Design of a Robust Design Experiment - Quick Tab

Click the Summary: Display design button to display a spreadsheet specifying the settings for each experimental run. Settings on this tab and the Display design tab affect the display of the design. To save the displayed design, use Save or Save As from the File menu.

Use the options in the Order of runs group box to (1) randomize the order of runs, or (2) display the design in standard order. Remember that for standard experiments, it is common practice to randomize the runs in order to avoid any biases due to fluctuations in the production process across runs. Note that when the Randomized order option button is selected, Statistica generates the random numbers based on a seed value. You can use the Seed field on the Display design tab to change the seed for the random number generator.

Click the Alias structure button to display a spreadsheet showing a matrix of main effects (rows) by two-way interaction effects (columns). If the entries in the matrix are blank, then the respective two effects (main effect and interaction effect) are not confounded with each other (they are orthogonal). Asterisks (*) in this matrix mark effects that are not orthogonal.

This option is useful for identifying whether or not certain two-way interactions are confounded with the main effects in the design, and, if so, with which ones. Some orthogonal arrays allow you to estimate some two-way interactions, provided that specific main effects are not used (i.e., that those columns in the design are not used). Please refer to Phadke (1989) or Taguchi (1987) to learn more about advanced methods of "manipulating" and "molding" orthogonal arrays to accommodate specific unusual design needs (via the use of linear graphs).