Page 320 - Six Sigma Advanced Tools for Black Belts and Master Black Belts
P. 320
OTE/SPH
OTE/SPH
Char Count= 0
3:6
August 31, 2006
JWBK119-19
Concluding Remarks 305
Start
Determine size of
regular design
Specify number of
effects needed and
select lean design
Run experiment
Calculate dummy
responses with
available data
Calculate needed
effects
Reduce number of
At least one dummy dummy responses
response in use?
by one
Yes
Replace one dummy
Necessary to refine Yes response by
No results? experimental value
from an addition run
No
Stop
Figure 19.1 Lean design for incremental experimentation.
nature inherent in the results generated at each stage is also consonant with the spirit
of ‘continuous improvement’ in an investigation.
19.7 CONCLUDING REMARKS
Experimental design is normally understood to be used in such a way that, regardless
of the type and size of experiment, no data analysis could commence before the full
set of response values is available. In other words, the investigator must wait for the
