Performing Gage R&R on Automated Thickness Measuring Machine

Six Sigma – iSixSigma Forums General Forums Methodology Performing Gage R&R on Automated Thickness Measuring Machine

This topic contains 5 replies, has 4 voices, and was last updated by  Chuck White 2 months, 1 week ago.

Viewing 6 posts - 1 through 6 (of 6 total)
  • Author
  • #56107

    Joe Kelly

    Hello all,

    I have a machine that is used to measure the thickness of flat pieces across the width and length. There are 5 thickness gages across the width (set up by an operator for the different widths of parts), and the system automatically collects a thickness data point, for each gage, every 6 inches down the length of the part.

    We want to look at the accuracy and repeatability of this system.

    My question is, should I perform a Gage R&R for each thickness sensor? Also, since I need a single data point to compare against, should I use an average of the collected data points, or should I use a common data point from piece-to-piece (e.g. 12 inches from the first edge for every part).



    Chris Seider

    Yes, since it sounds like the measurement is measuring up to 5 items across the width–do gage R&R across all separately.


    Mike Carnell

    @cseider Just a thought. If I do them all at the same time and the gage is acceptable I shouldn’t need to do it separately? It might not be a perfect answer but it would be faster and less work.

    If I have 5 gage readings every 6 inches I can see an opportunity for some simple hypothesis testing as well.


    Chris Seider

    Oh, I just meant to do the analysis of each separately–of course, I’d “gather” and do the physical gathering, etc at once. Glad you clarified to Joe K.


    Joe Kelly

    Thanks, guys!

    Follow-up question: If I wanted to do a gage R&R for each sensor, how should I compare part-to-part? Should I average the thickness data points from the sensor and use that to compare between parts (for part variability)?


    Chuck White

    I recommend choosing a common data point along the length. If you use an average, your study will underestimate the part to part variation, since the variance of sample averages is always less than the variance of individual measurements.

Viewing 6 posts - 1 through 6 (of 6 total)

You must be logged in to reply to this topic.