Based on the queuing theory model, this paper proposes that the utilization rate of charging piles is an important consideration in the site selection planning of charging stations. Based on this, the multi-objective site selection optimization model of the charging station is established by taking the user’s charge rejection rate, the utilization rate of the charging pile, and the total investment cost of operators as three objectives. Furthermore, this paper conducts experimental analysis based on real vehicle trajectory data and analyzes the actual charging demand of users by studying the resident behavior of vehicles while driving. The ArcGIS software is introduced to select suitable candidate charging stations from the map using its excellent spatial computing ability to plan candidate points. To solve the model, an improved multi-objective particle swarm optimization algorithm is proposed, and compared with other intelligent optimization algorithms, the rationality and effectiveness of the improved algorithm are verified. This paper further analyzes the influence of different charging station quantities and charging pile power on each index in the charging station planning process through simulation experiments. The results show that blindly building charging stations will not improve service efficiency when the number of charging stations reaches a certain scale. Different charging power has little influence on the rejection rate of charging stations, but it will significantly affect the utilization rate of charging piles.