Help: capability analysis for low volume and high mix products

Six Sigma – iSixSigma Forums General Forums General Help: capability analysis for low volume and high mix products

Viewing 5 posts - 1 through 5 (of 5 total)
  • Author
  • #53681


    Hi, There, if I have 10 similar products each different specifications, the quantity for each products is very low (less than 8) since they are customized products, how to measure the process capability? Thanks and Happy New Year!


    Robert Butler

    It sounds like you are facing the problem of short run process capability estimates – for starters you could look at pp.792-798 of Measuring Process Capability – Bothe. The section is titled “Short Run Process with Many Different Part Numbers”.



    Robert: I wish that I had the library that you have! Seems you have the right reference for any situation.

    BTW – Davis (Dave) and I live in the same quaint little town. Small world.


    Robert Butler

    Actually it isn’t that many – at least not as far as answering questions posed on this forum. My working statistics library is around 160 books but the number regularly cited by me on this forum is just a fraction of that.

    If you are interested – the books I most often cite here are, in no particular order:

    1. Measuring Process Capability – Bothe
    2. Applied Regression Analysis 2nd Edition – Draper and Smith
    3. Statistical Methods 7th Edition – Snedecor and Cochran
    4. The Cartoon Guide to Statistics – Gonick and Smith
    5. Statistical Models in Engineering – Hahn and Shapiro
    6. Regression Analysis by Example – Chatterjee and Price
    7. The Visual Display of Quantitative Information – Tufte
    8. Quality Control and Industrial Statistics 4th Edition – Duncan
    9. Understanding Industrial Designed Experiments – Schmidt and Launsby
    10. Statistics for Experimenters – Box, Hunter, and Hunter
    11. The Design and Analysis of Industrial Experiments 2nd Edition – Davies
    12. Categorical Data Analysis – Agresti
    13. Regression Methods in Biostatistics – Vittinghoff, Glidden, Shiboski, and McCulloch
    14. Statistical Methods for Rates and Proportions – 3rd Edition – Fleiss, Levin, and Paik
    15. Fundamental Concepts in the Design of Experiments – 2nd Edition – Hicks
    16. Fitting Equations to Data – 2nd Edition – Daniel and Wood

    Both #11 and #15 are old. However, they date from the pre-statistics software package era and consequently have lots of small data sets that are used again and again to illustrate points in the book. I have written and saved a number of these data sets to various Excel spread sheets and I use them to test new code, new ideas, and to illustrate points to individuals whom I may be either helping or tutoring.

    #3 also is old but it is great, easily understood, boilerplate – no theory just here-is-how-you-do-this-given-that-this-is-what-you-think-you-need.

    #5 is the cleanest discussion of distributions and distribution theory I’ve ever read.

    #6 is very readable and I often recommend it in conjunction with #4 when someone askes for “a little more” information about regression and statistical analysis.

    #9 is a nuts-and-bolts encyclopedia of design methods

    #14, 1, and 2 should have the titles “Everything you wanted to know about _________ but were afraid to ask.



    I’ve got to admit, I’ve got more than half of those, and it’s a great list.

    The one that I find very helpful in inititating newbies, is the Cartoon Guide to Statistics. My daughter just started college last fall and is going to be a math major. She laughed when I gave it to her, but said that it was the biggest help in understanding what the prof was going over in her stats concepts course.

    Two that I would add (albeit not specifically statistics related but more thinking process) would be The Goal and Critical Chain, both by Eli Goldratt. I have found the 5 focusing steps to be a powerful model for CI in general.

    Happy New Year to you Robert!

Viewing 5 posts - 1 through 5 (of 5 total)

You must be logged in to reply to this topic.