Resumen
Modeling emergency evacuation could help reduce losses and damages from disasters. In this paper, based on the system optimum principle, we develop a multimodal evacuation model that considers multiple transportation modes and their interactions, and captures the proper traffic dynamics including the congestion effects, the cooperative behaviour of evacuees, and the capacities of the transportation system and the shelters. We further develop a Method of Successive Average (MSA)-based sequential optimization algorithm for large-scale evacuations. Both the proposed model and the solution algorithm are tested and validated through a set of numerical tests on a small network, and a detailed case study on the Lower Manhattan network. The results of the paper can provide insight on modeling flow interactions of different transportation modes and useful guidance on developing evacuation strategies to reduce the system evacuation time and losses from disasters.