Bob H

Hi Shawn,
I’m a long-term “practitioner”, and I agree that Lean, SS, and Shainin all have a place in the tool-box, but there are some areas where Shainin is “better”…
If you have a product, that can be physically dis-assembled, and re-assembled multiple times, while changing individual components, then Shainin’s “Component Swap” technique is…[Read more]

Don’t throw it out yet.  P-values can be affected by normality, sample size, etc. so a 0.001 difference may itself not be significant.  Also, the 0.05 critical value is a standard but is not set-in-stone — if I was testing water quality or radiation levels, I might use a more conservative value of 0.10.  Such a change would increase the cha…[Read more]

Your data is continuous.  The problem is that you are settling for daily averages rather than getting the individual data points.  You may have a problem with high value outliers, the ones making your customers angry, which are hidden inside an average.  At a minimum, try to get a spread measurement to go with each average (e.g. range, in…[Read more]

Normality is not a requirement for X-Bar & R, but at times significant non-normality (bi-modals, heavy skewing, etc.) can cause some SPC test failiures.
If you want to run a capability analysis (I’m assumign you have variable data here), you need to run both normality and stability tests first.  You need the normality test to insure you are…[Read more]

Are you looking for qualities to look for in candidates, job description/expectations, or certification requirements?

According to my favorite reference, “Implementing Six Sigma” by Breyfogle, XmR is another name for I-MR…

Love the fist analogy. Will definitely use that in my next Green Belt
class, THANKS!

I did do a Box Cox transformation, but it looked a little too
“scrubbed”. I like the idea of using the median, rather than the mean.
How do I calculate the CIs and is hypothesis testing the same with the
median as with the mean?

SixSigmaGuy is correct, my distribution looks roughly like a Chi
Square distribution with 9 or so degrees of freedom.

Thanks for the clarification on the direction of my skew. ;-)My data is continuous, but there are tons of data points below the
60 minute mark. (Probably because there is an informal Service
Level Agreement at the 60 minute mark.) Most of these problems are solved in an acceptable amount of
time, (the mean is around 120 minutes over +2,500…[Read more]

I’m looking for a way to say, “we can say with 95% confidence, that
our 2007 numbers are a statistical improvement over our 2006
numbers”. I haven’t tried Mann-Whitney yet, because I’m looking for
something simple to execute and something I can explain to non-
statisticians.

But you forget that the software development shop is a customer to itself.  The problem for over 5 decades has been “we are so busy trying to satisfy the customer, we forget that what we produce has to be maintained and evolved by us”.  If we create “crap” that satisfies the customer requirements, we do so at great cost to create and maintain s…[Read more]

But you forget that the software development shop is a customer to itself.  The problem for over 5 decades has been “we are so busy trying to satisfy the customer, we forget that what we produce has to be maintained and evolved by us”.  If we create “crap” that satisfies the customer requirements, we do so at great cost to create and maintain s…[Read more]

Jaran S.
     It is true that ISO 9000 and 6S are complementary.  You need one more ingrediant……..   what are the “industry best practices” for your particular problem domain (manufacturing, software development etc.).
In software one can complement ISO and 6S with the Software Engineering Institute’s Capability Maturity Model.  6S can wor…[Read more]

Save your money and get with a company serious about Six Sigma.  Let them pay for your training and certification.  It sounds like your experience will result in significant ROI for any Six Sigma company that will hire you.
Just my 2 cents.
Bob H (originally from 25th and 10th ;-)

Like any new/emerging “technology”, training is expensive because there isn’t  a plethora of trainers.  The day will probably come when Leaning Tree type organizations will teach courses on Six Sigma at $800 (or less) a pop.  As more trainers get in the business of teaching Six Sigma, the price will come down.  However, the cost is cheap at twic…[Read more]

Marc et al,
    I found doing a Process Model (in IDEF0) before a six sigma process map to be very helpful.  Also the Process Model tool(s) let your organize process “data/attributes” in terms of diagrams of processes, ICOMs (inputs, controls, outputs and mechanisms/agents). Also the tool(s) have notes panels that allow you to put down ent…[Read more]

I have found that starting a six sigma project using IDEF0 really helps to get all team members to fully understand the process before proceeding down to lower levels of abstraction using process maps, flow charts etc.  IDEF0 really helps build a process asset library (PAL)for future use in the company.
Bob H.  (801)  525-9841
Call anytime ni…[Read more]