iSixSigma

How to create a control limit

Six Sigma – iSixSigma Forums Old Forums General How to create a control limit

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
    Posts
  • #42768

    boettler
    Member

    Dear Masters,
    I need your expertise since I am quite new in this area. If my process determined by spec:
    Average = 259, USL = 266 and LSL = 252, how to create control limits for both UCL and LCL  as well as the R(UCL) and R(LCL)
    I would be very grateful if anyone of you could give me the detail calculation on order to have those limits.
    Please help me,
    Thank you very much,
    Rob,

    0
    #135223

    Ken Feldman
    Participant

    Rob,
    Your specs have nothing to do with computing the control limits.  You haven’t given enough information to provide much help.  The formulas for control limits are easily found with a little research on this site and a simple Google.  Depending on whether you are using subgroups or individual data, your calculation of the Xbar and Rbar will vary.  Once you have figured out which one and calculate the Xbar and Rbar the formulas and constant values are easily found and computed.

    0
    #135240

    ezweld
    Participant

    Rob,
     
    For you to have a 1.33 CPK your standard deviation would have to be 1.75.
    That would give you a UCL=264.26 and LCL=253.75.
     
    Most people figure control limits by multiplying 3*Std dev +Average for UCL and Average – 3*Std dev for the LCL. Without knowing your sample subgroup it’s hard to come up with a control limit for your ranges.
     
    In less you are lucky you will find that your actual standard deviation is greater than 1.75.
    Most likely you based your specs off of your past measurements observation and not a statistical calculation. When you take measurement you will roughly see 75% to 80% of the process, this is misleading because 68% of your process that you will see about 90% of the time is plus or minus one standard deviation. So your next step is going to be a process capability study. Really, it should be a Gage R&R but I don’t want to preach. After you complete the capability study you will have to determine if your process is stable, if it is then determine if your process is capable. (75% of your product spec) If your process is stable and capable, you are really blessed.

    0
Viewing 3 posts - 1 through 3 (of 3 total)

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