A hybrid model for explainable university timetable optimization using machine learning and expert knowledge
2026, vol.18 , no.4, pp. 67-78
Article [2026-04-07]
Creating a university timetable is a complex multi-criteria problem that requires the processing of heterogeneous data, expert rules, resource constraints, and user preferences. This article proposes a hybrid knowledge-based decision-support model for explainable optimization of university timetables by integrating machine learning, expert knowledge, and mechanisms for explainable artificial intelligence. The model uses historical data, expert rules, and satisfaction metrics to evaluate and select the most appropriate timetable variant. A multi-criteria objective function has been developed that combines student and lecturer satisfaction, expert evaluation, and a predicted score from a machine learning model. An explainable artificial intelligence component for interpreting the results is included, as well as a closed-loop self-learning mechanism. The proposed approach lays the foundation for a transparent and adaptive information technology framework for decision-makers in optimizing timetables.
expert knowledge, explainable artificial intelligence, machine learning, multi-criteria optimization, university timetabling
https://doi.org/10.59035/NGWJ8757
Fatme Rashidova, Aldeniz Rashidov. A hybrid model for explainable university timetable optimization using machine learning and expert knowledge. International Journal on Information Technologies and Security, vol.18 , no.4, 2026, pp. 67-78. https://doi.org/10.59035/NGWJ8757