Efficient Heuristics for Timetabling Scheduling in Blended Learning

Abstract

Blended learning, an innovative educational approach adopted by the Saudi Electronic University, combines traditional face-to-face teaching with online methods to improve educational outcomes. However, implementing it effectively faces challenges, especially in shifting physical classes to virtual formats for faculty and students spread across distant campuses, potentially compromising the core principles of blended learning and reducing its effectiveness. Ideally, each student would attend one face-to-face and one virtual session per course weekly. Yet, operational challenges emerge due to geographical disparities between faculty and student locations, necessitating the conversion of face-to-face sessions into virtual ones. To address these challenges, this study proposes two novel heuristic strategies for scheduling timetables in blended learning contexts. The first heuristic, called Minimum Load Accumulation Heuristic (MLAH), aims to evenly distribute teaching loads among faculty members and time slots while maximizing the number of groups assigned to faculty members from the same campus. The second heuristic, called Average Load Accumulation Heuristic (ALAH), calculates the average load of all faculty members and time slots and reduces the number of iterations searching for the minimum load, as performed by MLAH. These strategies aim to minimize the conversion of face-to-face sessions to virtual, ensure fair distribution of teaching responsibilities, and maintain a balanced allocation of face-to-face and virtual classes throughout the academic year. The paper demonstrates the effectiveness of these algorithms in producing high-quality solutions comparable to those generated by CPLEX, with significantly reduced computational complexity.