A knowledge based approach to the faculty-course assignment problem

Abstract This paper considers the problem of assigning faculty to courses at a university. Traditional operations research methods emphasizing the use of mathematical models suffer several shortcomings, including poor handling of qualitative data, undue abstraction from the problem, difficulty of problem formulation, and the combinatorial problem. This paper presents a knowledge-based system (KBS) approach to the faculty-course assignment problem that is believed to offer advantages over previous mathematical scheduling methods. A variety of criteria such as course preferences by faculty, scheduled course offerings, faculty teaching load and maximum number of days an instructor teaches per week are considered in developing the model. This KBS was applied to a department in a major university, and real-world faculty-course assignment examples were used to illustrate the model's potential. The KBS's recommendations were subsequently validated by the course coordinators. The system demonstrates that KBSs can be an effective tool for faculty-course scheduling.