Walter Castellon CpE & EE Mohammad Amori CpE Josh Steele CpE Tri Tran CpE.
CPE 619 2 k-p Factorial Design
description
Transcript of CPE 619 2 k-p Factorial Design
![Page 1: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/1.jpg)
CPE 6192k-p Factorial Design
Aleksandar Milenković
The LaCASA LaboratoryElectrical and Computer Engineering Department
The University of Alabama in Huntsvillehttp://www.ece.uah.edu/~milenkahttp://www.ece.uah.edu/~lacasa
![Page 2: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/2.jpg)
2
PART IV: Experimental Design and Analysis
How to: Design a proper set of experiments
for measurement or simulation Develop a model that best describes
the data obtained Estimate the contribution of each alternative to the
performance Isolate the measurement errors Estimate confidence intervals for model parameters Check if the alternatives are significantly different Check if the model is adequate
![Page 3: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/3.jpg)
3
Introduction
2k-p Fractional Factorial Designs Sign Table for a 2k-p Design Confounding Other Fractional Factorial Designs Algebra of Confounding Design Resolution
![Page 4: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/4.jpg)
4
2k-p Fractional Factorial Designs
Large number of factors ) large number of experiments ) full factorial design too expensive ) Use a fractional factorial design
2k-p design allows analyzing k factors with only 2k-p experiments 2k-1 design requires only half as many experiments 2k-2 design requires only one quarter of the
experiments
![Page 5: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/5.jpg)
5
Example: 27-4 Design
Study 7 factors with only 8 experiments!
![Page 6: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/6.jpg)
6
Fractional Design Features Full factorial design is easy to analyze due to
orthogonality of sign vectors Fractional factorial designs also use orthogonal vectors That is
The sum of each column is zeroi xij =0 8 j
jth variable, ith experiment The sum of the products of any two columns is zero
i xijxil=0 8 j l The sum of the squares of each column is 27-4, that is, 8
i xij2 = 8 8 j
![Page 7: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/7.jpg)
7
Analysis of Fractional Factorial Designs
Model:
Effects can be computed using inner products
![Page 8: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/8.jpg)
8
Example 19.1
Factors A through G explain 37.26%, 4.74%, 43.40%, 6.75%, 0%, 8.06%, and 0.03% of variation, respectively Use only factors C and A for further experimentation
![Page 9: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/9.jpg)
9
Sign Table for a 2k-p Design
Steps:1. Prepare a sign table for a full factorial design with
k-p factors2. Mark the first column I3. Mark the next k-p columns with the k-p factors4. Of the (2k-p-k-p-1) columns on the right, choose p
columns and mark them with the p factors which were not chosen in step 1
![Page 10: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/10.jpg)
10
Example: 27-4 Design
![Page 11: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/11.jpg)
11
Example: 24-1 Design
![Page 12: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/12.jpg)
12
Confounding Confounding: Only the combined influence of two or more
effects can be computed.
![Page 13: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/13.jpg)
13
Confounding (cont’d)
) Effects of D and ABC are confounded. Not a problem if qABC is negligible.
![Page 14: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/14.jpg)
14
Confounding (cont’d) Confounding representation: D=ABC
Other Confoundings:
I=ABCD ) confounding of ABCD with the mean
![Page 15: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/15.jpg)
15
Other Fractional Factorial Designs A fractional factorial design is not unique. 2p different designs
Confoundings:
Not as good as the previous design
![Page 16: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/16.jpg)
16
Algebra of Confounding Given just one confounding, it is possible to list all other
confoundings Rules:
I is treated as unity. Any term with a power of 2 is erased.
Multiplying both sides by A:
![Page 17: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/17.jpg)
17
Algebra of Confounding (cont’d)
Multiplying both sides by B, C, D, and AB:
and so on. Generator polynomial: I=ABCDFor the second design: I=ABC.
In a 2k-p design, 2p effects are confounded together
![Page 18: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/18.jpg)
18
Example 19.7 In the 27-4 design:
Using products of all subsets:
![Page 19: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/19.jpg)
19
Example 19.7 (cont’d)
Other confoundings:
![Page 20: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/20.jpg)
20
Design Resolution
Order of an effect = Number of termsOrder of ABCD = 4, order of I = 0.
Order of a confounding = Sum of order of two termsE.g., AB=CDE is of order 5.
Resolution of a Design = Minimum of orders of confoundings
Notation: RIII = Resolution-III = 2k-pIII
Example 1: I=ABCD RIV = Resolution-IV = 24-1IV
![Page 21: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/21.jpg)
21
Design Resolution (cont’d) Example 2:
I = ABD RIII design. Example 3:
This is a resolution-III design A design of higher resolution is considered a better design
![Page 22: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/22.jpg)
22
Case Study 19.1: Latex vs. troff
![Page 23: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/23.jpg)
23
Case Study 19.1 (cont’d)
Design: 26-1 with I=BCDEF
![Page 24: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/24.jpg)
24
Case Study 19.1: Conclusions Over 90% of the variation is due to: Bytes, Program, and
Equations and a second order interaction Text file size were significantly different making it's effect
more than that of the programs High percentage of variation explained by the ``program £
Equation'' interaction Choice of the text formatting program depends upon the number of equations in the text. troff not as good for equations
![Page 25: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/25.jpg)
25
Case Study 19.1: Conclusions (cont’d) Low ``Program £ Bytes'' interaction ) Changing the file size
affects both programs in a similar manner. In next phase, reduce range of file sizes. Alternately, increase
the number of levels of file sizes.
![Page 26: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/26.jpg)
26
Case Study 19.2: Scheduler Design Three classes of jobs: word processing, data processing, and
background data processing.
Design: 25-1 with I=ABCDE
![Page 27: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/27.jpg)
27
Measured Throughputs
![Page 28: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/28.jpg)
28
Effects and Variation Explained
![Page 29: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/29.jpg)
29
Case Study 19.2: Conclusions For word processing throughput (TW): A (Preemption), B (Time
slice), and AB are important. For interactive jobs: E (Fairness), A (preemption), BE, and B
(time slice). For background jobs: A (Preemption), AB, B (Time slice), E
(Fairness). May use different policies for different classes of workloads. Factor C (queue assignment) or any of its interaction do not
have any significant impact on the throughput. Factor D (Requiring) is not effective. Preemption (A) impacts all workloads significantly. Time slice (B) impacts less than preemption. Fairness (E) is important for interactive jobs and slightly
important for background jobs.
![Page 30: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/30.jpg)
30
Summary
Fractional factorial designs allow a large number of variables to be analyzed with a small number of experiments
Many effects and interactions are confounded The resolution of a design is the sum of the order of
confounded effects A design with higher resolution is considered better
![Page 31: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/31.jpg)
31
Exercise 19.1
Analyze the 24-1 design:
Quantify all main effects Quantify percentages of variation explained Sort the variables in the order of decreasing importance List all confoundings Can you propose a better design with the same number of
experiments What is the resolution of the design?
![Page 32: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/32.jpg)
32
Exercise 19.2
Is it possible to have a 24-1III design? a 24-1
II design? 24-1IV
design? If yes, give an example.
![Page 33: CPE 619 2 k-p Factorial Design](https://reader035.fdocuments.us/reader035/viewer/2022062302/5681681b550346895ddda7d5/html5/thumbnails/33.jpg)
33
Homework Updated Exercise 19.1
Analyze the 24-1 design:
Quantify all main effects. Quantify percentages of variation explained. Sort the variables in the order of decreasing importance. List all confoundings. Can you propose a better design with the same number of
experiments. What is the resolution of the design?