ARTIFICIAL INTELLIGENCE-DRIVEN OPTIMIZATION MODELS FOR SUPPLY CHAIN MANAGEMENT EFFICIENCY

The challenges of demand volatility, increased logistics costs, and disruption risk are mounting pressure on global supply chains, revealing the weaknesses of classical operations-research and rule-based planning tools. In this paper, the authors discuss the application of artificial intelligence (AI)-based optimization models to forecasting, inventory management, transportation routing, and supplier management to enhance the efficiency of the supply chain overall. A hierarchical model is suggested that combines machine learning-driven demand forecasting, reinforcement learning-driven inventory management, metaheuristic route planning, and AI-enabled supplier risk evaluation into one multi-objective decision layer. The study provides a comparison of the use of performance evidence synthesized based on recent publications on operations-management and computer-science to compare traditional planning methods with AI-based models based on five main performance metrics: forecast accuracy, inventory cost, on-time delivery, order cycle time, and stockout rate. The findings show that AI-based optimization outperforms conventional heuristics by a significant margin, and the most significant relative improvements are in inventory cost, and stockout reduction. The paper also presents the issue of challenges related to implementation, such as data quality, the interpretability of the model, and the readiness of the organization, and it also ends by providing the direction of future research regarding explainable and sustainability-aware AI optimization in supply chains.