A multi-objective optimization based method for evaluating earthquake shelter location–allocation

Abstract

Constructing shelters is an important part of earthquake disaster management which involves selecting the location of them and assigning the evacuees to them. For this work, selecting suitable objectives and solution is important. Thus, in this paper, a multi-objective mathematical model with four groups of the objectives, allied with a modified particle swarm optimization algorithm, has been developed to solve the location–allocation problem for earthquake shelter. The four objective groups are: total shelter number (TSN) and total evacuation distance (TED), TSN and total weighted evaluation time (TWET), total shelter area (TSA) and TED, and TSA and TWET. The solutions of the model include the determination of the shelters from the candidates and how to allocate population to them. Then the solutions of the model with four objective groups are given and compared using safety, capacity and investment evaluation index with the case of Chaoyang district of Beijing, China. Related to government's preferences and future city planning, the most suitable model solutions can be chosen to help decide where it is suitable to construct shelters and how to allocate the population.