Use Select Optimal Design to select, add, exchange, or evaluate runs from a candidate set of experimental runs. Minitab's optimal design capabilities can be used with general full factorial designs, response surface designs, and mixture designs. Minitab provides two optimality criteria for the selection of design points, D-optimality and distance-based optimality. For more information, go to What is an optimal design?.

You can use Select Optimal Design to perform any of the following tasks.

- Select an optimal design
- Select design points from a candidate set of points to achieve an optimal design. Select Optimal Design is often used to reduce the number of experimental runs when the original design contains more points than are feasible due to time or financial constraints.
- For example, an engineer has two 3-level categorical factors and three 2-level categorical factors that require 72 runs for a single replicate of a general full factorial design. Instead, the engineer selects 24 points to form a D-optimal design that can estimate the main effects and some 2-way interactions.
- Augment an existing design
- Add design points to a D-optimal design. Usually, you augment a design when you have additional resources to collect more data after you already created a design and collected data. You may also want to augment a design to add runs that allow you to include additional terms in your model.
- For example, after the engineer runs a 24-run D-optimal design, the engineer finds that there are enough resources to add 4 more experimental runs.
- Improve the D-optimality of a design
- Add or exchange points to improve the D-optimality of the design. Usually, you exchange points before you collect data.
- For example, in the initial design, the 24 points in the D-optimal design estimate main effects and some two-way interactions. The engineer decides to select a different set of 24 design points that estimate only main effects.
- Evaluate a design
- Obtain optimality statistics for your design. You can use this information to compare designs or to evaluate changes in the optimality of a design if you change the model.
- For example, the engineer generates a D-optimal design that estimates all of the main effects and some 2-way interactions. The engineer wants to know how the D-optimality changes if the model contains only main effects.

- If you have a factorial design in the worksheet but cannot select an optimal design, first use Define Custom General Full Factorial Design. Then you can select optimal points from the design.
- If you have a response surface design in the worksheet but cannot select an optimal design, first use Define Custom Response Surface Design. Then you can select optimal points from the design.
- If you have a mixture design in the worksheet but cannot select an optimal design, first use Define Custom Mixture Design. Then you can select optimal points from the design.