Developing an integrated decision making approach to assess and promote the effectiveness of occupational health and safety management systems

Accepted Manuscript Developing an integrated decision making approach to assess and promote the effectiveness of occupational health and safety management systems Iraj Mohammadfam, Mojtaba Kamalinia, Mansour Momeni, Rostam Golmohammadi, Yadollah Hamidi, Alireza Soltanian PII:

S0959-6526(16)30195-0

DOI:

10.1016/j.jclepro.2016.03.123

Reference:

JCLP 6972

To appear in:

Journal of Cleaner Production

Received Date: 4 April 2015 Revised Date:

14 February 2016

Accepted Date: 16 March 2016

Please cite this article as: Mohammadfam I, Kamalinia M, Momeni M, Golmohammadi R, Hamidi Y, Soltanian A, Developing an integrated decision making approach to assess and promote the effectiveness of occupational health and safety management systems, Journal of Cleaner Production (2016), doi: 10.1016/j.jclepro.2016.03.123. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Abstract

M AN U

SC

RI PT

Occupational Health and Safety Assessment Series (OHSAS) 18001 standard has been recommended as a tool for managing and controlling occupational risks through a systematic and structured management. However, controversy exists as to the effectiveness of OHSAS 18001 in reducing occupational risks at workplace. The current study introduces an integrated decision making approach by merging two techniques including Analytical Network Process (ANP) and Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) to assess and improve the effectiveness of OHSAS 18001 standard. Our findings indicate that the most influential factors to be taken into account to improve the effectiveness of OHSAS 18001 standard are management commitment, workers’ participation, allocation financial resources, training, risk assessment, definite responsibility, communication and dissemination of occupational health and safety results and activities. The study offers this approach as a tool to evaluate and promote the effectiveness of OHSAS 18001 standard.

AC C

EP

TE D

Keywords: Safety management systems, Effectiveness, Decision making, Assessment

ACCEPTED MANUSCRIPT

Developing an integrated decision making approach to assess and promote the effectiveness of occupational health and safety management systems

3. 4. 5.

SC

2.

Department of Occupational Hygiene, School of Public Health and Research center for health sciences, Hamadan University of Medical Sciences, Hamadan, Iran Department of Occupational Hygiene, School of Public Health, Hamadan University of Medical Sciences, Hamadan, Iran Department of Management, University of Tehran, Tehran, Iran Department of Health Management School of Public Health, Hamadan University of Medical Sciences, Hamadan, Iran Department of Biostatistics, School of Public Health, Hamadan University of Medical Sciences, Hamadan, Iran

M AN U

1.

RI PT

Iraj Mohammadfam1, Mojtaba Kamalinia2*, Mansour Momeni3, Rostam Golmohammadi1, Yadollah Hamidi4, Alireza Soltanian5.

*corresponding author: Tel: +988118380025; Fax: +988118380509. Address: Hamadan Medical Science University, Hamadan, Iran, postal code: 689

AC C

EP

TE D

E-mail addresses: [email protected], [email protected]

1

ACCEPTED MANUSCRIPT

1. Introduction

SC

RI PT

Despite the undeniable efforts that have been invested in safety and health at workplace, occupational accidents and incidents continue to occur, at a large scale, globally (Cadieux et al., 2006; Cagno et al., 2014). Occupational accidents and illnesses have serious adverse effects on the workforce, assets, equipment, environment and economics. They harm the staff, damage the equipment and consequently decrease productivity, public reputation and market competitiveness of companies (Fernandez-Muniz et al., 2007).

TE D

M AN U

In industrialized countries occupational accidents and illnesses are responsible for 5 to 7% of all deaths. According to a report by the International Labor Organization (ILO), approximately 2 million fatal and 268 million non-fatal occupational accidents take place annually (ILO, 2003). Furthermore, the ILO estimates that production loss, absenteeism, medical treatments and compensation payments to injured employees would be equivalent to 4% of annual global gross domestic product (Takala et al., 2014). This means that occupational accidents and illnesses are serious problems in industries (Fernandez-Muniz et al., 2012a; Ramli et al., 2011).

AC C

EP

Therefore, devising new preventive strategies to reduce the accidents and their adverse impacts is a top priority in occupational health and safety (FernandezMuniz et al., 2012b, 2007). Within the past decades, newer methods for managing the occupational health and safety (OHS) areas in an effective manner have been introduced. These methods have focused on more complex aspects of OHS such as safety management systems rather than on specific aspects such as technical and human factors (Abad et al., 2013; Mengolini and Debarberis, 2008). Research findings on occupational health and safety have revealed that many OHS risks could be prevented or controlled via implementation of occupational health and safety management systems (OHSMSs). Nowadays, various OHS models or guidelines have been developed by national and international organizations and institutions in response to the demands of firms to manage occupational risks (Abad et al., 2013; Robson et al., 2007).

2

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

The most commonly used OHSMS for assessing health and safety management processes is OHSAS (Occupational Health and Safety Assessment Series) 18001 standard (Granerud and Rocha, 2011). It was first introduced by British Standard Institution (BSI) in April 1999 and was updated in July 2007 (Fernandez-Muniz et al., 2012b). OHSAS 18001 standard is based on a management system framework and involves Plan-Do-Check-Act cycle of Deming. Although its structure is similar to ISO 9001 and ISO 14001 standards, it has the advantages to identify, reduce and control work hazards (Vinodkumar and Bhasi, 2011; Zeng et al., 2007). Indeed, it includes OHSMS requirements such as policy, planning, implementation and operation, checking, management review, responsibility, documentation, audits, records, communication and continuous improvement (Robson et al., 2007; Zeng et al., 2007).

TE D

Thus, it can be viewed as a tool that can manage and control occupational risks through a systematic and structured management (Chen et al., 2009; Santos et al., 2013). This standard, due to its compatibility with quality and environmental management systems, has become more widespread (Fernandez-Muniz et al., 2012a; Santos et al., 2013). About 56251 companies in 116 countries have received OHSAS 18001 certification at the end of 2009 (Hasle and Zwetsloot, 2011). Despite the increased use of OHSMSs all over the world, no unanimous agreement exists as to their effectiveness in controlling and reducing occupational risks at workplace (Hasle and Zwetsloot., 2011).

AC C

EP

This might be due to improper audits and OHS progression measurement tools (Blewett and O’Keeffe, 2011; Cadieux et al., 2006; Chang and Liang, 2009). Auditors often use audit tools during evaluation process which are based on subjective and intangible data. These neither objectively assess the effectiveness of OHSMSs nor provide robust information about OHS practices. Therefore, paperwork of audits and auditor teams without considering OHSMSs performance aspect cause many managers feel that OHSMSs are just more paperwork and nothing more than another expenditure job (Chang and Liang, 2009; Podgorski, 2015). Most companies utilize lagging indicators to measure the status of the occupational health and safety (Laitinen et al., 2013; Podgorski, 2015). Performance indicators 3

ACCEPTED MANUSCRIPT

based on traditional data, such as number of Lost Time Injuries (LTI), number of sick leave days due to injuries, Injury Frequency Rate (IFR) and Injury Severity Rate (ISR), due to the inherent complexity of OHS cannot reflect the real OHS conditions (Chang and Liang, 2009; Hinze et al., 2013).

SC

RI PT

These indicators are retrospective and provide imprecise and unreliable information about improvement to management (Laitinen et al., 2013; Ramli et al., 2011). On the other hand, proactive indicators provide early warnings and signs in time in relation to the occurrence of occupational risks, irregularities, faults and forecast future status (Harms-Ringdahl, 2009; Øien et al., 2011; Shafiee, 2015).

M AN U

Thus, it is necessary to develop a comprehensive evaluation system that consists of both lagging and leading measurable and achievable indicators to continuously assess and improve the effectiveness of OHSMSs (Chang and Liang, 2009; Reiman and Pietikainen, 2012). Assessing the effectiveness of OHSMSs and identifying their influential factors enable managers to realize what the status of OHS is and how to make decisions for promoting OHS conditions ( Cagno et al., 2011).

EP

TE D

In this context, the main objectives of this study were as follows: (1) to develop an integrated decision making approach (ANP-TOPSIS) in order to evaluate and promote the effectiveness of occupational health and safety management systems (2) to identify appropriate criteria and key performance indicators and to quantitatively prioritize them as parts of an effective occupational health and safety management system and (3) to assess the effects of such criteria on performance of occupational health and safety management systems.

AC C

In the following section, the literature concerning the effectiveness of OHSMSs based on OHSAS 18001 standard is reviewed. Then the research methodology is described and demonstrated with a case study. Furthermore, the results are reported. The final section gives discussion and conclusion for promotion the effectiveness of OHSMSs.

4

ACCEPTED MANUSCRIPT

2. Literature review

M AN U

SC

RI PT

The mission of OHSAS 18001 standard is managing occupational risks and promoting work conditions through a systematic and structured approach (Vinodkumar and Bhasi, 2011). Most studies related to OHSMSs are mainly concerned with certification process, performance evaluation, motivation for the adoption of OHSMSs and benefits of OHSMSs, not directly dealing with the effectiveness of OHSMSs (Fernandez-Muniz et al., 2012a; Ismail et al., 2012; Santos et al., 2013; Vinodkumar and Bhasi, 2011). Few studies have tried to identify key factors contributing to success and the effectiveness of OHSMSs. Ramli et al. (2011) proposed a method to determine influential factors for successful planning and implementation of OHSMSs. This approach determined a limited number of influential factors such as policy and program, risk assessment and risk control. Chang and Liang (2009) developed a quantitative model to evaluate safety management system performance in manufacturing facilities using multi attribute decision making method.

TE D

Teo and Ling (2006) used Analytic Hierarchy Process (AHP) method to develop a model to measure the effectiveness of safety management systems of construction sites. Another research was carried out to measure the effectiveness of safety management system implementation on-board ships. In this study, authors utilized AHP and Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) techniques to monitor safety management system performance through a limited number of key performance indicators (KPIs) (Akyuz and Celik, 2014).

AC C

EP

Recently, a research attempting to select KPIs for components of OHSMSs has been performed (Podgorski, 2015). OHSMSs are effective and efficient when they achieve their own expected results. Thus, they should be monitored and interpreted continuously. Continuous improvement of OHSMSs requires an appropriate method for analysis to provide high quality results (Gallagher et al., 2001). Some researchers have used multi-criteria decision-making (MCDA) techniques, especially AHP, for assessing the effectiveness of OHSMSs. It is essential to know that OHSMSs are dynamic, complex and depend on many complicated factors. Thus, there are some interactions, interdependencies and feedback between components of OHSMSs. For instance, management with high commitment to health and safety communicates the importance of health and safety to workers through attending safety training courses, safety inspections, safety meetings and safety committees. Such measures can involve workers in 5

ACCEPTED MANUSCRIPT

safety processes such as hazard identification and reporting and this can lead to positive safety performance such as fewer work related accidents.

SC

RI PT

The literature review leads us to conclude that most researches have considered these factors in a hierarchy and independent from one another without paying attention to interactions, feedback and dependency among elements and the contributing factors of OHSMSs. The interdependencies and feedback between components of OHSMSs may never be considered by using AHP technique. However, the ANP technique has the capability to consider these interactions and feedback between OHS components.

M AN U

To the best of authors’ knowledge ANP technique has not previously been applied to assess the effectiveness of OHSMSs. In order to overcome these limitations, we proposed ANP-TOPSIS approach, which aims at exploring relevant factors related to OHSMSs effectiveness. 3. Methods

3.1. ANP technique

TE D

This paper proposes an ANP-TOPSIS-based approach in order to allow the evaluation of the effectiveness of OHSAS 18001 standard in workplaces and to improve its effectiveness. The following sections explain ANP-TOPSIS techniques and their utilizations.

AC C

EP

The ANP, as a generalization and complementary technique of the AHP, is an appropriate multi-criteria technique for solving complex decision-making problems. In some cases that there is interactions, feedback and independencies among various factors of upper level with lower level, the ANP technique for solving these decision making problems has been suggested (Hsu and Hu, 2009). In ANP, the problem of decision making is modeled in the form of a network structure in which goal, criteria and alternatives constitute clusters. ANP has been used in different domains such as risk assessment and decision analysis (Ergu et al., 2014; Huang et al., 2012), work system safety (Dagdeviren et al., 2008), hazards planning and emergency management (Levy and Taji, 2007), and environmental management systems selection (Guerrero-Baena et al., 2014). This method has not been used to assess the effectiveness of OHSMSs.

6

ACCEPTED MANUSCRIPT

Thus, in this study ANP technique was just used for computing the relative importance weights of criteria to assess and promote the effectiveness of OHSAS 18001 standard.

RI PT

3.2. TOPSIS technique

3.3. Proposed approach

TE D

M AN U

SC

TOPSIS is a multi-criteria technique that investigates multi attribute decision making problems with m alternatives as a geometric system in which has m points in the n dimensional space of criteria. It employs cost and benefit criteria to identify an alternative that has the shortest distance to the positive ideal solution and the longest distance to the negative solution. Decision makers are often faced with problems when they specify careful performance ranking to an alternative for the attributes under consideration (Sun, 2010). This technique is a particularly useful tool for comparing and ranking alternatives based on positive and negative ideal solutions. TOPSIS technique has been utilized by researchers in different areas such as water management, transportation, human resource management and safety management evaluation (Akyuz and Celik, 2014). The integrated ANPTOPSIS has been used in different sectors such as supplier selection and maintenance strategy (Lin et al., 2011; Pourjavad et al., 2013). However, ANPTOPSIS approach has rarely been applied to OHSMSs domain. In this study, the TOPSIS technique was performed to compute the effectiveness of OHSAS 18001 standard in different years.

EP

The quantified results of ANP and TOPSIS techniques are utilized and applied under the integrated study methodology framework to evaluate the effectiveness of OHSAS 18001 standard.

AC C

Initially, the ANP technique is applied to develop evaluation criteria weights. Then, pairwise comparison matrices are constructed. Subsequently, the relative importance weights of criteria are computed. Thereafter, TOPSIS technique is utilized to assess the effectiveness of OHSAS 18001 standard for different years. As a result, advantages of both ANP and TOPSIS techniques are applied to the integrated ANP-TOPSIS approach to measure and promote the effectiveness of OHSAS 18001 standard. Figure 1 illustrates the flow diagram of integrated research methodology for measuring the effectiveness of OHSAS 18001 standard. The structure of the proposed integrated decision making approach (ANP-TOPSIS) includes 11 steps: 7

ACCEPTED MANUSCRIPT

Step (1) Identifying the criteria and measurement indicators: In this step influential criteria and their measurement indictors are identified based on experts opinion, literature and data in order to establish comparison matrix of criteria for main elements of OHSAS 18001 standard.

SC

RI PT

Step (2) Developing decision making model and specifying interdependencies between the criteria under consideration: In this step, a decision making model based on ANP technique is constructed. Constructing ANP model process depends on decision structure, impact of criteria on each other and experts’ opinion in order to determine weights of criteria. All elements or factors that affect decision are placed in clusters (main elements) and are linked together (figure 2).

TE D

M AN U

Step (3) Constructing a pairwise comparison matrix: In order to determine the relative importance of criteria, a pairwise comparison matrix of clusters and a pairwise comparison matrix of criteria are constructed for each of the main elements. To compare the two cluster or criteria in these comparison matrices, a measurement 1-9 scale proposed by Saaty is used (Table 1). In this table,1 indicates that the two criteria have equal importance whereas 9 indicates that the row criterion is far more important (extreme importance) than the column criterion. If a row criterion is weaker than a column criterion, its score will be reciprocated from 1-9. For example, the fraction 1/9 indicates that the column criterion is much more important (extreme importance) than the row criterion.

AC C

EP

Step (4) Calculating criteria weights (priorities): The final priorities or weights of criteria are acquired from columns in the limit super matrix based on ANP model. In order to form limit super matrix, the three following stages are needed. a) First the unweighted super matrix in which eigenvectors of the pairwise comparison has been replaced is constructed. b) Then, the weighted super matrix is yielded by multiplying the unweighted super matrix by the priority weights from the clusters. c) Finally, the limit super matrix is constructed by raising the weighted super matrix to powers until the importance weights converge and remain stable. Priorities or normalized weights of criteria in cluster are found by dividing each entry in the column to the sum of entries in the column. General structure of super matrix is as follows: E E W E W  E W E W

E E E W W W W W W W W W W W W

(1)

8

ACCEPTED MANUSCRIPT

W denotes dependence and feedback where w is weights of relation network between elements (E , E , E , E ). More information about the theory of ANP can be found in Saaty’s book (Saaty, 2005).

C X X X ⋮ X

C X X X ⋮ X

C  X  X  X  ⋮  X 

SC

C X X X ⋮ X

(2)

M AN U

 A   A D=   A  ⋮ A 

RI PT

Step (5) constructing a decision matrix (D): This decision matrix comprises all available information on criteria including weight and numerical values of the data. The structure of a decision matrix is as follows;

Where x is the performance rating of alternative A with respect to criterion c .

Step (6) Calculating a normalized decision matrix: The decision matrix D is normalized by using the following equation. !"

% #∑ & !'( !"

, i = 1,2,3, . . m and j = 1,2,3. . , n.

TE D

r =

(3)

EP

Step (7) Calculating a weighted normalized decision matrix: In this step, the weighted normalized decision matrix 1v 3 can be computed by multiplying the normalized decision matrix r by its associated weight w . The calculation is as follows;

AC C

v = w . r , i = 1,2, … , n, j = 1,2, . . . , n

Where w is the weight of the ith criterion.

(4)

Step (8) Determining the positive and negative ideal solution: positive (A⁺) and negative (A‾) ideal solutions are determined through identifying the maximum and minimum values within the row of weighted normalized decision matrix based on the following equations.

9

ACCEPTED MANUSCRIPT

A⁺=61max v ⎸j ∈ J3or1min v ⎸j ∈ J < 3for i = 1, 2 , . . , m> = ?v@ , v@ , . . , v@ A And

AB =61min v ⎸j ∈ J3or1max v ⎸j ∈ J < 3for i = 1, 2 , . . , m> = ?vB , vB , . . , vB A

(5)

RI PT

Where J = 1, 2, 3, . . , n is related to benefit (positive criteria) and J < = 1, 2, 3, . . , n is related to cost (negative criteria).

SC

Step (9) Calculating the separation measure: the separation of each alternative to the positive (S @) and negative (S B ) ideal solutions can be estimated by the following equations. S@ = #∑ G(v − v @ ) , i = 1, 2, … , m

M AN U

(6)

And

SB = #∑ G(v − v B ) , i = 1, 2, … , m

(7)

C∗ =

IJ !

I!K @ I!J

TE D

Step (10) Calculating the relative closeness to the ideal solution: In this step, the relative closeness to the ideal solution is estimated by the following formula. , 0
(8)

AC C

EP

Step (11) Ranking the preference order (OHSMSs effectiveness assessment): In the final step, alternatives are ranked in accordance with the descending order of C∗ . In this step, the effectiveness of OHSAS 18001 standard in different years is compared.

[Insert Fig. 1. around here]

[Insert Fig. 2. around here]

10

ACCEPTED MANUSCRIPT

Table 1 Saaty’ scale 1-9 for pairwise comparisons. Numerical values

Definition

1

Moderate importance of one over another Essential or strong importance Very strong importance Extreme importance

3 5 7 9

2,4,6,8

M AN U

Intermediate values between the two adjacent judgments

SC

RI PT

Equal importance

3.4. Criteria and related key performance indicators

TE D

The principles of Plan-Do-Check-Act model philosophy (PDCA cycle of Deming) have been used in most OHSMSs especially in OHSAS 18001 standard (Chen et al., 2009; Robson et al., 2007). Therefore, it includes certain items on policy, objectives, practices, roles, procedures, functions and processes.

AC C

EP

A comprehensive review of OHS management systems and the previous literature on OHS management system was performed to define main elements and factors that have an impact on OHSAS 18001 standard, based on various aspects characterized in the OHSAS 18001 standard. Five components regarding the structure of OHSAS 18001 standard were identified as main elements namely “policy”, “planning”, “implementation”, “checking” and “management review”. Then, initial criteria and relevant indicators that would affect these five main elements were obtained. In order to identify whether or not each criterion and related indicator(s) were appropriate for each main element, a preliminary list of extracted criteria and related indicator(s) was investigated by five university professors in occupational health and safety, industrial management, health management and biostatistics. This resulted in some modifications in which suggested changes were performed via simplifying, rephrasing, replacing and eliminating redundant items twice.

11

ACCEPTED MANUSCRIPT

Then, 43 criteria and 80 performance indicators were assigned to the five main elements of OHSAS 18001 standard. Before including these criteria in the questionnaire, the reliability and content validity of the criteria and indicators were determined.

M AN U

SC

RI PT

For this purpose, a questionnaire which contained the required criteria as influential factors covering the areas of policy (8 criteria), planning (9 criteria), implementation (11 criteria), checking (11 criteria) and management review (4 criteria) was designed to consider opinion of OHS experts about the importance of individual criteria within an element. Then, the questionnaires were sent via email to 30 OHS experts, OHS managers, with at least 5 years of practical experience in OHSMSs. Each criterion was measured on a 5-point Likert scale. In this study, the respondents were required to give their level of agreement with importance of each criterion for each element on a 5-point Likert scale (1= very low, 2= low, 3= middle, 4=high and 5= very high). All of the 30 questionnaires were returned and the data were coded and transferred into Statistical Package for the Social Sciences (SPSS software, version 16).

AC C

EP

TE D

To facilitate interpretation of the data, the 5-point Likert scales measuring the importance of individual criteria within an element were aggregated into three categories (1= not essential, 2+3= useful and 4+5= essential). Using Lawshe’s equation, used to establish and quantify content validity, and reliability coefficient called Cronbach’s alpha, we checked the Content Validity Ratio (CVR) and the internal consistency of the criteria. Based on the number of panel members, Lawshe provides minimum CVR value with 95% confidence interval. With a panel size of 30 respondents, the CVR value of 0.33 was determined as the minimum CVR value for validating the criteria. A criterion with CVR value of less than 0.33 should not be included in the final questionnaire. All criteria had CVR values higher than the desired value of 0.33, thus all criteria were strongly accepted. Additionally, internal consistency between the criteria for each of the five elements and all criteria was calculated using reliability indicator called Cronbach’s alpha. Alpha value of 0.7 or above is required for demonstrating good reliability of established criteria. All criteria in each of the five elements had Alpha coefficients equal to 0.7 or higher. As a result, there was very good internal consistency between the criteria in each element. The values of content validity ratio and reliability coefficient of all of the criteria are summarized in table 4. 12

ACCEPTED MANUSCRIPT

SC

RI PT

The list of criteria and examples of KPIs assigned to each main element are provided in table 2. These criteria and KPIs present guidelines and information on how a system should be managed. These factors can also be used as predictors to improve system performance. Therefore, in order to monitor and justify the OHSMS performance and describe how far the OHSMS is from the ideal level, reference levels of value in the reference period should be determined through a panel of experts (a group of subject matter experts) for each indicator. Afterwards, the real value of the indicators should be compared with values of reference levels. If the real value is not far from its reference level, the satisfaction is good and if the real value is away from its reference level, the satisfaction is neutral (Bana e Costa et al., 2002; Basso et al., 2004).

M AN U

Table 3 provides an example for the indicators of the criteria “communicating OHS policy”, “preliminary risk assessment” and “workers’ participation” of the policy element described in table 2. Thus, pretty robust OHS performance measurement is vital for assessing the effectiveness of OHSAS 18001 standard.

TE D

A simple computational approach to obtain sound information about OHS management systems through measurement and statistical analysis of traditional data (i.e. number of injuries and ill-health) is often unreliable and insufficient due to dynamic nature, multidimensionality and inherent complexity of OHS. Hence, an appropriate approach which provides high quality results for continuous improvement of OHSMS is required (Ramli et al., 2011).

AC C

EP

The interactions, dependencies and feedback between OHS components and identifying their key criteria are important issues that should be considered during assessing the effectiveness of OHSMS. For this purpose, ANP technique presents these dependencies and interactions as a network and incorporates them in nodes and clusters. Thus, this technique has the capability of determining the relative importance weights of influencing factors of successful OHSMS and quantitatively prioritizing them. On the other hand, the gaps of alternatives between real performance values and the desired levels in each element and criterion can be improved using TOPSIS technique. Therefore, the integrated decision making approach (ANP-TOPSIS) via using specified criteria and their KPIs can measure the effectiveness of OHSAS 18001 standard.

13

ACCEPTED MANUSCRIPT

Table 2 List of criteria and some examples of KPIs assigned to individual OHSMS main elements Criteria

KPIs

PO1

A. Top management commitment

A1: The number of top managers walkthroughs per year

PO6 PO7 PO8 PL1 PL2

PL3

Planning

PL4 PL5 PL6 PL7

I. Workers’ participation in work posts risk assessment J. Encouraging workers due to their participation in risk assessment K. Recording and reporting OHS activities for risk assessment planning L. Communicating OHS activities M. Reviewing and up-dating risk assessment policies N. Using units' OHS data during OHS program development O. Deadline for OHS programs

I1: The number of OHS problems identified by workers J1: The number of rewards for OHS hazard reports by workers

(BSI, 2007; Costella et al., 2009; Podgorski, 2015)

K1: The number of units in which OHS report and record keeping systems exist

(BSI, 2007; Costella et al., 2009;Podgorski, 2015)

L1: The number of toolbox meetings per week M1: The number of job safety analyses which have been reviewed by team work N1: The number of OHS violations

(BSI, 2007; Chang and Liang, 2009) (Ramli et al., 2011; Podgorski, 2015)

O1: The number of OHS programs performed in a definite period P1: The number of OHS boards in work posts

(BSI, 2007; Costella et al., 2009; Podgorski, 2015)

PL9

P. Announcing OHS programs and objectives Q. Allocating financial resources to OHS programs

IM1

R. Training workers about OHS duties to ensure competence

R1: The number of work posts in which defined OHS responsibility and duties exist

S. Using risk assessment results during OHS training plan development T. Announcing OHS activities and issues to workers U. Workers’ participation in OHS activities V. Incentive for workers due to their participation in OHS activities W. OHS documentation and regulation X. Allocating financial resources to ERP Y. Emergency response drills based on risk assessment results Z. Practical emergency response drills based on procedures

S1: The number of work posts in which risk assessment has been performed and corrective action or changes have been made T1: The number of OHS posters, bulletin, newsletters published U1: The number of accidents due to unsafe act

AA. Provision of response equipment

AA1: The number of verified OHS procedures applied in purchase or practice

PL8

Implementation and operation

IM2 IM3 IM4 IM5 IM6 IM7 IM8 IM9 IM10

emergency and their

(BSI, 2007;, Fernandez-Muniz et al., 2007; Chang and Liang, 2009; Podgórski, 2015) (BSI, 2007; Fernandez-Muniz et al., 2007; Podgorski, 2015) (BSI, 2007; Jovasevic-Stojanovic and Stojanovic, 2009; Podgorski, 2015) (Basso et al., 2004; BSI, 2007; Chang and Liang, 2009; Jovasevic-Stojanovic and Stojanovic, 2009) (BSI, 2007; Jovasevic-Stojanovic and Stojanovic, 2009, Reiman and Pietikainen, 2012;Podgorski,2015)

SC

PO5

C1: The number of OHS policies which have been reviewed D1: Percentage of OHS regulations and standards that are applicable in work posts E1:The number of OHS suggestions proposed by workers F1: The number of workers that have good knowledge on OHS policy G1: The number of risk assessments carried out in units H1: The number of managers’ meetings for OHS issues

Q1: Financial resources allocated for OHS per year($)

EP

Policy

PO4

B1: The number of units in which OHS policy exists

M AN U

PO3

(BSI, 2007; Fernandez-Muniz et al., 2007;Robson et al., 2007; Chang and Liang, 2009; Costella et al., 2009; Reiman and Pietikainen ,2012; Podgorski, 2015) (BSI, 2007; Fernandez-Muniz et al., 2007; Chang and Liang, 2009; Podgorski,2015) (BSI, 2007; Ramli et al., 2011; Podgorski, 2015)

B. Communicating OHS policy and availability in work posts C. Reviewing and up-dating OHS policy D. Consistent with other organization's policies E. Workers’ participation for developing OHS policy F. Simplicity and understandability of OHS policy G. Preliminary risk assessment for developing OHS policy H. Supervision on OHS policy implementation

TE D

PO2

Reference

RI PT

Code

AC C

OHSMS main element

V1: The number of rewards for participating in OHS activities W1: The number of procedures changed or developed X1: The number of ERP maneuver performed Y1: The number of work posts that have ERP procedure Z1: The number of emergency response drills performed

14

(Fernandez-Muniz et al., 2007; Podgorski, 2015; Wu et al, 2010 )

(Basso et al., 2004; BSI, 2007)

(BSI, 2007; Robson et al., 2007, Chang and Liang, 2009; Costella et al., 2009; Podgorski, 2015) (BSI, 2007; Robson et al., 2007; Costella et al., 2009;Podgorski ,2015) (Basso et al., 2004; BSI, 2007; Fernandez-Muniz et al., 2007; Robson et al., 2007; Chang and Liang 2009, Podgorski 2015) (Jovasevic-Stojanovic and Stojanovic, 2009; Podgorski, 2015) (BSI, 2007; Fernandez-Muniz et al., 2007; Chang and Liang, 2009; Podgorski, 2015) (BSI 2007, Robson, Clarke et al. 2007, Costella, Saurin et al. 2009, Hamidi, Omidvari et al. 2012, Podgórski 2015) (Fernández-Muñiz, Montes-Peón et al. 2007)

(Basso et al., 2004; BSI, 2007; Reiman and Pietikainen, 2012;Podgorski, 2015) (Basso et al., 2004; Chang and Liang, 2009; JovasevicStojanovic and Stojanovic,2009;Podgorski, 2015) (BSi,2007;Chang and Liang, 2009; Podgorski, 2015) (BSI, 2007; Chang and Liang, 2009; Reiman and Pietikainen, 2012; Podgorski, 2015) (BSI, 2007; Podgorski, 2015; Wu et al., 2010)

ACCEPTED MANUSCRIPT

CH3 CH4

Checking

CH5 CH6 CH7 CH8 CH9 CH10 CH11

MA2 MA3 MA4

AK. Announcing accident investigation results to employees AL. Announcing corrective and preventive actions AM. Having record, report and analysis system for accidents AN. Having time-scale to review meetings AO. Available OHS activities results during reviewing AP. Having OHS indicator during reviewing AQ. Presence of responsible person in review meeting

(BSI,2007; Robson et al., 2007; Ramliet al., 2011; Podgorski, 2015) (BSI, 2007; Costella et al., 2009; Hamidi et al., 2012,) (BSI, 2007; Fernandez-Muniz et al., 2007; Costella et al., 2009;Podgorski, 2015) (BSI, 2007; Podgorski, 2015; Wu et al., 2010)

AI1: The number of workers trained on an accident investigation AJ1: The number of accidents reports investigated

(Basso et al., 2004;Ramli et al., 2011; Podgorski, 2015) (BSI, 2007; Chang and Liang, 2009; Costella et al., 2009; Podgorski, 2015) (Basso et al., 2004; BSI, 2007; Podgorski, 2015; Wuet al., 2010) (Basso et al., 2004; BSI, 2007; Fernandez-Muniz et al., 2007;Chang and Liang, 2009;Costella et al., 2009; Reiman and Pietikainen, 2012; Podgorski, 2015) (BSI, 2007; Podgorski, 2015; Wu et al., 2010)

AK1: The number of accident reports sent to units

AL1: The number of advertisement on corrective and preventive actions AM1: The number of accident, near miss documents

(BSI, 2007; Fernandez-Muniz et al., 2007; Chang and Liang, 2009; Costella et al., 2009; Podgorski, 2015) (BSI, 2007; Fernandez-Muniz et al., 2007; Robson et al., 2007; Podgorski, 2015)

AN1: The number of review meetings carried out

(Basso et al., 2004; BSI, 2007; Jovasevic-Stojanovic and Stojanovic, 2009)

AO1: The number of OHS performance reports of units AP1: The number of recommendations for continual improvement AQ1: The number of managers of units in review meeting

(Fernandez-Muniz et al., 2007; Chang and Liang, 2009; Costella et al., 2009; Podgorski, 2015) (BSI, 2007; Fernandez-Muniz et al., 2007; Podgorski, 2015; Wu et al., 2010) (Robson et al., 2007; Wu et al., 2010)

EP

TE D

Management review

MA1

AH. Continuous reviewing of OHS audits AI. Workers involvement in accident investigation AJ. Reviewing and up-dating accident investigation

AC1: The number of units that their OHS performance has been evaluated AD1: Frequency Rate (FR), Severity Rate (SR), Incident Rate (IR) AE1: The number of units that have OHS report systems AF1: The number of OHS meetings held for workers about OHS issues AG1: The number of audits performed at a given period AH1: The number of recommendations applied

RI PT

CH2

AC. Measuring and monitoring based on risk assessment AD. Measuring and monitoring based on Lagging indicators AE. Record and control systems for OHS activities AF. Announcing results of OHS audits to workers AG. Deadline for OHS audits

(BSI, 2007; Chang and Liang, 2009; Costella et al., 2009; Wu et al., 2010)

SC

CH1

AB1: The number of units that have OHS representatives

M AN U

IM11

regular inspection and testing AB. Establishing organizational structure for OHS

Table 3 Example of reference levels for performance indicators of policy criteria Performance indicator

AC C

Criterion

Good

Neutral

Communicating OHS policy

Existence of OHS policy

≥ 90% units have OHS policy

70-89% units have OHS policy

Preliminary risk assessment

Risk assessment

Hazard identified in units and appropriate control measures have been implemented

Hazard identified in units and quite appropriate control measures have not been implemented

Workers’ participation

OHS suggestions

Number of suggestions for OHS improvement in a given reporting period against total number of workers is equal or more than 90%

Number of suggestions for OHS improvement in a given reporting period against total number of workers is between 70-89%

15

ACCEPTED MANUSCRIPT

4. Application

SC

RI PT

In the following section the introduced integrated decision making approach (ANP-TOPSIS) applied to assess the effectiveness of OHSMSs based on OHSAS18001 in the Iranian Power Plant Projects Management Company (MAPNA) is described. The MAPNA Company is one of the biggest certified OHSAS 18001 standard construction company, a general contractor for constructing power plant projects such as gas power plants, combined cycle power plants, utility project power plants and steam of refinery and petrochemical industries in Iran. 4.1. Population and sample

4.2 Survey instrument

TE D

M AN U

This case study was carried out in a combined cycle power plant construction projects performed by MAPNA Company in one of the southeastern provinces of Iran. The data for assessing the effectiveness of the OHSAS 18001 standard was gathered through OHS documents and data records for three years from 2011 to 2013. OHS experts in academia and industries with the necessary in-depth and wide knowledge on OHSMSs constituted the expert profile to determine the relative importance weights of criteria.

AC C

EP

In order to determine the relative importance weights of criteria, a questionnaire was designed. The questionnaire comprised an individual pairwise comparison matrix for each main element. Then, the questionnaires with a presentation letter to inform respondents about the purpose, aim and method of the study were submitted via e-mail to 35 OHS experts with at least 5 years of experience in OHSMSs as an educator, consultant or OHS manager in universities, petrochemical or refinery industries. Iranian petrochemical and refinery industries were chosen because they were the best industries in which OHSMSs have been implemented and properly monitored. Therefore, their OHS managers were familiar with OHSMSs and had practical, detailed knowledge about them. They were asked to assign a value of 1 to 9 as the relative importance of each criterion in comparison with the other criteria in the same main element. Comparison scale was a measurement 1-9 scale proposed by Saaty.

16

ACCEPTED MANUSCRIPT

4.3. Data collection

RI PT

The data for this case study was obtained from OHS documents and data records of MAPNA Company for a 3 year period, from 2011 to 2013. These data were gathered for each criteria based on performance indicators depicted in table 2.

SC

A total of 32 questionnaires were recovered. Six out of those questionnaires had been incorrectly completed and excluded from the study. Thus, only 26 questionnaires were used. The priority weights of criteria can be calculated in accordance with ANP procedure mentioned in section 3.1.3. But as this procedure was time consuming and needed complex calculations, SuperDecision software was used to facilitate all ANP analyses.

TE D

M AN U

Upon completion of data collection, the geometric means of the expert judgments were calculated through multiplying the judgments by one another and taking the /Q 26th root based on the following equation; LMN< = (∏Q , where LMN< is PG LMNP ) geometric means of i th criterion compared to j criterion, (ij) compared criteria, I number of experts and k number of experts. Then, to estimate weight of criteria with SuperDecision software, a decision making model based on experts opinion and structure of OHSAS 18001 standards was constructed and the element and criteria were grouped in related clusters, nodes and the dependency among them was determined (Fig.3a.).

AC C

EP

As can be seen, a node shows a component (or cluster) with criteria inside it; the interactions between two criteria or elements are indicated with a straight line and the inner dependence of criteria within an element is shown with a loop. However, the geometric means of the expert judgments for each pair of criteria were entered in priority matrices which SuperDecision software had constructed based on a research model and sets of pairwise comparisons. Fig. 3b. depicts the matrix for the four criteria of management review (MA1-MA4). Each judgment reflects the importance of each pair criteria. In the matrix of software, if the row criterion is more important than each criterion in the column, the value in cell is blue. In contrast, if the criterion in the column is more important than each criterion in the row, the value in cell is red. Then, the software automatically computed the priorities and the inconsistency of the judgments for each matrix. It is noteworthy that the bar chart of the matrix indicated in fig. 3b. shows the numerical values of the relative weights of the management review’s criteria with respect to the management review element. The software shows a value for inconsistency of the judgments above the bar chart. The inconsistency index 17

ACCEPTED MANUSCRIPT

RI PT

should be less than 0. 1(Ergu et al., 2014). As this index was found to be 0.01 in our study, it is concluded that the judgments are quite consistent. However, after completing all the comparison matrices, the software established unweighted, weighted and limit super matrix based on step 4 mentioned in section 3.3. Part of the unweighted, weighted, limit super matrix and priorities calculated by software are presented in figs. 3c, 3d, 3e and 3f, respectively.

TE D

M AN U

SC

Finally, the numerical priority values of criteria and the main elements of OHSAS 18001 standard were obtained from limit super matrix and the weight of criterion was computed by normalizing limit priorities of criteria in the elements. Thereafter, a decision matrix was constructed for each main element based on Eq. (2). In this stage, the data gathered from MAPNA Company and priorities weight of criteria calculated by ANP technique were applied to establish initial decision matrix for each of the main elements. The data that have negative effects on OHSAS 18001 standard effectiveness assessment (cost criteria) should be placed as reciprocal. For instance, accidents due to unsafe acts (PL6) demonstrate weak workers’ participation in OHS activities. Therefore, its value for 2011 is inserted reciprocally (1/81.5). Table 4 displays relative importance weights of the criteria and data records provided by the company. The data reveal the means of the indicators that have been performed or happened in a year. For example, on average, the number of rewards because of workers’ participation in risk assessment was 2. On average, out of all OHS programs planed 7 were carried out.

EP

Then, using the Eq. (3) the normalized decision matrix was established for each main element. In the next step, weighted normalized decision matrix for five main elements were constructed using Eq.(4). Thereafter, for each main element positive ideal solution (PIS) and negative ideal solution (NIS) was calculated using Eq.(5).

AC C

Table 5 shows an example of decision matrices, PIS and NIS for planning element. Then, separation measurement for PIS and NIS was calculated in accordance with Eqs. (6) and (7) and relative closeness of main elements were determined using the Eq.(8). Eventually, the alternatives were ranked in accordance with the descending order of their relative closeness. Table 6 provides relative closeness values for main elements and illustrates the effectiveness of each main element in the three years. In order to determine the critical criteria that affect the effectiveness of main elements, the distance values of criteria in each main element for alternative years to PIS were calculated. This section provides useful information for decision 18

ACCEPTED MANUSCRIPT

makers about critical issues in the effectiveness of OHSAS 18001 standard figs. 48.

RI PT

[Insert Fig. 3. around here]

5. Results

M AN U

SC

Content validity was used to support the validity of the criteria for each of the five elements. The CVR values of all of the criteria are summarized in table 4. It can be seen that all the criteria have CVR values higher than the desired value of 0.33. This demonstrates strong content validity of the proposed criteria. The results presented in table 4 show that alpha coefficient for all of the criteria in each of the five elements and all criteria exceed 0.7. This confirms the presence of a very good internal consistency between the criteria in each element.

AC C

EP

TE D

The weights of the criteria and values of performance indicators for three years are presented in table 4. Management commitment, workers participation, reviewing and up-dating OHS policies have the highest values in policy element, respectively. The results prove that these criteria are key safety management practices. According to the data in table 4, announcing OHS programs and objectives and allocating financial resources to OHS programs are the most important criteria in the planning element. As can be seen in table 4, allocating financial resources for emergency response plan and workers’ participation in OHS activities are significant criteria in the implementation and operation element. As table 4 shows measuring and monitoring based on OHS indicators and having record, report and analyses systems for accidents have an important role in the checking element. Based on the results shown in table 4, available OHS activities results and the use of OHS indicators in review meetings have the highest weight in the management review element. Thus, they seem to be significant success factors on the effectiveness and continuous improvement of OHSAS 18001 standard. In order to evaluate the effectiveness of OHSAS 18001 standard, it would be better to review the results based on the main elements. Table 6 presents the 19

ACCEPTED MANUSCRIPT

RI PT

OHSAS 18001 standard effectiveness evaluation results for the main elements for the period of 2011 to 2013, separately. Based on the results, the effectiveness of the main elements of the OHSAS 18001 standard in 2013 compared with 2011 and 2012 has the highest value. It seems that the effectiveness of the OHSAS 18001 standard in 2013 has an increasing trend. However, there is a significant decrease in 2011 in terms of effectiveness of OHSAS 18001 standard.

AC C

EP

TE D

M AN U

SC

Figs. 4-8 illustrate distances of criteria to PIS for each of the main elements in alternative years. Apparently, in fig. 4 PO5 (workers’ participation) is the most significant criterion to take into consideration in 2011 and 2012 policies. Then, PO3 (reviewing and updating policy) is a significant criterion in policy 2013. Fig. 5 depicts that Pl4 (communication OHS activities) in 2011 and 2012 and PL 9 (financial resources) in 2011 are the significant criteria in planning element in order to improve the effectiveness of OHSAS 18001 standard. In 2011, IM1 (workers training on OHS responsibility), in 2012 IM3 (communicating OHS activities) and IM4 (workers’ participation) are found as critical aspects of implementation elements according to fig. 6. Analysis of fig. 7 shows the critical criteria in the checking element. The criteria such as CH2 (measurement and monitoring based on OHS indicators), CH7 (workers’ participation in accident investigation), CH9 (communication accident investigation results) and CH11 (having accident record, report and analysis system) in 2011 and CH7 and CH11 in 2012 are the key aspects in order to increase the effectiveness of OHSAS 18001 standard. Finally, fig. 8 points out that MA2 (having OHS results during review meetings) is the most remarkable criterion in 2011. As table 6 shows, the most successful year regarding the effectiveness of OHSAS 18001 elements is 2013. Almost all of the criteria in this year cumulate in the centre of figs. 4-8.

20

ACCEPTED MANUSCRIPT

Table 4. Priority, validity, reliability, and indicators’ data of OHSAS 18001 criteria

0.93 0.87 0.87 0.93 1 0.8 0.87 0.93 0.87

0.76

0.86

EP

0.8 1 0.8 0.93 1

0.93 1 1 0.87 0.93 0.93 1 1 0.87 0.73 0.73

Data on indicators priorities

2011

2012

RI PT

0.83

TE D

1 0.93 0.93 0.8 1 0.93

Total cronbach's α coefficient

0.89

0.95

2013

0.2158 0.1191 0.1748 0.0964 0.193 0.0957 0.0662 0.0383

0.65 0.75 0.75 1 0.5 0.5 0.3 1

0.925 0.75 0.665 1 1 0.5 0.5 1

0.95 1.3 0.125 1 2 0.6 0.5 1

0.0727 0.0966 0.0947 0.0716 0.0763 0.0438 0.0365 0.3361 0.1712

0.67 2 0.5 22.66 160.3 81.5 7 1 588.24

0.33 4 0.5 68 188.3 79.36 9 1 694.12

0.67 3 0.5 96.33 187.66 80.9 8 1 1088.24

0.0784 0.0667 0.0746 0.2282 0.0372 0.0251 0.2023 0.0413 0.0743 0.0765 0.0949

163.66 23 2 56.5 56 20.67 90.33 1 10 1 0.75

409 25 2 65 63 25 97 1 12 1 0.8

466.67 30 4 60 71 33.67 102.67 1 13 1 0.9

0.0345 0.1847 0.1195 0.0178 0.0206 0.1009 0.1297 0.0366 0.1135 0.0979 0.1437

0.1 102.3 0.1 1 3 5.5 5 4.5 0.5 0.5 0.4

0.15 71.713 0.1 1 3 5 7.5 5.5 1 0.5 0.5

0.2 75.82 0.1 1 3 6.5 10 8 1 0.5 0.7

0.199 0.415 0.218 0.166

3 0.2 10 0.4

3 0.3 10 0.3

3 0.3 12 0.4

SC

1 0.87 0.87 0.8 0.87 0.8 0.93 0.87

AC C

Policy(PO) PO1 PO2 PO3 PO4 PO5 PO6 PO7 PO8 Planning(PL) PL1 PL2 PL3 PL4 PL5 PL6 PL7 PL8 PL9 Implementation(IM) IM1 IM2 IM3 IM4 IM5 IM6 IM7 IM8 IM9 IM10 IM11 Checking(CH) CH1 CH2 CH3 CH4 CH5 CH6 CH7 CH8 CH9 CH10 CH11

Cronbach's α coefficient

M AN U

Content Validity Ratio (CVR)

Criteria Code

Management Review (MA) MA1 MA2 MA3 MA4

0.73 0.93 0.87 0.93

0.7

21

ACCEPTED MANUSCRIPT

Table 5 Calculation of weighted normalized decision matrix, positive and negative ideal solutions for planning element of OHSAS 18001 standard.

0.072768 0.096602 0.094765 0.071602 0.076369 0.043881 0.036574 0.336145 0.171293

2011 0.67 2 0.5 22.67 160.33 0.0123 7 1 588.24

2012 0.33 4 0.5 68 188.33 0.0126 9 1 694.12

2013 0.67 3 0.5 96.33 187.67 0.0124 8 1 1088.24

SC

PL1 PL2 PL3 PL4 PL5 PL6 PL7 PL8 PL9

Weighted normalized decision matrix 2011 2012 2013 0.048512 0.024256 0.048512 0.035877 0.071754 0.053816 0.054713 0.054713 0.054713 0.013516 0.040549 0.057445 0.039438 0.046325 0.046161 0.025047 0.02572 0.025233 0.018381 0.023633 0.021007 0.194073 0.194073 0.194073 0.071034 0.083821 0.131414

M AN U

Weight

Table 6

Positive and negative ideal solution Max Min 0.048512 0.024256 0.071754 0.035877 0.054713 0.054713 0.057445 0.013516 0.046325 0.039438 0.02572 0.025047 0.023633 0.018381 0.194073 0.194073 0.131414 0.071034

RI PT

Normalized decision matrix 2011 2012 2013 0.67 0.33 0.67 0.37 0.74 0.56 0.58 0.58 0.58 0.19 0.57 0.8 0.52 0.61 0.60 0.57 0.59 0.58 0.50 0.65 0.57 0.58 0.58 0.58 0.415 0.49 0.77

Initial decision matrix

Criteria code

Distance calculation, relative closeness and precedence of main elements of OHSAS 18001 standard for the period of 2011 to 2013. Planning

Implementation

TE D

Policy

Management Review

Checking

2012

2013

2011

2012

2013

2011

2012

2013

2011

2012

2013

2011

2012

2013

s+

0.141

0.095

0.108

0.083

0.056

0.018

0.054

0.038

0.008

0.08

0.04

0.006

0.092

0.035

0

SC* precedence

0.108

0.11

0.141

0.024

0.047

0.081

0.018

0.032

0.055

0.005

0.06

0.08

0.026

0.088

0.095

0.434

0.54

0.56

0.23

0.46

0.82

0.25

0.45

0.87

0.06

0.6

0.93

0.22

0.72

1

3

2

1

3

2

1

3

2

1

3

2

1

3

2

1

EP

2011

AC C

[Insert Fig. 4. around here]

[Insert Fig. 5. around here] [Insert Fig. 6. around here] [Insert Fig. 7. around here] [Insert Fig. 8. around here]

22

ACCEPTED MANUSCRIPT

6. Discussion

RI PT

Occupational health and safety management systems have been recommended as important tools for preventing and reducing work-related risks and promoting continuous improvement of the conditions of work environments. However, despite their wide application, concern exists as to their effectiveness in reducing occupational risks at workplace (Hasle and Zwetsloot., 2011).

SC

This might be due to improper audits with too much paper work and inappropriate measurement tools which are based on subjective and intangible data (Blewett and O’Keeffe, 2011; Cadieux et al., 2006; Chang and Liang, 2009).

M AN U

Auditors and certification institutes check systems with relevant criteria in order to ensure whether or not they comply with those criteria and do not perfectly consider key OHS performance aspects of OHSMSs. Robson et al. (2012) pointed out that audit methods used by organizations have differences in the management elements, time, cost, auditor training and organization activities. These cause wide variation and complexity in measuring and assessing of OHSMSs (Robson et al., 2012).

EP

TE D

Each aspect of OHS issues should be monitored and measured by appropriate criteria and indicators. Therefore, considering a single aspect of OHSMSs can be inadequate and misleading. It is like an illness which is associated with relevant specific symptoms through which the illness is recognized and treated. For instance, a person who has a broken leg, his or her blood pressure and body temperature may be normal while his or her leg is broken (Ale, 2009).

AC C

Thus, methods and tools are appropriate for assessing the effectiveness OHSMSs that comprise all of the measurement of OHSMSs performance approaches. The current study provided appropriate criteria and KPIs for assessing and improving the effectiveness of OHSAS 18001 standard. The validity and reliability of these criteria have been demonstrated.

As a result, these criteria and KPIs could be considered as useful means for evaluating the quality of a system in terms of development, implementation, and results. In addition, these criteria and KPIs might impress decision making in OHS 23

ACCEPTED MANUSCRIPT

issues and help in predicting future status and planning. Monitoring and comparing existing values of these criteria and KPIs with a reference level or determined target levels, allow managers to obtain a picture from the current status of a system and assist them to understand how the system operates.

TE D

M AN U

SC

RI PT

We found and ranked the main criteria in each of the elements of OHSAS 18001. Based on the obtained weights of criteria, the most influential factors to be taken into account to improve the effectiveness of OHSAS 18001 standard were found to be management commitment, workers’ participation, allocation of financial resources to OHS programs, announcement of the objectives and the activities of OHS programs, provision of OHS indicators, risk assessment, workers involvement in accident investigation, availability of record, report and analysis system for OHS issues, availability of OHS activities results and availability of OHS indicators during reviewing. The findings of this study are in line with the studies carried out by Abad et al. (2013) and Fernandez-Muniz et al. (2007). They also found that effective and successful implementation of OHSMS requires management commitment. This factor reflects the importance of management attitude and behavior toward OHS. In fact, the more committed management provides enough support and resources to safety, helps to create a safe environment, participates in meetings and injury investigation committees, communicates the importance of safety and so on (Wu et al., 2010).

AC C

EP

Similarly, workers’ participation is another decisive and important factor in successful implementation of OHSMS. Employees can help to create a safe condition through activities such as presence in safety training, participating in voluntary OHS activities and OHS related decision making and identifying OHS related issues (Vinodkumar and Bhasi, 2011). Our findings also indicate that comminucation and dissemination of information about OHS issues are effective components of OHSMS that might influence safety performance. A conclusion that other investigators have also reached (FernandezMuniz et al., 2007). These components of OHSMS increase workers’ awareness about risks and hazards of workplace and improve workers’ safety knowledge and safety skills

24

ACCEPTED MANUSCRIPT

(Wu Lin et al., 2010). Therefore, workers participate in OHS activities more actively and do safe acts and these have significant effects on success of OHSMS.

RI PT

As far as the issue of resource allocation was concerned, our results revealed that this issue is also an important component of OHSMS that can be taken into account to improve the effectiveness of OHSMSs and, successful implementation of OHSMS relies on allocating enough resources. This finding is also in full agreement with those of Wu et al.,(2010).

SC

Regarding risk assessment, our results are in accordance with those of Ramli et al. (2011) who found that hazard identification and risk assessment are major factors that affect successful implementation of OHSMS.

M AN U

Our findings also indicated that another influential factor to be taken into account to improve the effectiveness of OHSAS 18001 standard is the presence of an incident reporting and analysis system. Others have also emphasized on the importnce of proper incident reporting and analysing as an important tool in OHSMS (Frazier et al., 2013). This provides information that can be used in the learning process in order to avoid recurrence of incidents (Lindberg et al., 2010).

AC C

EP

TE D

Moreover, our findings are consistent with those of Fernandez-Muniz et al. (2007) who showed that controling and reviewing activities are key aspects of a good OHSMS. These factors are two important tools that provide information about OHS performance and quality of OHSMS. Hence, it is plausible to assume that the effectiveness of OHSMS is also strongly affected by these criteria. Therefore, these criteria and relevant indicators can be appropriate and effective benchmarks for assessing and promoting the effectiveness of OHSMSs. The criteria recognized to be less important due to their low weights have their own roles or effects on the effectiveness of OHSMSs and cannot simply be ignored. Our findings also indicate that assessing the effectiveness of occupational health and safety management systems should not be restricted to outcome results such as the frequency of accidents at work and occupational diseases, rather attention should be turned to intangible outcomes particularly safety culture such as internal communication about risks and hazards, workers’ participation, workers’ 25

ACCEPTED MANUSCRIPT

awareness of safety conditions and so forth. Similar views have been expressed by Abad et al., 2013.

RI PT

We believe that considering all of these factors by proper strategies and tactics will have a highly positive impact on the effectiveness of OHSMSs and will improve the functioning of OHSMSs through identifying deficiencies in OHS practices.

SC

One of the most important characteristics of this study is the fact that the proposed approach for evaluating and improving the effectiveness of OHSMS is proactive and covers broad scope of OHS issues. This could predict the future of OHSMSs and could help developing more efficient and effective actions and plans for achieving OHS objectives before accidents happen.

TE D

M AN U

The results of this case study also showed that the framework developed in this study is a comprehensive tool and enables managers to provide appropriate information for continuous improvement of OHSMSs from strategic issues ( lack of management commitment) to operational ones (workers training and communication OHS activities). They also demonstrate the importance of the discussed criteria in successful implementation of OHSMS. As shown in the results section, in 2013, OHSAS 18001 standard was more effective than the two preceding years because the number of criteria that had the highest values to PIS in 2013 was lower than those in 2011 and 2012.

AC C

EP

The results of the study also demonstrated that implementation of OHSMS alone is not adequate and it is not a remedy for all OHS issues, per se. Therefore, in order to increase and promote the effectiveness of OHSAS 18001 standard, the criteria which have the highest values to PIS should be taken into account and modified. It is necessary to take actions to monitor such criteria in the future. These criteria could be revised based on performance indicators depicted in table 3. The findings of this case study are also consistent with those of Zeng et al. (2008) and Abad et al. (2013) who found that implementation of OHSAS 18001 standard is a gradual process and its positive outcomes on OHS performance achieve in the long run. Finally, it is noteworthy that the application of the proposed method requires only a short time and it doesn’t need specific training for auditors. These important features of the proposed method are consistent with those suggested for a good method of OHSMSs auditing (Robson et al., 2012). 26

ACCEPTED MANUSCRIPT

6.1. Limitation of study and recommendation for further research

M AN U

SC

RI PT

The relative importance weights of criteria were determined by experts who reside in Iran. Therefore, the weights of criteria are expected to have been affected by the safety culture and practice in this region. Further research is clearly needed if the proposed method is to be applied elsewhere. Another limitation of this study is that it only considers OHSAS 18001 standard. Therefore, the criteria and relevant indicators should be developed for other OHSMSs. Considering other OHSMSs, it would allow researchers to provide criteria and indicators for their future work. The criteria and KPIs proposed in this study cannot be considered as definite criteria and KPIs for assessing the performance of OHSMS in industries.

7. Conclusion

AC C

EP

TE D

The aims of this research were to develop a decision making approach for assessing and improving the effectiveness of occupational health and safety management systems and identifying the influential factors and their effects on OHSMS effectiveness. In this study, the proposed method, ANP-TOPSIS, contributes to filling the gap regarding the lack of rigorous methodology for assessing and promoting the effectiveness of the occupational health and safety management systems by means of identifying and ranking influential factors of OHSMSs. The most influential factors of each OHSMS element were identified based on their relative importance weights. This work showed that the criteria with higher weight have large contributions to improving the effectiveness of OHSAS 18001 standard. The most important characteristic of the proposed approach is its proactiveness and its coverage of a broad scope of OHS issues. The approach verifies that for an OHSMS to be implemented successfully and to work properly, the criteria for each of the elements that have the highest distance values to PIS must be taken into account. The key results of the study indicate that determining values of criteria generate useful knowledge about OHSMS status and facilitate interpretation of OHSMS effectiveness. This case study also showed that the proposed method can be used for construction industry. As previous researches note that safety and health management systems are able to improve safety performance indicators such as the safety climate, safety culture, personal injuries, labor absenteeism etc, we suggest that the validity of the proposed method could 27

ACCEPTED MANUSCRIPT

RI PT

be further evaluated in a large number of same workplaces(companies) from several industrial sectors using safety performance data (safety climate , safety culture , personal injuries, labor absenteeism, total accident rate, lost hours per worker ,etc.) for a period of two to three years by future studies. Finally, the method developed in this study could be used to determine factors that affect the effectiveness of OHSMSs. Acknowledgments

SC

This work was supported by Grant No: 9112084399 from Hamadan University of Medical Sciences.

References

TE D

M AN U

The authors would like to thank the OHS managers of MAPNA Company and petrochemical industries in Assaloyeh for their honest cooperation in responding to the questionnaires. Similarly, we are grateful to Ms Rozhan Saaty for her skillful assistance in developing the ANP model.

Abad, J., Lafuente, E., Vilajosana, J., 2013. An assessment of the OHSAS 18001 certification process: Objective drivers and consequences on safety performance and labour productivity. Safety Science. 60, 47-56. http://dx.doi.org/10.1016/j.ssci.2013.06.011

EP

Aksorn, T., Hadikusumo, B., 2008. Critical success factors influencing safety program performance in Thai construction projects. Safety Science. 46, 709-727. http://dx.doi.org/10.1016/j.ssci.2007.06.006

AC C

Akyuz, E., Celik, M., 2014. A hybrid decision-making approach to measure effectiveness of safety management system implementations on-board ships. Safety Science. 68, 169-179. http://dx.doi.org/10.1016/j.ssci.2014.04.003

Ale, B., 2009. More thinking about process safety indicators. Safety Science. 47, 470-471. http://dx.doi.org/10.1016/j.ssci.2008.07.012

Bana e Costa, C. A., Correa, E. C., De Corte, J. M., Vansnick, J. C., 2002. Facilitating bid evaluation in public call for tenders: a socio-technical approach. Omega. 30(3), 227-242.doi:10.1016/S0305-0483(02)00029-4 Basso, B., Carpegna, C., Dibitonto, C., Gaido, G., Robotto, A., Zonato,C., 2004. Reviewing the safety management system by incident investigation and performance indicators. Journal of Loss Prevention in the Process Industries. 17(3), 225-231.doi:10.1016/j.jlp.2004.02.004 Blewett, V., O’Keeffe, V., 2011. Weighing the pig never made it heavier: Auditing OHS, social auditing as verification of process in Australia. Safety science. 49, 1014-1021. http://dx.doi.org/10.1016/j.ssci.2010.12.010 28

ACCEPTED MANUSCRIPT

BSI, 2007. OHSAS18001: 2007 Occupational Health and Safety Management Systems Requirements, Occupational Health and Safety Assessment Series. British Standards Institution. London.

RI PT

Cadieux, J., Roy, M., Desmarais, L., 2006. A preliminary validation of a new measure of occupational health and safety. Journal of Safety Research. 37, 413-419. http://dx.doi.org/10.1016/j.jsr.2006.04.008 Cagno, E., Micheli, G.J.L., Jacinto, C., Masi, D., 2014. An interpretive model of occupational safety performance for Small- and Medium-sized Enterprises. International Journal of Industrial Ergonomics. 44, 60-74. http://dx.doi.org/10.1016/j.ergon.2013.08.005

SC

Cagno, E., Micheli, G.J.L., Perotti, S., 2011. Identification of OHS-related factors and interactions among those and OHS performance in SMEs. Safety Science. 49, 216-225. http://dx.doi.org/10.1016/j.ssci.2010.08.002

M AN U

Chang, J.I., Liang, C.-L., 2009. Performance evaluation of process safety management systems of paint manufacturing facilities. Journal of Loss Prevention in the Process Industries. 22, 398402.http://dx.doi.org/10.1016/j.jlp.2009.02.004 Chen, C.-Y., Wu, G.-S., Chuang, K.-J., Ma, C.-M., 2009. A comparative analysis of the factors affecting the implementation of occupational health and safety management systems in the printed circuit board industry in Taiwan. Journal of Loss Prevention in the Process Industries. 22, 210-215. http://dx.doi.org/10.1016/j.jlp.2009.01.004

Costella, M.F., Saurin, T.A., de Macedo Guimaraes, L.B., 2009. A method for assessing health and safety management systems from the resilience engineering perspective. Safety Science. 47, 1056-1067. http://dx.doi.org/10.1016/j.ssci.2008.11.006

TE D

Dagdeviren, M., Yuksel, I., Kurt, M., 2008. A fuzzy analytic network process (ANP) model to identify faulty behavior risk (FBR) in work system. Safety Science. 46, 771-783. http://dx.doi.org/10.1016/j.ssci.2007.02.002

Ergu, D., Kou, G., Shi, Y., Shi, Y., 2014. Analytic network process in risk assessment and decision analysis.Computers and Operations Research. 42, 58-74. http://dx.doi.org/10.1016/j.cor.2011.03.005

AC C

EP

Fernandez-Muniz, B., Montes-Peon, J.M., Vazquez-Ordas, C.J., 2007. Safety management system: Development and validation of a multidimensional scale. Journal of Loss Prevention in the Process Industries. 20, 52-68. http://dx.doi.org/10.1016/j.jlp.2006.10.002 Fernandez-Muniz, B., Montes-Peon, J.M., Vazquez-Ordas, C.J., 2012a. Occupational risk management under the OHSAS 18001 standard: analysis of perceptions and attitudes of certified firms. Journal of Cleaner Production. 24,36-47. http://dx.doi.org/10.1016/j.jclepro.2011.11.008 Fernandez-Muniz, B., Montes-Peon, J.M., Vazquez-Ordas, C.J., 2012b. Safety climate in OHSAS 18001certified organisations: antecedents and consequences of safety behaviour. Accident Analysis and Prevention. 45, 745-758. http://dx.doi.org/10.1016/j.aap.2011.10.002 Frazier, C. B., Ludwig, T. D., Whitaker,B., Roberts, D. S., 2013. A hierarchical factor analysis of a safety culture survey. Journal of Safety Research. 45: 15-28.doi:10.1016/j.jsr.2012.10.015

29

ACCEPTED MANUSCRIPT

Gallagher, C., Underhill, E., Rimmer, M., 2001. Review of the effectiveness of OHS management systems in securing healthy and safe workplaces. a report prepared for the National Occupational Health and Safety Commission, March, Victorian University of Technology, Melbourne.

RI PT

Granerud, R.L., Rocha, R.S., 2011. Organisational learning and continuous improvement of health and safety in certified manufacturers. Safety Science. 49, 1030-1039. http://dx.doi.org/10.1016/j.ssci.2011.01.009 Guerrero-Baena, M.D., Gomez-Limon, J.A., Fruet, J.V., 2014. A multicriteria method for environmental management system selection: an intellectual capital approach. Journal of Cleaner Production. In press. http://dx.doi.org/10.1016/j.jclepro.2014.07.079

Hamidi, N., Omidvari, M., Meftahi, M., 2012. The effect of integrated management system on safety and productivity indices: Case study; Iranian cement industries. Safety Science. 50, 1180-1189. Harms-Ringdahl,

L.,

2009.

Dimensions

in

safety

http://dx.doi.org/10.1016/j.ssci.2008.07.019

SC

http://dx.doi.org/10.1016/j.ssci.2012.01.004

indicators.

Safety

Science.

47,481-482.

M AN U

Hasle, P., Zwetsloot, G., 2011. Editorial: Occupational Health and Safety Management Systems: Issues and challenges. Safety Science. 49,961-963.http://dx.doi.org/10.1016/j.ssci.2011.02.013 Hinze, J., Thurman, S., Wehle, A., 2013. Leading indicators of construction safety performance. Safety Science. 51(1), 23-28.doi:10.1016/j.ssci.2012.05.016 Hohnen, P., Hasle, P., 2011. Making work environment auditable – A critical case study of certified occupational health and safety management systems in Denmark. Safety Science. 49, 10221029.http://dx.doi.org/10.1016/j.ssci.2010.12.005

TE D

Hsu, C.-W., Hu, A.H., 2009. Applying hazardous substance management to supplier selection using analytic network process. Journal of Cleaner Production. 17, 255264.http://dx.doi.org/10.1016/j.jclepro.2008.05.004

EP

Huang, R.-H., Yang, C.-L., Kao, C.-S., 2012. Assessment model for equipment risk management: Petrochemical industry cases. Safety science. 50, 1056-1066.http://dx.doi.org/10.1016/j.ssci.2010.02.024 ILO, 2003. Safety in numbers: pointers for the global safety at work. International Labor Office. Geneva.Switzerland.

AC C

Ismail, Z., Doostdar, S., Harun, Z., 2012. Factors influencing the implementation of a safety management system for construction sites. Safety Science. 50, 418-423.http://dx.doi.org/10.1016/j.ssci.2011.10.001 Jovasevic-Stojanovic, M., Stojanovic, B., 2009. Performance indicators for monitoring Safety Management Systems in chemical industry. Chemical Industry and Chemical Engineering Quarterly. 15(1), 5-8. Laitinen, H., Vuorinen, M., Simola, A., Yrjanheikki, E., 2013. Observation-based proactive OHS outcome indicators–Validity of the Elmeri method. Safety science. 54, 6979.http://dx.doi.org/10.1016/j.ssci.2012.11.005 Levy, J.K., Taji, K., 2007. Group decision support for hazards planning and emergency management: A Group Analytic Network Process (GANP) approach. Mathematical and Computer Modelling. 46, 906917.http://dx.doi.org/10.1016/j.mcm.2007.03.001 30

ACCEPTED MANUSCRIPT

Lin, C.-T., Chen, C.-B., Ting, Y.-C., 2011. An ERP model for supplier selection in electronics industry. Expert Systems with Applications. 38,1760-1765.http://dx.doi.org/10.1016/j.eswa.2010.07.102 Lindberg, A-K., Hansson, S. O., Rollenhagen, C., 2010. Learning from accidents–what more do we need to know?. Safety Science. 48(6), 714-721.doi:10.1016/j.ssci.2010.02.004

RI PT

Mengolini, A., Debarberis, L., 2008. Effectiveness evaluation methodology for safety processes to enhance organisational culture in hazardous installations. Journal of hazardous materials. 155, 243252.http://dx.doi.org/10.1016/j.jhazmat.2007.11.078 Øien, K., Utne, I.B., Herrera, I.A., 2011. Building safety indicators: Part 1–theoretical foundation. Safety Science. 49,148-161.http://dx.doi.org/10.1016/j.ssci.2010.05.012

SC

Podgórski, D., 2015. Measuring operational performance of OSH management system– A demonstration of AHP-based selection of leading key performance indicators. Safety Science. 73, 146166.http://dx.doi.org/10.1016/j.ssci.2014.11.018

M AN U

Pourjavad, E., Shirouyehzad, H., Shahin, A., 2013. Selecting maintenance strategy in mining industry by analytic network process and TOPSIS. International Journal of Industrial and Systems Engineering. 15,171-192.http://dx.doi.org/10.1504/IJISE.2013.056095 Ramli, A.A., Watada, J., Pedrycz, W., 2011. Possibilistic regression analysis of influential factors for occupational health and safety management systems. Safety Science. 49,11101117.http://dx.doi.org/10.1016/j.ssci.2011.02.014

TE D

Reiman, T., Pietikainen, E., 2012. Leading indicators of system safety-monitoring and driving the organizational safety potential. safety science. 50(10), 1993-2000. http://dx.doi.org/10.1016/j.ssci.2011.07.015 Robson, L.S., Clarke, J.A., Cullen, K., Bielecky, A., Severin, C., Bigelow, P.L., Irvin, E., Culyer, A., Mahood, Q., 2007. The effectiveness of occupational health and safety management system interventions: A systematic review. Safety Science. 45, 329-353.http://dx.doi.org/10.1016/j.ssci.2006.07.003

AC C

EP

Robson, L.S., Macdonald, S., Gray, G.C., Van Eerd, D.L., Bigelow, P.L., 2012. A descriptive study of the OHS management auditing methods used by public sector organizations conducting audits of workplaces: Implications for audit reliability and validity. Safety Science. 50, 181189.http://dx.doi.org/10.1016/j.ssci.2011.08.006 Saaty, T.L., 2005. Theory and applications of the analytic network process: Decision Making with Benefits, Opportunities, Costs, and Risks. RWS publications. Pittsburgh, PA, USA. Santos, G., Barros, S., Mendes, F., Lopes, N., 2013. The main benefits associated with health and safety management systems certification in Portuguese small and medium enterprises post quality management system certification. Safety Science. 51, 29-36.http://dx.doi.org/10.1016/j.ssci.2012.06.014 Shafiee, M., 2015. A fuzzy analytic network process model to mitigate the risks associated with offshore wind farms. Expert Systems with Applications. 42, 2143-2152.http://dx.doi.org/10.1016/j.eswa.2014.10.019 Sun, C.-C., 2010. A performance evaluation model by integrating fuzzy AHP and fuzzy TOPSIS methods. Expert Systems with Applications. 37, 7745-7754.http://dx.doi.org/10.1016/j.eswa.2010.04.066

31

ACCEPTED MANUSCRIPT

Takala, J., Hamalainen, P., Saarela, K.L., Yun, L.Y., Manickam, K., Jin, T.W., Heng, P., Tjong, C., Kheng, L.G., Lim, S., Lin, G.S., 2014. Global estimates of the burden of injury and illness at work in 2012. Journal of Occupational and Environmental Hygiene. 11, 326337.http://dx.doi.org/10.1080/15459624.2013.863131

RI PT

Teo, E.A.L., Ling, F.Y.Y., 2006. Developing a model to measure the effectiveness of safety management systems of construction sites. Building and Environment. 41, 15841592.http://dx.doi.org/10.1016/j.buildenv.2005.06.005 Vinodkumar, M.N., Bhasi, M., 2011. A study on the impact of management system certification on safety management. Safety Science. 49, 498-507.http://dx.doi.org/10.1016/j.ssci.2010.11.009

SC

Wu, T. C., Lin, C.-H., Shiau, S. Y., 2010.Predicting safety culture: The roles of employer, operations manager and safety professional.Journal of safety research. 41(5), 423-431.doi:10.1016/j.jsr.2010.06.006 Wurzelbacher, S., Jin, Y., 2011. A framework for evaluating OSH program effectiveness using leading and trailing metrics. Journal of Safety Research. 42,199-207.http://dx.doi.org/10.1016/j.jsr.2011.04.001

M AN U

Zeng, S., Shi, J. J., Lou, G., 2007. A synergetic model for implementing an integrated management system: an empirical study in China. Journal of cleaner production. 15(18), 17601767.doi:10.1016/j.jclepro.2006.03.007

AC C

EP

TE D

Zeng, S., Tam, V.W., Tam, C.M., 2008. Towards occupational health and safety systems in the construction industry of China. Safety science. 46, 1155-1168.http://dx.doi.org/10.1016/j.ssci.2007.08.005

32

ACCEPTED MANUSCRIPT

List of figures captions

Fig.1. Schematic diagram of the proposed methodology for assessing the effectiveness of OHSMSs.

RI PT

Fig.2. ANP-based network model for evaluating weights of OHSAS 18001 criteria.

Fig.3. Example of weighting process of OHSAS 18001 criteria with SuperDecision software. Fig.4. OHSAS 18001 standard overview based on policy criteria for the period of 2011 to 2013.

SC

Fig.5. OHSAS 18001 standard overview based on planning criteria for the period of 2011 to 2013.

M AN U

Fig.6. OHSAS 18001 standard overview based on implementation criteria for the period of 2011 to 2013. Fig.7. OHSAS 18001 standard overview based on checking criteria for the period of 2011 to 2013.

AC C

EP

TE D

Fig.8. OHSAS 18001 standard overview based on management review criteria for the period of 2011 to 2013.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Highlights 1- We developed a comprehensive tool that enables managers to provide useful information for continuous improvement of OHSAS18001.

RI PT

2- The proposed method covers broad scope of OHS issues and could predict future of OHSAS18001.

3- We determined usable criteria and related indicators for assessing OHSAS18001 effectiveness.

AC C

EP

TE D

M AN U

SC

4- OHSAS18001 implementation is a gradual process.