The NYC Flights Dataset contains data for all flights that departed from New York City area airports (JFK, LGA, and EWR) in 2013. It is available as a package in R Studio.
#install.packages("nycflights13")
library(nycflights13)
library(dplyr)
Attaching package: 'dplyr'The following objects are masked from 'package:stats':
filter, lagThe following objects are masked from 'package:base':
intersect, setdiff, setequal, uniondata(flights)Checking how many airlines are represented in the data.
length(unique(flights$carrier))[1] 16?airlinesstarting httpd help server ... doneCreating a dataframe of just the carrier, arrival delay, and departure delay.
flights_delay <- flights[, c("carrier", "arr_delay", "dep_delay")]
flights_delay# A tibble: 336,776 × 3
carrier arr_delay dep_delay
<chr> <dbl> <dbl>
1 UA 11 2
2 UA 20 4
3 AA 33 2
4 B6 -18 -1
5 DL -25 -6
6 UA 12 -4
7 B6 19 -5
8 EV -14 -3
9 B6 -8 -3
10 AA 8 -2
# ℹ 336,766 more rowsCalculating the average departure delay
EndeavorAir <- filter(flights_delay, carrier == "9E")
EndeavorAir# A tibble: 18,460 × 3
carrier arr_delay dep_delay
<chr> <dbl> <dbl>
1 9E 11 0
2 9E -2 -9
3 9E -2 -3
4 9E -1 -6
5 9E -5 -8
6 9E -5 0
7 9E 5 6
8 9E 13 0
9 9E -8 -8
10 9E -33 -6
# ℹ 18,450 more rowsEndeavorAir_dd_mean <- mean(EndeavorAir$dep_delay, na.rm = TRUE)AmAir <- filter(flights_delay, carrier == "AA")
AmAir# A tibble: 32,729 × 3
carrier arr_delay dep_delay
<chr> <dbl> <dbl>
1 AA 33 2
2 AA 8 -2
3 AA 31 -1
4 AA -12 -4
5 AA 5 13
6 AA -3 -2
7 AA 14 -1
8 AA 48 0
9 AA 4 -4
10 AA -10 -3
# ℹ 32,719 more rowsAmAir_dd_mean <- mean(AmAir$dep_delay, na.rm = TRUE)AlaskaAir <- filter(flights_delay, carrier == "AS")
AlaskaAir# A tibble: 714 × 3
carrier arr_delay dep_delay
<chr> <dbl> <dbl>
1 AS -10 -1
2 AS -19 -7
3 AS -41 -3
4 AS 1 3
5 AS -18 -1
6 AS -9 2
7 AS 1 0
8 AS -29 -7
9 AS -19 0
10 AS -12 -12
# ℹ 704 more rowsAlaskaAir_dd_mean <- mean(AlaskaAir$dep_delay, na.rm = TRUE)JetBlue <- filter(flights_delay, carrier == "B6")
JetBlue# A tibble: 54,635 × 3
carrier arr_delay dep_delay
<chr> <dbl> <dbl>
1 B6 -18 -1
2 B6 19 -5
3 B6 -8 -3
4 B6 -2 -2
5 B6 -3 -2
6 B6 -4 0
7 B6 -7 0
8 B6 -6 1
9 B6 4 3
10 B6 -21 0
# ℹ 54,625 more rowsJetBlue_dd_mean <- mean(JetBlue$dep_delay, na.rm = TRUE)DeltaAir <- filter(flights_delay, carrier == "DL")
DeltaAir# A tibble: 48,110 × 3
carrier arr_delay dep_delay
<chr> <dbl> <dbl>
1 DL -25 -6
2 DL -8 -8
3 DL -8 -4
4 DL -9 0
5 DL -33 -7
6 DL -9 0
7 DL -8 -5
8 DL -18 -5
9 DL -14 -3
10 DL 5 -2
# ℹ 48,100 more rowsDeltaAir_dd_mean <- mean(DeltaAir$dep_delay, na.rm = TRUE)ExpressAir <- filter(flights_delay, carrier == "EV")
ExpressAir# A tibble: 54,173 × 3
carrier arr_delay dep_delay
<chr> <dbl> <dbl>
1 EV -14 -3
2 EV 29 -6
3 EV 12 24
4 EV -7 -6
5 EV -26 -4
6 EV 23 -2
7 EV -2 -8
8 EV -18 -4
9 EV -4 0
10 EV -14 0
# ℹ 54,163 more rowsExpressAir_dd_mean <- mean(ExpressAir$dep_delay, na.rm = TRUE)FrontierAir <- filter(flights_delay, carrier == "F9")
FrontierAir# A tibble: 685 × 3
carrier arr_delay dep_delay
<chr> <dbl> <dbl>
1 F9 32 -2
2 F9 -6 -14
3 F9 18 -8
4 F9 -1 -2
5 F9 0 0
6 F9 98 123
7 F9 -3 -1
8 F9 36 61
9 F9 -5 0
10 F9 -5 -4
# ℹ 675 more rowsFrontierAir_dd_mean <- mean(FrontierAir$dep_delay, na.rm = TRUE)AirTran <- filter(flights_delay, carrier == "FL")
AirTran# A tibble: 3,260 × 3
carrier arr_delay dep_delay
<chr> <dbl> <dbl>
1 FL 10 -3
2 FL 17 4
3 FL 2 -2
4 FL 8 -8
5 FL -5 -10
6 FL 6 -3
7 FL 6 -8
8 FL 14 0
9 FL 2 -11
10 FL -7 -10
# ℹ 3,250 more rowsAirTran_dd_mean <- mean(AirTran$dep_delay, na.rm = TRUE)HawaiiAir <- filter(flights_delay, carrier == "HA")
HawaiiAir# A tibble: 342 × 3
carrier arr_delay dep_delay
<chr> <dbl> <dbl>
1 HA -14 -3
2 HA -5 9
3 HA -26 14
4 HA -14 0
5 HA -11 -2
6 HA 28 79
7 HA 50 102
8 HA -26 1
9 HA 1272 1301
10 HA -41 -1
# ℹ 332 more rowsHawaiiAir_dd_mean <- mean(HawaiiAir$dep_delay, na.rm = TRUE)EnvoyAir <- filter(flights_delay, carrier == "MQ")
EnvoyAir# A tibble: 26,397 × 3
carrier arr_delay dep_delay
<chr> <dbl> <dbl>
1 MQ 12 0
2 MQ 16 -3
3 MQ 32 8
4 MQ 10 -6
5 MQ 27 -9
6 MQ 49 39
7 MQ -6 -10
8 MQ -4 -10
9 MQ 137 101
10 MQ -13 -4
# ℹ 26,387 more rowsEnvoyAir_dd_mean <- mean(EnvoyAir$dep_delay, na.rm = TRUE)SkyWest <- filter(flights_delay, carrier == "OO")
SkyWest# A tibble: 32 × 3
carrier arr_delay dep_delay
<chr> <dbl> <dbl>
1 OO 107 67
2 OO -5 -6
3 OO 27 13
4 OO -24 -8
5 OO -6 4
6 OO 3 1
7 OO -20 -9
8 OO 157 131
9 OO 3 -10
10 OO 140 154
# ℹ 22 more rowsSkyWest_dd_mean <- mean(SkyWest$dep_delay, na.rm = TRUE)UnitedAir <- filter(flights_delay, carrier == "UA")
UnitedAir# A tibble: 58,665 × 3
carrier arr_delay dep_delay
<chr> <dbl> <dbl>
1 UA 11 2
2 UA 20 4
3 UA 12 -4
4 UA 7 -2
5 UA -14 -2
6 UA -8 -1
7 UA -17 0
8 UA 14 11
9 UA 1 -4
10 UA 29 -2
# ℹ 58,655 more rowsUnitedAir_dd_mean <- mean(UnitedAir$dep_delay, na.rm = TRUE)USAir <- filter(flights_delay, carrier == "US")
USAir# A tibble: 20,536 × 3
carrier arr_delay dep_delay
<chr> <dbl> <dbl>
1 US 3 -8
2 US 0 -3
3 US -9 -1
4 US -11 -2
5 US -4 -7
6 US 10 -4
7 US -5 -2
8 US -29 -7
9 US 11 -5
10 US -15 -1
# ℹ 20,526 more rowsUSAir_dd_mean <- mean(USAir$dep_delay, na.rm = TRUE)VirginAir <- filter(flights_delay, carrier == "VX")
VirginAir# A tibble: 5,162 × 3
carrier arr_delay dep_delay
<chr> <dbl> <dbl>
1 VX 2 -2
2 VX -26 -1
3 VX -2 -1
4 VX -6 2
5 VX -22 1
6 VX -2 3
7 VX -26 3
8 VX -17 -3
9 VX -40 -3
10 VX -5 0
# ℹ 5,152 more rowsVirginAir_dd_mean <- mean(USAir$dep_delay, na.rm = TRUE)Southwest <- filter(flights_delay, carrier == "WN")
Southwest# A tibble: 12,275 × 3
carrier arr_delay dep_delay
<chr> <dbl> <dbl>
1 WN -19 -1
2 WN 33 -1
3 WN 26 -3
4 WN 30 4
5 WN 15 -2
6 WN 7 -3
7 WN -14 -1
8 WN 5 -5
9 WN -4 -2
10 WN 23 0
# ℹ 12,265 more rowsSouthwest_dd_mean <- mean(Southwest$dep_delay, na.rm = TRUE)MesaAir <- filter(flights_delay, carrier == "YV")
MesaAir# A tibble: 601 × 3
carrier arr_delay dep_delay
<chr> <dbl> <dbl>
1 YV -20 -7
2 YV -23 -11
3 YV -13 -5
4 YV 75 89
5 YV -15 -8
6 YV -18 -5
7 YV -1 -6
8 YV -22 -3
9 YV 5 -7
10 YV -16 -3
# ℹ 591 more rowsMesaAir_dd_mean <- mean(MesaAir$dep_delay, na.rm = TRUE)delays_by_airline <- data.frame(
carrier = c("Endeavor Air", "American Airlines","Alaska Airlines",
"JetBlue Airways", "Delta Air Lines", "ExpressJet Airlines",
"Frontier Airlines", "AirTran Airways", "Hawaiian Airlines",
"Envoy Air","SkyWest Airlines", "United Air Lines", "US Airways",
"Virgin America", "Southwest Airlines", "Mesa Airlines"),
mean_dep_delay = c(EndeavorAir_dd_mean, AmAir_dd_mean, AlaskaAir_dd_mean,
JetBlue_dd_mean, DeltaAir_dd_mean, ExpressAir_dd_mean,
FrontierAir_dd_mean, AirTran_dd_mean, HawaiiAir_dd_mean,
EnvoyAir_dd_mean, SkyWest_dd_mean, UnitedAir_dd_mean,
USAir_dd_mean, VirginAir_dd_mean, Southwest_dd_mean,
MesaAir_dd_mean),
flight_frequency = c(length(EndeavorAir$carrier), length(AmAir$carrier),
length(AlaskaAir$carrier),length(JetBlue$carrier),
length(DeltaAir$carrier),length(ExpressAir$carrier),
length(FrontierAir$carrier), length(AirTran$carrier),
length(HawaiiAir$carrier), length(EnvoyAir$carrier),
length(SkyWest$carrier), length(UnitedAir$carrier),
length(USAir$carrier), length(VirginAir$carrier),
length(Southwest$carrier), length(MesaAir$carrier))
)
delays_by_airline carrier mean_dep_delay flight_frequency
1 Endeavor Air 16.725769 18460
2 American Airlines 8.586016 32729
3 Alaska Airlines 5.804775 714
4 JetBlue Airways 13.022522 54635
5 Delta Air Lines 9.264505 48110
6 ExpressJet Airlines 19.955390 54173
7 Frontier Airlines 20.215543 685
8 AirTran Airways 18.726075 3260
9 Hawaiian Airlines 4.900585 342
10 Envoy Air 10.552041 26397
11 SkyWest Airlines 12.586207 32
12 United Air Lines 12.106073 58665
13 US Airways 3.782418 20536
14 Virgin America 3.782418 5162
15 Southwest Airlines 17.711744 12275
16 Mesa Airlines 18.996330 601library(treemap)
library(RColorBrewer)
treemap(delays_by_airline, index="carrier", vSize="flight_frequency",
vColor="mean_dep_delay", type="manual",
title = "Mean Departure Delays and Number of Flights (2013)",
fontsize.title = 18,
title.legend = "Mean Departure Delay",
force.print.labels = TRUE,
palette="PuRd", aspRatio = NA)
Source: Federal Aviation Administration
The airlines responsible for the most flights out of NYC in 2013 had mean departure delays closest to the median. The exception was ExpressJet Airlines, which had one of the highest mean departure delays, while third in the number of flights flown to and from NYC airports. Airlines operating fewer flights to and from NYC were more likely to have a very high or very low mean departure delay.