• Sadly, I saw study materials from a course at a university MBA program state that 6 sigma is because of the +/- 3 sigma.  Amazing how diluted the teachings are from the early 90s.

    I wasn’t in the class but told the student that wasn’t six sigma meant.

      2 weeks ago

  • I’d say fix the measurement system.

    Just saying AFTER fixed, it would probably allow the Anderson Darling test to come to another conclusion. 2 weeks, 3 days ago

  • @mike-carnell “what about 4.5 sigma”?  :) 2 weeks, 4 days ago

  • the correct ones?

    My curiosity is piqued. 2 weeks, 4 days ago

  • If you followed the DMAIC methodology you would have checked your measurement system.  My educated guess is it won’t pass a good variable gage R&R.  This is PART of the reason your normality tests are failing but don’t fret too much if it doesn’t pass…median shifts are just as valid for statistical confidence–yes I know how my friend R. B…[Read more]

  • Have you considered following the classic DMAIC process?  You should already have gathered data on some of the X’s and looked for POTENTIAL/PROBABLE X’s in the measure/analyze phases.

    If you are trying to create a DOE and can’t get support–are you 1. making sure the process participants are part of the team and 2. giving updates to the team and…[Read more]

  • Yes, your service industry definitely has many peaks and valleys–especially if in the retail arena.

    Some thoughts to consider

    1.  How “easy” is past data for the metric that was agreed by the project sponsor/project leader?  If easy, see if any sign of seasonality and that may help you decide on what time frame for getting an average or s…[Read more]

  • Control charts are great…just always don’t folks to think they are doing capability analysis on means which is where CLT applies.  Capability analysis is always done on individual readings. 2 months ago

  • I think you’ve forgotten an important thing.  Do a gage R&R for both devices before going first. 5 months, 2 weeks ago

  • @fausto.galetto   :)  “Right” is spelled the way I typed….not the way in your cute survey.

    Yea, I’m “poking the bear” but I think you broke some china in the shop beforehand.   FYI, I’ll restate….doing a control chart on MTBF would seem to be a silly tool for practical use.  If the valve is truly causing a plurality of downtime–break it do…[Read more]

  • MTBF with a control chart seems like a stretch to me–what kind of control plan would you have?  I’d look at pareto of lost time and drill down to the component(s) and fix the issue(s) and see if the pareto of lost time showed a difference. 6 months, 3 weeks ago

  • Is this twitter? :)

    Happy new year to my buds such as @mike-carnell and @rubtler and @darth 6 months, 3 weeks ago

  • I can only imagine your approach COULD work IF…

    1.  The model was validated.  No model is 100% so direction better be correct and an acceptable magnitude of error isn’t large relative to predictive changes in results.  The model implies you apply the factors AS IS and get “close” to actual results.  Don’t ask me online (LOL) what close is but…[Read more]

  • Happy Holidays to all this season 7 months, 4 weeks ago

  • Yes, the power of statistics is to use RANDOM sampling to get population mean estimates and see if you can shift those estimates after process changes.  Yes, it could be the same for variation. 8 months ago

  • It shouldn’t be any different.  There are losses to attack. 9 months, 1 week ago

  • OEE is a great tool but not easy to “measure” if loss capturing isn’t credible.

    I’ve helped many clients improve 10% output on poorly performing lines by finding either 1) easy to tackle significant loss buckets or 2) just providing instant feedback (even if just hourly) of how production is occurring so management and supervision has a chance to…[Read more]

  • Consider what you mean by target of 100%.

    Your control chart would be great if no losses were assumed and the theoretical output was included–not planned output. 9 months, 1 week ago

  • Load More