Key points are not available for this paper at this time.
The chemical industry involves the production, storage, and use of many flammable, explosive, toxic, and other hazardous chemicals. Once an accident occurs, it will cause serious harm to human and economic activities. In order to prevent chemical accidents, this paper combines Interpretive Structural Modeling (ISM) and Bayesian network (BN) to quantitatively study the relationship and interaction strength among accident risk factors in chemical industry. Through the analysis of accident cases and questionnaire survey, 21 accident risk factors in chemical industry are selected. According to the decision of experts, the influence relationship between risk factors is determined, and a multi-level directed graph of ISM is obtained. And the ISM model is transformed into a quantitative BN model. The BN model is applied to forward reasoning, sensitivity analysis, and reverse reasoning. The results indicate that there is a positive correlation between various risk factors and chemical accidents, and the supervision mechanism has the highest probability of occurrence in production activities. Illegal operation has the highest sensitivity and the greatest impact on chemical accidents. Inherent hazards of materials and products is the most likely cause of accidents. Based on the research results, feasible measures have been proposed to improve safety management in the chemical industry.
Ma et al. (2023) studied this question.