Global supply chains are increasingly characterized by complexity, uncertainty, and vulnerability to disruptions, creating a pressing need for intelligent, adaptive systems that support decentralized decision-making and real-time control. This paper develops a new framework that integrates Multi-Agent Systems (MAS) with Case-Based Reasoning (CBR) to address these challenges. The model leverages autonomous agents representing suppliers, manufacturers, distributors, retailers, and coordinators that negotiate through defined protocols while embedding CBR mechanisms to retrieve and adapt historical supply chain cases for enhanced responsiveness. An optimization layer, guided by both agent heuristics and case-driven initial solutions, targets key objectives such as cost minimization, lead-time reduction, and resilience improvement. Simulation experiments were conducted under both static and dynamid environments with disruptions including supplier failures and demand fluctuations. Results demonstrate that the proposed framework achieves convergence up to 34- 41% faster than heuristic-only baselines (p<0.05) and sustains solution quality with supply chain sizes increasing from 50 to 500 agents, indicating near-linear scalability. Comparative analysis further highlights adaptability in dynamic contexts and robustness under uncertainty. A case study illustrates practical deployment and validates its effectiveness. The findings provide evidence of a powerful synergy between MAS and CBR, with implications for next-generation supply chain intelligence.