View in EDS

Balanced ranking method for constrained optimization problems using evolutionary algorithms.

Saved in:
Bibliographic Details
Title: Balanced ranking method for constrained optimization problems using evolutionary algorithms.
Authors: Rodrigues, Max de Castro1 max.castro.rodrigues@gmail.com, de Lima, Beatriz Souza Leite Pires1 beatriz@poli.ufrj.br, Guimarães, Solange1 sol@coc.ufrj.br
Source: Information Sciences. Jan2016, Vol. 327, p71-90. 20p.
Subjects: Constraint satisfaction, Evolutionary algorithms, Mathematical optimization, Search algorithms, Wilcoxon signed-rank test
Abstract: This work presents a new technique to handle constraints in the solution of optimization problems by evolutionary algorithms – the Balanced Ranking Method (BRM). In this method the fitness function is based on two rankings, for feasible and infeasible solutions respectively. The rankings are merged according to deterministic criteria that consider the status of the search process and specific properties of the population. The focus of the BRM method is to comprise a constraint-handling technique (CHT) that is not coupled to the optimization algorithm, and thus can be implemented into different algorithms. The method is compared with other well-known CHTs that follow this same uncoupled approach, all implemented into a canonical Genetic Algorithm. Two well-known suites of benchmark functions and five engineering problems are used as case studies. The performance of the different CHTs is assessed by nonparametric statistical tests, including the Sign test and the Wilcoxon Signed-Ranks test. The results indicate that the BRM presents a good performance, being reliable and efficient, while maintaining its uncoupled characteristic leading to an easy implementation and hybridization with any search algorithm. [ABSTRACT FROM AUTHOR]
Copyright of Information Sciences is the property of Elsevier B.V. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
Description
Abstract:This work presents a new technique to handle constraints in the solution of optimization problems by evolutionary algorithms – the Balanced Ranking Method (BRM). In this method the fitness function is based on two rankings, for feasible and infeasible solutions respectively. The rankings are merged according to deterministic criteria that consider the status of the search process and specific properties of the population. The focus of the BRM method is to comprise a constraint-handling technique (CHT) that is not coupled to the optimization algorithm, and thus can be implemented into different algorithms. The method is compared with other well-known CHTs that follow this same uncoupled approach, all implemented into a canonical Genetic Algorithm. Two well-known suites of benchmark functions and five engineering problems are used as case studies. The performance of the different CHTs is assessed by nonparametric statistical tests, including the Sign test and the Wilcoxon Signed-Ranks test. The results indicate that the BRM presents a good performance, being reliable and efficient, while maintaining its uncoupled characteristic leading to an easy implementation and hybridization with any search algorithm. [ABSTRACT FROM AUTHOR]
ISSN:00200255
DOI:10.1016/j.ins.2015.08.012