Modern supply chains have grown so complex that they require intelligent systems to dynamically optimize supply allocation according to changing demand. This survey looks at the most recent breakthroughs in AI-driven supply chain planning, which are focused on predictive analytics, machine learning, and optimization techniques to improve demand forecasting and supply allocation. We consider AI methods, including deep neural networks, reinforcement learning, constraint programming, and hybrid models that improve the agility, operational efficiency, and profitability of the supply chain. Through a critical review of the literature and industry case studies, we derive performance benchmarks achieved by AI-based supply chain systems. The studies reveal that AI-based planners can improve order fulfillment rates by 15–20%, revenue by 10–15%, and demand-fluctuation insensitivity by over 20%, significantly outperforming the traditional rule-based methods. Furthermore, AI facilitates real-time decision-making, reducing computational overhead and response time in dynamic market scenarios. The paper synthesizes research findings and industry applications of AI-based methods for benchmarking the methods used in demand-supply matching and concludes with upcoming trends, challenges, and research areas for AI-based optimization in supply chain systems.