iSixSigma

P chart blues

Six Sigma – iSixSigma Forums Old Forums General P chart blues

Viewing 9 posts - 1 through 9 (of 9 total)
  • Author
    Posts
  • #39067

    Z
    Member

    Heres the problem. We manufacture small parts which usually hold to tolerance. when it comes to final inspection a lot of them are scrapped since they fail visual inspections (attributes). right now we have no criteria for passing or failing a part. its subjective. We need to set up nP charts. Company is against setting a criteria since it feels that setting up criteria for pass/fail will put blinders on inspectors who will not catch defects other than those mentioned in the criteria. please advise
    thanks in advance

    0
    #118035

    ancien opus
    Participant

    We donΒ’t know what we donΒ’t know. If we canΒ’t measure it, we really donΒ’t know much about it. If we donΒ’t know much about it, we canΒ’t control it. If we canΒ’t control it, we are at the mercy of chance.

    0
    #118577

    Payne
    Participant

    Is there anyway to collect what is scrapped and show management the lost revenue due to scrapping “good” parts ?Β  There must be some kind of standard already since the parts are checked.Β  granted it could be totally subjective but that is very rare.Β  Try to document what makes a part defectiveΒ and document with visuals.Β  I had a similar situation with the visual inspection of solder joints.Β  We needed to have pictures of good solder joints and bad solder joints.Β  Employees liked having the standards shown and we didn’t have a quality melt down; we caught more defective joints and were able to reapir on the fly.

    0
    #118579

    Kelly,
    I don’t know what is ‘standard.’ I try to use what makes sense for my particular situation – obviously, it makes sense to use statistics when you’re not in a position to describe each part!
    My approach has been to use an np chart. The reponse is just the number of parts lost. According to Taguchi, a loss is due to a lack of quality. Now in you’re operation ought to have a target – the number of parts you should be able to manufacture per day.
    Now you’re in a position to record all your losses per day – some losses will be due to scrap, some to component unavailability, some due to operator sickness, etc. It is a simple matter to convert these to financial losses and may even justify moving to Ikko Nagashi, or single flow, which takes a different approach. I have a VB program that does all this and more – unfortuantely it is written in VB3, but could be converted with new ‘components.’
    You can also use SPC limits since losses are always a proportion of the target, so that the sample size remains constant. If you do this you’ll find that some bad weeks and good weeks are just random variation around a given natural process performance; then when you make a real change, you’ll see the benefit.
    You might also be able to justify moving to Ikko Nagashi or single flow, or even Lean (TPS) provided you have sufficient discipline to stop the line – that really makes things happen!!!
    If this doesn’t help, I’m sure someone else will be able to describe a more suitable method.
    Good luck,
    Andy

    0
    #118593

    Scott
    Member

    Z’s response is correct. It is a copout of leadership not to provide operational definitions of quality and expectations of everyone in achieving quality. And, the resources for them to do their job.
    Β 
    In addition, the quality should be assessed by the persons in the process. They should have the training and resources to assess their won quality.
    If this is not convincing enough, collect some data.
    The data to collect is the ability of the “Experts” to repeat and reproduce their judgments of quality. Perform an attribute agreement analysis (Minitab has a really easy process) with these experts. Use samples that represent the difficulty of finding multiple defects per service/unit. Share the results. If the experts fail, lesson learned. If they pass, then use AAA to certify the people in the process are capable.
    Β 
    Β 
    Β 
    To convince

    0
    #118596

    Kelly’s answer is correct. Sometimes we have to take responsiblity ourselves and provide a solution. The next time the leadership will be more supportive.
    Andy

    0
    #118597

    Marz
    Participant

    A couple of comments:
    1.Β  By default you have standards in place for pass fail.Β  The problem is that they are not sytematic or repeatable in their current form.Β 
    2.Β  Also if most of the initial inspection rejects forΒ attribute failure are later dispositioned as “Use-As-Is” then you are just waisting time and energy.
    Recommendations:
    1.Β  Take photographs of the rejects and later group the rejects that were dispositioned as Scrap, Rework, Repair, Use-As-Is.Β  This will provide a more consistent guide and should make the process more efficient.Β  It can also be a great training tool for newer employees and great material for refresher training, if you do that kind of thing.
    2.Β  Classify the document as an inspection guide and not a standard.Β  This will leave the inspectors and operators with the feeling that they are not bound to the insturction and can reject material for other/defferent/new defects as they occure.Β  Also management might feel less constrained with this approach.
    Hope this helps – feel free to ask any additional questions
    Kevin
    Β 

    0
    #118614

    Jonathon Andell
    Participant

    First you need to establish – from a customer’s viewpoint – what differnetiates an acceptable from an unacceptable product. Then you can create inspectio criteria, train your inspectors, and validate their ability to conduct consistent inspections. Only then is it worth your while to chart the data.

    0
    #118617

    DANG Dinh Cung
    Participant

    Bonjour,Thank you, ancien opus, for those four sentences I immediately write down in my pocket-book for remembering them by heart.The point of Z’s boss is acceptable. But it is not quality control because you can say something is OK or not only by referencing to a standard.Regards,DANG Dinh Cung
    [email protected]

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

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