Intelligent Supplier Selection and Disruption-Resilient Pharmaceutical Distribution Network Design: A Data-Driven Hybrid Framework

Abstract

The selection of an appropriate supplier in the pharmaceutical supply chain plays a crucial role in reducing costs, ensuring quality, and enhancing the resilience of the network due to the high sensitivity and the need for timely delivery of medications. The complexities and uncertainties within this chain further necessitate the use of innovative, data-driven approaches for supplier selection and decision-making improvement. In this study, a combined data-driven framework for supplier selection and pharmaceutical distribution network design under uncertainty is presented. Initially, key criteria were dynamically identified using a decision tree algorithm based on real-world supplier data, and then weighted using the Analytic Hierarchy Process (AHP). Subsequently, supplier ranking was performed using an improved version of the CoCoSo method, named CoCoFISo, which, by refining the normalization process and redefining the aggregation strategy, resulted in higher accuracy and stability compared to conventional methods. Finally, the outcomes of these stages were integrated into a two-stage stochastic optimization model to simultaneously optimize three objectives: minimizing total cost, maximizing supplier scores, and maximizing supply chain reliability. The implementation of the model on real-world data demonstrated that the proposed approach, especially the use of CoCoFISo, outperformed the TOPSIS and CoCoSo methods. This superiority was particularly evident in cost reduction, improved supplier selection, and increased resilience of the pharmaceutical network. The proposed framework can serve as an effective tool for decision-makers in the pharmaceutical industry to design sustainable and resilient networks capable of coping with demand fluctuations and operational disruptions.