Example of Create Response Surface Design (Box-Behnken)

Food scientists are developing a new healthier breakfast cereal that still tastes good. They want to identify the combination of three sweeteners that optimize taste. Because previous research has determined that the relationship is curvilinear, the scientists start by creating a Box-Behnken design.

  1. Choose Stat > DOE > Response Surface > Create Response Surface Design.
  2. Under Type of Design, select Box-Behnken.
  3. From Number of continuous factors, select 3.
  4. Click Designs, and then click OK.
  5. Click Results. Select Summary table and design table.
  6. Click OK in each dialog box.

Interpret the results

The first table gives a summary of the design, which includes the total number of factors, runs, blocks, and replicates.

The design table displays the factor settings for each experimental run using coded factor names and levels. For example, in the first run, Factor A and C are at level −1 and Factors B is at level 0. With 3 factors, the design has 15 runs. In the worksheet, Minitab displays the names of the factors and the levels.

Note

Minitab randomizes the design by default, so when you create this design, the run order will not match the order in the example output.

Design Summary

Factors:3Replicates:1
Base runs:15Total runs:15
Base blocks:1Total blocks:1
Center points: 3

Design Table (randomized)

RunBlkABC
11-10-1
21-101
31-1-10
411-10
5110-1
610-1-1
71101
81-110
91011
101110
1110-11
121000
131000
141000
15101-1