DfR Solutions Reliability Designed and Delivered

Which Reliability Calculation Should you Choose? Deciding Between MTBF and a Physics of Failure Approach

Posted by Natalie Hernandez on Jun 27, 2017 12:00:42 PM

deciding-between-mtbf-and-physics-of-failure.jpgWhen carrying out reliability testing calculations, there are two popular methods that are frequently chosen: Mean Time Between Failure (MTBF) and Physics of Failure (PoF). While both approaches provide answers regarding a product’s predicted reliability, there are several key differences between the two that could have a significant impact on product development. A PoF approach offers several unique advantages over MTBF, but first, here’s a quick overview of how both methods evolved.

History of Reliability Calculations


Reliability calculations started as basic equations that describe the behavior of the physical world. They have evolved over the last 70 years but dated tactics are still commonplace in the industry because new requirements have not been set. The most widely known and used reliability prediction handbook, MIL-217, sets the standard for reliability prediction using calculated Failure Rate and MTBF numbers for the individual components, equipment and the overall system.

Starting in 1957, the Ad-Hoc Group on Reliability of Electronic Equipment (AGREE) recommended MTBF as the preferred testing method for reliability. Back then, empirical evidence based on previous tests and existing field information was widely believed to be the most reliable information available. 

In 1960, the Rome Air Development Center (RADC) recognized a need for more testing on the physical processes that are known to cause failure. Two years later, in 1962, the first ever Physics of Failure conference was held in Chicago, IL, and shortly after, the International Reliability Physics Symposium (IRPS) held their inaugural conferencewhich continues today. PoF takes reliability equations to another level based on scientific knowledge, without needing field information.

Current Problems with MTBF Calculations

MTBF calculations don’t tell the whole product story. Data gathered from physical product testing helps to shape reliability expectations, but it doesn’t offer any insights on improvements that can be made to lengthen product lifetime. MTBF calculations can help identify weak areas and provide predictions on when they’ll fail, but you’ll need to perform other tests to determine how to fix them. This means additional time as well as added costs to your product development plans. 

A Physics of Failure Solution

The scientific nature of a PoF approach allows engineers to more accurately pinpoint and predict potential failures in their products. PoF uses degradation algorithms to calculate how electrical, physical, thermal, and other mechanisms fail over time, and the science behind these algorithms is always improving. 

PoF can help identify potential failures in a product before a physical prototype is even created, thus giving you the opportunity to resolve issues in the design phase, accelerate product development and minimize costs. You can also estimate product lifetime under actual operating conditions using simulation and modeling software, which has become a key part of PoF reliability testing today. Programs like Sherlock Automated Design Analysis™ software use a PoF-based approach to simulate how your product would perform under specific operating conditions, rather than conducting lengthy, expensive physical testing or designing several prototypes.

The science and technology behind a PoF approach is getting more sophisticated each year, while MTBF continues to provide basically the same dated analysis first recommended by AGREE in 1957! To learn more about the benefits of using a PoF testing method, download our free webinar Integrating Design and Reliability: The Power of Physics of Failure.


Topics: Sherlock, Physics of Failure