Doc
was active Not recently activeActivity

Doc replied to the topic ISO vs. Six Sigma in the forum General 14 years, 11 months ago
ISO9000 and TS16949 (auto industry) defines systems that need to be in place in order to ensure high quality designs, processes, and production. This includes management responsibilities, design systems, purchasing, document control, auditing, corrective actions, etc… Just one of these areas is continuous improvement.
Six Sigma is a method of…[Read more] 
Doc replied to the topic Degree of Freedom:Pooled estimate of variance in the forum General 14 years, 11 months ago
Pooled Var = [V1(n11) + V2(n21) + V3(n31)] / [(n11) + (n21) + (n31))
You can think of this as a weighted average of the variances … weighted using the degrees of freedom for each group. Of course, if the groups all have the same sample size, then the weighting is equal, and the pooled variance is simply the mean of the va…[Read more] 
Doc replied to the topic MSA Help Needed in the forum General 14 years, 11 months ago
The problem with using the same parts for both studies is that the linearity parts are typically handselected so that the parts filled the expected operating range. These parts can’t provide an estimate of the process variation.
The GR&R parts, on the other hand, should be randomly selected hoping to provide a good representation of the…[Read more] 
Doc replied to the topic Shanin Quality Principles in the forum General 15 years, 1 month ago
you said…
“If we did the PPAP’s and got the capability and controls that we pretend to, hardly any of this would be necessary.”
If “we” were able to do that, none of US would be necessary as problem solvers, now would we…
Fact is, PPAP’s are usually just a list of items to check off, and the “meat” of what underlies each of those check…[Read more] 
Doc replied to the topic Measures of central tendency in the forum General 15 years, 4 months ago
My mistake. I thought the original post said “binomial” instead of “bimodial”.
For bimodal distribution most people would probably use the 50th percentile, though a better strategy would be to try to determine the cause of the two modes, isolate them, and then characterize each cause separately. 
Doc replied to the topic Measures of central tendency in the forum General 15 years, 4 months ago
There is a lot of confusion about the binomial distribution in this thread.
The binomial distribution involves the random variable x, where x is number of “successes” from n trials where each trial involves a probability of success equal to p.
The potential values of x are 0, 1, …, n, and the binomial distribution gives the values for each of…[Read more] 
Doc replied to the topic Confidence and Reliability in the forum General 15 years, 5 months ago
1. When they say 95% reliability, they are talking about the lifespan at which 5% of the units are expected to have failed (95% survive). This is the 5th percentile for the distriubution of expected lifespans (in cycles, in your case).
2. When they say “at 90% confidence” they are usually referring to the 90% lower confidence limit for the 5th…[Read more] 
Doc replied to the topic Linearity and Bias study in the forum General 15 years, 6 months ago
The AIAG Measurement System Analysis Reference Manual calls for at least 5 parts whose measurements cover the operating range of the gage, and then to measure each part at least 10 times by an operator who normally uses the gage.

Doc replied to the topic Statistics in the forum General 16 years, 7 months ago
Descriptive Statistics (mean, var, …) and Inferential Statistics (tests)

Doc replied to the topic statapult in the forum General 16 years, 9 months ago
Word Mark: STATAPULT
Goods & Services: IC 009. US 021 023 026 036 038. G & S: SCIENTIFIC AND TEACHING APPARATUS COMPRISED OF WOOD AND RESEMBLING A CATAPULT FOR USE IN DEMONSTRATING AND TEACHING MATHEMATICAL CONCEPTS, PROCESSES, METHODS, AND CAUSE AND EFFECT, AND FOR CONDUCTING MATHEMATICAL MODELING EXPERIMENTS. FIRST USE: 19891231. FIRST USE IN…[Read more] 
Doc replied to the topic Sigma 17.4 with N=33. What is the standard deviation in the forum General 16 years, 11 months ago
You want:
Mean +/ Z[(c+1)/2]*Sigma/SQRT(n)
Where Z[(c+1)/2] is the 100*(c+1)/2th percentile of the standard normal distribution and c is the confidence, such as 83.3. By the way, Z[(c+1)/2] is the same as Z[1(alpha/2)]
So, you look up Z[(.833+1)/2]=Z(0.9165) in a standard normal table and you find 1.38. This is the value of z for which 91.65%…[Read more] 
Doc replied to the topic Gage RR % of contribution in the forum General 16 years, 11 months ago
I’m not sure where he got 2% and 8%.
I suspect these are mutated acceptance criteria taken from the 10%^2=1% and 30%^2=9% metrics I mentioned in the earlier post. These are calculated using the AIAG MSA Reference Manual acceptance criteria for %GRR and %Tolerance.
Some people have rounded the 9% up to 10% for easeofuse/presentaiton. 
Doc replied to the topic Gage RR % of contribution in the forum General 16 years, 11 months ago
Careful here … you are mixing up on the indices.
The AIAG MSA Reference Manual uses the following accpetance criteria for %GRR and %Tolerance. Note that these are standard deviationbased metrics:
<10% – generally acceptable
10%30% – may be acceptable based upon importance of application, cost of measurement device, cost of repairs,…[Read more] 
Doc replied to the topic How does minitab….? in the forum General 17 years, 5 months ago
Go to the Minitab Help.
Click on the help for Graphical Summary. Select “see also”. Select “Methods and formulas”. Scroll down to find the formulas you’re seeking. The form for the median is not as simple as that of the mean. Read up on it though. 
Doc replied to the topic (n1) degrees of freedom in the forum General 17 years, 7 months ago
n people have just finished eating at a Chinese restaurant and the waiter gives them n fortune cookies on a single plate in the center of the table.
One person picks up the plate, chooses one of the cookies, and then passes the plate to the next person.
That person chooses one of the remaining cookies, and then passes the plate to the next…[Read more] 
Doc replied to the topic Using the term Six Sigma in an article in the forum General 17 years, 9 months ago
It would be good form to say something like “Six Sigma is a registered trademark of Motorola, Inc.”, BUT you certainly don’t need permission to use it in an article.
I have heard some rumors that Motorola has been known to contact some companies using the term “for profit”, possibly discussing licensing of the trademark, but haven’t gotten…[Read more] 
Doc replied to the topic Reliability statistics in the forum General 17 years, 9 months ago
Consider Statistical Methods for Reliability Data by Meeker & Escobar. This book is more focused on the analysis of reliability data.
O’Connor’s book is more focused on how to setup engineering and management systems that result in reliable products, though it does cover some basic analysis techniques.
Also consider the online resource…[Read more] 
Doc replied to the topic Six Pack Analysis in Gage RR using Minitab in the forum General 17 years, 10 months ago
Yes, there is.
See the StatGuide 
Doc replied to the topic DFSS to potentially change pricing in the forum General 17 years, 11 months ago
Stan & RubberDude – Here is an acronym translation:
DFSS – Design for Six Sigma
DMEDI – Define, Measure, Evaluate, Design, Integrate (similar to DMAIC, but applies to DFSS instead)
QFD – Quality Function Deployment – used to help determine CTQ’s
CTQ – Critical to Quality characteristics – highest priority benefits the customer is looking for…[Read more] 
Doc replied to the topic Reliability Testing in the forum General 17 years, 11 months ago
With no censored data, you really should be using maximum likelihood estimation instead of least squares regression. It is preferred when there is little (or no) censoring and the sample size isn’t tiny.
 Load More