A data-driven and knowledge graph-based analysis of the risk hazard coupling mechanism in subway construction accidents

Abstract

With the development and expansion of subway construction, accidents usually cause economic losses and seriously threaten personnel safety. Exploring the experience and lessons from accident reports is beneficial for promoting subway construction safety management. Therefore, this study proposes a data-driven and knowledge graph-based approach to reveal the coupling mechanism of risk hazards in subway construction. A total of 231 subway construction accident reports in China from 2001 to 2023 are collected as the database of the knowledge graph. By developing a subway construction safety accident knowledge graph (SCSAKG), the potential paths for accidents can be revealed. To further investigate the internal features of the knowledge graph, topological indicators are proposed to explore the relationships between hazards, hazard types, hazard frequencies, hazard occurrence times, accidents, and accident consequences. Prevention strategies for key risk hazards in subway construction are developed based on the proposed coupling mechanism. The proposed methods that combine knowledge graphs with accident reports can provide decision-making references for subway construction safety management.