iSixSigma

Defects and opportunities

Six Sigma – iSixSigma › Forums › Old Forums › General › Defects and opportunities

Viewing 24 posts - 1 through 24 (of 24 total)
  • Author
    Posts
  • July 17, 2007 at 2:08 pm #47590

    GRG
    ★★★★★★★★★★
    Participant

    Hi all,
    I’m new to six sigma and work in a plastic and metal custom manufacturing shop.  If I have a customer requirement for one metal frame that contains 25 individual metal components that are welded together and each component contains 15 features (through holes, thread holes, slots, etc), then are my opportunites 1 (unit) x 25 individual components x 15 features for a total of 375 opportun ites?
    I appreciate any feedback!
     
     
      
     

    0
    July 17, 2007 at 2:49 pm #158764

    Mikel
    ★★★★★★★★★★
    Member

    Do you do create the metal components or are they purchased parts?

    0
    July 17, 2007 at 3:02 pm #158766

    GRG
    ★★★★★★★★★★
    Participant

    Stan,
    Thanks for replying.  We purchase the raw material and manufacture each metal component.
     Gary

    0
    July 17, 2007 at 5:47 pm #158776

    Ron
    ★★★★★★★★★★
    Member

    To avoid gamesmanship we have adopted the following as our opportunity count.
     
    Three times the bill of material. This is a standrard and we maintain consistency which is what really counts.
     
     

    0
    July 17, 2007 at 6:11 pm #158777

    GRG
    ★★★★★★★★★★
    Participant

    Thanks Ron,
      Sounds like a viable option.  What kind of industry do you serve?
     

    0
    July 17, 2007 at 10:33 pm #158787

    Mike Carnell
    ★★★★★★★★★★
    Participant

    GRG,
    The old formula was parts plus connections. That meant you got one count for picking the correct part and another opportunity for each connection i.e. solder connection, screw, nail, etc.
    You need to use the generalized rules with some degree of caution. They tend to overlook indigenous processes that may be an issue. The up side is to have a serious effect on your DPMO calulation by moving the denominator (opportunity count) you have to have a pretty significant change in the number of opportunities.
    Good luck

    0
    July 26, 2007 at 7:04 pm #159141

    Rick Haynes
    ★★★★★★★★★★
    Member

    Greg, that is a question that has been seldom asked from most six sigma efforts since the DMAIC cam about.
    If a student asked me, I would then ask why do you ask?
    Generally the use of opportunities and DPMO is intended to provide a baseline where multiple products/process quality performance can be combined into a rolled up scorecard metric.  Specifically when the multiple products/processes have significant differences in complexity and or maturity.  The basis for DPMO is a expectation that every defined opportunity has an equal probability of error, which leads to the process/product opportunity count is proportional to complexity. 
    The DPMO process management worked well because it allowed for a quick identification of quality problems (yields) that do not match what is predicted based on the opportunity count.
    Now to your question, as others answered;  Opportunies were originally defined by Motorola as a count of components, and actions taken that can effect yields.  A circuit board opportunities would be 1 per component, 1 for the board, 1 for every solder joint, 1 for every circuit line in the board. 
    As you may guess, this was the most controversial aspect of early six sigma (right after the sigma level).  The dontroversy revolved around people changing the opportunity count as a method to improve DPMO.  If you choose this path, the most important issue is to create a business policy on opportunity counting and then always follow it.  Now when every DPMO reduces, it is better for the business.
    I hope this helps.

    0
    July 26, 2007 at 7:40 pm #159146

    TC
    ★★★★★★★★★★
    Member

    There are valid reasons for reducing part counts, such as improved reliability, but opportunities should not be arbitrarily reduced to improve DPMO score. 
    I agree that you must create a set of ground rules and stick to them.  Changing opportunities should only be allowed when there is an ACTUAL change to the process, such as redesign to reduce components.

    0
    July 26, 2007 at 10:08 pm #159157

    GRG
    ★★★★★★★★★★
    Participant

    TC,
      Great points, but what I suspect is happening is just the reverse.  I have a Customer that is using the formula to rate our quality.  I suspect they are using a part as a unit, but not adding opportunities. 
      I’ll provide an example.  We have a large frame unit that consists of approximately 25 different lengths of stainless steel tubing, 4 plates and 4 casters.  The frame consists of approximately 35-50 welds and has at least 300 threaded and non-threaded holes machined into different locations on the frame.  Would it be correct to assume that there are 383 opportunities?  Wouldn’t a greater number of opportunites work to our benefit, especially if they had an issue with one threaded hole?
      Thanks
      Gary
     
     

    0
    July 26, 2007 at 11:47 pm #159159

    Mike Carnell
    ★★★★★★★★★★
    Participant

    GRG,
    From your customers standpoint the count of 1 makes sense – as much as using DPMO makes sense in that situation. They are not concerned with your manufacturing opportunities. Each unit is an opportunity to supply a good part or a defective (not defects) part.
    The count you are refering to (383) would be something that would make sense internally if your product was being compared with dissimilar products. The operation count is a way of off setting the effects of complexity between various products i.e. if you want to evaluate the performance difference between a sector making light bulbs, another making loans and another making aircraft engines.
    As far as improvement projects on the line it would make more sense to run dpu. If you play with the opportunity counts and Sigma you will find they are pretty insensitive so it is difficult to make them move much. dpu sensitive enough to measure the effects of improvements you are making.
    Just my opinion.
    Good luck.

    0
    July 27, 2007 at 5:59 am #159174

    Chris Butterworth
    ★★★★★★★★★★
    Participant

    Hi GRG,
    I would like to add that the metric is not so important. As Mr. Carnell mentioned, it does not make much difference when you fiddle with the denominator. When a defect is identified (particularly by a customer), the next task is to remove it, not to get hung up on the math.
    There’s a great line from Jack Nicholson in As Good As It Gets – “I’m drowning here and you’re describing the water.”
    Chris

    0
    July 27, 2007 at 6:18 am #159175

    Omashi Sabachi
    ★★★★★★★★★★
    Participant

    You should  not ignore  the  Root-Causation”.

    0
    July 30, 2007 at 1:01 pm #159285

    GRG
    ★★★★★★★★★★
    Participant

    Hi Chris,
      I agree that it’s very important to remove the problem – but our customer ties our PPM into our contract and our supplier evaluation.  It’s important that I understand the industry standard on how the PPM is calulated.  If the industry standard is one part equals one opportunity, regardless of the components or operations, then I’ll have to live with that.
    GRG
        
     

    0
    July 30, 2007 at 1:55 pm #159288

    Mikel
    ★★★★★★★★★★
    Member

    For a supplied part, you get one opportunity. It’s right or it’s not.What PPM level do you need to achieve?

    0
    July 30, 2007 at 2:12 pm #159289

    GRG
    ★★★★★★★★★★
    Participant

    6500 PPM.

    0
    July 30, 2007 at 2:28 pm #159291

    Mikel
    ★★★★★★★★★★
    Member

    What is your current performance and how complex is the product?

    0
    July 30, 2007 at 2:54 pm #159293

    GRG
    ★★★★★★★★★★
    Participant

    Stan,
      We are all over the place.  This years data  – June-4505, May-86957, April-15789, March-3929, Feb-15775, Jan-6596.
      We are a custom fabricator, so our product mix with this customer ranges from a simple part to a frame assembly with 35 components, multiple welds, and 250-300 other features (threaded holes, slots, etc).
    GRG
         
     
     
     

    0
    July 31, 2007 at 4:06 pm #159325

    Lebowski
    ★★★★★★★★★★
    Participant

    Omashi Hibachi,
    Are reading what you post? The post is around the measurement of the defect level. Nobody is talking about causation.
    This must be Marlon Brando morphed into some new level of cosmic unconciousness.
    Lebowski

    0
    July 31, 2007 at 4:21 pm #159326

    iSixSigma-Editorial
    ★★★★★★★★★★
    Keymaster

    As  I  could  remember  Marlon Brando was  a  famous  actor , who has  passed  away  before  few  years.I  think  you  are  mixing up?

    0
    July 31, 2007 at 5:33 pm #159331

    Jim Shelor
    ★★★★★★★★★★
    Participant

    GRG,
    What are the possible consequences for not meeting the DPM requirement?

    Is there a penality clause in your contract?
    Is it a comparison between you and other suppliers that supply the same or similar complicated parts?
    Is it a goal to be met with no particular consequences except in the case of being so much higher than everybody else the customer starts looking for another supplier?
    How is the customer finding the defects they are reporting to you?

    Receipt inspection?  If so, make your shipping inspection at least as rigorous as their receipt inspection.
    During assembly?  If so, is there a way for you to do a test fitup prior to shopping the part?
    You talk a lot about holes (threaded and non-threaded) being issues.  Can you test fit the fastener that will be used in that hole prior to shipping to test for fit and alignment?
    You need to take the possible consequences of failure to meet the DPM requirement into account and decide how much inspection money you are willing to spend to avoid the customer finding defects.
    Regards,
    Jim Shelor

    0
    July 31, 2007 at 5:36 pm #159333

    Mikel
    ★★★★★★★★★★
    Member

    How about finding root cause and putting in a few poke-yokes?

    0
    July 31, 2007 at 5:38 pm #159335

    Mikel
    ★★★★★★★★★★
    Member

    Achieving the target PPM while reducing inspections ought to be fairly straghtforward.

    0
    July 31, 2007 at 5:39 pm #159334

    Jim Shelor
    ★★★★★★★★★★
    Participant

    Stan,
    Another good idea as usual.  Thanks for the help.
    Jim

    0
    August 1, 2007 at 7:01 am #159362

    Sparhawk
    ★★★★★★★★★★
    Member

    In the automotive industry, the PPM standard is for each unit you supply to the customer. As previously stated you had 1 opportunity to supply a good part and you failed.For internal PPM calc, I have had many discussions, again as mentioned previously pick a standard way that makes sense to your organisation and the level of detail you require, and stick to it across the board.

    0
  • Author
    Posts
Viewing 24 posts - 1 through 24 (of 24 total)

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