King's College London

Research portal

Integrated particle swarm and evolutionary algorithm approaches to the quadratic assignment problem

Research output: Chapter in Book/Report/Conference proceedingConference paperpeer-review

Ayah M. Helal, Enas Jawdat, Islam Elnabarawy, Donald C. Wunsch, Ashraf M. Abdelbar

Original languageEnglish
Title of host publication2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1-8
Number of pages8
ISBN (Electronic)9781538627259
DOIs
Published2 Feb 2018
Event2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017 - Honolulu, United States
Duration: 27 Nov 20171 Dec 2017

Publication series

Name2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017 - Proceedings
Volume2018-January

Conference

Conference2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017
Country/TerritoryUnited States
CityHonolulu
Period27/11/20171/12/2017

King's Authors

Abstract

This paper introduces three integrated hybrid approaches that apply a combination of Hierarchical Particle Swarm Optimization (HPSO) and Evolutionary Algorithms (EA) to the Quadratic Assignment Problem (QAP). The approaches maintain a single population. In the first approach, Alternating HPSO-EA (AHE), the population alternates between applying HPSO and EA in successive generations. In the second, more integrated approach, Integrated HPSO-EA (IHE), each population element chooses to apply one of the two algorithms in each generation with some probability. An element applying HPSO in a given generation can be influenced by an element applying EA in that generation, and vice versa. Thus, within the same generation, some elements act as HPSO particles and others as EA population members, and yet the entire population still cooperates. In the third approach, we present a Social Evolutionary Algorithm (SEA), in which the population applies EA, and each population element can choose to apply the PSO-style social mutation operator in each generation with some probability. The three approaches are compared to HPSO and EA using 31 instances of varying size from the QAP instance library.

View graph of relations

© 2020 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454