1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.
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.

Quantile Cpk

Discussion in 'Capability - Process, Machine, Gage …' started by Pongsakorn, Oct 8, 2018.

  1. Pongsakorn

    Pongsakorn Active Member

    Joined:
    Aug 10, 2015
    Messages:
    72
    Likes Received:
    16
    Trophy Points:
    7
    JESD557C para 5.3 defines the "Quantile Cpk" formula to be use for non-normal and and non-log-normal data but I still do not understand clearly, please advise as follows:
    - What is the different between "Cpk" and "Quantile Cpk"?
    - When to use "Quantile Cpk"

    upload_2018-10-8_17-55-39.png
     

    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:
    601
    Likes Received:
    658
    Trophy Points:
    92
    Location:
    Maine
    The quantile approach is an attempt to align a non-normal distribution standard deviation to a defect rate estimation.
    it's theoretically OK. there are many approaches to deal with non-normal distributions - this is one of them. Quantiles is a non-parametric approach that is theoretically distribution-less. you simple divide the data into 4 segments with 25 of the data in each 'chunk'. the chunks will have varying lengths. (google 'box plot' it uses the same concept)
    If you have a Normal distribution you just calculate the Cpk in a straight forward manner.


    HOWEVER: this is all a bunch of hoo-ha. lots of math to no valuable end. it only serves to check some box for a Customer or internal QA person who doesn't understand variation. Cpk is a zombie theory - it just won't die. no matter how much logic and reason and science we throw at it. if you don't have to do it, don't. total waste of time.
     
    Pongsakorn likes this.
  3. Pongsakorn

    Pongsakorn Active Member

    Joined:
    Aug 10, 2015
    Messages:
    72
    Likes Received:
    16
    Trophy Points:
    7
    Bev, thanks for your advice and comment.