steve-ongera/ccba_forecast

AI-powered demand forecasting platform for Coca-Cola Bottlers Africa beverage supply chain. Predicts product demand using historical sales + weather + holidays + events, trains ML models (Linear Regression, Random Forest, XGBoost), and surfaces forecasts via a React dashboard. # CCBA Demand Forecast — AI-Powered Demand Forecasting System AI-powered demand forecasting platform for Coca-Cola Bottlers Africa beverage supply chain. Predicts product demand using historical sales + weather + holidays + events, trains ML models (Linear Regression, Random Forest, XGBoost), and surfaces forecasts via a React dashboard. ## Stack - Backend: Django 5 + Django REST Framework, single core app - ML: scikit-learn, pandas, numpy, xgboost (Celery task or management command driven training) - DB: PostgreSQL - Frontend: React + Vite, Bootstrap Icons, custom CSS design system (`main_ccba.css`) - Auth: JWT (SimpleJWT), role-based (Admin, Analyst, Manager) ## Project Structure ``` ccba_forecast/ ├── backend/ │ ├── manage.py │ ├── requirements.txt │ ├── .env.example │ ├── ccba_forecast/ # project (main urls/settings) │ │ ├── __init__.py │ │ ├── settings.py │ │ ├── urls.py │ │ ├── asgi.py │ │ └── wsgi.py │ └── core/ # the one core app │ ├── __init__.py │ ├── admin.py │ ├── apps.py │ ├── models.py │ ├── serializers.py │ ├── views.py │ ├── urls.py │ ├── permissions.py │ ├── ml/ │ │ ├── __init__.py │ │ ├── train.py # model training (RF + XGBoost) │ │ ├── predict.py # generate forecasts │ │ └── features.py # feature engineering (weather/holiday/seasonality) │ ├── management/ │ │ └── commands/ │ │ ├── seed_data.py │ │ └── run_forecast.py │ └── migrations/ │ └── frontend/ ├── index.html # includes Bootstrap Icons CDN link ├── package.json ├── vite.config.js ├── src/ │ ├── main.jsx │ ├── App.jsx │ ├── context/ │ │ └── AuthContext.jsx │ ├── services/ │ │ └── api.js # all endpoint calls (axios instance) │ ├── components/ │ │ ├── Sidebar.jsx │ │ ├── Navbar.jsx │ │ ├─ …