ChowdriFurkhan07/autonomous-route-optimization-engine

🚚 Autonomous Fleet Route Optimizer (Bengaluru VRP)

A Streamlit-based Vehicle Routing Problem (VRP) optimizer built using a custom Genetic Algorithm.

Designed for Bengaluru urban logistics simulation.

🚀 Features

• —Multi-vehicle routing
• —Vehicle capacity constraints
• —Traffic simulation (peak hour multiplier)
• —Genetic Algorithm optimization
• —GA convergence tracking
• —Route comparison vs random solution
• —Interactive Bengaluru map (Folium)
• —CSV route export
• —Docker deployment ready
• —Hugging Face Spaces compatible

🧠 Problem Solved

This app solves a real-world Vehicle Routing Problem (VRP) where:

• —Vehicles start and end at a depot
• —Each customer has demand
• —Vehicles have capacity constraints
• —Goal is to minimize total travel distance

Distance is calculated using Haversine formula.

📂 Project Structure

. ├── app.py ├── requirements.txt ├── Dockerfile ├── README.md ├── data/ │ └── rawlocations.csv └── src/ ├── dataloader.py ├── optimizer.py ├── constraints.py └── vrp_service.py

🐳 Docker Deployment (Local)

Build image:

docker build -t vrp-app .

Run container:

docker run -p 7860:7860 vrp-app

Open in browser:

http://localhost:7860
## 🤗 Hugging Face Deployment

Runtime: Docker

Exposed Port: 7860

Dockerfile included

No extra config required

⚙ Configuration Options (Sidebar)

Number of Stops

Traffic Hour

Number of Vehicles

Vehicle Capacity

Peak traffic multiplier:

8–11 AM

5–8 PM

📊 Output

Optimized Distance

Random Distance Comparison

Vehicle Routes

GA Convergence Graph

Interactive Map

Downloadable CSV

🛠 Tech Stack

Python

Streamlit

Genetic Algorithm (Custom)

Folium

NumPy

Pandas

Docker

📌 Future Improvements

Real road routing API (OpenRouteService / Google)

Time window constraints

Live traffic integration

Cloud database support

Multi-depot VRP

Production REST API version

👩‍💻 Author

Chowdri Furkhan
AI/ML & Data Science Enthusiast
Bengaluru, India