Resilience assessment for urban utility tunnels under gas explosion scenarios

Abstract

Urban utility tunnels (UUTs), essential for unified pipeline management, face heightened safety risks during gas leak and explosions. This paper presents a resilience assessment method that integrates Decision-making Trial and Evaluation Laboratory (DEMATEL), Interpretive Structural Modeling (ISM), and Bayesian Network (BN) to enhance system impact resistance and recovery. Expert surveys and literature reviews identify key gas explosion factors, forming a direct influence matrix to quantify their interrelationships. ISM stratifies these factors, highlighting critical indicators such as ventilation, alarm systems, pressure relief, and emergency response. The hierarchical structure is mapped to a BN, where node probabilities are assigned using conditional probability table (CPT), expert evaluations, and fuzzy set theory for dynamic resilience assessment. Case studies reveal a maximum resilience score of 73.8/100 and significant coupling effects among key factors, while sensitivity analysis confirms the model's robustness and accuracy. Based on diagnostics, the paper recommends strategies including optimizing ventilation design, adjusting alarm layouts, refining pressure relief configurations, and reducing emergency response times, thereby offering both theoretical and practical guidance for enhanced safety management.