Browse by author
Lookup NU author(s): Professor Elaine Martin,
Emeritus Professor Julian Morris
Full text for this publication is not currently held within this repository. Alternative links are provided below where available.
A recursive Exponentially Weighted Principal Component Analysis (EWPCA) methodology is proposed for the monitoring of dynamic manufacturing processes that are subject to both drift over time and operational changes that are part of normal process behaviour. The process representation, that is based on principal component analysis, is updated following the measurement of each new observation, with the weighting of each new observation being determined by current process variation. The number of principal components included in the model, and the control limits for the monitoring chart, are determined recursively. The methodology is demonstrated by application to an industrial sheet forming process.
Author(s): Martin EB; Morris AJ; Lane S
Publication type: Article
Publication status: Published
Journal: Pulp and Paper Canada
ISSN (print): 0316-4004
Publisher: Business Information Group