Aerospace equipment reliability is usually specified by the aircraft manufacturer when they set the requirements for a particular piece of equipment.
MTBF values of 50,000 to 100,000 hours (given a particular operating environment) are not uncommon for a box, which is ultimately what the aircraft manufacturer cares about as if a box fails, the box will be replaced.
This brings up the concept of a Line Replaceable Unit (LRU) and a Shop Replaceable Unit.
The actual operating environment drives what can be achieved.
If a box is in a relatively benign environment (maximum temperatures below 55C in normal operation), it is not difficult to achieve very high reliability values; if it is hanging off the edge of a wing (such as a wing hardpoint), things become more difficult.
When calculating reliability, it is usual to use MIL-HBK-217 to build up a complete picture based on the parts proposed for a particular PCB, and the box level reliability calculated from the individual PCB level numbers.
The failure rate of the devices in question is an input to the failure prediction, if it is available.
In terms of longevity of the kit, assuming no refresh has been made (true for older aircraft), I know that kit is still being repaired that was designed over 50 years ago. The actual box is not expected to last for 50 years of continuous operation without repair.