In this study, a new optimization approach for robust design, design for multi-objective six sigma, has been developed and applied to three robust optimization problems. The design for multi-objective six sigma builds on the ideas of design for six sigma, coupled with multiobjective evolutionary algorithm, for an enhanced capability to reveal tradeoff information considering both optimality and robustness of design. While design for six sigma requires careful input parameter setting, design for multi-objective six sigma needs no such prior tuning, plus it can reveal the tradeoff information in a single optimization run. Three robust optimization problems were taken as to demonstrate the capabilities of design for multi-objective six sigma. Results indicate that design for multi-objective six sigma has a more practical and more efficient capability than the design for six sigma to reveal tradeoff design information considering both optimality and robustness of design.
|Number of pages||19|
|Journal||Journal of Aerospace Computing, Information and Communication|
|Publication status||Published - 2008 Aug|
ASJC Scopus subject areas
- Aerospace Engineering
- Computer Science Applications
- Electrical and Electronic Engineering