Abstract
With the development of intelligent coal mine construction, China’s coal production safety has been greatly improved, but coal mine gas explosion accidents still cannot be completely avoided and the unsafe acts of miners are an important cause of the accidents. Therefore, this study firstly collected 100 coal mine gas explosion cases in China, improved the framework of human factors analysis and classification system (HFACS) and used it to identify the causes of miners’ unsafe acts in detail. A hierarchy of the impacting factors is established. Then, combining with the interpretive structural model (ISM), the correlation between the impacting factors among different levels, especially among non-adjacent levels, is qualitatively analyzed through expert judgment. Then, the correlation among the contributing factors was quantitatively tested by chi-square test and odds ratio (OR) analysis. On this basis, a Bayesian network (BN) is constructed for the impacting factors of miners’ unsafe acts. The results show that the probability of coal mine gas explosion accident is 20% and 52%, respectively. Among the leading factors, the government’s insufficient crackdown on illegal activities had the greatest impact on miners’ violations, with a sensitive value of 13.2%. This study can provide reference for evaluating the unsafe acts of miners in coal mine gas explosion accidents by the probabilistic method.