Cascading failures in interdependent transportation networks: A generalized coupled map lattice model with intermodal dynamics

Urban public transportation networks constitute critical infrastructure essential for daily mobility. During emergencies, the failure of stations or lines may trigger cascading failures. In multimodal transportation networks, disturbances influence both intramodal and intermodal operations. While existing research has examined the cascading failure in multimodal networks by considering network topology and travel demand, the inherently complex interdependencies within such networks remain a challenge. To address this, this study improves the weighted coupled map lattice model to analyze failure propagation. The improved model integrates network structure and travel demand to determine inter-node weights, effectively capturing coupling effects between transportation modes. An interdependence factor that integrates both spatial and functional interdependencies is introduced into the node state evolution function. Spatial interdependence reflects static intermodal connections based on the distance between coupled nodes, while functional interdependence represents dynamic interactions measured by variations in betweenness centrality and demand following disturbances. To enhance model generalizability, two discrimination parameters are introduced. Finally, the proposed model is applied to a bus-subway network in Tianjin, using smart card transaction data to construct a weighted multimodal transportation network. Results reveal that the improved CML model effectively captures the intermodal and intramodal node dynamics. The residual capacity plays a significant role in enhancing the network robustness. Furthermore, an appropriate coupling distance should be established in a multimodal network, which can provide convenience for passengers and reduce the losses during cascading failures. This study offers a theoretical foundation for transportation managers to develop emergency strategies and mitigate disruptions.