Robust design modeling with correlated quality characteristics using a multicriteria decision framework

Research output: Contribution to journalReview article

43 Citations (Scopus)

Abstract

Robust design (RD) is a popular design methodology for minimizing the variability of product performance, thereby, improving the quality of a product. Although RD principles have been widely implemented in many industries, and a number of new RD models have been reported in the literature, there is ample room for improvement. First, a number of RD models consider a single quality characteristic (QC). In reality, however, the judgmental basis of a product from the perspective of the customer is characterized by the assessment of multiple QCs, which are often correlated. Second, previous models often fail to emphasize customer preferences in the modeling and optimization phases. To rectify these shortcomings, this paper proposes a compromise programming model for synthesizing Pareto solutions that represent compromised trade-offs between multiple correlated QCs based on the customer's preferences. The Tchebycheff-metric-based compromise programming method employed to determine Pareto solutions is superior to the simplistic weighted sum and goal programming approaches that may fail to explore the full set of Pareto points for a multi-objective optimization problem.

Original languageEnglish
Pages (from-to)423-433
Number of pages11
JournalInternational Journal of Advanced Manufacturing Technology
Volume32
Issue number5-6
DOIs
Publication statusPublished - Mar 2007

Fingerprint

Multiobjective optimization
Industry

Keywords

  • Correlated multiple quality characteristics
  • Entropy method
  • Multi-attribute weight assessment
  • Multi-objective
  • Pareto solutions
  • Robust design optimization

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Software
  • Mechanical Engineering
  • Computer Science Applications
  • Industrial and Manufacturing Engineering

Cite this

@article{79246f675cbb474896016c69f7b3fdb9,
title = "Robust design modeling with correlated quality characteristics using a multicriteria decision framework",
abstract = "Robust design (RD) is a popular design methodology for minimizing the variability of product performance, thereby, improving the quality of a product. Although RD principles have been widely implemented in many industries, and a number of new RD models have been reported in the literature, there is ample room for improvement. First, a number of RD models consider a single quality characteristic (QC). In reality, however, the judgmental basis of a product from the perspective of the customer is characterized by the assessment of multiple QCs, which are often correlated. Second, previous models often fail to emphasize customer preferences in the modeling and optimization phases. To rectify these shortcomings, this paper proposes a compromise programming model for synthesizing Pareto solutions that represent compromised trade-offs between multiple correlated QCs based on the customer's preferences. The Tchebycheff-metric-based compromise programming method employed to determine Pareto solutions is superior to the simplistic weighted sum and goal programming approaches that may fail to explore the full set of Pareto points for a multi-objective optimization problem.",
keywords = "Correlated multiple quality characteristics, Entropy method, Multi-attribute weight assessment, Multi-objective, Pareto solutions, Robust design optimization",
author = "Govindaluri, {Srikrishna Madhumohan} and Cho, {Byung Rae}",
year = "2007",
month = "3",
doi = "10.1007/s00170-005-0349-6",
language = "English",
volume = "32",
pages = "423--433",
journal = "International Journal of Advanced Manufacturing Technology",
issn = "0268-3768",
publisher = "Springer London",
number = "5-6",

}

TY - JOUR

T1 - Robust design modeling with correlated quality characteristics using a multicriteria decision framework

AU - Govindaluri, Srikrishna Madhumohan

AU - Cho, Byung Rae

PY - 2007/3

Y1 - 2007/3

N2 - Robust design (RD) is a popular design methodology for minimizing the variability of product performance, thereby, improving the quality of a product. Although RD principles have been widely implemented in many industries, and a number of new RD models have been reported in the literature, there is ample room for improvement. First, a number of RD models consider a single quality characteristic (QC). In reality, however, the judgmental basis of a product from the perspective of the customer is characterized by the assessment of multiple QCs, which are often correlated. Second, previous models often fail to emphasize customer preferences in the modeling and optimization phases. To rectify these shortcomings, this paper proposes a compromise programming model for synthesizing Pareto solutions that represent compromised trade-offs between multiple correlated QCs based on the customer's preferences. The Tchebycheff-metric-based compromise programming method employed to determine Pareto solutions is superior to the simplistic weighted sum and goal programming approaches that may fail to explore the full set of Pareto points for a multi-objective optimization problem.

AB - Robust design (RD) is a popular design methodology for minimizing the variability of product performance, thereby, improving the quality of a product. Although RD principles have been widely implemented in many industries, and a number of new RD models have been reported in the literature, there is ample room for improvement. First, a number of RD models consider a single quality characteristic (QC). In reality, however, the judgmental basis of a product from the perspective of the customer is characterized by the assessment of multiple QCs, which are often correlated. Second, previous models often fail to emphasize customer preferences in the modeling and optimization phases. To rectify these shortcomings, this paper proposes a compromise programming model for synthesizing Pareto solutions that represent compromised trade-offs between multiple correlated QCs based on the customer's preferences. The Tchebycheff-metric-based compromise programming method employed to determine Pareto solutions is superior to the simplistic weighted sum and goal programming approaches that may fail to explore the full set of Pareto points for a multi-objective optimization problem.

KW - Correlated multiple quality characteristics

KW - Entropy method

KW - Multi-attribute weight assessment

KW - Multi-objective

KW - Pareto solutions

KW - Robust design optimization

UR - http://www.scopus.com/inward/record.url?scp=33847752190&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33847752190&partnerID=8YFLogxK

U2 - 10.1007/s00170-005-0349-6

DO - 10.1007/s00170-005-0349-6

M3 - Review article

VL - 32

SP - 423

EP - 433

JO - International Journal of Advanced Manufacturing Technology

JF - International Journal of Advanced Manufacturing Technology

SN - 0268-3768

IS - 5-6

ER -