21.4.3.12. Two-Level Orthogonal Design
Two-level orthogonal design is called a \({{2}^{k-p}}\) fractional factorial design. These designs require the selection of p independent generators. Although the Plackett-Burman design is more efficient than two-level orthogonal design, it can be inefficient for some cases such as 15-factors. It causes from the failure of balance.
AutoDesign provides the strength-II orthogonal design such as \({{2}^{k-p}}\) fractional factorial design, which generates the table automatically. Table 21.10 lists the number of trials for available factors.
Trials |
Factors |
Trials |
Factors |
4 |
2 - 3 |
128 |
64 - 127 |
8 |
4 - 7 |
256 |
128 - 255 |
16 |
8 - 15 |
512 |
256 - 511 |
32 |
16 - 31 |
1024 |
512 - 1023 |
64 |
32 - 63 |
Reference
Peter W.M. John, 1998, Statistical Design and Analysis of Experiments, SIAM, Philadelphia.
Douglas C Montgomery, 2000, Design and Analysis of Experiments, John Wiley & Sons, New York.