Restaurant Analytics is a professional-grade, data-driven platform designed to optimize restaurant operations through predictive intelligence. By bridging the gap between historical sales data and future inventory needs, it empowers restaurateurs to minimize waste, prevent stockouts, and maximize profitability.
- AI-Driven Predictions: Utilizes Random Forest regression models to predict demand for top-selling menu items over the next 7 days.
- Context-Aware Learning: Features engineering that considers seasonality, day-of-week patterns, and weekend surges.
- Automated Recipe Mapping: Dynamically translates predicted dish demand into precise ingredient quantities.
- Restock Planning: Generates optimized procurement lists based on predicted usage and existing stock levels.
- Shelf-Life Awareness: Intelligent reordering logic that prioritizes ingredients based on perishability and lead times.
- Real-time Tracking: A unified dashboard to monitor current stock levels and ingredient health.
- Visual KPIs: Instant visibility into total revenue, transaction counts, and category performance.
- Interactive Charts: Powered by Recharts for clear, actionable data visualization.
- FastAPI: High-performance asynchronous API framework.
- Scikit-learn: Machine learning models for demand forecasting.
- Pandas/NumPy: Robust data manipulation and numerical processing.
- Joblib: Efficient model persistence and retrieval.
- React 19: Modern component-based UI.
- Vite: Lightning-fast build tool and dev server.
- Tailwind CSS: Utility-first styling for a premium, responsive design.
- Lucide React: Clean and consistent iconography.
- Recharts: Responsive charting library.
- Python 3.9+
- Node.js 18+ & npm
cd Kitchen_eda
# Create virtual environment
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
# Install dependencies
pip install -r requirements.txt
# Run the API
python app.pyThe API will be available at http://localhost:8000
cd Kitchen_eda/client
npm install
npm run devThe dashboard will be available at http://localhost:5173
Restuarant-Analytics/
├── Kitchen_eda/ # Core Application Directory
│ ├── app.py # FastAPI Entry Point
│ ├── prediction_engine.py # ML Logic & Ingredient Mapping
│ ├── requirements.txt # Python Dependencies
│ ├── dataset/ # CSV/Excel Data Sources
│ ├── models/ # Trained Joblib Models
│ └── client/ # React Frontend Project
│ ├── src/
│ │ ├── pages/ # Dashboard, Inventory, Predictions
│ │ └── components/ # Reusable UI Elements
│ └── package.json
└── README.md # Project Documentation
- Data Ingestion: Upload historical sales CSVs via the Settings page.
- Training: The engine trains individual models for top-performing items to understand their unique demand patterns.
- Prediction: The system forecasts dish demand for the upcoming week.
- Explosion: Dish forecasts are "exploded" into ingredient requirements using the Recipe Mapping master.
- Reordering: The final output is a restock plan that ensures you only buy what you'll use before it spoils.
- Support for Multi-outlet management.
- Integration with POS systems (Toast, Square).
- Deep Learning (LSTM) models for long-term seasonality.
- Automated supplier order generation (Email/API).
This project is licensed under the MIT License - see the LICENSE file for details.
Developed with ❤️ for the Culinary Industry.