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 language | English |
---|---|
Pages (from-to) | 423-433 |
Number of pages | 11 |
Journal | International Journal of Advanced Manufacturing Technology |
Volume | 32 |
Issue number | 5-6 |
DOIs | |
Publication status | Published - Mar 2007 |
Externally published | Yes |
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