Adapting Genetic Algorithms for Combinatorial Optimization Problems in Dynamic Environments

Adapting Genetic Algorithms for Combinatorial Optimization Problems in Dynamic Environments

by Abdunnaser Younes

Type Book
Rate 0.0 /5

The island based model proves to be effective under different dynamics. Although statistical analysis suggest that these benefits are not significant under some problem dynamics, this model can be more rewarding if several processors are employed. With each island allocated to a different processor, the per processor computational costs are reduced significantly. The problem of parameter tuning is aggravated with dynamic environments, as a result of the increased problem complexity and the incr…

Read More ↓
Details
Files
Reviews
Download

Title Statements

  • Full title :Adapting Genetic Algorithms for Combinatorial Optimization Problems in Dynamic Environments.

Summary

The island based model proves to be effective under different dynamics. Although statistical analysis suggest that these benefits are not significant under some problem dynamics, this model can be more rewarding if several processors are employed. With each island allocated to a different processor, the per processor computational costs are reduced significantly. The problem of parameter tuning is aggravated with dynamic environments, as a result of the increased problem complexity and the increased number of algorithm parameters; however, by using diversity to control the EA parameters, the models developed in this article had significantly reduced tuning efforts. There are several ways in which the developed models can be applied and improved:? The effectiveness of the developed methods on the TSP and FMS problems encourages their application to other problems, such as intelligent transportation systems, engine parameter control, scheduling of airline maintenance, and dynamic network routing.? Diversity controlled models can use operator-specific diversity measures so that each operator is controlled by its respective diversity measure, i.e., based on algorithmic distance. Future work that is worth exploring involves using adaptive limits of diversity for the models presented in this article.

Read More ↓

Edition Statements

  • Publication Date:2008.

Publication Statements

The Physical Object

  • Description:1 online resource

Edition Identifiers

  • ISBN:9789537619114, 9537619117
  • oclc:ocn884023742

Subjects

Open Access Collection

Writers

Abdunnaser Younes Otman Basir Paul Calamai Shawki Areibi

Rate & Review

0.0/5
(0 people Voted)

No reviews yet. Be the first to rate this work.

Files & Downloads

  • Advances in evolutionary algorithms -- (.djvu)

    DJVU · 6.8 MB · updated
    Login →

  • Advances in evolutionary algorithms -- (.djvu)

    DJVU · 6.0 MB · updated
    Login →

  • Advances in Evolutionary Algorithms -- (.pdf)

    PDF · 46.9 MB · updated
    Login →

  • Advances in evolutionary algorithms -- (.djvu)

    DJVU · 6.4 MB · updated
    Login →
Log in to request a copy