The Gist of Reliability

Abstract

Chris and Fred discuss the ‘gist’ of reliability. Fred was recently asked to be part of a panel to a number ‘hardware startups’ and was asked … what is the ‘gist’ of reliability? We try and answer that question as simply as possible in this podcast. The thing is … reliability will happen. Ignoring it doesn’t mean it won’t happen. Reliability is an outcome of your decisions and the design choices you make. The harder you work on it, the closer reliability will be to what you expect. So do you want to learn what the ‘gist’ of reliability is? Well, listen to this podcast.

Key Points

Join Chris and Fred as they discuss what the ‘gist’ of reliability is. That is … what do you tell a small startup how to ‘get’ reliability? … what do they have to do? If you ask a design engineer how fast their car goes, they give you all the statistics (top speed, time to accelerate from 0 to 60 miles per hour and so on). If you ask that same design engineer if their car is reliability … they will say yes. But how reliable is this? … and what does it need to be? So can you even ask how to ‘get’ reliability?

We talk about a number of things in this podcast, but we boil the ‘gist’ of reliability down to three things:

Focusing on the vital few reliability drivers – not the trivial thousands. Always working on trying to understand how your system ‘sucks’ is what you need to do. And the good thing is that there is typically only a small number of failure mechanisms that drive reliability.
Your organization needs to celebrate (finding) failure. Let us explain! Every failure found in a laboratory replaces, tens, hundreds, or thousands of failures experienced by customers. If design engineers don’t feel ‘safe’ in raising concerns about the design which may require time and money to do something about now … then you don’t celebrate finding failure. And because reliability is a random process, engineers also need to feel comfortable in expressing their uncertainty about something as well.
Resource your ‘celebration.’ Celebrating failure discovery is one thing. Having the budget to do something about it is another. If you don’t have the money, time, resources, expertise and support to rectify the failures you find in the laboratory or testbed … then you will very quickly go back to an organization that doesn’t celebrate finding failure – but represses it. If you want to assess where your organization stands in regard to the celebration of (finding) failure, then have a look at your design review meetings. Are they ‘dog and pony shows’ that are basically forums of shared adulation within a small group of like-minded design engineers. Design reviews need to focus exclusively on weak points and areas with minimal margin.

Reliability engineers need to work out what will fail … and when. This can be depressing! People in the design space want to make something that works. They are in the ‘success’ realm. Or so it seems. Because focusing on success ‘now’ means you aren’t focused on success ‘in the future.’ Reliability engineers (unfortunately) need to work in the ‘failure now’ realm. Designers need to understand the stresses that the product will experience, and how they will lead to failure. And this is a design decision space.

And let’s not forget -reliability helps the development process and budget. You find less issues during design, so you spend less time resolving unexpected failures in prototypes, which also saves money and reduces time to market.

Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques, to field data analysis approaches.