12.9. A variation of Example 12.1. In Example 12.1

(which utilized data from Example 6.2), we found that one of the process variables (B $ pressure) was not important.

Dropping this variable produces two replicates of a 23 design.

The data are as follows:

Assume that C and D are controllable factors and that A is a noise variable.

(a) Fit a model to the mean response.

(b) Fit a model to the ln(s2) response.

(c) Find operating conditions that result in the mean filtration rate response exceeding 75 with minimum variance.

(d) Compare your results with those from Example 12.1, which used the transmission of error approach. How similar are the two answers?

C D A(+) A (-) y 45,48 71,65 57.75 121.19 +- 68, 80 60, 65 68.25 72.25 + 43,45 100, 104 73.00 1124.67 + 75,70 86,96 81.75 134.92