Minimizing Distance Between Classes in School Timetabling Problem Using a Mixed-Integer Programming Model

Abstract

A school timetabling problem is a challenging optimisation problem. A set of subjects need to be assigned into a class schedule while these subjects are related to both students and lecturers. The number of students required to attend a class leads to a room's capacity requirement. Moreover, the time to assign any subject depends on lecturers' availability. This paper aims to minimize the distance between classes for students as they have to move to another room. A mixed-integer programming model is developed. The model is also used to investigate the difference in total distance between classes when the subject is taught in a single session and multiple sessions. Computational results on 60 generated instances indicate that arranging a class schedule in multiple sessions can lead the increase of total distance from arranging in single session as high as 1,868%.