Artikel

A novel hybrid multi-objective metamodel-based evolutionary optimization algorithm

Optimization via Simulation (OvS) is an useful optimization tool to find a solution to an optimization problem that is difficult to model analytically. OvS consists in evaluating potential solutions through simulation executions; however, its high computational cost is a factor that can make its implementation infeasible. This issue also occurs in multi-objective problems, which tend to be expensive to solve. In this work, we present a new hybrid multi-objective OvS algorithm, which uses Kriging-type metamodels to estimate the simulations results and a multi-objective evolutionary algorithm to manage the optimization process. Our proposal succeeds in reducing the computational cost significantly without affecting the quality of the results obtained. The evolutionary part of the hybrid algorithm is based on the popular NSGA-II. The hybrid method is compared to the canonical NSGA-II and other hybrid approaches, showing a good performance not only in the quality of the solutions but also as computational cost saving.

Sprache
Englisch

Erschienen in
Journal: Operations Research Perspectives ; ISSN: 2214-7160 ; Volume: 6 ; Year: 2019 ; Pages: 1-14 ; Amsterdam: Elsevier

Klassifikation
Wirtschaft
Thema
Optimization via simulation
Metamodel
Multi-Objective optimization
Kriging
NSGA-II

Ereignis
Geistige Schöpfung
(wer)
Baquela, Enrique Gabriel
Olivera, Ana Carolina
Ereignis
Veröffentlichung
(wer)
Elsevier
(wo)
Amsterdam
(wann)
2019

DOI
doi:10.1016/j.orp.2019.100098
Handle
Letzte Aktualisierung
10.03.2025, 11:45 MEZ

Datenpartner

Dieses Objekt wird bereitgestellt von:
ZBW - Deutsche Zentralbibliothek für Wirtschaftswissenschaften - Leibniz-Informationszentrum Wirtschaft. Bei Fragen zum Objekt wenden Sie sich bitte an den Datenpartner.

Objekttyp

  • Artikel

Beteiligte

  • Baquela, Enrique Gabriel
  • Olivera, Ana Carolina
  • Elsevier

Entstanden

  • 2019

Ähnliche Objekte (12)