Causal risk models of air transport - NLR-ATSI

caused by overrotation and subsequent wing stall. These problems were solved by a
modified wing leading edge. Regulation was also changed as a result of these accidents,
requiring demonstration of a minimum unstick speed with maximum angle of attack.
[EASA 2003a]. But then in 1954 a Comet disintegrated while climbing to its cruising
altitude, and the British Overseas Airways Corporation BOAC grounded its fleet of jet
aircraft. However, no defects were found and the aircraft were put back into service.
Sixteen days later, another Comet broke-up in mid-air, and the aircraft’s Certificate of
Airworthiness was suspended. After extensive testing, it was discovered that the accidents
had been caused by fatigue cracking of the fuselage skin. This was a new phenomenon, first
encountered by the Comet because its high cruise altitude resulted in relatively large stress
cycles to the fuselage (pressure cabin) skin. Production of the Comet was halted. A
modified version, the Comet 4, flew in 1958 but by that time competing aircraft were on the
market and the Comet 4 was not a commercial success.
As a result of these developments, air travel today is astonishingly safe. For air carriers in
Europe or the US, the fatal accident rate per million flights is approximately 0.2 [CAA-NL
2007]. If a person were to take a flight each day, he would, on average, experience a fatal
accident after 13,700 years. Even then, the likelihood of being killed in that accident is less
than the probability of survival; an analysis of fatal accidents by US domestic airlines in
1987-1996 showed that, on average, a fatal accident kills 44% of the occupants of the
aircraft [Barnett & Wang 2000]. A fatal accident is defined as an event in which at least one
of the occupants is killed, so even in the case of a fatal accident the majority of the
occupants may survive. Does this mean that a sufficient level of safety has been reached?
Certainly not from the viewpoint of the regulator. Aircraft accidents cost lives and are a
cause of emotional trauma to victims’ relatives and friends, to people working within the
industry, to communities. Crashes consume the attention of the media in a unique way; any
fatal airline crash will always be a headline story. As a result, it is feared, the general public
may loose confidence in the air transport system. The extent to which ‘air crashes’ are
reported in the media has always been of concern to the industry and has been seen as a
possible limiting factor to growth for at least 60 years. In 1943, a confidential ‘report’
produced by the Curtis Wright Corporation on behalf of the US Government stated that if
air safety did not improve from the level achieved during the 1930s, the rapid expansion in
air travel expected after the war would result in an unacceptable level of air crashes and
would limit growth. These views have been repeated many times over the years, for
instance by the US Secretary of Transport at the Aviation Safety Conference in Washington
D.C. in 1995 [FAA 1995b] and are still regularly voiced today. Perceived public demand
for safe air travel is counterbalanced by a demand from within the industry to remain
profitable. As early as 1959, Walter Tye (UK ARB – Air Registration Board) suggested
that an optimum balance may have been achieved between airline flight safety and costs.
He argued that any further marked increase in safety by (then) existing methods would cost
far more than passengers or the industry would be willing or able to pay (as quoted in
[Bland 1962]). This view was repeated, at the 1961 FSF Seminar, by M.G. Beard,
American Airlines, who said he believed that the major airlines had already reached a
position of ‘diminishing returns’ in that they had (in 1961) reached a very high degree of
safety and further improvements would cost more and more. However, the costs associated
with an accident can also be enormous, and while much of this is absorbed by insurance,
the effect of an accident on the airline’s stock price and ticket sales is uninsured. Some
airlines have actually gone bankrupt after an accident, examples are Birgenair 24 and Helios
24
On February 6, 1996 a Birgenair Boeing 757 crashed shortly after take-off from Puerto
Plata in the Dominican Republic, killing all 189 occupants. Birgenair was a Turkish holiday
39