DSpace at Saint Louis University >
Different School Researches >
School of Computing and Information Sciences >

Please use this identifier to cite or link to this item: http://hdl.handle.net/123456789/33

Title: An automated class scheduler for SLU : a genetic algorithm approach
Authors: Berto, Calire B.
Keywords: Scheduling
Timetabling
Genetic Algorithms
Issue Date: Apr-2000
Abstract: Summary: This study is all about solving the problem of class time tabling in local setting. The study is founded on the premise that time tabling is such an arduous task, which would actually be eased by computer automation. In so doing, genetic algorithm was chosen from other approaches because it offers more advantages compared to the other two candidate approaches, namely simulated annealing and tabu search. In order to prove the effectiveness of the GA approach to the local time tabling problem, a rapid prototype was developed. 8.2 Findings: The study presented the prototype based on a genetic algorithm approach. Based on several data comparisons, it was shown that the prototype was able to improve the fitness of its initially generated population while maintaining the feasibility of the schedules. It was also shown that the generation of class schedules takes lesser effort and lesser time to accomplish as compared with the manual scheduling. 8.3 Conclusion: The GA based prototype was successful in providing a solution to Saint Louis University’s class timetabling problem. Through the automation, the long and arduous task of building a class schedule was considerably trimmed and reduced. The prototype also gave evidence to the effectiveness of the GA approach in solving Saint Louis University’s class timetabling problem. 8.4 Recommendations: With the complexities surrounding class timetabling and genetic algorithms, there is much room for improvement and investigation. Thus, the prototype may be extended from different angles. It could be extended to include more scheduling constraints or a comparison can be done on the results when different parameters are used.
Description: Thesis (M.S.I.T.)--Saint Louis University, Baguio City, 2000.
URI: http://hdl.handle.net/123456789/33
Appears in Collections:School of Computing and Information Sciences

Files in This Item:

File Description SizeFormat
AN AUTOMATED CLASS SCHEDULER FOR SLU_A GENETIC ALGORITHM APPROACH.pdf19.71 kBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

 

Valid XHTML 1.0! DSpace Software Copyright © 2002-2008  The DSpace Foundation - Feedback