Bayesian Statistics and Sample Sizes

Abstract

Andre and Fred discussing the benefits of using what you know as you approach setting sample sizes using Bayesian statistics.

Key Points

Join Andre and Fred as they discuss a few papers at the recent ARDC conference that included topics based on Bayesian statistics. We quickly spent time talking about the role of Bayesian statistics and test sample samples.

Topics include:

A basic description of Bayesian statistics as your current knowledge change the probability of an event occurring.
A brief description of Weibayes and what it is and is not.
A bit of a debate on the differences between frequentist and Bayesian approaches.

Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques, to field data analysis approaches.

SOR 232 Bayesian Statistics and Sample SizesAndre Kleyner

Download Audio RSS

Show Notes

About Andre Kleyner

Andre Kleyner has 30 years of engineering, research, consulting, and managerial experience specializing in reliability of electronic and mechanical systems designed to operate in severe environments. He received the doctorate in Mechanical Engineering from University of Maryland, and Master of Business Administration from Ball State University. Dr. Kleyner is Global Reliability Engineering Leader with Delphi Electronics & Safety and an adjunct professor at Purdue University. He is an ASQ Fellow, a CRE, CQE, and Six Sigma Black Belt. He also holds several US and foreign patents and authored multiple professional publications including three books on the topics of reliability, statistics, warranty management, and lifecycle cost analysis. Andre Kleyner is also the editor of the Wiley book Series in Quality and Reliability Engineering published by John Wiley & Sons.

Comments

John Nelson says
July 25, 2017 at 5:39 AM
Andre, can you provide me with the Title of your book. Where I work they won’t let me have nice tools like WeiBayes because I might hurt myself e.g. engineer with scissors..lol
Regards,
John
Reply
- Andre Kleyner says
  July 25, 2017 at 7:26 AM
  John,
  The book is ‘Practical Reliability Engineering’ Ed.5 by O’Connor and Kleyner and it has some basic info on Bayesian statistics and reliability applications. However, if you are looking for the specific application of Bayesian statistics to reduce the sample size, I can send you couple of articles we wrote on the subject. Let me know your email address or we can ask Fred to upload the papers to the Accendo website.
  Reply
John Nelson says
July 25, 2017 at 5:57 AM
Fred, I’ve got a nuts and bolts book on Bayesian. Send me an e-mail and I’ll provide the title and some excel examples. Its also a history on Lord Bayes
Regards,
John
Reply
- Fred Schenkelberg says
  July 25, 2017 at 6:42 AM
  Hi John, Great really appreciate the lead to a good reference. I’m at fms@accendoreliability.com cheers, Fred
  Reply
David Coit says
July 28, 2017 at 6:33 PM
I know of two textbooks on Bayesian Reliability, and I recommend both, although they may not offer the quick practical explanations some people want. For the last 30 years, one of the most intelligent and thoughtful advocates of Bayesian reliability is Nozer Singpurwalla and he has a book on the subject. Then there is a very old textbook by Harry Martz, although it is likely to be out of print now.
Reply
David Coit says
July 28, 2017 at 6:38 PM
I have always been somewhat sceptical of Bayesian statistics for reliability, because it can be used to manipulate the data to get a desired answer. However, Bill Meeker at a recent RAMS tutorial gave a very convincing and intelligent explanation and defense of Bayesian reliability. Meeker is not known as Bayesian, so if he is a believer … I will be too
Reply
- Mark Powell says
  July 31, 2017 at 1:07 PM
  David,
  You might want to take a look at this blog post Fred had me do a few years back on noMTBF.com. http://nomtbf.com/2012/07/whats-all-the-fuss-about-bayesian-reliability-analysis-2/
  You get the full derivation there that is not in any text anywhere.
  Mark Powell
  Reply
DonMacArthur7 says
October 17, 2017 at 11:20 AM
would like to hear/learn more about bayesian stats for reliability engineering. thanks
Reply