The increasing rate of accidents in recent years Tabibi J,et al. Accidents are the outcomes of unsafe mechanical conditions or unsafe practices, so usually, they are not the product of a single error of one person but instead develop through a series of interactive elements involved at various levels of the system Tabibi J,et al.
Over the years, many theories have been propounded to specifically explain the cause of accidents; a brief review on the evolution of most notable theories in this regard is presented below: A The era of technical errors: the development of mechanical systems brought about a rapid growth in rates of work accidents. B The era of human errors: In this period, the investigations on the cause of accidents showed the importance of human-related factors. The accident in a nuclear power plant in Iceland revealed the shortcomings of human factors, most notably the psychological weakness.
In this period, the "human" element was identified as the weak link, or the component directly responsible for unsafe practices Rochlin, G. C The era of technical-social errors: In this period, the reaction of humans to "technical" errors was declared as the cause of fatal errors and subsequent accidents.
D The era of organizational culture: over the recent years, a new approach, which puts great emphasis on organizational culture, has emerged. It should be noted that due to the limited ability of human cognition, decision-makers and managers working in complex systems tend to develop a silo mentality Sterman, ; Senge, ; Forrester, Meadows, Silo approach encourages managers to solve the problems in their area of responsibility by focusing on internal components.
These solutions are often ineffective in the long run and may create counter-intuitive behavior in the entire system Checkland and Scholes, ; Senge, So to gain a proper understanding of the system and conduct a precise OHS performance evaluation, all components of the organization and their interactions should be assessed in tandem with each other. To achieve this goal, the researcher must first identify the variables and elements affecting the safety culture and performance.
The UK Advisory Committee on nuclear safety has published a report on influence of Human Factors on this issue, and therein it has defined a set of parameters as the key indices of safety culture; these parameters include: management commitment, management style, management visibility, communication, pressure for production, training, housekeeping, job satisfaction and workforce composition Flin,et al.
In a study conducted by the Irish Aviation Administration , aspects such as safety objectives and policies, safety risk management, commitment to safety, and safety promotion were considered as the main components of organizational safety culture State Safety Plan What that reflects the ambiguity of OHS performance evaluation is the still very high count of industrial accidents despite extensive efforts to improve safety conditions.
The main purpose of this study was to use system dynamics model and particularly quantitative analysis via simulation and scenario building to examine the effect of safety performance and culture on the rate of accidents.
This method is one of the Appropriate methods to simulate complex systems is based on causal relationships that enabling the system to provide appropriate learning by providing an environment for testing various scenarios. This method can be also used to analyze the problems with the qualitative approach causal analysis and quantitative approach stock and flow. In order to solve a problem by system dynamics we need to pursue five steps below: 1. Identifying and defining the problem 2. Mapping cause and effect diagrams 3.
Developing the mathematical model stock and flow diagram 4. Model simulation and validation 5. Scenarios generation and evaluation, then selecting and implementing the most appropriate solution System dynamics is often employed to analyze complex social and economic systems; as these systems dynamically change due to many unknown causes.
Sterman describes the steps of system dynamics modeling according to figure 1. So to identify the important parameters in an accurate manner, this study used variables provided by literature and especially those identified by Yang Miang Goh et al. Experts contributing to this study included: 2 safety managers with 10 and 15 years of experience, safety manager of Organization A, 3 safety expert with more than 15 years of experience, 3 safety experts of organization A, and 1 safety inspector with 7 years of experience.
The authors, who were familiar with the system dynamics literature, provided these experts with several articles about safety culture dynamics. After a period of 10 days, during which articles were read by experts, a meeting was held at the office of Organization A. Therefore, after discussion and addition and removal of a number of variables, the main variables of the study were identified.
In addition, to convert a qualitative model to a quantitative one, we needed to formulate the relationships and quantify the qualitative variables. To determine a level for each qualitative variable and turn it into a quantity, we needed to perform a survey at the organizational level. For this survey, the training manager, the production manager, and the members of HSE committee were added to the team of experts.
This survey team acquired all documents related to production, hazard detection, risk assessment, statistics, reports, and analyses pertaining to the accident of past 10 years, and training programs of last 5 years, and the list of budgets planned and allocated to OHS applications. Survey plan was announced to all units. Survey process included field studies and analysis of documents. Given the size and scope of activities, the survey process took about 3 days. At this point, a quantification process was needed to determine the current status of constant qualitative variables.
So, after the survey, members of the survey team held another meeting. After discussing the results of the survey, this team proposed a scheme for quantifying the qualitative variables. This scheme is shown in Table 1. Table 1. Organization A had employees, who worked hours per year effective hours of work. So, Organization A had an accident severity rate of 4. According to the survey team, the score pertaining to the current status of accidents was 70, which represented the poor condition.
Values of other variables needed to be defined by formulating the relationship between them; this process will be described in later sections. The current values of mentioned variables were determined by the results of survey and comments of the survey team. Considering the currently poor condition of the organization, the objective was to reduce the rate of accidents and its quantitative index from 70 to the acceptable value of While due to increased volume of orders in the past few months the production rate was extremely high.
Members of the survey team also recommended the value of 0. At the time, communication channels between employees and safety managers included accidents and near misses record box, a suggestion system, phone lines and a safety site. Note that values of other variables must be determined through the definition of their relationships; this process will be described in the simulation section.
Figure 2 shows the diagram developed by Vensim software. In addition, focus on safety will also be affected by production pressure. In a very busy production schedule, there will be little time to focus on safety issues. Allocation of more resources requires the availability of resources. More resources provide the groundwork for more training as well as the development of new safety measures.
It should be remembered that the development of new safety measures also depends on the availability of technology.
Training and development of new safety measures increase the safety competency which in turn increases the safety culture and reduce the rate of accidents. Increasing the safety competency is a time-consuming objective and will have long delayed results Loop B1. Note that intensified focus on B1 and B2 control loops can close the gap between current status and target status.
The loop R1 is a potential factor of risk growth. The occurrence of an accident decreases the confidence in the current safety measures. These circumstances lead to decreased motivation to use current safety measures and therefore decreased thoroughness of risk assessment. Decreased thoroughness of risk assessment leads to inadequate reporting of accidents and subsequent failure to address safety issues, and all these consequently lead to higher levels of risk.
The important point is that communication channels are the main instruments of reporting, so improving these channels can improve the status of hazard reporting. In addition, eliminating the reported hazards depends on the status of available safety resources.
Without a doubt, increased risk of accidents is directly related to increased rate of accidents. Loop R1. Here, the status of accidents is a level variable, while risk and safety culture are both rate variables influencing the status of accidents. The risk is directly related to the rate of accidents and an increased risk leads to an increased accident severity rate.
On the other hand, safety culture reduces the rate of accidents. Other variables affect both risk and safety culture.
Here, fixed variables are Target, Available resources, Production pressure, Existing technology, and Communication channels, all having a constant fixed effect on safety performance and culture. The next step after developing the flow diagram was to conduct the simulation but to simulate the model we needed to define the relationships between variables. The flow diagram for safety performance and culture 5. In fact, the quality of relationships between variables and effects and feedbacks between them were determined earlier in the causal loop diagram.
Here, the opinions of experts were again used to define the quantitative relationships between these variables. Table 2. It seems that due to delayed effects of safety performance and culture on OHS performance, year period provides an acceptable scope for gauging this effect. In fact, a continuation of current trends is not acceptable, and some strategy must be adopted to change this situation. An interesting point is that despite a slight increase in safety culture over these years, the risk will still see an increasing trend, and this will, in turn, leads to more accidents.
It seems that the reason behind this type of behavior of safety performance and culture should be traced back to variables such as production pressure. In fact, it seems that a variable such as production pressure can essentially foil all efforts made to increase the safety culture.
To prove this hypothesis and investigate the effect of other factors on the occurrence of accidents, in the next section, we will present the results of a simulation conducted based on a series of defined scenarios. The simulation results obtained for the defined scenarios are shown in Figures 6 to 9.
As these graphs show, Scenarios 1, 2 and 3 have reduced the accidents, but the extent of reductions made by scenarios 1 and 3 is minimal, so they will not make a significant contribution to the improvement of the situation over the next 25 year.
The fourth scenario, i. In fact, the implementation of this scenario alone has reduced the level of accidents to an acceptable level. This shows the importance of reducing production pressure through measures such as adequate staffing, setting up extra work shifts and improving the management of orders.
However, the results of simulation and scenario building indicate that rapid and timely action is essential since the lack of proper attention to this issue will cause irreparable damage to the organization. This simulation showed that over the next 25 years the trend of accidents in the studied organization will be far from acceptable.
In the next step, a series of scenarios were defined to identify the critical variables. Of course, it is essential to also pay due attention to other variables and does not limit the efforts to production pressure reduction programs. Overall, it seems that in cases where assessed parameters are heavily influenced by both environmental and behavioral factors, the use of system dynamics model can clarify the uncertainties and facilitate the process of problem-solving.
The interesting subjects for future research regarding this topic could be the development of model introduced in this paper, determination of effects of safety standards on OHS performance, and dynamic modeling of the effects of organizational performance on other HSE aspects. Methods of behavior based safety. Checkland, P. Soft Systems Methodology in Action.
Wiley, Chichester. Cooke, D. A system dynamics analysis of the Westray mine disaster. System Dynamics Review 19 2 : Learning from incidents: from normal accidents to high reliability. System Dynamics Review 22 3 : Flin, R. Measuring safety climate: identifying the common features. Safety Science 34 1—3 , — Forrester, J. System dynamics — a personal view of the first fifty years. System Dynamics Review 23, — The textual approach - risk and blame in disaster sensemaking.
Academy of Management Journal 36 6 : Gherardi S, Nicolini D. Journal of Management Studies. Goh, Y. Applying systems thinking concepts in the analysis of major incidents and safety culture. Safety Science 48, — Organizational accidents: a systemic model of production versus protection.
Journal of Management Studies, in press, doi Hynes, T. Patterns of 'mock bureaucracy' in mining disasters: An analysis of the Westray coal mine explosion. Journal of Management Studies 34 4 : Jasanoff, S. Learning from Disaster: Risk Management after Bhopal.
University of Pennsylvania Press, Philadelphia, Sarriegi, and J. Leveson, N. Applying systems thinking to analyze and learn from events. The effects of safety climate on vessel accidents in the container shipping context. Marais, K. Saleh, and N. Archetypes for organizational safety. Zohar, ; Yule et al. In a survey conducted by a questionnaire called safety climate scale, it has been concluded that safety culture is a collective understanding about the way the OHS management is implemented in the workplace.
Research on safety climate has determined that safety climate has a correlation of 0. Although there is an important relationship between safety climate and OHS performance, the strength of this relationship is debatable.
Safety climate researchers have been inclined to study the different aspects of safety climate via hierarchical linear methods Y. Goh et al, and have paid less attention to the interaction between various components of the system. Meanwhile, few papers that have assessed the subject of dynamics of safety culture and performance have only expressed and analyzed the causal relationships of this subject. For example, Y. Goh et al. The measurement capability is a critically required feature of performance evaluation systems, but the mentioned study lacks quantitative analyses, simulations, and scenario building required to explore the leverage variables.
So in the present paper, we tried to use the variables provided in literature and especially those identified by Y. The case study assessed in this paper was a company with employees, of which about were employed in manufacturing department and other were working in administrative sections. Due to failure in acquiring the licenses required to use the name of this company, hereinafter it will be called Organization A.
This company is a manufacturer of office equipment. Perrow measured interactive complexity based on the number of ways system components can interact and tightness of coupling is measured based on the responsiveness of system components to a change in the system. Due to the characteristics of organizational accidents, Goh, Brown and Spickett suggested that traditional accident investigation tools Sklet should be complemented by system dynamics tools to reveal the systemic structure associated with the organizational accident.
In recent years, there has been several system dynamics analysis of organizational accidents. However, the methods used are not consistent. Some placed more emphasis on qualitative system dynamics tools Goh, Brown, and Spickett ; Leveson ; Marais, Saleh, and Leveson and others focus on the stock and flow simulation Cooke ; Cooke and Rohleder ; Rudolph and Repenning ; Salge and Milling In comparison with traditional organizational accident research, which usually relies on textual analysis to elicit evidence to support theories on accident causation e.
Gephart ; Hynes and Prasad , system dynamics analysis of organizational accidents is a departure from the usual. Work-related accidents account for a significant portion of annual mortality in Iran as well as in the world Mohammadfam I, et al. The increasing rate of accidents in recent years Tabibi J,et al. Accidents are the outcomes of unsafe mechanical conditions or unsafe practices, so usually, they are not the product of a single error of one person but instead develop through a series of interactive elements involved at various levels of the system Tabibi J,et al.
Over the years, many theories have been propounded to specifically explain the cause of accidents; a brief review on the evolution of most notable theories in this regard is presented below: A The era of technical errors: the development of mechanical systems brought about a rapid growth in rates of work accidents. B The era of human errors: In this period, the investigations on the cause of accidents showed the importance of human-related factors.
The accident in a nuclear power plant in Iceland revealed the shortcomings of human factors, most notably the psychological weakness. In this period, the "human" element was identified as the weak link, or the component directly responsible for unsafe practices Rochlin, G.
C The era of technical-social errors: In this period, the reaction of humans to "technical" errors was declared as the cause of fatal errors and subsequent accidents. D The era of organizational culture: over the recent years, a new approach, which puts great emphasis on organizational culture, has emerged.
It should be noted that due to the limited ability of human cognition, decision-makers and managers working in complex systems tend to develop a silo mentality Sterman, ; Senge, ; Forrester, Meadows, Silo approach encourages managers to solve the problems in their area of responsibility by focusing on internal components.
These solutions are often ineffective in the long run and may create counter-intuitive behavior in the entire system Checkland and Scholes, ; Senge, So to gain a proper understanding of the system and conduct a precise OHS performance evaluation, all components of the organization and their interactions should be assessed in tandem with each other.
To achieve this goal, the researcher must first identify the variables and elements affecting the safety culture and performance. The UK Advisory Committee on nuclear safety has published a report on influence of Human Factors on this issue, and therein it has defined a set of parameters as the key indices of safety culture; these parameters include: management commitment, management style, management visibility, communication, pressure for production, training, housekeeping, job satisfaction and workforce composition Flin,et al.
In a study conducted by the Irish Aviation Administration , aspects such as safety objectives and policies, safety risk management, commitment to safety, and safety promotion were considered as the main components of organizational safety culture State Safety Plan What that reflects the ambiguity of OHS performance evaluation is the still very high count of industrial accidents despite extensive efforts to improve safety conditions. The main purpose of this study was to use system dynamics model and particularly quantitative analysis via simulation and scenario building to examine the effect of safety performance and culture on the rate of accidents.
This method is one of the Appropriate methods to simulate complex systems is based on causal relationships that enabling the system to provide appropriate learning by providing an environment for testing various scenarios. This method can be also used to analyze the problems with the qualitative approach causal analysis and quantitative approach stock and flow. In order to solve a problem by system dynamics we need to pursue five steps below: 1.
Identifying and defining the problem 2. Mapping cause and effect diagrams 3. Developing the mathematical model stock and flow diagram 4. Model simulation and validation 5. Scenarios generation and evaluation, then selecting and implementing the most appropriate solution System dynamics is often employed to analyze complex social and economic systems; as these systems dynamically change due to many unknown causes.
Sterman describes the steps of system dynamics modeling according to figure 1. So to identify the important parameters in an accurate manner, this study used variables provided by literature and especially those identified by Yang Miang Goh et al. Experts contributing to this study included: 2 safety managers with 10 and 15 years of experience, safety manager of Organization A, 3 safety expert with more than 15 years of experience, 3 safety experts of organization A, and 1 safety inspector with 7 years of experience.
The authors, who were familiar with the system dynamics literature, provided these experts with several articles about safety culture dynamics. After a period of 10 days, during which articles were read by experts, a meeting was held at the office of Organization A.
Therefore, after discussion and addition and removal of a number of variables, the main variables of the study were identified.
In addition, to convert a qualitative model to a quantitative one, we needed to formulate the relationships and quantify the qualitative variables. To determine a level for each qualitative variable and turn it into a quantity, we needed to perform a survey at the organizational level. For this survey, the training manager, the production manager, and the members of HSE committee were added to the team of experts. This survey team acquired all documents related to production, hazard detection, risk assessment, statistics, reports, and analyses pertaining to the accident of past 10 years, and training programs of last 5 years, and the list of budgets planned and allocated to OHS applications.
Survey plan was announced to all units. Survey process included field studies and analysis of documents. Given the size and scope of activities, the survey process took about 3 days. At this point, a quantification process was needed to determine the current status of constant qualitative variables. So, after the survey, members of the survey team held another meeting.
After discussing the results of the survey, this team proposed a scheme for quantifying the qualitative variables. This scheme is shown in Table 1. Table 1. Organization A had employees, who worked hours per year effective hours of work. So, Organization A had an accident severity rate of 4. According to the survey team, the score pertaining to the current status of accidents was 70, which represented the poor condition.
Values of other variables needed to be defined by formulating the relationship between them; this process will be described in later sections. The current values of mentioned variables were determined by the results of survey and comments of the survey team.
Considering the currently poor condition of the organization, the objective was to reduce the rate of accidents and its quantitative index from 70 to the acceptable value of While due to increased volume of orders in the past few months the production rate was extremely high.
Members of the survey team also recommended the value of 0. At the time, communication channels between employees and safety managers included accidents and near misses record box, a suggestion system, phone lines and a safety site. Note that values of other variables must be determined through the definition of their relationships; this process will be described in the simulation section. Figure 2 shows the diagram developed by Vensim software. In addition, focus on safety will also be affected by production pressure.
In a very busy production schedule, there will be little time to focus on safety issues. Allocation of more resources requires the availability of resources. More resources provide the groundwork for more training as well as the development of new safety measures. It should be remembered that the development of new safety measures also depends on the availability of technology. Training and development of new safety measures increase the safety competency which in turn increases the safety culture and reduce the rate of accidents.
Increasing the safety competency is a time-consuming objective and will have long delayed results Loop B1. Note that intensified focus on B1 and B2 control loops can close the gap between current status and target status.
The loop R1 is a potential factor of risk growth. The occurrence of an accident decreases the confidence in the current safety measures. These circumstances lead to decreased motivation to use current safety measures and therefore decreased thoroughness of risk assessment. Decreased thoroughness of risk assessment leads to inadequate reporting of accidents and subsequent failure to address safety issues, and all these consequently lead to higher levels of risk.
The important point is that communication channels are the main instruments of reporting, so improving these channels can improve the status of hazard reporting.
In addition, eliminating the reported hazards depends on the status of available safety resources. Without a doubt, increased risk of accidents is directly related to increased rate of accidents. Loop R1. Here, the status of accidents is a level variable, while risk and safety culture are both rate variables influencing the status of accidents.
The risk is directly related to the rate of accidents and an increased risk leads to an increased accident severity rate. On the other hand, safety culture reduces the rate of accidents. Other variables affect both risk and safety culture. Here, fixed variables are Target, Available resources, Production pressure, Existing technology, and Communication channels, all having a constant fixed effect on safety performance and culture. The next step after developing the flow diagram was to conduct the simulation but to simulate the model we needed to define the relationships between variables.
The flow diagram for safety performance and culture 5. In fact, the quality of relationships between variables and effects and feedbacks between them were determined earlier in the causal loop diagram. Here, the opinions of experts were again used to define the quantitative relationships between these variables. Table 2. It seems that due to delayed effects of safety performance and culture on OHS performance, year period provides an acceptable scope for gauging this effect.
In fact, a continuation of current trends is not acceptable, and some strategy must be adopted to change this situation. An interesting point is that despite a slight increase in safety culture over these years, the risk will still see an increasing trend, and this will, in turn, leads to more accidents. It seems that the reason behind this type of behavior of safety performance and culture should be traced back to variables such as production pressure.
In fact, it seems that a variable such as production pressure can essentially foil all efforts made to increase the safety culture. To prove this hypothesis and investigate the effect of other factors on the occurrence of accidents, in the next section, we will present the results of a simulation conducted based on a series of defined scenarios.
The simulation results obtained for the defined scenarios are shown in Figures 6 to 9. As these graphs show, Scenarios 1, 2 and 3 have reduced the accidents, but the extent of reductions made by scenarios 1 and 3 is minimal, so they will not make a significant contribution to the improvement of the situation over the next 25 year.
The fourth scenario, i. In fact, the implementation of this scenario alone has reduced the level of accidents to an acceptable level. This shows the importance of reducing production pressure through measures such as adequate staffing, setting up extra work shifts and improving the management of orders.
However, the results of simulation and scenario building indicate that rapid and timely action is essential since the lack of proper attention to this issue will cause irreparable damage to the organization. This simulation showed that over the next 25 years the trend of accidents in the studied organization will be far from acceptable. In the next step, a series of scenarios were defined to identify the critical variables.
Of course, it is essential to also pay due attention to other variables and does not limit the efforts to production pressure reduction programs. Overall, it seems that in cases where assessed parameters are heavily influenced by both environmental and behavioral factors, the use of system dynamics model can clarify the uncertainties and facilitate the process of problem-solving.
The interesting subjects for future research regarding this topic could be the development of model introduced in this paper, determination of effects of safety standards on OHS performance, and dynamic modeling of the effects of organizational performance on other HSE aspects. Methods of behavior based safety. Checkland, P. Soft Systems Methodology in Action. Wiley, Chichester.
Cooke, D. A system dynamics analysis of the Westray mine disaster. System Dynamics Review 19 2 : Learning from incidents: from normal accidents to high reliability. System Dynamics Review 22 3 : Flin, R.
Measuring safety climate: identifying the common features. Safety Science 34 1—3 , — Forrester, J. System dynamics — a personal view of the first fifty years.
0コメント