iSixSigma

What is Cpk in Laymans Terms

Six Sigma – iSixSigma Forums Old Forums General What is Cpk in Laymans Terms

Viewing 60 posts - 1 through 60 (of 60 total)
  • Author
    Posts
  • #27262

    LPM
    Participant

    I am having difficulty describing Cpk when asked, the problem being I know what it is, what it does, how its used, I just cannot describe it to my financial manager in Laymans terms!
    I always come up with some mad statistical bull*&*& that makes him lose the plot straight from the get go!
    Is there anyway I can tell him within a brief sentence, possible on a slide show, what it is so he can understand!

    0
    #66555

    sundeep
    Member

    Cpk means that how much mean is shifted from ur target.

    0
    #66556

    Neil Polhemus
    Participant

    Cpk is an index (a simple number) which measures how close a process is running to its specification limits, relative to the natural variability of the process. The larger the index, the less likely it is that any item will be outside the specs. If you like, you can then relate specific values of Cpk to the proportion of items which would be beyond the specs. Don’t start talking about standard deviations. If he has an MBA, he will be used to indices, which are designed to hide the complexity behind a calculation and come up with a single number which is easy to understand.

    0
    #66561

    DRAGOS BANDUR
    Participant

    Cpk is not a simple number, it is a statistic.

    0
    #66562

    Yadav
    Participant

    Niel / LPM,

    The explanation given by Niel is pretty much good, that’;s the way one can explain it to a non-statistician.

    The only correction is : Cpk measures how close you r to your target and how consistant u r to around your avaerage performance. A person may be performing with minimum variation, but he can be away from his target towards one of the specification limit, which indicates lower Cpk, whereas Cp will be high. On the other hand, a person may be on average exactly at the target, but the variation in performance is high (but still lower than the tolerance band (i.e. specification interval). In such case also Cpk will be lower, but Cp will be high. Cpk will be higher only when you r meeting the target consistently with minimum variation.

    For practical example:
    Consider Tiger woods playing a long shot in golf game: (though I’m not familiar with terminology of the game, but I’ll try to explain with this game itself).

    whenever he takes up teh long shot, the ball lands on the ground, after it’s long flight in the air crossing a water barrier), it always lands near to the hole (is that called birdie?) within a radius of say 30cm, whereas the green (smooth surface) around the hole is about a radius of 5m. In such scenario Woods has very high Cpk (may be 6 or more). On the otehr hand, another player’s ball lands up away from hole within a circualr area (excluding the hole) of 30cm, and this circle is say 1m away from hole. In such case the other guy has same Cp as that of Woods, but lesser Cpk (may be only 1 or so).

    On the other hand in another day, if Woods’ ball lands up always within a circle of radius 1m, then his Cpk and as well Cp will be lesser than his ealrier day. Since his region of variability has increased.

    0
    #66566

    Savage
    Participant

    I would only add that, assuming a process is in control, you can make predictions about how much data is theoretically outside the specification limit(s). The points I am making are that you can have a “good” Cpk yet still have data outside the spec. and that the process needs to be in control before evaluating Cpk.

    0
    #66570

    Neil Polhemus
    Participant

    Good point. “Simple” only in the sense that it is a dimensionless quantity which does not depend on the units of measurement.

    0
    #66576

    Joe Perito
    Participant

    CpK is a performance measurment that is affected by the amount of variation in your process and how close you are to target. The greated the amount of variation and/or the further you are off target the worse the performance. Take target shooting: Bullets outside of the target is undesirable variation. Bullets all over the target is better but still has a lot of variation and is undesireable. A nice tight pattern of bullets in the third ring of the target shows little variation…that’s an improvement, but it’s still not on target. A tight pattern “on” the bullseye is the best performance of all. The smaller the variation and the closer you are to target, the higher the CpK factor is. You must have a CpK of 1.33 or higher to satisfy most customers.

    0
    #66580

    Leung
    Participant

    Consider a car and a garage. The garage defines the specification limits; the car defines the output of the process.

    If the car is only a little bit smaller than the garage, you had better park it right in the middle of the garage (center of the specification) if you want to get all of the car in the garage.

    If the car is wider than the garage, it does not matter if you have it centered; it will not fit.

    If the car is a lot smaller than the garage (six sigma process), it doesn’t matter if you park it exactly in the middle; it will fit and you have plenty of room on either side.

    If you have a process that is in control and with little variation, you should be able to park the car easily within the garage and thus meet customer requirements. Cpk tells you the relationship between the size of the car, the size of the garage and how far away from the middle of the garage you parked the car.

    Hope this helps.

    0
    #66820

    Ken K.
    Participant

    I like the garage analogy.

    The value itself can be thought of as the amount the process (car) can widen before hitting the nearest spec limit (garage door edge).

    Cpk=1/2 means you’ve crunched nearest the door edge (ouch!)
    Cpk=1 means you’re just touching the nearest edge
    Cpk=2 means your width can grow 2 times before touching
    Cpk=3 means your width can grow 3 times before touching

    etc…

    0
    #66824

    “Ken”
    Participant

    Ken K.,

    Great analogy! I’ve used it on serveral occasions. However, this analogy is best used for Cp instead of Cpk. The width of the car refers to the variability of the system. Cpk is a measure of both variability and centering of the process as Neil described earlier.

    Ken

    0
    #66827

    Tommy
    Member

    If you hunt our shoot targets with bow, darts, or gun try this analogy. If your shots are falling in the same spotforming a good group this is a high cP, and when the sighting is adjusted so this tight group of shots is landing on the bullseye, you now have a high cpK

    0
    #66828

    Ilene D
    Participant

    For those of us quality professionals on the West Coast, CPK means only one thing – California Pizza Kitchen. Yummmmm!

    0
    #66830

    “Ken”
    Participant

    Irene,

    It could also mean “Cabbage Patch Kids”.

    0
    #66845

    kony
    Participant

    … so it means that if Cpk (or as Ken posted Cp) is 1,33 even your wife can park a car without hitting the garage?!

    0
    #66937

    “Ken”
    Participant

    Kony,

    Cp tells us what you wife has the potential of doing. Cpk tells us what she has actually been doing. It’s possible for your wife to have a potential 4-sigma, Cp=1.33, parking capability. However, you may still find some dents in the sides of her car, because her actual capability, Cpk, is less than one. This would mean you need to get into the car with her, and guide her to keep the car centered between the door edges of the garage. Then, after you are successful at helping to improve her actual capability, go out and buy her a new pair of glasses! The glasses would be the corrective action to process improvement…

    Ken

    0
    #72444

    Luis C
    Participant

    Your example is very clear for Cpk but can you tell me where is the Cp in you garage example, also Ppk and Pp?
     
    regards

    0
    #72570

    Luis C
    Participant

    I perfectly understand the Cp and Cpk with the car example. Does someone have the example for Pp and Ppk

    0
    #79903

    Concerned Student
    Participant

    Dear all,
    Please be serious when you post up or answer a question.  As a student, it’s very frustrated to have someone give us wrong information or taking something serious and turn into a joke.
    To all with appropriate reply.  Thank you

    0
    #79905

    abeerwart
    Participant

    Another example besides the garage/parking to explain Cpk is in American football, consider a goal kicker.  Cpk explains how well does the kicker place the ball between the goal posts (upper and lower specs).  A negative Cpk means the kicker can’t get it between the goals (but still could play for the Bears) and a positive Cpk means that the kicker is getting the ball between the goal at least some of the time.  The higher the Cpk number, the less variation (spread) in where the ball goes.  Cp and Cpk are for describing the kicker’s game, where Pp and Ppk can describe the kicker’s career.

    0
    #79911

    Sambuddha
    Member

    Andy(?),
    That’s very well put! Specially the game Vs. career part.
    Best,
    Sambuddha

    0
    #82927

    Iker Etxarri
    Participant

    What do I do if the model I study is not based in a normal distribution. Example: to get a hole of diam.11 +/- 0,2 mm, I use a tool of diam. 11, 95 mm. The result is:
    11.15;11.915;11,14;11.13;11.15;11.14
    everage:11.143
    if I calculate Cpk it will be less than 1.33, but the process it capable, because the tool can only lose sharpenes and then the hole will be smaller. Distance (11.143-10.8)=0.343. So I have to calculate Cpk with the samllest baound not with the closest. 
     
        

    0
    #82931

    Hemanth
    Participant

    Hi
    Cpk is an index used to measure capability of process. Capability is defined as the ability of the process to perform within the specification limit. Hence higher the capability better it is. Now for a finance guy this might seem irrelevant as he would say as long as my process is behaving within the specs I am not incurring any loss. Thats when you need to tell him about taguchi loss function. More you move away from the target more is the loss and this follows the parabolic curve. Thats why higher capability means most of your processes are near the target and hence result in minimum loss. If you need any info, please do let me know.
    Hope this helped
    Hemanth

    0
    #82940

    vidyut
    Member

    right.
    Cpk is about hitting in the bull’e eye.

    0
    #82941

    Allegood
    Member

    Cpk is about hitting the bulls eye and how far are you from not hitting it.

    0
    #82942

    Heebeegeebee BB
    Participant

    Precision vs. accuracy
     

    0
    #82951

    Gabriel
    Participant

    In that analogy, the spread of the shots is an indication of the precision. The smaller the area affected by the shots, the better the precision. The “average position” of the shots is an indication of the accuracy. The closer the center that area affected by the shots is from the center of bull’s eye, the better the accuracy.
    Cp is precision.
    But Cpk is a mix between precision and accuracy. For a good Cpk you need BOTH a good precision and a good accuracy.

    0
    #82955

    Heebeegeebee BB
    Participant

    Some like to view CP as “hitting the target” and CPK as tight “shot groups” on target.
    I’ve also heard just the opposite in these forums.
    In my opinion, a tight “shotgroup” is demonstrative of precision and repeatability/reproduceability regardless of where, on the target the grouping lies.   I can see your point for both Accuracy and Precision in CPK though.

    0
    #82957

    Robert Butler
    Participant

      Since Cpk requires data normality it is inappropriate for data that does not have a normal distribution.  To get a measure of capability the recommended procedure is to compute an equivalent capability which is based on the .135 and 99.865 points of the normal probability distribution.
      The method is as follows:
       Plot your data on normal probability paper.  Curve fit your data points and identify the .135% and 99.865% points on the graph.  The equivalent capability is expressed as:
           (USL – LSL)/(x(99.865)-x(.135) )  = Tolerance/Equivalent 6 sigma spread.
    If you want further details check Chapter 8 of Measuring Process Capability by Bothe which is titled Capability for Non-Normal Distributions

    0
    #92358

    Pramod Bapat
    Participant

    Ajit,
    You have selected a very good theme for your example; but I am not clear about the conclusion. Can you re-explain with sketches?
    Regards..Pramod Bapat, Pune (INDIA)…11/11/2003

    0
    #96408

    Pipkin
    Participant

    Tommy,
    I liked your analogy on Cp and Cpk with the arrows and target.  Can a similar analogy be made for Pp and Ppk?
     
    Jack

    0
    #105497

    Ushan
    Member

    Can somone explain the connection between Cp, Cpk and Sigma values to actual ppm figures.  
    Ex: For a tollarance value to be capable whithin 4sigma, what will the Cp and Cpk values need to be.  and what the ppm’s be if the sigma is changed from 4 to 3?
    Thank you,
    Ushan 

    0
    #109716

    Pradeep Palat
    Participant

    Is it true that in order to get a Ppk value of 1.33 the Cpk value need to be 1.66?

    0
    #109721

    Ron
    Member

    The simplest answer is that it shows you how centered your process is to the specification.
     

    0
    #113720

    delio
    Participant

    this means if you have lots of things to do that requires attention you must try to not answer the phone too much or rather do just one thing and make good on it so as to have a higher CPK factor. joke.

    0
    #114975

    ENRIQUE PEREZ
    Participant

    Could some of you tell me what does mean the “K” in the Cpk acronymun,
    Thanks,

    0
    #114978

    Mark Almeter
    Participant

    Basiclly, Cp describes how capable a process is relative to customer specs with regard to variation.  If my process variation is small compared to the requirements given by the customer, then my process is said to be capable.  However, customers not only care about measures of spread (variation) in the process, but measures of central tendency of the process as well (such as mean) and where the mean falls with respect to the target value given by the customer.  As such Cp only tells part of the story.
    CpK takes into account the process mean relative to the customer target value.  You can have a very capable process (low variation) but if the mean is far from the target value for the customer, then it fails to meet customer expectations. It is analogous to an archer that shoots a tight group of arrows (low process spread and good Cp) but way outside the bulls eye (bad CpK). 
    Hope this helps….
    Mark

    0
    #115797

    Dmitry Kiskatchi
    Participant

    When we are saying that Cpk value should be 1.33 and more to satisfy most customers, do we include 1.33 into the range of acceptance?  In other words, if I got Cpk for my process equals exactly 1.33, is it acceptable? Or it needs to be at least 1.34? And another thing: above what level of Cpk we need to be kind of “suspicious” about reliability of collected data (in this case we might say: “Oh, it is too good to be true”). I have heard opinions about Cpk>6.00. May be this level is already too high?

    0
    #130450

    dhonde
    Participant

    Good one ..

    0
    #130452

    Habib
    Participant

    what is a good one?

    0
    #130739

    Julio Romero
    Participant

    Which are the difwrence between Cpk and Ppk, and how can i use those parameters on DMAIC

    0
    #132111

    Edwards
    Participant

    I like the CPK analogy with the Arrows. Tight group but away from the bulls eye, draws a nice visual of what CPK might be. So, if the CPK number is higher, does that mean the tight group of arrows is further away from the bulls eye and is it linear, meaning the greater the distance the higher the number?
    As for CP, I need to look at the other examples. It almost sound the same as CPK.
     
    Regards,
        David

    0
    #132263

    good example
    Participant

    good example

    0
    #132264

    Prabhat
    Participant

    good example

    0
    #141197

    Sujeet Mishra
    Member

    1) Why Cp & Cpk > 1.33 ?
    2) Where from this Figure 1.33 came into picture ( Calculation part )?
    3) What happens when Cp > 1.33 but Cpk < 1.33 ?
     
     

    0
    #141198

    Mike Carnell
    Participant

    Sujeet Mishra,
    There is no hard ans fast rule that a product/process must operate >1.33 or 4 standard deviations from the closest specification limits. The are product and market specific opportunities that will cause a customer and or a producer to go to market for a product with less than a 1.33 Cpk. The opposite is true in a mature and competitive market. The customers may demand >1.33.
    We worked with a pacemaker company in Japan. In that situation it is a good idea to be considerably above the 1.33 Cpk mark. We have worked with a racketball manufacturer (at the time there was only one). Obviously they drove on efficency for the sake of the business but interms of customer demand – there was only one option.
    Cpk > 1.33 was probably popularized by the automotive industry. That is a guess. They require it and have a fairly large influence throughout their supply base.
    What happens if you are greater than or less than 1.33. You get the famous “it depends.” If you have selected a characteristic that has some value in the marketplace and the customers comprhend the value it may mean more revenue. If it is a characteristic that customers don’t value you may have wasted some money. If it is purely an internal issue it may save you some cost to be >1.33.
    This is not a one size fits all metric. It needs context to have meaning.
    Good luck

    0
    #141656

    Desmond
    Participant

    “You must have a Cpk of 1.33 [4 sigma] or higher to satisfy most customers.” Joe Perito
    How is this figure of Cpk=1.33 linked to the 4 sigma as in above statement?
    Can someone help? Thx.
    Rgrds,
    Desmond.

    0
    #152056

    lknn
    Participant

    ????????????? what are you talking about ??????????

    0
    #160908

    Segaran Soomu
    Member

    Why we must get Cpk 1.33. If i get Cpk 2 is this good process or weak process.
    If i get Cpk less than 1.33 an my process is control is this call poor process.

    0
    #160909

    fake accrington alert
    Participant

    If  Cp is  equal  to  2 then you  have a  SS score of 6
    If Cp is

    0
    #160910

    fake accrington alert
    Participant

    If  your Cp is  above 1 then  you  have  SS score equal  to 3
    If  your Cp is  equal  to  1.33 then  you  have SS score  equal  to 4
    You  can  talk  about  about  a  poor  process  if  your  Cp score is  below  1
    You  are mixing  up  between  Cp and  Cpk,Cpk  is  an  indicator  of  your process  location (centered  or  skewed). 

    0
    #177421

    p k joshi
    Participant

    Hallo,
    You know when we set at target cp and cpk are same. cp = T/6s for 1.33 it will be 8/6 means Tis equivalent to 8s  hence half of tolerance or target is equale to 4s  Cpk= USL-Target/3s means 4s/3s=1.33
    I hope it clarify how cpk 1,33 is equivalent to 4s
    Regards
    P K Joshi
     

    0
    #177422

    Nolan
    Participant

    Thanks PK. Now it’s totally clear.
    Just like Obama’s plan…”We’re going to Change.” Yeah, what are we going to change? “Well, everything. We’re going to change everything because this country’s ready for change.”

    0
    #177426

    Mikel
    Member

    Les honey,You lost and maybe you should look in the mirror and ask yourself
    why you would want more of the same BS we’ve had for the last 8
    years.Nothing has gone well and the Republicans have presided over the
    biggest expansion of the federal gov’t in history. We’re in a
    quagmire in Iraq and there is no good way out. We are not winning
    and we cannot win.Wake up and do something positive for your country.

    0
    #177435

    Craig
    Participant

    We should have listened to Rocky Balboa a long time ago. (We can change!!)
    I recommend a LSS initiative on the government. (or even a Kaizen Blitz). Tell you one thing, I’d sure hate to see one of their spaghetti diagrams. We’d have a world shortage on pasta.

    0
    #177444

    Saherngu
    Participant

    Consistency around the mean, not within upper and lower limits (cp)

    0
    #182546

    DUF
    Participant

    The streets are full of 6Sigma workers that were replaced by workers from Asian countries not beeing able to deliver only 1,3Sigma. Cheers

    0
    #182547

    Mikel
    Member

    Maybe where you live, but not here.

    0
    #182560

    BritW
    Participant

    What was the record for responding to a previous post?  This was 2yrs 10 mo……

    0
    #182561

    Mikel
    Member

    There have been some > 5 years, but not many this insightful and
    enlightening.

    0
Viewing 60 posts - 1 through 60 (of 60 total)

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