Ken Myers


  • Ken Myers replied to the topic Reduced model in DOE in the forum General 16 years ago

    Hi Anderson,
    Here’s my take on basic empirical model development with classical DOE’s:
    1a. When developing an empirical model you want to include only the terms that are statistically significant.  These terms are the ones that have the greatest likelihood of providing an explanation of the observed effects.  Any additional non-significant t…[Read more]

  • Ken Myers replied to the topic control limits in the forum General 16 years ago

    When computing the limits for process control charts it’s common practice to use a short-term estimate of variation.  This is done by computing the average variation observed within a subgroup. 
    When using a Range chart you would compute the average Range, and then multiply by the control chart constant A2.  You would then add and su…[Read more]

  • Ken Myers replied to the topic Could you explain….? in the forum General 16 years ago

    I happend to drop in to look at the postings after a long working stent in Europe.  It’s good to see that many appreciate a few of my past ramblings(postings). 
    I referred to Davis Bothe in the past concerning the explanation of using a +/-1.5SD shift in the process mean for computing the sigma qualify level.  I should mentioned that his boo…[Read more]

  • Ken Myers replied to the topic measureable quality in the forum General 18 years, 10 months ago

    Russ,Thanks for your candid response. However, I’m somewhat confused with it… The metric and calculations you used do not appear to be associated with a unit production volume in order to baseline the true improvement opportunity. That not withstanding, the COPQ in your example before improvement was $100,000 and after $70,000. I agree it…[Read more]

  • Denton,You and I have had discussions of point in past postings on this site. I too have enjoyed the interchange. You sound like a very knowledgeable guy, and your guidance is usually accurate. However, between the two of us a mid-course correction concerning the points raised in this thread might be helpful. I’m not too fluent in religious…[Read more]

  • Ken Myers replied to the topic how shift is calculated? in the forum General 19 years ago

    Victor,You know, there are some imaginative answers in this thread… :-)) While I’m not a practicing statistician, I know enough to get myself in trouble. Mike’s explanation above is very colorful, but I’ve never seen anything like it in journals, trade articles, or textbooks. Great stuff none the less.The use of a 1.5 standard deviation…[Read more]

  • Ken Myers replied to the topic Capability Studies in the forum General 19 years ago

    Darren,A rational subgroup is one which provides you with the lowest calculated short-term(within group) standard deviation. This is done to allow you to detect changes in your process average over time using the control limits on an SPC chart. In your question above you said that you run 88 components in BATCH sizes of 4. Therefore, one batch…[Read more]

  • Ken Myers replied to the topic sampling in the forum General 19 years ago

    Marc,What’s your reason for the sample?Ken

  • Ken Myers replied to the topic Capability Studies in the forum General 19 years ago

    Darren,If your queue size per batch is in fact 4 units, then per my original discussion the natural sample per subgroup would be as the others above suggested, n=4. Again, this selection should yield the lowest estimated standard deviation among all the possible sampling possibilities. Selecting samples on this basis would be termed “rational”…[Read more]

  • Ken Myers replied to the topic Making FMEA fun in the forum General 19 years ago

    Don’t know if there’s anything out there to make FMEA’s fun… But, to make them a bit easier you can use Dyadem’s product called FMEA 5. It takes a log of the tedium out of the process, and allows you to focus more on the process. Catch them at

  • Ken Myers replied to the topic Ppk Calculation in the forum General 19 years ago


    Thanks for your input above. The use of within subgroup standard deviation over the range to compute Cp is fine. Using Ppk with an unstable process provide little useable information, but it appears you’re not using the process indices to drive your improvement efforts. Instead, as I understand you are using other run indicators to…[Read more]

  • Ken Myers replied to the topic Ppk Calculation in the forum General 19 years ago

    Well, it looks like Bernie opened yet another lively discussion supporting Capability vs. Performance indices. I truely chuckle when I look at some of the responses in this message thread, because I’m sure some of you are still very confused about all of this stuff. :-?

    Matt, to respond to your posting dated the 11th, no one, and I mean no one…[Read more]

  • Ken Myers replied to the topic Ppk Calculation in the forum General 19 years ago

    You know guys, I seem to have been running for over 25 years with the wrong understanding about this stuff… Before the AIAG invented Process Performance Indices there existed short versus long term variation. These lay terms were typically used to describe within(short-term) versus between(long-term) subgroup variation. These terms have…[Read more]

  • Stan,
    To quote Wheeler years after Shewhart, “3-Sigma limits are NOT probability limits.”  Shewhart considered many criterion before selecting 3-Sigma limits to observe process change.  Jim above suggests there are situations where the Individuals or other control charts might be less effective at signaling a change in the process.  I would cert…[Read more]

  • Gary,
    I looked over Wheeler’s work on the Discrimination ratio, and two observations popped out in comparing it to your response above.  First, in Wheeler’s reference one(1) is SUBTRACTED from the ratio of variances instead of ADDED to the ratio as indicated above.  Second, the subtraction of 1 from the ratio comes about as a natural algebraic d…[Read more]

  • In limited support of Denton’s use of the sqrt(2) for defining the resolution point of the image between two light sources, which is not correct, I found something in support of the possible use of a normal distribution of light to explain how a continuum of measures could be broken up into distinct levels for describing its resolving power. This…[Read more]

  • Denton,I too think your explanation is interesting…However, I was not able to locate quite the same definition for Rayleigh’s criterion as you did. That being the minimum angular separation between two point light sources able to be resolved by an optical system. Additionally, the image produced through a lense from a point light source is…[Read more]

  • Ken Myers replied to the topic Variance in the forum General 19 years ago

    Tom,Thanks for setting the record straight. Interesting response from the others… Wow!

  • Stan,When working with Individuals data you are correct the Central Limit Theorem does not apply. Some Individuals data may not distribute normally while others may. But, consider even if when working with Average data the Subgroup Ranges on the Range Chart are not distributed normally. However, the control limits for the Average Chart are…[Read more]

  • Ken Myers replied to the topic Define the Opportunity in the forum General 19 years ago

    Anon,In your example above you asked to determine the number of opportunities observed with a stock number… But, is stock number a deliverable to your customer? How many value added areas or components do you add to stock number prior to it being delivered? Each one of areas would be considered an opportunity where you could find a defect.…[Read more]

  • Load More