Optimization strategy of cross-border e-commerce supply chain network based on machine learning

Abstract

The rapid development of cross-border e-commerce increases the complexity of supply chain management, and machine learning-based supply chain network optimization strategies are essential for improving efficiency and reducing costs. The study first analyzes the cross-border supply chain network topology and determines the multilayer network structure. Subsequently, an optimization model based on particle swarm algorithm (PSO) is proposed, including the mathematical model of the algorithm and improvement measures. The practical effect of the optimization strategy was verified through example analysis. It is found that the supply chain optimized with particle swarm algorithm has significant Improvement in terms of shipping accuracy, surface transportation ratio, and unit transportation cost. For example, the shipping accuracy of product A in 2022 increased to 95.3% compared to 2021, the proportion of surface transportation increased to 96.5%, and the unit transportation cost decreased by RMB 2.175 per kilogram. This study shows that the particle swarm algorithm can effectively optimize the cross-border e-commerce supply chain network, which is significant in achieving efficient supply chain management.