Fuzzy- logic multi-agent system for solving course time tabling problems

Tan, Li June (2020) Fuzzy- logic multi-agent system for solving course time tabling problems. Masters thesis, Universiti Malaysia Sabah.

	Text 24 PAGES.pdf Download (956kB)
	Text FULLTEXT.pdf Restricted to Registered users only Download (11MB)

Abstract

Course timetabling problems are defined to assign a set of courses in suitable timeslot and room subject to a set of constraints. This research aims to solve the course timetabling problems of Universiti Malaysia Sabah Labuan International Campus (UMSLIC). In UMSLIC, course timetable is manually prepared by timetabling officer. The timetabling process is cumbersome as it involves different kinds of elements such as students, lecturers, courses, timeslots and rooms. Therefore, it is very difficult for timetabling officer to produce a feasible course timetable manually. Usually, they will go through several times of drafts and discussions to make sure the final timetable is ready-to-use. However, the outcomes are often infeasible timetables. Besides, timetabling officer does not take those students who retake subject into consideration during the timetabling process. Once the timetable is officially published, those students need to deal with the lecturer personally. This research presents a framework which integrates fuzzy logic and multi-agent system to generate feasible course timetable. The main algorithm used in the proposed framework is Sequential Constructive Algorithm. The concept of sequential constructive algorithm is to schedule a set of events based on the heuristic orderings. There are different types of heuristic orderings and this research only uses three of them: Largest Degree (LD), Largest Enrolment (LE), and Saturation Degree (SD). Fuzzy Logic is applied to combine multiple heuristic orderings and generate three different fuzzy approaches: Fuzzy LDLE, Fuzzy SOLE, and Fuzzy SOLD. The fuzzy approaches are used to generate initial feasible solutions and further improve the quality of the solutions. The performance between the fuzzy approaches are compared to investigate which of them solves UMSLIC course timetabling problems well. After that, develop the proposed framework with integrating the fuzzy approaches with multi-agent system. The main objective of the proposed framework is to investigate how the communication between fuzzy approaches generates a high quality and feasible timetable for UMSUC. Besides, the framework is also used to conduct experiments with benchmark datasets to investigate the applicability of the framework. The experimental results show that the proposed framework produced comparable results to the literature.

Item Type:	Thesis (Masters)
Keyword:	Course timetabling, Fuzzy logic, Multi-agent system, Sequential Constructive Algorithm, Heuristic orderings
Subjects:	Q Science > QA Mathematics > QA1-939 Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science > QA76.75-76.765 Computer software
Department:	FACULTY > Faculty of Computing and Informatics
Depositing User:	DG MASNIAH AHMAD -
Date Deposited:	17 Jan 2025 14:19
Last Modified:	17 Jan 2025 14:19
URI:	https://eprints.ums.edu.my/id/eprint/42630

Actions (login required)

View Item