Medical equipment suppliers play an indispensable role in sustaining healthcare supply chains because of their direct influence on patient health and safety. This study introduces a data-driven and intelligent framework for evaluating supplier performance with an emphasis on resilience and agility. To construct the framework, a set of critical indicators—namely quality consistency, response time, delivery speed, total cost, and regulatory compliance—was identified as the foundation for evaluation. These indicators were analyzed through the application of multiple machine learning algorithms to ensure accuracy, robustness, and interpretability. Comparative analysis showed that the CatBoost algorithm delivered the most reliable outcomes, achieving approximately 92% accuracy while maintaining balanced performance across key metrics such as F1 score, Precision, and Recall. Other methods, including Support Vector Machines (SVM) and Decision Trees, demonstrated moderate results but did not match CatBoost’s superior predictive capability. A sensitivity analysis was further conducted to uncover the most influential determinants of supplier performance. Findings revealed that quality consistency, responsiveness, and timely delivery were the strongest drivers, while cost and compliance played secondary yet important roles. These insights underscore that supplier evaluation cannot rely solely on traditional cost-based metrics but must integrate dynamic performance factors that directly affect resilience and agility. The study highlights the managerial implications of adopting data-driven evaluation models. Beyond improving accuracy in supplier assessment, such frameworks enable healthcare managers to allocate resources more effectively, select resilient partners, and mitigate supply risks. Consequently, the proposed model contributes to enhancing both efficiency and sustainability across healthcare supply chains.
Aryana Amiri (Wed,) studied this question.