Hi Reed: If you are from GE, then were you trained as a Six Sigma Black Belt or Six Sigma Green Belt? Just curious if you went through the formal training. I applaud your desire to "look at a number of attributes at the same time and lets you distinguish between each of them". These advanced statistical technique are not widely known when people seek out experimental design to advance knowledge gained. I trained both of my sons on how to do these DOE type of experiments when they where in grade 7-8. Refer to following web site links: www.asq.org/pub/qualityprogress/past/1101/news.html or www.Golf54.com "Press", then "Quality Progress, Nov. 2001" for the same article. Refer to the article below, that describes the difference between OFAT and DOE type of experiments: Reed, you are definitely on the right track with your test strategy for improvement. If you are from GE, then you might consider partnering with one of the golfing Black Belts to design the experiments then data crunch. Reed, this stuff is real powerful, so you may be surprised by what you learn! Thanks HarryS (Director of Quality, ASQ trained Six Sigma Black Belt)
BACKGROUND INFORMATION ON DOE / TAGUCHI METHODS® Design of Experiments (DOE) · Design of Experiments is the simultaneous study of several variables (causes / input factors). · Typical one-factor-at-a-time (OFAT) experiment limits knowledge generation and wastes time. · By combining several variables in one study, instead of creating a separate study for each, the amount of testing is dramatically reduced and the experiment yields greater knowledge. · The major disadvantage of the one-factor-at-a-time strategy is that it fails to consider any possible interaction between the factors. An interaction example is two medications, each of which is good for a patient, but when both are taken together yields a new strong side effect. · Design of Experiments is a valuable tool for gaining knowledge through experimentation, since it is possible to extract large amounts of knowledge in a matrix, from a small fraction of data. Taguchi Methods® Robust Engineering · Robust Engineering is the creation of world-renowned engineering genius Dr. Genichi Taguchi. · Dr. Taguchi defines “robustness” as, “the state where the technology, product, or process performance is minimally sensitive to factors causing variability and ageing at lowest unit cost. · Dr. Taguchi specifies the use of a S/N Ratio (Signal-to-Noise Ratio) to measure robustness. Parameter Design and Control Factors are used to optimize process / product for robustness. NASA Case Study · Jet Propulsion Laboratory, NASA and California Institute of Technology, USA applied Taguchi Methods® to the qualification and optimization of NASA flight NiCd rechargeable batteries. · In recent years, several spacecraft NiCd battery anomalies occurred that drastically affected spacecraft life. There were many variables and levels of each variable that affects reliability. · Using conventional experimenting techniques of studying one variable at a time, this study at NASA would have taken 11.2 years to complete. Using a modified L16 Taguchi Array, a total of five Input Factors at four levels each were studied. Testing was reduced to several months. · The application of Taguchi Methods® at NASA resulted in improved battery voltage performance by 96%. COPYRIGHT © 2001 - 2002 BY HARRY F. SCHIESTEL (www.Golf54.com). ALL RIGHTS RESERVED. -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of REED GRANT Sent: Tuesday, December 24, 2002 10:11 PM To: [EMAIL PROTECTED] Subject: Re: ShopTalk: shaft flex v.s. frequency Dave I appreciate you asking me for some comments. Just as each of you experienced indivduals, are doing, this spring, too damn cold now(23* this morning), I want to apply Six Sigma testing capabilties to shaft spining-frequenzy. I feel I am not good enough a player to be consistantly repetitive to distinguish testing. But I do have those around me that are good enough that are worthy of testing with good results. Six Sigma is designed as a testing medium that looks like it would be perfect for this type of testing. It lets you look at a number of attributes at the same time and lets you distinguish between each of them. And I do want to load some good shafts with some knowingly predictable (bad) shafts. I want to know exactly what does a bad shaft do. And then I want to know what does a good shaft do. I want to classify good shafts, so that I can predict what part shaft orientation plays. I know that many of you already have and are doing this now. I would appreciate any help into how to lay out the test and any info that you have already derived. Also I want to test club heads designs so that I know which club heads are best for what kind of fairways. As you can see, I have many quetions with few answers. Building clubs started out as a hobby, and now has turned into a passion for knowledge. I guess this shows my ignorance, but that has shown many times before. So what's new. Thanks Reed