DocumentCode :
2344767
Title :
The Effects of Room Slot Address (RSA) Selection Technique in a Modified PSO Algorithm to Solve Class Scheduling Problems
Author :
Aziz, Mohd Azri Abdul ; Taib, Mohd Nasir ; Hussin, Naimah Mohd
Author_Institution :
Fac. of Electr. Eng., Univ. Teknol. MARA, Shah Alam, Malaysia
fYear :
2010
fDate :
28-30 Sept. 2010
Firstpage :
143
Lastpage :
148
Abstract :
The selection of slot and room in the process of improving timetable solution plays an important role in ensuring a good quality of timetable can be produced in a shorter time. A good selection strategy or technique will be able to remove the unnecessary repeated selection process due to the rejected assignments. This paper study the Room Slot Address (RSA) selection technique with three variations, namely Random RSA selection, Earliest RSA selection and Semi-Random RSA selection techniques. The role of RSA selection technique is to select an RSA before it can be assigned to an event selected in the earlier process. Five sets of data from International Timetabling Competition were used in the experiment. The experimental results shows that Semi-Random RSA selection technique used the shortest execution time while random RSA selection technique was found to be using the longest execution time (about three times longer). In terms of percentage penalty reduction (%PR), Semi-Random RSA selection technique outperformed the other two techniques by producing the highest %PR of 3 out of 5 datasets and therefore can be concluded as the best RSA selection technique to be used in the proposed Modified PSO algorithm in solving class scheduling problems.
Keywords :
educational institutions; particle swarm optimisation; scheduling; International Timetabling Competition; class scheduling problems; earliest RSA selection; modified PSO algorithm; room slot address selection technique; semi-random RSA selection; timetable solution; Particle Swarm Optimization; Timetabling; scheduling;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computational Intelligence, Modelling and Simulation (CIMSiM), 2010 Second International Conference on
Conference_Location :
Bali
Print_ISBN :
978-1-4244-8652-6
Electronic_ISBN :
978-0-7695-4262-1
Type :
conf
DOI :
10.1109/CIMSiM.2010.35
Filename :
5701836
Link To Document :
بازگشت