<p>With the significant intensification of competition in today’s global marketplace, the importance of supply chain management has been dramatically emphasized. Within this context, one of the most critical managerial tasks is the evaluation of potential suppliers’ performance. In recent years, the restrictions on in-person purchasing caused by the COVID-19 pandemic have further highlighted the growing importance of online shopping and e-commerce businesses. Therefore, this study focuses on the supplier evaluation process in the e-commerce sector by simultaneously considering two crucial perspectives: viability and customer-based dimensions. To achieve this goal, the main evaluation criteria and alternatives are first identified. Then, a white-box, machine learning–based decision support model is proposed by integrating the Random Forest Regression (RFR) and Particle Swarm Optimization (PSO) methods to assess suppliers’ performance based on the selected criteria. The applicability of the developed model is examined through a real-world case study. The obtained results indicate a significant relationship between the quality and robustness indicators. Furthermore, the most influential factors affecting supplier and vendor scores are found to be quality, delivery time, robustness, and cost. In addition, the performance of the proposed model is compared with other benchmark methods, and the results confirm its superior efficiency. Notably, the optimized RFR–PSO model achieves a prediction accuracy of 95.5%, outperforming conventional approaches such as Decision Tree and Artificial Neural Network. Finally, several managerial insights and practical implications are provided based on the findings.</p>