Shimuk

Thanks again. By what extent the sensitivity of the X mR chart increases by turning on rules 5 and 6 as well ? Without the 5 and 6, an X mR can detect sigma shifts of 3 or more only.

Thanks….what you say makes sense. One question though….how would X mr chart of the data with additional rules turned on compare with the alternative approach of doing a EWMA chart instead ?

As I mentioned in the post, I am envisaging subgrouping by day and NOT by week……that will give me 3 data points for each “day” over the span of 3 months data. Contrary to washing out, it should highlight the day to day variation if any. However, this way, the subgroups are not time ordered….I am picking 4 observations each separated by a…[Read more]

Thanks Quality Colorado, it does. Do you have a web reference to the kind of VSM map you are referring to ….process along the top and variation along the bottom ?
Also what you seem to imply is that a single process map cannot really be drawn under these circumstances. The exercise will at least lead to standardization of the process after…[Read more]

I totally agree with you that one must go see the process in action. Also, that the process must be depicted, warts and all.
But how?……is it practically feasible to go and see how 50 different operators are working say a credit assessment process ? And how do you depict the differences on each step…….obviously you can’t draw 50 different…[Read more]

Hi Jim. You say Anova is robust to non-normality….so if I have 5 data variables with say 10 observations each and two of these variables do not show normality, should I do an Anova or a Kruskal Wallis ?

Thanks Eric, GLM, Jim. Can someone provide an unbalanced anova dataset with covariates please ? I would like to use it to see how th Glm workd out on it.

I checked minitab and found that unbalanced Anova experiments are to be handled by applying GLM. Any comments or thoughts ?

Hi lean and Zip
Would be helpful if you can send those at [email protected] please. Would like to modify that for a back office transational environment.
 

Hi Sridhar. I have worked in a hospital before and your papers on FMEA application in hospital setting will be very much appreciated.
 
Regards
Shimuk

Thank you folks for your useful feedback. Didn’t find the harmonic series in JMP though. I could find the Spectral density plotting option.

Faster results ofcourse! Also, greater productivity of six sigma resources.

Lots of thought provoking ideas. Thanks. But how do you level demand? How do you bring in takt because hourly call load is not always predictable? The idea about calling customers to resolve their problem sounds like a demand leveling and takt like approach but it is nothing short of a revolutionary idea.

Is a timeline for this evolution available somewhere?