Artificial Intelligence-Based Supply Chain Optimization Models Challenges and Future Directions

Abstract

Artificial Intelligence (AI) is becoming a key driver of change in the supply chain, particularly in its ability to predict, adapt and optimise in complex, uncertain environments. The review covers AI-based supply chain optimization models, applications, performance, challenges and future research directions. The study covers the key AI techniques, such as Machine Learning (ML), Deep Learning (DL), Reinforcement Learning (RL), Multi-Agent Reinforcement Learning (MARL), Generative AI and Large Language Models (LLMs), and hybrid AI–metaheuristic methods. Such methods are used for demand forecasting, inventory optimization, supplier selection, procurement, production scheduling, transportation, routing and supply chain network design. The review notes benefits of AI-based optimization, such as better forecasting accuracy, cost efficiency, resource utilization, adaptability, operational responsiveness, and supply chain resilience. Despite these advances, there are major hurdles such as data quality, interoperability, limited model explainability, cybersecurity and privacy issues, computational demands, scalability, and lack of empirical validation. The results highlight the importance of the development of integrated, reliable, explainable, and adaptive AI optimization frameworks that are able to assist with real-time decision-making. There is a need for further research on Generative AI, multi-agent systems, digital twins, sustainable optimization, privacy-preserving AI, and large-scale empirical validation, to further enhance practical supply chain implementation.