Answers are grounded in this paper and cite their sources.
: Flood emergencies place severe strain on urban transport systems, particularly when evacuation depends on organised transport rather than private vehicles, as is the case for many transit-dependent groups such as older people, individuals with disabilities, and residents of care facilities. Planning under these conditions is challenging because road networks are often disrupted, congestion is highly uncertain, and warning times can be limited. This paper proposes an integrated transit-based evacuation planning framework that combines agent-based traffic simulation with scenario-based stochastic optimisation to support decision-making under such uncertainty. The framework uses simulation to generate realistic, congestion-aware travel times under different flood scenarios and embeds these directly into a stochastic mixed-integer optimisation model that accounts for shared transport resources, receiving centre capacities, and evacuation time windows. To ensure the approach remains computationally practical, a logic-based Benders decomposition solution method is developed and compared with a monolithic formulation. The framework is demonstrated through a case study in a flood-prone regional area of Australia, showing that it can produce feasible and robust evacuation plans while substantially improving solution stability and runtime performance. The results also highlight the strong influence of evacuation time windows and inter-facility collaboration on evacuation outcomes. Overall, the study offers a practical and scalable decision-support approach for improving the planning of transit-based evacuations in flood-prone communities.
No takes yet. Share an insight, caveat, or question.
Khalili et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: