Artificial Intelligence-Driven University Timetabling: A Comprehensive Review of Conflict-Free Timetable Generation Techniques

Abstract

Academic timetabling remains a critical and inherently complex task within higher education institutions, formally classified as an NP-hard optimization problem. It involves allocating courses, resource persons, classrooms, and student groups under numerous hard constraints such as resource persons availability and classroom capacity, and soft constraints such as minimizing gaps between student lectures or optimizing room usage. Traditional manual or rule-based scheduling methods often struggle with these complexities, resulting in conflicts, underutilized resources, and increased administrative workload. In response, Artificial Intelligence and metaheuristic-based approaches including Simulated Annealing, Genetic Algorithms, Tabu Search, and hybrid frameworks have emerged as effective solutions for automated, conflict-free timetable generation. This review critically examines five influential studies that exemplify methodological innovations in hyper-heuristics, genetic algorithms, and hybrid metaheuristics, highlighting the transition from static heuristics to adaptive learning-based models. The study identifies key challenges in scalability, interpretability, real-time adaptability, and integration with institutional systems. The findings suggest that hybrid AI frameworks and adaptive heuristic selection mechanisms are among the most promising approaches, enabling efficient schedule generation and supporting informed administrative decision-making.