Multi-Supply Chains Optimization Mechanism Based On Machine Learning And Double Auctions

Every enterprise in the supply chain will participate in managing the supply chain. The decisions made by each party will affect the future decisions of other members in the supply chain and themselves. There are trade-offs since the upstream and downstream of the multi-supply chain tackle problems from different perspectives based on their interests. With the increasing competition and cooperation among enterprises in multi-supply chains, game theory is widely used to analyze the competition and collaboration among enterprises. This study combines game theory and auction theory to obtain the relative optimal allocation through double auctions for multi-objective optimization. Nevertheless, there are a few issues, including the number of iterations and the potential for local monopoly in double auctions. To this end, the generalized genetic particle swarm optimization (GGPSO) algorithm is introduced to improve the double auctions mechanism, and finally, the global optimization of the supply chain is achieved. The simulation results show that the proposed method can efficiently complete the distribution and pricing among multi-supply chains and compensate the competing buyers with higher average quotations and the competing sellers with lower average quotes. Additionally, the GGPSO algorithm presented in this paper has a good performance in terms of the time needed to obtain the optimal solution, which is conducive to the global optimization of the supply chain.