iSixSigma

Control chart for groups of data

Six Sigma – iSixSigma › Forums › Old Forums › General › Control chart for groups of data

Viewing 16 posts - 1 through 16 (of 16 total)
  • Author
    Posts
  • June 19, 2008 at 9:59 am #50332

    Richard
    ★ 10 Years ★
    Member

    I have 6 lots of data from a process and would like to plot using IM-R chart or similar.  I was wondering is it possible to put all the data onto the same chart and have limiits calculated for each section (x6)? So I can show before improvement, during and after all on the same chart? Do I just need to use subscipts for each group and if so how do I then chart on 1 sheet?
     
    Thanks

    0
    June 19, 2008 at 10:26 am #172975

    Kannan
    ★ 10 Years ★
    Participant

    Whats the harm in using different charts. It would look neat as well
     

    0
    June 19, 2008 at 11:57 am #172979

    Bower Chiel
    ★ 10 Years ★
    Participant

    Hi Richard
    What software are you using?
    Bower Chiel
     
     

    0
    June 19, 2008 at 12:08 pm #172980

    Richard
    ★ 10 Years ★
    Member

    Hi Bower,
    Minitab, Yes sorry that was an after thought.
     
    Richard

    0
    June 19, 2008 at 12:28 pm #172983

    Bower Chiel
    ★ 10 Years ★
    Participant

    Hi Richard
    You can achieve what I think you require by having one column for your individual data values and another for the lot number: – 1, 2, 3, ….  .  Under I Chart Options the Stages tab enables you to define stages using the lot number column.
    Good Luck!
    Bower

    0
    June 19, 2008 at 12:46 pm #172984

    Kannan
    ★ 10 Years ★
    Participant

    Good one Bower

    0
    June 19, 2008 at 3:05 pm #172997

    Richard
    ★ 10 Years ★
    Member

    Hi Bower,
    Thanks for that. I had tried that and didn’t give em exactly what I was after? Its along the right line but I would like to disply it on a continuous  single chart with limits recalculated for each section in sequence? Perhaps you cannot do it but I’m sure I’ve seen it done somewhere. I’ll keep trying, any other ideas?
    Richard 

    0
    June 19, 2008 at 3:21 pm #172999

    Bower Chiel
    ★ 10 Years ★
    Participant

    Hi RichardIf you post an e-mail address I’ll send you some more details.Bower Chiel

    0
    June 21, 2008 at 5:05 am #173069

    Saleem
    ★ 10 Years ★
    Member

    can u send me the details
     
    my id is [email protected]

    0
    June 21, 2008 at 5:26 am #173070

    Venugopal G
    ★ 10 Years ★
    Member

    Hi Richard,
    Can U send me the e-mail ID so that I can post an example. If I understood correctly, U are looking for a single chart with both before and after situations and also the control limits have changed indicating the improvement and there is a shift in the mean.
    Regards,
    G. Venugopal
     

    0
    June 21, 2008 at 5:33 am #173071

    Viswa
    ★ 10 Years ★
    Member

    Hi Richard!
    You will have one more Tab in the I-MR option called “Display”. Under it just tick the “display control limit/center line labels for all stages”. Click OK. u can see the means and comtrol limits at all stages.
     
    Regards
    Viswa

    0
    June 24, 2008 at 9:02 pm #173172

    Saherngu
    ★ 10 Years ★
    Participant

    I prefer using one chart showing the before and after with the computed ucl and lcl based on pre change.

    0
    June 25, 2008 at 7:29 am #173180

    Richard
    ★ 10 Years ★
    Member

    Hi Venugopal / Bower
    My email address: [email protected]
    Thanks for your help
    Richard

    0
    June 25, 2008 at 11:02 am #173182

    Ron
    ★ 10 Years ★
    Member

    You have described the requirement for the most powerful shewhart chart..the Xbar & R chart. It displays within and between process variaiton.
     

    0
    June 25, 2008 at 11:09 am #173184

    Richard
    ★ 10 Years ★
    Member

    Hi Thanks everyone I’ve cracked it! it was as suggested by Bower the Stges under IMR Option tab. I must of had other items selected whilst rying to find the right solution that was confusing issues.
     
    Cheers

    0
    June 25, 2008 at 4:35 pm #173198

    Ron
    ★ 10 Years ★
    Member

    ricjard,
    6 groups of data is not sufficient to calculate fixed control limits. Typically a minimum of 25 to 30 sample lots of data is required statistically.

    0
  • Author
    Posts
Viewing 16 posts - 1 through 16 (of 16 total)

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