Gauge R&R Head-Scratcher

Six Sigma – iSixSigma Forums General Forums Tools & Templates Gauge R&R Head-Scratcher

This topic contains 2 replies, has 3 voices, and was last updated by  Mike Carnell 5 months, 1 week ago.

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
  • #56069


    I recently conducted a Measurement Systems Analysis exercise regarding the measuring of racking locations in a Warehouse. A team of operators are measuring the dimensions of the racking for a project we are working on.

    The team did very well for repeatability and reproducibility, and, even without running the numbers, it was clear that they were very good. With only two exceptions (where there was an operator/part interaction effect caused by a location being unusually tall and an operator unusually small), every single measurement was identical both across operators and on the two measurements.

    During the exercise, I took great care to control for collusion – ensuring that the operators couldn’t observe their colleagues measuring, or see the numbers they were writing down. I’m confident I did a robust, correct test.

    I went away feeling very cheered. Then I saw the measurements which had been taken by the same operators of the same locations during the actual measuring exercise. I was shocked to see that these measurements were significantly different from the measurements taken during the MSA exercise – by about 2cm in each case.

    This has left me scratching my head. I’m thinking either that this was a Hawthorne effect – where the operators measured particularly carefully because I was observing them; or else, the actual dimensions of the racking have changed in some way (not as daft an idea as it sounds).

    How should I respond to this? What experience do people have of similar situations?



    What tools were used to make the measurements? 2cm is not an untypical error if measuring across a large warehouse. If tape measures then bad alignment or being inconsistent with push/pull leeway could easily account for this.


    Mike Carnell

    @8sigma The MSA is a very controlled environment. You need to verify your spaces are the same size they were previously. I wouldn’t even worry about it until then.

    Remember you are qualifying the measurement system (device and operator) You did that and know that the device you used and the people you used were good. That is a fact. It is capable.

Viewing 3 posts - 1 through 3 (of 3 total)

You must be logged in to reply to this topic.