The development of e-commerce has brought the public into the era of e-commerce logistics 4.0. Therefore, optimizing logistics network location and path planning is important to improve logistics efficiency and reduce operating costs. To deal with the increasingly complex market demand and the changing distribution environment, the diversity enhancement mechanism and neighborhood search strategy are introduced into the particle swarm optimization algorithm. This optimized algorithm is combined with the density peak clustering algorithm. Then, the e-commerce logistics network location and path planning algorithm is proposed. The optimal path on the grid map only produced 11 inflection points and passed 24 grids. It was the best in the comparison model. Finally, the total route length in modeling and testing was 32.51 km, the average loading rate was 97.85%, the total cost was 668.18 ¥, and the calculation time was 25.34s. The proposed model performs well in many key performance indicators, which can realize the efficient operation of logistics network. The research aims to provide new ideas and methods for network location and path planning of e-commerce logistics 4.0, which has theoretical and practical application value.