Enhancing supply chain visibility with knowledge graphs and large language models

In today's globalised economy, comprehensive supply chain visibility is crucial for effective risk management. Achieving visibility remains a significant challenge due to limited information sharing among supply chain partners. This paper presents a novel framework leveraging Knowledge Graphs (KGs) and Large Language Models (LLMs) to enhance supply chain visibility, without direct reliance on stakeholder information sharing. Our zero-shot, LLM-driven framework enables the automated extraction of diverse and domain-specific supply chain information from publicly available sources, and constructs structured knowledge graphs that capture complex, multi-tier interdependencies across supply chain entities, geographic locations, ownership structures, and product flows. We employ zero-shot prompting for Named Entity Recognition (NER) and Relation Extraction (RE) tasks, eliminating the need for extensive domain-specific training. We validate the framework through both quantitative evaluations and a case study on EV supply chains, specifically focussing on tracking critical minerals for battery manufacturing. The results demonstrates the effectiveness of the framework in supply chain mapping, which extends visibility beyond tier-2 suppliers. Additionally, the framework reveals critical dependencies and alternative sourcing options, enhancing risk management and strategic planning in case of disruptions. With high precision in NER and RE tasks, it provides an effective tool for understanding complex, multi-tiered supply networks. This research offers a novel framework for constructing domain-specific supply chain knowledge graphs, addressing longstanding challenges in visibility and paving the way for advancements in digital supply chain surveillance.