iSixSigma

Variation

This topic contains 3 replies, has 3 voices, and was last updated by  Michael Schlueter 15 years, 6 months ago.

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
    Posts
  • #32856

    Anonymous
    Participant

    I have a question for the bright minds who read this forum:Formed parts are made, 20 at a time from a die that forms a compound into disks. These disks are then fed to another piece of equipment that processes the part into the next step.Variation may occur across the disk and out of spec thickness at any spot can cause a shutdown at the next station. QA people normally measure the disk in 12 spots around the perimeter and toward the center of the disk because of the nature of the compound and the “grain” orientation.The question of sample size came up and someone suggested that we would need Five (5) samples. If we assume five per cavity, then we realistically have a total sample size of 1,200 when we consider the total measurements.An alternative was suggested: we consider each measured spot on a disk as a sample (therefore 12 samples per disk). A single forming operation would provide 20 disks, and therefore a total sample size of 240.This will provide insight into variation within the die and within each disk. It will not provide information regarding variation within a cavity.Because the experts in the process expect that little variation exists from disk to disk (within the same forming cavity) outside of special causes, this may be the best way to go initially.Any thoughts would be appreciated.

    0
    #88226

    Rick Pastor
    Member

     
    Anonymous:  Let me tell you what I would do assuming that the following has not already been done.  Measure the 13 point (12 perimeter, 1 center), keeping track of their (x,y) position in the disk and the (x’,y’) position on the perimeter or (r,q).  You could make a series of contour plots that would give you a visual impact of the uniformity across the die.  Making these measurements on multiple shots provide a statistical comparison of the locations within the die.  (260 measurements per shot with the selection of shots taking place over a period of time)
    The pointed that I am making is that I would want to be sure that there was no systematic behavior in the machining of the die that would cause some disk to go out of control more than others.  This is a lot of work; but it is better to improve the process and remove defects than capture them afterwards.  If you cannot improve the die, then you could at least look more carefully at high risk disk.
     
        

    0
    #88241

    Michael Schlueter
    Participant

    Hi,
    Two remarks.
    Sounds like your experts >assume< little disk to disk variations. Assumptions can be wrong and should be verified. I myself dealt with products which had little in-batch but high batch-to-batch variability.
    I think you are mainly interested about risks, are you? If so, I find Dr. Taguchi’s Online Quality Engineering very useful. It is a loss-based estimate, which takes into account cost factors, like for measurement or belt-stop, sampling frequency, allowed tolerances etc. You can use it to balance quality with total cost. It will show you, how good (or bad) the suggestions for sample sizes are in terms of quality and money. – Ususally one aspect is overdone (e.g. taking too many samples), while another is ignored (e.g. taking samples too frequently). It depends on your specific situation, which the estimate will show you.
    This technique has been applied, e.g., at ITT, Nissan or for adjusting environmental measurement stations.
    Best regards, Michael Schlueter

    0
Viewing 3 posts - 1 through 3 (of 3 total)

The forum ‘General’ is closed to new topics and replies.