How does the age of the pilot affect the probability of having an accident?



In [1]:

    
%load_ext autoreload
%autoreload 2
%matplotlib inline



In [2]:

    
import sqlite3
import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
import matplotlib as mpl

from flight_safety.queries import get_flight_crew_accidents



In [3]:

    
con = sqlite3.connect('data/avall.db')



In [4]:

    
flight_crew = get_flight_crew_accidents(con)

Age distribution of pilots who have had an accident



In [5]:

    
flight_crew.crew_age.plot(kind='box');



In [6]:

    
age_bins = np.arange(14, 85, 5)
flight_crew['crew_age'].hist(bins=age_bins, ec='white', grid=False)
plt.xlabel('age')
plt.ylabel('number of accidents');

Number of licenses per age

Active FAA Pilot Certificates Held by Category and Age Group of Holder according to GAMMA databook



In [7]:

    
licenses_by_age_airline = np.array(
    [0, 572, 5199, 12003, 15507, 18337,
     23058, 25882, 24220, 16824, 10184,
     4284, 1766, 787]
)

plt.plot(age_bins[1:], licenses_by_age_airline, 'o-');

Accidents per license rate



In [8]:

    
age_group = pd.cut(flight_crew['crew_age'], age_bins)
gby_age = flight_crew['ev_id'].groupby(age_group).count()



In [9]:

    
accident_rate_age = gby_age / licenses_by_age_airline
accident_rate_age.plot.bar()
plt.ylabel('accidents / licenses')
plt.xlabel('crew age');

It seems that some people have investigated this issue before:

[...] the accident rate of airline transport rated (ATR) pilots aged 55–59 is approximately one-third of that of pilots with the same rating who are aged 20–24.

Handbook of Aviation Human Factors, Second Edition. John A. Wise,V. David Hopkin,Daniel J.