Your Reliability Engineering Professional Development Site
All models are wrong, some are useful. ~ George E. P. Box
If you know me, you know I do not like MTBF. Trying to predict MTBF, which I consider a worthless metric, is folly.
So, why the article on predicting MTBF?
Predicting MTBF or creating an estimate is often requested by your customer or organization. You are being specifically asked for MTBF for a new product.
You have to come up with something.
Recently received a request for my opinion concerning the calculation of system availability using the classic formula
$$ \displaystyle \large A=\frac{MTBF}{MTBF+MTTR}$$
The work is to create a set of goals for various suppliers and contractors to achieve. The calculation values derive from vendor data sheets and available information concerning MTBF and MTTR. The project is in the design phase; thus, they do not have working systems available to measure actual availability.
How would you go about improving on this approach? [Read more…]
What good is your service if your livelihood consists of providing MTBF upon request?
Sure, you earn some money, yet did the customer receive value in the transaction? As you know, or should know, MTBF is so commonly misunderstood that the customer is likely to be confused about what they want, reliability or MTBF. Providing them with MTBF does not answer their question.
Worse, the customer thinks they got something of value and blithely heads off with rather meaningless information.
My contention is to provide MTBF because the customer’s request is wrong. We know better. Those performing predictions, doing data analysis, and other reliability engineering work know that MTBF is a faulty and rather meaningless metric often confused with reliability, R(t). (probability of success over a duration). [Read more…]
Let’s say we have a set of numbers, {2.3, 4.2, 7.1, 7.6, 8.2, 8.4, 8.7, 8.9, 9.0, 9.1} and that is all we have at the moment.
How many ways could you analyze this set of numbers? We could plot it a few different ways, from a dot plot, stem-and-leaf plot, histogram, probability density plot, and probably a few other ways as well. We could calculate a few statistics about the dataset, such as mean, median, standard deviation, skewness, kurtosis, and so on. [Read more…]
As regular readers know, MTBF by itself is misleading. It can also be deceptive when representing actual data. Just because you have a high MTBF value doesn’t mean it is reliable.
In a previous article, 10 Reasons to Avoid MTBF, I mentioned that it is possible to have a relatively high MTBF value when the actual reliability is low. Ashley sent me the following note:
Hi Fred, i love reading your articles they are very informative. I have a question about something you said in a comment which i am hoping you will be able to clarify for me. You said products with higher MTBF can actually be less reliable than products with a lower MTBF
I have tried to find information on how this is possible online, and tried to do the maths myself to make this happen but i have to admit i am struggling.
No worries, Ashley, let’s work out an example to illustrate what I meant. [Read more…]
This was a follow-up question in a recent discussion with Alaa concerning using a metric other than MTBF.
The term ‘Weibull’ in some ways has become a synonym for reliability. Weibull analysis = life data (or reliability) analysis. The Weibull distribution has the capability to describe a changing failure rate, which is lacking when using just MTBF. Yet, is it suitable to use ‘Weibull’ as a metric? [Read more…]
The lognormal distribution has two parameters, μ and σ. These are not the same as mean and standard deviation, which is the subject of another post, yet they describe the distribution, including the reliability function.
$$ \displaystyle R(t)=1-\Phi \left( \frac{\ln (t)-\mu }{\sigma } \right)$$
Where Φ is the standard normal cumulative distribution function, and t is time. [Read more…]
The Linkedin NoMTBF group is growing and while not very active does have an occasional interesting discussion. Join the discussion and maybe relate how you have raised awareness around the proper use of MTBF.
https://www.linkedin.com/groups/1857182/
[Read more…]
It took us years to move from top-down to people-centric management, but the journey was worth it Peter Drucker once said, “Culture eats strategy for breakfast.” Throughout my career, I have spent a lot of time in the strategy room. But no matter what we decided there, until we got the organizational culture right, we were unsuccessful in executing our goals.
[Read more…]
The following is a recent discussion on the sister Linkedin NoMTBF Group. It was and may continue to be a great discussion. Please take a look and comment on where you stand. Do you have some form of the Arrhenius reaction rate equation in your reliability engineering work? Join the discussion here with a comment or on the Linkedin group conversion.
Fred
[Read more…]
An excellent short white paper by Craig Hillman that is worth reading. It underscores why I claim HALT is the second worst 4 letter acronym in our profession.
[Read more…]
Hello, can anyone advise me how to calculate temperature acceleration factor for a complex system including cards, RF elements, cables, motors and moving parts? Is the Arrhenius model valid for such systems, or there are more precise models? Thank you!
[Read more…]Fred i was asked this question and wanted to know what your thoughts were on this. R and D asked me what was the criteria to decide if to test at a component level or at a system level , my answer was that it should depend on what is the reliability and confidence level of the component
your thoughts?
thanks
sd
Just a short post to point to a paper on the accuracy of part count prediction techniques. A few years ago, I recalled seeing a paper that studied the difference between various parts count methods and actual results.
[Read more…]
With the kind permission of Wayne Nelson and Robert Abernathy, we are posting an article on the analysis of repair data. As you may know, the assumptions made when using simple time-to-failure analysis of repairable systems may provide misleading results. Using the analysis method outlined by Wayne is one way to avoid those costly mistakes.
[Read more…]