TY - CONF TI - Leveraging Multi-agent System Powered by Large Language Model to Improve Transparency and Reliability in Automated Supply Chain Coordination C1 - Osaka and Kyoto, Japan C3 - Proceedings of the 33rd Annual Conference of the International Group for Lean Construction (IGLC 33) SP - 681 EP - 692 PY - 2025 DO - 10.24928/2025/0102 AU - Zhao, Xianxiang AU - Kuang, Wenyi AU - Kim, Yong-Woo AD - Ph.D. Student, Department of Construction Management, University of Washington, Seattle, USA, seanzhao@uw.edu, https://orcid.org/0009-0000-8378-0644 2 Researcher, Khoury College of Computer Sciences, Northeastern University, Boston, USA, kuang.w@northeastern.edu, orcid.org/0000-0002-2008-5924 3 Professor, Department of Construction Management, University of Washington, Seattle, USA, yongkim@uw.edu, https://orcid.org/0000-0001-7020-0700 AD - Researcher, Khoury College of Computer Sciences, Northeastern University, Boston, USA, kuang.w@northeastern.edu, orcid.org/0000-0002-2008-5924 AD - Professor, Department of Construction Management, University of Washington, Seattle, USA, yongkim@uw.edu, https://orcid.org/0000-0001-7020-0700 ED - Seppänen, Olli ED - Koskela, Lauri ED - Murata , Koichi AB - Lack of transparency and information reliability in supply chain management have been persistent challenges. Based on the LangChain and LangGraph frameworks, this research proposed a Large Language Model (LLM)-based Multi-Agent System (MAS) specifically designed to enhance information reliability and transparency in construction supply chain coordination. A prototype system composed of multiple autonomous agents was designed and developed capable of working collaboratively, sharing information, and supporting decision-making. The system comprises Supplier Agents and General Contractor Agents capable of engaging in natural language interactions. These agents coordinate the supply chain by facilitating communication about material deliveries and project progress. The prototype demonstrated the potential of LLM-based MAS in improving supply chain transparency and reliability. This research not only validated the feasibility of applying large language models in automated supply chain coordination but also offered insights for the design and implementation of future systems. KW - Lean construction KW - Supply Chain Coordination KW - Large Language Model KW - Multi Agent System. PB - T2 - Proceedings of the 33rd Annual Conference of the International Group for Lean Construction (IGLC 33) DA - 2025/06/02 CY - Osaka and Kyoto, Japan L1 - http://iglc.net/Papers/Details/2306/pdf L2 - http://iglc.net/Papers/Details/2306 N1 - Export Date: 02 June 2025 DB - IGLC.net DP - IGLC LA - English ER -