stats chap8.1

stats chap8.1 - Chapter 8 Two-Level Fractional...

This preview shows pages 1–13. Sign up to view the full content.

Chapter 8 Two-Level Fractional Factorials (cont.)

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Complementary Half-Fraction : In Example 8-1, a 2 4-1 Fraction was constructed using the generator; I = ABCD This is called the principal fraction. An alternate (complementary) one-half can be constructed using I = - ABCD . This will consist of runs a , b , c , abc , d , abd , acd and bcd .
Alternate Alias Structure : [ ] [ ] on. . so and ACD B B BCD A A - - Thus, the two designs can be augmented (in blocks) to de-alias the main effects by adding and subtracting estimates from the two fractions.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Example 8-3 Consider the alternative fraction of Example 8-1, with filtration rates shown in p. 301
[ ] [ ] [ ] on. . so and ABD C C ACD B B BCD A A 75 . 5 4 23 75 . 4 4 19 25 . 24 4 97 = = - = = - = = -

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
These estimates are combined with those calculated from the principal fraction (see Table 8-4, p. 296) to obtain: [ ] [ ] [ ] [ ] 625 . 2 ) 25 . 24 0 . 19 ( 2 1 ) ( 2 1 625 . 21 ) 25 . 24 0 . 19 ( 2 1 ) ( 2 1 - = - = - = + = + BCD A A A A A and so on (see p. 301).
Half Normal plot Half Normal % probability |Effect| 0.00 5.41 10.81 16.22 21.63 0 20 40 60 70 80 85 90 95 97 99 A C D AC AD

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
The ANOVA Table : Source SS df MS F 0 P-value Blocks 7.56 1 7.56 Model 5535.81 5 1107.16 53.13 < 0.0001 A 1870.56 1 1870.56 89.76 < 0.0001 C 390.06 1 390.06 18.72 0.0019 D 855.56 1 855.56 41.05 0.0001 AC 1314.06 1 1314.06 63.05 < 0.0001 AD 1105.56 1 1105.56 53.05 < 0.0001 Error 187.56 9 20.84 Total 5730.94 15
Conclusion : Results of the alternative fraction offers a confirmation to our initial conclusions regarding the two-factor interaction effects.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Projection of Fractions : Every 2 k-1 fractional factorial design can be projected into: Full factorial in any (k-1) of the original k factors, Two replicates of a full factorial in any subset of (k-2) factors, Four replicates of a full factorial in any subset of (k-3) factors, and so on.
Projection of a 2 3-1 design: B A C a c b abc

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
The One-Quarter Fractional Factorial : These designs require two generators (P,Q) and their generalized interaction PQ. The
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 05/03/2010 for the course ME 250-750 taught by Professor Signer during the Summer '10 term at Wichita State.

Page1 / 42

stats chap8.1 - Chapter 8 Two-Level Fractional...

This preview shows document pages 1 - 13. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online