Methods in Mahalanobis-Taguchi (MT) System

Mahalanobis-Taguchi (MT) system is a collection of multivariate analysis methods developed for the field of quality engineering. MT system consists of two families depending on their purpose. One is a family of Mahalanobis-Taguchi (MT) methods (in the broad sense) for diagnosis (see Woodall, W. H., Koudelik, R., Tsui, K. L., Kim, S. B., Stoumbos, Z. G., and Carvounis, C. P. (2003) ) and the other is a family of Taguchi (T) methods for forecasting (see Kawada, H., and Nagata, Y. (2015) ). The MT package contains three basic methods for the family of MT methods and one basic method for the family of T methods. The MT method (in the narrow sense), the Mahalanobis-Taguchi Adjoint (MTA) methods, and the Recognition-Taguchi (RT) method are for the MT method and the two-sided Taguchi (T1) method is for the family of T methods. In addition, the Ta and Tb methods, which are the improved versions of the T1 method, are included.


MTSYS

Travis-CI Build Status CRAN_Status_Badge Coverage Status

MTSYS provides a collection of multivariate analysis methods in Mahalanobis-Taguchi System (MTS), which was developed for the field of quality engineering. MTS consists of two families depending on their purpose. One is a family of Mahalanobis-Taguchi (MT) methods (in the broad sense) for diagnosis and the other is a family of Taguchi (T) methods for forecasting.

Overview

The following methods are implemented.

  • MT method
  • MTA method
  • RT method

A family of T methods

  • T(1) method
  • Ta method
  • Tb method

For details, see the following referenses.

Installation

Install the release version from CRAN:

install.packages("MTSYS")

Or the development version from github

# install.packages("devtools")
devtools::install_github("okayaa/MTSYS")

Example

library(MTSYS)
 
# 40 data for versicolor in the iris dataset
iris_versicolor <- iris[61:100, -5]
 
unit_space_MT <- MT(unit_space_data = iris_versicolor)
 
# 10 data for each kind (setosa, versicolor, virginica) in the iris dataset
iris_test <- iris[c(1:10, 51:60, 101:111), -5]
 
diagnosis_MT <- diagnosis(unit_space = unit_space_MT, newdata = iris_test, 
                          threshold = 4)
 
(diagnosis_MT$le_threshold)
#>     1     2     3     4     5     6     7     8     9    10
#> FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#>   51    52    53    54    55    56    57    58    59    60   
#> TRUE  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE
#>  101   102   103   104   105   106   107   108   109   110   111 
#> TRUE  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE 

References

  • Woodall, W. H., Koudelik, R., Tsui, K. L., Kim, S. B., Stoumbos, Z. G., and Carvounis, C. P. (2003) A review and analysis of the Mahalanobis-Taguchi system. Technometrics, 45(1), 1-15. <doi:10.1198/004017002188618626>
  • Kawada, H., and Nagata, Y. (2015) An application of a generalized inverse regression estimator to Taguchi's T-Method. Total Quality Science, 1(1), 12-21. <doi:10.17929/tqs.1.12>

News

MTSYS v1.2.0

In this version,

  • Mistakes were fixed in the Tb method.
  • A Parameter "subtracts_V_e" was added in functions for a family of T methods.

MTSYS v1.0.0

Initial release in CRAN

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.

install.packages("MTSYS")

1.2.0 by Akifumi Okayama, 9 months ago


https://github.com/okayaa/MTSYS


Report a bug at https://github.com/okayaa/MTSYS/issues


Browse source code at https://github.com/cran/MTSYS


Authors: Akifumi Okayama [aut, cre], Masato Ohkubo [ctb], Yasushi Nagata [ctb]


Documentation:   PDF Manual  


MIT + file LICENSE license


Imports stats

Suggests testthat, covr


See at CRAN