Principal Component Analysis (PCA) has been widely applied to identify the sources of dimensional variation in automotive body assembly; however, current PCA methods based on a covariance matrix are not appropriate for dealing with high-dimensional data, noisy data, and missing data. Due to its adaptive nature and fast computational capability, an artificial neural network has been considered as a new alternative to overcome the statistical analysis problems. In this paper, an artificial neural network using a nonlinear transfer function will be introduced to overcome current data analysis problems in an auto assembly process. A case study is used to demonstrate the application of the proposed approach.
Improving principal component analysis (PCA) in automotive body assembly using artificial neural networks
Journal of Manufacturing Systems ; 20 , 3 ; 188-197
2001
10 Seiten, 8 Bilder, 2 Tabellen, 11 Quellen
Article (Journal)
English
Improving automotive dimensional quality by using principal component analysis
Tema Archive | 1996
|Image compression using principal component neural networks
British Library Online Contents | 2001
|Neural Networks for Inverse Problems Using Principal Component Analysis and Orthogonal Arrays
British Library Conference Proceedings | 2003
|