Please use this identifier to cite or link to this item: http://hdl.handle.net/1959.3/148534

Download PDF (Published version) (Adobe Acrobat PDF, -1 bytes)

Title

Parallel constraint handling in a multiobjective evolutionary algorithm for the automotive deployment problem

Author(s)

Montgomery, James;  Moser, Irene

Abstract

The component deployment problem is a complex multiobjective optimisation task faced by engineers in the automotive industry. Thus far, the best-known solutions to this problem have been achieved using the NSGA-II algorithm combined with a constraint handling method based on repairing solutions that have been rendered infeasible by the genetic operators. It can reasonably be assumed that an approach that repairs solutions immediately after a change has limited coverage of the infeasible space. Exchanging solutions with other algorithms may help enhance the search space coverage. However, we observe an improvement in performance through parallelisation only after increasing the complexity of the problem.

Publication type

Conference paper

Research centre

Swinburne University of Technology. Faculty of Information and Communication Technologies

Source

Proceedings of the 6th IEEE International Conference on e-Science Workshops (e-ScienceW 2010), Brisbane, Queensland, Australia, 07-10 December 2010, pp. 104-109

Publication year

2010

FOR Code(s)

010303 Optimisation;  080108 Neural, Evolutionary and Fuzzy Computation;  0805 Distributed Computing

Keyword(s)

Automotive deployment;  Constraint handling;  Evolutionary algorithms;  Multiobjective problems;  Parallel optimisation

Publisher

IEEE

ISBN

9780769542959

Publisher URL

http://dx.doi.org/10.1109/eScienceW.2010.26

Publisher URL

http://ieeexplore.ieee.org/

Copyright

Copyright © 2010 IEEE. Published version of the paper reproduced here in accordance with the copyright policy of the publisher. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.

Full text

Peer reviewed