Modelling Risk Factors of Urban Gas Accidents for Enhanced Energy Security: An Association Rule Mining Approach Integrating Causality Theory, 24‐Model, and HFACS

Urban gas systems are core components of global energy infrastructure, underpinning residential, commercial, and industrial energy supply while supporting low‐carbon goals. However, the growing frequency of urban gas accidents—from pipeline leaks to explosions—poses severe threats to energy security, public safety, and sustainable urban development, largely due to insufficient systematic analysis of root causes. This study addresses this gap by integrating accident causality theory, the 24‐Model, and the Human Factor Analysis and Classification System (HFACS) to investigate 100 urban gas accidents, combining Chinese (2014–2023) and global datasets. Using Apriori‐based Association Rule Mining (ARM), we identified 9 key risk factors, 19 critical association rules, and a universal causal chain, with institutional negligence (inadequate regulatory frameworks and weak supervision) confirmed as the predominant root cause. The key causal chain is: “lack of targeted energy safety regulations → poor supervision of gas infrastructure → unaddressed safety hazards → equipment failure in delivery systems → gas leakage (and subsequent ignition) → accident occurrence.” To mitigate these risks, we propose three actionable strategies: tailored safety regulations for urban gas networks, AI‐enhanced real‐time supervision systems, and competency‐based training programs. These insights not only optimize the resilience of China's gas energy systems but also provide a transferable framework for global urban energy safety management, aligning with goals for secure and sustainable urban energy.