An AI‑powered Mixed Integer Linear Program (MILP) optimization agent. “OptimizationAgent” is designed to assist in formulating, solving, and analyzing MILP problems, combining operational research techniques with agent frameworks.
Live Demo: optimizationagent.onrender.com
Formulate MILP problems (variables, constraints, objective)
Solve using backend solvers (Pulp)
Interpret and present solutions in human‑readable form
Agent architecture to assist with dynamic decision making
Web UI for interactive usage
Modular code structure (agents, API, utils)
├── agents/ # Core agent modules and logic
├── api/ # API endpoints (Flask / FastAPI or similar)
├── static/ # Static files (CSS, JS)
├── templates/ # HTML / Jinja templates
├── utils/ # Utility functions, helpers
├── app.py # Main application bootstrap
├── main.py # Launch entry point
├── run.py # Auxiliary runner script
├── tests.py # Test suite
├── requirements.txt # Python dependencies
├── roadmap.md # Planned enhancements & roadmap
├── Procfile # For deployment (e.g. Heroku / Render)
├── .gitignore
└── .gitattributes
Prerequisites
Python 3.8+
A MILP solver (Pulp)
Virtual environment setup (recommended)
Installation
git clone https://github.com/PARKER-X/OptimizationAgent.git
cd OptimizationAgent
python3 -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
pip install -r requirements.txt
Running Locally
# Using main entrypoint
python main.py
# Or with run helper
python run.py