Paper Info
Title
Evolutionary many-objective optimization based on linear assignment problem transformations.
Abstract
The selection mechanisms that are most commonly adopted by multi-objective evolutionary algorithms (MOEAs) are based on Pareto optimality. However, recent studies have provided theoretical and experimental evidence regarding the unsuitability of Pareto-based selection mechanisms when dealing with problems having four or more objectives. In this paper, we propose a novel MOEA designed for solving many-objective optimization problems. The selection mechanism of our approach is based on the transformation of a multi-objective optimization problem into a linear assignment problem, which is solved by the Kuhn–Munkres’ (Hungarian) algorithm. Our proposed approach is compared with respect to three state-of-the-art MOEAs, designed for solving many-objective optimization problems (i.e., problems having four or more objectives), adopting standard test problems and performance indicators taken from the specialized literature. Since one of our main aims was to analyze the scalability of our proposed approach, its validation was performed adopting test problems having from two to nine objective functions. Our preliminary experimental results indicate that our proposal is very competitive with respect to all the other MOEAs compared, obtaining the best results in several of the test problems adopted, but at a significantly lower computational cost.
Year
DOI
Venue
2018
10.1007/s00500-018-3164-3
Soft Comput.
Keywords
Field
DocType
Multi-objective optimization, Many-objective optimization, Evolutionary computation
Mathematical optimization,Performance indicator,Evolutionary algorithm,Computer science,Evolutionary computation,Theoretical computer science,Multi-objective optimization,Assignment problem,Optimization problem,Pareto principle,Scalability
Journal
Volume
Issue
ISSN
22
16
1432-7643
Citations 
PageRank 
References 
0
0.34
21
Authors
3
Name
Order
Citations
PageRank
Luis Miguel Antonio1583.92
José A. Molinet Berenguer220.71
C. A. Coello Coello35799427.99