Cascading failure in urban rail transit network considering demand variation and time delay

Abstract

To study the cascading failure in real-world urban rail transit network, the cascading failure model is proposed by considering two essential characteristics of demand variation and time delay, which has rarely been considered in most existing studies. Specifically, the model utilizes a simulation-based approach where passengers in the rail transit network are abstracted as packets moving through the network. The movement of these packets, including their response to station failures, is planned to model the travel behavior of passengers. The Shanghai rail transit network is chosen as the case study for this research. Comparative experiments suggest that the proposed model may produce very different results for stations with high closeness and enough travel demand. The proposed model can depict the dynamic failure process and yield a longer and more dynamic failure process, which is beneficial for managers to identify the period for prevention. Besides, sensitivity analysis shows that higher travel demand is more likely to cause a cascading failure, while time delay slightly impacts the final cascade size but affects propagation duration. In summary, travel demand plays a vital role in cascading failures, and the presence of time delays complicates the cascading failure propagation process. The research results provide references and support for studying cascading failure in real complex environments.