Type 1 MSA on tape measure

Jayfaas

Involved In Discussions
Hello all. I came across a predicament and wanted to get some feedback on it. Our internal documents allow for us to use tape measures for dimensions where the tolerances are 5 mm total range or above. Therefore, we have to run MSA studies on it. Our gauge studies are required to be, at least type one and also type two or three depending on the machine. Doing a type two study would not be a problem, but how would you run a type one study on a tape measure where the likelihood of any variation is a little to none. If we are going to be using a Reference value, I figured we could just grab one of our ceramic gauge blocks or something and use it, but I get the feeling that we will likely not have any variation and therefore will not get any results from the study.

Normally, you would say that you don’t have enough resolution, but the tolerances on the features we measure with the tape measure are usually +/-5 mm. it would seem silly to go out and have to buy tape measures with half millimeter resolution for measuring that kind of tolerance range. Any thoughts on this? Are there ever times where a type one is not feasible because of cases like this?
 

Miner

Forum Moderator
Leader
Admin
I would not use a Type 1 study because many people do not use a tape measure correctly. Resolution is critical to getting an accurate R&R study. If the resolution is poor, the R&R results are basically worthless.
 

Jayfaas

Involved In Discussions
Is there something out there that governs study types in the auto industry? I wish we could just use calibration as a factor and say a Type 1 is not required, but I am not sure if we can do that
 

Miner

Forum Moderator
Leader
Admin
There is the AIAG MSA manual, but you seem to have a misunderstanding of the difference between calibration and R&R. Calibration addresses accuracy (sometimes called bias) and possibly linearity if the calibration program was well designed. R&R addresses precision or measurement variation. These are two different aspects of a measurement. Accuracy is the mean of a measurement distribution while precision is the spread of the measurement distribution. So, you can have a very accurate (calibrated) measurement devices with totally unacceptable variation (R&R).
 

Mikey324

Quite Involved in Discussions
Maybe wrong here, but for a tape measure, I would be inclined to use an attribute study. Looking at appraiser system % effectiveness score Vs reference, Kappa, and crosstabulation.
I think a tape is accurate enough to tell the difference between 400mm and 410mm. Just not enough to trust 400mm and 400.70. So I would say the measurement gives you more of a "yes it is x"; than a "x = 400mm"
 
Top Bottom