Global and well-distributed Pareto frontier by modified normalized normal constraint methods for bicriterion problems

Important efforts have been made in the last years to develop methods for the construction of Pareto frontiers that guarantee uniform distribution and that exclude the non-Pareto and local Pareto points. Nevertheless, these methods are susceptible of improvement or modifications to reach the same le...

Full description

Saved in:
Bibliographic Details
Published in:Structural and multidisciplinary optimization Vol. 34; no. 3; pp. 197 - 209
Main Authors: Martínez, M., Sanchis, J., Blasco, X.
Format: Journal Article
Language:English
Published: Heidelberg Springer Nature B.V 01.09.2007
Subjects:
Computational efficiency
Construction costs
Construction methods
Dependence
Genetic algorithms
Optimization
Pareto optimum
ISSN:1615-147X, 1615-1488
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Important efforts have been made in the last years to develop methods for the construction of Pareto frontiers that guarantee uniform distribution and that exclude the non-Pareto and local Pareto points. Nevertheless, these methods are susceptible of improvement or modifications to reach the same level of results more efficiently. This paper presents some of these possibilities, based on two types of techniques: those based on nonlinear optimization and those based on genetic algorithms. The first provides appropriate solutions at reasonable computational cost though they are highly dependent on the initial points and on the presence or absence of local minima. The second technique does not present such dependence although computational cost is higher. Since the construction of the Pareto frontier is usually off-line, that computational cost is not a restrictive factor. Goodness of the improvements proposed in the paper are shown with two bicriterion examples.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ObjectType-Article-2
ObjectType-Feature-1
content type line 23
ISSN:1615-147X
1615-1488
DOI:10.1007/s00158-006-0071-5