Gage R R Is somebody hearing
Six Sigma – iSixSigma › Forums › Old Forums › General › Gage R R Is somebody hearing
 This topic has 6 replies, 5 voices, and was last updated 18 years, 5 months ago by Arend.

AuthorPosts

December 17, 2003 at 4:29 am #34120
We are using Gage R & R for the Call Monitoring Executives.
The total Gage R & R has been thus reduced from a unbelievable 85 % to 35 %
Can somebody share the Benchmark for this particular case ( I am not talking about generic benchmark on Gage R & R ) – I mean for the Quality Monitoring Executives.0December 17, 2003 at 5:44 am #93640What is “call monitoring Executives” and how do you measure it?
0December 17, 2003 at 10:30 am #93648Hi G,
I am not aware of the benchmark for this case of yours. But that’s great! You have reduced it by 50%. Have you noted all the key factors impacting gage performance? It could be well associated with people factor, skills, experience and on top of it attitude.
Second most important thing is that Gage R & R is dependent on the process performance, that is, process sigma level to be precise. If you consider the old rule of thumb (which I feel can be applied to any situation), for Gage R & R performance according to the tolerance – R & R 30% is unacceptable and you need to find problem, revisit the fishbone diagram, and remove root causes. But, the problem is that this Old rule of thumb is valid only if we are trying to measure our process with Z = 6. What if your call monitoring process is performing low? In other words “Have you considered what is required for measuring lower performing systems?”
You must know without thinking about the tolerance, what is the relationship between the actual process deviation, gauge deviation and the observed process deviation. Variance ratios (Process variation/Gage variation) of 2 and 20 represent another rule of thumb. With a variance ratio of less than 2, the gauge is unacceptable. With a variance ratio of greater than 20, the gauge is acceptable. In between 2 and 20, be careful! If the measurement variability is close to the same as the observed process variability, dont assume that you have a great process. That conclusion is highly prone to error! (It is statistically demanding to confidently calculate a small difference in two similar, large numbers.) Note that at a Z short term of 6, the 2 and 20 rule for variance ratios is the same as the 10% and 30% rule for gauge performance according to tolerance.
However, as the process being measured degrades, this demands less performance of the measurement system. It could be a good exercise if you havent done if you obtain a relationship between Zshort term and Gage R & R, then plotting graphs (Gage R & R vs. Zst) to get the zones of good, marginal and poor gage R & Rs. Also, note that as the process degrades higher values of Gage R &R expressed as percent of tolerance becomes acceptable. For e.g., with Cp=1.33 and Zst = 4.00, the range for acceptable gage could be 14% – 46%. So, if your process is at 4.00 sigma, then the improvement is truly significant. But if your process is performing at Z=2.00 (Cp=0.67), then the range for acceptance becomes 29% – 91%, in which case it will appear that you have significant improvement (gage r & r reduced to 35%), but in reality you have added little through your efforts.
Benchmarking of Gage R & R is not appropriate I think in any situation. But if we do have sufficient survey data to justify any particular value then I will be happy to know that.
Good luck!Stathem
0December 17, 2003 at 4:09 pm #93653Where do you get this stuff? Everything starting with the Z = 6 is wrong.
0December 17, 2003 at 4:18 pm #93655Let me see your justification please. To arrive that as I have said one needs to work.
0December 17, 2003 at 6:14 pm #93662Say what?
0December 18, 2003 at 2:33 am #93669Dear Stathem,I would like to bring one (often overlooked) point forward regarding the discussion of process performance versus required gage R&R. This point is that it really depends on the purpose of the measurement:
when a measurement is used to compare performance against specification limits (tolerance) the R&R performance needs to be judged against these limits, not against the process. This is called the tolerance% in the R&R output results (at least in Minitab, that is). In this case, there is no link between the required R&R and the process performance.
when a measurement is used to detect differences between products (being calls) , the R&R needs to be judged against the actual process spread. This is called study% in the R&R output results. In this case, there is a clear link between the process performance and the required R&R. This requirement usually is more severe; at least you may hope that your process spread is narrower than the tolerance ;).I think it would be very good if mr. G explained a bit more about the purpose of the measurement for which he is checking the R&R and why he wants to do benchmarking. That would help in better answering the original question.kind regards,
Arend0 
AuthorPosts
The forum ‘General’ is closed to new topics and replies.