Published Paper Details:

Classroom and Timetable Scheduling has become an increasing part of the educational environment, as modern educational institutions have found that efficient use of classrooms and automating the timetable improves the overall academic activities and institutional productivity. The increased number of students and the diversity in the structure of courses demand a flexible and accurate timetable generation mechanism in the universities, colleges, and training institutes. This work proposes integrating the optimisation and AI-assisted models to generate timetables based on the institutional constraints. The models use the input datasets such as faculty availability, classroom capacity, program structure, student group distribution, and departmental regulations to successfully generate conflict-free timetables for the academic year. The motivation for this project comes from the challenges faced while manually or semi-automatically scheduling timetables. The issues with manual scheduling occur mainly because of the high complexity of the constraints, dynamic changes that may occur, and the lack of uniformity across different institutions. When it comes to automatic scheduling, the issues remain the same, but in addition to that, the system fails to provide a rationale for the generated schedule. The data inconsistencies and low generalizability of the timetable generation across institutions also pose a challenge. Another issue is the lack of a system that can bridge the gap between timetable generation and actual classroom management. The proposed system incorporates an AI-assisted optimisation with explainability and dynamic constraint handling. The proposed design also proposes a classroom-allocation system based on the utilisation of the resource, academic load balancing, and scheduling frequency.