When to Calculate, Lock, and Recalculate Control Limits

Quick Links

SPC for Excel Software

Visit our home page

SPC Training

SPC Consulting

Ordering Information

Thanks so much for reading our publication. We hope you find it informative and useful. Happy charting and may the data always support your position.


Dr. Bill McNeese
BPI Consulting, LLC

View Bill McNeese

Connect with Us

Comments (21)

  • AnonymousAugust 31, 2011 Reply

    Best news letter on control limits Thanks

  • AnonymousSeptember 3, 2011 Reply

    The control chart the way details are given is excellent!

  • AnonymousSeptember 3, 2011 Reply

    Hi Sir,
    I thank you very much for your letters.

  • AnonymousSeptember 12, 2011 Reply

    Very nice article and very informative.
    Just a question,how about for Xbar and R charts that are being monitored twice a day,frequency is per shift(morning and evening).how many data points is applicable before I set a control limits and when shall I recalculate.Thanks

    • billSeptember 13, 2011 Reply

      Thanks. I think you still go with the information in the article. Start with 5 or 6 points – then recalculate each time until you have 20 points. Dont' recalculate unless there is a need to.

  • Jonathan UdeNovember 15, 2015 Reply

    In this article, it was stated that if we have a run of points above the average and it is not what we want but we can't find the cause, we can change the process to bring the average down – but "this is not the right thing to do". My question is in this situation please what is the right thing to do since we can't find the cause. Overall this is a very useful article and it has been of immense help to me.Thank you very much

    • billNovember 17, 2015 Reply

      If you can't find the special cause, you don't have a choice but to adjust the average back down – what I meant by not the right thing to do is that you didn't find the special cause.  It might go away and the process average changes again – so you will probably have more opportunities to find it.  Glad you like the article.  Bill

  • MuraliKrishnaJanuary 4, 2016 Reply

    Hi Bill,Thanks a lot. The article is very nice and clear.While recalculating control limits I have encountered following situations.Request you to clrarify.9 points in row are above central line,1) I found one of them was way beyond UCL.I preferred to omit that data point while re calculating the control limits. Hope this is fine2) I found most of them beyond UCL, indicating a bigger process shift.I considered all the 9 points, for recalculating the control limits.pl confirm this is correct.3) Now there was situation,5 are within control limits and 4 are above.Not sure how to go about recalculating control limits, pl guide for below two possibillities a. Process shift happened –reason knownb. Process shift happened – reason unknown 4) Is there any rule of thumb for guidance in situation given aboveThanks a lotMurali Krishna K

    • billJanuary 4, 2016 Reply

      The rule of thumb is that you can elminate an out of control point if you know the reason for the out of control point – what caused it to occur.  In reality if you have lots of data, the out of control point will not have that big of an impact.  But you can delete it if you wish – particularly if it is way beyond the control limits. I would have to see the chart to answer you other questions.  Please feel free to send me the data.

  • AmitFebruary 19, 2016 Reply

    This article is extremely helpful. Thanks a lot for writing this article. It answered almost all doubts I had on SPC. Remaining doubts I am writing here:1. How do we calculate 1sigma, 2sigma control limits on control chart ?2. Should we remove out of control points when special cause is analyzed & eliminated so that control limits can be recalculated?Thanks!

    • billMarch 28, 2019 Reply

      Thanks for the kind words.
      The control limit calcluations depend on the type of control chart you are using.  Please refer to the various publications in the SPC Knowledge Base for the type of control chart you are using.  For example, for the individuals chart, the UCL is given by Xbar + 2.66Rbar.  So, the 3 sigma limits are set at 2.66Rbar.
      Yes, if you eliminate the special cause, you can elminate the point and recalculate the control limits.

  • DuyMarch 28, 2019 Reply

    Very usefull. I'm learning a lot when reading your blog. Thank you very muchIf I want to contribute money for this website, what should I do now ? 

    • billMarch 28, 2019 Reply

      Thank you for your kind comments. We offer these articles in our SPC Knowledge Base free of charge. It is our gift to you. Of course, if you need software to help you with all these techniques, please fee free to purchase our SPC for Excel software at this link: /ordering-information

  • SeraphimAugust 29, 2020 Reply

    Thanks for these great articles – a real treasure trove! I have a question about recalculating the limits after each point until you have 20 points — for example, on an X-chart. Let's say you have collected 8 data points, and you have calculated limits based on that. And let's say the 9th data point is outside those limits. Do you count this as a signal of special-cause variation? I am guessing that yes, you would count that as a signal of special-cause variation.  (Otherwise, what's the point of calculating the limits after each data point up to 20?)  Is this correct?Thanks again!  This is very helpful material, and your Excel product looks great!

    • billAugust 29, 2020 Reply

      Thanks for the kind comments.  Yes, you would treat it as out of control point.  It is a signal based on the data.

  • riazDecember 8, 2020 Reply

    I have a question if the average control chart has been in use for over a week and you noticed that the last 6 data points have all been increasing. what should I do?

    • billDecember 8, 2020 Reply

      NOthing yet, there is not a single until you get 7 or more points in a row trending up.  If you get that, then you should try to find the reason why the process is trending up and correct it.  

  • John seeking knowledgeApril 6, 2021 Reply

    Hello – Can you briefly review the pros and cons of each – fixed verus process/formula or suggest and article to read? I am also very curious on “SPC can be for checking the stability of the imbalance”. Do you have articles on sability of imbalance?

    • billApril 6, 2021 Reply

      You definitely want to fix your control limits at some point and monitor versus those limits into the future.  Not sure what you mean by stability of the imbalance.  

  • [email protected]October 10, 2021 Reply

    hi BillVery useful article.Do you think we need to lock the Variation Chart also (Range or S chart), not just only Average Chart?Thanks

    • billOctober 11, 2021 Reply

      Thank you for the kind comment.  Yes, you should lock both.

Leave a Reply

Your email address will not be published. Required fields are marked *