Re: Air Safety (was Re: A3xx vs B747-600)

Date:         13 Dec 96 04:26:03 
From:         Todd Curtis <>
Organization: AirSafe Services
References:   1 2
Followups:    1
Next article
View raw article
  or MIME structure

Dear Karl,

Thanks to a friend of mine, I was made aware of an ongoing debate about
my site.  I am the sole author of The Air Safety Home Page, and I am
willing and able to respond to any question about it.  When I started the
site, I did so with the intent of taking a passenger eye view of risk.
When I am a passenger, I care most about the chance that I will be
seriously injured or killed while I am a passenger.  From my perspective
as a passenger (usually from the point of view of the cheapest seat I can
buy) I don't care about whether the aircraft is damaged or if the crew is
hurt or killed.  Airplanes are machines that can be fixed or replaced and
any risk the crew faces is an occupational hazard.  My expectation is to
arrive at my destination and to go about my business.  Also from my
perspective, dead is dead.  It does not matter if my death is due to crew
error, acts of nature, or some politically motivated act of violence; I
would desire to avoid that risk when I fly.

Given that fundamental perspective, it seemed to me that a logical first
step would be to find the data that would give me or any other airline
passenger an historical view of passenger fatal events on the aircraft
models that are most commonly used around the world.  I only have a few
rules concerning what is or is not included in my listings.  If it is a
scheduled or non-scheduled flight on which the general public could book
passage, then it makes the first cut.  If at least one passenger dies
during the flight and the fact that the person was on that flight was
related to their death, then it makes the second cut.  Certainly there
are some judgement calls that I make and those are certainly ripe for
debate, but on the whole I have tried to be as consistent as possible
when including fatal events.

The fatal event rates that I give are certainly measures of risk. Risk is
simply defined as being a product of a consequence and a probability.  On
my site, the consequence is at least one passenger dies as a result of
being on that flight.  The probability is based on the estimated number
of passenger flights that aircraft model has made.

Unlike risk, safety does not lend itself to objective measures.  In
general, an activity is considered safe if the risk is deemed acceptable.
 Risk is acceptable on a personal or a social level depending on several
factors, so I avoid labeling my fatal event rates as some sort of safety
measurement.  If one wishes to define "safety" as being equivalent to
risk, so be it.  I prefer in my site to gather up event data based on
risk definitions that I consider relevant, and then letting the reader
decide for themselves what it means in terms of safety.

I welcome debate on what I have said either in this forum or directed to

- Todd Curtis