A new proposal to solve the autocorrelation problem for monitoring processes in the cutlery industry

Abstract

Traditional multivariate control charts consider the observations to be independent. In many industrial applications, however, a certain autocorrelation level can normally be observed. This can be due to the nature of the process itself (inertia in the process), or to the way in which the sample is obtained. In the latter case, the problem involved in the autocorrelation existence was treated by the individual modelling of each series or the application of vector autoregression (VAR) models. In the former case, the analysis of the cross correlation structure between the variables is altered. In the latter one, if the cross correlation is not strong, the filtering process may modify the weakest relations. In order to improve these aspects, state-space models have been introduced into multivariate statistical process control. In the present paper, we consider a cutlery industry in order to develop a complete scheme for the efficient construction of control charts for innovations and the “inertial factor”. This is the main difference from other proposals in the literature; that is, consideration of the inertial factor in the statistical process control.