As an engineer I am regularly faced with the challenge of defining how much effort needs to go into an element of a project. When NASA decided to start using the pen developed by Paul Fisher of the Fisher Pen Company it was not because they had spare cash lying around and wanted to show off or had a tiff with their standard supplier. It is because they worked out the consequences of a dud pen or pencil in orbit and decided it was not a risk they could tolerate. What you need depends a lot on what you do.
Sitting at my desk pen failure results in a binned biro and a gentle roll over to the cupboard where I get the next one out and keep going. In space pen failure could equal mission failure. If an astronaut needs to make a note, mark a point or do some quick hand calculations you can bet that they need that tool to work to provide a quick solution to whatever instigated the need. Having to scramble around inside the space craft for the next pen may result in wasted time they can’t afford. Subsequently the consequences of failure justify the very high level of reliability.
In our every day purchasing or design decisions we have to make sure we know what this thing is going to be used for before we commit. The accuracy of this assessment is the key to finding the sweet spot between money well spent on high reliability and a false economy from a poor understanding of the end user. Spending time finding out the end user’s needs has a similar level of importance to design that reconnaissance does to planning military operations. As all good military leaders know, time in reconnaissance is seldom wasted. The designer’s equivalent would be that time in testing is certainly time well spent.
What allowances have been made for testing on your current project/s?