Quantifying and classifying the robustness of bus transit networks

Abstract

This study adopts Complex Network Theory in the context of bus transit network. The study aims at quantifying the topological characteristics and assessing the validity of static-robustness metrics as expressive measures of transit networks robustness. In addition, dynamic-robustness indices, that consider transit operational profile, are utilized to measure the cascading impacts of disruptive events. The analysis is based on a dataset of 40 bus transit networks. The results indicate that bus networks don't conform to any major network class: scale-free, small-world, or random. Furthermore, the static-robustness metrics produced contradictory results, which raises valid concerns on their applicability. The dynamic-robustness indices developed in the current study indicated significant cascading impacts resulting from node removal relative to the removal of links. This behaviour was further examined through a two-step cluster analysis, which resulted in three distinct network clusters: small node-sensitive; small link-sensitive; and medium less-sensitive networks. These findings are directed to inform a robustness-based design of bus transit networks.

Topics & Concepts