iSixSigma

assembly metrics

Six Sigma – iSixSigma Forums Old Forums General assembly metrics

Viewing 8 posts - 1 through 8 (of 8 total)
  • Author
    Posts
  • #28395

    Ovidiu Contras
    Participant

    Problem definition : You have a complex assembly process . The process is not measured , therefore ,no sigma level attached . What kind of metrics can be used in this situation ? Your oppinions are very welcomed !

    0
    #70637

    TCJ
    Member

    Try establishing the Rolled Throughput Yield of the process.  This will help you determine the “hidden factory”, which can be benificial in stratifying areas for improvement.  If the process is not being measured now it is crtical that you establish a solid data collection plan to ensure information is collected correctly and consistently.  Hope this helps.

    0
    #70640

    Ovidiu Contras
    Participant

    So , the data should include something like # of assembly defects . Is there anything else ?

    0
    #70657

    Jaran S.
    Participant

    What’s about time?  Cycle time.
    Measure your assembly time and compare it against expectation of your customers (or expectation of your boss).
    What do you think? Please let me know.
    Jaran S.

    0
    #70659

    Ovidiu Contras
    Participant

    Yes , Cycle Time is an excellent metric . Should I add $$$ spent as a metric ?

    0
    #70671

    Jaran S.
    Participant

    Yes. It is a good idea to add $ spent or $ produced as a metric.
    We use houly rate ($ produced / employee) in our company.
    Jaran S.

    0
    #70678

    Nobrert
    Participant

    It looks like you have a “Boiling the ocean” type project.
    To determine your global process sigma, simply count the good and bad products over a specific sample time (select your time range to include all variability (shift, day of week, …).
    But that does not give your a manageable project.
    To unload a truck of bricks by hand, take a brick at a time.
    Here is a quick method to find the most interesting brick to unload: identify the most important weak link in your subprocesses and make that your project.
    Here is an idea how this can be done quickly in a complex environment:
    Tag your input parts – if possible, number them sequentially. Then identify how many source parts made it thru the end. If that is hard to do, then see if you can find the answer … in the trash cans!
    For each part that didn’t made it into good inventory, identify the cost of manufacture and multiply by the number of failures in your sample.
    Now, focus on the failure that costs most to the company and translate this into your ‘Y’. You will not only have a manageable project with a maximal return now, but likely a wealth of future projects to tackle thereafter, with the right priorities set.
    Go for it, tiger !
     

    0
    #70680

    Ovidiu Contras
    Participant

    Thank you ,everybody ! That was really awesome !

    0
Viewing 8 posts - 1 through 8 (of 8 total)

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