Earthquakes are devastating natural disasters that necessitate the design of reasonable emergency shelters for areas likely to be severely damaged by earthquake. In this paper, we describe our use of a particle swarm optimization algorithm to study urban and rural emergency shelter planning. We combined the particle swarm optimization theory of the annealing algorithm with the planning of emergency shelters in urban and rural residential areas. In the case of many constraints, the emergency shelter is regarded as a particle, and the objective function is to plan emergency shelters, i.e., the scale planning and design of emergency shelters in urban and rural residential areas. Next, we designed a model of the content and location of the emergency sites, and identified the factors influencing the layout of urban and rural demand points. We then built different emergency shelter facility points. On this basis, we achieved the overall planning of emergency shelters that provided coverage of all the demand points. The simulation results show that the proposed particle swarm algorithm has a good planning efficiency and can ensure that after urban and rural residential areas are damaged by earthquake, victims will have access to immediate shelter.