airlinesafety.Rmd

---
title: "Airline Safety"
output: html_document
editor_options: 
  chunk_output_type: console
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```

```{r message=FALSE, warning=FALSE}
library(tidyverse)

airline_safety <- read_csv("https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2018/2018-08-07/week19_airline_safety.csv") %>%
  select(-X1)

theme_set(theme_light())
```

## What are the airlines in this dataset?

```{r}
airline_safety %>%
  count(airline, sort = TRUE)
```

## Which airlines have had the most incidents, fatalities and fatal accidents?

```{r}
airline_safety %>%
  filter(type_of_event == "incidents") %>%
  group_by(airline) %>%
  summarise(total_incidents = sum(n_events)) %>%
  arrange(desc(total_incidents))

airline_safety %>%
  filter(type_of_event == "fatalities") %>%
  group_by(airline) %>%
  summarise(total_fatalities = sum(n_events)) %>%
  arrange(desc(total_fatalities))

airline_safety %>%
  filter(type_of_event == "fatal_accidents") %>%
  group_by(airline) %>%
  summarise(total_fatal_accidents = sum(n_events)) %>%
  arrange(desc(total_fatal_accidents))

```

## Have the total number of airline incidents decreased over time?

```{r fig.width=7, fig.height=6}
airline_safety %>%
  filter(type_of_event != "fatalities") %>%
  group_by(airline, year_range) %>%
  summarise(total_incidents = sum(n_events)) %>%
  ggplot(aes(factor(year_range, levels = c("85_99", "00_14")), total_incidents, fill = year_range)) +
  geom_boxplot(show.legend = F) +
  labs(x = "Year",
       y = "Total Incidents",
       title = "Number of airline incidents and fatal accidents between 1985-99 and 2000-14",
       subtitle = "Significant outlier is Aeroflot, with a total of 90 incidents between 1985 and 1999")
```


```{r}
airline_safety %>%
  filter(type_of_event != "fatalities") %>%
  mutate(events_ASK = n_events * 1000000 / avail_seat_km_per_week) %>%
  arrange(desc(events_ASK)) %>%
  head(15) %>%
  mutate(airline = fct_reorder(airline, events_ASK)) %>%
  ggplot(aes(x = airline, y = events_ASK, fill = airline)) +
  geom_col(show.legend = FALSE) +
  coord_flip() +
  labs(x = NULL,
       y = "Number of events per million kilometres")
```