Project APOLLO

Predictive Process Simulation & Operational Forecasting Engine

🏆 Submission for the IIT Madras x Appian AI Application Challenge 2026.

APOLLO is a cutting-edge application designed to revolutionize operational management by moving from reactive problem-solving to proactive prediction and optimization. Leveraging a Hybrid Neuro-Symbolic Architecture, APOLLO provides a "time machine" for business processes, allowing managers to foresee SLA breaches, simulate "what-if" scenarios, and optimize resource allocation before issues even arise.

---

✨ Our Solution: The Four Pillars of Prophecy

APOLLO integrates advanced AI and simulation techniques to deliver unparalleled operational intelligence:

1. Digital Twin Engine (SimPy): A living, breathing replica of your Appian workflows, accurately modeling queues, resources, and process logic.
2. Hybrid Forecasting Engine (Prophet + TFT): An intelligent oracle that predicts future case volumes with high accuracy, combining the power of statistical and deep learning models.
3. Monte Carlo Risk Engine: Runs thousands of parallel simulations to quantify risk, providing probabilistic forecasts of SLA breaches and bottlenecks.
4. What-If Sandbox & Optimizer (PuLP): An interactive control panel allowing managers to test resource allocation changes and receive data-driven recommendations for optimal strategies.

---

🚀 Getting Started (Local Development)

Follow these steps to get APOLLO up and running on your local machine.

1. Prerequisites

• Docker Desktop: Ensure Docker is installed and running on your system.
• Node.js & npm: Required for frontend development (though Docker handles most of this).
• Python 3.11+: For backend development (though Docker handles most of this).

2. Clone the Repository

git clone https://github.com/your-username/appian-apollo.git
cd appian-apollo

3. Generate Synthetic Data

Our backend requires historical data to train its forecasting model.

# Navigate to the data directory
cd data

# Run the Python script to generate sample_operation_log.csv
python synthetic_log_generator.py

# Go back to the project root
cd ..

4. Configure OpenAI API Key

The GenAI Analyst Layer requires an OpenAI API key.

• Create a file named .env in the root directory of the project (appian-apollo/.env).
• Add your OpenAI API key to this file:

  OPENAI_API_KEY="your_openai_api_key_here"
  

5. Run the Application with Docker Compose

This command will build the Docker images for both the backend and frontend, and then start both services.

docker-compose up --build

• The first build might take a few minutes.
• Once running, you can access the application:
  • Frontend UI: http://localhost:3000
  • Backend API Docs (Swagger UI): http://localhost:8000/docs

---

🛠️ Tech Stack

• Backend: Python, FastAPI, SimPy, Prophet, NumPy, Pandas, PuLP, OpenAI API
• Frontend: React, TypeScript, Recharts, React Router
• Containerization: Docker, Docker Compose

---

🤝 Contributing

(Optional section for future contributions)

---

📄 License

This project is licensed under the MIT License.