https://github.com/robinlovelace/cyclehirelocationslondon
https://github.com/robinlovelace/cyclehirelocationslondon
Last synced: about 1 month ago
JSON representation
- Host: GitHub
- URL: https://github.com/robinlovelace/cyclehirelocationslondon
- Owner: Robinlovelace
- Created: 2021-03-05T14:54:33.000Z (about 4 years ago)
- Default Branch: main
- Last Pushed: 2021-03-05T15:41:02.000Z (about 4 years ago)
- Last Synced: 2025-04-03T16:12:19.643Z (about 2 months ago)
- Size: 197 KB
- Stars: 1
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.Rmd
Awesome Lists containing this project
README
---
output: github_document
---
# Cycle Hire Locations in London
This repo contains reproducible code to get London cycle hire data.
We will get them using R with the following packages loaded:
```{r, message=FALSE, warning=FALSE}
library(tidyverse)
library(osmdata)
library(tmap)
tmap_mode("view")
```
## Official data
TfL's official dataset on cycle hire locations is no longer available: https://data.london.gov.uk/dataset/tfl-cycle-hire-locations
However, there was a FOI made in 2020, leading to the publication of cycle hire location data, which can be accessed from R as follows:
```{r}
u = "http://foi.tfl.gov.uk/FOI-2898-1920/FOI-2898-1920.xlsx"
f = basename(u)
download.file(u, f)
stations_tfl_2020 = readxl::read_excel(f)
stations_tfl_2020
nrow(stations_tfl_2020)
```
As the above code shows, that contains 802 docking stations.
We believe these are stations that are currently active.
We set the id to a character to match the OSM data:
```{r}
stations_tfl_2020$official_ref = as.character(stations_tfl_2020$Station.Id)
```
We will remove the pop-up stations, which only exist for a few days, as follows:
```{r}
stations_tfl_2020 = stations_tfl_2020 %>%
filter(!stringr::str_detect(string = StationName, pattern = "Pop"))
```
## OSM data
The OSM community has been active in mapping many things, including cycle hire docking stations.
The following commands get this data:
```{r}
stations_osm_list = opq("london") %>%
add_osm_feature(key = "amenity", value = "bicycle_rental") %>%
osmdata_sf()
stations_osm = stations_osm_list$osm_points
stations_osm = stations_osm %>% select(osm_id, name, brand, capacity, start_date, network, official_name, official_ref)
stations_osm = stations_osm %>%
filter(str_detect(string = network, "tfl") | str_detect(string = brand, "antand"))
```
## Comparing TfL and OSM sources
There are 8 more stations in the official dataset.
We can join the OSM data with the official data as follows, after a few checks:
```{r}
summary(duplicated(stations_osm$official_ref))
summary({osm_in_official = stations_osm$official_ref %in% stations_tfl_2020$official_ref})
nrow(stations_tfl_2020) - nrow(stations_osm)
stations_combined = left_join(
stations_tfl_2020,
stations_osm %>% sf::st_drop_geometry()
)
```
```{r, echo=FALSE}
# stations_2010 = spData::cycle_hire_osm
# stations_2010$year = 2010
# stations_combined = left_join(
# stations_combined,
# stations_2010 %>% sf::st_drop_geometry() %>% select(name, year)
# )
```
```{r}
stations_combined$capacity = as.numeric(stations_combined$capacity)
```
We need to make this combined dataset a geographic class for mapping:
```{r}
# stations_sf = sf::st_as_sf(stations_combined, coords = c("longitude", "latitude"), crs = 4326)
stations_sf = sf::st_as_sf(stations_combined, coords = c("Easting", "Northing"), crs = 27700)
```
```{r plot}
stations_area = stations_sf %>%
sf::st_buffer(1000) %>%
sf::st_union()
qtm(stations_area, fill = NULL) +
tm_shape(stations_sf) +
tm_dots(size = "capacity", scale = 0.3, alpha = 0.5)
```
## Updating the data
There are many ways we could improve the data.
The most obvious is the start date.
If anyone has ideas on that, please let us know on the issue tracker.
## Publishing open data
We hope to collaborate with others to improve and eventually publish data data as open data.