Topological assessment of recoverability in public transport networks

Reducing the impact of disruptions is essential to provide reliable and attractive public transport. In this work, we introduce a topological approach for evaluating recoverability, i.e., the ability of public transport networks to return to their original performance level after disruptions, which we model as topological perturbations. We assess recoverability properties in 42 graph representations of metro networks and relate these to various topological indicators. Graphs include infrastructure and service characteristics, accounting for in-vehicle travel time, waiting time, and transfers. Results show a high correlation between recoverability and topological indicators, suggesting that more efficient networks (in terms of the average number of hops and the travel time between nodes) and denser networks can better withstand disruptions. In comparison, larger networks that feature more redundancy can rebound faster to normal performance levels. The proposed methodology offers valuable insights for planners when designing new networks or enhancing the recoverability of existing ones. Public Transport Networks often suffer from disruption that needs to be reduced and mitigated. The authors study the correlation between topological and recoverability indicators for 42 transport networks in various cities, and find that denser and more efficient networks can better withstand disruptions, while larger networks with additional redundancy can rebound faster to normal performance levels during the recovery process.