Urban Flood Risk Assessment and Optimization of Shelter Site Selection Following a Pre-Expropriation System

Abstract

Shelter site selection is a key topic in research on the preparedness and response phases of disaster management, significantly impacting the availability of safe, temporary residential refuges and rescue locations during disasters. The pre-expropriation system presents a novel direction for the government to reduce response time and improve efficiency. However, the question arises: how can a government establish a pre-expropriation–emergency–expropriation–compensation system targeting collective economic organizations (a kind of public ownership economic organization that the means of production are owned by part of the laborers), enterprises, and institutions by comprehensively considering the correlation between disaster risk and shelter location under a pre-expropriation system? This study explores a three-level networked distribution center–shelter–disaster point structure and establishes a model for selecting shelter sites that is based on a pre-expropriation system and flood risk assessment. Furthermore, we developed a multiobjective mixed-integer stochastic programming model, with the objectives of achieving the shortest distribution time and evacuation time, the least number of people not evacuated in time, the lowest disaster risk, and the smallest government compensation payment. The proportion of the evacuated population at each disaster point was considered as a random variable. Given the model’s high-dimensional nature, a multiobjective artificial bee colony algorithm based on crossvariation was designed for model solving. By comprehensively balancing the interests of governmental agencies, expropriation locations, and disaster victims, this study considers multiple objective factors and constructs a regional relative flood risk assessment model and a shelter location model grounded in multiobjective mixed-integer stochastic programming. These provide scientific and effective theoretical support for optimal pre-expropriation shelter site selection. The case study offers recommendations for selecting optimal pre-expropriation shelters based on risk scenarios, demonstrating that the multiobjective framework constructed provides decision makers with better flexibility and applicability.