README.md

metadata

title: Beer Game Experiment
emoji: 🍺
colorFrom: purple
colorTo: green
sdk: streamlit
sdk_version: 1.32.0
app_file: app.py
pinned: false
license: mit
python_version: '3.10'

🍺 Beer Game — Human & AI Collaboration

This is a Beer Game experiment interface built with Streamlit and hosted on Hugging Face Spaces.
The goal is to study human behaviour in supply chain decision-making when collaborating with AI agents (Factory, Wholesaler, Retailer powered by OpenAI GPT-4o-mini).

---

🎮 Roles

• Distributor: Human participant — you control the order quantity each week.
• Factory / Wholesaler / Retailer: AI agents making decisions using LLMs.

---

📊 Features

• Classic Beer Game with 24 weeks, including a demand shock at week 6.
• Distributor inputs weekly order quantity; AI agents respond automatically.
• Optional Information Sharing toggle: reveal historical customer demand to all roles.
• Multi-participant support: each participant gets a unique participant ID and per-participant CSV log.
• Rich data logging: timestamps, inventory, backlog, orders, info-sharing toggles, and raw AI output.
• Local CSV export & optional cloud upload (S3) for persistent storage.

---

🛠️ Running the Experiment

1. On Hugging Face Spaces

1. Deploy this repo as a Streamlit Space.

2. Add Secrets (Settings → Secrets):

   • OPENAI_API_KEY — your OpenAI API key
   • Optional for S3 export:
     • AWS_ACCESS_KEY_ID
     • AWS_SECRET_ACCESS_KEY
     • S3_BUCKET
     • S3_REGION (default: us-east-1)

3. Participants open the Space and either:

   • Enter a custom participant ID, or
   • Use the auto-generated ID shown in the app.

Example URL with participant ID:

https://huggingface.co/spaces//?participant_id=ABC123

---

2. Locally

pip install -r requirements.txt
streamlit run app.py