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.

Lower control limits

Discussion in 'SPC - Statistical Process Control' started by Davide, Jan 20, 2020.

  1. Davide

    Davide New Member

    Joined:
    Jan 20, 2020
    Messages:
    1
    Likes Received:
    0
    Trophy Points:
    1
    Hello,
    why do we need a lower control limit for range, in a R chart?
    Similarly, why do we need it for S in an S chart?
    Since we want to minimize variability, the LCL should be 0.
    Thanks
     
  2. _Zeno_

    _Zeno_ Member

    Joined:
    Sep 3, 2015
    Messages:
    46
    Likes Received:
    20
    Trophy Points:
    7
    It might indicate a process improvement which would prompt a re-calculation of the limits.
     
  3. Bev D

    Bev D Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    497
    Likes Received:
    540
    Trophy Points:
    92
    Location:
    Maine
    Davide - remember that control limits are not goals and variation will NEVER be zero unless your measurement system is incapable of measuring the characteristic of interest.

    The intent of control limits is to tell you when a true change has occurred (as opposed to random variation being mistaken for a real change). This not only applies to ‘things gone wrong’ but for ‘things gone right’. Too often people will claim victory whenever a single point goes in the direction they want. This includes the within subgroup variation. So if you made a change to reduce the variation then the lower control limit adn it’s associated rules would tell you that you have indeed made a real change to the variation - or it will tell you that you have not. (As Zeno said)

    The other reason for the lower control limit is to warn of an unexpected reduction in variation which may be the result of ‘data manipulation’ (intended and unintended).
     
    Andy Nichols likes this.

Share This Page