AI-DRIVEN DEMAND FORECASTING AND INVENTORY OPTIMIZATION IN SUPPLY CHAINS

Abstract

This dissertation delves into the revolutionizing impact of Artificial Intelligence (AI) in streamlining demand forecasting and inventory management in retail supply chains. As conventional forecasting models continue to become more and more insufficient in the complex, omnichannel world of current retailing, AI-driven methods—such as LSTM, neural networks, and assembling techniques—provide real-time, data-driven solutions. Based on case studies, empirical comparison of models, and a thorough review of the literature, this thesis assesses the effect of AI on forecast accuracy, cost savings, inventory turnover, and customer satisfaction. A conceptual framework is developed that connects AI architecture to retail KPIs, and directions for future work are described with a focus on explainability, ethics, and scalability.