THURSDAY, AUGUST 28, 2014
Font Size

MBBinWI

  • I wish you had taken a training model that had you do a project DURING your time in classroom (e.g. demonstrating tools/strategy in between lectures).

    Someone must be doing some continuous improvement in your organization. It sounds like no one thinks you’d be adding much assistance to a problem solving team without needing a lot of…[Read more]

  • I’m glad @rbutler has spent time on this.

    I would have said to use a tool like Minitab and simply rerun the experiment and fully randomize if possible. If you know which materials you’re using, then your blocks are easily identifiable and you can set up a 2^(6-2) with 4 blocks and get some information about how to proceed if you need a more…[Read more]

  • Nicely stated @MBBinWI . I often give the clunky answer of long term sigma has most/all of the special causes which isn’t as intuitive of a statement.

  • LOL @Marke I smile when you imply baggage claim process doesn’t need to be as high of a sigma level…I guess you haven’t had airlines forget to load your bags onto the plane. :)
    Of course I understand your thoughts.

    Cheers.

    Jose, as a side note….careful just multiplying Cpk x 3 to get the OVERALL sigma level…don’t forget there are defects…[Read more]

  • Cpk gives you a defect rate at the closest spec although I prefer the ppm overall defective from both sides of a 2 sided spec.

    Look into DPMO on this website.

  • Consider an excellent book out there…

    Note that I get no personal or financial benefit for sales of the link below.

    http://www.amazon.com/5S-Operators-Pillars-Workplace-Organization/dp/1563271230/ref=sr_1_1?s=books&ie=UTF8&qid=1408566025&sr=1-1&keywords=5s+for+operators

  • You are the only one who EVER measures this characteristic? Does the facility shut down when you are on leave for vacation, sickness, etc?

    Do you have more than one device that measures CH2O VOC’s? They can be your “operators” but I’d advise you to find another person who measures in your place and look at the variation across the instruments/personnel.

  • @acohen

    “If the data is large enough you could disregard the normality issue ….” I don’t understand your statement. I don’t agree with it–it’s too broad of a statement to those newer to analysis tools.

  • It is always a good idea to look at your data using graph’s. I’ve never had an instance where the AD showed my data to be normal, but my graph showed non-normal. BUT, I have, in the past, been able to look at the data graphically, after determining the data to be non-normal, and find outliers, which caused the distribution to be non-normal. I…[Read more]

  • @cprao101 It looks like this is a homework question. You are unlikely to get a response unless you share your thought process first. What do you think the answer is? Why? What’s giving you trouble and why? The iSixSigma audience is willing to help but they are not here to do your work for you.

  • Chris Seider replied to the topic Black Belt? in the forum Training 1 week, 5 days ago

    @ShelbyJarvis

    Nice comments.

  • @yazannajjar

    Have you defined your problem statement? Don’t just jump into a bunch of metrics.

    My own 2 cents.

  • Pick an area where other problems obviously exist….inventory, cost, accidents, throughput. You will get synergies if you look hard enough. Good luck.

  • Really @CFB? Doesn’t one need to know type of distribution to select statistical tests? I think so.

  • We’re assuming you are talking about a test like the Anderson Darling normality test?

    It’s always wise to look at the graphical summary (if using Minitab) of your data and see how the shape looks and be wary of solely looking at the p-value from a normality test because there are numerous reasons why the AD test may fail the hypothesis that…[Read more]

  • Just gather randomly samples from each machine, measure offline, and then do your capability analysis for EACH machine separately.

    You can’t get a sense of ppm defective just looking at max’s and min’s.

  • Chris Seider replied to the topic Cpk and Ppk in the forum Tools & Templates 2 weeks, 1 day ago

    Let’s all not forget that Measurement System Analysis might be impacting the results of Cp being less than Ppk which we wouldn’t expect. Also, how many samples (observations) are we using to estimate s.d.?

    @MarcShaeffers
    How can one state that someone isn’t taking subgroups correctly? How one defines subgroups varies depending on the sources…[Read more]

  • Load More

Login Form