iSixSigma

Selecting a measure for a DOE

Six Sigma – iSixSigma Forums Old Forums General Selecting a measure for a DOE

Viewing 9 posts - 1 through 9 (of 9 total)
  • Author
    Posts
  • #30464

    JB
    Participant

    I had a guy come to me the other day desperately wanting to have a DOE done on his sonic welder. The problem is with scrap, he says the line has 2.5% scrap caused by sonic burns. My question for you guys would be what can I measure to eliminate the burn but still get a good sonic weld? Right now I am looking at measuring how much “pull strength” it takes to pull the fabric off of the substrate (thats what we are sonic welding). However, this measure really doesn’t seem like it will eliminate the burn issue. Any help would be greatly appreciated.

    0
    #79364

    Robert Butler
    Participant

    Your description of the problem suggests that there are at least two measured responses: occurrence of burn – (this either being a Yes, No response or, if there is some continuous measure of degree of burn, whatever that measure may be) and pull strength of the sonic weld.
    ย  Your design would focus on those variables that you have good reason to believe will impact weld strength and burn occurrence. After the design is run you would build separate models for burn occurrence and for weld pull strength. You can then use these models to hopefully identify a region of minimum burn occurrence and maximum weld strength.

    0
    #79365

    Jack Stuart
    Participant

    The operator already has a measure for Sonic Burn or he would not be able to scrap 2.5%.ย  Unless the process is discrete (You have no Sonic burn at all, then you do), the amount of Burn is continuous.ย  The process, then, requires first an MSA to determine if the “Sonic Burn” is objective or subjective.ย  If subjective, can it be objectivetized (That may be a made up word) or does it need to be rated by several people and the value averaged to make it look objective?ย  In your MSA, somewhere you will have to determine if the sonic burn is a good predictor of bad parts and that would have to be done by showing that the “pull off” test is correlated with the Sonic burn.ย ย  Regression should do that.ย 

    0
    #79372

    JB
    Participant

    Robert,
    Hey sounds like I was somewhat on the right track! My question now is, I am using Minitab to construct my DOE so besides having a column for pull strength (results) do I also make a column for Go-No Go for the burning issue? I haven’t been exposed to DOE’s using attribute data like that.

    0
    #79375

    Robert Butler
    Participant

    ย  JB, that’s correct.ย  You will have two response columns one for the pull strength and a column for yes/no responses.ย  If you can only run the design once you can run a regression against just the -1/1 or 0/1 responses but you should use the results of the regression only as a guide to give you some sense of variables imapacting burn and not as a final definitive predictive equation.ย  If you can replicate the entire design several times you should be able to convert the yes/no responses into probabilities (i.e. say for the first experiment which was run 5 times you have as responses yes,yes, yes,no,no then you can use this to convert these responses into the probability of occurrence of a yes).
    ย  There are a number of caveats concerning either approach which are too lengthly to go into in a single post.ย  If you want to read more on the subject I’d recommend Analysis of Binary Data by Cox and Snell sections 1.1 to 1.4 of the first chapter.

    0
    #79565

    RonJ
    Member

    It seems to me that burn could be directly related to temperature. Can you measure the temperature seen at each weld? Just a thought.

    0
    #79663

    Matson
    Participant

    Be careful!In my experience with sonic welding and burns, it was imperfections in the material, not the welding process.

    0
    #79665

    Mike Carnell
    Participant

    Bruce,
    I am not disagreeing with you that you had material imperfections as a cause. If you are going to help him out he will probably need to know how you determined there were imperfections in the material.
    The other side of that is if you can prove that the imperfections are the cause then the supplier should be doing two things:1. pay for the scrap 2. fix the material.
    In parallel if you need to run the DOE and find the setting that make your process most robust against the material imperfections and you can sort material so you don’t have imperfections and have an optimized process so when your supplier gets “Good” material you are ready to go.
    Good luck.

    0
    #135043

    wg
    Member

    Do you have a way of measuring the output of your sonic welder during the welding operation? You may have intermittent spikes or dips in the amount of power applied to the weld– and that may have a root cause in the welder or in your power supply.

    0
Viewing 9 posts - 1 through 9 (of 9 total)

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