An evolutionary approach to the solution of multi-objective min-max problems in evidence-based robust optimization

Simone Alicino, Massimiliano Vasile

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

4 Citations (Scopus)

Abstract

This paper presents an evolutionary approach to solve the multi-objective min-max problem (MOMMP) that derives from the maximization of the Belief in robust design optimization. In evidence-based robust optimization, the solutions that minimize the design budgets are robust under epistemic uncertainty if they maximize the Belief in the realization of the value of the design budgets. Thus robust solutions are found by minimizing, with respect to the design variables, the global maximum with respect to the uncertain variables. This paper presents an algorithm to solve MOMMP, and a computational cost reduction technique based on Kriging metamodels. The results show that the algorithm is able to accurately approximate the Pareto front for a MOMMP at a fraction of the computational cost of an exact calculation.


LanguageEnglish
Title of host publicationProceedings of the 2014 IEEE Congress on Evolutionary Computation, CEC 2014
Pages1179-1186
Number of pages8
DOIs
Publication statusPublished - 6 Jul 2014
Event2014 IEEE Congress on Evolutionary Computation, CEC 2014 - Beijing, United Kingdom
Duration: 6 Jul 201411 Jul 2014

Conference

Conference2014 IEEE Congress on Evolutionary Computation, CEC 2014
CountryUnited Kingdom
CityBeijing
Period6/07/1411/07/14

Fingerprint

Min-max Problem
Robust Optimization
Computational Cost
Epistemic Uncertainty
Pareto Front
Robust Design
Kriging
Metamodel
Cost reduction
Maximise
Minimise
Evidence
Design
Costs
Beliefs

Keywords

  • evolutionary computation
  • statistical analysis
  • epistemic uncertainty
  • evolutionary approach
  • computational cost reduction technique

Cite this

Alicino, S., & Vasile, M. (2014). An evolutionary approach to the solution of multi-objective min-max problems in evidence-based robust optimization. In Proceedings of the 2014 IEEE Congress on Evolutionary Computation, CEC 2014 (pp. 1179-1186) https://doi.org/10.1109/CEC.2014.6900286
Alicino, Simone ; Vasile, Massimiliano. / An evolutionary approach to the solution of multi-objective min-max problems in evidence-based robust optimization. Proceedings of the 2014 IEEE Congress on Evolutionary Computation, CEC 2014. 2014. pp. 1179-1186
@inproceedings{3fc70de18d1d4cb08d802290d28c790d,
title = "An evolutionary approach to the solution of multi-objective min-max problems in evidence-based robust optimization",
abstract = "This paper presents an evolutionary approach to solve the multi-objective min-max problem (MOMMP) that derives from the maximization of the Belief in robust design optimization. In evidence-based robust optimization, the solutions that minimize the design budgets are robust under epistemic uncertainty if they maximize the Belief in the realization of the value of the design budgets. Thus robust solutions are found by minimizing, with respect to the design variables, the global maximum with respect to the uncertain variables. This paper presents an algorithm to solve MOMMP, and a computational cost reduction technique based on Kriging metamodels. The results show that the algorithm is able to accurately approximate the Pareto front for a MOMMP at a fraction of the computational cost of an exact calculation.",
keywords = "evolutionary computation, statistical analysis, epistemic uncertainty, evolutionary approach, computational cost reduction technique",
author = "Simone Alicino and Massimiliano Vasile",
year = "2014",
month = "7",
day = "6",
doi = "10.1109/CEC.2014.6900286",
language = "English",
isbn = "9781479914883",
pages = "1179--1186",
booktitle = "Proceedings of the 2014 IEEE Congress on Evolutionary Computation, CEC 2014",

}

Alicino, S & Vasile, M 2014, An evolutionary approach to the solution of multi-objective min-max problems in evidence-based robust optimization. in Proceedings of the 2014 IEEE Congress on Evolutionary Computation, CEC 2014. pp. 1179-1186, 2014 IEEE Congress on Evolutionary Computation, CEC 2014, Beijing, United Kingdom, 6/07/14. https://doi.org/10.1109/CEC.2014.6900286

An evolutionary approach to the solution of multi-objective min-max problems in evidence-based robust optimization. / Alicino, Simone; Vasile, Massimiliano.

Proceedings of the 2014 IEEE Congress on Evolutionary Computation, CEC 2014. 2014. p. 1179-1186.

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

TY - GEN

T1 - An evolutionary approach to the solution of multi-objective min-max problems in evidence-based robust optimization

AU - Alicino, Simone

AU - Vasile, Massimiliano

PY - 2014/7/6

Y1 - 2014/7/6

N2 - This paper presents an evolutionary approach to solve the multi-objective min-max problem (MOMMP) that derives from the maximization of the Belief in robust design optimization. In evidence-based robust optimization, the solutions that minimize the design budgets are robust under epistemic uncertainty if they maximize the Belief in the realization of the value of the design budgets. Thus robust solutions are found by minimizing, with respect to the design variables, the global maximum with respect to the uncertain variables. This paper presents an algorithm to solve MOMMP, and a computational cost reduction technique based on Kriging metamodels. The results show that the algorithm is able to accurately approximate the Pareto front for a MOMMP at a fraction of the computational cost of an exact calculation.

AB - This paper presents an evolutionary approach to solve the multi-objective min-max problem (MOMMP) that derives from the maximization of the Belief in robust design optimization. In evidence-based robust optimization, the solutions that minimize the design budgets are robust under epistemic uncertainty if they maximize the Belief in the realization of the value of the design budgets. Thus robust solutions are found by minimizing, with respect to the design variables, the global maximum with respect to the uncertain variables. This paper presents an algorithm to solve MOMMP, and a computational cost reduction technique based on Kriging metamodels. The results show that the algorithm is able to accurately approximate the Pareto front for a MOMMP at a fraction of the computational cost of an exact calculation.

KW - evolutionary computation

KW - statistical analysis

KW - epistemic uncertainty

KW - evolutionary approach

KW - computational cost reduction technique

UR - http://www.scopus.com/inward/record.url?scp=84908576807&partnerID=8YFLogxK

U2 - 10.1109/CEC.2014.6900286

DO - 10.1109/CEC.2014.6900286

M3 - Conference contribution book

SN - 9781479914883

SP - 1179

EP - 1186

BT - Proceedings of the 2014 IEEE Congress on Evolutionary Computation, CEC 2014

ER -

Alicino S, Vasile M. An evolutionary approach to the solution of multi-objective min-max problems in evidence-based robust optimization. In Proceedings of the 2014 IEEE Congress on Evolutionary Computation, CEC 2014. 2014. p. 1179-1186 https://doi.org/10.1109/CEC.2014.6900286