ADVANCED OPERATIONS OPTIMIZATION ANALYSIS USING MACHINE LEARNING AND MANAGEMENT SCIENCE

Abstract

Operations optimization in modern supply chains requires the integration of predictive and prescriptive analytics to address uncertainty and complexity. Traditional management science methods provide structured optimization frameworks but often rely on deterministic assumptions, while machine learning offers improved forecasting capabilities without directly supporting decision-making. This study develops an integrated framework that combines machine learning and management science to enhance operational performance. A structured dataset comprising demand, inventory levels, production costs, transportation costs, lead time, and service level is utilized. Machine learning models are applied to forecast demand, and these predictions are incorporated into an optimization model aimed at minimizing total operational cost under capacity and service constraints. The approach treats forecasts as probabilistic inputs, improving robustness in decision-making. The results demonstrate that the integrated framework enhances efficiency by reducing stockouts, improving capacity utilization, and balancing cost–service trade-offs. The study highlights the importance of linking prediction with optimization for effective decision-making.