<![CDATA[This research explores the use of AI-driven techniques to optimize supply chain decision-making by integrating demand forecasting, inventory management, and logistics optimization. The main objective is to enhance predictive accuracy while minimizing overall supply chain costs through the application of machine learning and reinforcement learning methods. The research design involves the development of a comprehensive mathematical model that combines AI-based demand forecasting with cost optimization in inventory and transportation. A machine learning model is employed to predict demand, while optimization techniques are used to minimize inventory and logistics costs. Reinforcement learning is introduced as a method for real-time decision-making, allowing the system to continuously adapt and improve. The methodology involves testing the model through a numerical example, where predicted demand is used to optimize inventory and logistics costs. The main results show that the AI-based model achieves a demand forecasting accuracy with a Mean Squared Error (MSE) of 50, resulting in a total supply chain cost of 760 units, which includes both inventory and transportation costs. Despite the initial prediction error, the model demonstrates the potential for cost savings and operational efficiency through better alignment of supply chain components. The research concludes that while the AI-driven approach offers significant improvements in supply chain management, further refinement of the predictive model and the practical application of reinforcement learning are necessary to fully realize its benefits. Future research should focus on enhancing model accuracy and scalability in real-world supply chain environments]]>
Copyrights © 2024
