This paper presents a predictive forecasting framework designed to enhance inventory and logistics efficiency in consumer goods supply chains. Addressing the challenges of demand variability, lead time uncertainties, and complex operational dynamics, the framework integrates advanced machine learning techniques with comprehensive data collection and preprocessing strategies. Through the identification of key variables influencing supply chain performance and the application of robust predictive analytics, including ensemble and time series models, this framework offers a unified architecture for simultaneous inventory replenishment and logistics optimization. Rigorous model validation using industry-standard metrics demonstrates significant improvements in forecast accuracy, leading to cost reductions, improved service levels, and enhanced operational agility. The findings underscore the framework’s potential to provide supply chain managers with actionable insights, facilitating real-time decision-making and fostering collaboration across stakeholders. Limitations regarding data quality and model interpretability are discussed, along with future research directions to expand the framework’s applicability and responsiveness to market dynamics.