Intelligent Scheduling Systems for Practice-Based Higher Education Programs

Abstract

This study develops an intelligent scheduling system for practice-based higher education programs, focusing on how AI-assisted optimization can improve the allocation of instructors, specialized rooms, equipment, and repeated practice sessions. The study is designed to collect scheduling and resource-use data from approximately 25 practice-based programs across 8 universities over two academic semesters, including around 18,000 course-booking records, 6,500 room-use logs, 4,200 instructor availability records, 3,800 equipment reservation records, and 12,000 student practice requests. Key variables include room occupancy rate, instructor workload balance, equipment utilization rate, booking conflict frequency, student waiting time, schedule adjustment frequency, course completion continuity, and learner satisfaction. The study applies mixed-integer linear programming, genetic algorithms, constraint satisfaction modeling, and reinforcement learning-based scheduling simulation to compare traditional manual scheduling with intelligent scheduling scenarios. Model performance is evaluated through resource utilization rate, conflict reduction rate, average waiting time, instructor workload variance, schedule stability, and student access equity. The innovation of this study lies in moving practice-based program management from experience-based scheduling to data-driven resource coordination, providing a measurable optimization model for institutions where learning quality depends heavily on limited spaces, instructors, and equipment.