Location-allocation model for earthquake shelter solved using MPSO algorithm

Constructing shelters in suitable quantities, with adequate capacities and at the right locations is essential for evacuees under earthquake disasters. As one of the disaster management methods, constructing shelters can help to significantly reduce disruption and devastation to affected population. Mathematical models have been used to solve this problem allied with a heuristic optimization algorithm. The optimization of evacuation efficiency, as one of the most important objectives, has many expressive forms, such as minimizing evacuation distance and evacuation time. This paper proposes a new model that aims to minimize evacuation time with a new calculation method and to maximize total evacuees' comfort level. The modified particle swarm optimization (MPSO) algorithm is employed to solve the model and the result is compared with a model that calculated evacuation time differently and a model without distance constraint, respectively.