iSixSigma

Stable process: what is it?

Six Sigma – iSixSigma Forums Old Forums General Stable process: what is it?

Viewing 19 posts - 1 through 19 (of 19 total)
  • Author
    Posts
  • #27394

    SAMIR MISTRY
    Member

    a stable process in simple terms is a process of which all the causes of variations are known and are acted upon and the process is then governed by common causes of variations, where the output of the process is fairly predictable. management decision requires to further increase the capability of the process.

    0
    #66999

    Ken K.
    Participant

    I wouldn’t go so far as to say that “all the causes of variations are known”. That is a pretty extreme statement.

    I would tend to say a stable process is one that is comprized of mostly common cause variation, as opposed to special cause variation. As you hinted, that common cause variation will be comprized of a whole bunch of sources of variation, some will be knowable and some won’t.

    The whole idea of process improvement is to understand many of those sources of variation and try to remove/control them.

    0
    #82436

    Bill Letcher
    Participant

    Yes, I liked this, plain, simple and to the point – not a lot that can be argued about here!
    Bill.

    0
    #82437

    Mike Carnell
    Participant

    Ken,
    I am going along with you on this. There are a lot of control charts out there where as long as something is classified as common cause nobody pursues it because they believe it is unknowable or uncontrollable.
    Relating stability to a control chart doesn’t make any sense to me. A process that has a trend run shift or cycle can be stable. A chart that has seasonal shifts can be stable.
    There are and have been a lot of people who look at a control chart – classify it as unstable – and don’t do anything.
    If you go back to the Allied & GE deployments we never required stability to initiate a project. There were a lot of successful projects. I have never concerned myself over this. Sometimes things are a little strange but that is what the job is – sorting it out. Stability is over rated and mor often than not an excuse.
    Good luck.

    0
    #82446

    Schultz
    Participant

    Mike
    People can always find a reason or excuse for doing nothing. But just because the control chart indicates the process is not stable, does not mean nothing can be. First, determine the assignable causes, i.e. special causes, for the lack of control. Once these have been identified and corrected, then we can use the Six Sigma Statistical Tools to reduce the variation. The search for special causes is different than reducing variation. The tools are different. Many times I have seen people trying to use statistical tools in the wrong application. I have also seen people try to use statistical tools on process that are unstable and they are never able to repeat their results.

    0
    #82451

    Mike Carnell
    Participant

    Charlie,
    Thanks for the tip on how to use to use a control chart to reduce variation. I’ll have to keep that in mind.
    The most commonly misapplied tool I run into is the control chart ie. last week I was in a business with attribute data plotted on individuals charts.
    Ther is nothing more frustrating for to see than some person wrapped around the axle because somebody told him that the process had to appear under control on a chart. If you are running a DOE or hypothesis test you can fix (freeze) an input to get piece of information you need. I see more of the DOE’s and hypothesis tests go south because the measurement system on the input variables and response fails a MSA (that typically doesn’t get run).
    If you are running confirming runs you will probably pick up the lack of repeatability. It is all calculated risk – that is why you use stats. Speed is undeniably a key factor in a project identified as financially viable for a SS project (listen to the whinning about the time that it takes to do a SS project). I have an analysis of a company that loses >$10 per second on COPQ (their numbers) – how much of a risk do think they are willing to trade for speed?
    Good luck.

    0
    #89547

    Chandran
    Participant

    Does it make sense to create metrics baselines out of an unstable process?

    0
    #89549

    Praveen Gupta
    Participant

    Well, it depends  upon what do you plan to do with metrics. Unstable process could mean assignable variables present in the process or larger variance. With the knowledge of variance, one can have metrics for specified purpose.

    0
    #89559

    Manish Pharasi
    Participant

    Stability of a process is also governed by the consistency of its output over a period of time. Thus, the dimension of Time is a crucial factor. This also underlines what others have said that all special causes of variation have been eliminated or controlled.

    0
    #89595

    Chandran
    Participant

    hi Praveen,
    Thanks for your response. The purpose that I had in my mind is using those baselines for future estimation. I am of the opinion that one needs to identify and eliminate special causes of variation before coming up with such baselines. what do you think?

    0
    #93004

    senthilkumar
    Member

    Thank

    0
    #93005

    Talaid
    Participant

    Do you mean to say that stability need not be checked?

    0
    #93007

    Talaid
    Participant

    What are the tools that you are talking about?

    0
    #130757

    Robert Smith
    Member

    What is variation between in relation to control charts?
    What is variation within in relation to control charts?

    0
    #136009

    odra
    Participant

    Have you read any of W. Edwards Deming’s books. Out of the Crisis or The New Economics give excellent examples of a stable system. Not to mention the differences between specail causes and common causes.

    0
    #136010

    odra
    Participant

    Once you have a stable system with control limits you can determine if the problems you are having are the system.  If it falls outside the control limits then it is more than likely a specail problem or one that can reoccur.  This is a problem that can be tracked and corrected once the source is identified.

    0
    #136689

    joshi pk
    Participant

    What is process stability ? how to assess it? why +-3 sigma conisidered for capability & not any other thing

    0
    #170064

    VM
    Member

    Here is what I feel the basic requirement is. A target line. If the target line is outside the control limits, 3/4th of the war is won. Post that all I am bothered is what can help me maintain it that way or what can supposedly ruin it. This fanda helped me stabilize my process as even my Team leads understand this.

    0
    #181377

    Sankar
    Participant

    I have no doubt about the process capability or process in statistical process control means you used all your design, pilot plans in the process to remove all your assignable causes and took out production with Cp or Cpk of atleast 1.33 and above or e1.66 and above to meet the customer’s requirements. That means the process is established before taking out production and proved its consistency for a prolonged time of production. This has to be the beginning and historically delivered to the customers. Any new assignable or common cause which distrubed this long range capability alone should be taken as a Six sigma project and claimed. Anything else is a misuse of QA principles as anyone can start a sixsigma project for a routine QA problems and apply all the diagonstic techniques like a hospital to tell the patient tha he had a common cold due to rains. I think Walter Shewarts method and book is very clear on setting the limits etc., 

    0
Viewing 19 posts - 1 through 19 (of 19 total)

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