Select Optimal Design - Methods
main topic see also
Select Optimal Design - Methods
Stat > DOE > Factorial, Response Surface, or Mixture > Select Optimal Design > choose Select optimal design > Methods
Allows you to specify whether the initial design is generated using a sequential or random algorithm, or a combination of both methods and to choose the search procedure for improving the initial design.
Initial Design
Generated by sequential optimization: Choose to have all design points selected sequentially.
Percentage of design points to be selected randomly: Choose the percentage of design points to be selected randomly. You can choose from 10% to 100% in increments of 10.
Number of random trials: Enter the number of trials for the optimization procedure. The default is 10.
Base for random data generator: Enter a base for the random data generator. By entering a base for the random data generator, you can control the randomization so that you obtain the same pattern every time.
Search Procedure for Improving Initial Design
Exchange method with number of exchange points: Choose to improve the initial design using the exchange method. Minitab will first add the best points from the candidate set, and then drop the worst points until the D-optimality of the design cannot be improved further. Then, enter the number of points to be exchanged.
Fedorov's method: Choose to improve the initial design using Fedorov's method. Minitab will simultaneously switch pairs of points. This is accomplished by adding one point from the candidate set and dropping another point so that the switch results in maximum improvement in D-optimality. This process continues until the design cannot be improved further.
None: Choose to suppress improvement of the initial design.