Documentos de Académico
Documentos de Profesional
Documentos de Cultura
Multi-Factored Experiments
Engineering 9516
Dr. Leonard M. Lye, P.Eng, FCSCE, FEC
Professor and Associate Dean (Graduate Studies)
Faculty of Engineering and Applied Science, Memorial University of
Newfoundland
St. Johns, NL, A1B 3X5
L. M. Lye
DOE Course
DOE - I
Introduction
L. M. Lye
DOE Course
L. M. Lye
DOE Course
Assumptions
You have
DOE Course
Others
Raymond Myers, J. S. Hunter, W. G. Hunter, Yates,
Montgomery, Finney, etc..
L. M. Lye
DOE Course
L. M. Lye
DOE Course
References
D. G. Montgomery (2008): Design and Analysis of
Experiments, 7th Edition, John Wiley and Sons
one of the best book in the market. Uses Design-Expert
software for illustrations. Uses letters for Factors.
DOE Course
DOE Course
Strategy of Experimentation
Strategy of experimentation
Best guess approach (trial and error)
can continue indefinitely
cannot guarantee best solution has been found
L. M. Lye
DOE Course
10
L. M. Lye
DOE Course
11
DOE Course
12
DOE Course
13
L. M. Lye
DOE Course
14
INPUTS
(Factors)
X variables
OUTPUTS
(Responses)
Y variables
People
Materials
PROCESS:
Equipment
responses related
to performing a
service
Policies
responses related
to producing a
produce
Procedures
A Blending of
Inputs which
Generates
Corresponding
Outputs
responses related
to completing a task
Methods
Environment
L. M. Lye
Illustration of a Process
DOE Course
15
INPUTS
(Factors)
X variables
OUTPUTS
(Responses)
Y variables
Type of
cement
compressive
strength
Percent water
PROCESS:
Type of
Additives
Percent
Additives
Mixing Time
modulus of elasticity
Discovering
Optimal
Concrete
Mixture
modulus of rupture
Poisson's ratio
Curing
Conditions
% Plasticizer
L. M. Lye
DOE Course
16
INPUTS
(Factors)
X variables
OUTPUTS
(Responses)
Y variables
Type of Raw
Material
Mold
Temperature
Holding
Pressure
PROCESS:
% shrinkage from
mold size
Holding Time
Gate Size
thickness of molded
part
Manufacturing
Injection
Molded Parts
number of defective
parts
Screw Speed
Moisture
Content
L. M. Lye
17
INPUTS
(Factors)
X variables
OUTPUTS
(Responses)
Y variables
Impermeable layer
(mm)
Initial storage
(mm)
PROCESS:
Coefficient of
Infiltration
Coefficient of
Recession
Rainfall-Runoff
Model
Calibration
R-square:
Predicted vs
Observed Fits
Soil Moisture
Capacity
(mm)
L. M. Lye
Model Calibration
DOE Course
18
INPUTS
(Factors)
X variables
OUTPUTS
(Responses)
Y variables
Brand:
Cheap vs Costly
PROCESS:
Taste:
Scale of 1 to 10
Time:
4 min vs 6 min
Power:
75% or 100%
Height:
On bottom or raised
Making the
Best
Microwave
popcorn
Bullets:
Grams of unpopped
corns
DOE Course
19
Boiling water
Factors: Pan type, burner size, cover
Response: Time to boil water
D-day
Factors: Type of drink, number of drinks, rate of drinking, time
after last meal
Response: Time to get a steel ball through a maze
Mailing
Factors: stamp, area code, time of day when letter mailed
Response: Number of days required for letter to be delivered
L. M. Lye
DOE Course
20
More examples
Cooking
Factors: amount of cooking wine, oyster sauce, sesame oil
Response: Taste of stewed chicken
Sexual Pleasure
Factors: marijuana, screech, sauna
Response: Pleasure experienced in subsequent you know what
Basketball
Factors: Distance from basket, type of shot, location on floor
Response: Number of shots made (out of 10) with basketball
Skiing
Factors: Ski type, temperature, type of wax
Response: Time to go down ski slope
L. M. Lye
DOE Course
21
Basic Principles
Statistical design of experiments (DOE)
the process of planning experiments so that
appropriate data can be analyzed by statistical
methods that results in valid, objective, and
meaningful conclusions from the data
involves two aspects: design and statistical
analysis
L. M. Lye
DOE Course
22
DOE Course
23
Randomization
cornerstone of all statistical methods
average out effects of extraneous factors
reduce bias and systematic errors
Blocking
increases precision of experiment
factor out variable not studied
L. M. Lye
DOE Course
24
DOE Course
25
DOE Course
26
DOE Course
27
L. M. Lye
DOE Course
28
DOE (II)
Factorial vs OFAT
L. M. Lye
DOE Course
29
L. M. Lye
DOE Course
30
DOE Course
31
Factorial Designs
In a factorial experiment, all
possible combinations of
factor levels are tested
The golf experiment:
L. M. Lye
DOE Course
32
Factorial Design
L. M. Lye
DOE Course
33
L. M. Lye
DOE Course
34
DOE Course
35
L. M. Lye
DOE Course
36
DOE Course
37
OFAT
2 factors: 4 runs
2 factors: 6 runs
3 effects
2 effects
3 factors: 8 runs
3 factors: 16 runs
7 effects
3 effects
5 factors: 32 or 16 runs
5 factors: 96 runs
31 or 15 effects
5 effects
127 or 63 effects
L. M. Lye
7 effects
DOE Course
38
OFAT
high
high
Factor B
low
low
low
high
low
high
A
Factor A
DOE Course
39
DOE Course
40
DE SI GN-E A SE Pl ot
Ln (f)
X = A : Rey nold 's #
Y = B : k /D
Des ign Po i nt s
B - 0. 000
B + 0.00 1
In te ra c ti o n G ra p h
- 3 .4 2 0 3 8
k /D
2
2
L n ( f)
- 3 .6 4 1 5 5
- 3 .8 6 2 7 2
2
- 4 .0 8 3 8 9
- 4 .3 0 5 0 7
4 .0 00
4 .5 00
5 .0 00
5 .5 00
6 .0 00
R e yn o l d 's #
L. M. Lye
DOE Course
41
DE S IGN -E A S E P lot
Ln (f)
X = A : R e yno ld 's #
Y = B : k /D
De sig n P oin ts
L n (f)
0 .00 1 0
0 .00 0 8
k /D
-3 . 5 6 7 8 3
-3 .8 6 2 7 2
-3 . 7 1 5 2 8
0 .00 0 6
-4 .0 1 0 1 7
0 .00 0 3
-4 . 1 5 7 6 2
0 .00 0 1
4 .0 0 0
4 .50 0
5 .00 0
5 .50 0
6 .0 0 0
R e yn o l d 's #
L. M. Lye
DOE Course
42
DES I GN-E A SE P l ot
Ln(f)
X = A : Reyn ol d's #
Y = B : k/ D
-3 . 4 2 0 3 8
-3 . 6 4 1 5 5
-3 . 8 6 2 7 2
L n (f)
-4 . 0 8 3 8 9
-4 . 3 0 5 0 7
0.0010
0.0008
0.0006
6.000
k/ D
5.500
0.0003
5.000
4.500
0.0001
L. M. Lye
4.000
DOE Course
R e y n o l d 's #
43
DOE Course
44
DESI GN-EXP ER T Pl ot
Log10(f )
X = A: RE
Y = B: k/ D
Design Poin ts
B- 0.00 0
B+ 0. 001
Inte ra c ti o n G ra p h
B : k /D
- 1 .4 9 5
L o g 1 0 ( f)
- 1 .5 6 7
- 1 .6 3 9
- 1 .7 1 2
- 1 .7 8 4
4 .2 9 3
4 .6 4 6
5 .0 0 0
5 .3 5 4
5 .7 0 7
A: R E
L. M. Lye
DOE Course
45
DE SI GN -E XP ERT Pl ot
L og10(f )
X = A:R E
Y = B : k /D
-1 . 5 5 4
-1 . 6 1 1
L o g 1 0 (f)
-1 . 6 6 8
-1 . 7 2 5
-1 . 7 8 3
0.0008828
0.0007414
. 0D0 0 6 0 0 0
B : 0k/
0.0004586
0.0003172
4.293
4.646
5.000
5.354
5.707
A: RE
L. M. Lye
DOE Course
46
L o g 1 0 (f)
0 .0 00 8 8 2 8
0 .0 00 7 4 1 4
B : k /D
-1 . 6 6 8
0 .0 00 6 0 0 0
-1 . 7 0 6
-1 . 5 9 2-1 . 6 3 0
-1 . 7 4 4
0 .0 00 4 5 8 6
0 .0 00 3 1 7 2
4 .2 9 3
4 .6 4 6
5 .0 00
5 .3 5 4
5 .7 0 7
A: R E
L. M. Lye
DOE Course
47
P re d i c te d vs . A c tu a l
- 1 .4 9 4
P r e d ic te d
- 1 .5 6 6
- 1 .6 3 9
- 1 .7 1 1
- 1 .7 8 3
- 1 .7 83
- 1 .7 1 1
- 1 .63 9
- 1 .5 66
- 1 .4 9 4
A c tu a l
L. M. Lye
DOE Course
48
DOE (III)
Basic Concepts
L. M. Lye
DOE Course
49
L. M. Lye
50
Modified Mortar
(Formulation 1) y1 j
Unmodified Mortar
(Formulation 2)
y2 j
16.85
17.50
16.40
17.63
17.21
18.25
16.35
18.00
16.52
17.86
17.04
17.75
16.96
18.22
17.15
17.90
16.59
17.96
10
16.57
18.15
L. M. Lye
DOE Course
51
17.3
16.3
Form 1
L. M. Lye
Form 2
DOE Course
52
Box Plots
Boxplots of Form 1 and Form 2
(means are indicated by solid circles)
18.5
17.5
16.5
Form 1
L. M. Lye
Form 2
DOE Course
53
DOE Course
54
H1 : 1 2
L. M. Lye
DOE Course
55
Estimation of Parameters
1 n
y yi estimates the population mean
n i 1
n
1
2
2
S
( yi y ) estimates the variance
n 1 i 1
2
L. M. Lye
DOE Course
56
Summary Statistics
Formulation 1
Formulation 2
New recipe
Original recipe
y1 16.76
y2 17.92
S 0.100
S 0.061
S1 0.316
S2 0.247
2
2
2
1
n2 10
n1 10
L. M. Lye
DOE Course
57
2
n
This suggests a statistic:
2
y
Z0
L. M. Lye
y1 y2
12 22
n1 n2
DOE Course
58
2
2
2
1
2
2
y1 y2
2
1
2
2
S
S
n1 n2
2
2
1)
S
(
n
1)
S
2
1
2
2
Sp 1
n1 n2 2
L. M. Lye
DOE Course
59
n1 n2
Values of t0 that are near zero are consistent with the null
hypothesis
Values of t0 that are very different from zero are consistent
with the alternative hypothesis
t0 is a distance measure-how far apart the averages are
expressed in standard deviation units
Notice the interpretation of t0 as a signal-to-noise ratio
L. M. Lye
DOE Course
60
0.081
n1 n2 2
10 10 2
2
p
S p 0.284
y1 y2
16.76 17.92
t0
9.13
1 1
1 1
Sp
0.284
n1 n2
10 10
The two sample means are about 9 standard deviations apart
Is this a "large" difference?
L. M. Lye
DOE Course
61
L. M. Lye
DOE Course
62
DOE Course
63
DOE Course
64
Mean
StDev
SE Mean
Form 1
10
16.764
0.316
0.10
Form 2
10
17.922
0.248
0.078
-1.158
DOE Course
65
Checking Assumptions
The Normal Probability Plot
Tension Bond Strength Data
ML Estimates
Form 1
99
Form 2
Goodness of Fit
95
AD*
90
1.209
1.387
Percent
80
70
60
50
40
30
20
10
5
1
16.5
17.5
18.5
Data
L. M. Lye
DOE Course
66
DOE Course
67
L. M. Lye
DOE Course
68
An Example
Consider an investigation into the formulation of a
new synthetic fiber that will be used to make ropes
The response variable is tensile strength
The experimenter wants to determine the best level
of cotton (in wt %) to combine with the synthetics
Cotton content can vary between 10 40 wt %; some
non-linearity in the response is anticipated
The experimenter chooses 5 levels of cotton
content; 15, 20, 25, 30, and 35 wt %
The experiment is replicated 5 times runs made in
random order
L. M. Lye
DOE Course
69
An Example
DOE Course
70
L. M. Lye
DOE Course
71
DOE Course
72
DOE Course
73
2
2
(
y
y
)
[(
y
y
)
(
y
y
)]
ij .. i. .. ij i.
i 1 j 1
i 1 j 1
a
i 1 j 1
SST SSTreatments SS E
L. M. Lye
DOE Course
74
SST SSTreatments SS E
A large value of SSTreatments reflects large differences in
treatment means
A small value of SSTreatments likely indicates no differences in
treatment means
Formal statistical hypotheses are:
H 0 : 1 2 L a
H1 : At least one mean is different
L. M. Lye
DOE Course
75
If the treatment means are equal, the treatment and error mean squares will be (theoretically) equal.
If treatment means differ, the treatment mean square will be larger than the error mean square.
an 1 a 1 a (n 1)
MSTreatments
L. M. Lye
SSTreatments
SS E
, MS E
a 1
a (n 1)
DOE Course
76
F0 F ,a 1,a ( n 1)
L. M. Lye
DOE Course
77
R-Squared
Adj R-Squared
Pred R-Squared
Adeq Precision
DOE Course
0.7469
0.6963
0.6046
9.294
78
L. M. Lye
DOE Course
79
DE S IGN-E X P ERT P lo t
St r
e ngt h
X = A: Co t on We ight %
Desig n P oi nts
O n e F a c to r P lo t
25
S tr e n g th
2 0.5
2
2
16
1 1.5
15
20
25
30
35
A : C o tto n W e i g h t %
L. M. Lye
DOE Course
80
L. M. Lye
DOE Course
81
Examination of residuals
Design-Expert generates
the residuals
Residual plots are very
useful
Normal probability plot
of residuals
N o rm a l p lo t o f re s i d u a ls
99
N o r m a l % p r o b a b i l i ty
95
90
80
70
50
30
20
10
5
1
- 3 .8
- 1 .5 5
0 .7
2 .9 5
5 .2
R e s id u a l
L. M. Lye
DOE Course
82
R e s id ua ls vs . R un
R e s id ua ls vs . P re d ic te d
5.2
5.2
2.95
2.95
R e s id u a ls
R e s id u a ls
2
0.7
0.7
- 1.55
- 1.55
2
2
2
- 3.8
- 3.8
9.80
12.75
15.70
18.65
21.60
10
13
16
19
22
25
P re d ic te d
L. M. Lye
DOE Course
83
DOE Course
84
Design-Expert Output
Treatment Means (Adjusted, If Necessary)
Estimated
Standard
Mean
Error
1-15
9.80
1.27
2-20
15.40
1.27
3-25
17.60
1.27
4-30
21.60
1.27
5-35
10.80
1.27
Mean
Treatment Difference
1 vs 2
-5.60
1 vs 3
-7.80
1 vs 4
-11.80
1 vs 5
-1.00
2 vs 3
-2.20
2 vs 4
-6.20
2 vs 5
4.60
3 vs 4
-4.00
3 vs 5
6.80
4 vs 5
10.80
L. M. Lye
DF
1
1
1
1
1
1
1
1
1
1
Standard
Error
1.80
1.80
1.80
1.80
1.80
1.80
1.80
1.80
1.80
1.80
DOE Course
t for H0
Coeff=0
-3.12
-4.34
-6.57
-0.56
-1.23
-3.45
2.56
-2.23
3.79
6.01
85
VIF
9.03
1.00
9.03
86
DES IGN-EX PE RT Pl ot
S trengt h
X = A: Cot ton We ight %
Desi gn P oi nt s
2
2
S tr e n g th
20.5
16
11.5
15.00
L. M. Lye
DOE Course
20.00
25.00
30.00
A : C o tto n W e ig h t %
35.00
87
DE S IGN-EX P ERT P lo t
De si ra bi li t y
X = A: A
Desi gn Po in ts
O n e F a c to r P lo t
1 .0 0 0
Pr e d ic t 0 .7 7 2 5
X
2 8 .2 3
D e s ir a b ili ty
0 .7 5 0 0
5
5
0 .5 0 0 0
0 .2 5 0 0
6
6
0 .0 0 0 0
1 5 .0 0
2 0 .0 0
2 5 .0 0
3 0 .0 0
3 5 .0 0
A: A
L. M. Lye
DOE Course
88
L. M. Lye
DOE Course
89
L. M. Lye
DOE Course
90
DOE (IV)
General Factorials
L. M. Lye
DOE Course
91
L. M. Lye
DOE Course
92
40 52 20 30
21
2
2
30 52 20 40
B yB yB
11
2
2
52 20 30 40
AB
DOE Course 1
2
2
A y A y A
L. M. Lye
93
50 12 20 40
A y A y A
1
2
2
40 12 20 50
B yB yB
9
2
2
12 20 40 50
AB
29
2
2
L. M. Lye
DOE Course
94
y 0 1 x1 2 x2
12 x1 x2
The least squares fit is
y 35.5 10.5 x1 5.5 x2
0.5 x1 x2
35.5 10.5 x1 5.5 x2
L. M. Lye
DOE Course
95
L. M. Lye
DOE Course
96
2.
L. M. Lye
DOE Course
97
DOE Course
98
i 1, 2,..., a
DOE Course
99
i 1
j 1
2
2
2
(
y
y
)
bn
(
y
y
)
an
(
y
y
)
ijk ...
i.. ...
. j. ...
i 1 j 1 k 1
i 1 j 1
i 1 j 1 k 1
SST SS A SS B SS AB SS E
df breakdown:
abn 1 a 1 b 1 (a 1)(b 1) ab(n 1)
L. M. Lye
DOE Course
100
DOE Course
101
Design-Expert Output
Response:
Life
ANOVA for Selected Factorial Model
Analysis of variance table [Partial sum of squares]
Source
Model
A
B
AB
Pure E
C Total
L. M. Lye
Sum of
Squares
59416.22
10683.72
39118.72
9613.78
18230.75
77646.97
DF
8
2
2
4
27
35
Mean
F
Square Value
7427.03 11.00
5341.86
7.91
19559.36 28.97
2403.44
3.56
675.21
R-Squared
Adj R-Squared
Pred R-Squared
0.7652
0.6956
0.5826
PRESS
Adeq Precision
8.178
32410.22
DOE Course
Prob > F
< 0.0001
0.0020
< 0.0001
0.0186
102
Residual Analysis
DESIGN-EXPERT Plo t
Life
DE SIG N -EXP ER T Pl o t
L i fe
N o rm a l p lo t o f re s id ua ls
R e s id ua ls vs . P re d ic te d
45.25
99
18.75
90
80
R e s id u a ls
N o rm a l % p ro b a b ility
95
70
50
30
- 7.75
20
10
- 34.25
5
1
- 60.75
49.50
- 60.75
- 34.25
- 7.75
18.75
102.62
129.19
155.75
P re d ic te d
R e s id u a l
L. M. Lye
76.06
45.25
DOE Course
103
Residual Analysis
DESIG N-EXPERT Plo t
Li fe
R e s id ua ls vs . R un
45.25
R e s id u a ls
18.75
- 7.75
- 34.25
- 60.75
11
16
21
26
31
36
L. M. Lye
DOE Course
104
Residual Analysis
D ES IG N-E XP ER T P lo t
DESIGN-EXPERT Pl ot
Life
L i fe
R e s id ua ls vs . T e m p e ra ture
45.25
45.25
18.75
18.75
R e s id u a ls
R e s id u a ls
R e s id ua ls vs . M a te ria l
- 7.75
- 34.25
- 34.25
- 60.75
- 60.75
T e m p e ra tu re
M a te ria l
L. M. Lye
- 7.75
DOE Course
105
Interaction Plot
Inte ra c tio n G ra p h
A : M a te ria l
188
L ife
146
104
2
2
62
20
15
70
125
B : T e m p e ra tu re
L. M. Lye
DOE Course
106
DOE Course
107
4.009E+005
Linear
2FI
Quadratic
Cubic
Residual
Total
49726.39
3
2315.08
2
76.06
1
7298.69
2
18230.75 27
4.785E+005 36
F
Value
Prob > F
19.00
1.36
0.086
5.40
< 0.0001
0.2730
0.7709
0.0106
4009E+005
16575.46
1157.54
76.06
3649.35
675.21
13292.97
Suggested
Aliased
"Sequential Model Sum of Squares": Select the highest order polynomial where the
additional terms are significant.
L. M. Lye
DOE Course
108
DOE Course
Candidate model
terms from DesignExpert:
Intercept
A
B
B2
AB
B3
AB2
109
Source
Linear
Sum of
Squares
9689.83
DF
Mean
Square
F
Value
Prob > F
1937.97 2.87
0.0333
2FI
7374.75 3
Quadratic 7298.69 2
Cubic
0.00 0
Pure Error 18230.75 27
2458.25 3.64
3649.35 5.40
0.0252
0.0106
Suggested
Aliased
675.21
"Lack of Fit Tests": Want the selected model to have insignificant lack-of-fit.
L. M. Lye
DOE Course
110
Source
Std.
Dev.
Adjusted
Predicted
R-Squared R-Squared R-Squared PRESS
Linear
29.54
0.6404
0.6067
0.5432
35470.60 Suggested
2FI
29.22
Quadratic 29.67
Cubic
25.98
0.6702
0.6712
0.7652
0.6153
0.6032
0.6956
0.5187
0.4900
0.5826
37371.08
39600.97
32410.22 Aliased
"Model Summary Statistics": Focus on the model maximizing the "Adjusted RSquared"
and the "Predicted R-Squared".
L. M. Lye
DOE Course
111
Life
L. M. Lye
Mean
F
Square Value
7427.03 11.00
5341.86 7.91
39042.67 57.82
76.06
0.11
1157.54
1.71
3649.35
5.40
675.21
Prob > F
< 0.0001
0.0020
< 0.0001
0.7398
0.1991
0.0106
R-Squared
Adj R-Squared
Pred R-Squared
0.7652
0.6956
0.5826
PRESS
Adeq Precision
8.178
32410.22
DOE Course
112
L. M. Lye
DOE Course
113
X = B: Temperat ure
Y = A: Materi al
A1 A1
A2 A2
A3 A3
Inte ra c tio n G ra p h
A : M a te ria l
188
L ife
146
104
2
2
62
20
15.00
L. M. Lye
42.50
70.00
97.50
B : T e m p e ra tu re
DOE Course
125.00
114
SST SS A SS B L SS AB SS AC L
SS ABC L SS ABL K SS E
L. M. Lye
DOE Course
115
DOE Course
116