https://github.com/hq969/deepfleet-ai
DeepFleet-AI is an AI-powered fleet route optimization platform that combines machine learning, geolocation algorithms, and cloud infrastructure to enhance delivery efficiency and reduce operational costs. It offers a modular architecture with a FastAPI backend, Flask APIs, machine learning models, a React frontend (optional), and AWS integration
https://github.com/hq969/deepfleet-ai
aws fastapi reactjs terraform
Last synced: 2 months ago
JSON representation
DeepFleet-AI is an AI-powered fleet route optimization platform that combines machine learning, geolocation algorithms, and cloud infrastructure to enhance delivery efficiency and reduce operational costs. It offers a modular architecture with a FastAPI backend, Flask APIs, machine learning models, a React frontend (optional), and AWS integration
- Host: GitHub
- URL: https://github.com/hq969/deepfleet-ai
- Owner: hq969
- License: mit
- Created: 2025-08-03T06:06:52.000Z (11 months ago)
- Default Branch: main
- Last Pushed: 2025-08-04T06:10:12.000Z (11 months ago)
- Last Synced: 2025-09-02T15:01:30.750Z (10 months ago)
- Topics: aws, fastapi, reactjs, terraform
- Language: Python
- Homepage:
- Size: 43.9 KB
- Stars: 1
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# DeepFleet-AI 🚚📦
**AI-Powered Logistics Fleet Optimization System**
DeepFleet-AI is an end-to-end, production-grade fleet management and route optimization platform built using **Machine Learning**, **AWS Cloud**, and **React**. It enables intelligent delivery routing, ETA prediction, and real-time fleet visibility using historical delivery data and live tracking.
---
## 🌐 Tech Stack
### 🧠 Backend & ML
- **Python**, **Flask**
- **Scikit-learn**, **Pandas**, **NumPy**, **XGBoost**
- **Geopy**, **Haversine**, **Folium**
### 🌍 Frontend (optional)
- **React.js**, **Tailwind CSS**, **Recharts**
### ☁ Cloud/DevOps
- **AWS EC2**, **S3**, **CloudWatch**, **Lambda**
- **Docker**, **GitHub Actions**
---
## 📁 Project Structure
```bash
DeepFleet-AI/
├── backend/
│ ├── main.py # FastAPI entrypoint
│ ├── routes/
│ │ ├── fleet.py
│ │ └── auth.py
│ ├── services/
│ │ └── optimizer.py # Route optimization logic
│ ├── models/
│ │ └── delivery_model.py # SQLAlchemy models
│ └── utils/
│ └── geo_utils.py # Geolocation, Haversine formula
│
├── ml/
│ ├── model.py # Fleet routing ML model
│ ├── train.py # Training script
│ ├── predict.py # Inference logic
│ └── data/
│ └── delivery_logs.csv # Sample training data
│
├── frontend/ # Optional React Dashboard
│ └── src/
│ └── components/
│ └── pages/
│ └── Dashboard.jsx
│
├── docker/
│ └── Dockerfile
│ └── docker-compose.yml
│
├── scripts/
│ └── seed_db.py # Populate DB
│ └── scheduler.py # Fleet update scheduler (Lambda)
│
├── infrastructure/
│ └── terraform/ # AWS setup
│ └── sagemaker-deploy.tf
├── api/ # Flask backend APIs
│ ├── app.py # API entrypoint
│ ├── routes/
│ │ └── delivery_routes.py
│ └── utils/
│ └── helpers.py
├── README.md
└── requirements.txt
````
---
## 🚀 Features
* ✅ Predict ETA (Estimated Time of Arrival) using ML
* ✅ Route Optimization using Haversine/Mapbox APIs
* ✅ Vehicle Type Classification
* ✅ Data ingestion from CSV or REST
* ✅ Scalable deployment with Docker
---
## 📦 Setup Instructions
### 1. Clone the repo
```bash
git clone https://github.com/hq969/DeepFleet-AI.git
cd DeepFleet-AI
```
### 2. Backend Setup
```bash
cd api/
python -m venv venv
source venv/bin/activate
pip install -r ../requirements.txt
python app.py
```
### 3. Frontend Setup (Optional)
```bash
cd frontend/
npm install
npm run dev
```
### 4. Run Docker (Alternative)
```bash
docker build -t deepfleet-backend .
docker run -p 5000:5000 deepfleet-backend
```
---
## 📊 Sample API
**Endpoint:** `/predict_eta`
**POST Body:**
```json
{
"origin": [28.6139, 77.2090],
"destination": [28.5355, 77.3910],
"vehicle_type": "van"
}
```
**Response:**
```json
{
"eta_minutes": 42.5
}
```
---
## 📂 Dataset
Sample delivery logs can be found in:
```
ml/data/delivery_logs.csv
```
You can expand this with more real-world logs.
---
## 📌 To-Do
* [ ] Integrate Mapbox/Google Directions API
* [ ] Add MongoDB or PostgreSQL for delivery log storage
* [ ] Real-time GPS data streaming via AWS Kinesis
---
## 👨💻 Contributors
* **Harsh Sonkar** — [LinkedIn](https://www.linkedin.com/in/harsh-sonkar-232573250)
---
## 📄 License
MIT License. Free to use and modify with attribution.
---