InvAgent: A Large Language Model based Multi-Agent System for Inventory Management in Supply Chains

2024-07-16

arXiv.org

AI Summary

Existing supply chain inventory management methods struggle to simultaneously achieve efficiency, robustness, and interpretability in dynamic, volatile environments—leading to stockouts, excessive holding costs, and suboptimal inter-organizational coordination. To address this, we propose InvAgent: the first large language model (LLM)-based multi-agent framework for collaborative inventory decision-making. InvAgent treats LLMs as autonomous, zero-shot agents—requiring no fine-tuning—while leveraging chain-of-thought (CoT) reasoning for transparent, interpretable inference and real-time demand responsiveness. By integrating multi-agent systems (MAS) architecture with joint modeling of inventory policies across echelons, InvAgent enables adaptive, decentralized coordination. Empirical evaluations demonstrate that InvAgent significantly reduces stockout rates and inventory holding costs under stochastic demand, thereby enhancing supply chain resilience. It consistently outperforms conventional heuristic and reinforcement learning baselines across diverse operational scenarios.

Technology Category

• Multiagent Systems: Coordination and Collaboration
• Planning, Routing, and Scheduling: Planning with Language Models
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Supply chain management (SCM) involves coordinating the flow of goods, information, and finances across various entities to deliver products efficiently. Effective inventory management is crucial in today's volatile and uncertain world. Previous research has demonstrated the superiority of heuristic methods and reinforcement learning applications in inventory management. However, the application of large language models (LLMs) as autonomous agents in multi-agent systems for inventory management remains underexplored. This study introduces a novel approach using LLMs to manage multi-agent inventory systems. Leveraging their zero-shot learning capabilities, our model, InvAgent, enhances resilience and improves efficiency across the supply chain network. Our contributions include utilizing LLMs for zero-shot learning to enable adaptive and informed decision-making without prior training, providing explainability and clarity through chain-of-thought, and demonstrating dynamic adaptability to varying demand scenarios while reducing costs and preventing stockouts. Extensive evaluations across different scenarios highlight the efficiency of our model in SCM.

Problem

• Supply Chain Management
• Inventory Optimization
• Collaborative Systems

Innovation

• InvAgent
• Multi-agent Warehouse Systems
• Language Model for Autonomous Decision-making