STAT 503, Fall 2005
Homework Solution 6
Due: Nov. 3, 2005
6-1 An engineer is interested in the effects of cutting speed (A), tool geometry (B),
and cutting angle on the life (in hours) of a machine tool. Two levels of each factor
are chosen, and three replicates of a 23 factorial design are run. The results follow:
Treatment
A
Combinatio
n
(1)
ab
ac
bc
abc
Replicat
e
II
I
2
2
3
31
2
3
5
5
34
5
4
4
4
45
0
6
0
3
50
III
2
5
2
43
9
5
0
4
47
6
3
8
3
37
6
5
4
4
41
(a) Estimate the factor effects. Which effects appear to be large?
From the normal probability plot of effects below, factors B, C, and the AC interaction
appear to be significant.
DE SI G N- EX PE RT P lo t
Li fe
A: Cut t in g Spe ed
B: Too l G eom et r y
C: Cut t in g Ang le
N o rm a l p lo t
N o rm a l % p ro b a b ility
99
95
90
80
70
50
30
20
10
5
1
AC
-8 .8 3
-3 .7 9
1 .2 5
6 .2 9
1 1 .3 3
E ffe c t
(b) Use the analysis of variance to confirm your conclusions for part (a).
The analysis of variance confirms the significance of
interaction.
Design Expert Output
Response:
Life
in hours
ANOVA for Selected Factorial Model
Analysis of variance table [Partial sum of squares]
Sum of
Mean
Source
Squares
DF
Square
Model
1612.67
7
230.38
A
0.67
1
0.67
B
770.67
1
770.67
F
Value
7.64
0.022
25.55
6-1
Prob > F
0.0004
0.8837
0.0001
factors B, C, and the AC
significant
STAT 503, Fall 2005
C
AB
AC
BC
ABC
Pure Error
Cor Total
280.17
16.67
468.17
48.17
28.17
482.67
2095.33
Homework Solution 6
1
1
1
1
1
16
23
280.17
16.67
468.17
48.17
28.17
30.17
9.29
0.55
15.52
1.60
0.93
Due: Nov. 3, 2005
0.0077
0.4681
0.0012
0.2245
0.3483
The Model F-value of 7.64 implies the model is significant. There is only
a 0.04% chance that a "Model F-Value" this large could occur due to noise.
The reduced model ANOVA is shown below.
hierarchy.
Design Expert Output
Response:
Life
in hours
ANOVA for Selected Factorial Model
Analysis of variance table [Partial sum of squares]
Sum of
Mean
Source
Squares
DF
Square
Model
1519.67
4
379.92
A
0.67
1
0.67
B
770.67
1
770.67
C
280.17
1
280.17
AC
468.17
1
468.17
Residual
575.67
19
30.30
Lack of Fit
93.00
3
31.00
Pure Error
482.67
16
30.17
Cor Total
2095.33
23
Factor A was included to maintain
F
Value
12.54
0.022
25.44
9.25
15.45
1.03
Prob > F
< 0.0001
0.8836
< 0.0001
0.0067
0.0009
0.4067
significant
not significant
The Model F-value of 12.54 implies the model is significant. There is only
a 0.01% chance that a "Model F-Value" this large could occur due to noise.
Effects B, C and AC are significant at 1%.
(c) Write down a regression model for predicting tool life (in hours) based on the
results of this experiment.
yijk 40.8333 0.1667 x A 5.6667 xB 3.4167 xC 4.4167 x A xC
Design Expert Output
Coefficient
Factor
Estimate
DF
Intercept
40.83
1
A-Cutting Speed
0.17
1
B-Tool Geometry
5.67
1
C-Cutting Angle
3.42
1
AC
-4.42
1
Final Equation in Terms of Coded Factors:
Life
+40.83
+0.17
+5.67
+3.42
-4.42
Standard
Error
1.12
1.12
1.12
1.12
1.12
95% CI
Low
38.48
-2.19
3.31
1.06
-6.77
=
*A
*B
*C
* A* C
Final Equation in Terms of Actual Factors:
Life
+40.83333
+0.16667
+5.66667
+3.41667
=
* Cutting Speed
* Tool Geometry
* Cutting Angle
6-2
95% CI
High
43.19
2.52
8.02
5.77
-2.06
VIF
1.00
1.00
1.00
1.00
STAT 503, Fall 2005
Homework Solution 6
-4.41667
Due: Nov. 3, 2005
* Cutting Speed * Cutting Angle
The equation in part (c) and in the given in the computer output form a hierarchial
model, that is, if an interaction is included in the model, then all of the main effects
referenced in the interaction are also included in the model.
(d) Analyze the residuals. Are there any obvious problems?
Normal plot of residuals
Residuals vs. Predicted
11.5
95
90
6.79167
80
70
Res iduals
Norm al % probability
99
50
2.08333
30
20
10
5
-2.625
1
-7.33333
-7.33333
-2.625
2.08333
6.79167
11.5
27.17
33.92
Res idual
40.67
47.42
54.17
Predicted
There is nothing unusual about the residual plots.
(e) Based on the analysis of main effects and interaction plots, what levels of A, B,
and C would you recommend using?
Since B has a positive effect, set B at the high level to increase life. The AC
interaction plot reveals that life would be maximized with C at the high level and A at
the low level.
DE SI GN - EX PE RT P lo t
DE SI GN - EX PE RT P lo t
Li fe
Li fe
X = A: C ut t in gS pee d
Y = C: C ut t in gA ngl e
X = B: T oo l G eom et r y
Act ual Fact or s
A: C ut ti n gS pee d = 0. 00
C: C ut ti n gA ngl e = 0. 00
C- - 1. 00 0
C+ 1. 00 0
Act ual Fact or
B: T ool G eom et r y= 0 .0 0
Inte ra c tio n G ra p h
O ne F a c to r P lo t
C u ttin g A n g le
60
60
5 0 .5
L ife
L ife
5 0 .5
41
41
3 1 .5
3 1 .5
22
22
-1 .0 0
-0 . 5 0
0 .0 0
0 .5 0
1 .0 0
-1 . 0 0
C u ttin g S p e e d
-0 .5 0
0 .0 0
0 .5 0
T o o l G e o m e tr y
6-3
1 .0 0
STAT 503, Fall 2005
Homework Solution 6
Due: Nov. 3, 2005
6-3 Find the standard error of the factor effects and approximate 95 percent
confidence limits for the factor effects in Problem 6-1. Do the results of this analysis
agree with the conclusions from the analysis of variance?
SE( effect )
1
n2 k 2
Variabl
A
B
AB
C
AC
BC
ABC
S2
3 2 3 2
Effect
0.333
11.33
-1.667
6.833
-8.833
-2.833
-2.167
30.17 2.24
*
*
*
The 95% confidence intervals for factors B, C and AC do not contain zero.
agrees with the analysis of variance approach.
This
6-8 A bacteriologist is interested in the effects of two different culture media and
two different times on the growth of a particular virus. She performs six replicates of
a 22 design, making the runs in random order. Analyze the bacterial growth data that
follow and draw appropriate conclusions. Analyze the residuals and comment on the
models adequacy.
Culture Medium
Medium
1
2
Time
12 hr
18 hr
21
22
25
26
23
28
24
25
20
26
29
27
37
39
31
34
38
38
29
33
35
36
30
35
Design Expert Output
Response:
Virus growth
ANOVA for Selected Factorial Model
Analysis of variance table [Partial sum of squares]
Sum of
Mean
Source
Squares
DF
Square
Model
691.46
3
230.49
A
9.38
1
9.38
B
590.04
1
590.04
AB
92.04
1
92.04
Residual
102.17
20
5.11
Lack of Fit
0.000
0
Pure Error
102.17
20
5.11
Cor Total
793.63
23
F
Value
45.12
1.84
115.51
18.02
The Model F-value of 45.12 implies the model is significant. There is only
a 0.01% chance that a "Model F-Value" this large could occur due to noise.
Values of "Prob > F" less than 0.0500 indicate model terms are significant.
In this case B, AB are significant model terms.
6-4
Prob > F
< 0.0001
0.1906
< 0.0001
0.0004
significant
STAT 503, Fall 2005
Homework Solution 6
Normal plot of residuals
Due: Nov. 3, 2005
Residuals vs. Predicted
4.66667
95
90
2.66667
80
70
Res iduals
Norm al % probability
99
50
0.666667
30
20
10
5
-1.33333
1
-3.33333
-3.33333
-1.33333
0.666667
2.66667
4.66667
23.33
26.79
Res idual
30.25
33.71
37.17
Predicted
Growth rate is affected by factor B (Time) and the AB interaction (Culture medium
and Time). There is some very slight indication of inequality of variance shown by
the small decreasing funnel shape in the plot of residuals versus predicted.
DE SI GN - EX PE RT P lo t
Vi ru sgr o wt h
X = A: C ul t ur eM e di um
Y = B: T i m e
Desi gn P oi nt s
B- 1 2. 000
B+ 18 .0 00
Inte ra c tio n G ra p h
39
T im e
2
V iru s g ro w th
3 4 .2 5
2 9 .5
2 4 .7 5
20
1
C u ltu re M e d i u m
6-5
