Re: Probability of Two Engines Failing

Date:         19 Aug 98 00:57:47 
From: (Malcolm Weir)
Organization: Little to None
References:   1
Next article
View raw article
  or MIME structure

On 06 Aug 98 11:26:15 , caused to appear as if it was

>When the big twins first started transatlantic operation this was the formula
>that the mathematicians produced to predict the first twin ditching through
>multiple engine failure.(The original date I don't recall) Needless to say
>that time passed a long time ago.
>The probability of a double power unit shutdown from independent causes is
>expressed as (2 P1 P2 T Y )where P1 is the probability of a single engine
>failure in cruise per hour,  P2 is the probibility of second engine failure, T
>is the flight time duration, and Y is the diversion time at single engine
>cruise speed to the nearest suitable aerodrome.
>P1 is conservatively assessed by the ICAO study group as the recorded
>shut-down rate fo the mature aircraft -engine combination.
>P2 the probability of second engine failure from an independent cause, will
>be higher than P1 because the second engine will be operating at higher
>power; sot the study group recommends that P2 is 2P1.

One thing that I've never been very clear on is the proportion of in-flight
engine shutdowns that are, as it were, "optional".

It is clear that there are two basic varieties of shutdown: those that are,
essentially, precautionary, and those that were necessary as a result of
(say) the engine leaving large chunks of its fan on the runway (like the
GE-90 did on BA's 777 at LHR).

By contrast, Cathay's Trent 700 problems on their A330s were precautionary,
apparently motivated by a desire to avoid turning a multi-million dollar
engine into a pile of junk (and, of course, the realization that as soon as
said engine becomes said junk, you now have the same situation as BA had...)

Which leads to the (operational, not regulatory) question: speaking purely
theoretically, what is the likelihood that in the event that an ETOPS
aircraft did have unrelated problems with both engines, how far could the
aircraft travel in a "bugger the resale value, I want to get home" mode?

(This is partly prompted by the infamous Eastern L-1011 incident, which got
home *because* the number 2 engine had been shut down, and so had some
remaining run time before it seized; numbers 1 & 3 seized virtually
simultaneously, as I recall, and well short of the runway).

In other words, from an statistical standpoint, shouldn't P2, which really
ought to be defined as the probability of a total loss of power given that
you have one engine failure, be *less* than P1, to account for the cases
where the Captain basically decides that if the company doesn't like it,
they can damn well take the cost of the engine out of his pay... if he