2015 Flights Delay Data analysisfrom LAX to SEA

Dataset: 2015 Flights Delay Data from LAX to SEA

Exploratory Analysis

Hypothesis here is that most of the arrival delays for route from LAX to SEA are due Airline or airplane related delays and not due to weather or security.
Secondly, certain airlines perform better than other while flying from LAX to SEA based on arrival delay.

Summary/Conclusion for exploratory analysis

For airline route from LAX to SEA

Departure delay has significant impact on Arrival delay.
Weather, Security, Air System delays does not have impact on arrival delay.
Skywest and United airlines are least reliable and Alaska airlines is most reliable in terms of arrival delay.

Context in relation to Flights dataset

The U.S. Department of Transportation’s (DOT) Bureau of Transportation Statistics tracks the on-time performance of domestic flights operated by large air carriers. Summary information on the number of on-time, delayed, canceled, and diverted flights is published in DOT’s monthly Air Travel Consumer Report and in this dataset of 2015 flight delays and cancellations.

Fetch data from flights.csv

Flights.data  <- read.table("~/Desktop/Flight Delays 2015/flights.csv", sep =",", header = TRUE)

head(Flights.data)

##   YEAR MONTH DAY DAY_OF_WEEK AIRLINE FLIGHT_NUMBER TAIL_NUMBER
## 1 2015     1   1           4      AS            98      N407AS
## 2 2015     1   1           4      AA          2336      N3KUAA
## 3 2015     1   1           4      US           840      N171US
## 4 2015     1   1           4      AA           258      N3HYAA
## 5 2015     1   1           4      AS           135      N527AS
## 6 2015     1   1           4      DL           806      N3730B
##   ORIGIN_AIRPORT DESTINATION_AIRPORT SCHEDULED_DEPARTURE DEPARTURE_TIME
## 1            ANC                 SEA                   5           2354
## 2            LAX                 PBI                  10              2
## 3            SFO                 CLT                  20             18
## 4            LAX                 MIA                  20             15
## 5            SEA                 ANC                  25             24
## 6            SFO                 MSP                  25             20
##   DEPARTURE_DELAY TAXI_OUT WHEELS_OFF SCHEDULED_TIME ELAPSED_TIME AIR_TIME
## 1             -11       21         15            205          194      169
## 2              -8       12         14            280          279      263
## 3              -2       16         34            286          293      266
## 4              -5       15         30            285          281      258
## 5              -1       11         35            235          215      199
## 6              -5       18         38            217          230      206
##   DISTANCE WHEELS_ON TAXI_IN SCHEDULED_ARRIVAL ARRIVAL_TIME ARRIVAL_DELAY
## 1     1448       404       4               430          408           -22
## 2     2330       737       4               750          741            -9
## 3     2296       800      11               806          811             5
## 4     2342       748       8               805          756            -9
## 5     1448       254       5               320          259           -21
## 6     1589       604       6               602          610             8
##   DIVERTED CANCELLED CANCELLATION_REASON AIR_SYSTEM_DELAY SECURITY_DELAY
## 1        0         0                                   NA             NA
## 2        0         0                                   NA             NA
## 3        0         0                                   NA             NA
## 4        0         0                                   NA             NA
## 5        0         0                                   NA             NA
## 6        0         0                                   NA             NA
##   AIRLINE_DELAY LATE_AIRCRAFT_DELAY WEATHER_DELAY
## 1            NA                  NA            NA
## 2            NA                  NA            NA
## 3            NA                  NA            NA
## 4            NA                  NA            NA
## 5            NA                  NA            NA
## 6            NA                  NA            NA

Load Required functions

require(dplyr)

## Loading required package: dplyr

## Warning: package 'dplyr' was built under R version 3.4.2

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

library(MASS)

## 
## Attaching package: 'MASS'

## The following object is masked from 'package:dplyr':
## 
##     select

require(repr)

## Loading required package: repr

## Warning in library(package, lib.loc = lib.loc, character.only = TRUE,
## logical.return = TRUE, : there is no package called 'repr'

require(ggplot2)

## Loading required package: ggplot2

require(gridExtra)

## Loading required package: gridExtra

## 
## Attaching package: 'gridExtra'

## The following object is masked from 'package:dplyr':
## 
##     combine

require(glmnet)

## Loading required package: glmnet

## Warning in library(package, lib.loc = lib.loc, character.only = TRUE,
## logical.return = TRUE, : there is no package called 'glmnet'

Reduced and cleansed data

flights.delay.sub <- Flights.data[,c('YEAR','MONTH','DAY','DAY_OF_WEEK','AIRLINE','ORIGIN_AIRPORT','DESTINATION_AIRPORT',
                                     'SCHEDULED_DEPARTURE','SCHEDULED_ARRIVAL','DEPARTURE_DELAY','ARRIVAL_DELAY',
                                     'AIR_SYSTEM_DELAY','SECURITY_DELAY','AIRLINE_DELAY','LATE_AIRCRAFT_DELAY',
                                     'WEATHER_DELAY')]
flights.delay.sub <- flights.delay.sub[complete.cases(flights.delay.sub), ]
flights.delay.sub.orig.lax.dest.sea <- flights.delay.sub[ which(flights.delay.sub$ORIGIN_AIRPORT=='LAX' & 
                                                                flights.delay.sub$DESTINATION_AIRPORT=='SEA'), ]
numcols <- c('DEPARTURE_DELAY','ARRIVAL_DELAY','AIR_SYSTEM_DELAY','SECURITY_DELAY','AIRLINE_DELAY',
             'LATE_AIRCRAFT_DELAY','WEATHER_DELAY')

Exploratory Analysis

Peek into the data and guage it

head(flights.delay.sub.orig.lax.dest.sea)

##       YEAR MONTH DAY DAY_OF_WEEK AIRLINE ORIGIN_AIRPORT
## 6135  2015     1   1           4      AS            LAX
## 9942  2015     1   1           4      VX            LAX
## 13707 2015     1   1           4      AS            LAX
## 17449 2015     1   2           5      AS            LAX
## 19327 2015     1   2           5      AS            LAX
## 21837 2015     1   2           5      AS            LAX
##       DESTINATION_AIRPORT SCHEDULED_DEPARTURE SCHEDULED_ARRIVAL
## 6135                  SEA                1300              1534
## 9942                  SEA                1705              1945
## 13707                 SEA                2225               106
## 17449                 SEA                 840              1127
## 19327                 SEA                1030              1314
## 21837                 SEA                1300              1534
##       DEPARTURE_DELAY ARRIVAL_DELAY AIR_SYSTEM_DELAY SECURITY_DELAY
## 6135               82            96               14              0
## 9942               20            16                0              0
## 13707              62            60                0              0
## 17449             104           105                1              0
## 19327               8            28               20              0
## 21837               0            19               19              0
##       AIRLINE_DELAY LATE_AIRCRAFT_DELAY WEATHER_DELAY
## 6135              0                  82             0
## 9942             16                   0             0
## 13707             0                  60             0
## 17449           104                   0             0
## 19327             0                   8             0
## 21837             0                   0             0

summary(flights.delay.sub.orig.lax.dest.sea)

##       YEAR          MONTH             DAY         DAY_OF_WEEK   
##  Min.   :2015   Min.   : 1.000   Min.   : 1.00   Min.   :1.000  
##  1st Qu.:2015   1st Qu.: 4.000   1st Qu.: 9.00   1st Qu.:2.000  
##  Median :2015   Median : 8.000   Median :16.00   Median :4.000  
##  Mean   :2015   Mean   : 7.185   Mean   :15.82   Mean   :3.898  
##  3rd Qu.:2015   3rd Qu.:11.000   3rd Qu.:22.00   3rd Qu.:5.000  
##  Max.   :2015   Max.   :12.000   Max.   :31.00   Max.   :7.000  
##                                                                 
##     AIRLINE    ORIGIN_AIRPORT DESTINATION_AIRPORT SCHEDULED_DEPARTURE
##  AS     :618   LAX    :1221   SEA    :1221        Min.   : 600       
##  DL     :250   10135  :   0   10135  :   0        1st Qu.:1050       
##  VX     :212   10136  :   0   10136  :   0        Median :1515       
##  OO     :123   10140  :   0   10140  :   0        Mean   :1460       
##  UA     : 17   10141  :   0   10141  :   0        3rd Qu.:1840       
##  B6     :  1   10146  :   0   10146  :   0        Max.   :2355       
##  (Other):  0   (Other):   0   (Other):   0                           
##  SCHEDULED_ARRIVAL DEPARTURE_DELAY  ARRIVAL_DELAY    AIR_SYSTEM_DELAY
##  Min.   :   1      Min.   :-12.00   Min.   : 15.00   Min.   :  0.00  
##  1st Qu.:1255      1st Qu.:  8.00   1st Qu.: 21.00   1st Qu.:  0.00  
##  Median :1719      Median : 27.00   Median : 31.00   Median :  6.00  
##  Mean   :1621      Mean   : 37.68   Mean   : 44.03   Mean   : 11.26  
##  3rd Qu.:2100      3rd Qu.: 52.00   3rd Qu.: 52.00   3rd Qu.: 18.00  
##  Max.   :2358      Max.   :444.00   Max.   :451.00   Max.   :103.00  
##                                                                      
##  SECURITY_DELAY   AIRLINE_DELAY    LATE_AIRCRAFT_DELAY WEATHER_DELAY     
##  Min.   : 0.000   Min.   :  0.00   Min.   :  0.0       Min.   :  0.0000  
##  1st Qu.: 0.000   1st Qu.:  0.00   1st Qu.:  0.0       1st Qu.:  0.0000  
##  Median : 0.000   Median :  0.00   Median :  0.0       Median :  0.0000  
##  Mean   : 0.086   Mean   : 11.61   Mean   : 20.3       Mean   :  0.7682  
##  3rd Qu.: 0.000   3rd Qu.:  8.00   3rd Qu.: 29.0       3rd Qu.:  0.0000  
##  Max.   :38.000   Max.   :334.00   Max.   :444.0       Max.   :128.0000  
##

str(flights.delay.sub.orig.lax.dest.sea)

## 'data.frame':    1221 obs. of  16 variables:
##  $ YEAR               : int  2015 2015 2015 2015 2015 2015 2015 2015 2015 2015 ...
##  $ MONTH              : int  1 1 1 1 1 1 1 1 1 1 ...
##  $ DAY                : int  1 1 1 2 2 2 2 2 2 2 ...
##  $ DAY_OF_WEEK        : int  4 4 4 5 5 5 5 5 5 5 ...
##  $ AIRLINE            : Factor w/ 14 levels "AA","AS","B6",..: 2 13 2 2 2 2 13 11 2 13 ...
##  $ ORIGIN_AIRPORT     : Factor w/ 628 levels "10135","10136",..: 483 483 483 483 483 483 483 483 483 483 ...
##  $ DESTINATION_AIRPORT: Factor w/ 629 levels "10135","10136",..: 585 585 585 585 585 585 585 585 585 585 ...
##  $ SCHEDULED_DEPARTURE: int  1300 1705 2225 840 1030 1300 1705 1935 1940 1955 ...
##  $ SCHEDULED_ARRIVAL  : int  1534 1945 106 1127 1314 1534 1945 2221 2220 2230 ...
##  $ DEPARTURE_DELAY    : int  82 20 62 104 8 0 20 32 12 6 ...
##  $ ARRIVAL_DELAY      : int  96 16 60 105 28 19 20 41 31 24 ...
##  $ AIR_SYSTEM_DELAY   : int  14 0 0 1 20 19 4 9 19 0 ...
##  $ SECURITY_DELAY     : int  0 0 0 0 0 0 0 0 0 0 ...
##  $ AIRLINE_DELAY      : int  0 16 0 104 0 0 0 8 12 24 ...
##  $ LATE_AIRCRAFT_DELAY: int  82 0 60 0 8 0 16 24 0 0 ...
##  $ WEATHER_DELAY      : int  0 0 0 0 0 0 0 0 0 0 ...

Check for correlation and covariance between different kind of delays

cor(flights.delay.sub.orig.lax.dest.sea[,numcols])

##                     DEPARTURE_DELAY ARRIVAL_DELAY AIR_SYSTEM_DELAY
## DEPARTURE_DELAY         1.000000000   0.944965755      -0.26801529
## ARRIVAL_DELAY           0.944965755   1.000000000      -0.06030859
## AIR_SYSTEM_DELAY       -0.268015286  -0.060308594       1.00000000
## SECURITY_DELAY         -0.005092528  -0.006099835      -0.02887722
## AIRLINE_DELAY           0.549427204   0.549944125      -0.18845956
## LATE_AIRCRAFT_DELAY     0.680104218   0.652664761      -0.30548226
## WEATHER_DELAY           0.048874405   0.064121013      -0.05445003
##                     SECURITY_DELAY AIRLINE_DELAY LATE_AIRCRAFT_DELAY
## DEPARTURE_DELAY       -0.005092528    0.54942720          0.68010422
## ARRIVAL_DELAY         -0.006099835    0.54994412          0.65266476
## AIR_SYSTEM_DELAY      -0.028877223   -0.18845956         -0.30548226
## SECURITY_DELAY         1.000000000   -0.01734383         -0.02433607
## AIRLINE_DELAY         -0.017343826    1.00000000         -0.16159407
## LATE_AIRCRAFT_DELAY   -0.024336074   -0.16159407          1.00000000
## WEATHER_DELAY         -0.006465666   -0.04121311         -0.05107731
##                     WEATHER_DELAY
## DEPARTURE_DELAY       0.048874405
## ARRIVAL_DELAY         0.064121013
## AIR_SYSTEM_DELAY     -0.054450030
## SECURITY_DELAY       -0.006465666
## AIRLINE_DELAY        -0.041213113
## LATE_AIRCRAFT_DELAY  -0.051077308
## WEATHER_DELAY         1.000000000

cov(flights.delay.sub.orig.lax.dest.sea[,numcols])

##                     DEPARTURE_DELAY ARRIVAL_DELAY AIR_SYSTEM_DELAY
## DEPARTURE_DELAY        1974.2148763  1646.9518636     -172.9264463
## ARRIVAL_DELAY          1646.9518636  1538.6337388      -34.3519562
## AIR_SYSTEM_DELAY       -172.9264463   -34.3519562      210.8673232
## SECURITY_DELAY           -0.3633524    -0.3842228       -0.6733758
## AIRLINE_DELAY           739.7570293   653.6841033      -82.9286388
## LATE_AIRCRAFT_DELAY    1066.6558270   903.6690794     -156.5821572
## WEATHER_DELAY            13.8288060    16.0167351       -5.0351076
##                     SECURITY_DELAY AIRLINE_DELAY LATE_AIRCRAFT_DELAY
## DEPARTURE_DELAY        -0.36335240   739.7570293          1066.65583
## ARRIVAL_DELAY          -0.38422282   653.6841033           903.66908
## AIR_SYSTEM_DELAY       -0.67337576   -82.9286388          -156.58216
## SECURITY_DELAY          2.57866436    -0.8439642            -1.37943
## AIRLINE_DELAY          -0.84396423   918.2553403          -172.84578
## LATE_AIRCRAFT_DELAY    -1.37942965  -172.8457774          1245.95760
## WEATHER_DELAY          -0.06611753    -7.9528564           -11.48116
##                     WEATHER_DELAY
## DEPARTURE_DELAY       13.82880600
## ARRIVAL_DELAY         16.01673514
## AIR_SYSTEM_DELAY      -5.03510761
## SECURITY_DELAY        -0.06611753
## AIRLINE_DELAY         -7.95285643
## LATE_AIRCRAFT_DELAY  -11.48115627
## WEATHER_DELAY         40.55197299

options(repr.plot.width=12, repr.plot.height=12)
require(car)

## Loading required package: car

## Warning: package 'car' was built under R version 3.4.3

## 
## Attaching package: 'car'

## The following object is masked from 'package:dplyr':
## 
##     recode

suppressWarnings(
scatterplotMatrix(~ flights.delay.sub.orig.lax.dest.sea$DEPARTURE_DELAY 
                  + flights.delay.sub.orig.lax.dest.sea$ARRIVAL_DELAY 
                  + flights.delay.sub.orig.lax.dest.sea$AIR_SYSTEM_DELAY 
                  + flights.delay.sub.orig.lax.dest.sea$SECURITY_DELAY
                  + flights.delay.sub.orig.lax.dest.sea$AIRLINE_DELAY
                  + flights.delay.sub.orig.lax.dest.sea$LATE_AIRCRAFT_DELAY
                  + flights.delay.sub.orig.lax.dest.sea$WEATHER_DELAY
                  , data = flights.delay.sub.orig.lax.dest.sea))

options(repr.plot.width=12, repr.plot.height=6)
library(ellipse)

## 
## Attaching package: 'ellipse'

## The following object is masked from 'package:car':
## 
##     ellipse

R = cor(flights.delay.sub.orig.lax.dest.sea[,numcols], method = 'kendal')
print(R)

##                     DEPARTURE_DELAY ARRIVAL_DELAY AIR_SYSTEM_DELAY
## DEPARTURE_DELAY          1.00000000   0.638419794     -0.395051900
## ARRIVAL_DELAY            0.63841979   1.000000000     -0.069167017
## AIR_SYSTEM_DELAY        -0.39505190  -0.069167017      1.000000000
## SECURITY_DELAY           0.00494000   0.012005478     -0.002923588
## AIRLINE_DELAY            0.25712412   0.161440788     -0.294412287
## LATE_AIRCRAFT_DELAY      0.49181928   0.372501994     -0.377238395
## WEATHER_DELAY           -0.01706855   0.004183648     -0.067211387
##                     SECURITY_DELAY AIRLINE_DELAY LATE_AIRCRAFT_DELAY
## DEPARTURE_DELAY        0.004940000   0.257124119          0.49181928
## ARRIVAL_DELAY          0.012005478   0.161440788          0.37250199
## AIR_SYSTEM_DELAY      -0.002923588  -0.294412287         -0.37723840
## SECURITY_DELAY         1.000000000  -0.004308466         -0.01642175
## AIRLINE_DELAY         -0.004308466   1.000000000         -0.20565278
## LATE_AIRCRAFT_DELAY   -0.016421746  -0.205652777          1.00000000
## WEATHER_DELAY         -0.012135734  -0.060608298         -0.06160339
##                     WEATHER_DELAY
## DEPARTURE_DELAY      -0.017068551
## ARRIVAL_DELAY         0.004183648
## AIR_SYSTEM_DELAY     -0.067211387
## SECURITY_DELAY       -0.012135734
## AIRLINE_DELAY        -0.060608298
## LATE_AIRCRAFT_DELAY  -0.061603394
## WEATHER_DELAY         1.000000000

plotcorr(R, col = colorRampPalette(c("firebrick3", "white", "navy"))(10))

For the flights from LAX to SEA and based on above correlation and covariance analysis, there is a very significant correlation between departure and arrival delays, however airline delay and late aircraft delay have a moderately significant impact on arrival delays while other delays like weather, air system, security have least impact (weak correlation) on flight arrival delay.

One discovery made here was that departure delay is most impacted by same airline delay and late aircraft delays, which in turn impacts arrival delay.

Arrival Delay Analysis

Based on above analysis, lets concentrate on Arrival delays by understanding which Airlines had highest delays. So, lets reduce the data only to DEPARTURE and ARRIVAL delays by MONTH/DAY_OF_WEEK/AIRLINE.

flights.delay.sub.new <- Flights.data[,c('YEAR','MONTH','DAY','DAY_OF_WEEK','AIRLINE','ORIGIN_AIRPORT',
                                         'DESTINATION_AIRPORT','DEPARTURE_DELAY','ARRIVAL_DELAY')]
head(flights.delay.sub.new)

##   YEAR MONTH DAY DAY_OF_WEEK AIRLINE ORIGIN_AIRPORT DESTINATION_AIRPORT
## 1 2015     1   1           4      AS            ANC                 SEA
## 2 2015     1   1           4      AA            LAX                 PBI
## 3 2015     1   1           4      US            SFO                 CLT
## 4 2015     1   1           4      AA            LAX                 MIA
## 5 2015     1   1           4      AS            SEA                 ANC
## 6 2015     1   1           4      DL            SFO                 MSP
##   DEPARTURE_DELAY ARRIVAL_DELAY
## 1             -11           -22
## 2              -8            -9
## 3              -2             5
## 4              -5            -9
## 5              -1           -21
## 6              -5             8

flights.delay.sub.new <- flights.delay.sub.new[complete.cases(flights.delay.sub.new), ]
flights.delay.sub.orig.lax.dest.sea <- flights.delay.sub.new[ which(flights.delay.sub.new$ORIGIN_AIRPORT=='LAX' & 
                                                                    flights.delay.sub.new$DESTINATION_AIRPORT=='SEA'), ]
numcols <- c('DEPARTURE_DELAY','ARRIVAL_DELAY')

Generating DATE column derived from YEAR, MONTH and DAY

flights.delay.sub.orig.lax.dest.sea$DATE <- as.Date(paste(flights.delay.sub.orig.lax.dest.sea$YEAR, 
                                                          flights.delay.sub.orig.lax.dest.sea$MONTH,
                                                          flights.delay.sub.orig.lax.dest.sea$DAY , sep = "."), 
                                                    format = "%Y.%m.%d")

## Warning in strptime(x, format, tz = "GMT"): unknown timezone 'zone/tz/
## 2017c.1.0/zoneinfo/Asia/Shanghai'

head(flights.delay.sub.orig.lax.dest.sea)

##      YEAR MONTH DAY DAY_OF_WEEK AIRLINE ORIGIN_AIRPORT DESTINATION_AIRPORT
## 233  2015     1   1           4      AS            LAX                 SEA
## 896  2015     1   1           4      AS            LAX                 SEA
## 2183 2015     1   1           4      AS            LAX                 SEA
## 2258 2015     1   1           4      DL            LAX                 SEA
## 3202 2015     1   1           4      OO            LAX                 SEA
## 3799 2015     1   1           4      AS            LAX                 SEA
##      DEPARTURE_DELAY ARRIVAL_DELAY       DATE
## 233               -5            -9 2015-01-01
## 896               -5           -15 2015-01-01
## 2183              -1            -7 2015-01-01
## 2258              -3           -13 2015-01-01
## 3202             -13           -12 2015-01-01
## 3799              -7           -18 2015-01-01

summary(flights.delay.sub.orig.lax.dest.sea)

##       YEAR          MONTH             DAY         DAY_OF_WEEK   
##  Min.   :2015   Min.   : 1.000   Min.   : 1.00   Min.   :1.000  
##  1st Qu.:2015   1st Qu.: 4.000   1st Qu.: 8.00   1st Qu.:2.000  
##  Median :2015   Median : 7.000   Median :16.00   Median :4.000  
##  Mean   :2015   Mean   : 6.611   Mean   :15.69   Mean   :3.911  
##  3rd Qu.:2015   3rd Qu.: 9.000   3rd Qu.:23.00   3rd Qu.:6.000  
##  Max.   :2015   Max.   :12.000   Max.   :31.00   Max.   :7.000  
##                                                                 
##     AIRLINE     ORIGIN_AIRPORT DESTINATION_AIRPORT DEPARTURE_DELAY 
##  AS     :4582   LAX    :7699   SEA    :7699        Min.   :-25.00  
##  DL     :1644   10135  :   0   10135  :   0        1st Qu.: -6.00  
##  VX     : 889   10136  :   0   10136  :   0        Median : -3.00  
##  OO     : 495   10140  :   0   10140  :   0        Mean   :  3.79  
##  UA     :  88   10141  :   0   10141  :   0        3rd Qu.:  2.00  
##  B6     :   1   10146  :   0   10146  :   0        Max.   :444.00  
##  (Other):   0   (Other):   0   (Other):   0                        
##  ARRIVAL_DELAY          DATE           
##  Min.   :-48.000   Min.   :2015-01-01  
##  1st Qu.:-12.000   1st Qu.:2015-04-14  
##  Median : -4.000   Median :2015-07-03  
##  Mean   :  1.495   Mean   :2015-07-04  
##  3rd Qu.:  7.000   3rd Qu.:2015-09-17  
##  Max.   :451.000   Max.   :2015-12-31  
##

str(flights.delay.sub.orig.lax.dest.sea)

## 'data.frame':    7699 obs. of  10 variables:
##  $ YEAR               : int  2015 2015 2015 2015 2015 2015 2015 2015 2015 2015 ...
##  $ MONTH              : int  1 1 1 1 1 1 1 1 1 1 ...
##  $ DAY                : int  1 1 1 1 1 1 1 1 1 1 ...
##  $ DAY_OF_WEEK        : int  4 4 4 4 4 4 4 4 4 4 ...
##  $ AIRLINE            : Factor w/ 14 levels "AA","AS","B6",..: 2 2 2 4 10 2 13 2 4 2 ...
##  $ ORIGIN_AIRPORT     : Factor w/ 628 levels "10135","10136",..: 483 483 483 483 483 483 483 483 483 483 ...
##  $ DESTINATION_AIRPORT: Factor w/ 629 levels "10135","10136",..: 585 585 585 585 585 585 585 585 585 585 ...
##  $ DEPARTURE_DELAY    : int  -5 -5 -1 -3 -13 -7 -1 82 6 -2 ...
##  $ ARRIVAL_DELAY      : int  -9 -15 -7 -13 -12 -18 -7 96 -3 -7 ...
##  $ DATE               : Date, format: "2015-01-01" "2015-01-01" ...

Lets look into the density of Arrival Delays

library(ggplot2)
ggplot(flights.delay.sub.orig.lax.dest.sea, aes(ARRIVAL_DELAY)) + geom_density() + 
xlab('Arrival Delay in minutes') + ylab('Density of Arrival Delay') + ggtitle('Density charr arrival delay by Airline for 2015 flights from LAX to SEA')

Above density chart shows me that most of the flights either reach before time or on-time. But, there are delays. Now lets check which airlines are delayed the most.

options(repr.plot.width=12, repr.plot.height=8)
ggplot(flights.delay.sub.orig.lax.dest.sea, aes(x = factor(AIRLINE), y = ARRIVAL_DELAY)) + geom_boxplot() + 
   ylab('Arrival Delay in minutes') + xlab('Airline Abbreviations') + 
   ggtitle('Arrival delay by Airline for 2015 flights from LAX to SEA')

ggplot(flights.delay.sub.orig.lax.dest.sea, aes(x = factor(AIRLINE), y = ARRIVAL_DELAY)) + 
    geom_violin(trim = TRUE, draw_quantiles = c(0.25, 0.5, 0.75)) +
    ylab('Arrival Delay in minutes') + xlab('Airline Abbreviations') + 
    ggtitle('Arrival delay by Airline for 2015 flights from LAX to SEA')

Above violin chart provides insight into the data that AS (Alaska Airlines) has more delays but apart from longer and more delay’s, Alaska Airlines overall record is better than other airlines accept B6 (Jetblue), which should not be counted as there is only one record for B6.

ggplot(flights.delay.sub.orig.lax.dest.sea, aes(DEPARTURE_DELAY, 
                                                ARRIVAL_DELAY)) + geom_point(aes(color = factor(AIRLINE),
                                                                                 size  = ARRIVAL_DELAY, alpha = 0.5)) + 
  xlab('Departure delay in minutes') + ylab('Arrival delay in minutes') +
  ggtitle('Relationship between Departure delay and Arrival delay with Airline shown')

Scatter plot above provides view of density as well as distribution of departure versus arrival delay, proving the hypothesis that most of the arrival delays occurs due departure delay when flying from LAX to SEA based on data from 2015.

ggplot(flights.delay.sub.orig.lax.dest.sea, aes(DAY_OF_WEEK, 
                                                ARRIVAL_DELAY)) + geom_point(aes(color = factor(AIRLINE), 
                                                                                 shape = factor(AIRLINE),
                                                                                 size  = ARRIVAL_DELAY, alpha = 0.5)) + 
  xlab('Day of week in numbers (MON=1...SUN=7)') + ylab('Arrival delay in minutes') +
  ggtitle('Relationship between Departure delay and Arrival delay with Airline shown')

Based on above plot, least amount of delay’s happen on Tuesday’s, Wednesday’s and Saturday’s while there is a changes flight being delayed on Monday’s, Thursday’s and Sunday’s if you fly from LAX to SEA based on 2015 Flights data.

2015 Flights Delay Data analysisfrom LAX to SEA

Dataset: 2015 Flights Delay Data from LAX to SEA

Exploratory Analysis

Summary/Conclusion for exploratory analysis

For airline route from LAX to SEA

Context in relation to Flights dataset

Fetch data from flights.csv

Load Required functions

Reduced and cleansed data

Exploratory Analysis

Peek into the data and guage it

Check for correlation and covariance between different kind of delays

One discovery made here was that departure delay is most impacted by same airline delay and late aircraft delays, which in turn impacts arrival delay.

Arrival Delay Analysis

Based on above analysis, lets concentrate on Arrival delays by understanding which Airlines had highest delays. So, lets reduce the data only to DEPARTURE and ARRIVAL delays by MONTH/DAY_OF_WEEK/AIRLINE.

Generating DATE column derived from YEAR, MONTH and DAY

Lets look into the density of Arrival Delays

Above density chart shows me that most of the flights either reach before time or on-time. But, there are delays. Now lets check which airlines are delayed the most.

Above violin chart provides insight into the data that AS (Alaska Airlines) has more delays but apart from longer and more delay’s, Alaska Airlines overall record is better than other airlines accept B6 (Jetblue), which should not be counted as there is only one record for B6.

Scatter plot above provides view of density as well as distribution of departure versus arrival delay, proving the hypothesis that most of the arrival delays occurs due departure delay when flying from LAX to SEA based on data from 2015.

Based on above plot, least amount of delay’s happen on Tuesday’s, Wednesday’s and Saturday’s while there is a changes flight being delayed on Monday’s, Thursday’s and Sunday’s if you fly from LAX to SEA based on 2015 Flights data.

Conclusion of Exploratory Analysis

1. Departure delay has significant impact on Arrival delay.

2. Weather, Security, Air System delays does not have impact on Arrival delay.

3. Skywest and United airlines are least reliable and Alaska airlines is most reliable in terms of arrival delay.