chapter 8 design and analysis of experiments 1 8e 2012
TRANSCRIPT
![Page 1: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/1.jpg)
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
1
![Page 2: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/2.jpg)
The One-Quarter Fraction of the 2k
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
2
![Page 3: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/3.jpg)
The One-Quarter Fraction of the 26-2
Complete defining relation: I = ABCE = BCDF = ADEF
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
3
![Page 4: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/4.jpg)
The One-Quarter Fraction of the 26-2
• Uses of the alternate fractions
• Projection of the design into subsets of the original six variables
, E ABC F BCD= ± = ±
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
4
original six variables
• Any subset of the original six variables that is not a word in the complete defining relation will result in a full factorial design
– Consider ABCD (full factorial)
– Consider ABCE (replicated half fraction)
– Consider ABCF (full factorial)
![Page 5: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/5.jpg)
A One-Quarter Fraction of the 26-2:
Example 8.4, Page 336
• Parts manufactured in an injection molding process are showing
excessive shrinkage.
• Design Factors:
– mold temperature (A),
– screw speed (B), –– screw speed (B),
– holding time (C),
– cycle time (D),
– gate size (E), and
– holding pressure (F)
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
5
![Page 6: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/6.jpg)
A One-Quarter Fraction of the 26-2:
Example 8.4, Page 336
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
6
![Page 7: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/7.jpg)
A One-Quarter Fraction of the 26-2:
Example 8.4, Page 336
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
7
![Page 8: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/8.jpg)
A One-Quarter Fraction of the 26-2:
Example 8.4, Page 336
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
8
![Page 9: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/9.jpg)
The General 2k-p Fractional
Factorial Design
• Section 8.4, page 340
• 2k-1 = one-half fraction, 2k-2 = one-quarter fraction, 2k-3 = one-eighth fraction, …, 2k-p = 1/ 2p fraction
• Add p columns to the basic design; select p
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
9
• Add p columns to the basic design; select pindependent generators
• Important to select generators so as to maximizeresolution, see Table 8.14
• Projection – a design of resolution R contains full factorials in any R – 1 of the factors
![Page 10: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/10.jpg)
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
10
![Page 11: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/11.jpg)
The General 2k-p Design: Resolution may
not be Sufficient
• Minimum aberration designs
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
11
Minimum aberration: minimum number of words in the
defining relation that are of minimum length.
![Page 12: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/12.jpg)
Resolution III Designs: Section 8.5,
page 351
• Designs with main effects aliased with two-factor
interactions
• Used for screening (5 – 7 variables in 8 runs, 9 - 15
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
12
• Used for screening (5 – 7 variables in 8 runs, 9 - 15
variables in 16 runs)
• A saturated design has k = N – 1 variables
• See Table 8.19, page 351 for a 7 42III
−
![Page 13: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/13.jpg)
Resolution III Designs
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
13
![Page 14: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/14.jpg)
Resolution III Designs
• Sequential assembly of fractions to separate aliased effects (page 354)
• Switching the signs in one column provides estimates of that factor and all of its two-factor interactions
• Switching the signs in all columns dealiases all main effects from their two-factor interaction alias chains –called a full fold-over
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
14
called a full fold-over
• Defining relation for a fold-over (page 356)
• Be careful – these rules only work for Resolution III designs
• There are other rules for Resolution IV designs, and other methods for adding runs to fractions to dealias effects of interest
• Example 8.7, eye focus time, page 354
![Page 15: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/15.jpg)
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
15
![Page 16: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/16.jpg)
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
16
![Page 17: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/17.jpg)
Remember that the full fold-over technique
illustrated in this example (running a “mirror
image” design with all signs reversed) only
works in a Resolution III design.
Defining relation for a fold-over design – see
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
17
Defining relation for a fold-over design – see
page 356.
![Page 18: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/18.jpg)
Resolution IV and V Designs (Page 366)
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
18
A resolution IV design must have at least 2k runs.
![Page 19: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/19.jpg)
Sequential Experimentation with Resolution
IV Designs – Page 367
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
19
We can’t use the full fold-over procedure given previously for
Resolution III designs – it will result in replicating the runs in the
original design.
Switching the signs in a single column allows all of the two-factor
interactions involving that column to be separated.
![Page 20: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/20.jpg)
The spin coater experiment – page 368
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
20
![Page 21: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/21.jpg)
[AB] = AB + CE
We need to dealias these
interactions
The fold-over design
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
21
The fold-over design
switches the signs in
column A
![Page 22: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/22.jpg)
The aliases from the complete design following the fold-
over (32 runs) are as follows:
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
22
Finding the aliases involves using the alias matrix.
Aliases can also be found from computer software.
![Page 23: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/23.jpg)
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
23
![Page 24: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/24.jpg)
A full fold-over of a Resolution IV design is usually not necessary,
and it’s potentially very inefficient.
In the spin coater example, there were seven degrees of freedom
available to estimate two-factor interaction alias chains.
After adding the fold-over (16 more runs), there are only 12
degrees of freedom available for estimating two-factor interactions
(16 new runs yields only five more degrees of freedom).
A partial fold-over (semifold) may be a better choice of follow-up
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
24
A partial fold-over (semifold) may be a better choice of follow-up
design. To construct a partial fold-over:
![Page 25: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/25.jpg)
Not an
orthogonal
design – but
that’s not
such a big
deal
Correlated
parameter
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
25
parameter
estimates
Larger
standard
errors of
regression
model
coefficients
or effects
![Page 26: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/26.jpg)
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
26
There are still 12
degrees of freedom
available to estimate
two-factor interactions
![Page 27: Chapter 8 Design and Analysis of Experiments 1 8E 2012](https://reader031.vdocuments.net/reader031/viewer/2022020707/61ff9254c2136b2a6a2efb5c/html5/thumbnails/27.jpg)
Resolution V Designs – Page 373
We used a Resolution V design (a 25-2) in Example 8.2
Generally, these are large designs (at least 32 runs) for six or
more factors
Non-regular designs can be found using optimal design
construction methods
Chapter 8 Design and Analysis of Experiments
8E 2012 Montgomery
27
construction methods
JMP has excellent capability
Examples for k = 6 and 8 factors are illustrated in the book