In today’s highly competitive retail and industrial landscape, multiproduct retail systems face growing challenges due to complex operations, fluctuating demand, and market uncertainty. This paper presents a data-driven framework for optimizing integrated decision-making and enhancing operational efficiency. By utilizing historical transaction data and advanced analytical techniques, the model combines key operational functions—including demand forecasting, inventory management, and resource allocation—to support real- time, data-informed decisions. The approach employs predictive modeling and optimization algorithms to minimize operational costs while maintaining product availability and service level targets. The initial model features five interconnected components: inspection, distribution, disposal, recovery, and retail centers. However, it currently excludes forward logistics, fleet operations, and is limited to a single product and planning period. To address supplier uncertainty, a deterministic equivalent formulation is introduced, relying on the estimation of statistical moments from limited data. Since supplier selection is critical to effective sourcing strategies, improving this process directly enhances supply chain performance. The study highlights that accurately identifying and modeling operational uncertainties is essential for achieving robust and optimal outcomes in retail environments.
Aryza et al. (Thu,) studied this question.