The robustness of London metro network from network perspectives

Authors: YUERONG ZHANG*, , Stephen Marshall, University College London, Ed Manley, University of Leeds
Topics: Transportation Geography
Keywords: robustness, network, metro system, percolation, community detection
Session Type: Paper
Day: 4/7/2020
Start / End Time: 10:15 AM / 11:30 AM
Room: Terrace, Sheraton, IM Pei Tower, Terrace Level
Presentation File: No File Uploaded


The robustness of metro systems in adapting to changes is becoming increasingly important with its growing complexities and uncertainties. However, little is known about the specific roles station plays in maintaining a robust network. Drawing on London Underground station data and passenger origin-destination data, the paper explores the robustness of the London Underground network under various attack strategies through the percolation approach at city level and modular community analysis at sub-city level. Specifically, the involvement of stations within and between movement communities is estimated. The analysis shows that the London Underground is vulnerable to targeted attacks with relatively low fault tolerance and emphasises the significance of understanding metro robustness as the result of interaction between passengers’ demand and transport services supply rather than topological characteristics. It also points to the need to include modular community analysis into robustness analysis of the metro network. The results show that the same centrality have variable interactions with other parts of the network, resulting in heterogeneous levels of impact across the network. This finding lends weight to the argument that when identifying the central stations, it is crucial to gains insights into the specific roles stations play within and between movement communities instead of relying merely on the centrality measures. Besides, the analysis finds that the compositions of the top ten important stations at city and district levels are distinct. These findings not only enables us to gain deeper insights of centrality but also provide details about the likely vulnerable passenger groups.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login