Composition method of weighted dispatching rules for multi objective dynamic job scheduling

Xili Chen, Yang Lu, Tomohiro Murata

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Multiple dispatching rule combination method is applied widely to solve multiple objective scheduling problem in practice for its simplicity. It mostly combines particular dispatching rules with linear manner corresponding to multiple scheduling objectives. However, the effectiveness of this method has limited practical value for several reasons. It takes considerable effort to find the optimal weights of dispatching rules, and there is not guarantee these weights would be the right ones for a new problem with its updated system parameters. In this paper, a new heuristic method based on unified ranking measure of the job priority is proposed to deal multiple objective scheduling problem. The priority of jobs regarding to respective dispatching rules are calculated and the unified priority of the job priority is calculated by using analytic hierarchical process with adjusted weights corresponding to scheduling objectives. The simulation result demonstrates that the proposed heuristics is feasible and flexible method to solve multiple objective scheduling problem. The sensitivity analysis indicates how throughput and job weight impact on system effectiveness.

Original languageEnglish
Title of host publicationLecture Notes in Engineering and Computer Science
Pages2155-2160
Number of pages6
Volume2
Publication statusPublished - 2007
EventInternational MultiConference of Engineers and Computer Scientists 2007, IMECS 2007 - Kowloon
Duration: 2007 Mar 212007 Mar 23

Other

OtherInternational MultiConference of Engineers and Computer Scientists 2007, IMECS 2007
CityKowloon
Period07/3/2107/3/23

Fingerprint

Scheduling
Chemical analysis
Heuristic methods
Sensitivity analysis
Throughput

Keywords

  • AHP
  • Dispatching rule
  • Multi-objective
  • Priority
  • Scheduling

ASJC Scopus subject areas

  • Computer Science (miscellaneous)

Cite this

Chen, X., Lu, Y., & Murata, T. (2007). Composition method of weighted dispatching rules for multi objective dynamic job scheduling. In Lecture Notes in Engineering and Computer Science (Vol. 2, pp. 2155-2160)

Composition method of weighted dispatching rules for multi objective dynamic job scheduling. / Chen, Xili; Lu, Yang; Murata, Tomohiro.

Lecture Notes in Engineering and Computer Science. Vol. 2 2007. p. 2155-2160.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Chen, X, Lu, Y & Murata, T 2007, Composition method of weighted dispatching rules for multi objective dynamic job scheduling. in Lecture Notes in Engineering and Computer Science. vol. 2, pp. 2155-2160, International MultiConference of Engineers and Computer Scientists 2007, IMECS 2007, Kowloon, 07/3/21.
Chen X, Lu Y, Murata T. Composition method of weighted dispatching rules for multi objective dynamic job scheduling. In Lecture Notes in Engineering and Computer Science. Vol. 2. 2007. p. 2155-2160
Chen, Xili ; Lu, Yang ; Murata, Tomohiro. / Composition method of weighted dispatching rules for multi objective dynamic job scheduling. Lecture Notes in Engineering and Computer Science. Vol. 2 2007. pp. 2155-2160
@inproceedings{3512f0330a8540ddb6d0783d618c45cc,
title = "Composition method of weighted dispatching rules for multi objective dynamic job scheduling",
abstract = "Multiple dispatching rule combination method is applied widely to solve multiple objective scheduling problem in practice for its simplicity. It mostly combines particular dispatching rules with linear manner corresponding to multiple scheduling objectives. However, the effectiveness of this method has limited practical value for several reasons. It takes considerable effort to find the optimal weights of dispatching rules, and there is not guarantee these weights would be the right ones for a new problem with its updated system parameters. In this paper, a new heuristic method based on unified ranking measure of the job priority is proposed to deal multiple objective scheduling problem. The priority of jobs regarding to respective dispatching rules are calculated and the unified priority of the job priority is calculated by using analytic hierarchical process with adjusted weights corresponding to scheduling objectives. The simulation result demonstrates that the proposed heuristics is feasible and flexible method to solve multiple objective scheduling problem. The sensitivity analysis indicates how throughput and job weight impact on system effectiveness.",
keywords = "AHP, Dispatching rule, Multi-objective, Priority, Scheduling",
author = "Xili Chen and Yang Lu and Tomohiro Murata",
year = "2007",
language = "English",
isbn = "9789889867171",
volume = "2",
pages = "2155--2160",
booktitle = "Lecture Notes in Engineering and Computer Science",

}

TY - GEN

T1 - Composition method of weighted dispatching rules for multi objective dynamic job scheduling

AU - Chen, Xili

AU - Lu, Yang

AU - Murata, Tomohiro

PY - 2007

Y1 - 2007

N2 - Multiple dispatching rule combination method is applied widely to solve multiple objective scheduling problem in practice for its simplicity. It mostly combines particular dispatching rules with linear manner corresponding to multiple scheduling objectives. However, the effectiveness of this method has limited practical value for several reasons. It takes considerable effort to find the optimal weights of dispatching rules, and there is not guarantee these weights would be the right ones for a new problem with its updated system parameters. In this paper, a new heuristic method based on unified ranking measure of the job priority is proposed to deal multiple objective scheduling problem. The priority of jobs regarding to respective dispatching rules are calculated and the unified priority of the job priority is calculated by using analytic hierarchical process with adjusted weights corresponding to scheduling objectives. The simulation result demonstrates that the proposed heuristics is feasible and flexible method to solve multiple objective scheduling problem. The sensitivity analysis indicates how throughput and job weight impact on system effectiveness.

AB - Multiple dispatching rule combination method is applied widely to solve multiple objective scheduling problem in practice for its simplicity. It mostly combines particular dispatching rules with linear manner corresponding to multiple scheduling objectives. However, the effectiveness of this method has limited practical value for several reasons. It takes considerable effort to find the optimal weights of dispatching rules, and there is not guarantee these weights would be the right ones for a new problem with its updated system parameters. In this paper, a new heuristic method based on unified ranking measure of the job priority is proposed to deal multiple objective scheduling problem. The priority of jobs regarding to respective dispatching rules are calculated and the unified priority of the job priority is calculated by using analytic hierarchical process with adjusted weights corresponding to scheduling objectives. The simulation result demonstrates that the proposed heuristics is feasible and flexible method to solve multiple objective scheduling problem. The sensitivity analysis indicates how throughput and job weight impact on system effectiveness.

KW - AHP

KW - Dispatching rule

KW - Multi-objective

KW - Priority

KW - Scheduling

UR - http://www.scopus.com/inward/record.url?scp=84888358888&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84888358888&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84888358888

SN - 9789889867171

VL - 2

SP - 2155

EP - 2160

BT - Lecture Notes in Engineering and Computer Science

ER -