iSixSigma

Gauge RR

Viewing 8 posts - 1 through 8 (of 8 total)
  • Author
    Posts
  • #67753

    Neil Polhemus
    Participant

    The MSA manual recommends the use of the ANOVA method unless a computer is not available. One possible explanation for a dramatic difference between the results of the 2 methods is the optional inclusion of an appraiser by part interaction in the R&R term estimated by the ANOVA method. It would be surprising to see a large difference between the two analysis methods if the interaction term was not included. The presence of outliers could also impact the analyses, but would usually cause the Average and Range method to give a larger R&R estimate.

    0
    #67754

    Neil Polhemus
    Participant

    I did some experimenting, and outliers actually seem to have a bigger impact on the ANOVA method than the Average and Range method (much bigger in the examples I tried).

    0
    #27574

    Anonymous
    Participant

    Is there any reason to ever use the Xbar and R method rather than the Anova method for a gauge r&r study? I have recently gotten results that indicate an effective system from Xbar and a poor system from anova on the same data set. Obviously, the anova method is more accurate and as such, why would I ever want to use Xbar?Thank you in advance for your input.

    0
    #67773

    Anonymous
    Participant

    The only time we used XBar and R Charts was when we didn’t have statistical software.  Then we used an Excell Spreadsheet.  With proper software, I see no reason to use anything other than ANOVA.  Outliers should be examined and perhaps repeated if possible, or if you have more than 3 operators, I sometimes calculate R&R after deleting one operator.  Of course remember you want to know measurement system variation as it actually exists. 

    0
    #67845

    Murray
    Participant

    What defines an outlier in a R&R study?
    Our opinion? The data, if entered properly, is the data.

    0
    #67846

    Neil Polhemus
    Participant

    Now that’s an intriguing question. If someone screws up, is that part of the measurement process? It may be.

    0
    #68300

    kbower50
    Participant

    If there is a significant interaction effect, the Xbar-R computations give lousy estimates of the reproducibility aspect.  Using ANOVA we have ways around it (e.g. dropping the interaction term and refitting the model, as Minitab does if the p-value is >0.25 for the op*part effect.)  You may want to check out my gage r&r paper at http://www.minitab.com/company/virtualpressroom/Articles/index.htm and the references therein, esp. the Montgomery and Runger paper from Quality Engineering. 
    Hope this helps.

    0
    #68313

    anon
    Participant

    If you only have Excel, still use ANOVA. The formulas are straight forward and the graphs are easy. I’ll sell anyone an addin that gives the same output and graphics as Minitab (they are just duplicating the AIAG manual anyway) for $50.

    0
Viewing 8 posts - 1 through 8 (of 8 total)

The forum ‘General’ is closed to new topics and replies.