2015 QPRC

Principal components regression-based control charts for monitoring plastic plywood process

Danilo Marcondes Filho
Statistical Department, Federal University of Rio Grande do Sul. Av. Bento Gonçalves, 9500 – Porto Alegre, RS 91509-900, Brazil. Tel: +55 (51) 3308-6225 , Fax: +55 (51)3308-7301 . Email: marcondes.danilo@gmail.com


Ângelo Márcio Oliveira Sant’Anna
Department of Mechanical Engineering, University of Bahia, Rua Aristides Novis, 02, Federação, Salvador, BA 40210-630, Brazil. Tel: +55 (71) 3283-9703 / Fax: +55 (71) 3283-9702 . Email: angeloms@gmail.com

Control charts based on regression models are appropriate for monitoring in which the quality characteristics of products vary depending on adjustments in process variables (or input variables). Its use enables monitoring the correlation structure between input variables and the response variable through residuals from the fitted model according to historical process data. However, such strategy is restricted to data from input variables which are not significantly correlated. Otherwise, collinear variables that hold substantial information on the variability of the response variable might be absent in the regression model adjustment. This paper proposes a strategy for monitoring count data in plastic plywood process, combining the Poisson regression and principal component analysis. In such strategy, collinear variables are turned into uncorrelated variables by multivariate analysis and a Poisson regression is performed on principal components scores. A deviance residual control chart from fitted model is then used to evaluate the laminated plastic plywood manufacturing process.

Keywords: Statistical processes control; control charts; principal components analysis; Poisson regression; plywood manufacture.