IEEE - Aerospace and Electronic Systems - August 2023 - 8

Airbus A32x Versus Boeing 737 Safety Occurrences
Figure 10.
Proportions of accidents coded by phase of flight. Total counts:
Airbus ¼ 214, Boeing ¼ 263.
to support the finding that Boeing has more safety occurrence
than expected during landing by 9%. It should be
noted that Boeing had 24 more RE events than expected
and 30 more occurrences at landing than expected. As
such, it is possible the increased number ofBoeing landing
events is likely due to the aerodromes where the aircraft
are operated, as previously noted.
Figure 7 shows the proportions of all safety occurrences
coded in terms of the injury level. The chi-squared
test is not statistically significant (a ¼ 853, b ¼ 985,
x2 ¼ 2.83, p ¼ 0.42). Looking at fatalities (shown in
Figure 8), relative to the total number of safety occurrences,
the number of observed fatalities in Boeing occurrences
is 28% more than expected, while for Airbus they
are 26% less than expected. This result is statistically significant
(a ¼ 773, b ¼ 1251, x2 ¼ 144,p < 0.001).
ACCIDENTS
Figure 9 shows the breakdown of accidents coded in terms
of ICAO occurrence categories. When looking at only
accidents rather than all safety occurrences, the observed
difference in distribution is not statistically significant
(a ¼ 182, b ¼ 283, x2 ¼ 10.8, p ¼ 0.29). The proportions
for abnormal runway contact and turbulence are higher for
Figure 12.
Number of fatalities for Airbus and Boeing accidents (expected
based on the number of accidents).
Airbus (a) while the system component failure nonpowerplant
and runway excursions are higher for Boeing (b);
although, not statistically significantly different.
Figure 10 shows the proportions of accidents coded
in terms of phase of flight. The chi-squared test shows
the difference in proportions between Airbus (a) and
Boeing (b) are not statistically significantly different
(a ¼ 214, b ¼ 263, x2 ¼ 7.2, p ¼ 0.41). The follow up
test comparing landing to all other phases of flight,
between Airbus and Boeing was also not statistically significant
(x2 ¼ 0.05, p ¼ 0.82).
Figure 11 shows the proportions of all the safety
occurrences coded in terms of the injury level. The
chi-squared test is not statistically significant (a ¼ 853,
b ¼ 985, x2 ¼ 2.83, p ¼ 0.42). In terms of fatalities relative
to the number of accidents (shown in Figure 12), the
number of observed fatalities in Boeing accidents is 18%
more than expected, while for Airbus they are 20%
less than expected. This result is statistically significant
(a ¼ 771, b ¼ 1249, x2 ¼ 70,p < 0.001).
FATAL ACCIDENTS
Finally, we can consider just fatal accidents. Figure 13
shows the distribution of the Airbus (a) and Boeing (b)
Figure 11.
Proportions of accidents coded by injury level. Total counts: Airbus
¼ 853, Boeing ¼ 985.
8
Figure 13.
Proportions of fatal accidents coded by occurrence category.
Total counts: Airbus ¼ 11, Boeing ¼ 25.
IEEE A&E SYSTEMS MAGAZINE
AUGUST 2023
IEEE - Aerospace and Electronic Systems - August 2023

Table of Contents for the Digital Edition of IEEE - Aerospace and Electronic Systems - August 2023
