Dismiss Notice
You must be a registered member in order to post messages and view/download attached files in this forum.
Click here to register.

Capability/ Stability Interpretation

Discussion in 'SPC - Statistical Process Control' started by Kristof, Oct 28, 2019.

  1. Kristof

    Kristof Member

    Joined:
    Feb 27, 2017
    Messages:
    25
    Likes Received:
    0
    Trophy Points:
    1
    Location:
    Poland
    Hello All,

    Could somebody help me with capability/ stability Interpretation?

    I generated some SPC statistics, but I have problems with the interpretation.

    Please find attached SPC.

    I appreciate your quick response.

    Gr. Kristof
     

    Attached File(s): 1. Scan for viruses before using. 2. Report any 'bad' files by reporting this post. 3. Use at your own Risk.:

  2. Bev D

    Bev D Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    453
    Likes Received:
    488
    Trophy Points:
    62
    Location:
    Maine
    what are your questions?
    have you had any training in SPC?
     
  3. Kristof

    Kristof Member

    Joined:
    Feb 27, 2017
    Messages:
    25
    Likes Received:
    0
    Trophy Points:
    1
    Location:
    Poland
    Hello Bev,

    I’m so fresh in SPC. The SPC training I have planned in January.

    Our customer requires SPC on 15th CTQ’s.

    My question is related to those Analysis in the scope of the interpretation.

    Thank you in advance.
     
  4. Kristof

    Kristof Member

    Joined:
    Feb 27, 2017
    Messages:
    25
    Likes Received:
    0
    Trophy Points:
    1
    Location:
    Poland
    Could you recommend me some SPC books with SPC interpretation?
     
  5. Miner

    Miner Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    344
    Likes Received:
    260
    Trophy Points:
    62
    Location:
    Greater Milwaukee USA
  6. Bev D

    Bev D Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    453
    Likes Received:
    488
    Trophy Points:
    62
    Location:
    Maine
    SPC is complicated and to be able to interpret the results - and more importantly improve your process - you must have a solid understanding of how SPC works, how to design the subgrouping and frequency correctly and then you need to understand the underlying processes in order to correlate the results to action.

    each of your plotted process is 'out of statistical control'. technically then the capability results are not useful because the processes are not stable.

    my questions would be:
    what is the characteristic?
    what is the process that creates the characteristic
    how is the characteristic measured?
    what is the subgrouping scheme?
    how frequently do you take subgroups
    what types of adjustments rea the operators allowed to make?
    what rules are in place that would trigger an adjustment?
    are there any logs that might help indicate what happened coincident with the out of control events?
     
    tony s likes this.
  7. Miner

    Miner Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    344
    Likes Received:
    260
    Trophy Points:
    62
    Location:
    Greater Milwaukee USA
    Studies 10 to 15 inclusive have insufficient gage resolution to capture the actual variation. It can either over or under estimate the variation.

    Showing spec limits on a control chart is a bad practice, even when it is an individuals chart. The red spec limits overshadow the pale control limits making it difficult to judge, but many of these processes appear to be unstable, which make the capability numbers unreliable.

    I recommend that you pick one of these studies about which to ask specific questions. It is impossible to generalize on 15 studies.
     
    tony s and Bev D like this.
  8. tony s

    tony s Well-Known Member

    Joined:
    Sep 10, 2015
    Messages:
    1,038
    Likes Received:
    777
    Trophy Points:
    112
    Location:
    Laguna Philippines
    It is also noticeable that the measuring equipment used in CTQ14 and CTQ15 doesn't have enough discrimination to detect variation in the measured values. It is not recommended that your measuring equipment can only provide pass/fail or hi/medium/low or less than 5 data categories for a variable data study (e.g. charts for individuals and moving range).
     

Share This Page