Page 254 - Six Sigma Advanced Tools for Black Belts and Master Black Belts

OTE/SPH
OTE/SPH
Char Count= 0
3:5
JWBK119-16
August 31, 2006

16

A Glossary for Design of

Experiments with Examples

H. S. Yam

Fundamental concepts in the design of engineering experiments are usually conveyed
through factorial designs and fractional factorial designs which allow for the study of two
or more treatment factors in reduced time and at reduced cost. This chapter pro-
k
vides a rudimentary outline of factorial experimental designs -- comparing 2 designs
k
vs. 3 designs, coded values vs. original values, standard order vs. run order, and
main effects vs. interactions. It also includes terms used in the analysis of results,
residual analysis and types of design. This is followed by fractional factorial designs,
alias structure and resolution of an experiment; all of which are illustrated with ex-
amples. Finally, basic terms used in robust design such as Taguchi’s outer array of
noise factors, inner array of control factors, repeated measures, replicates and blocking are
presented with examples.

16.1 FACTORIAL DESIGNS

Factorial designs allow for the simultaneous study of two or more treatment factors
and their interaction effects. This facilitates the collection of relevant information
in reduced time and at reduced cost. In the following we first present some basic
terminology.

k
k
16.1.1 2 Designs vs. 3 Designs
k
Ina2 factorial design, each factor is run at only two levels -- usually the upper and
k
lower ends of the current process window. In a 3 factorial design, each factor is run
at the upper and lowers ends, as well as the center point between them.
Six Sigma: Advanced Tools for Black Belts and Master Black Belts L. C. Tang, T. N. Goh, H. S. Yam and T. Yoap
C 2006 John Wiley & Sons, Ltd
239