A multiple perspective method for urban subway network robustness analysis

Abstract

Most network research studying the robustness of critical infrastructure networks focuses on a particular aspect and does not take the entire system into consideration. We develop a general methodological framework for studying network robustness from multiple perspectives, i.e., Robustness assessment based on percolation theory, vulnerability analysis, and controllability analysis. Meanwhile, We use this approach to examine the Shanghai subway network in China. Specifically, (1) the topological properties of the subway network are quantitatively analyzed using network theory; (2) The phase transition process of the subway network under both random and deliberate attacks are acquired (3) Critical dense areas that are most likely to be the target of terrorist attacks are identified, vulnerability values of these critical areas are obtained; (4) The minimum number of driver nodes for controlling the whole network is calculated. Results show that the subway network exhibits characteristics similar to a scale-free network with low robustness to deliberate attacks. Meanwhile, we identify the critical area within which disruptions produce large performance losses. Our proposed method can be applied to other infrastructure networks and can help decision makers develop optimal protection strategies.