IEEE Spectrum May, 2015 - 28

NUMBERS DON'T LIE_BY VACLAV SMIL

OPINION

KEEP ON FLYING
700

Fatalities on scheduled
commercial flights

655

Accidents on scheduled
commercial flights

626

600

Fatal accidents on scheduled
commercial flights
500

Source: 2014 Safety Report,
International Civil Aviation
Organization

400

372

388

300

200

173
100

102

104

118

2009

2010

2011

2012

2013

YOU MIGHT THINK 2014 WAS A BAD YEAR FOR FLYING.
There were four highly publicized accidents-the still-mysterious
disappearance of Malaysia Airlines Flight 370 in March, the
shooting down of Malaysia Airlines Flight 17 over Ukraine
in July, AirAsia Flight QZ8501 falling into the Java Sea in December,
and finally, in July, the Air Algérie Flight 5017 crash in Mali, for a total
of 815 dead. * But according to Ascend, the consulting branch of Flightglobal
that monitors aircraft accidents, 2014 in fact had the best accident rate in
history: one fatality per 2.38 million flights, compared to the previous best
of one per 2.37 million in 2012. True, Ascend did not count the downing
of MH Flight 17, which was an act of war, not an accident. * In any case,
it's better to personalize the problem by putting it in terms of the risk per
passenger per hour of flight. The necessary data are in the annual safety
report by the International Civil Aviation Organization, which covers large
jetliners as well as smaller commuter planes. * In 2013, 32.1 million domestic and international flights carried 3.1 billion people, logged 5.8 trillion
passenger-kilometers, and experienced 90 accidents, causing 173 fatalities. With the mean flight time at about 2.2 hours, this implies roughly
6.8 billion of passenger-hours or 2.5 × 10-8 fatalities per person per hour
in the air. For large jetliners-dominated by Airbuses and Boeings and
regional jets made by Canada's Bombardier and Brazil's Embraer-the risk
28

MAY 2015

NORTH AMERICAN

SPECTRUM.IEEE.ORG

that year was just 1 × 10-8. In 2014, large
jetliner accidents (excluding the MH 17)
would have pushed the latter rate to
about 8 × 10-8, but the mean for the past
decade remains at historic lows.
In short, flying has never been safer.
The obvious measuring stick is
general mortality, the annual death
rate per 1,000 people. In affl uent nations that rate now ranges between 7
and 11; I will use 9 as the mean. Because the year has 8,760 hours, this average mortality prorates to 0.000001
or 1 × 10-6 deaths per person per hour
of living. This means that the average
additional chance of dying while flying
is just 1/100th of the risk of simply being alive. Smoking risks are 100 times
as high; ditto for driving in a car.
Obviously, age-specific mortality
for older people is much higher. For
individuals of my age group-I'm 71-
it is about 25 per 1,000 or 3 × 10 -6 per
hour (that means that of a million of
us, three will die every hour). In 2014,
I flew more than 100,000 kilometers,
spending more than 100 hours aloft in
large jets belonging to four major airlines whose last fatal accidents were,
respectively, in 1983, 1993, 1997, and
2000. In every hour aloft the probability of my demise wasn't even 1 percent
higher than it would have been had I
stayed on the ground.
Of course, I've had white-knuckle
moments. The most recent one was
in October 2014, when my Air Canada
Boeing 767 headed into the turbulent
fringes of a megatyphoon that was
crossing over Japan.
But I never forget that quiet hospital
rooms are what should really be avoided:
The latest assessment shows that preventable medical errors push the risk
of premature death to at least 5 × 10 -5
and perhaps as high as 1 × 10-4 per every
hour of hospital stay, easily more than
doubling or even tripling the probability
of dying for those of my age group who
cannot stay out of a hospital. So keep on
flying, and avoid hospitals! ■

http://SPECTRUM.IEEE.ORG

Table of Contents for the Digital Edition of IEEE Spectrum May, 2015