|
|
Issue:
Volume 6, Number 4
Date:
April 2006
From:
Mark J. Anderson, Stat-Ease,
Inc. (http://www.statease.com)
"Statistics Made
Easy" (tm) Blog http://statsmadeeasy.net
Dear Experimenter,
Here's another set of frequently asked questions (FAQs) about doing
design of experiments (DOE), plus alerts to timely information and
free software updates. If you missed the previous DOE FAQ Alert, please click on the links at the bottom of this page.
If you have a question that needs answering, click the Search tab
and enter the key words. This finds not only answers from previous
Alerts, but also other documents posted to the Stat-Ease web site.
Feel free to forward this newsletter to your colleagues. They can
subscribe by going to http://www.statease.com/doealertreg.html.
If this newsletter prompts you to ask your own questions about DOE,
please address them via mail to:[email protected].
This plain text message may be easier to read
if you do not remove the line breaks. If you use Microsoft Outlook,
press the bar labeled "Extra line breaks in this message were
removed." In newer versions select "Restore line breaks."
Here's an appetizer to get this Alert off to a good start: http://www.amstat.org/publications/jse/v13n3/richardson.html
an in-class experiment that compares the flavor longevity of
two different brands of chewing gum. Gum manufacturers seek the longest-lasting
flavor about 12 minutes is typical according to statistics
cited by the authors. (After that time is up it goes under the classroom
desk, so ideally the flavor could be stretched longer than the lecture
and students would keep their gummy where their mouths are.)
I found a more sophisticated experiment on how long gum lasts at http://www.che.utexas.edu/cache/newsletters/fall2003_active.pdf.
It suggests a two-level factorial design on:
A. Flavor Fruit Juice versus Double Mint
B. Gender Female vs Male
C. Meal Before vs After.
The author, Gerardine G. Botte (currently at Ohio University http://webche.ent.ohiou.edu/faculty/botte.html),
suggested students at the University of Minnesota Duluth chew on this
as homework for their Design of Engineering Experiments class. Many
other fun ideas for DOE exercises are provided by Professor Botte.
I wish I had her as a statistics teacher!
PS. Breaking news: Chewing gum makes you smarter and good-looking!
Wrigley gum researchers hope to prove this by investing in unbiased
(?) experimentation on these possible benefits to masticating their
product. See http://tinyurl.com/papg8
for their publicity on this research initiative.
Here's what I cover in the body text of this DOE FAQ Alert (topics
that delve into statistical detail are designated "Expert"):
1. FAQ: Why is there no intercept in models for mixture designs?
2. FAQ: Categorical factor in a response surface method (RSM)
3. Info Alert: Balloon car simulation (submitted by reader)
4. Submission: Photo of "RSM Simplified" book at
Center Point
5. Events Alert: First European DOE User Meeting (2nd notice!)
6. Workshop Alert: "Crash Course on DOE for Sales &
Marketing"
PS. Quote for the month: How to get the Net Generation engaged
in statistics classes.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1. FAQ: Why is
there no intercept in models from mixture designs?
-----Original Question-----
From: Pennsylvania
"Why do I not have the intercept as in a regular factorial
model in my analysis of a mixture experiment set up on Design-Expert®
software?"
Answer:
We make use of a mixture model developed by Scheffe, which is characterized
by not having an intercept or squared terms for the quadratic form.
For the mathematical details, see these lecture notes by Professor
Steve Buyske (link provided with his express permission) from the
Statistics Department of Rutgers University: http://www.stat.rutgers.edu/~buyske/591/lect09.pdf.
As he states "...Design-Expert will analyze mixture designs
with aplomb..."
(Learn more about design and analysis of mixture experiments by
attending the three-day computer intensive workshop "Mixture
Design for Optimal Formulations." For a course description,
click http://www.statease.com/clas_mix.html
and link from there to a schedule of public presentations and on-line
enrollment.)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2. FAQ: Categorical factor in a
response surface method (RSM)
-----Original Question-----
From: Ohio
"I saw your DOE FAQ Alert electronic newsletter on the
internet and I was wondering if I could ask you a question
about an issue I am having with setting up a DOE design. I
am trying to set up a four-factor, three-level design. Since
it is three-level, (I know that the response is nonlinear),
I wanted to use a central composite design (CCD), but one
of my factors, "types of acids", is qualitative.
I have three different types of acids that I want to try.
The other three factors are quantitative. I realize that the
CCD design is only for quantitative factors, but since I have
three different acids types can I still use the CCD design?"
Answer:
Yes, but it must be a three-factor CCD replicated on each
of the three acids, or, more efficiently, a D optimal fraction
aimed at fitting the four-factor quadratic model the
backbone of response surface methods (RSM). Our software Design-Expert
Version 7 provides the necessary tools (free trial available
at http://www.statease.com/soft_ftp.html).
(Learn more about RSM by attending the three-day computer-intensive
workshop "Response Surface Methods for Process Optimization."
See http://www.statease.com/clas_rsm.html
for a complete description. Link from this page to the course
outline and schedule. Then, if you like, enroll online.)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3. Info alert: Balloon
car simulation (submitted by reader)
-----Original Submission (edited)-----
From: Anish Patani, GE Energy, Master Black Belt - Lean
& Six Sigma Program Manager
"I haven't had an opportunity to use Design-Expert but the
reviewer you cited last month is right on by saying that "Stat-Ease
is known for the jokily accessible examples." (See http://www.scientific-computing.com/scwfebmar06computational.html.)
Any form of humor to make statistics fun is always welcome. Reading
your post "DOE It Yourself, Fun Science Projects" at
http://www.statease.com/pubs/doe-self.pdf
on in-class DOEs, I got the drag racing software, but it took
too much time for students to learn. A student in my class suggested
the Balloon Car Simulation (http://pbskids.org/zoom/games/ballooncar/)
that they had used in a MBA class. Each student was required to
run all possible combinations (60) to get the optimum answer.
The simulation is very simple you can vary four factors:
— Tire size (2)
— Axle length (2)
— Valve diameter (5)
— Side wall height (3)
The objective is to increase speed and maximize distance. By running
a half-factorial DOE you can get two transfer functions for speed
and distance which yields an optimum or close to optimum answer.
Then one can perform validation runs for the remaining eight runs
in addition to creating a full-factorial design.
Tips:
— If you ask the student to find the best speed and distance without
teaching DOE, the chances are they will reach the optimum in 15
trials but they will continue trying further experiments without
realizing they have reached the optimum design.
— The two transfer functions contradict each other and now the
solution becomes a question about optimization: What will you
compromise speed or distance?
— No interactions are significant, hence using one factor at a
time or sticking with the winner approach works in this case,
but if there were interactions the student needs to understand
DOE.
— For the factor valve diameter the response is not linear, hence
if you selected two extreme settings, the chances are you will
not get optimum transfer function, but if you selected the 2nd
and 4th levels it will be better, yet not the best. This should
take you into a discussion on selection of regions such that the
assumption of linearity is valid, or consideration for a central
composited design (CCD) for generating a non-linear response surface.
Caution since the simulation is very simple so is the learning.
It is good simulation for those who are doing DOE for the first
time. For more advanced learning, one can use the simulation for
drag racing that you suggest in "DOE It Yourself."
Answer:
I took a look at this it promises to be fun! I plan to
actually make some balloon cars to see what they will do.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4. Submission: Photo of "RSM Simplified" book at Center
Point
-----Original Submission (edited)-----
From: Ian A. McCulloh, Major, Chemical Corps, Instructor
"Mark, I teach RSM at the U.S. Military Academy at West Point.I
bought your "RSM Simplified" book* for my students this
semester. A tradition here at the Academy is for students to get
a spring break picture with their text book. The idea is to try
and make them at least take a text book with them. There is some
obvious RSM/DOE humor, but I thought you might like to put it in
the Stat-Ease newsletter."
Answer:
Click the following link to see how CDT Dusty Turner carried on
the U.S. Military Academy tradition for spring break: http://www.statease.com/images/turner_at_center_point.jpg.
According to Major McCulloh "he is a second semester junior
or COW as we call it here. He was born and raised in Center Point,
TX." I found 18 "Center Point"s listed for the USA
in Microsoft "Streets and Trips" . I suppose that wherever
one is must be considered the center point of that person's universe.
*(See http://www.statease.com/rsm_simplified.html
for details on this book and link from there to order online.)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
5. Events alert: First European DOE User Meeting (2nd notice!)
The "First European DOE User Meeting" for Design-Expert
and Design-Ease Users" will be held at the Faculty Club in
Leuven, Belgium on April 24 and 25. This DOE conference is sponsored
by Stat-Ease with hosting by CQ Consultancy and participation from
other European resellers of Design-Expert and Design-Ease software.
See details at http://www.cq.be/DOEconference/.
The meeting is a must for European users of Stat-Ease software and
perhaps the perfect excuse for those elsewhere to take in the sights
in Leuven and surroundings. I lunched at the city center some years
ago it was very beautiful!
Also, coming up soon are:
— ASQ's "World Conference on Quality & Improvement"
in Milwaukee, Wisconsin on May 1-3, where I will help staff the
Stat-Ease exhibit at Booth 424.
— PSI (statisticians in the pharmaceutical industry) 29th Annual
Conference 2006 in Bristol, UK on May 14-17, where reseller and
DOE expert in his own right, Alan Collins, will exhibit Stat-Ease
software via his company, QD Consulting.
Click http://www.statease.com/events.html
for a complete list of appearances by Stat-Ease professionals. We
hope to see you sometime in the near future!
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
6. Workshop alert: Crash Course on DOE for Sales & Marketing
(If you are a corporate technical professional, please pass this
news on to your business people!) Quickly identify those factors
which affect your sales and marketing results. Learn how they interact
and apply that knowledge to make breakthrough increases in sales
and profits. Attend the newly expanded two-day "Crash Course
on DOE for Sales & Marketing" at the Stat-Ease training
center in Minneapolis on April 19 20. See the course description
and links to the syllabus and online enrollment at http://www.statease.com/clas_smdoe.html.
Note that we will also present a one-day version of DOE for Sales
& Marketing at a hotel near O'Hare airport in Chicago this June.
See http://www.statease.com/clas_pub.html
for schedule and site information on all Stat-Ease workshops open
to the public. To enroll, click the "register online"
link on our web site or call Stat-Ease at 1-612-378-9449. If spots
remain available, bring along several colleagues and take advantage
of quantity discounts in tuition, or consider bringing in an expert
from Stat-Ease to teach a private class at your site. Call us to
get a quote.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
I hope you learned something from this
issue. Address your general questions and comments to me at: [email protected].
Sincerely,
Mark
Mark J. Anderson, PE, CQE
Principal, Stat-Ease, Inc. (http://www.statease.com)
2021 East Hennepin Avenue, Suite 480
Minneapolis, Minnesota 55413 USA
PS. Quote for the
monthhow to get the Net Generation (see http://en.wikipedia.org/wiki/Generation_Y)
engaged in statistics classes:
"Their fast-paced world of
action movies, rapid-fire TV commercials and video games does not
prepare todays students to sit and absorb a lecture, especially
on a supposedly dull subject like statistics. To capture the interest
of these students, teaching must move away from lecture and listen
to innovative activities that engage students in the learning process."
R. Scheaffer (http://www.stat.ufl.edu/personnel/usrpages/scheaffer.shtml)
Trademarks: Design-Ease, Design-Expert and Stat-Ease are registered trademarks of Stat-Ease, Inc.
Acknowledgements to contributors:
Students of Stat-Ease training and users of Stat-Ease software
Fellow Stat-Ease consultants Pat Whitcomb, Shari Kraber and
Wayne Adams (see http://www.statease.com/consult.html
for resumes)
Statistical advisor to Stat-Ease: Dr. Gary Oehlert (http://www.statease.com/garyoehl.html)
Stat-Ease programmers, especially Tryg Helseth and Neal Vaughn
(http://www.statease.com/pgmstaff.html)
Heidi Hansel, Stat-Ease marketing director, and all the remaining
staff
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Interested
in previous FAQ DOE Alert e-mail newsletters?
To view a past issue, choose it below.
#1 Mar 01,
#2 Apr 01,
#3 May 01,
#4 Jun 01,
#5 Jul 01
, #6 Aug 01,
#7 Sep 01,
#8 Oct 01,
#9 Nov 01,
#10 Dec 01,
#2-1 Jan 02,
#2-2 Feb 02,
#2-3 Mar 02,
#2-4 Apr 02,
#2-5 May 02,
#2-6 Jun 02,
#2-7 Jul 02,
#2-8 Aug 02,
#2-9 Sep 02,
#2-10 Oct
02, #2-11
Nov 02, #2-12
Dec 02, #3-1
Jan 03, #3-2
Feb 03, #3-3
Mar 03, #3-4
Apr 03, #3-5
May 03, #3-6
Jun 03, #3-7
Jul 03, #3-8
Aug 03, #3-9
Sep 03 #3-10
Oct 03, #3-11
Nov 03, #3-12
Dec 03, #4-1
Jan 04, #4-2
Feb 04, #4-3
Mar 04, #4-4
Apr 04, #4-5
May 04, #4-6
Jun 04, #4-7
Jul 04, #4-8
Aug 04, #4-9
Sep 04, #4-10
Oct 04, #4-11
Nov 04, #4-12
Dec 04, #5-1
Jan 05, #5-2
Feb 05, #5-3
Mar 05, #5-4
Apr 05, #5-5
May 05, #5-6
Jun 05, #5-7
Jul 05, #5-8
Aug 05, #5-9
Sep 05, #5-10
Oct 05, #5-11
Nov 05, #5-12
Dec 05, #6-01
Jan 06, #6-02
Feb 06, #6-03
Mar 06, #6-04 Apr 06 (see above)
Click here to add your name to the DOE
FAQ Alert newsletter list server.
Statistics Made Easy
DOE FAQ Alert ©2006 Stat-Ease, Inc.
All rights reserved.
|
|
|