By Tasbeeh Takrori
This is an AI-powered bakery management prototype built using Streamlit and SQLite, designed to support small bakeries in making smarter daily decisions. It includes features like AI demand forecasting, stock tracking, and raw material consumption estimation.
⚠️ Note: The system uses fake/generated data for testing and demo purposes.
- Add, edit, and delete products and their ingredients
- Manage raw materials and monitor stock levels
- Real-time alerts for low stock
- Predict product demand using Facebook Prophet
- Demand is affected by:
- 📍 Weather conditions (temperature, humidity, wind)
- 🚧 Checkpoint conditions (5 dynamic barriers)
- Daily and 7-day forecast per product
- Visualized predictions with interactive charts
- Estimate the expected daily demand for each raw material
- Show "Days to depletion" per ingredient based on product predictions
- Color-coded risk indicators (e.g., red for < 3 days)
- Add new customer orders
- Automatically:
- Reduce product stock
- Reduce raw materials based on product recipe
- Prevent order placement if stock is insufficient
- Product demand trends over time
- Daily product orders (last 7 days)
- Raw material usage and remaining stock
- Compare predicted vs. actual demand
- View MAE, MAPE, and Accuracy per product
- Interactive error tables and accuracy progress bars
- Built-in AI-based text generation assistant
- Helps bakery owners write promotional content (e.g., social media posts, product descriptions)
- Generates Arabic content tailored to bakery context
- Clone the repo
git clone https://github.com/TasbeehTakrori/Smart-Bakery-Management-System.git
cd Smart-Bakery-Management-System- Install requirements
pip install -r requirements.txt- Run the app
streamlit run main.py| Technology | Description |
|---|---|
| Streamlit | For building the interactive user interface |
| SQLite | Lightweight embedded database for storage |
| Prophet | Time series forecasting model for demand prediction |
| Pandas | Data manipulation and analysis |
| Plotly | Interactive charts and visualizations |
| SQLAlchemy | ORM for database interactions |
| Joblib | Saving and loading trained AI models |
| Thinkstack (GPT-based) | Used to build the AI-powered chatbot, leveraging OpenAI's language models (ChatGPT) to generate bakery-specific marketing content in Arabic |
- All AI models are trained on generated/fake data for demonstration purposes only.
- Demand forecasting considers external factors like weather conditions and checkpoint delays.
- Raw materials page includes "days to depletion" insights to assist with inventory planning.
- The AI chatbot is integrated using Thinkstack, built on top of OpenAI's GPT technology, to help bakery owners write better Arabic promotional content.
- This system is a prototype developed for educational and demo use.
Tasbeeh Takrori
Graduate Student in Artificial Intelligence
An-Najah National University