🚌 Accra Transit Optimizer

AI-powered transit optimization for Accra’s public transport (Trotros) Built with a sleek Flutter app (Cupertino-style) + FastAPI backend + ML models for route and demand prediction.

---

📦 Project Structure

accra-transit-optimizer/
│
├── backend/                  # FastAPI backend
│   ├── api/                  # API routes and logic
│   ├── models/               # Trained ML models
│   ├── utils/                # GTFS & preprocessing
│   ├── data/                 # GTFS data & generated features
│   ├── main.py               # API entry point
│   └── requirements.txt      # Python deps
│
├── frontend/                 # Flutter app
│   ├── lib/
│   │   ├── screens/          # Main UI (home, analytics, routes)
│   │   ├── services/         # API integration
│   │   ├── providers/        # State management
│   │   └── main.dart         # App entry point
│   ├── pubspec.yaml
│   └── assets/               # Branding, splash

---

🚀 Features

✅ GTFS stop parsing & mapping ✅ ML demand prediction model ✅ AI-suggested optimal routes ✅ Dynamic analytics dashboard ✅ Realtime GPS location detection ✅ Sleek Cupertino-inspired UI ✅ Auto-filled coordinates with override ✅ Toast-based error handling ✅ Hosted FastAPI backend (Render)

---

🧠 Tech Stack

Layer	Technology
Backend	FastAPI, Python, Scikit-learn, XGBoost
Frontend	Flutter 3.x (Cupertino), Provider
Mapping	flutter_map + OpenStreetMap
Data	GTFS Accra 2016 + synthetic GPS
ML Models	Route optimization, demand prediction

---

🔧 Setup Instructions

▶️ Backend (FastAPI)

cd backend/
python -m venv venv
source venv/bin/activate   # Windows: venv\Scripts\activate
pip install -r requirements.txt
python train_models.py
uvicorn main:app --reload

• API: http://localhost:8000/docs
• Health: /health

📱 Frontend (Flutter)

cd frontend/
flutter pub get
flutter run -d chrome       # Or `-d android`

---

📡 API Endpoints (Deployed on Render)

Method	Endpoint	Description
GET	/api/v1/stops	All stops (with optional demand)
POST	/api/v1/predict_demand	Predict demand for stop & time
GET	/api/v1/suggest_routes	AI-optimized route suggestions
GET	/api/v1/suggest_routes_from_to	Suggest based on start → end
GET	/api/v1/analytics	Returns analytics summary
POST	/api/v1/submit_gps	Submit live GPS data

📄 Docs: https://accra-transit-optimizer.onrender.com/docs

---

✨ App Highlights

• 📍 GPS Location Detection

  • Auto-prefills coordinates
  • Toasts: e.g., “Location Detected: 5.6, -0.18”

• 🗺️ Flutter Map UI

  • Displays GTFS stops & route markers
  • Highlights viable paths with polylines

• 📊 Analytics Screen

  • Fetches live analytics from /analytics
  • Stylish cards: peak hours, high-traffic stops, etc.

• 🔄 Route Suggestions

  • From start → end (user input or GPS)
  • Filters: demand threshold & viability

---

💅 Design & UX

• 🍎 Cupertino-style (rounded, elegant)
• 🧭 Responsive layouts
• 🎨 Splash screen with minimal branding
• 🔔 Toast feedback for all actions

---

🔮 Future Enhancements

• 🔄 Live trotro tracking via real-time GPS
• 🧠 Model auto-retraining
• 📊 Admin dashboard (Web)
• 🚏 Stop clustering & demand heatmaps
• 📅 Schedule optimization (OR-Tools)

---

🙌 Credits

• Lead Developer: [Caleb Botchway / DEVKD.]
• Data Source: 2016 Accra GTFS
• Frameworks: Flutter, FastAPI, Scikit-learn, OSMnx, GTFS-Kit

---

📜 License

MIT License — free to use, modify & contribute.

---

🚀 Optimizing Accra’s future, one route at a time.

---