Title: MPP search method using limited radial importance sampling

Authors: Mahmoud Awad, Nanua Singh, Agus Sudjianto

Addresses: PTO Quality Office, Ford Motor Company, Dearborn, MI 48129, USA. ' Department of Industrial and Manufacturing Engineering, Wayne State University, Detroit, MI 48201, USA. ' Bank of America Charlotte, NC 28255, USA

Abstract: Competitiveness and current market conditions push product development to strive for more product functionality, fewer design cycles, reduced cost and better quality. One practical approach to achieve that is through building robustness in early design stages utilising Design for Six Sigma (DFSS) disciplines. In this paper, we propose a new search method for the Most Probable Point (MPP), which is a good metric for measuring product robustness. The new method is based on a simulation within a limited radial sampling region determined by an initial uniform DOE that guarantees a good coverage of design space. Numerical non-linear examples are used to compare and verify the validity of the new method.

Keywords: design for six sigma; DFSS; reliability estimation; radial sampling; importance sampling; most probable point; MPP; product development.

DOI: 10.1504/IJPD.2006.009896

International Journal of Product Development, 2006 Vol.3 No.3/4, pp.305 - 318

Published online: 01 Jun 2006 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article