Central Composite Design Creation

Creates standard central composite, non-factorial, response surface designs.

General

Element Name Description
Type of design Select a type of central composite design; after selecting the design type, select the design from the options below. In the standard designs, the cube portion of the design is typically of resolution V (or higher). This is, however, not necessary, and in cases when the experimental runs are expensive, or when it is not necessary to perform a statistically powerful test of model adequacy, then one could choose for the cube portion designs of Resolution III. For example, it could be constructed from highly fractionalized Plackett-Burman designs. These designs are referred to as Small design.
Standard design Select a standard central composite design (Factors, Blocks, Runs); the selection from this menu is used only if the Type of design is set to Standard design. These types of designs are typically constructed from a resolution V (or higher) 2(k-p) design, which is enhanced by center and star points.
Small design Select a small central composite design (Factors, Blocks, Runs). These are constructed from resolution III (highly fractionalized) 2(k-p) designs or Plackett-Burman designs, which are enhanced by center and star points. These designs are very useful when each observation or run for the experiment is very costly.
Design characteristics/criterion Determines which axial distance will be used in the design, and whether center points will be added.
Ordering of runs Specifies ordering method for runs in design. Random, Standard (sequential), or by Blocks (the latter only if blocks are selected).
Random number seed Optional, if random ordering or order by blocks is selected; provide here a seed from 1 to 99999999 to create the random sequence of runs for the design.
Number of replicates Add to design option: Number of replicates to add to the design (minimum 0, maximum 300).
Number of center points Add to design option: Number of Center Points (per block) to add to the design (minimum 0, maximum 50).
Number of blank columns Add to design option: Number of blank columns to add to the design spreadsheet (minimum 0, maximum 300).
Correlations of effects Displays a spreadsheet that contains the correlations for the effects in the standard second-order model and the blocking variables (if the current design is blocked). Note that these correlations are based on the current choice of axial distance and the current number of center points (per block) in the design.