Measure the uncertainty

Six Sigma – iSixSigma Forums Old Forums General Measure the uncertainty

Viewing 2 posts - 1 through 2 (of 2 total)
  • Author
  • #26997


    i have a measurement system and i want to know the uncertanity of the gage
    is it the bias-the mean of the data?

    thank you very much


    Jim Parnella

    You need to be a little more specific here on what you are looking for.

    You mentioned the “mean of the data”, so I can tell you this. If the mean of your data (say on a reference sample) is the equal to the accepted value of the reference sample (within some pre-defined +/- tolerances), then you can say that the measurement system is “accurate”.

    You used the term “bias”. If the measurement system is accurate (as defined above), then it is also unbiased. I like to think of bias in a measurement system as one in where the average value is CONSISTENTLY AND PREDICTABLY HIGHER OR LOWER than the accepted value.

    With respect to “uncertainty” which you also mentioned, this deals with precision (both repeatability and reproducibility), for which you will have to do a “gage study” (Gage R&R), or better yet, implement a control chart for the measurement process.

    If you don’t know how to do a R&R study, you might be able to search the internet using these search phrases:
    Gage R&R
    Gauge R&R
    Repeatability and Reproducibility
    Gage Capability
    Gauge Capability
    Precision to Tolerance Ratio

    A number of quality and statistics textbooks have chapters on these techniques. Also, some software has the capability to perform the analysis, and their Help files might provide enough information for you to get by.

    Good Luck!

Viewing 2 posts - 1 through 2 (of 2 total)

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