Integrating Graph Retrieval-Augmented Generation With Large Language Models for Supplier Discovery

This research article presents a methodology that integrates large language models (LLMs) and knowledge graphs (KGs) to enhance supplier discovery in manufacturing supply chains. The document outlines the challenges faced by supply chains, including a lack of resilience and visibility, which can result in disruptions and inefficiencies. The proposed methodology aims to transform unstructured supplier capability data into a harmonized KG, thereby improving the accessibility and findability of manufacturing suppliers. It details an ontology-driven graph construction process that combines KGs with retrieval-augmented generation techniques, enhancing the quality of supplier information. A case study is included to demonstrate the effectiveness of this integrated approach, which not only improves the visibility of small- and medium-sized manufacturers but also increases agility and provides strategic insights into supply chain management. The article concludes with a discussion of limitations and future directions for this research.