Advertisement

Introduction

  • Frumen Olivas
  • Fevrier Valdez
  • Oscar Castillo
  • Patricia Melin
Chapter
Part of the SpringerBriefs in Applied Sciences and Technology book series (BRIEFSAPPLSCIENCES)

Abstract

Optimization is a process where an algorithm creates several variations of solutions to a problem and in an intelligent way can improve these solutions iteratively, using meta-heuristics based on nature like evolution of species, the way in which ants find their food from their nest, or the way in which flock of birds fly together, more known as collective intelligence.

Optimization is a process where an algorithm creates several variations of solutions to a problem and in an intelligent way can improve these solutions iteratively, using meta-heuristics based on nature like evolution of species, the way in which ants find their food from their nest, or the way in which flock of birds fly together, more known as collective intelligence.

Bio-inspired optimization algorithms can be applied to a wide variety of problems but lack the ability to change dynamically their parameters to self-adapt into different problems; in this book, we present a methodology for parameter adaptation using fuzzy logic.

The proposed methodology can be applied to any optimization method that meets the specification criteria from the general procedure of the proposed methodology. Also, another metrics can be applied for more specific optimization methods.

There is a huge problem in setting the correct parameters for an optimization method when it is applied to a new problem; we can just use the recommended parameters for most common problems but these are not always the best parameters and depending on the problem can be the worst; this is why there are methodologies for parameter adaptation using mathematical methods, random parameters, and lastly fuzzy logic.

The use of fuzzy logic brings the tools needed to model a complex problem with easy to understand concepts like membership functions, for small, medium, or high values of a variable from the problem and use if-then rules to create complex knowledge about the problem in an easy way.

With a dynamic parameter adaptation, an optimization algorithm can obtain better quality results when compared with its original counterpart, also it can improve the diversity of solutions and the convergence of the algorithm, just by controlling one or more parameters dynamically over the iteration process.

Copyright information

© The Author(s) 2018

Authors and Affiliations

  • Frumen Olivas
    • 1
  • Fevrier Valdez
    • 1
  • Oscar Castillo
    • 1
  • Patricia Melin
    • 1
  1. 1.Division of Graduate StudiesTijuana Institute of TechnologyTijuanaMexico

Personalised recommendations