90 Percent of DOE Is Half Planning

Don’t just teach statistics; teach how to solve problems

Wed, 05/18/2016 - 15:14

I’ve mentioned that design of experiments (DOE) is one of the few things worth salvaging from typical statistical training, and I thought I’d talk a bit more about DOE in the next couple of columns. The needed discipline for a good design is similar when using rapid-cycle plan-do-study-act (PDSA).

ADVERTISEMENT

Doing a search on the current state of DOE in improvement education, I observed that curricula haven’t changed much in the last 10 years and still seem to favor factorial designs or orthogonal arrays as a panacea.

The main topics for many basic courses remain:
• Full and fractional factorial designs
• Screening designs
• Residual analysis and normal probability plots
• Hypothesis testing and analysis of variance (ANOVA)

The main topics for advanced DOE courses usually include:
• Taguchi signal-to-noise ratio
• Taguchi approach to experimental design
• Response-surface designs
• Hill climbing
• Mixture designs

 …

Want to continue?
Log in or create a FREE account.
By logging in you agree to receive communication from Quality Digest. Privacy Policy.

Comments

Thank you Davis, the Quality Improvement Through Planned Experimentation is on it's way to my door now. Could you go a little more in depth as to why the 15 experiments is the best approach. Obviously I would need to see the data to form more of an opinion, but you say you have 25 maxiumum, 3 variables, and 1 output. Please go into more detail as to how the 15 experiments would be the most efficient use of my time.

I'm currently using AIAG DOE's similar to Gage RR's, to test a bunch of gages in a short business trip to Japan, the purpose is machine and gage trials. Obviously time is my budget for this experiment. I have 42 days and many gages, I found that planning my experiments before I left has helped alot, but the experiments are still tedious and time consuming.

Otherwise, I completely agree that there should be more emphasis on DOE and specifically hypothesis testing in engineering statistics and problem solving. DOE has been my most powerful tool, and it shows as I am very successful and my company sends me to Japan to test out fun new machines.

Cheers, great article.

Hi, Ken,

Thank you so much for your kind comments.  I'm glad the article was useful for you.  It was definitely a "to be continued..." and the next two or so columns will continue using the tar scenario to expand on some very basic concepts of DOE.

BUT...here is another article I wrote for QD 10 years ago using the same scenario that goes more in depth and should answer your questions, especially about the 15-run design:

http://www.qualitydigest.com/feb06/articles/02_article.shtml

Please contact me if you have any more questions.

Thanks again for reading,

Davis

Add new comment