Comprehensive Risk System Analysis and Factor Coupling in Underground Railway Space

Abstract

The rapid expansion and intensification of urbanization have led to increased growth and utilization of underground spaces. This trend has raised safety concerns, necessitating additional preventive and mitigation measures. This study presents a comprehensive and systematic assessment of safety risks in urban underground railway spaces. Various research techniques were employed, including literature review, case study, risk matrix, Borda ordinal, and Bayesian network methods. The ISO 31000:2018 risk management standard served as the framework for risk management. Based on existing knowledge, three main risk categories, consisting of 17 constituent risk factors, were identified: rail system; passenger behavior; and environmental hazards. This study critically analyzed the interrelationships and couplings of risks among these categories. A Bayesian network is constructed based on the collected accident information, and the probability that the level of risk is located at low, medium, and high levels is calculated to be 42%, 39%, and 20%, respectively. This finding is verified with a real case of an in-operation subway. Inverse reasoning and sensitivity analyses were conducted using Genie software to predict the occurrence probability and severity of high-risk factors. The findings highlighted significant risks to underground railway spaces, such as arson and accidental fires. In response, authorities proposed measures to enhance risk management strategies. The results provide a theoretical foundation, diverse analytical approaches, and practical guidelines to improve risk management capabilities in urban underground spaces, thus facilitating the development of informed risk management strategies.