<![CDATA[Accurate demand forecasting is a critical determinant of efficiency and resilience in industrial supply chains. Increasing market volatility, shortened product life cycles, and complex global networks have exposed the limitations of traditional statistical forecasting approaches. Artificial Intelligence (AI), particularly machine learning and deep learning, has emerged as a transformative solution capable of processing large-scale, heterogeneous data and capturing nonlinear demand patterns. This study aims to systematically analyze and synthesize empirical and conceptual evidence on AI-based demand forecasting in industrial supply chains. Using a structured literature-based analytical method, this research reviews and integrates findings from peer-reviewed journal articles, conference proceedings, and authoritative preprints published between 2020 and 2025. The results demonstrate that AI-based forecasting models—such as neural networks, ensemble learning, hybrid ARIMA–LSTM architectures, and predictive analytics—consistently outperform traditional methods in terms of accuracy, adaptability, and responsiveness. Moreover, AI-driven forecasting contributes significantly to improved inventory optimization, cost reduction, and supply chain resilience. However, challenges related to data quality, implementation cost, system integration, and skill gaps remain substantial barriers. The study concludes that AI-based demand forecasting is not merely a technological enhancement but a strategic capability for industrial supply chains. Practical implications and directions for future research are discussed to support broader and more effective adoption of AI-driven forecasting systems.]]>
