Conducting Simple Factorial Designs

Definition and Principles

 Many experiments involve the study of the

effects of two or more factors. Factorial designs
are most efficient for this type of experiment.
 In a factorial design, all possible combinations of
the levels of the factors are investigated in each
replication.
 If there are a levels of factor A, and b levels of
factor B, then each replicate contains all ab
treatment combinations.

1
Types of Factorial Designs

Types of Factorial Designs

FD
FD

Crossed Nested
Fixed Random Mixed
(FD) (ND)

FD

Full Fractional

2
Example: Maximizing Adhesion of Primer Paint

A manufacturer uses primer to improve the

adhesion of automobile paint. The process
improvement team wants to determine the primer
type and spray pressure level that maximize the
adhesion

Data Collection

The team decides to run full factorial experiment to test

two levels of spray pressure on each primer type.
Each combination of primer type and spray pressure is
replicated four times.
Operators apply primer to an aluminum surface using a
designated spray pressure. After they paint each
specimen, they apply a finish paint and measure the
adhesive force.
Variable Description
Pressure Pressure of the primer supplied the nozzle
(310, 380 kPa)
PrimeType Formulation of the primer (One, Two)
Adhesion Force required to pull the paint from the base
metal (Mpa)

3
 Tasks

1. Create Factorial Design

2. Display Design
3. Key in the data

4.52 4.55 5.05 4.88 4.48 4.37 4.89 4.75 4.29 4.43 5.14 4.91 4.60 4.25 4.95 4.71

4. Analyze Factorial Design

5. Check residual plots
6. Generate Factorial Plots, Cube Plot to find best
setting
7. Use Response Optimizer to maximize the adhesion

Exercise 1
A manufacturer of package cake mixes must determine the
recommended baking time and temperature to print on the
box. They want to find the factor settings that result in a
mean moisture content of 26%±1%.

Food scientists replicate a full factorial design twice. They

bake all cakes in a single oven in a single day, so no
blocking variable is required. They bake the cakes for 30 or
50 minutes, at a temperature of 325 ºF or 375 ºF. The
scientists calculate moisture content by measuring the mass
of the material before and after baking.

Dataset: Cake.MTW

4
 Exercise 1: Instruction
1. Create a full FD with 2 factors, time and temperature.
Replicate the design twice.
2. Open the Project Cake.MTW, which contains the response
data
3. Fit an appropriate model to the data
4. Check the model assumptions using residual plots
5. Create factorial plots and use the Response Optimizer to
determine the best settings for time and temperature to
obtain the target moisture content of 26%, with a lower limit
of 25% and upper limit of 27%.

Exercise 2 (Optional)
The yield of a chemical process is being studied. The two most important
variables are thought to be the pressure and the temperature. Three levels of
each factor are selected, and a factorial experiment with two replicates is
performed. The yield data are as follow:

Pressure
Temperature 200 215 230
90.4 90.7 90.2
150
90.2 90.6 90.4
90.1 90.5 89.9
160
90.3 90.6 90.1
90.5 90.8 90.4
170
90.7 90.9 90.1

Determine the best combination of Temperature and Pressure that maximizes

the yield
Verify the assumptions using residual plots