[S] Acting on low DOE confidence

Kim Niles (KSNiles@safeskin.com)
Fri, 27 Feb 1998 08:33:15 -0800


Can anyone assure me that I'm not just playing the lottery when I act on
low statical confidence from DOE's performed on manufacturing processes
at work.

My take on this is that if I don't have any idea which way to turn the
knob so as to produce the best product then I might as well go with the
direction pointed by DOE's even if it is lower than 50% confidence. So
far it's worked out fine but I afraid someone is going to pin me on it.

Since confidence lower than 50% worked, does this mean that confidence
was low just due to confounding or poor selection of factor ranges or
does it mean I was lucky? Does 50% confidence really mean a 50/50
situation in this case?

Thanks,
,,,
(o o) Kim Stewart Niles - Process Engineer
----oOO--(_)--OOo-----------------------------------------------
Tactyl Technologies,.......Phone: 760.598.2233
A Safeskin Corporation...Fax: 760.598.1012
2595 Commerce Way,.... mailto:ksniles@safeskin.com
Vista, CA 92083............. http://www.safeskin.com
----(=====O=====)----------------------------------------------
San Diego Chairman of Society for the Advancement
of Material and Process Engineering (SAMPE;97-98).
http://www.znet.com/~sdsampe/sampe.htm
mailto:sdsampe@znet.com (checked every other day).
Over 72 Designed Experiments Performed.
oooO Oooo
====( )==( )=============================
\ ( ) /
\_) (_/

"The problems that exist in the world today cannot be solved by the
level of thinking that created them" - Albert Einstein.

"Each quandary is a call for experimentation. Each experiment a
question put to the world. Each answer a journey. Let life plan the
itinerary. Our job is to pack light and bring a camera." - Max Elmore
in The Max Strategy by Dale Dauten; $12 Great Book.

-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news