Resilience of urban metro rail networks globally guided by mesoscale and connectivity attributes

Abstract Urban metro rail systems are crucial lifelines that play key roles in mobility, accessibility, and economic activity, serving millions daily. However, they are vulnerable to a range of natural and human-induced disruptions. In this study, we simulate 7 different failure and recovery scenarios for a set of 45 global urban metro rail networks around the world to examine how network topology influences infrastructure resilience. Existing studies seldom integrate varied failure/recovery strategies and topology at cross system scale. Our analysis of 25 network attributes, their interaction, and collective behavior reveals their influence on fragility and restoration. Our findings indicate that a consistent set of grouped network attributes, particularly those capturing connectivity, redundancy, and loop structure, and modular organization are critical for identifying structural vulnerabilities and enhancing resilience. These insights have practical value for transport infrastructure design, retrofitting, emergency management, and urban planning. We show the value of these insights via a proof-of-concept retrofitting experiment aimed at improving failure resilience and accelerating recovery through targeted topological interventions.