R
radarredux
Guest
> <i><font color="yellow">Yes, the 40% probability is faulty.</font>/i><br /><br />No, the statistics are quite correct [1].<br /><br />The probability of failure is<br /> Pr(failure) = 1.0 - Pr(success)<br /><br />The probability of success for one flight is 0.982; the probability of success for two flights is 0.982*0.982; three flights is 0.982*0.982*0.982; and in general, for 'X' consecutive flights:<br /> Pr(success) = 0.982^X<br /><br />Thus, the probability for failure is<br /> Pr(failure) = 1 - 0.982^28 = 0.399 (or 40%)<br /><br /><br />[1] The primary problem with the analysis is we really don't know what the probability of success is. 0.982 is just an estimate based on 2 failures for 114 flights. But from a statistics point of view (1) the numbers are way too low to make a reliable estimate, and (2) the conditions are always changing (e.g., new procedures, changes to the system, aging fleet, etc.).<br /><br />But, if 0.982 success rate is near accurate, then 40% probability of losing another shuttle before completing the ISS (and thus probably not completing the ISS) is correct. Also a bit sobering.</i>