iSixSigma

Control Chart for Growth Rates (percentages)

Six Sigma – iSixSigma Forums Old Forums General Control Chart for Growth Rates (percentages)

Viewing 9 posts - 1 through 9 (of 9 total)
  • Author
    Posts
  • #35429

    lbl627
    Participant

    How do I control chart percentages?  I have monthly growth rates and I’m not sure if control charts are necessary or even doable.  Please help!
     

    0
    #99677

    Tim Folkerts
    Member

    In a business setting, I would tend to say that a control chart for growth rate wouldn’t make sense.  To make a control chart, the first requirement is that the precess is “in control”.  For most companies, growth is anything but in control.  Every month it seems there is a new “special cause” of variation – a supplier can’t send a shipment on time; a customer needs a rush order; a salesman lands a big new customer; you release a new product; your competition releases a new product; ….
     
    Tim F

    0
    #100667

    Markert
    Participant

    1. Tim F. – The purpose of a control chart is to detect when a process is out of control! Why would you therefore avoid charting data which was out of control? How would you know?
    2. Any data can be charted on a control chart (although it may not always be rational to do so). Suggest you take a look at “Making Sense of Data” by Don Wheeler (http://www.spcpress.com). This has good stuff on charting trended data, and also data subject to seasonal variation, which may or may not be an issue for this application.
    Hope this helps
    Phil

    0
    #100677

    austin
    Participant

    Yeah the control chart will also let you know if you have a trend happening … be it good or bad.  You can do a lot with control charts

    0
    #100681

    Mauricio Salinas
    Participant

    If you have history, you can use time series to chart all the data that you already have and forecast future growth rate. With this forecast you calculate confidence interval for 70% or 90% confidence, depending on what you want, and if the actual data falls inside this interval you can say with reasonable confidence that your growht rate is in control.
    Of course, for the forecast, you should probably analize your data for any dominant factor (change in economy, any change in the market or industry, etc), trend and/or seasonality so that you can use a good model to make your prediction.
    This is the closest form I know to make a control chart with this kind of data.

    0
    #100701

    Sphynxras
    Member

    While we are on the subject of control charts and percentages, I am trying to figure out the best method for charting recovery(yield) rates for a mfg. process-aluminum extrusion. Since I am dealing with a non-normal distribution that encompasses data from 4 alloy types, with subgroup size=1 and independant, can I use the Tukey test? I considered using an IR chart, but the fact that the data failed normality testing made me consider other options.
    Should I use Tukey, or does anybody have any other suggestions?
    thanks!!

    0
    #100703

    Savage
    Participant

    Your data doesn’t have to be normal to use control charts.  Check Wheeler and Montgomery and Grant and they’ll all say the same thing.  You can also check the mathematical proof yourself.

    0
    #100704

    Sphynxras
    Member

    Thanks for the reply, Matt! This will make life much easier! :)

    0
    #100893

    ALEK DE
    Participant

    Control charts can always be used for growth rate . I’ll also suggest that decide your band for past data & future data . Draw charts keeping the control limits same as for the past data (using past mean & sigma ). You will really know whether growth rate has significantly changed or not .
     
    Rgds
    Alek

    0
Viewing 9 posts - 1 through 9 (of 9 total)

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