I know that capability and stability of a process are completely different concepts but I wonder if cp and cpk values can be accepted as a measure for deciding to start a SPC study or not? If yes, then what cp-cpk value is enough to be in safe region. (especially for automotive industry)
For example, if my cp and cpk values are 3, is it really necessary to perform a SPC study of that process?
I don’t exactly know what you mean by “SPC Study”.
SPC is a control tool. If your CpK is 3, then you still need to control the process. How much, how often, SPC or another tool? You have not given enough information for a good evaluation.
Ok you’re right. ”SPC” is enough. There is no need to say “SPC Study”.
I aggree that SPC is a control tool which gives us information about stability of the process. As all we know, LCL and UCL are computed by using measured values only. That means LCL and UCL is not related with the tolerance. Result: SPC examines the process only within itself.
I quoted a phrase from isixsigma.com below. Ken describes the situation perfectly
“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” Ken K.
If you have a hangar, does size of your car really matter by means of parking difficulty? (Hangar: Specification limits, Size: variation within the process)
The short answer to your question is no. The reason is because your Cp, Cpk values are meaningless unless you have shown your process is in control (stable).
Jsev607 is correct: your process will be in only one of four states – stable & capable, unstable and capable, stable and incapable, unstable and incapable. A capability assessment of unstable data means you are looking at data that is unpredictable….which means the data you are using in your capability assessment will change….hence, your capabilty measures will change. That is why LSS teaches that you prove stability first, capability second.
Really helpful explanations, thank you all.
Second Case: Let’s assume that my process was proven to be stable in the long term. And also pp and ppk values has reached 10. is it still necessary to continue SPC?
I ask these questions to utilize my spc team’s workforce efficiently
1. Is it possible Cpk and Ppk value give a negative value?
2. What does this means? Is it wrong calculations or spec issue?
This is quite possible to have a Cpk or Ppk value as negative. This means that process is in bad shape as average of the data against capability study falls beyond specification limit.When cpk is zero this means that process average and one of the specification limits are same i.e these are getting superimposed.
The more capable your process (and I would call a process with a Cpk of 10 extremely capable) the less SPC is required. What is sufficient a sufficient level of control will be a business decision: what is the cost of SPC efforts vs the result of a shift in your process mean? Go from there.
Deming and Shewhart are quite clear on the subject that SPC is not just a monitoring / control tool – it is a continual process improvement tool. If we take the limited view that is is just for process control and monitoring, we miss out its their true power: variation reduction and improvement.
Just my $0.02 worth,
The process needs to be shown to be in control before the CPk has any meaning. I have seen data that shows an acceptable CPk value that is making parts out of spec (the process was out of control btw). The problem comes from small within variation and large between variation giving you a unrealistically low std deviation. Bottom line, you should chart your process, demonstrate that it is stable and then report its capability.
If Cp & Cpk is 3 then there is no need to study.
Excellent point “Six Sigma Shooter!”
Having a stable process is one of the assumptions that goes into your process capability assessment.
Now SPC is not just a stability/control tool, it can also be used for root cause analysis and evaluating solutions independently of assessing the capability of the process.
It’s all part of using the tools to solve problems.
SPC should not be a yes / no question. Do you ever want to give up control? Your real issue should be: when Cpk shows excellent capability, what adjustments to sampling rate can I tolerate to reduce control costs without jeopardizing quality of delivery. With a Cpk of 3 you might make some small reductions in sampling frequency, while with Cpk of 10 you might make some larger reductions. HOWEVER, make sure you establish some signal points to return to higher sampling rates when Cpk decreases.
Thank you for your time and support. Really helpful and focused answers.
We have so many machines with less cp & cpk , so what are rhe steps are to be taken to improve these values? this problem regards to grinding machines
Just curious as to the percentage of out-of-spec material you are producing across all grinding machines? Do any of the grinding machines have acceptable cpk values?
Some basic first steps:
Look at your data and see what the major sources of variation are ( within piece, piece to piece, etc) If you didn’t collect your data to show this….design a multi-vari study and run it.
It is difficult to help you with such limited information.
I could just say “Apply six sigma methodologies!” I assume the M phase is complete and you know that your measurement system is capable!
A process or machine capability study should be performed upon initial purchase of a piece of equipment and after any significant maitenance operation perfomre oh that equipment prior to putting it back into production service.
As a general rule I try to run a study each quarter and monitor the process with SPC chartage in the periods between running the studies.
If you follow this refgimen you will be continuously running someform of “SPC” .
Are you serious? If the Cpk, or more appropriately Ppk, was 3ish, then why run SPC on that process parameter? Obviously the process has enough controls–unless it’s a case of overly lax standards by the customer.
I’ve successfully eliminated wallpaper SPC on processes for those with Ppk above 2 if reasons for success were known–the control system being in place.
Just my 2c, it still depends on the RPN. If you’re controlling a safety critical component, ie severity 9 or 10, you may want to continue SPC.
Hi, I don’t know if your post was directed specifically at me, but I think it deserves an answer anyway.
I am a strong believer in “Economic Design of Control ~Charts” I have written a lot on that subject.
SPC is not for a process that does not vary. (Is there any such thing?)
SPC is not wasted on “good” processes. It depends on the frequency and magnitude of process shifts. If you have a good way to identify and react to them consistently, across operators, shifts, machines without using a chart, that is fine with me.
The forum ‘General’ is closed to new topics and replies.