iSixSigma

Measurement System Analysis

Six Sigma – iSixSigma Forums General Forums Methodology Measurement System Analysis

Viewing 7 posts - 1 through 7 (of 7 total)
  • Author
    Posts
  • #55281

    Zineb
    Participant

    Hi,

    How can we analyze our measurement system by using historical data ?

    thanks

    0
    #199430

    Sam
    Participant

    Hi Zineb: What are you trying to achieve by analyzing the historical data? Do you question the gauges you used to collect those data?
    I assume you saw something that you didn’t like in your data and want to know if it is coming from your “old” measurement device and for some reason you cannot find that out by doing a study on the current measurement device. If that is the case you have to give more details about your data. What type of data you have? Can you split the data you already have using the operators who measured it? Do you have multiple measurements of each data by the same old gauge? give us more details

    0
    #199432

    Zineb
    Participant

    Hi Peach,

    My data concerns logistics process cycle time, i have historical emails which can help me to determine the start and the end points of my process, so i can count the number of days, it means the cycle time.
    Now, I’m at the measure phase and i have to analyze the measurement system, so i want to perform the attribute gage R&R on the number of days calculated by using historical emails

    thanks

    0
    #199434

    Everton da Silva Paula
    Participant

    I am sorry! I wouldn’t do that this way!

    Perform the tests according AIAG Manual!

    Hugs!

    0
    #199436

    Chris Seider
    Participant

    Did you ask a similar question earlier in the week?

    Look for my past posts on cycle time MSA’s.

    0
    #199440

    Zineb
    Participant

    Yes, i posted it, but it wasn’t clear for me

    0
    #199473

    Prasad
    Guest

    @ga.zineb
    I think the past data will not hold good for better MSA. Is the data contains repetitions (which is critical for such MSA as we generally don’t find in historical) I feel for such attribute type data we require master (SEM) & subsequent trials with repetitions with appraisers to get better kappa. For this fresh data will be of more use. Have a quick data runs and try attribute agreement analysis.

    0
Viewing 7 posts - 1 through 7 (of 7 total)

You must be logged in to reply to this topic.