21.4.3.13. Bose’s Orthogonal Design
Even though the level-balanced descriptive design is efficient, it gives an aliased result in the effect analysis. As explained in Level Balanced Descriptive Design, it is the limitation of the strength-I orthogonal design. For higher level design, AutoDesign provides the Bose’s orthogonal design that is a strength-II design. This design gives a \({{q}^{2}}\) design for the \(q+1\) factors and \(q\) levels that must be a prime number. Table 21.11 lists the number of trials for available factors.
Trials |
Level |
Factors |
4 |
2 |
2-3 |
9 |
3 |
2-4 |
25 |
5 |
2-6 |
49 |
7 |
2-8 |
121 |
11 |
2-12 |
169 |
13 |
2-14 |
In the above table, the trial number 25 is well known as Taguchi’s \({{L}_{25}}\) design.
Reference
Hedayat, A.S., Sloane, N.J.A. and Stufken, J., Orthogonal Arrays: Theory and Applications, Springer, Newyork, 1999.