How to Analyze Non-normal Data and Indicate the Capability of the Process?

Six Sigma – iSixSigma Forums General Forums Tools & Templates How to Analyze Non-normal Data and Indicate the Capability of the Process?

Viewing 5 posts - 1 through 5 (of 5 total)
  • Author
  • #239998


    In an earlier discussion i have mentioned that we sampled a certain process and have collected a data that appeared to be out of control in several locations, but now after closely examining the process we determined the problem and worked in order to fix and have succeeded and now all data are in control but exhibit a non normal distribution as shown in the attachments.

    We would like now to analyze these data and examine the capability of the process and if these data are indeed valid for the process, What should we do?

    1. You must be signed in to download files.
    2. You must be signed in to download files.

    Robert Butler

    I’m not trying to be funny at your expense but if you really have a process like this then then next Nobel Prize in Physics will be yours on demand because you have found a process that exhibits zero variation which means you have invalidated the second law of thermodynamics – something that heretofore was deemed physically impossible.

    Seriously, it appears you have a measurement system that is completely inadequate with respect to detecting changes in your process. This, in turn, eliminates any possibility of computing process capability. I realize you said in your post to the other thread you started that you couldn’t afford a better measuring device but what I don’t understand is how, with the same measuring device, you could go from the plots you provided in the other thread to the plots you provided in this one.

    If both of these plots are of the same process with the only difference between the two sets of data being some kind of “correction” then I would first want to re-examine the correction process because whatever it is doing is wrong.

    One additional thought – in both of your threads you do not give any sense of the elapsed time for taking the measurements you have illustrated in your plots. I assumed your first plots in the other thread represented independent samples taken over a “reasonable” period of time (at a minimum multiple days). If the time interval is very short then, not only are you at risk of not having independent samples but, if it is short enough it may explain the odd plot you just provided – i.e. the interval was so short that what variation was exhibited was indeed too small to be detected by your measurement device.

    • This reply was modified 1 year, 4 months ago by Robert Butler.
    • This reply was modified 1 year, 4 months ago by Katie Barry.

    Robert Ratcliffe

    Karim, the I-MR chart looks as if you don’t have sufficient discrimination in the units of measure; therefore, you may have an issue with the Measurement System in terms of how finely tuned the machine is.

    Another thing you can do with variable individual data points is to sub-group your data (if feasible) in an X-bar/R (subgroups of 3-8) or X-bar/S chart (subgroups of 9+). This method takes uses the Central Limit Theorem to potentially smooth out the “between” variability that was present in your first example in the other thread (not pictured here).




    Firat of all i really appreciate the time you give to answer, and let me try to explain, the machine we are dicussing is an acid levelling machine used in automotive battery industry, its function is to level the acid inside the battery after being charged this level is controlled in the machine and have a certain specification given by the operator to the machine, in the first thread these samples were taken after there has been claims by some customers that there has been issues with some batteries when operating the car and tha acid was to overflow from the battery and ultimatly the battery was defected, so after examining the while process( not only this machine but every possible reason for this issue to happen) we came to a conclusion that this machine is the problem and after sitting with the maintenance department inside the factory and brainstorm about the possible soultion we decided to conduct a full maintenance to the machine by replacing some of the parts in it to a new one, and after that we take samples( with the same method used eralier and the same time intervals but less number of samples)
    Our explanation that there is no variation exisiting is that the tool we are using is not accurate enough to exhibit any variation, So is the soultion now that we have to use a more accurate tool??


    Chris Seider

    You’ve just described a gage R&R. go for it and see what the folks are suggesting–the gage is not precise.

Viewing 5 posts - 1 through 5 (of 5 total)

You must be logged in to reply to this topic.