1. Propuesta de Carta de Control Multivariada utilizando la Media Winsorizada basada en la Carta |S| / Proposed Multivariate Control Chart using Winsorized Mean based on the |S| Chart
- Author
-
Barreto Pombo, Rister Junior, Ariza Guerrero, Angellys Paola, and Herrera Acosta, Roberto José
- Subjects
quality ,outliers ,Extremos ,determinants ,Sensibilidad ,Variability ,sensibility ,Determinantes ,Variabilidad ,Calidad - Abstract
The global quality of a product is often subject to the simultaneous evaluation of several of its features; under these circumstances, it becomes necessary to implement a multivariate analysis. The variance-covariance matrix is one of the most important quality control statistics in multivariate settings and it can be monitored using different methods, such as the Shewhart, CUSUM and EWMA charts. In this article, a construction method for the multivariate control chart |S| is proposed using the Winsorized Mean as the mean vector estimator. It was demonstrated that this modification was more efficient when the sampled data hat outliers. This study shows a comparison between the traditional multivariate control chart and the proposed chart, which was found to have more sensitivity. El monitoreo global de la calidad de un producto está sujeto a la evaluación simultánea de varias de sus características; es necesario bajo estas condiciones la implementación de las cartas de control tipo multivariadas. La variabilidad, en este caso la matriz de varianza covarianza, es sin duda el más importante de los estadísticos desde la perspectiva multivariada, que puede ser monitoreada con distintas cartas. Entre éstas se encuentran: las cartas Shewhart, CUSUM y EWMA. En este artículo se desarrolla una metodología de implementación de la Media Winsorizada en la carta de control multivariada de varianza efectiva |S|, encontrando una gran utilidad en procesos con valores extremos. El estudio muestra además una comparación entre la carta de control tradicional multivariante y la carta propuesta, que muestra mayor sensibilidad. Abstract— The global quality monitoring of a product is often subject to the simultaneous evaluation of several of its features; under these circumstances it is necessary to implement multivariate control charts. Variability, in this particular case, the variance-covariance matrix is indisputably the most important of the statistics from the multivariate perspective and it can be monitored with different charts, among these: Shewhart, CUSUM and EWMA. This article develops the Winsorized Mean in the effective variance multivariate control |S|-chart implementation methodology and it was demonstrated that the modification was more efficient when the sample hat outliers. This study shows a comparison between the traditional multivariate control chart and a proposed chart which was found to have more sensitivity.  
- Published
- 2018