A novel dynamic risk assessment method for the petrochemical industry using bow-tie analysis and Bayesian network analysis method based on the methodological framework of ARAMIS project

Xingguang Wu,Huirong Huang,Jianyu Xie,Meixing Lu,Shaobo Wang,Wang Li,Yixuan Huang,Weichao Yu,Xiaobo Sun
DOI: https://doi.org/10.1016/j.ress.2023.109397
2023-05-26
Abstract:In order to obtain the actual risk level and assess the performance of the safety barriers, bow-tie (BT) approach proposed by ARAMIS project is applied to accident scenarios identification and risk analysis, but this approach has limitations in dynamic risk assessment due to its static nature. This study takes the floating roof tank as the research object, and further proposes a dynamic risk assessment method based on the methodology proposed by the ARAMIS project. The BT model of the major leak of external floating roof tank was established by comprehensive consideration of the barrier functions and the logical relationship between causes and events. Furthermore, The BT to Bayesian network (BN) mapping algorithm was proposed to characterize system dynamics and uncertainty. The analysis results show that not only the dynamic assessment of the risk level for safety barriers in different states can be implemented, but also whether the current overall safety protection capability of the system is sufficient can be judged. The proposed method and findings can help managers identify safety barriers that play a key role in various accident risks, and provide effective support for risk management decision-making and implementation of preventive strategies.
engineering, industrial,operations research & management science
What problem does this paper attempt to address?