Web-Based Scheduling System using Genetic Algorithm with Facial Recognition

Aim/Background This study aims to develop a web-based scheduling system that integrates a genetic algorithm for timetable optimization and facial recognition for secure authentication. The project addresses the inefficiencies and security limitations of traditional manual scheduling in educational institutions. Methods The system was designed and implemented using Agile Software Development methodology, enabling iterative feedback and improvements. Genetic algorithms were applied to automatically generate conflict-free schedules based on user-defined constraints such as teacher availability and classroom capacity. Facial recognition, powered by OpenCV and TensorFlow, was integrated to ensure biometric-based secure login access for users. User experience and system performance were evaluated using functionality, accuracy, usability, reliability, and security metrics. Results The system achieved a significant reduction in manual scheduling time (by 80%) and consistently generated conflict-free timetables. User evaluation yielded high scores across all performance categories, with usability and security rated highest (Mean = 4.00). Facial recognition accuracy during system testing reached 97.5%. Conclusion The integration of genetic algorithms and facial recognition in a web-based scheduling system demonstrated improvements in efficiency, accuracy, and data security. Future enhancements may include improving facial recognition in low-light conditions and expanding system compatibility for mobile devices.