A Framework for Quantitative Assessment of Socio-Technical Resilient Systems using Multiple Criteria Decision Making (MCDM): A Case Study in an Oil Site

Today, the major problem in the field of industrial safety is that most traditional risk assessment and safety management methods alone do not meet the needs of modern technologies and complex industries. In other words, traditional methods, such as risk analysis and probability safety assessment, are in practice capable of do not provide many of the solutions needed for todaychr('39')s industries. There are various reasons for this claim, the most important of these reasons are their roots in the very simple models of the incident as a cause-effect chain. Therefore, there is an urgent need for new approaches in risk assessment and safety management, and resilience engineering (RE) has been proposed as a remedy to satisfy that need. Resilience Engineering is the new window which uses the insights from research on failures in complex systems, organizational contributors to risk, and human performance to develop engineering practices including measures of sources of resilience, decision support for balancing production/safety tradeoffs, and feedback loops that enhances the organization’s ability to monitor/revise risk models and to target safety investments. Therefore, RE was defined as: "The intrinsic ability of a system to adjust its functioning prior to, during, or following changes and disturbances, so that it can sustain required operations under both expected and unexpected conditions."However, the purpose of this study is to provide a framework for assessing the resilience of socio-technical systems based on the Multi Criteria Decision Making (MCDM).

In this research, after identifying the indicators of resilience through reviewing the literature and interviewing experienced industry experts, information on these indicators in the operating units of the industry was obtained through a semi-structured interview with 16 experienced operators whose average working experience was more than 10 years. Semi-structured interview is a meeting in which the interviewer does not strictly follow a formalized list of questions. They will ask more open-ended questions, allowing for a discussion with the interviewee rather than a straightforward question and answer format. Interview topics included buffering capacity with 29, safety margins with 24, system tolerance with 10, cross-scale interactions with 5, learning from accidents, incidents and normal works with 16, system flexibility with 24, anticipation of expected and unexpected events with 6, attention to problems in the system with 10 and proper and timely response to them with 11 topics. Then, the Principal Component Analysis (PCA) was used to determine the weight of the indicators and the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) method was used to determine the ranking of different units. The Minitab version 16 and MATLAB software was used to process the collected data.  

The results showed that the three indicators of buffering capacity (0.43), Attention (0.39) and learning culture (0.36) have highest weight and rank and the indicators of tolerance (0.26) and cross-scale interactions (0.21) have the lowest weight and rank. In addition to strengthening the weak indicators, the management of the organization should also make a lot of efforts to promote and develop the desired indicators in accordance with the principles of resilience engineering. For example, identifying safety margins, reducing stress, reducing uncertainties in work processes, identifying worn-out systems and equipment and types of threats can improve the systemchr('39')s tolerance index. In this context, the cross-scale interactions can also be improved through effective interactions between individuals, groups, and different sections, attention to competency-based education, training and experience and positive and effective communication. Accordingly, operational units 2, 3 and 11 also ranked the best in terms of resilience indicators. Also by examining the rankings based on the similarity to ideal solution, it was found that the lowest distance from the positive ideal solution and the highest distance from the negative ideal solution was related to unit 11 and unit 6, respectively. Therefore, unit 11 get less priority to improve than other units, and unit 6, which has the lowest similarity to ideal solution, has a higher priority for improvement.  

The study showed that considering the nature of risk and its complexity in socio- technical systems, it is no longer possible to identify those using traditional methods. Hence, there is an enormous need for further development in risk assessment tools and safety management systems. This requires to an insight that pushes management beyond traditional counts of negative occurrences and finds new types of leading indicators that reflect critical aspects of an organizationchr('39')s resilience. Furthermore, the results of indicators analysis and unit ranking showed that PCA and TOPSIS may be a reasonable and practical alternative for evaluating resilience of complex systems. Because they can identify the strengths and weaknesses of the system in terms of resilience and based on input information. Therefore, the managers and decision makers of the industry will be able to use the results of this research as an important step in improving and enhancing the industrychr('39')s resilience and safety.