A Process Capability Index for Multivariate Normal and Non-Normal Distributions of Correlated and Uncorrelated Variables

• Autor: Salazar, Erasmo Rafael
• Formato: artículo
• Fecha de Publicación: 2020
• Institución: Universidad de Lima
• Repositorio: Revistas - Universidad de Lima
• Lenguaje: español
• OAI Identifier: oai:revistas.ulima.edu.pe:article/4765
• Enlace del recurso: https://revistas.ulima.edu.pe/index.php/Ingenieria_industrial/article/view/4765
• Nivel de acceso: acceso abierto
• Materia: Multivariate process capability index; Univariate capability index; Non-normal distribution; Normal distribution; Statistical methods; Índice de capacidad de proceso multivariado; Índice de capacidad univariado; Distribución no normal; Distribución normal; Métodos estadísticos

DOI: https://doi.org/10.26439/ing.ind2020.n038.4814 The multivariate process capability analysis includes many indices that are only used when the data is normal and others, when the data is not normal. The same occurs with correlated and uncorrelated quality variables. In this research work, a CPME multivariate capability index was developed by initially using a univariate index—depending on whether the data was o not normal—and any correlation between variables, and then through a characteristic function for the multivariate data. This index may be used in all the aforementioned cases. Some examples of this alternative are presented on a set of real and simulated data, where a broad performance of our proposed index was found against other similar capability indices.