Let’s think of this as a crowdsourced project. The first version of this book is a compilation of NoMTBF.com articles. It lays out why we do not want to use MTBF and what to do instead (to some extent).
With your input on success stories, how to make progress using better metrics, and input of examples, stories, case studies, etc., the next version of the book will be much better and much more practical.
What Should Be in the Book?
The Reliability Metric book was published in 2015, and since then, we have sold a few books every month. While not a NYT Best Seller, it does help readers make a difference.
The table of contents of the current book is:
Forward 1
Introduction 3
Eradicate the MTBF goal 13
Issues Surrounding MTBF 15
Solutions 37
Call to action 63
Reliability: The best alternative 89
Last Words 115
I would like to add a chapter on success stories, breakthroughs, and progress in moving away from MTBF use. I’ve heard about changes in policy, setting vendor expectations, requiring reading of the NoMTBF blog in college courses, etc.
I’d like to add your stories with or without credit/company mention (your choice)
I’d also like to add a few examples of how decisions can go terribly wrong when using MTBF. If you have such examples, please send on over to me.
What else should be included? What should we remove?
Send Your Examples and Stories
This is an action item for you. Let’s create a practical, useful, and impactful update to this book.
The deadline for input, at least for this round, is March 31st, 2025.
Send your comments, suggestions, ideas, and stories to me, Fred, at fms@nomtbf.com
or just leave a comment below
Fred: As much as I don’t want to use MTBF, what are options to initiate a discussion to configure large scale super computers…. I would love to hear your take on this
Hi Nathan,
I’m a bit confused by the question. MTBF really should have nothing to do with the configuration of large-scale super computers. If availability is a goal, then using reliability and redundancy (reliability block diagrams for models) than MTBF has no role, use reliability with the appropriate time to failure distributions. Hard drives, memory, capacitors, etc all do not follow the exponential distribution for time to failure patterns…thus using MTBF obscures the very information you need to make design decisions.
Cheers,
Fred
ASQ monograph “Credible Reliability Prediction,” was in fact a success story of using field reliability observations to predict field reliability of a new product. Despite designers’ and engineers’ claims about new and improved, new products resemble old, their designs use some old parts, their production processes are probably similar, and their customers are probably similar. Used Cox’ proportional hazards model.