A full-stack logistics dashboard that solves a variation of the Vehicle Routing Problem (VRP). It takes a pool of unassigned customer orders and uses a Genetic Algorithm to intelligently assign them to a fleet of vehicles, visualizing the optimized routes on an interactive map.
- Genetic Algorithm Engine: Utilizes heuristic evolutionary mechanics to generate, evaluate, and mutate thousands of possible routes, returning a near-optimal sequence to minimize total fleet travel distance.
- Interactive Geospatial Dashboard: Built with React-Leaflet, translating complex mathematical coordinates into a human-readable map of New Delhi.
- Fleet Management Interface: A dynamic carousel that calculates real-time metrics including Estimated Time of Arrival (ETA), total distance, and sequence of customer stops.
- Smart Highlighting: Clickable route isolation that dims unselected routes and places directional flow arrows (➤) on the active vehicle's path.
- Frontend: React.js, Vite, React-Leaflet
- Backend: Python, FastAPI
- Database: SQLite
- Algorithm: Custom Genetic Algorithm (Stochastic Optimization)
Open a terminal and navigate to the backend folder:
cd delivery-optimizer-backend
source venv/bin/activate
uvicorn main:app --reload