Bayesian network and bow tie to analyze the risk of fire and explosion of pipelines



This article is corrected by:

  1. Errata: Corrigendum Volume 36, Issue 3, 318, Article first published online: 17 July 2017


As accidents were and still one of the main reasons standing behind the increasing rates of casualties such as death, injuries, and evacuations, the constant improvement of safety measures especially in the field of hydrocarbons remains a major concern. Therefore, in the work in hand, we attempt to shed the light on the ways of developing a method for the evaluation of risks of fire and explosions of pipelines. The causes of the latter and consequences are, in one hand, analyzed by means of fault tree and bow tie methods. On the other hand, a quantitative analysis implementing the Bayesian networks is used to estimate the probability of occurrence of the adverse event. Moreover, 72 basic events were found to be of the primary causes provoking the occurrence of undesirable events. However, some experts often find it difficult to precisely determine the probabilities of occurrence of basic events of the tree. For the purpose of evaluating the occurrence of each basic event, we used the fuzzy logic. Hence, at the end of the study, we were able to develop a model that could help us evaluate the risks accompanied the fires and pipelines explosion as well as the consequences. © 2016 American Institute of Chemical Engineers Process Saf Prog 36: 202–212, 2017