If you look around you day to day you can see a lot of examples of reliability testing.
When you are at the hardware store and looking at new power tools, notice which have broken switches, triggers, handles. Many of us often just think it’s a defective one. But it’s not. It’s the weakest model of the bunch.
If all the models were put on display at similar times they have all been going through life cycle testing. Every person that walks by picks it up and pulls the trigger, flips the switches bangs it around a bit.
It’s getting its life cycle testing done right in front of you. The one with the broken features is the model with the shortest life. 
I recently was on vacation in Death Valley Grand Canyon area. I rented a Dodge Charger with a Hemi!
I also grew up watching “The Dukes of Hazard”. I was in the desert and two hours away from my destination. The map showed a dirt road that would eliminate a significant portion of the journey, that is if I kept my speed similar to the speed I would go on the highway.
So I put on my “Safety Third” T-shirt with a picture of a guy on a motorcycle holding a beer and we turned off onto the dirt road.
About every 2 miles is something called a cattle guard. It’s a rise in the road that has a metal grate that the cattle do not like to walk on.
It’s a virtual fence because cows think you can’t walk on a grate and people know you can. Which also explains our order on the food chain. It’s also good for catching some serious air.
Whistling Dixie!, Safety Third!, HALT testing! All in the name of science.
Now something totally unrelated. I am a big advocate of HALT testing. HALT testing has a very simple principle of
- Stress to failure
- Analysis of the failure
- Improve the design
- Rinse, repeat
Many development teams define their use case, test to that use case, make design improvements, and release.
A few questions about this: How was your use case defined? Was the customer use studied? Was data collected from legacy product us?
Some of our best products have simply come from development teams that successfully integrated field studies data and legacy field data as design input.
Tesla is likely to have far greater success in the development of their autonomous and other car features because all of their products stream data back from the field.
My Dodge Charger could have collected a lot of good data for Dodge that day. I sure hope it didn’t.
Note: If you are the car rental facility in Las Vegas with a super dirty Charger on the lot, this is a fictional story.
Leave a Reply