Reproducible methods for network simplification

Robin Lovelace

Leeds Institute for Transport Studies, University of Leeds, UK

Zhao Wang

Leeds Institute for Transport Studies, University of Leeds, UK

Will Deakin

Digital, Data and Technology services, Network Rail, UK

Josiah Parry

Environmental Systems Research Institute, Redlands, CA, USA

Details

  • For listen-along audio, download and open this: https://github.com/nptscot/networkmerge/releases/download/v1/networkmerge-talk-gisruk.mp4
  • For the associated GISRUK extended abstract, see here: https://github.com/nptscot/networkmerge/releases/download/v1/gisruk.pdf
  • For the code, see here: https://github.com/nptscot/

The problem

(a)
(b)
Figure 1: Two parallel ways in Leeds, UK, with similar flow but different widths.

Issues and live demo

Illustration of issues associated with route network-level results containing multiple parallel ways on the same corridor: it is not clear from the visualisation that the corridor shown in the right hand figure has greater flow than the corridor shown in the left.. Source: open access Propensity to Cycle Tool results available at www.pct.bike.

Demo: https://www.pct.bike/m/?r =west-yorkshire

Solution + demo

 

Figure 2: The Network Planning Tool for Scotland, showing the network results for central Edinburgh without simplification (left) and with simplification (right). Note that the values on Princes Street (highlighted) are hard to interpret without simplification. Demo: www.npt.scot.

Prior work

Route network aggregation generates estimates of flow on the network, rasterisation help with visualisatioin (Morgan and Lovelace 2020)

Existing simplification algorithms

Buffering

Buffer rasterisation

skeletonization

Knot removal

The merge stage

Rust implementation

See https://github.com/nptscot/rnetmatch/blob/main/paper.qmd

The code

The code underlying the results presented in this paper are available from the following repositories:

  • The nptscot/networkmerge repository contains the reproducible paper.
  • The parenx Python for image skeletonization and Voronoi diagram-centreline identification is available on PyPI in the GitHub repo anisotropi4/parenx.
  • The rnetmatch R package for network simplification is available on GitHub in the repo nptscot/rnetmatch.

References

Morgan, Malcolm, and Robin Lovelace. 2020. “Travel Flow Aggregation: Nationally Scalable Methods for Interactive and Online Visualisation of Transport Behaviour at the Road Network Level.” Environment & Planning B: Planning & Design, July. https://doi.org/10.1177/2399808320942779.