iSixSigma

Gage RR Control charts: x bar chart

Six Sigma – iSixSigma Forums Old Forums General Gage RR Control charts: x bar chart

Viewing 13 posts - 1 through 13 (of 13 total)
  • Author
    Posts
  • #46478

    Rob Duivis
    Member

    I am looking for a good explanation why the plotted points in a gage r&r control x-bar chart should be out of the upper and lower control limits, indicating gage r&r is low. I know it is due to the overall measurement system variation.

    0
    #153664

    Bower Chiel
    Participant

    The way, perhaps naively, that I think of it is this: – When you use Xbar and R charts to monitor a process that has been behaving in a stable, predictable manner and on target then signals on the Xbar chart provide potentially bad news – they provide evidence that the process is “out of control” and in need of some attention. With the Gage R&R scenario you have to have a different mind set. Signals on the Xbar chart are now good news because you want your measurement system to signal differences between parts. The louder and clearer the signals the better the measurement system performance and the lower the %Gage R&R.

    0
    #153668

    Rob Duivis
    Member

    Bower,
    I understand what you are saying;it’s the mind set here, let say i think it is mathematically difficult to understand that a proces what is beyond the upper and lower control limits is actually performing well.
     
    Rob.

    0
    #153688

    accrington
    Participant

    Extracted from the AT & T handbook:
    The Xbar chart shows the discriminating power of the instrument. The control limits, being derived from the error of measurement chart, represent the inability of the instrument to tell one piece from another – i.e., the area over which the instrument is not capable of discriminating. If the Xbar points stayed in control, it would mean that the measuring instrument could see no difference between the pieces of product.
    A good measuring instrument, suitable for measuring this product, should have such narrow control limits that all or most of the Xbar points are out of control. Note this is quite different from the ordinary control chart. In an error of measurement study we want the Xbar points to go out of control.
    Another way of putting it is that the region within the control limits represents measurement system ‘noise’, and points beyond the cotrol limits represent ‘signals’
    You might consider investing in a copy of the AIAG MSA handbook, or a copy of Evaluating the Measurement Process by Wheeler (cross yourself when his name is mentioned in this forum) and Lyday.
    Good Luck!
     

    0
    #153690

    Rob Duivis
    Member

    accrincton,
    Thanks for this explanation, makes it all clear to me.
     
    Rob.

    0
    #153710

    The Force
    Member

    At least most of the data points should be outside the control limits for an xbar chart in a gage rnr study to ensure the ability to detect resolution and/or discrimination

    0
    #177162

    George Lee Sye
    Participant

    Here’s another way to think about it. The data points and the control
    limits are not really connected. The data points show variation
    through the range of part measurements by each operator.
    By contrast, the control limits are related to variation within the
    measurements of ‘each’ part. Not much variation in measurement
    system equates to closer control limits.When it appears out of control, it means the part to part variation (the
    data points) is greater than variation within part measurements (the
    control limits), the desired state.

    0
    #177167

    Craig
    Participant

    The xbar and r chart is monitoring “between” group variation and “within” group variation. The limits are based on the “within” varation, which is based on the range of the subgroups.
    You want to see evidence of assignable cause on the xbar chart, which is the part to part variation that you hopefully designed into the study. If you don’t see assignable cause, this means you picked a poor selection of parts (all the same), or your measurement process has poor R/R and the “within” variation is large.

    0
    #177168

    boettler
    Member

    Thanks George,
    I have been waiting 19 months for your answer and it finally came. Appreciate the feedback. Now, I can get on with my project.

    0
    #177170

    Swaggerty
    Participant

    Rob,Let me know when you have questions about process capability. I’ll
    get back to you in 2010 or 2011.

    0
    #177171

    boettler
    Member

    Thanks George. I will alert the staff here at the nursing home so they allow me to use the computer.

    0
    #177200

    George Leesye
    Participant

    My pleasure Rob, sorry I only read your original message yesterday. I
    was just writing some material on this topic for a re-write of a book I
    published in 2004 and up came your message when I did a quick
    search. Good luck

    0
    #177202

    Mikel
    Member

    George,Not to burst your bubble but that was really Rob. Believe or not there are people on here that like to joke.

    0
Viewing 13 posts - 1 through 13 (of 13 total)

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