Food waste is a significant global challenge, contributing to environmental degradation and economic losses. This research paper explores the potential of predictive modeling and AI-driven inventory management systems in reducing food waste across the supply chain. By leveraging advanced machine learning algorithms, these systems can forecast demand with high accuracy, optimize inventory levels, and enhance decision-making processes in procurement, storage, and distribution. The study integrates various data sources, including historical sales data, seasonal trends, and consumer behavior patterns, to develop predictive models that anticipate fluctuations in demand. These models are further enhanced with real-time data from IoT sensors, enabling dynamic adjustments to inventory levels. The research also examines the role of AI in optimizing supply chain logistics, ensuring that perishable goods are distributed efficiently, and minimizing the likelihood of spoilage. Through case studies and simulations, the paper demonstrates the effectiveness of these AI-driven solutions in reducing food waste at both retail and consumer levels. The results indicate a significant reduction in overstocking, expiration, and spoilage, leading to improved sustainability and profitability for businesses. The findings underscore the importance of integrating predictive analytics and AI into inventory management systems as a proactive approach to mitigating food waste.