iSixSigma

MSA Analysis of Insulation Thickness of Pipe

Six Sigma – iSixSigma Forums General Forums Methodology MSA Analysis of Insulation Thickness of Pipe

This topic contains 9 replies, has 4 voices, and was last updated by  yasemin 5 months, 2 weeks ago.

Viewing 10 posts - 1 through 10 (of 10 total)
  • Author
    Posts
  • #209369

    yasemin
    Participant

    We have an Eddy-current device for thickness measurement of insulation of pipes. We take measurements from 6 different distances and 4 different angles (0,90,180 and 270 degrees) total of 24 values. The nature of insulation has an undulating surface and every point of measurement gives us 24 different values. We would like to conduct a measurement system analysis it but we are not sure how to do it. We put all values on minitab and executed Gage R&R Study (crossed) but results are not acceptable. Then we separated every distances and angles and executed again. This time results are close to perfect.

    Is there a specific method to conduct an MSA for this kind of instrument and part with undulated surfaces?

    Attachments:
    0
    #209373

    Mike Carnell
    Participant

    Sample #2 has 1 distinct category. It doesn’t think it can see the difference in the parts.

    0
    #209376

    Andrew
    Participant

    @MikeCarnell would this mean that the measurement system resolution needs to be improved by a factor of 10?

    @yaseminin I am struggling to understand the example – how many different parts are used, how many times is each measurement repeated? Could you please share the Gage R&R graph if available?

    0
    #209383

    yasemin
    Participant

    @andrew we use 10 parts with 3 operators and 2 repeats. Every part has 24 measurement points (6 distances and 4 angles) and every one is different because of the waviness of the surface.

    0
    #209384

    Chris Seider
    Participant

    consider posting the data also or the graphical output

    0
    #209387

    yasemin
    Participant

    @cseider sorry I forgot to add it

    Attachments:
    0
    #209389

    Mike Carnell
    Participant

    Andrew with 1 distinct category it has no resolution at all. There is no reason to look at any other numbers in the analysis. This is what happens (maybe not in this case) when you run a MSA with parts that are all the same size.

    0
    #209406

    yasemin
    Participant

    @MikeCarnell These result is from all 24 values for one part. If we divided into one point (for example at 50mm and 90 degrees) and analyse it we get acceptable result with more than 1 distinct categories. we confused that one by one result are acceptable but all values together shows us inefficient system

    Attachments:
    0
    #209408

    Mike Carnell
    Participant

    Yasemin If you report on individual positions and can reject on individual positions then I would run the analysis by individual position. I would also go back and stratify the data by position and thickness and look for differences i.e is 50 degrees different other angles.

    0
    #209410

    yasemin
    Participant

    @MikeCarnell thank you Mike. We report as individual values. What about all values in one analysis? Why are they unacceptable in this way? Do you think it is not a right system analysis method?

    0
Viewing 10 posts - 1 through 10 (of 10 total)

You must be logged in to reply to this topic.