Gage R&R using only 2 appraisers and 2 readings??

ScottBP

Involved In Discussions
#1
Hi all,

We are a commercial calibration lab, and one of our customers is requesting that we do a R&R study on everything they send in, and they want it done using only 2 appraisers and 2 readings each.

1. Every GR&R I've ever seen has at least 3 appraisers for 5 readings each (3 x 5 for short.) Is it possible to even get valid results using only a 2 x 2 study?

2. This is the first time we've ever had to do a GR&R for a customer. Since our calibration techs are NOT the same people who will be using the gages in the field, doing it in the first place seems kinda pointless. Maybe the customer is trying to gage our effectiveness in calibrating their test equipment? First we calibrate the gages, then we do the R&R study. Shouldn't they be wanting us to give them an uncertainty report instead? After all, we are an ISO 17025 accredited lab.

3. Among the usual things (e.g. micrometers, scales, etc.), they are wanting us to do R&R studies on digital multimeters and insulation testers. How the heck do you do a R&R study on a high voltage insulation tester?? :mg:

Sounds to me like the customer has been told they need to do GR&R studies by an auditor, and doesn't have the resources (or know-how for that matter). The guy who has been bringing their equipment to us for years admits even he doesn't know why his company is wanting us to do GR&Rs for them or how we are to do it. Is there some sage advice we can give them, like what equipment is normally subject to a GR&R, the minimum number of readings to take, who gets to do it, etc.? :frust:
 
Elsmar Forum Sponsor

Stijloor

Staff member
Super Moderator
#2
Hi all,

We are a commercial calibration lab, and one of our customers is requesting that we do a R&R study on everything they send in, and they want it done using only 2 appraisers and 2 readings each.

1. Every GR&R I've ever seen has at least 3 appraisers for 5 readings each (3 x 5 for short.) Is it possible to even get valid results using only a 2 x 2 study?

2. This is the first time we've ever had to do a GR&R for a customer. Since our calibration techs are NOT the same people who will be using the gages in the field, doing it in the first place seems kinda pointless. Maybe the customer is trying to gage our effectiveness in calibrating their test equipment? First we calibrate the gages, then we do the R&R study. Shouldn't they be wanting us to give them an uncertainty report instead? After all, we are an ISO 17025 accredited lab.

3. Among the usual things (e.g. micrometers, scales, etc.), they are wanting us to do R&R studies on digital multimeters and insulation testers. How the heck do you do a R&R study on a high voltage insulation tester?? :mg:

Sounds to me like the customer has been told they need to do GR&R studies by an auditor, and doesn't have the resources (or know-how for that matter). The guy who has been bringing their equipment to us for years admits even he doesn't know why his company is wanting us to do GR&Rs for them or how we are to do it. Is there some sage advice we can give them, like what equipment is normally subject to a GR&R, the minimum number of readings to take, who gets to do it, etc.? :frust:
Is your customer an automotive supplier?

If so, they can not outsource this obligation because the MSA study is supposed to be performed according Customer Specific (PPAP) Requirements using their own parts made during manufacturing process validation, using their own appraisers, and their own measurement system.

The purpose is to assess variation in the organizations OWN measurement system, not somebody else's.

Does this help?

Stijloor.
 
D

Dr. Electron

#3
2 x 2 is way too small. The cardinal rule is Operators X Sample X Trials > 60. Recommended is 3 x 10 x 3.

I guess if you did 15 trials you'd at least get something that showed you whether you're in the ballpark or not and how much variation you're getting. As a formal GRR study though...I'd still be weary...as 2 x 2 doesn't present enough degrees of freedom to produce reliable numbers.

Also agree with the previous post...the appraisers should be the ones at the manufacturer who are actually measuring the product with the actual machines present there. I suspect, however, that they just want a "reality check" to see what your machine repeatability is like and are calling it a Gage R&R.
 

Miner

Forum Moderator
Staff member
Admin
#4
There are a multitude of reasons why the customer should not ask this of you:

  • You are not measuring the production parts, so your results will not be representative of actual results
  • You will not have the estimate of Product Variation, so no %GRR (%PV)
  • You will not have the Tolerances, so no P/T Ratio (%TV)
  • Your operators will not be representative of the customer's operators

A GR&R study makes no sense. A Linearity or stability study would make sense if they had to have some type of MSA.
 

Stijloor

Staff member
Super Moderator
#5
Friends,

Excellent responses form Dr Electron and Miner! :applause:

Now I am curious what the OP will say. Is their Customer indeed an automotive supplier? :mg:

Stijloor.
 

bobdoering

Stop X-bar/R Madness!!
Trusted Information Resource
#6
I agree - a calibration house really has no place doing gage R&Rs. Lab-to-lab, yes. Not gage R&Rs. I think ScottPB hit the nail on the head.

Sorry, some things you just can't outsource. You can have somebody else get you a beer while you are watching the game, but you are on your own to handle the end results of drinking it. :cool:
 

Bev D

Heretical Statistician
Staff member
Super Moderator
#7
Agree that the request from your customer is odd.

that said the use of 2 readings per part per operator is far better than the "standard" method promoted by AIAG and others. Provided of course that enough parts are measured.

let's think about what we're trying to estimate here. Measurement error is a measure of variation (standard deviation or variance) it is not a mean. The accuracy of an estimate of a standard deviation is related to the sample size just as an estimate of the mean is. However, the estimate of a standard deviation is far more inaccurate than for a mean with the same sample size. The improvement in accuracy of the estimate when going form 2 readings per part to 3 readings per part is negligable when compared to the improvement that is gained by taking 30 samples (parts) of 2 readings rather than 10 samples (parts) of 3 readings.

we could improve our estimate of the standard deviation by taking 30-50 repeated measurements of a part BUT we would absolutely lose two things: The first is that we would miss any interactions between the part size and the measurement technique - and this isn't a trivial interaction in many industries, processes and products; ya gotta think past the automotive metal and plastic stuff. Secondly, it becomes difficult to get truly independent values for manual processes when there are very few parts and many readings - everybody knows what value they are supposed to get...so all in all 2 readings from 30 or so parts gets us around the accuracy problem.

The number of operators is a different story - we're trying to assess operator to operator differences. how many operators do you have? how many do you need to test before you know if a difference exists? is 3 better than 2? what if you have 20 operators? For operator qualification of critical measurement systems I test them all. annually. For non critical systems, I only qualify the system. depending on the complexity and the number of operators that may use it I will test 2-4 operators to ensure that no significant operator to operator variation is inherent in the measurement system and then I rely on simple training of how to use the system to maintain reproducibility.


There are plenty of articles (that have been peer reviewed) written that explain the alternative approach to what the automotive industry has perpetuated over the years. These articles are clear about the superiority and utility of the approach (as opposed to the shanting mantra of "AIAG says so"...). My experience with both methods and emperical side by side studies that I've run (to assure myself) I have found the 2 measurements per 30 parts plotted on Youden chart to be teh most value add and relevent test...

by the way a long time ago in a galaxy far, far away Western Electric described a similar 'control chart' method in their SQC book...
 

ScottBP

Involved In Discussions
#8
Hey thanks for the input, everybody. This is the sort of info we need before we get in too deep over our heads... Our customer is a large government contractor that manufactures land and marine military vehicles, so you would think they would know better. Like Dr. Electron says, I think they want to know how good our neasurement process is, but that's why we pay an accreditation body to come in every year and gage our methods for ISO 17025 accreditation. But then again, government contractors :tunnel: have very strange requirements, like the ones who want all the data taken on their gages to be hand-written onto their forms (with carbon paper in between duplicates) instead of accepting the computer printouts we normally provide. :eek: :bonk:
 
Thread starter Similar threads Forum Replies Date
B Informational Expanded Gage R&R Analysis using Minitab Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 18
Q How can you justify using a more accurate Pin Gage class? General Measurement Device and Calibration Topics 16
B Gage R & R on Visual Inspection using the Cross Tabulation Method IATF 16949 - Automotive Quality Systems Standard 4
S Destructive Gage RR - Using Crossed - want your thoughts Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 18
F Comparison of 2 CMM machines using Gage R&R Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 7
K Gage R on a 10 Piece Sample with a CMM using Minitab Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 2
J Implementing "Gage Control Software" Is anyone using this program? Calibration and Metrology Software and Hardware 12
M Experimental Design at Nested Gage R&R using Minitab 16 Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 2
A Gage R&R without Operators (using Sensors) using Minitab Using Minitab Software 2
C Using a Pin Gage to check True Position of Inline Holes, Design, Tolerance. Etc Inspection, Prints (Drawings), Testing, Sampling and Related Topics 1
J Variation (using ANOVA method) Required for Gage R&R Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 3
Q Mitutoyo USB Cables for Variables Data Collection using a Dial Snap Gage Statistical Analysis Tools, Techniques and SPC 2
M Gage R&R for Big Parts - Approving a Variable Gage using only 1 Part Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 7
B Attribute GR&R (Gage R&R) using a CMM (Coordinate Measuring Machine)? General Measurement Device and Calibration Topics 3
B Minitab Gage R&R using Defect Count per Unit Area Using Minitab Software 1
E Gage R&R or Anova two way? Using Minitab Using Minitab Software 1
S Using the results from GR&R (Gage R&R) Analysis - One sided dimension Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 25
Miner Intro to MSA of Continuous Data – Part 7: R&R using Wheeler’s Honest Gage Study Imported Legacy Blogs 17
Proud Liberal Using Three (3) Gage Pins to Check Hole Inspection, Prints (Drawings), Testing, Sampling and Related Topics 5
M Checking a male taper using a female air gage fixture - Air Gaging question Inspection, Prints (Drawings), Testing, Sampling and Related Topics 5
M Gage R&R (Reproducibility and Repeatability) using Minitab 14 Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 3
C Bridgeport Mill as a Gage - Verify gage dimensions using a Bridgeport Mill Measurement Uncertainty (MU) 4
J Gage R&R on Attribute Data Using Minitab 14 - Data from Automated Test Systems Quality Tools, Improvement and Analysis 2
P Attribute Gage Study using MiniTab Using Minitab Software 2
N Gage R&R - Using Tolerance method, or part to part variation? Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 1
J Using Gage Pins - Inside diameter - Interference fit General Measurement Device and Calibration Topics 4
O Gage R&R when using Tolerance - Really bad Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 18
S % of Contribution or % of Study Variation? Gage R&R using Minitab Using Minitab Software 26
K Gage R&R - Rockwell Scale - Hardness using Standard Test blocks Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 5
G Gage Calibration - Using a CMM and Gage Blocks for In-House Calibration ISO 9000, ISO 9001, and ISO 9004 Quality Management Systems Standards 1
J Gage R&R Software - Does anyone have experience using Mitutoyo's "MeasurLink"? Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 5
T Gage R&R Using a Unilateral Tolerance - Only a minimum limit is specified Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 1
R Guardbands - Using Gage Studies to define a guardband Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 1
T Plug Gage Calibration Calibration and Metrology Software and Hardware 1
L Gage R&R TMV Acceptance Criteria Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 4
S MSA for attribute relation gage Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 2
E Zero part to part variation - Gage R&R project Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 15
M Gage R&R and right way to measure Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 16
L Gage RandR on automated equipment. IATF 16949 - Automotive Quality Systems Standard 7
B Gage calibration frequency, ISO and IATF - What are the requirements Calibration Frequency (Interval) 3
D Difference between Test Method Validation and Gage R&R Qualification and Validation (including 21 CFR Part 11) 18
R Determining Uncertainty from Gage R&R Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 1
C Correct Calibration Method for Dial Depth Gage General Measurement Device and Calibration Topics 6
C Gage Block Wringing General Measurement Device and Calibration Topics 3
B Gage R&R with NDC=1 Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 19
S Type 1 Gage R&R or something else? Reliability Analysis - Predictions, Testing and Standards 6
G Should I perform Gage R&R only at the beginning of a new project? Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 6
V Thread Plug Gage Pitch GO Diameter out of spec AQL - Acceptable Quality Level 8
G Gage R&R - Where am I going wrong? Part of a FAIR submission (Aerospace) Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 2
R Gage R&R Excel templates Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 3

Similar threads

Top Bottom