Causes Prediction of Coal Mine Gas Explosion Accidents Based on Fault Tree and Bayesian Network

Coal mine gas explosions occur frequently, posing a severe threat to mine safety production and the life and property security of personnel. However, traditional single analytical methods suffer from significant shortcomings, including low accuracy in risk assessment and inadequate connectivity between cause identification and risk prediction. To address these issues, this study constructs an integrated risk assessment and prediction model combining Fault Tree Analysis, Analytic Hierarchy Process, and Bayesian Network, aiming to improve the risk management capacity for coal mine gas explosion accidents. This study uses 64 coal mine gas explosion accidents that occurred in China from 2016 to 2026 as research samples. Firstly, six key causal indicators are identified and their importance coefficients are calculated via FTA; subsequently, a judgment matrix is constructed by integrating AHP with probability importance, and after passing the consistency test, the weight quantification of each indicator is completed. Finally, the comprehensive weights obtained from AHP are incorporated into the BN model to realize quantitative assessment, dynamic prediction, and causal traceability analysis of gas explosion accidents. The research results indicate that ventilation system failure, excessive gas concentration, and unsafe operations by personnel form the core causal system, whereas electrical spark generation, monitoring equipment failure, and safety management deficiencies serve as indirect causes. The importance hierarchy of each causal factor is clearly defined. This integrated model effectively overcomes the limitations of traditional single methods, significantly improving the accuracy, reliability, and dynamic adaptability of risk analysis for gas explosions in complex mining systems. It provides scientific quantitative analysis methods and important technical support for the precise prevention and control of coal mine gas explosions, refined mine safety management, and optimization of the prevention and control system.