Design and Development of a Machine Learning-Based Supply Chain Optimization Model: A Framework for Multiobjective Smart Decision Making

Abstract

This paper presents a machine learning-based framework for optimizing multiobjective supply chain networks, focusing on improving decision-making in complex and dynamic environments. The proposed model integrates predictive analytics, clustering, and risk assessment through advanced machine learning techniques to enhance various supply chain functions, including demand forecasting, inventory management, and transportation planning. Using these data-driven insights, a multiobjective optimization model is developed, balancing key performance criteria such as cost, service level, and delivery time. The model incorporates uncertainty by employing robust and probabilistic methods, ensuring reliability in fluctuating market conditions. A case study is conducted to validate the framework, demonstrating significant improvements in operational efficiency and decision quality. The results highlight the potential of combining machine learning with optimization techniques to address the growing complexity of modern supply chains, offering a scalable solution for smart supply chain management in diverse industries.