Data-Driven Forecasting and Inventory Optimization using Machine Learning Models and Methods

The growing availability of big data and machine learning has led to a focus on data-driven solutions in Operations and Supply Chain Management (OSCM). This research aims to develop predictive and prescriptive analytics models to address challenges in supply chain forecasting and inventory optimization. The proposed Demand Forecasting Model (DFM) combines big data analytics and machine learning to provide accurate forecasts for operational and tactical decisions. A novel Demand Forecasting Decision Model (DFDM) is developed to provide long-term forecasts where big data models are not applicable. The research also explores the applicability of human judgment in the era of artificial intelligence. A novel supply chain forecasting framework is proposed, combining temporal and cross-sectional hierarchical forecasting to generate coherent forecasts at all levels of a supply chain. A data-driven inventory optimization model is developed, solving the multi-item inventory optimization problem with limited capacity/budget using machine learning and optimization. The proposed models and methodologies have numerous applications across various industries, including fashion, food, energy, electronics, and media. The proposed models and methodologies have numerous applications across supply chains of various industries.