• APQC has some benchmarks in this area (Accounts Payable and Expense Reimbursement).  I am leading similar effort (as a contractor) for a governement entitity.  Our process is very immature and has a number of pain points.  We are still working this through a DMAIC but have focused our Measure and Analyze phases around a Value Stream Mapping exerc…[Read more]

  • Is the primary interaction by phone – can you call monitor?  A lot of financial services organizations have the ability to log into service or sales  lines, or even select segments or call types.  Usually this is totally unobtrusive providing not indication to the customers (or the service reps – assumes that the “your calls may be moni…[Read more]

  • Hi Morning Li,
    You are correct in that the conventional control charts will not work in your situation. Fortunately, there are other types of charts for high-mix/low-volume production. Unfortunately, they are a little more difficult to apply than the conventional charts.
    The following two books explain these other kinds of control charting…[Read more]

  • Hi Sharan,
    In the first edition of AIAG’s SPC manual, it is unclear whether you need stability for estimating Pp and Ppk. However, in AIAG’s second edition of this manual (published late in 2005), they make it very clear that the process should be in control before making any estimates of process capability. Here’s one of the quotes (page…[Read more]

  • Hi Ken,
    You could try using the analysis of means (ANOM) to compare the three sample average percentages. This technique can handle different sample sizes.

  • Hi Weeze,
    Check out post number 84034, which is my reponse to a similar question.

  • Loehr replied to the topic process capabilty in the forum General 14 years, 7 months ago

    Hi Pradeep,
    Like the other responders, I am not sure what you mean by an “open tolerance,” but if you are referring to a feature that does not have specification limits then, as already mentioned, the traditional capability metrics such as Cp, Cpk, Pp, and Ppk will not work as they require specification limits.
    However, there are some n…[Read more]

  • Loehr replied to the topic Cpk=Cp(1-k) in the forum General 14 years, 7 months ago

    Hi Nivra,
    The derivation of that formula is given in Bothe’s book, starting on the bottom of page 338.
    I hope this helps.

  • Hi Weeze,
    You have asked a lot of good questions about capability metrics, in fact, many of the same ones that I have wondered about.  A book that has helped me find my way out of the capability maze is Measuring Process Capability, by Davis Bothe.
    This book explains all the various capability indices you mentioned (and even many more!) and a…[Read more]

  • Loehr replied to the topic Cp Vs Pp in the forum General 14 years, 9 months ago

    Hi Nicolas,
    There has been much confusion over the years about capability indices due to the double meaning of the phrases “short term” and “long term.”
    Many practitioners assume that a capability metric like Cpk is called “short-term” because the data used to estimate it were collected over a short period of time (for example, a short run of…[Read more]

  • Hi Pat,
    We have worked with the Asia-Pacific Research Centre in Singapore and have had good results (both black belt and green belt).  Contact Mr. Chin Hak Wong at [email protected] .
    I hope they will be able to help you.

  • Hi Robert,
    The story I have read is that Larry Bossidy, CEO of Allied Signal and former GE executive, was the person who first introduced the concept of six sigma to Jack Welch.  Welch at that time (around 1994) was a member of the board of directors for Allied Signal.
    However, I don’t know if that means Bossidy was the person who actually t…[Read more]

  • Loehr replied to the topic Attribute Capability in the forum General 15 years, 4 months ago

    Hi Scott,
    I learned about these formulas from a book by Davis Bothe called Measuring Process Capability.  There is a full chapter (Chap. 11) of about 120 pages devoted to calculating confidence bounds for all the various capability metrics, both for variable data as well as attribute data. He also provides a very long list of references at the…[Read more]

  • Loehr replied to the topic Attribute Capability in the forum General 15 years, 4 months ago

    Hi Scott,
    One test for percentages you might wish to consider is the one known as “success” testing.  If you want to demonstrate the the process average percentage nonconforming is p or less with 1 – alpha confidence, then you must run at least m pieces off this process without seeing a single nonconforming item.  m is found with this f…[Read more]

  • Hi Bob,
    You bring out an excellent point.  There are often different needs for a “capability study”.  To understand the behavior of this process, you should take consecutive pieces out a single cavity and plot these results on a control chart for just that cavity.  A separate chart would be needed for each cavity.
    This approach will allow pe…[Read more]

  • Hi,
    The goal was to center the process at the middle of the tolerance and get a Cpk index of 1.33.  If the average is centered in the middle of the tolerance, then mu – LSL will equal USL – mu, both of which will equal 1/2 (USL – LSL), where mu is the process average and USL – LSL is the width of the tolerance.
    If we let 8sigma = USL – LSL,…[Read more]

  • Hi Miguel,
    Which value to use depends on what you want to do with the gage.
    If you are planning to use the gage for inspection purposes, then the % of gage variation to tolerance value should be used to determine if the gage is adequate or not for sorting good parts from bad.
    If you are planning to use the gage for process improvement or for…[Read more]

  • Hi Devang,
    The control limits for an X-bar chart are based on short-term (sometimes called within-subgroup or piece-to-piece) variation.  Because subgroups are taken at different times, the variation between subgroup averages plotted on the X-bar chart represents long-term (sometimes called overall or time-to-time) variation.
    If all the s…[Read more]

  • Hi Devang,
    Assuming one cavity is involved, it would be best to collect five consecutive shots.  Here is the reason why.  Before estimating process capability, you should check to see if the process is in a reasonably good state of control.  This can be done with a control chart (in your case, an X-bar, R chart with a subgroup size eqaul to 5)…[Read more]

  • Hi Tim,
    Assuming the average is centered at the middle of the tolerance, sigma for the length of the first component (sigma1) is found as follows (given a tolerance of  +- .075 mm and a Cpk of 1.33, which means 8sigma1 = USL – LSL);
    8sigma1 = USL – LSL
    8 sigma1 = .075 – -.075
    8 sigma1 = .150
    sigma1 = .15 / 8
    The sigmas for the second co…[Read more]

  • Load More