تدوین و رتبه بندی شاخص های ارزیابی عملکرد ایمنی با استفاده از شبکه بیزی و تحلیل سلسله مراتبی مطالعه موردی: فعالیت کار در ارتفاع فاز ساخت پالایشگاه های نفت و گاز

Background and aims: The safety and health performance measurement, designed to provide the necessary information on the concern to progress and the current state of the organization's strategies, processes and activities. The purpose of this study is to present a new model for the development of safety performance indicators using the probability risk assessment model and applying experts' opinions. Methods: This descriptive-analytic study was carried out in 3 steps: categorize of the construction face activities and its related hazards identification; formation of the accident causal network occurred in the oil and gas refinery construction phase using the Bayesian network and the selection of key performance indicators using AHP method. Results: The statistical analysis of the recorded accidents showed that 21% of the incidence rate is related to the falling. Among all the construction phase activities, using the WBS analysis of the project, 27 activities with work at height risk of were identified.  18 active performance indicators were extracted the accident causal network  that using SMART criteria and occurrence probability rate  Calculated from the Bayesian network was selected as 5 active key performance indicators. Conclusion: Determining the leading performance indicators is influenced by various organizational, managerial, operational and other factors. As the project progresses, the nature and level of risk of the operation of construction projects is changing. Therefore, indicators of safety performance measurements in these projects should be sensitive to rapid changes. For this reason, active indicators with a short-term measurement period are more effective in measuring the safety performance of construction operations.