The rapid development of the e-commerce industry has brought new challenges to logistics networks. Constrained clustering algorithms have been introduced as an optimization tool to solve the construction problem of e-commerce positive and reverse logistics networks. This research is based on a constrained clustering algorithm. First, the spatial relationship between e-commerce suppliers, customers and warehouses is clustered and analyzed to determine the best logistics node. Then, the path planning algorithm is used to determine the optimal logistics route to reduce the transportation distance and time. At the same time, reverse logistics processing methods are applied to optimize the return and exchange process of goods to reduce reverse logistics costs. The research results show that compared with the traditional method, this method reduces the logistics cost by about 1,000 yuan. Under the same logistics requirements and constraints, the e-commerce forward and reverse logistics network based on the constrained clustering algorithm can significantly reduce the logistics time.