wafer
AT&T Wafer Experiment
Description
The wafer data frame has 18 rows and 13 columns, of which 8 contain
factors, 4 contain responses, and one is the auxiliary variable N .
It is a design object based on an orthogonal-array design for an
experiment in which
two integrated circuit wafers were made for each
combination of factors, and on each wafer the pre- and post-etch line
widths were measured five times. The response
variables are the mean and deviance of the measurements.
As 3 of the wafers were broken, the auxiliary variable N gives the
number of measurements actually made.
This data frame contains the following columns:
Arguments
| maskdim |
ordered factor with the inheritance attribute "category" , levels are 2 < 2.5
|
| visc.tem |
factor, levels are 204,90 , 206,90 , and 204,105
|
| spinsp |
ordered factor with the inheritance attribute "category" , levels are low < normal < high.
|
| baketime |
ordered factor with the inheritance attribute "category" , levels are 20 < 30 < 40.
|
| aperture |
factor, levels are 1 , 2 , and 3 .
|
| exptime |
ordered factor with the inheritance attribute "category" ,
levels are `-20% < normal < +20%'.
|
| devtime |
an ordered factor with the inheritance attribute "category" ,
levels are ` 30 < 45 < 60'.
|
| etchtime |
ordered factor with the inheritance attribute "category" ,
levels are ` 13.2 < 14.5 < 15.8'.
|
| pre.mean |
numeric vector giving the mean of the pre-etch line width
|
| pre.dev |
numeric vector giving the deviance of the pre-etch line width
|
| post.mean |
numeric vector giving the mean of the post-etch line width
|
| post.dev |
numeric vector giving the deviance of the post-etch line width
|
| N |
numeric vector giving the number of measurements.
|
Source
Phadke, M.S., Kackar, R.N., Speene, D.V., and Grieco, M.J. (1983)
Off-line Quality Control in Integrated Circuit Fabrication Using
Experimental Design. Bell System Technical Journal
Vol. 62, pp. 1273--309.
John M. Chambers and Trevor J. Hastie, (eds.) Statistical Models in S,
Wadsworth and Brooks, Pacific Grove, CA 1992, pg. 147.
Examples
data(wafer)
wpm <- wafer[, c(1:9)]
waov1 <- aov(pre.mean ~ . , wpm)