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.

Interpretation of control charts

Discussion in 'SPC - Statistical Process Control' started by leomac83, Apr 13, 2020.

  1. leomac83

    leomac83 Member

    Joined:
    Aug 22, 2019
    Messages:
    10
    Likes Received:
    0
    Trophy Points:
    1
    Hi everyone, I need some information. Is there a book that allows me to better understand how to interpret control charts X – R, X – S, etc.? All the books I read always explain what the control charts are and the related formulas to build them but they did not explain how to interpret a data series and the approach to an out of range or to an out of trend or in general to the problem. Do you know a book which provides me this kind of information and which can help me?
    Thanks
     
    Last edited by a moderator: Apr 13, 2020
  2. Bev D

    Bev D Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    472
    Likes Received:
    513
    Trophy Points:
    92
    Location:
    Maine
    One of the first books and still the best for the user:
    Western Electric, "Statistical Quality Control Handbook", 1956, 2nd Edition, 10th Printing, May 1984, Western Electric SQC Manual

    The link is to a FREE PDF of the book. (yep - it's that old)


    These pages speak directly to interpretation: Part B, pp. 23-30 and Part F, pp. 149-183
     
  3. leomac83

    leomac83 Member

    Joined:
    Aug 22, 2019
    Messages:
    10
    Likes Received:
    0
    Trophy Points:
    1
    Thanks Bev D. Are there other books, articles or web sites to increase my knowledge regarding the data interpretation or which you or the other user use to keep yourself updated or usefull?
    Thanks again.
     
  4. Bev D

    Bev D Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    472
    Likes Received:
    513
    Trophy Points:
    92
    Location:
    Maine
    Anything written by Donald Wheeler. You can find his FREE articles at qualitydigest.com and his SPC Press website... Book
    I also recommend “innovative Control Charting” by Steven Wise and Douglas Fair. It covers how to handle non-homogenous processes and how to rationally subgroup.
     
    Last edited by a moderator: Apr 14, 2020
    Mike S. likes this.
  5. Mike S.

    Mike S. Member

    Joined:
    Apr 20, 2020
    Messages:
    37
    Likes Received:
    7
    Trophy Points:
    7
    As Bev said, anything written by Donald Wheeler -- IMO he is THE guru of SPC, bar none. His book, Understanding Variation is a good start.

    Also, Davis Balestracci has published many good articles and a book called Data Sanity to help figure out control charts and how to proceed to solve out-of-control conditions.
     
    Last edited by a moderator: Apr 20, 2020
  6. Miner

    Miner Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    392
    Likes Received:
    295
    Trophy Points:
    62
    Location:
    Greater Milwaukee USA
    All good suggestions. I will add two more things that often get missed. Donald Wheeler talks about both of these in his short articles, but there are a lot of articles, so they might get missed.
    • There is a right way and many wrong ways to determine the standard deviation used to calculate control limits. Some software packages and homegrown Excel templates use the wrong way. Some books and consultants also teach the wrong way.
    • Almost all textbooks teach that the control limits are probability based on the Normal distribution. They are not. Shewhart developed them empirically to balance the risk of a false alarm with the risk of not reacting. The myth of probability started early on and both Deming and Wheeler fought against that understanding.
    I will add one more from my personal experience. There are many versions of extended rules out there starting with the Western Electric rules that Bev cited. Do not use all of these rules or you will experience a lot of false alarms. Use only those rules that make sense for your particular process. Also, the typically last two rules are simply diagnostic rules intended to check whether you truly have rational subgroups. Once you have done this, you no longer need those two rules.
     
    Bev D, Mike S. and Andy Nichols like this.

Share This Page