Header menu link for other important links
X
An adaptive one-factor-at-a-time method for robust parameter design: Comparison with crossed arrays via case studies
, Frey D.D., Persons J.B.
Published in American Society of Mechanical Engineers (ASME)
2008
Volume: 2006
   
Pages: 87 - 98
Abstract
This paper presents a conceptually simple and resource efficient method for robust parameter design. The method varies control factors according to an adaptive one-factor-at-a-time plan while varying noise factors using a two-level resolution III fractional factorial array. The method is assessed and compared against a crossed array method which uses fractional factorial inner arrays. The comparison is made by looking at the improvements in robustness across four case studies. The proposed method improves system robustness to more than 80% of the potentially attainable improvement on average. This figure improves to over 90% if prior knowledge of the system is used to define a promising starting point for the search. The results vary across the case studies, but, in general, both the average amount of improvement and the consistency of the results are better than those provided by a crossed array approach as long as experimental error is low enough. This is despite the fact that the proposed method typically uses fewer, and at most equal, number of experiments as the crossed array method. The case studies reveal that the proposed method provides these benefits by exploiting both control by noise interactions, and higher order effects such as control by control by noise interactions, with probability substantially better than random chance. Copyright © 2006 by ASME.
About the journal
JournalProceedings of the ASME Design Engineering Technical Conference
PublisherAmerican Society of Mechanical Engineers (ASME)
Open AccessNo