Identifying and Analyzing Key Risk Factors for Subway Construction Accidents: An Integrated Approach Using Association Rules and Complex Network Theory

Abstract With the continuous expansion of the scale of subway construction, the associated risks and uncertainties have been increasing accordingly. Once an accident occurs, it may pose a substantial threat to personnel safety. Therefore, to effectively prevent subway construction accidents, it is necessary to explore the risk evolution mechanisms of such accidents through appropriate technical approaches and to identify key risk factors for targeted prevention and control. Accordingly, this study proposes an integrated research framework that combines text mining, the 24Model, association rule mining, and complex network theory to conduct an in-depth analysis of textual data from subway construction accident cases. This approach aims to systematically identify subway construction accident risk factors and to explore the interaction relationships and relative importance among these factors. First, text mining techniques and the 24Model were applied to analyze 124 collected accident investigation reports, through which 93 subway construction accident risk factors were identified. Second, association rules among the risk factors were extracted using the Apriori algorithm, and a complex network model of subway construction accident risks was constructed on this basis. Finally, key risk factors in subway construction accidents were determined through network topological indicators and robustness analysis. Corresponding risk control measures were proposed, and the applicability of the constructed network model was further validated. The results indicate that 14 risk factors—including defects in equipment and facility management, unreasonable personnel allocation, and failure to implement safety production responsibilities—constitute the key risk factors in subway construction accidents. This study provides a new perspective for the data-driven identification and analysis of key risk factors in subway construction accidents.