Featured Product
This Week in Quality Digest Live
Operations Features
Laurel Thomas
Trust your co-worker, but remember its video camera is running
Ryan E. Day
FARO products help Composiflex deliver industry-leading service
Susan Whitehead
Cut training time in half while reducing scrap and rework
DP Technology
Seamlessly integrates with milling center, enabling company to make full use of five-axis machining capabilities
Isaac Maw
A conversation with former Apple engineer Anna Shedletsky

More Features

Operations News
A guide for practitioners and managers
Examine nine ways to help understand context, diffuse politics, heal relationships, and find the right work at the right time
Provides eight operating modes and five alarms
Making lean Six Sigma easier and adaptable to current workplaces
Deep Reality Viewer creates stereo, high-definition 3D images without using a monitor
April 25, 2019 workshop focused on hoshin kanri and critical leadership skills related to strategy deployment and A3 thinking
Makes it faster and easier to find and return tools to their proper places
Process concerns technology feasibility, commercial potential, and transition to marketplace
Identifying the 252 needs for workforce development to meet our future is a complex, wicked, and urgent problem

More News

Fred Schenkelberg


How Did Reliability Become Confused With MTBF?

When asking for reliability information, do you ask for MTBF?

Published: Tuesday, August 1, 2017 - 12:02

Our customers, suppliers, and peers seem to confuse reliability information with mean time between failure (MTBF). Why is that?

Is it a convenient shorthand? Maybe I’m the one confused, maybe those asking or expecting MTBF really want to use an inverse of a failure rate. Maybe they aren’t interested in reliability.

MTBF is in military standards. It is in textbooks and journals and component data sheets. MTBF is prevalent.

If one wants to use an inverse simple average to represent the information desired, maybe I have been asking for the wrong information. Given the number of references and formulas using MTBF, from availability to spares stocking, maybe people ask for MTBF because it is necessary for all these other uses.

What I don’t get is why

When someone asks me for the MTBF, I ask them what they want to know.

The standard answer is they want to know the chance an item will survive over some duration. Or they say they want to know the reliability. They ask for MTBF expecting to learn something about an item’s reliability.

Why not just ask for reliability, R(t), the reliability function. The time to failure distribution describing the pattern of failure over time. When I provide R(t), sometimes the next question is, “How do I covert this to MTBF?”

Why would anyone want to do that?

References and habit

Nearly any reference on reliability includes some discussion about MTBF, often an extensive part of a manual, book, or standard that only uses examples with mean time to failure (MTTF) or MTBF. This may be why so many ask for MTBF exclusively.
• Reliability apportionment tools often request MTBF values for subsystems.
• Reliability prediction methods request the MTBF of components (which we then convert to failure rate to add, then convert the result to MTBF).
• Reliability requirements are often written in term of MTBF, to which we compare our estimates.
• Reliability testing has formula after formula to determine sample size and test time, two essential elements for planning.

Those that ask for MTBF may be reading only the MTBF related sections of the book, manuals, and standards. For them, the world of reliability engineering revolves around MTBF. So, limiting.

“Everyone requests and uses MTBF.” It seems to be habit forming. If you are taught to “do” reliability by asking for MTBF and estimating MTBF, then MTBF is what you “do.” It is the way we learn the craft and then teach others. We also seem to stop learning, to stop checking assumptions, to stop noticing that the actual results are nothing like the promised MTBF.

MTBF is not reliability

Reliability is the probability an item will function over a duration within a specified environment. This is in most if not all reliability related textbooks, glossaries, and standards. Right before the introduction of MTBF.

MTBF is the mean time before (some use between) failure. A common method to estimate MTBF is to calculate the inverse of the average failure rate. Total time divided by the number of failures.

Often when we ask someone about the reliability information of an item, say a flat-screen TV, the response is a duration. “The TV has a five-year life.” That is only a duration and assumes we include a low chance of failure over the five years.

If we say the TV has a five-year MTBF, that is not the same as saying the TV is very likely to survive five years. I does mean that on average the TV will fail sometime before or after five years, and given no other information and assuming a constant hazard rate, the chance of failure is roughly 18 percent for any given year.

Part of the confusion between MTBF and reliability is that we’re are often interested in how long an item will survive, a duration, in units of time or cycles. Because MTBF is given in units of time (time per failure, actually), many confuse MTBF stated in units of time as a duration. The duration over which we expect items to survive.

Duration alone is not reliability; it is only one element of reliability. MTBF is actually a probability or the chance of failure per unit of time. If the MTBF is stated as 50,000 hours, that literally means there is a 1 in 50,000 chance of failure every hour. It is not a duration. The probability or failure, like the probability of success, is also not reliability. It is just one element of reliability.

So, just to be clear: MTBF is not reliability.


I wonder if the confusion between MTBF and reliability information will ever really go away. I have to believe it will; thus, I continue to maintain the NoMTBF site. With your help, we can clear up this confusion and get on with the important task of creating reliable products that meet our customers actual reliability needs and not their requested MTBF.


About The Author

Fred Schenkelberg’s picture

Fred Schenkelberg

Fred Schenkelberg is an experienced reliability engineering and management consultant with his firm FMS Reliability. His passion is working with teams to create cost-effective reliability programs that solve problems, create durable and reliable products, increase customer satisfaction, and reduce warranty costs. Schenkelberg is developing the site Accendo Reliability, which provides you access to materials that focus on improving your ability to be an effective and influential reliability professional.


Aren't Reliability and MTBF related?

Hi Fred,

Aren't reliability and MTBF related?

This is what I understand. Please correct me if I am wrong.


*] Reliability is a probability, therefore, it must be between 0 and 1 or (0% and 100%). It is unitless.

*] Mean-Time-Between-Failure is not a probability. It has the unit of time, typically hours. It is a measure of life.


Assuming that my life measurements are distributed according to the exponential distribution, then once I have calculated the mean value for life (mu) or MTBF, I also have the standard deviation (sigma) for the distribution. With mu and sigma I could calculate k*sigma bounds to capture various proportion of measurements that make up the distribution.

The proportion of measurements, i.e. area under the curve, above some lower specification value is the reliability. No?

If I have MTBF, I could estimate the proportion of life measurements above some minimum life value and thus estimate the reliability. Couldn't I?

Regards,Shrikant Kalegaonkar (twitter: @shrikale, blog: https://shrikale.wordpress.com)


I don't see the word efficiency anywhere or am I misunderstanding something