Interstellar 4 – Reliability

So now we have a little interstellar probe on its way to Alpha Centauri on about a century-long trip. We have a few mechanical things that have run for a century – clocks for example. We have no complex things that have ever run that long, even with continuous maintenance. We have some antique airplanes and steam trains and a few historic ships that are 100+ years old and still working, but they have been continuously maintained and even rebuilt over that period.

Engineers know that the reliability of a collection of linearly dependent parts (that is, all are in series and all have to work for the entire system to work) is the product of their individual reliability. Estimating the reliability of real-world devices is usually a bit more complex because of redundancies, but as a first approximation, just to understand the magnitude of the problem, let’s assume that we have only 100 components (transistors, resistors, capacitors, motors, solenoids, heaters, etc), all of which have to work for our probe to survive for a century and do its job when it gets to Alpha Centauri

Assume that each component has a 99.9999% chance of working perfectly for a century in the hostile cold and radiation-flooded environment of interstellar space. That is one chance in a million of failing over the whole century. That would be unbelievable reliability in today’s world, but let’s assume it anyway. Then the chances the whole assembly of 100 components will work for the whole century is 99.9999¹⁰⁰, or 9.999, just under 10%. That is, despite how few critical parts there are and how incredibly reliable they each are, the odds the whole thing will still work when it gets to Alpha Centauri are about 1 in 10.

Realistically a probe would probably have far more critical components, and the reliability of each would probably be significantly less, but even under our wildly optimistic assumption, the odds of success are pretty low.

So we have a long way to go before we could probably build a probe that would work reliably, without human maintenance, for a century in the hostile environment of interstellar space. That is our fourth problem.