iSixSigma

OUT OF CONTROL : DOUBT

Six Sigma – iSixSigma Forums Old Forums General OUT OF CONTROL : DOUBT

Viewing 11 posts - 1 through 11 (of 11 total)
  • Author
    Posts
  • #45845

    jayasuresh
    Participant

    HI all,

    This is surya rom INDIA.

    I have a small problem with out of control. 

    Suppose I have the data for 20 months,

    I applied P chart.

    But 2 points are above the UCL and 5 points are below the LCL. 

    After removing the Above UCL(2 points), All teh points including bellow LCL points.

    What should I take for Stable data.

    Shall I remove All 7 points/ Only above LCL?

    NOte : LCL is very close to 0 (i.e, 100% Quality )

    Please give the reply….

    0
    #150679

    Rohan Kadam
    Member

    Ideally one should not consider points beyond UCL.

    0
    #150685

    Dennis Craggs
    Participant

    The p chart indicates you have an unstable process. The basic assumption behind the p chart is that there is a stable % defective that describes the process. Eliminating undesireable points (beyond the control limits) will not improve the process or the analysis. Determine what the root cause of the defective units and remove the root cause.

    0
    #150686

    Charles H.
    Participant

    I would add that you should also investigate the five OOC conditions on the LCL.  Find out why those happened and make it the SOP for the process, conduct training or whatever needs to be done to make it happen all the time.  The key point is spot on – investigate the OOC conditions and find out why they happened.

    0
    #150696

    6Sigma Newbie
    Participant

    Agree with Charles..Basically identify (by askin team members) and establish the diff b/w the process/ what happending which resulted in 2 points which went over UCL and the one which resulted in points almost hitting 100% quality. Try and bridge the Gap accordingly by setting controls.

    0
    #150698

    Theo
    Member

    The other posts are on the right track but first you should check your data for the assumptions of p charts.  If these assumptions are not met, the increased sensitivity of p charts may be creating out of control points.
    There is no need to remove points.  Your process appears to be unstable.
    Read Wheeler “Understanding SPC”.

    0
    #150701

    p chart
    Participant

    What is your total sample size per month? You may get these types of erratic out-of-control points simply because you are throwing very large datasets into your SPC package. As a result your control limits will shrink and you’ll run into the type of problem you are describing. Control charts are designed for the active review of processes not the retroactive reporting of defects!

    0
    #150702

    lin
    Participant

    It may be that your subgroup size is too large.  This makes the control limits very tight and generate out of control points.  You can move to weekly data and keep the p control chart.  There is also a good possibility that, if you are monitoring more than one defect, that the conditions for using the p control chart are not met.
    Consider using an individuals control chart for the %.

    0
    #150703

    Ropp
    Participant

    This idiot shows obvious signs of being a six sigma black belt.
    p charts don’t have sub-groups you moron.

    0
    #150704

    p chart
    Participant

    It is absolutely amusing to see a rude and unprofessional response go up in the flames of its own ignorance: factually, there is a difference between a subgroup and a rational subgroup, and the term subgroup is used for both attribute and variable charts. Rational subgrouping, on the other hand, is used only in the context of variable charts. And that claim has been read into a response that clearly states a valid response to the answer at hand.

    0
    #150705

    lin
    Participant

    Hello Dave,
    As stated by the previous poster, p charts do have subgroups.  For example, suppose you are monitoring the fraction of invoices that are priced correctly each day.  You randomly select 100 invoices a day.  This is your subgroup and n is the subgroup size – in this case 100.  Suppose 98 of these are priced correctly.  This is np.  You can then calculate p for this subgroup as follows:
    p = np/n = 98/100 = 0.98
    So, there are subgroups with p charts.
    Best of luck to you. 

    0
Viewing 11 posts - 1 through 11 (of 11 total)

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