Addressing Examination Timetabling Problem Using a Partial Exams Approach in Constructive and Improvement

The paper investigates a partial exam assignment approach for solving the examination timetabling problem. Current approaches involve scheduling all of the exams into time slots and rooms (i.e., produce an initial solution) and then continuing by improving the initial solution in a predetermined num...

Full description

Bibliographic Details
Main Authors: Ashis Kumar Mandal, M. N. M. Kahar, Graham Kendall
Format: Article
Language:English
Published: MDPI AG 2020-05-01
Series:Computation
Subjects:
Online Access:https://www.mdpi.com/2079-3197/8/2/46
id doaj-a9eea8a2d9f64ee58d529d968542624d
record_format Article
spelling doaj-a9eea8a2d9f64ee58d529d968542624d2020-11-25T02:15:29ZengMDPI AGComputation2079-31972020-05-018464610.3390/computation8020046Addressing Examination Timetabling Problem Using a Partial Exams Approach in Constructive and ImprovementAshis Kumar Mandal0M. N. M. Kahar1Graham Kendall2Graduate School of Software & Information Science, Iwate Prefectural University, Iwate 020-0693, JapanFaculty of Computer Systems & Software Engineering, Universiti Malaysia Pahang, Kuantan 25000, Pahang, MalaysiaUniversity of Nottingham Malaysia Campus, Semenyih, Selangor 43500, MalaysiaThe paper investigates a partial exam assignment approach for solving the examination timetabling problem. Current approaches involve scheduling all of the exams into time slots and rooms (i.e., produce an initial solution) and then continuing by improving the initial solution in a predetermined number of iterations. We propose a modification of this process that schedules partially selected exams into time slots and rooms followed by improving the solution vector of partial exams. The process then continues with the next batch of exams until all exams are scheduled. The partial exam assignment approach utilises partial graph heuristic orderings with a modified great deluge algorithm (PGH-mGD). The PGH-mGD approach is tested on two benchmark datasets, a capacitated examination dataset from the 2nd international timetable competition (ITC2007) and an un-capacitated Toronto examination dataset. Experimental results show that PGH-mGD is able to produce quality solutions that are competitive with those of the previous approaches reported in the scientific literature.https://www.mdpi.com/2079-3197/8/2/46examination timetabling problemgraph heuristic orderingsgreat deluge algorithmmeta-heuristics
collection DOAJ
language English
format Article
sources DOAJ
author Ashis Kumar Mandal
M. N. M. Kahar
Graham Kendall
spellingShingle Ashis Kumar Mandal
M. N. M. Kahar
Graham Kendall
Addressing Examination Timetabling Problem Using a Partial Exams Approach in Constructive and Improvement
Computation
examination timetabling problem
graph heuristic orderings
great deluge algorithm
meta-heuristics
author_facet Ashis Kumar Mandal
M. N. M. Kahar
Graham Kendall
author_sort Ashis Kumar Mandal
title Addressing Examination Timetabling Problem Using a Partial Exams Approach in Constructive and Improvement
title_short Addressing Examination Timetabling Problem Using a Partial Exams Approach in Constructive and Improvement
title_full Addressing Examination Timetabling Problem Using a Partial Exams Approach in Constructive and Improvement
title_fullStr Addressing Examination Timetabling Problem Using a Partial Exams Approach in Constructive and Improvement
title_full_unstemmed Addressing Examination Timetabling Problem Using a Partial Exams Approach in Constructive and Improvement
title_sort addressing examination timetabling problem using a partial exams approach in constructive and improvement
publisher MDPI AG
series Computation
issn 2079-3197
publishDate 2020-05-01
description The paper investigates a partial exam assignment approach for solving the examination timetabling problem. Current approaches involve scheduling all of the exams into time slots and rooms (i.e., produce an initial solution) and then continuing by improving the initial solution in a predetermined number of iterations. We propose a modification of this process that schedules partially selected exams into time slots and rooms followed by improving the solution vector of partial exams. The process then continues with the next batch of exams until all exams are scheduled. The partial exam assignment approach utilises partial graph heuristic orderings with a modified great deluge algorithm (PGH-mGD). The PGH-mGD approach is tested on two benchmark datasets, a capacitated examination dataset from the 2nd international timetable competition (ITC2007) and an un-capacitated Toronto examination dataset. Experimental results show that PGH-mGD is able to produce quality solutions that are competitive with those of the previous approaches reported in the scientific literature.
topic examination timetabling problem
graph heuristic orderings
great deluge algorithm
meta-heuristics
url https://www.mdpi.com/2079-3197/8/2/46
work_keys_str_mv AT ashiskumarmandal addressingexaminationtimetablingproblemusingapartialexamsapproachinconstructiveandimprovement
AT mnmkahar addressingexaminationtimetablingproblemusingapartialexamsapproachinconstructiveandimprovement
AT grahamkendall addressingexaminationtimetablingproblemusingapartialexamsapproachinconstructiveandimprovement
_version_ 1724895977643442176