iSixSigma

GRR, on process variation or on tolerance range?

Six Sigma – iSixSigma Forums Old Forums General GRR, on process variation or on tolerance range?

Viewing 17 posts - 1 through 17 (of 17 total)
  • Author
    Posts
  • #53296

    Saherngu
    Participant

    Have a good day,
    About %GRR: When I have to use %GRR based on process variation or based on tolerance range? Thanks.
    Respecto al %GRR: ¿Cuándo debo usar el %GRR basado en la variación del proceso y cuando, basado en la tolerancia total? Gracias.

    0
    #189503

    Mikel
    Member

    Tolerance – knowing your measurement’s capability for product
    acceptance.Variation – knowing your measurement’s capability to assist in
    reducing variation.

    0
    #189507

    Paco
    Participant

    Yes, the G&RR it’s a useful tool to find whose  responsible for the variation on your process or product (tools, man, etc), the tolerance is define by requirements that you have to meet.
     
    Asi es James, el estudio de RR es para conocer a quien se le atribuye la variabilidad que ocurre en tu producto y la tolerancia son los requerimientos limites de tu rpoducto.
     
     

    0
    #189508

    Saherngu
    Participant

    Thanks for your response.
    The fact is this: I can analyse %GRR from two ways: 1) Based on the tolerance range, and 2) Based on process variation. My question is: Wich way (1 or 2) have I to use and why? Thankns again. 
    Gracias por su respuesta.
    El caso es: Yo puedo analizar el %GRR de dos formas: 1) Basado en la tolenacia total, y 2) Basado en la variación del proceso. Mi pregunta es: ¿Cuál de las dos formas debo usar y porque? Gracias de nuevo.

    0
    #189510

    DY
    Participant

    Why would you want to analysis based on process variation? If your process varies then there is no way to have reproducibility or repeatability.

    0
    #189511

    DY
    Participant

    Why would you want to analysis based on process variation? If your process varies then there is no way to have reproducibility or repeatability.

    0
    #189512

    Mikel
    Member

    The first response answered your question. see if you can deduce the
    reasoning.

    0
    #189516

    Mikel
    Member

    DY,What remarkable insight. Do you always respond to things you don’t have a clue about?

    0
    #189520

    Paco
    Participant

    I don’t see why you analyze based on tolerance, since the limits your taking are part of your speficications or ranges to have your outcome. You should use GRR based on process variation due to is what are you looking for to eliminate. If you consider the tolerance in the study that mean is part of the variation of your process.
    No veo el motivo de analizar basado enla tolerancia, debido a que la tolerancia es parte de los limites considerados en las especificaciones o rangos de lo que buscas. Deberias considerar GRR basado en la variacion del procesos debido a que es lo que bsucas eliminar. Si consideras la tolerancia en el estudio siginifca que esa tolerancia es parte de la variaciond el proceso.
      

    0
    #189521

    Venerable Bede
    Member

    James:
    Stan had it right from the beginning; his first (somewhat terse) comment says all that needs to be said, but I will take this opportunity to expand (a lot) on what he is saying.  Look at it this way:
    %Contribution or %Study are supposed to examine (assuming you have selected your study parts properly) a reasonable expected range of part-to-part variation resulting from your process and the ratio of variation in your measurement system versus variation among your selected parts.  Knowing this will help you understand how useful your gauge is for collecting data to study process variation.  So if your intent is to characterize process variation for the purpose of process improvement use %Contribution or %Study.
    %Tolerance is important if you want to run a capability analysis.  This metric tells you the ratio of variation contributed by your measurement system versus your tolerance range.  In general if your customer is requesting a GRR or capability analysis, this is the metric you will want to focus on.
    Of course it is possible to have good %Contribution/%Study and bad %Tolerance.  This means your Cp is probably bad and while your measurement system is marginally good enough to understand process variation, the tolerance range is very small compared to your process variation.  It also tells you that your measurement system variation is too large to run a meaningful capability analysis.  Eventually, if you intend to improve your process variation in order to significantly improve your process capability, it will be necessary to improve your measurement system too.  That’s why I say “your measurement system is marginally good enough to understand process variation” in this case.
    If your %Tolerance is good but the %Contribution/%Study is bad, then your Cp is probably pretty good and your measurement system (and process) variation is small compared to the tolerance.  This means your measurement system is adequate for capability analysis, but if you have reason to believe that process variation is costing you significant money or your Cpk is bad, then you will need to improve your measurement system in order to effectively study and eliminate that process variation or move the process mean.
    Hope this helps.
    Bede

    1
    #189522

    GB
    Participant

    Venerable Bede is in the house! Welcome back!!

    0
    #189524

    Saherngu
    Participant

    All your responses, answered my question.
    This is the story:
    We were evaluating Cpk and Ppk index from a CNC machine and we were trying to understand how measurement equipment affects these index. Also, MSA manual on page 116 has a note regarding process variation and tolerance.
    Sus respuestas contestaron mi pregunta.
    Ésta es la historia:
    Estábamos evaluando los indicadores Cpk y Ppk de una máquina CNC y queríamos entednder como afecta a éstos indicadores el valor de GRR. También el manual MSA en la página 116 tiene una nota refrente al análisis del GRR basado en la tolernaci y en la variación del proceso.

    0
    #189525

    Mikel
    Member

    Very nice. Could not have said it better.Welcome back

    0
    #189526

    Mikel
    Member

    You analyze based on tolerance to know if your measurement is
    adequate for product acceptance. I assume you know that is important.

    0
    #189529

    Mike Carnell
    Participant

    He was at the Usual Suspects Q&A session in Miami. The one on the table mooning you when you were reading Haikus. You probably didn’t catch that from where you were sitting.

    0
    #189539

    Venerable Bede
    Member

    Thanks for your kind words, HB (and later Stan and Mike Carnell).  It is nice to be back in the swift current  with big fish like you, Stan, Mike Carnell, Darth, Stevo, Bobby Butler, and others.  I recognize that the rest of us are simply drafting in your collective wake.

    0
    #189552

    Mike Carnell
    Participant

    Bede,
    WTF? Are you going to do that discussion about prime numbers or the origin of the octive or Englebrook Messerschmitt.
    Regards

    0
Viewing 17 posts - 1 through 17 (of 17 total)

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