Supplier selection is a critical decision in sustainable supply chain management, where firms must balance environmental, economic, and social objectives while managing delay risks that threaten performance and resilience. Traditional evaluation methods struggle to address the scale and complexity of supplier data, motivating the use of intelligent analytics. This study develops a comprehensive framework that applies machine learning techniques which is a subset of artificial intelligence to assess suppliers across sustainability dimensions while explicitly incorporating delay risk. Supervised learning algorithms including support vector machine, decision tree, random forest, XGBoost, and multilayer perceptron are implemented, with neural networks delivering the most reliable performance. The analysis demonstrates how intelligent models uncover hidden patterns in supplier performance data and generate robust, explainable rankings that support strategic and tactical decision-making. The framework provides prescriptive insights for prioritizing suppliers, enabling firms to align selection processes with sustainability goals and strengthen resilience against disruptions. By integrating sustainability and delay risk management into a unified decision-support system, this research advances the role of intelligent analytics in building efficient, responsible, and adaptive supply chains. • Apply intelligent analytics to select sustainable suppliers and reduce delay risks in supply chains. • Automate supplier evaluation using machine learning for sustainability and resilience in operations. • Deliver prescriptive analytics that customizes supplier rankings by sustainability performance. • Use large-scale supplier data to generate reliable insights for sustainable supply chain management. • Provide analytical and managerial guidance on economic, social, and environmental supplier impacts.
Ziari et al. (Sun,) studied this question.