Kwame-23/SchedulAI

# SchedulAi **SchedulAi** is a hybrid algorithmic and AI-powered web application for academic timetabling and resource allocation. Designed to alleviate the administrative burden on university registries, it automatically generates conflict-free semester schedules that satisfy both hard constraints (e.g., no room or lecturer overlaps) and soft constraints (e.g., preferred teaching times). ## 🚀 Features - AI-powered timetable generation using Greedy and Genetic Algorithms - Real-time schedule editing with drag-and-drop functionality - Conflict detection and resolution with suggestions - Dynamic analytics dashboard (room utilization, lecturer workload, anomaly detection) - Export to Excel/CSV for integration with platforms like MyCamu - Multi-role access: Registry, Department Heads, Lecturers ## 🏗 Architecture SchedulAi follows a **microservices architecture**: - **Frontend**: Flask + Jinja2 + Bootstrap + Chart.js - **Backend**: Flask + Flask-SocketIO - **Database**: MySQL - **Scheduler Engine**: Greedy Heuristic & Genetic Algorithm via PyGAD - **Deployment**: Dockerized on AWS EC2 (github.com) ## 📦 Installation * 1. Clone the repo: ```bash git clone github.com cd schedulai * 2. Set up virtual environment: ```bash python -m venv venv source venv/bin/activate pip install -r requirements.txt ``` * 3. Configure `.env` with DB credentials and Flask settings. * 4. Initialize MySQL database using provided schema in `/db`. * 5. Run the app: ```bash flask run ``` ## 📊 Usage * Access the dashboard at `http` * Navigate through data input → scheduling → validation → analytics * Use the drag-and-drop matrix to manually adjust sessions * Export the final timetable to CSV ## 🧠 Algorithms * **Greedy Heuristic**: Prioritizes high-enrollment sessions and large rooms * **Genetic Algorithm**: Evolves candidate timetables using crossover/mutation * **Conflict Checker**: Detects over …