Flood-induced traffic congestion and accessibility loss for urban road networks using agent-based simulation

Abstract

This study proposes a methodology for assessing the impact of flood-induced functional disruptions on urban road networks through an agent-based traffic simulation. Network functionality is altered by reducing roads’ free-flow speeds using a risk-based approach, and traffic is appraised considering agent-based traffic dynamics. MATSim, an open-source transport simulator, is employed to model dynamic traffic redistribution and congestion under both baseline (non-flood) and flood conditions of the urban road transportation network. The methodology is applied to the city of Bristol, UK, which is chosen for its complex road layout and flood susceptibility. Key indicators, including travel speed ratios, redistribution ratios, changes in agent count, and time-based isochrones, are used to assess variations in congestion and accessibility under both baseline and flood conditions. This study further advances existing approaches by comparing the spatial shifts of congestion hotspots before and after flooding, and by integrating hazard scenarios to predict potential future congestion patterns and their subsequent impacts on the accessibility of critical facilities, such as the Bristol Royal Infirmary. Results indicate a substantial redistribution of traffic from flood-affected minor roads to central arterial routes, leading to increased congestion and reduced accessibility, which can be particularly detrimental to emergency services that require rapid access to affected areas. The findings highlight the importance of simulating agent-level behavioural responses to network disruption caused by flooding and provide a transferable framework for assessing urban transport resilience during flood events.