Home
Resource
Multi-objective optimization algorithms considering objective preferences and solution clusters
PDF
PDF format is widely accepted and good for printing.
Plug-in required
PDF-1(3.47 MB)
Citation & Export
View Usage Statistics
Staff View

Citation & Export
Hide

Simple citation

Taboada, Heidi. Multi-objective optimization algorithms considering objective preferences and solution clusters. Retrieved from https://doi.org/doi:10.7282/T3W66M56

Export

Click here for information about Citation Management Tools at Rutgers.

Statistics
Hide

Description

Uniform TitleMulti-objective optimization algorithms considering objective preferences and solution clusters
NameTaboada, Heidi (author); Coit, David (chair); Albin, Susan (internal member); Luxhøj, James (internal member); Chaovalitwongse, Wanpracha (internal member); Boile, Maria (outside member); Rutgers University; Graduate School - New Brunswick
Date Created2007
Other Date2007-10 (degree)
SubjectIndustrial and Systems Engineering, Mathematical optimization, Multiple criteria decision making, Evolutionary computation
Extentxx, 226 pages
DescriptionThis thesis presents the development of new methods for the solution of multiple objective problems. One of the main contributions of this thesis is that it presents new approaches that provide a balance between the determination of single solutions and a set of Pareto-optimal solutions. Existing solution methodologies for multiple objective problems can generally be categorized as single solution methods or Pareto optimality methods. However, for many problems and decision-makers, a balanced approach is more appropriate, and this thesis provides new approaches to meet those needs. Other main contributions are that several novel multi-objective evolutionary algorithms are presented, which offer distinct advantages compared to existing algorithms.
Two different new approaches are introduced which can efficiently determine an attractive Pareto set or organize and reduce the size of the Pareto-optimal set. This makes it easier for the decision-maker to comparatively analyze a smaller set of solutions, and finally, select the most desirable one for system implementation.
In the first approach, the developed algorithm has the capability to automatically identify an optimal number of clusters in the Pareto-optimal set and provide the decision-maker with representative solutions of each cluster. The second approach is a method that yields efficient results for any user who can prioritize the objective functions. In this method, the objective functions are ranked ordinally based on their importance to the decision-maker, and a reduced Pareto set is determined based on randomly generated weight sets, reflecting the decision-maker preferences.
Different new multiple objective evolutionary algorithms (MOEAs) are designed as the result of this research and they are described and tested. New ideas have been incorporated into these MOEAs to provide the research community with new alternatives. One of the developed MOEAs is MoPriGA, a multi-objective prioritized genetic algorithm. MoPriGA incorporates the knowledge of the decision-maker objective function preferences directly within the evolutionary algorithm. The idea behind this algorithm is to more intensely focus on the region of the Pareto set of interest to the decision-maker.
NotePh.D.
NoteIncludes bibliographical references (p. 211-224).
Genretheses, ETD doctoral
Persistent URLhttps://doi.org/doi:10.7282/T3W66M56
LanguageEnglish
CollectionGraduate School - New Brunswick Electronic Theses and Dissertations
Organization NameRutgers, The State University of New Jersey
RightsThe author owns the copyright to this work.
Version 8.5.5
Rutgers University Libraries - Copyright ©2024