Optimizing University Course Timetabling Using Constraint Satisfaction Models

Abstract

The giving out of the course timetable is a problem found at ever increasing complexity with the growth of universities.
This problem demands the prompt allocation of courses along with instructors, rooms and times while satisfying a
number of academic and logistical criteria. In this paper we propose a systematic approach using CSP which helps
solve the problem efficiently. Our measures consist of hard constraints such as no instructor or student conflict, room
constraints, facility restrictions, and other requirements for courses with soft constraints like preferred slots, balance
course distribution, etc. A system prototype was built in Python and tested with real world data from a mid-tier
university. Results when compared with traditional manual scheduling and other relative methods like simple
algorithm showed that use of CSP design framework enhances timetable user satisfaction alongside feasibility. A bar
graph and a performance table enhance the fact positive improvements were made in conflict resolution alongside
improvement of scheduling efficiency. The study proves that model designs are highly scalable, adaptable and
transparent proving great value to academic institutions esp. in terms of university timetabling. With the use of
machine learning focused on constraint prediction and optimization, future work has the potential to broaden system
boundaries further.