Statistically Comparing Technician Techniques

[NCYankee]

Is there a good way to compare two technicians measurements against each other?

The situation I have is this: I have one technician with whom management is not convinced in her measurement techniques. Right now, as a QE, mgmt. is making me go back in behind her and measure the same part again. The big problems here are that we're measuring an already machined die cast transmission main case and we're using a Romer arm to lay it out. I really have many other things I could be doing rather than going in behind a technician and remeasuring.

Is there some mathematical way to compare her measurements vs. my measurements, and maybe against the drawing specifications? Thanks in advance.

[Stijloor]

Originally Posted by NCYankee

Is there a good way to compare two technicians measurements against each other?

The situation I have is this: I have one technician with whom management is not convinced in her measurement techniques. Right now, as a QE, mgmt. is making me go back in behind her and measure the same part again. The big problems here are that we're measuring an already machined die cast transmission main case and we're using a Romer arm to lay it out. I really have many other things I could be doing rather than going in behind a technician and remeasuring.

Is there some mathematical way to compare her measurements vs. my measurements, and maybe against the drawing specifications? Thanks in advance.

Hello NCYankee,

Welcome to The Cove Forums!

This appears more like a competency issue. Is there evidence that this person is indeed competent to perform the required tasks?

In addition, you may want to perform a Gage R&R. Search this Forum for applicable examples.

Stijloor.

[Jim Wynne]

Originally Posted by NCYankee

Is there a good way to compare two technicians measurements against each other?

The situation I have is this: I have one technician with whom management is not convinced in her measurement techniques. Right now, as a QE, mgmt. is making me go back in behind her and measure the same part again. The big problems here are that we're measuring an already machined die cast transmission main case and we're using a Romer arm to lay it out. I really have many other things I could be doing rather than going in behind a technician and remeasuring.

Is there some mathematical way to compare her measurements vs. my measurements, and maybe against the drawing specifications? Thanks in advance.

Are you familiar with the general concept of Measurement System Analysis (MSA) and GR&R (Gauge Repeatability and Reproducibility)? It's what you're looking for. Do a search of the Cove--there's lots of information on the subject here. You may also want to avail yourself of a copy of the AIAG MSA manual, which details the generally-accepted procedures for conducting measurement system studies.

[NCYankee]

Thank you both, Stijloor & Mr. Wynne. I am familiar with MSAs and GR&R. If I have to do it using one of the previously mentioned methods, I will. However, I was hoping there was something I could use with the data we already have in hand: Same piece, same dimensions checked, but only two persons, not the three people usually done by MSA & GR&R.

[CarolX]

Originally Posted by NCYankee

However, I was hoping there was something I could use with the data we already have in hand: Same piece, same dimensions checked, but only two persons, not the three people usually done by MSA & GR&R.

Hi NCYankee,

By all means - use the data you have on hand. Most of my customers who request a GR&R only require 2 operators.

[NCYankee]

Thanks CarolX. I'm presently searching for the GR&R & MSA stuff as previously suggested.

[Jim Wynne]

Originally Posted by NCYankee

Thank you both, Stijloor & Mr. Wynne. I am familiar with MSAs and GR&R. If I have to do it using one of the previously mentioned methods, I will. However, I was hoping there was something I could use with the data we already have in hand: Same piece, same dimensions checked, but only two persons, not the three people usually done by MSA & GR&R.

You can do GR&R for two operators, and use existing data, so long as you can compare the results for both operators measuring the same part each time. If you don't want to do a new study (which I recommend), you can simply chart the data you already have to compare the measurement results of the two operators. This will probably just tell you what you already know, however. You really should compare the operators' results measuring the same parts, however.

[NCYankee]

We are measuring the same parts. Thanks again Mr. Wynne.