Comparison analysis on complex topological network models of urban rail transit: A case study of Shenzhen Metro in China

To study the topological complexity of urban rail transit (URT) networks with the multi-line transfer stations from different perspectives, Shenzhen Metro (SZM) is taken as an example and Space L & Space P models are established in this study. Then, based on multiple evaluation parameters and key nodes ranking, the differences of network topological complexity in two models are deeply explored and compared quantitatively. Some meaningful results have been obtained: (i) The characteristics of scale-free networks in Space L and Space P are proved through the eigenvector centrality distribution and truncated power-law distribution of cumulative degree. Scale-free networks show both robustness against random faults and vulnerability against deliberate attacks. The daily safety management at 16.87% of hub stations in Space P and 17.47% of hub stations in Space L should be taken seriously by metro managers in case of emergency events. (ii) Since the WS small-world effect in Space P model is more evident than that in Space L model, the connections among stations and OD accessibility of passenger are enhanced in Space P network. (iii) The important and risk nodes are concentrated in Space L and are more decentralized in Space P. P model has the stronger overall anti-attack capability than L model, which is more beneficial to the resilience of network. This study can realize the deeper understanding of URT system with different models and it can provide theoretical support for complex network analysis of URT system. © 2020 Elsevier B.V.