In ITSLeeds/roadworksUK: Accessing Processing and Visualising Data on UK Roadworks

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)

roadworksUK

The goal of roadworksUK is to enable you to access, process and visualise data on UK roadworks, particularly Electronic Transfer of Notifications (EToN) records.

Installation

install.packages("roadworksUK")

Install the package hosted on GitHub with:

# install.packages("devtools")
devtools::install_github("ITSLeeds/roadworksUK")

Load the package with:

library(roadworksUK)

What's in the package?

There are a number of functions for processing EToN records, as shown by:

x <- library(help = roadworksUK)
x$info[[2]]

The datasets provided by the package include kent10, a minimal dataset containing data from Kent, htdd_ashford, ~981 records from Ashford, Kent. The characteristics of these datasets is demonstrated below:

data("htdd_ashford")
dim(htdd_ashford) # a larger dataset
htdd_ashford[1:3, 1:5]

file_path = system.file("extdata", "kent10.csv.gz", package = "roadworksUK")
file.exists(file_path)
htdd_example = rw_import_elgin_htdd(file_path = file_path)
ncol(htdd_example)
nrow(htdd_example) # a small dataset with 10 rows

Example: common roadworks in Ashford

This section explores roadworks htdd_ashford, a dataset that is provided by the package: it's made available when you load roadworksUK. It relies on the tidyverse:

library(tidyverse)

A good way to 'get the measure' of potentially large spatio-temporal datasets is to find their size (in MB/GB/TB and number of rows/columns) and their temporal extent (we'll plot its spatial extent soon). We can find all of these things for the htdd_ashford dataset as follows:

pryr::object_size(htdd_ashford) # less than 2 MB - good test dataset
nrow(htdd_ashford)
ncol(htdd_ashford)
range(htdd_ashford$e__date_created)

A summary table of the contents of this dataset can be generated as follows:

htdd_ashford %>% 
  select(id, responsible_org_name, responsible_org_sector, description) %>% 
  slice(1:10) %>% 
  knitr::kable()

The following commands can find-out who reports road works in Ashford, using the dplyr package (part of the tidyverse):

htdd_ashford %>% 
  group_by(publisher_name) %>% 
  summarise(n = n()) %>% 
  arrange(desc(n))

It's mostly Kent County Council. There are a handful of reports by HE in the region also. Find out who does the work with the following commands (this finds the top 5 organisations, change then n parameter to see more organisations):

htdd_ashford %>% 
  group_by(responsible_org_name) %>% 
  summarise(n = n()) %>% 
  arrange(desc(n)) %>% 
  top_n(n = 5, wt = n)

Let's take a look at the temporal distribution of roadworks in the example dataset:

plot(htdd_ashford$e__date_created, htdd_ashford$e__duration_days)

This distribution is characteristic of roadworks data: it's not usually logged when it begins but after it ends. A log of actual reporting dates is illustrated in the next plot:

plot(htdd_ashford$e__date_updated, htdd_ashford$e__duration_days)

This shows the log comes from a single month (June) in 2018. We can do more sophisticated plots building on these examples and using packages such as ggplot2. For now, we will move on to plot the spatial extent of the object:

library(tmap)
tmap_mode("view")
tm_basemap(server = leaflet::providers$OpenTopoMap) +
  qtm(htdd_ashford$i__location_point)

library(tidyverse)
htdd_small = htdd_ashford %>% 
  select(description)

ITSLeeds/roadworksUK documentation built on May 31, 2019, 5:18 p.m.