Use Create 2-Level Split-Plot Design to create a designed experiment when complete randomization of the runs is difficult due to time or cost constraints. You can create a full and fractional split-plot design with up to 7 factors. For more information, see Split-plot designs in design of experiments and What is a hard-to-change factor?.
When you create a design, Minitab stores the design information in the worksheet, which shows the order in which data should be collected. Once you have collected your data, use Analyze Factorial Design to analyze the data.
For example, pastry chefs at a large-scale bakery are designing a new brownie recipe. The scientists test two types of chocolate, two amounts of sugar, and two different oven temperatures. The oven temperature is a hard-to-change factor because it takes too much time for the temperature to stabilize after it is changed. Thus, the chefs use a split-plot design. They set the oven temperature and bake all the combinations of chocolate and sugar in the oven at the same time. Then, the chefs change the temperature and bake all the combinations of chocolate and sugar at that temperature in random order.
This Minitab worksheet shows a portion of the split-plot design. The chefs perform the experiment by collecting data using the order shown in the RunOrder column, which contains the randomized order of the runs.
After collecting the data, a chef enters the response data in an empty column in the worksheet and then analyzes the design.
Many of the choices you make when you create a design depend on your overall experimental plan. For more information, go to Phases of a designed experiment.