Advanced topics

1 Visualising large datasets

Install and try visualising datasets from previous sessions with:

  • {mapgl}: an R interface to Mapbox GL JS for rendering large spatial datasets.
  • {tmap.gl}: Use your existing tmap code, that builds on mapgl
  • The {nptscot/pmtiles} GitHub package provides tools for creating vector tiles in R (advanced).

2 Route network integration

  • Take a read of the networkmerge paper and try to reproduce some of the code
  • Search for and download the Ordnance Survey’s OpenRoads data product
  • Check-out the stplanr function rnet_join() for joining different route networks, e.g. walking and cycling networks.
  • Check-out the anime package for astonishingly fast network joining (advanced).

3 Deploying your work as web applications

  • Think about the project that you’re working on that would have the maximum benefit from being deployed as a web application.
  • Sketch out the user interface and user experience (UI/UX) of the application.
  • Sketch the data flow and architecture of backend systems.
  • Read-up on the targets package for reproducible pipelines.
  • Take a look at the source code of some deployed web applications for inspiration
  • Check out the code in the nptscot/nptscot and nptscot/nptscot.github.io repositories, plus the resulting app at npt.scot.
  • Look for the source code associated with other apps including pctni.org, pct.bike and other apps of your choice.

3.1 Examples of R-based Transport Web Apps

3.1.1 Propensity to Cycle Tool (PCT)

The PCT is an open-source planning support system for sustainable transport, providing evidence-based cycling scenarios across England and Wales.

(Interactive map screenshot: Cycling potential layers over England and Wales, showing commute scenarios with color-coded routes)

  • Built with Shiny and Leaflet for interactive maps
  • Scenarios include Government Target, Go Dutch, Ebike; layers for commute/school trips
  • Best practices: Integrate open data (census, OSM), enable scenario modeling for policy impact, use open-source for community contributions
  • Source: pct.bike, GitHub npct/pct; see paper for methodology

3.1.2 Propensity to Cycle Tool for Northern Ireland (PCTNI)

An alpha version adapted for Northern Ireland, building on PCT and NPT.

(Interactive map screenshot: Route network layers for Northern Ireland, highlighting fastest/quietest cycling routes)

  • Features route networks (fastest/quietest), cycle infrastructure layers
  • Based on 2011 Census for work trips; active development with OSM integration
  • Best practices: Iterative releases with issue tracking, validate against local data, emphasize limitations for responsible use
  • Source: pctni.org, GitHub pctni

Reuse