Technical Communications (Publishing) Ltd. Sii, H.S., Ruxton, T., Wang, J., 2001. A fuzzy-logic-based approach to qualitative safety modelling for marine systems. Reliab. Eng. Syst. Saf. 73, 19e34. Sii, H.S., Wang, J., 2002. Safety Assessment of FPSO’s- the Process of Modelling System Safety and Case Studies, Report of the Project-“The Application of Approximate Reasoning Methodologies to Offshore Engineering Design. Liverpool John Moores University, U.K. EPSRC GR/R30624 and GR/R32413. Starr, C., 1969. Social benefits versus technological risks. Science 165 (3899), 1232e1238. Tweeddale, M., 1997. Risk Management Handbook for the Mining Industry. New South Wales Department of Mineral Resources, Sydney. U.S. Bureau of Labor Statistics, 2011. U.S. Department of Labor available online: http://www.bls.gov/iif/oshwc/cfoi/cfch0008.pdf (accessed 22.09.12.). _ _ ¸ aatlarda Yapı Makinaları Kullanımında Is _ ¸ Güvenlig i Risk Uzun, I.M., 2012. Ins erlendirmesi (Risk Assessment of the Construction Equipment Utilization in Deg the Construction Works). Master of Science Thesis. Graduate School of Science Engineering and Technology, Technical University of Istanbul. Vredenburgh, A.G., 2002. Organizational Safety: which management practices are most effective in reducing employee injury rates? J. Saf. Res. 33, 259e276. Wang, J., 1997. A subjective methodology for safety analysis of safety requirements specifications. IEEE Trans. Fuzzy Syst. 5 (3), 418e443. Zohar, D., 2000. A group-level model of safety climate: testing the effect of group climate on micro-accidents in manufacturing jobs. J. Appl. Psychol. 85 (4), 587e596.