Defined as the probability that a device will perform its required function under specific conditions for a defined period of time, MTBF is used broadly across industries. Reliability in avionic and automotive applications hinges on predicting failure — specifically the expected time between two failures for a repairable system. This meantime between failure (MTBF) serves as a basis for a number of formulaic calculations:
Classic MTBF from a test or field reporting perspective:
Predictive perspective (summing a number of failure rates of electronic components):
Repair time prediction:
Generally speaking MTBF simplifies the math around reliability. It uses one simple number to demonstrate the number of hours a well-manufactured and screened product will be defect-free and ultimately last before wearing out. For example, an avionic or automotive component may run for 153,000 hours before failure — a large, understandable number that does away with miniscule percentages that could complicate equations or muddy results when MTBF is used for predicting reliability in the design phase, or when it’s used to extrapolate reliability based upon existing events.
MTBF is ubiquitous in many industries for reliability prediction. The Federal Aviation Administration (FAA), for example, uses MTBF to feed safety and maintenance assessments and to demonstrate compliance with business critical requirements. It’s a logical progression in this instance, with one important caveat: the accuracy of reliability prediction is completely dependent upon the assumptions that there are no defects in the components, the product is well-designed and that the failure rate is constant — and the likelihood of this combination is negligible. MTBF is a basic approach that does not account for variations which can give users a false sense of security about reliability predictions on which they need to be able to depend.
Although MTBF is not the most accurate option, it’s easier to follow the status quo than introduce a new approach. The biggest justification for sticking with MTBF in avionics is that there is no business case to eliminate it at this time because of its link to airworthiness and the FAA. This being true, it is important to know there are other options that provide better insights. Stay tuned for our next blog where we will go through the more sophisticated and preferred alternatives.
To explore better, more precise ways to understand, measure and predict reliability, leverage Physics of Failure (PoF) and tools like Sherlock Automated Design Analysis™ software for accuracy, consistency and time- and money-saving solutions that occur in the design phase. Learn more in Integrating Design and Reliability: The Power of Physics of Failure. Click the button to download your free copy.