Re: MD-80: Lawn Dart or Efficient Design?

From:         Pete Mellor <pm@csr.city.ac.uk>
Date:         23 Aug 96 13:45:05 
References:   1 2 3 4 5
Followups:    1
Next article
View raw article
  or MIME structure

>>Apologies if I missed any.
>
> Yeah, you missed some.  It's computed on deaths or accidents per RPM,
> revenue passenger mile, or something.  There are over 1,000 MD-80
> racking up RPM's right now.
>
> The listing I saw was at a Boeing presentation, where they were
> discussing USAir 427 and the 737's fourth place.

In any kind of service reliability or safety assessment, one is
always faced with the choice of which incidents to record over
what measure of operation.

With air accidents, the two types of occurrence which are most
often cited are:-
a) deaths, and
b) hull losses.

Possible measures of operation are:-
i)   revenue passenger miles (RPM),
ii)  revenue passenger hours,
ii)  revenue flight hours, and
iii) revenue flight cycles (take-off plus (hopefully!) landing).

There are no prizes for spotting that one obtains completely
different statistics depending on the choice of measure. Not only
that, but the ranking of two models of aircraft with respect to
"safety" will often vary.

If you want to make air transport the safest ever mode of travel,
quote "hull losses per RPM". If you want a slightly less optimistic
(more realistic?) picture, quote "deaths per flight cycle".

I discount another measure of "safety" that I have seen ("hull losses
per number of examples in service") as being totally misleading, and
not a well-defined measure of anything.

Peter Mellor, Centre for Software Reliability, City University, Northampton
Square, London EC1V 0HB, UK. Tel: +44 (171) 477-8422, Fax: +44 (171) 477-8585
E-mail: p.mellor@csr.city.ac.uk
-----------------------------------------------------------------------------