meanDI {desire} | R Documentation |
Computes the weighted mean of a number of desirability functions.
meanDI(f, ..., weights = 1)
f,... |
desirability functions. |
weights |
vector of weights. Weights do not need to sum to one. |
The Desirability Index was introduced by Harrington (1965), and the concept was extended by Derringer and Suich (1980). It is a means for multicriteria (quality) optimization in industrial quality management. All desirability functions of the quality criteria are combined into a univariate global quality criterion which has to be optimized. The Weighted Mean Desirability Index is related to the concept of utility functions.
The function can be used for Harrington as well as Derringer and Suich desirability functions.
meanDI(f, ..., weights)
returns a function object of the Weighted Mean Desirability Index.
Heike Trautmann trautmann@statistik.tu-dortmund.de, Detlef Steuer steuer@hsu-hamburg.de and Olaf Mersmann olafm@statistik.tu-dortmund.de
J. Harrington (1965): The desirability function. Industrial Quality Control, 21: 494-498.
G.C. Derringer, D. Suich (1980): Simultaneous optimization of several response variables. Journal of Quality Technology 12 (4): 214-219.
harrington1
and harrington2
for Harrington type desirability functions;
derringerSuich
for desirability functions of Derringer and Suich;
geometricDI
,minimumDI
for other types of Desirability indices.
h1 <- harrington1(-2, .9, 2, .1)
h2 <- harrington2(0, 2, 2)
di <- meanDI(h1, h2,weights=c(0.2,0.8))
di(c(0, 1))
## Desirability Index of vector input:
h <- harrington2(3,7,1)
g <- harrington1(-2, .1, 2, .9)
d <- meanDI(h, g,weights=c(0.3,0.7))
m <- matrix(c(seq(2, 8, 0.1), seq(-2, 4, 0.1)), ncol=2, byrow=FALSE)
apply(m, 1, d)