FieldsEducation
Experiments here usually have about 3 factors and 14 runs. Most are full factorials.
- Full factorial
- Fractional factorial or screening
- Central composite or Box-Behnken
Show the numbers
| Kind of design | Experiments | Factors | Runs |
|---|---|---|---|
| Full factorial | 6 | 3 | 11 |
| Fractional factorial or screening | 2 | 5.5 | 14 |
| Central composite or Box-Behnken | 2 | 3 | 23 |
Factors and runs are the medians of each kind.
Full factorial
Every combination of the factor levels.
- An after-school program's enrollment: class length L and program weeks N, and a fifth run outside the cubeDunn (2015) · 2 factors, 5 runs
- Marlissa's accounting sections: delivery D, attempts N and practice quiz P, eight runs and two moreDunn (2015) · 3 factors, 10 runs
- Four spring toys, two inclines and a child or adult operator: an unreplicated 4 x 2 x 2 general factorialAnderson and Whitcomb (2007) · 3 factors, 16 runs
- Tabletop hockey: puck weight, stick length, wind-up and puck place in a 2^4, the distance analysed in logsAnderson and Whitcomb (2007) · 4 factors, 16 runs
- Time for three spring toys to walk down a plank at two inclines: a 3 x 2 general factorial, replicatedAnderson and Whitcomb (2007) · 2 factors, 12 runs
Fractional factorial or screening
A planned part of the combinations, to find the factors that matter.
Central composite or Box-Behnken
Extra runs that fit curvature, to find an optimum.
- Flight time of paper helicopters against wing area, wing ratio, base width and base length in a central composite design (Box and Liu 1999) (Problem 11.34)Montgomery (2013) · 4 factors, 30 runs
- Flight time of paper confetti: a 2^2 with four centre points shows curvature, then augmented to a CCDAnderson and Whitcomb (2007) · 2 factors, 16 runs