Design and Analysis of Experiments Dr. Tai-Yue Wang Department of Industrial and Information Management National Cheng Kung University Tainan, TAIWAN, ROC 1/33
Factorial Experiments Dr. Tai-Yue Wang Department of Industrial and Information Management National Cheng Kung University Tainan, TAIWAN, ROC 2/33
Outline Basic Definition and Principles The Advantages of Factorials The Two Factors Factorial Design The General Factorial Design Fitting Response Curve and Surfaces Blocking in Factorial Design 3
Basic Definitions and Principles Factorial Design—all of the possible combinations of factors’ level are investigated When factors are arranged in factorial design, they are said to be crossed Main effects – the effects of a factor is defined to be changed Interaction Effect – The effect that the difference in response between the levels of one factor is not the same at all levels of the other factors. 4
5 Basic Definitions and Principles Factorial Design without interaction
6 Basic Definitions and Principles Factorial Design with interaction
7 Basic Definitions and Principles Average response – the average value at one factor’s level Average response increase – the average value change for a factor from low level to high level No Interaction:
8 Basic Definitions and Principles With Interaction:
9 Basic Definitions and Principles Another way to look at interaction: When factors are quantitative In the above fitted regression model, factors are coded in (-1, +1) for low and high levels This is a least square estimates
10 Basic Definitions and Principles Since the interaction is small, we can ignore it. Next figure shows the response surface plot
11 Basic Definitions and Principles The case with interaction
12 Advantages of Factorial design Efficiency Necessary if interaction effects are presented The effects of a factor can be estimated at several levels of the other factors
13 The Two-factor Factorial Design Two factors a levels of factor A, b levels of factor B n replicates In total, nab combinations or experiments
14 The Two-factor Factorial Design – An example Two factors, each with three levels and four replicates 3 2 factorial design
15 The Two-factor Factorial Design – An example Questions to be answered: What effects do material type and temperature have on the life the battery Is there a choice of material that would give uniformly long life regardless of temperature?
16 Statistical (effects) model: means model The Two-factor Factorial Design
Hypothesis Row effects: Column effects: Interaction:
The Two-factor Factorial Design -- Statistical Analysis 18
The Two-factor Factorial Design -- Statistical Analysis 19 Mean square: A: B: Interaction:
The Two-factor Factorial Design -- Statistical Analysis 20 Mean square: Error:
The Two-factor Factorial Design -- Statistical Analysis 21 ANOVA table
The Two-factor Factorial Design -- Statistical Analysis 22 Example
The Two-factor Factorial Design -- Statistical Analysis 23 Example
The Two-factor Factorial Design -- Statistical Analysis 24 Example
The Two-factor Factorial Design -- Statistical Analysis 25 Example STAT ANOVA--GLM General Linear Model: Life versus Material, Temp Factor Type Levels Values Material fixed 3 1, 2, 3 Temp fixed 3 15, 70, 125 Analysis of Variance for Life, using Adjusted SS for Tests Source DF Seq SS Adj SS Adj MS F P Material Temp Material*Temp Error Total S = R-Sq = 76.52% R-Sq(adj) = 69.56% Unusual Observations for Life Obs Life Fit SE Fit Residual St Resid R R R denotes an observation with a large standardized residual.
The Two-factor Factorial Design -- Statistical Analysis 26 Example STAT ANOVA--GLM
The Two-factor Factorial Design -- Statistical Analysis 27 Example STAT ANOVA--GLM
The Two-factor Factorial Design -- Statistical Analysis 28 Estimating the model parameters
The Two-factor Factorial Design -- Statistical Analysis 29 Choice of sample size Row effects Column effects Interaction effects D:difference, :standard deviation
The Two-factor Factorial Design -- Statistical Analysis 30
The Two-factor Factorial Design -- Statistical Analysis 31 Appendix Chart V For n=4, giving D=40 on temperature, v 1 =2, v 2 =27, Φ 2 =1.28n. β =0.06 nΦ2Φ2 Φυ1υ1 υ2υ2 β
The Two-factor Factorial Design -- Statistical Analysis – example with no interaction 32 Analysis of Variance for Life, using Adjusted SS for Tests Source DF Seq SS Adj SS Adj MS F P Material Temp Error Total S = R-Sq = 64.14% R-Sq(adj) = 59.51%
The Two-factor Factorial Design – One observation per cell 33 Single replicate The effect model
The Two-factor Factorial Design – One observation per cell 34 ANOVA table
The Two-factor Factorial Design -- One observation per cell 35 The error variance is not estimable unless interaction effect is zero Needs Tuckey’s method to test if the interaction exists. Check page 183 for details.
The General Factorial Design 36 In general, there will be abc…n total observations if there are n replicates of the complete experiment. There are a levels for factor A, b levels of factor B, c levels of factor C,..so on. We must have at least two replicate (n ≧ 2) to include all the possible interactions in model.
The General Factorial Design 37 If all the factors are fixed, we may easily formulate and test hypotheses about the main effects and interaction effects using ANOVA. For example, the three factor analysis of variance model:
The General Factorial Design 38 ANOVA.
The General Factorial Design 39 where
The General Factorial Design -- example 40 Three factors: pressure, percent of carbonation, and line speed.
The General Factorial Design -- example 41 ANOVA
Fitting Response Curve and Surfaces 42 When factors are quantitative, one can fit a response curve (surface) to the levels of the factor so the experimenter can relate the response to the factors. These surface could be linear or quadratic. Linear regression model is generally used
Fitting Response Curve and Surfaces -- example 43 Battery life data Factor temperature is quantitative
44 Example STAT ANOVA—GLM Response life Model temp, material temp*temp, material*temp, material*temp*temp Covariates temp Fitting Response Curve and Surfaces -- example
45 General Linear Model: Life versus Material Factor Type Levels Values Material fixed 3 1, 2, 3 Analysis of Variance for Life, using Sequential SS for Tests Source DF Seq SS Adj SS Seq MS F P Temp Material Temp*Temp Material*Temp Material*Temp*Temp Error Total S = R-Sq = 76.52% R-Sq(adj) = 69.56% Term Coef SE Coef T P Constant Temp Temp*Temp Temp*Material Temp*Temp*Material Two kinds of coding methods: 1.1, 0, , 1, -1 coding method: -1, 0, +1 Fitting Response Curve and Surfaces -- example
46 Final regression equation: Fitting Response Curve and Surfaces -- example
47 Tool life Factors: cutting speed, total angle Data are coded Fitting Response Curve and Surfaces – example –3 2 factorial design
48 Fitting Response Curve and Surfaces – example –3 2 factorial design
49 Fitting Response Curve and Surfaces – example –3 2 factorial design Regression Analysis: Life versus Speed, Angle,... The regression equation is Life = Speed Angle Angle*Angle Speed*Speed Angle*Speed Angle*Speed*Speed Angle*Angle*Speed Angle*Angle*Speed*Speed Predictor Coef SE Coef T P Constant Speed Angle Angle*Angle Speed*Speed Angle*Speed Angle*Speed*Speed Angle*Angle*Speed Angle*Angle*Speed*Speed S = R-Sq = 89.5% R-Sq(adj) = 80.2%
50 Fitting Response Curve and Surfaces – example –3 2 factorial design Analysis of Variance Source DF SS MS F P Regression Residual Error Total Source DF Seq SS Speed Angle Angle*Angle Speed*Speed Angle*Speed Angle*Speed*Speed Angle*Angle*Speed Angle*Angle*Speed*Speed
51 Fitting Response Curve and Surfaces – example –3 2 factorial design
52 We may have a nuisance factor presented in a factorial design Original two factor factorial model: Blocking in a Factorial Design Two factor factorial design with a block factor model:
53 Blocking in a Factorial Design
54 Blocking in a Factorial Design - - example Response: intensity level Factors: Ground cutter and filter type Block factor: Operator
55 Blocking in a Factorial Design - - example General Linear Model: Intensity versus Clutter, Filter, Blocks Factor Type Levels Values Clutter fixed 3 High, Low, Medium Filter fixed 2 1, 2 Blocks fixed 4 1, 2, 3, 4 Analysis of Variance for Intensity, using Sequential SS for Tests Source DF Seq SS Adj SS Seq MS F P Clutter Filter Clutter*Filter Blocks Error Total S = R-Sq = 91.88% R-Sq(adj) = 87.55%
56 Blocking in a Factorial Design - - example General Linear Model: Intensity versus Clutter, Filter, Blocks Term Coef SE Coef T P Constant Clutter High Low Filter Clutter*Filter High Low Blocks