InvAgent is a novel approach leveraging large language models (LLMs) to manage multi-agent inventory systems. It enhances resilience and improves efficiency across the supply chain network through zero-shot learning capabilities, enabling adaptive and informed decision-making without prior training. For more detailed information, please check our paper.

1. Clone the repository:

   git clone https://github.com/zefang-liu/InvAgent.git
   cd InvAgent

2. Install the required packages:

   pip install -r requirements.txt

• To run the AutoGen experiments, use notebooks/autogen.ipynb. Note that an OPENAI_API_KEY is required as an environment variable.

• The main environment setup is found in src/env.py.
• Configure the environment settings in src/config.py.
• Implement custom inventory management policies in src/baseline.py.
• For specific implementations of IPPO and MAPPO, refer to src/ippo.py and src/mappo.py, respectively.

If you find this repository useful in your research, please consider citing our paper:

@article{quan2024invagent,
  title={InvAgent: A Large Language Model based Multi-Agent System for Inventory Management in Supply Chains},
  author={Quan, Yinzhu and Liu, Zefang},
  journal={arXiv preprint arXiv:2407.11384},
  year={2024}
}

For more information or any inquiries, please feel free to raise an issue or contact us directly.

This project is licensed under the Apache-2.0 license. See the LICENSE file for details.