Computer-Aided Resilience Simulation Analysis for Disaster Scenarios in Transportation Infrastructure Engineering

Transportation infrastructure systems are increasingly challenged by compound disaster scenarios and cascading failures, especially in geologically complex and high-risk regions. These challenges are exacerbated by climate change, rapid urbanization, and the increasing interconnectedness of infrastructure systems. A growing number of catastrophic events in recent years, including floods, earthquakes, and landslides, have highlighted the urgent need for comprehensive resilience analysis frameworks capable of simulating both damage propagation and recovery. This study proposes a computer-aided simulation framework for evaluating system resilience under multi-source disturbances and dynamically evolving scenarios. By integrating a directed scenario network with a time-dependent node health evolution model, the method captures both direct and cascading failure processes across interconnected infrastructure components. The framework incorporates Monte Carlo-based simulation and statistical inference to identify critical risk sources, quantify system degradation and recovery, and support data-driven resilience enhancement strategies. It allows for simulating uncertainties in hazard onset and propagation, capturing both average trends and variance in system behavior. Additionally, the framework supports scalable model adaptation across different infrastructure types and regional settings, promoting replicability and robustness. The approach also allows for the evaluation of recovery timelines, spatial influence patterns, and system-wide performance indicators. Applied to a representative section of the Sichuan–Tibet Railway, the framework demonstrates its effectiveness in simulating compound disaster chains, identifying vulnerability hotspots, and supporting proactive decision-making for mitigation and emergency planning. The results provide both theoretical and practical contributions to resilience-based infrastructure management in complex environments. In summary, this research presents a novel methodology that contributes to closing the gap between traditional risk models and dynamic, data-informed resilience assessment.