A Study on Pre-Semester Scheduling Disruption Recovery in Higher Education

The expansion of higher education has brought scheduling disruptions to the forefront for universities and colleges. In contrast to previous studies focused solely on minimizing disruptions, this research introduces an optimization model that minimizes disruptions while maximizing teacher and student satisfaction. To effectively address this model, a reinforcement learning-enhanced genetic algorithm is employed to enhance optimization capabilities. Simulation experiments reveal that the approach proposed in this study generates scheduling solutions that significantly enhance student satisfaction, thereby improving efficiency and satisfaction during scheduling disruption recovery. This study not only formulates an optimization model but also refines a genetic algorithm, including encoding methods, initial population generation, and fitness function design. Additionally, a solution algorithm is developed and validated through simulations, offering fresh insights into enhancing the quality of class scheduling disruption recovery.