Vulnerability Assessment and Optimization of Urban Rail Systems Against Extreme Perturbations: an MIP-Based Approach

Zehai Liu,Jiateng Yin,Andrea D'Ariano,Tao Tang

DOI: https://doi.org/10.1109/itsc57777.2023.10422487

2023-01-01

Abstract:Urban rail networks are expanding rapidly, due to its convenience and high capacity. However, urban rail networks are facing an increasing number of disruptions, in-cluding extreme weather events, equipment faults, and security threats. As these systems transport thousands of passengers daily, assessing their vulnerability to such perturbations is crucial. Existing vulnerability assessment approaches rely on expert judgment or simulation. In this study, we propose a network-flow-based model to quantitatively evaluate and optimize system vulnerability. Specifically, we first construct a graph model that captures the relationships among factors leading to disruptions, using nodes, edges, and assigned edge weights. Then, a mixed integer programming (MIP) model based on the graph is developed to identify the most vulnerable factors against perturbations. Case study conducted using the Zhengzhou Metro Line 5 accident report, which occurred in July 2021 during a flood, validate the ability of our model to identify critical accident causes and provide early warnings against perturbations.