Professional Documents
Culture Documents
Chapter 6 - Part 1
Chapter 6 - Part 1
of Experiments
Chapter 6
Note:
(1): treatment at A- and B-
a: treatment at A+ and B-
b: treatment at B+ and A-
ab: treatment at A+ and B+
Analysis Procedure for a
Factorial Design
• Estimate factor effects
• Formulate model
– With replication, use full model
– With an unreplicated design, use normal
probability plots
• Statistical testing (ANOVA)
• Refine the model
• Analyze residuals (graphical)
• Interpret results
Estimation of Factor Effects
Main effect of A
ab a b (1)
A y A y A
2n 2n
1
[ab a b (1)]
2n contrast
A
Main effect of B
ab b a (1)
B yB yB
2n 2n
1
[ab b a (1)]
2n contrast
B The effect estimates are:
Interaction effect: A = 8.33, B = -5.00, AB = 1.67
Practical interpretation?
ab (1) a b 1
AB [ ab (1) a b]
2n 2n 2n contrast
AB
Determining the sums of squares
Sum of square of A :
1
SS A [ab a b (1)]2
4n contrast
A
Sum of square of B :
1
SS B [ab b a (1)]2
4n contrast
B
The F-test for the “model” source is testing the significance of the
overall model; that is, is either A, B, or AB or some combination of
these effects important?
Regression Model
Regression model :
y 0 1 x1 2 x2 12 x1 x2 ,
8.33
1 (main effect of A) /2 =
2
5
2 (main effect of B) /2=
2
1.67
12 (interaction AB) / 2
2
0 ((1) + a + b + ab) / 4n
x1 , x2 are scaled in [-1 +1] [low high] 8.33 5
yˆ 27.5 x1 x2
2 2
11
Regression Model for the Process
y =
27.5
4.166667 * A
-2.5 * B
y =
18.33333
0.833333 * Concentration
-5 * Catalyst
Residuals and Diagnostic
Checking
The Response Surface
The 23 Factorial Design
Effects in The 23 Factorial Design
A y A y A
B yB yB
C yC yC
etc, etc, ...
Example 6.1: wafers manufacturing
1
Effect of A [(1) a b ab c ac bc abc]
4n
1
Effect of B [(1) a b ab c ac bc abc], n :number of replicates
4n
Properties of the Table
• Except for column I, every column has an equal number of +
and – signs
• The sum of the product of signs in any two columns is zero
• Multiplying any column by I leaves that column unchanged
(identity element)
• The product of any two columns yields a column in the table:
A B AB
AB BC AB 2C AC
• Orthogonal design
• Orthogonality is an important property shared by all factorial
designs
Estimation of Factor Effects
ANOVA Summary – Full Model
Model Coefficients – Full Model
Refine Model – Remove Nonsignificant
Factors
Model Coefficients – Reduced
Model
The Regression Model
Response surface and contour plot
Model Summary Statistics for
Reduced Model
Model Summary Statistics for
Reduced Model
• R2 and adjusted R2 (Đánh giá độ chính xác của mô
hình)
SS 5.106 10 5
R 2 Model 0.9608
SST 5.314 10 5
• PRESS 37080.44
2R2 for prediction (based on PRESS: PRediction Error Sum
R Pred 1 1 0.9302
SS T 5.314 10
of Squares) (Đánh giá độ chính xác dự5 đoán giá trị mới)
The General 2k Factorial Design
The General 2k Factorial Design
k k k 1 k i 1
i i i 1 1
Thực hành chương 6