Demonstrating the PCT, for education and reproducibility

Welcome to pct-demo, a small repo with a little code and data to how how the pct works.

The starting point is that regional data has already been generated. See pct-scripts or an academic paper on the subject (Lovelace et al. 2017).

We also assume your have R and RStudio installed and running on your computer. There is plenty of guidance online, notably on the RStudio website: rstudio.com.

The first step is to download the the repo https://github.com/npct/pct-demo/archive/master.zip - that contains both code and data. Note that it contains a .qgs file that can be opened with the open source program QGIS - another powerful tool for analysing the geographical distribution of cycling potential.

The remainder of this tutorial is based on R code, which can be found in the code folder. We encourage you to look over these scripts. The script to load-in the data, for example, can be opened with the following command:

file.edit("code/load-data.R")

Once that file is open we can run it line-by-line, e.g. by pressing Ctl-Enter. Alternatively you can entirety of a script file with the source() function.

There are some dependencies: you need to have some packages installed.

source("code/set-up.R")

## Loading required package: sp

## Linking to GEOS 3.5.1, GDAL 2.2.1, proj.4 4.9.2, lwgeom 2.3.3 r15473

To load the input data, we can run the following script:

source("code/load-data.R")

Zones, centroids, lines, routes

The input data we have just loaded can be seen by looking in the Environment tab in RStudio. Alternatively you can use the function ls():

ls()

## [1] "cents"      "l"          "pkgs"       "rf"         "rnet"      
## [6] "rq"         "to_install" "z"

This shows that we have loaded the following data objects:

• z, administrative zones about which we have cycling data
• cents, population-weighted centroids, one per zone
• l, straight 'desire lines' representing travel between one zone and another
• rf route-level data, like l but allocated to the road network

We can plot all these layers interactively as follows:

library(tmap)
tmap_mode("view")

## tmap mode set to interactive viewing

qtm(z, "bicycle") +
  qtm(cents) +
  qtm(l) +
  qtm(rf, col = "green")