 |
|
D-Optimal designs are particularly useful when classical factorial and orthogonal designs cannot be used because they require too many runs or because the design space is constrained. This algorithm choose a given number of the initial designs such that the determinant of the information matrix XTX is maximized.
This problem is very complex so there is no guarantee that the chosen subset is actually the D-optimal one, we can only say that is the best combination found so far.
Three parameters should be defined:
Number of Designs a positive integer number between the design space dimension and the number of initial valid designs.
Number of Iterations maximum number of iterations used to find a good solution. The more this number is big, the more the algorithm will run.
Random Generator Seed a positive integer number, used for sequence repeatability. If the seed value is 0, the sequence is automatically seeded to a value based on the current time.
