FLOOD EVACUATION OPTIMIZATION UNDER MULTIPLE TRANSPORTATION MODES

Abstract

In recent years, huge typhoons and heavy rains has been reported frequently, and the importance of advance evacuation planning for residents has been increasing. In this study, we develop an optimal evacuation planning model for inundation disasters that can be useful as reference information for district disaster prevention plans. The proposed model solves a linear programming problem to optimize the total evacuation time and the sum of service levels of evacuation shelters. The optimization problem represents evacuee behavior unique to flooding disaster evacuation, such as shelter in place and evacuation to private facilities, and provides a more realistic method for evacuation planning. The performance of the model was verified on a virtual network, and it was confirmed that the model is capable of planning evacuation plans that take into account the congestion rate of evacuation shelters, and that it can analyze the impact of the capacity of private facilities for evacuation shelters. Application of the model to the road network of Sapporo City confirmed that the computational load was small enough for a general-purpose computer.