Supply chain coordination using machine learning

Efficient supply chain management is a critical yet challenging objective in modern business, aiming to align product delivery with demand while minimizing costs. Despite its importance, coordination and collaboration among supply chain partners are often inadequate, leading to significant inefficiencies. Research highlights substantial annual waste due to poor coordination, particularly in industries like the U.S. food sector. This study addresses these challenges by developing innovative tools to optimize pricing strategies and enhance manufacturer-retailer coordination, focusing on the integration of bi-level programming models and Artificial Neural Networks (ANN).Building on existing models such as the Single-level Mixed-integer Linear Programming Model with Strictly Complementary Conditions and Participation Constraints (SMLP-SCP), this research introduces a Bi-level Optimization Model with ANN Approximation (BL-ANN). The BL-ANN model leverages ANN to predict retailers' ordering decisions and optimize production plans, aiming to maximize manufacturers' profits and enhance overall supply chain efficiency.Through comparative analysis, BL-ANN demonstrates superior computational efficiency and scalability, achieving near-optimal profitability with minimal discrepancies in wholesale pricing decisions compared to SMLP-SCP. This study's findings suggest that BL-ANN is a robust, scalable, and efficient alternative for modern supply chain management, offering significant advantages in computational speed and decision-making accuracy. Future research will focus on integrating real-time data analytics and adaptive learning to further enhance the model's predictive capabilities and responsiveness to market dynamics. This study contributes to advancing supply chain management by combining advanced optimization techniques and artificial intelligence, paving the way for innovative solutions in the field.