pg_0108

Process Improvement

5.3.

Choosing an experimental design

5.3.3.

How do you select an experimental design.

5.3.3.6.

Response surface designs

Response

surface

models may

involve just

main effects

and

interactions

or they may

also have

quadratic

and possibly

cubic terms

to account

for curvature

Earlier, we described the

response surface method

(RSM) objective. Under

some circumstances, a model involving only main effects and interactions

may be appropriate to describe a response surface when

Analysis of the results revealed no evidence of "pure quadratic"

curvature in the response of interest (i.e., the response at the center

approximately equals the average of the responses at the factorial

runs).

The design matrix originally used included the limits of the factor

settings available to run the process.

Equations for

quadratic

and cubic

models

In other circumstances, a complete description of the process behavior might

require a quadratic or cubic model:

Quadratic

Cubic

These are the full models, with all possible terms, rarely would all of the

terms be needed in an application.

5.3.3.6. Response surface designs

http://www.itl.nist.gov/div898/handbook/pri/section3/pri336.htm (1 of 6) [5/7/2002 4:02:00 PM]