The use of principal components and univariate charts to control multivariate processes

Imagem de Miniatura

Data

2008-01-01

Título da Revista

ISSN da Revista

Título de Volume

Editor

Resumo

In this article, we evaluate the performance of the T2 chart based on the principal components (PC chart) and the simultaneous univariate control charts based on the original variables (SU X̄ charts) or based on the principal components (SUPC charts). The main reason to consider the PC chart lies on the dimensionality reduction. However, depending on the disturbance and on the way the original variables are related, the chart is very slow in signaling, except when all variables are negatively correlated and the principal component is wisely selected. Comparing the SU X̄, the SUPC and the T 2 charts we conclude that the SU X̄ charts (SUPC charts) have a better overall performance when the variables are positively (negatively) correlated. We also develop the expression to obtain the power of two S 2 charts designed for monitoring the covariance matrix. These joint S2 charts are, in the majority of the cases, more efficient than the generalized variance |S| chart.

Descrição

Palavras-chave

Multivariate process control, Principal component, Simultaneous univariate control charts

Como citar

Pesquisa Operacional, v. 28, n. 1, p. 173-196, 2008.