Control Chart?

Six Sigma – iSixSigma Forums Old Forums General Control Chart?

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
  • #49068


    I am beginning to look at a process where plastic packages are sealed.  I gathered data from the last two months for one of our machines.  The seal quality (no good metric on that has been established here yet) is believed to be a function of two variables, the seal temperature and the seal time.  I have 50 pairs of the seal time and temperature to work with, all on the same machine, same film type.  Although the seal formation is done by a machine (the process is relatively stable, no heater malfunctions are evident).  The relationship between the time and the temperature is dismal (r^2 about 0.05 with a linear fit, r^2 = 0.25 even with a 4th order quadratic fit – a fit that is not credible based on the process fundamentals).
    So, the perhaps brute force way to begin to bring order would be to create two Run Control Charts — but that is more work that the production floor might be willing to accept.
    Alternatively, I could try to get persons to fix one of the variables and create a Run Control Chart of the remaining variable.  Likely that is more palatable to operations.
    Another alternative, if it exists:  Is there a control chart designed for two variables at one time?
    Any recommendations on how to bring a bit of order to this situation?  At this time there is no production agreement on how to test the seal quality, nor support that more than experienced judgment / qualitative methods are needed.  Yes, we do have seal failures at a later stage of processing however it is not a major issue.



    To answer your last question first, yes there is a class of control charts called multivariable. Hotellings T is an example. Google that and you’ll get more info than you ever want.
    Have worked with sealed blister packs in the distant past. Somewhat confused by your statement about relationship between time and temperature? What were you doing?
    Seal quality is an issue. We did visuals (with all the attendant attribute data problems) and also a sample of a vacuum test (also attribute pass/fail). Don’t remember the details. The measurement system is very suspect in this.
    BTW, while it makes engineering sense that seal time/temp should be the drivers; they might not be the KPIVs. When you say, “is believed to be a function of two variables”, it sounds like a tribal knowledge thing. While probably correct, I would want to do some experiments on this.
    I always thought that some kind of a pull apart test should be able to be developed to give a force measurement (variable data), but I left the job before we got that going.
    Hope this helps.   



    Thanks for the name to search on, that should help in looking at the possibility for two variables control charts.
    You are right, about the operation.  Meat is being sealed in a pouch composed of two pieces of plastic.
    I agree that the seal is a quality issue, but many here only see it as an operational issue (aside from the hidden factor as the meat is re-sealed).
    Right now I’m just getting the basics of the process.  It could be that time and temperature are not KIVs.  I have removed the variables of plastic types, the machine number, machine manufacturer, meat type, die size, etc. when I collected the 50 data pairs.  I do yet have to investigate the operator (shift) as a potential uncontrolled variable.
    Currently some people test the package seal qualitatively, as not everyone believes that the vacuum testing is adequate — we do have the tester though,  I looked into a pull test machine a bit.  I found that they are designed for Lab use, not the production floor environment, cost was certainly in line with Lab equipment, space is an issue, and training would have been an issue.

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

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