In the multivariate analysis of capacity many aspects even not exist resolved around some indexes and their application. For example, when the variables of quality are correlated and when they are not correlated. In this work we seek to propose the construction of a multivariate index of capacity that works for both previous cases, and to present an example of application of this alternative on a real dataset. A wide acting of our index is obtained proposed in front of other similar indexes of capacity.

In the multivariate analysis of capacity there are many aspects, not even resolved yet, around some indexes such as the non-normality of the data and if the variables of quality are correlated or not. In this work we intend to propose the development of a multivariate capacity index (ICPM, for its acronym in Spanish) that works for both previous cases through simulated data examples. It is worth mentioning that we find a wide performance of the proposed index in comparison with other similar capacity indexes.

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.

In the multivariate analysis of capacity many aspects even not exist resolved around some indexes and their application. For example, when the variables of quality are correlated and when they are not correlated. In this work we seek to propose the construction of a multivariate index of capacity that works for both previous cases, and to present an example of application of this alternative on a real dataset. A wide acting of our index is obtained proposed in front of other similar indexes of capacity.

In the multivariate analysis of capacity there are many aspects, not even resolved yet, around some indexes such as the non-normality of the data and if the variables of quality are correlated or not. In this work we intend to propose the development of a multivariate capacity index (ICPM, for its acronym in Spanish) that works for both previous cases through simulated data examples. It is worth mentioning that we find a wide performance of the proposed index in comparison with other similar capacity indexes.

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.