ADXCCD: Construct Central Composite Designs

Macros for the Design and Analysis of Experiments

ADXCCD: Construct Central Composite Designs

%adxccd(ds, nf, nr, nc, alpha, nb)

where

ds is the name of the output SAS data set for the constructed design. This data set contains variables for design factors and for a block factor, if the design uses blocking. Default names are used. The default names for design factors are T1, T2, T3, and so on; the default name for a block factor is BLOCK.

nf is the number of design factors.

nr is the number of points in the factorial portion of the design and must be a power of 2. There must be a resolution 5 design in nf factors for the value of nr you specify. For example, if nf=5, then nr can be 16, but not 8.

nc is the number of centerpoints in the design. If you do not use blocking, nc is a nonnegative integer. If you use blocking, nc is two integers separated by a slash. The first integer gives the number of centerpoints in each block in the factorial portion of the design, and the second integer gives the number of centerpoints for the axial block.

alpha is the value of the axial extreme.

nb is the number of blocks in the design. Since the blocks for a design are the blocks for the factorial portion together with the axial block, the number of blocks must be of the form 2ⁿ+1 for some n. If the design does not use blocking, nb can be omitted.

The ADXCCD macro constructs a central composite design according to the parameters you specify and outputs the design to a SAS data set.

Suppose you want to examine a response surface over 5 factors. You can use the ADXPCC macro to find that there is a design in 33 runs and 2 blocks. One block is a half-fraction of the full 2⁵ design, and the other block is the axial portion of the design. You can produce the output from the ADXPCC macro and construct the design with the following statements:

   %adxgen
   %adxff
   %adxcc
   %adxinit
   %adxpcc(5)
   %adxccd(a,5,16,6/1,2.0000,2)

The design is output to data set A. Note that 6/1 is used for nc. Since the design involves blocking, this specifies that the factorial block contains 6 centerpoints and the axial block contains 1 centerpoint.

You can produce a design (output to data set B) with just one centerpoint in the axial block with the following statements:

   %adxgen
   %adxff
   %adxcc
   %adxinit
   %adxccd(b,5,16,0/1,2.0000,2)

Chapter Contents
Previous
Next
Top

ds	is the name of the output SAS data set for the constructed design. This data set contains variables for design factors and for a block factor, if the design uses blocking. Default names are used. The default names for design factors are T1, T2, T3, and so on; the default name for a block factor is BLOCK.
nf	is the number of design factors.
nr	is the number of points in the factorial portion of the design and must be a power of 2. There must be a resolution 5 design in nf factors for the value of nr you specify. For example, if nf=5, then nr can be 16, but not 8.
nc	is the number of centerpoints in the design. If you do not use blocking, nc is a nonnegative integer. If you use blocking, nc is two integers separated by a slash. The first integer gives the number of centerpoints in each block in the factorial portion of the design, and the second integer gives the number of centerpoints for the axial block.
alpha	is the value of the axial extreme.
nb	is the number of blocks in the design. Since the blocks for a design are the blocks for the factorial portion together with the axial block, the number of blocks must be of the form 2ⁿ+1 for some n. If the design does not use blocking, nb can be omitted.