Originally Posted by
tdracer
Or to put it a little differently - the 767 and 747-400 were designed with roughly the same level of technology and safety, but the 767 (and 757) are statistically safer than the 747-400 (by roughly a factor of 2).
Apologies, I understand the wording may have been simplifed. It seem to depend on what is meant by
(shown to be) _
statistically safer_ in this context. Is it that
- the number of IFSD over (say 10 k) departures was less on the 75&6 compared to the 747?
There are other metrics to view an IFSD.
- the number of events where 50% thrust loss resulted from an IFSD / (10k departures) (menitoned already)
and
- the number of events of IFSD where the resulting configuration was 1 (single) engine away from total loss of propulsion / (10k departures)
Okay, now the last one is clearly tailored to drive a certain point which had been already disproven by the decades of catastrophy-free twin jet operation. It's recognizable spin doctoring. Saying "the 767/757 ended up statiscically 2x safer than the 747 of equivalent technical design" - - that is equally awful. I remain convinced the bolded metric is a reasonable measure. It takes more bullets to hit more hares, and more luck to take down 4 in sequence than 2. P(n) = p^n.
The disagreement extends against arguments proposed at post #16, all of them to some extent are wagging the dog. To paraphrase, the classical mistake of supporting a good cause with weak examples and hollow logic.
For instance, #16 mentions finite probability of engine fire turning catastrophic (and the suggestions 4 installed increases exposure to the risk). Over the decades while ETOPS / LROPS existed, there was not any. See the lucid post number #21 saying
_that (other discussed) failure was not supposed to happen_ and that's agreed. Same needs to apply to an uncontained rotor failure used to create a (not so) convincing probability calculation at # 16.
Nicely demonstrated with the AUA in-flight reverser deployment. Freak (uncontained by design at the time - absolutely love your insights!) failure mode where a fault on one engine takes down the whole ship. With that it certainly applies - having 3, 4, 8 increases the probabilty. That is mathematically valid but no longer relevant as it wouldn not happen again.
A conspiracy theorist would say this all reaks of groupthink and cherrypicking / modelling the probability equations for a pre-determined result. Perhaps in an era when launch of the T7 in twin configuration needed quasi-scientific support? I'm glad they did!
(personally I fly a twin and used to be rated on an ETOPS-180 one too. The marketing campaign 4engines 4the long-haul from a certain manufacturer to battle the T7 seizing her rightful place was pure hypocrisy)
Arcing over towards the dramatic now: in case of a complex failure, is a 2 man-crew
safer than a 3 man one because there is less humans to make mess of it? I know, I know ...engines don't correct each other.
and a three-captain cockpit is a recipe for a disaster!
Ooops, not so much of light relief.
Stay healthy everyone.