Within Subgroup Variations in DOE

Six Sigma – iSixSigma Forums General Forums Methodology Within Subgroup Variations in DOE

Viewing 4 posts - 1 through 4 (of 4 total)
  • Author
  • #55599

    Nitin Verma

    How is within subgroup variation addressed in full factorial DOE.
    Eg if a variable X is varied over lower spec and upper spec , the outcomes will impact the result Y in DOE model.
    However the variable X itself has inherent variation which will result in different outcomes in different trial runs.
    How do we address this process variation in variable X?



    Why don’t you provide your thoughts and rationale and we can confirm or guide you where you are wrong. We don’t do homework.


    Robert Butler

    Hint: Find a good book on regression and check the index for the topic concerning regression when there is variation in both the X and the Y.


    Mike Bonnice

    Record the actual value of x in the worksheet in addition to the actual y. After the analysis, make a plot of residuals versus the actual x values. Or residuals versus the difference between actual x and nominal x.

    The bigger problem is that your process is in trouble if the y variable shows reproducible variation across the tolerance range of x. The point of well-chosen tolerances is that the customer will not notice a performance difference across the range of tolerance. Maybe the tolerances are too loose?

Viewing 4 posts - 1 through 4 (of 4 total)

You must be logged in to reply to this topic.