Assessing dynamic congestion risks of flood-disrupted transportation network systems through time-variant topological analysis and traffic demand dynamics

Abstract

Amid escalating climate change and extreme weather events, urban road networks face increasingly severe flood risks. Current congestion risk analyses on transportation systems, largely grounded in static network characteristics, fall short of capturing the dynamic impacts of evolving flood events on network topology. A notable gap in integrating dynamic flood models with detailed temporal network property measurements impedes a deep understanding of the cascading mechanisms of traffic disruptions caused by floods. The research introduces an innovative framework for assessing dynamic congestion risks within transportation networks under flooding conditions. This framework comprises four distinct layers. The Data and Context layer collates and contextualizes primary data, encompassing road networks, commuter behaviour patterns and hydrological risks. This layer establishes a baseline for understanding the structure of the transportation system and its environmental vulnerabilities. The Computation layer focuses on calculating context-based betweenness centrality values to capture the importance of road segments within the network, setting a static indicator. Concurrently, this layer examines the dynamic exposure flag under flood conditions, providing a dynamic indicator that accounts for the fluctuating susceptibility of road networks to flooding. The Integration layer bridges the topological analysis and the functional performance of the transportation system. It creates a dynamic series that describes the topological properties of the network, reflecting its vulnerability to disruptions. Simultaneously, the time-variant traffic demand series are tracked to capture the system’s operational response to flood-induced conditions. Finally, the Evaluation layer synthesizes data and analyses from the previous modules. This synthesis involves a novel dual assessment: generating a time-variant vulnerability index series and evaluating the time-variant congestion risk map on the transportation system under flooding. The London case study is developed based on this proposed framework to identify critical areas of congestion risks. This study offers valuable data-driven insights for urban planning and disaster management, aiding policymakers, planners and emergency teams to enhance urban resilience against flood-induced transport disruptions.