Originally Posted by
Capot
But if the second failure is entirely independent of the first, it is a random event, and as such the probability of it happening is exactly the same at any point on the time line of the diversion flight, or indeed at any time, period. The length of the diversion flight makes no difference.
But Capot, is it really accurate to characterize the probabilistic assessment of a second engine failure as such failure being an utterly random event? Even for ETOPS-approved engines, does not the measurement of Mean-Time-Between-Failures (MTBF) still apply? And even if MTBF is an obsolete statistical measurement for this purpose (which I do not claim to know), still, is it not the case that the more hours an engine has been operated in total, the probability of a failure increases?
(I'm not saying your analysis is incorrect - what little knowledge of Statistics I might once have had, I had only for a social science degree, not as Mathematics.)