A1263
Title: Extending a novel class of control charts for sequential monitoring to the multivariate framework
Authors: Claudio Giovanni Borroni - University of Milano - Bicocca (Italy) [presenting]
Manuela Cazzaro - University of Milano-Bicocca (Italy)
Abstract: The change-point paradigm has been successfully applied to statistical process control by building many parametric and nonparametric charts aimed at the sequential monitoring of some univariate process characteristics. It is known, however, that the quality of a process is often measured by multivariate variables and also that the dependence structure of such variables cannot be ignored to provide a prompt signal of the drift of the process from the in-control situation. In the extension of univariate control tools to the multivariate case, the so-called interpoint distances play a key role and can be variously defined. A suitable version of such distances is sought to fit a non-conventional approach to the change-point methodology, which has been studied in the univariate setting so far. The focus is on the efficacy of the resulting multivariate control charts but also on the simplicity of their application.
