Dynamic PDCA-ACO Model for Adaptive Course Scheduling in Educational Institutions

Abstract

Academic scheduling is the core work of teaching management in colleges and universities, involving the optimization of complex spatio-temporal constraints on multi-dimensional resources such as courses, teachers, classrooms, and classes. Traditional class scheduling methods face challenges such as low efficiency of large-scale combination optimization and difficulty in coping with dynamic adjustment needs, which restrict the quality and management efficiency of class scheduling. In order to solve this problem, this study combines bioheuristic algorithms and dynamic process management ideas to propose an ACO-driven PDCA dynamic adjustment model for circular academic scheduling: ACO is improved by contrast enhancement, information entropy control and random perturbation, and PDCA is implemented in stages - Plan clearly constrains and initializes ACO parameters (α∈[1.2,1.8], etc.), does deploy the scheme and responds to interference, checks for monitoring anomalies, and Act activates ACO to locally optimize to form a closed-loop optimization. In order to verify the effectiveness, an experiment was designed based on real data from universities (80 teachers, 60 classrooms) to compare traditional GA and static ACO. The results showed that the success rate of the initial class scheduling was 96.5% (90.1% super-contrast mean, and the test difference was significant). In the 6 dynamic adjustments, the conflict resolution time was reduced by 42% (to 49 minutes), the conflict course was reduced by 63%, the classroom utilization rate increased by 5.8%, and the teacher satisfaction increased by 11.3%, and the stability of the algorithm was 2-4 times that of the comparison. The model highlights the advantages of efficient handling of constraints and rapid response adjustment, providing a feasible path for intelligent educational management.