DesignExperimental Matrix

Possibilities:

Factorial with centerpoint
Box-Behnken
Central composite design

Runexperimental matrix; collect data

Analyze datausingMultiple Linear Regression, withsecondorder equation:

Main Effects
X1
X2
X3

Second Order Effects
X1*X1
X2*X2
X3*X3

Interactions
X1*X2
X2*X3
X1*X3

Response

=

A

+B*X1+C*X2+D*X3
+E*X1^2+F*X2^2+G*X3^2
+I*X1*X2+J*X2*X3+K*X1*K3

GenerateResponse Surfaces, using model from multiple linear regression

-
-

Contour plots
Mesh plots

INPUT VARIABLES

RESPONSE

bs Implant Dose(E11)

Blanket Implant Dose(E11)

Vt–p (mV)

2.71

1.84

–

1088

16.2

1.84

–

1282

2.71

7.44

–

577

16.2

7.44

–

881

23.6

3.70

–

1257

1.87

3.70

–

913

6.64

9.94

–

402

6.64

1.38

–

1187

6.64

3.70

–

1012

B > let c11 = c1 • c1

B > let c12 = c1 • c2

B > let c22 = c2 • c2

B > regress c10

5

c1

c2

c11

c22

c12;

BC > coeff c20;

BC > resid c30.

B > nscores

c30

c31

B > plot

c30

c31

B > corr

c30

c31

B > WRITE ‘VT VS 12 RSM.COEFF’

C20

B > NAME C101 ‘SUBS’ C102‘BLANKET’ C103 ‘VT’

B > GRID C101= 1.8:2.4, C102 = 1.3:10
B > LET C103= 1189 + 16.2*C101 – 77.6*C102 – 0.258*C101**2– 2.28*C102**2
B > LET C103 = C103 + 1.69*C101*C102

B > CONTOUR C103

C101

C102