Despite the increasing importance of collaboration in achieving cost-related advantages for companies, existing studies lack a systematic framework for determining how multiple supply chains can collectively facilitate strategic decision-making. In this study, we propose a multi-network collaborative mixed integer programming model and a reinforcement learning enhanced evolutionary algorithm to optimize cross-enterprise supply chains. The model facilitates collaborative decisionmaking among supply chains of different enterprises by incorporating collaborative cost management, partner sharing, collaborative transportation, and horizontal logistics. Our algorithm integrates an adaptive reinforcement learning process and an evolutionary structure with multi-branch tree encoding, allowing for the effective accumulation of solving experiences in different states to enhance the efficiency and accuracy of the solving process. We conducted extensive experiments using real data collected from electric automotive manufacturing supply chains. The experimental results show that the obtained solution quality is close to optimal with negligible margin for both small- and large-scale instances. Overall, our proposed approach enables the joint optimization of cross-enterprise collaborative supply chains and holds the potential for improving supply chain management in various industries. [ABSTRACT FROM AUTHOR]