Application of recommender systems in supply chain management: a state-of-the-art review

Recommender Systems (RSs) are increasingly adopted in Supply Chain Management (SCM) to support complex decision-making under uncertainty, data heterogeneity, and operational constraints. This study presents a systematic literature review and bibliometric analysis of RS applications in SCM, covering peer-reviewed studies published between 2006 and April 2024. The review examines how RSs are used to optimize critical supply chain functions, including supplier selection, demand forecasting, inventory management, logistics, and sustainability-oriented decision-making. Major RS approaches—Content-Based Filtering (CBF), Collaborative Filtering (CF), Hybrid Filtering, and Knowledge-Based Filtering (KBF)—are systematically categorized and analyzed across key application domains such as pharmaceuticals, e-commerce, logistics, procurement, and green supply chains. The results reveal that Hybrid Filtering is the most dominant and effective approach in complex SCM environments, particularly in supplier management and logistics, due to its ability to integrate heterogeneous data sources and mitigate challenges such as data sparsity and cold-start problems. Knowledge-Based Filtering emerges as especially effective in sustainability-driven and high-risk domains, where transparency, explainability, and constraint-based reasoning are critical. The analysis further shows that advanced methods, including reinforcement learning, deep learning architectures, and blockchain-enhanced RSs, demonstrate superior performance in dynamic and trust-sensitive supply chains by enabling real-time adaptation, traceability, and multi-objective optimization. Bibliometric and thematic analyses indicate a clear research shift toward AI-driven, decentralized, and sustainability-oriented RS frameworks, particularly after 2020.