711949_en-US

liliconcom

GRR result from JMP is not matching web example

I was following the example described in this link

https://www.spcforexcel.com/knowledge/measurement-systems-analysis/variable-measurement-systems-part-4-gage-rr/

attempting to use their dataset with JMP MSA.

1. Copied their sample data table into JMP data table

2. Stacked the data table so the parts are in column "Label" and readings in column "Data"

3. Analyze > Quality and Process > Measurement Systems Analysis

4. Response = Data, Part = Label, Grouping = App, MSA Method = Gauge R&R, Model Type = Crossed, Sigma = 6, Alpha = 0.05, Analysis = Use REML. Show metadata = No, Use spec limits = Off

5. From Variability Gause Analysis for Data menu > Gause Studies > Gauge R&R

6. Tolerance Range = left blank

 

Results for "% of Total":

Guage R&R = 10.94

Repeatability = 5.22

Reproducibility = 5.72

Part-to-Part = 89.06

 

GRR 10.94% does not match the web page example of 43.3%

What am I missing?

Thanks in advance.

P.S. Using JMP v17.0.0

Basic Data Analysis and ModelingRe: GRR result from JMP is not matching web example

I didn't manage to get that big of a difference with the results (there was a small difference which could be maybe caused by rounding in that website). Are you sure you are looking at the correct results?

Re: GRR result from JMP is not matching web example

I have a typo above, 43.2% should have been 34.4%

If I had selected Bayesian analysis, then the % Gauge R&R matches.

Perhaps I am not understanding the JMP results...

34.3% Gauge R&R vs 11.79 % of Total?

Re: GRR result from JMP is not matching web example

My selections

Right click on the table box under Gauge R&R and enable % Total (it might also be a good idea to show and compare variance components)

 

If there are something you aren't understanding, select "?" tool from toolbar and click on some part of the platform and it will open Help page for you  Quality and Process Methods > Variability Gauge Charts > Variability Gauge Analysis Report Options > Gauge R&R Report  and then go through the documentation to understand the report better. Most likely you are comparing values which aren't the same (% Total vs Sqr % Total).

 

Re: GRR result from JMP is not matching web example

From the reference you cite:

"Note to our readers: this publication discusses the average and range Gage R&R methodology.  It is still used today by some, but more accurate information is gained by using the Evaluating the Measurement Process developed by Dr. Donald Wheeler or the ANOVA method.  More information on these techniques can be found in the list on the right-hand side of this page."

JMP uses REML to estimate the variance components.