Against the backdrop of economic globalization, supply chain management has emerged as a critical strategic tool for enterprises to enhance their competitiveness. This paper briefly analyzes the multi-agent game dynamics within logistics and supply chains, and based on this analysis, constructs a mathematical generative model with operational costs and customer satisfaction as optimization objectives. To effectively solve the model, the honeypot algorithm is modified. The performance of the improved honeypot algorithm is evaluated by computing convergence curves and mean-variance metrics for 10 benchmark functions, followed by case studies. The mean values of the improved algorithm are generally lower than those of the comparison algorithms across all functions, ranging from 2.89E-10 to 5.58E-9, while exhibiting superior convergence. The optimal solution identified allocated regional shares of 56.32%, 26.88%, and 16.80% to the three logistics service providers. Compared to the original plan, this solution reduced operational costs by 38.57% and increased customer satisfaction by 28.16%, validating the effectiveness of the dynamic optimization model. This provides enterprises with a decision-optimization method for logistics and supply chain management.
