A review and comparative analysis of predictive models for supply chain demand forecasting

Accurate demand forecasting is critical to supply chain optimization, influence inventory management, production scheduling, and customer satisfaction. This paper presents a comparative analysis of traditional forecasting models and machine learning approaches for supply chain demand prediction. This study reviews key techniques, including statistical time-series models, supervised and unsupervised learning algorithms, ensemble methods, and deep learning architectures. Empirical evidence from retail, manufacturing, healthcare, and food sectors demonstrates their relative performance and practical applicability. Challenges such as data quality, model interpretability, and system integration are analysed, along with emerging trends in real-time adaptability, hybrid modelling, and explainable AI. By synthesizing current research and implementation insights, this work provides a comprehensive evaluation of existing methods, identifying their strengths, limitations, and future research directions to enhance data-driven demand forecasting in modern supply chains.