Title: Development and application of a reliability-based multivariate model validation method

Authors: Zhenfei Zhan; Yan Fu; Ren-Jye Yang; Yinghong Peng

Addresses: School of Mechanical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China ' Research and Advanced Engineering, Ford Motor Company, Dearborn, MI 48121, USA ' Research and Advanced Engineering, Ford Motor Company, Dearborn, MI 48121, USA ' School of Mechanical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China

Abstract: Validation of computational models with multiple correlated functional responses requires the consideration of multivariate data correlation and uncertainty, and objective robust metrics. This paper presents a reliability based model validation method together with Probabilistic Principal Component Analysis (PPCA) to address these critical issues. The PPCA is employed to address multivariate correlation and to reduce the dimensionality. The reliability assessment method is used to quantitatively assess the quality of multivariate dynamic systems. In addition, physics-based thresholds are defined and transformed for reliability assessment. A rear seat child restraint dynamic system with multiple functional responses is used to demonstrate this new approach.

Keywords: model validation; reliability assessment; PPCA; probabilistic PCA; principal component analysis; child restraint systems; vehicle safety; multivariate modelling; rear seats; passenger safety; child safety.

DOI: 10.1504/IJVD.2012.050079

International Journal of Vehicle Design, 2012 Vol.60 No.3/4, pp.194 - 205

Received: 24 Feb 2011
Accepted: 01 Jun 2011

Published online: 26 Oct 2012 *

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