Cause factors in emergency process of fire accident for oil–gas storage and transportation based on fault tree analysis and modified Bayesian network model

Abstract

From the perspective of the safety of emergency process, in this paper, we put forward a new analysis method of cause factors based on fault tree analysis and modified Bayesian network in emergency process of fire accident for oil–gas storage and transportation. Nineteen cause factors are found based on the statistical analysis of actual accident cases. We adopt fault tree analysis method and Bayesian network model to analyze cause factors qualitatively and quantitatively. In order to more accurately determine the quantitative influencing degree of each cause factor, the conditional probabilities in Bayesian network are modified by using expert scoring method and modified Bayesian network model is established. Finally, the proposed method is applied to analyze cause factors in emergency process of two practical accident cases. Research results have an important scientific significance to reveal evolution mechanism of secondary accidents in emergency process and ensure system safety in whole life cycle.