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.
Skip Nav Destination
ASME 2006 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference
September 10–13, 2006
Philadelphia, Pennsylvania, USA
Conference Sponsors:
- Design Engineering Division and Computers and Information in Engineering Division
ISBN:
0-7918-4258-4
PROCEEDINGS PAPER
An Adaptive One-Factor-at-a-Time Method for Robust Parameter Design: Comparison With Crossed Arrays Via Case Studies
Daniel D. Frey,
Daniel D. Frey
Massachusetts Institute of Technology, Cambridge, MA
Search for other works by this author on:
Nandan Sudarsanam,
Nandan Sudarsanam
Massachusetts Institute of Technology, Cambridge, MA
Search for other works by this author on:
Jeffrey B. Persons
Jeffrey B. Persons
Massachusetts Institute of Technology, Cambridge, MA
Search for other works by this author on:
Daniel D. Frey
Massachusetts Institute of Technology, Cambridge, MA
Nandan Sudarsanam
Massachusetts Institute of Technology, Cambridge, MA
Jeffrey B. Persons
Massachusetts Institute of Technology, Cambridge, MA
Paper No:
DETC2006-99593, pp. 87-98; 12 pages
Published Online:
June 3, 2008
Citation
Frey, DD, Sudarsanam, N, & Persons, JB. "An Adaptive One-Factor-at-a-Time Method for Robust Parameter Design: Comparison With Crossed Arrays Via Case Studies." Proceedings of the ASME 2006 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference. Volume 4a: 18th International Conference on Design Theory and Methodology. Philadelphia, Pennsylvania, USA. September 10–13, 2006. pp. 87-98. ASME. https://doi.org/10.1115/DETC2006-99593
Download citation file:
3
Views
Related Proceedings Papers
Related Articles
Formal Solution of N-Type Robust Parameter Design Problems with Stochastic Noise Factors
J. Mech. Des (June,1994)
An Adaptive One-Factor-at-a-Time Method for Robust Parameter Design: Comparison With Crossed Arrays via Case Studies
J. Mech. Des (February,2008)
Robust Design Optimization of Component Parameters for DMDEB Powertrain System Based on Taguchi Method
J. Dyn. Sys., Meas., Control (September,2022)
Related Chapters
Methods to Select and Compound Noise Factors
Taguchi Methods: Benefits, Impacts, Mathematics, Statistics and Applications
Mash 2-1 Multi-Bit Sigma-Delta Modulator for WLAN
International Conference on Future Computer and Communication, 3rd (ICFCC 2011)
Efficient Transmission Images with QAM Modulation over an AWGN Channel
International Conference on Computer and Electrical Engineering 4th (ICCEE 2011)