iSixSigma

Missing or incomplete data on control charts and effect on limits

Six Sigma – iSixSigma Forums General Forums General Missing or incomplete data on control charts and effect on limits

Viewing 2 posts - 1 through 2 (of 2 total)
  • Author
    Posts
  • #53629

    Savage
    Participant

    I am looking for reference material on the best way to calculate control limits for X-bar and R charts as well as X-bar and Sigma charts. I think I remember seeing an ASTM reference years ago, but cannot locate this. Is there an issue in the Journal of Quality Technology that might discuss this?
    The specific situation is that my sample size is typically five, occasionally I have three or four observations and I want to ensure that the control limits are correct. (No, I do not use Minitab.)

    0
    #190926

    Strayer
    Participant

    The UCL is the mean plus 3 times the standard deviation. The LCL is the mean minus 3 times the standard deviation. You can use functions in Excel. With such small sample sizes your level of confidence will be very low. The numbers may be “correct” in the context of your samples but very off the mark for the process you are measuring. Yeah, there are some references on extrapolation from incomplete or missing data but I don’t think they do much to improve confidence. Look up how to calculate Z values and understand that you want at least 95% confidence level.

    0
Viewing 2 posts - 1 through 2 (of 2 total)

You must be logged in to reply to this topic.