od2net R package: prepare datasets for network generation with od2net

This R package provides functions to prepare OD data for network generation with the od2net tool, as illustrated in the example below.

Installation

# Install pak if not already installed:
if (!requireNamespace("pak", quietly = TRUE)) {
  install.packages("pak")
}
pak::pkg_install("Urban-Analytics-Technology-Platform/od2net/r")

Example

Imagine you want to generate a route network with values on the links representing the number of pupils/parents on their way to school each school day. The following code shows how to prepare the data for the od2net tool.

The following functions could be useful when you’re preparing the data, and show the kind of data preparation that is required.

# Aim: generate input data for od2net with R

#' Generate a 'zones.geojson' file
#' 
#' This function requires a zones file, e.g.
#' "https://raw.githubusercontent.com/nptscot/npt/main/data-raw/zones_edinburgh.geojson"
#' or a file on your computer.
#' It will generate a file in the input/ folder
#' 
#' @param file Location or URL of zones file
make_zones = function(file) {
  zones = sf::read_sf(file)[1]
  names(zones)[1] = "name"
  sf::write_sf(zones, "input/zones.geojson", delete_dsn = TRUE)
}

make_od = function() {
  od = readr::read_csv("https://raw.githubusercontent.com/nptscot/npt/main/data-raw/od_subset.csv")
  od = od |>
    dplyr::transmute(from = geo_code1, to = geo_code2, count = bicycle)
  readr::write_csv(od, "input/od.csv")
}
#' Get elevation data
#' 
#' This function downloads elevation data from a source such as
#' https://play.abstreet.org/dev/data/input/shared/elevation/UK-dem-50m-4326.tif.gz
#' or https://assets.od2net.org/input/LisboaIST_10m_4326.tif
#' 
#' @param url Full URL of the elevation dataset if available
#' @param file File name if hosted on a known site
#' @param base_url Base URL associated with the 'file' argument
#' 
make_elevation = function(
    url = NULL,
    file = "UK-dem-50m-4326.tif.gz",
    base_url = "https://play.abstreet.org/dev/data/input/shared/elevation/"
    ) {
  if (is.null(url)) {
    url = paste0(base_url, file)
  }
  is_gzip = grepl(pattern = "gz", url)
  # Download the file
    if (!file.exists("input/elevation.tif") && is_gzip) {
      download.file(
          url = url,
          destfile = "input/elevation.tif.gz"
      )
      R.utils::gunzip("input/elevation.tif.gz", destname = "input/elevation.tif")
    } else {
      download.file(
        url = url,
        destfile = "input/elevation.tif"
      )
    }
}

dir.create("input", showWarnings = FALSE)
# Get some zones from a URL:
uz = "https://github.com/acteng/netgen/raw/main/input/zones_york.geojson"
make_zones(uz)
sf::write_sf(zones, "input/zones.geojson", delete_dsn = TRUE)
make_osm(zones_file = "input/zones.geojson", output_file = "input/input.osm.pbf")
make_origins(
  osm_file = "input/input.osm.pbf",
  query = "SELECT osm_id FROM multipolygons WHERE building IS NOT NULL",
  output_file = "input/buildings.geojson"
)
make_elevation()
destinations = simodels::destinations_york # Provided in the R package
names(destinations)[1] = "name"
destinations = destinations[1]
class(destinations$name) = "character"
sf::write_sf(destinations, "input/destinations.geojson", delete_dsn = TRUE)
od_geo = sf::read_sf("https://github.com/acteng/netgen/releases/download/v0.1.0/res_output.geojson")
# Save the OD dataset:
od = od_geo |>
  sf::st_drop_geometry() |>
  dplyr::transmute(from = O, to = as.character(D), count = round(trips_modelled))
readr::write_csv(od, "input/od.csv", quote = "all")

Then create a config.json file, e.g. with the following content:

readLines("config.json") |> cat(sep = "\n")

{
    "requests": {
      "description": "Test data for SchoolRoutes project.",
      "pattern": {
        "ZoneToPoint": {
          "zones_path": "zones.geojson",
          "destinations_path": "destinations.geojson",
          "csv_path": "od.csv",
           "origin_zone_centroid_fallback": false
        }
      },
      "origins_path": "buildings.geojson",
      "destinations_path": "destinations.geojson"
    },
    "cost": "Distance",
    "uptake": "Identity",
    "lts": "BikeOttawa",
    "elevation_geotiff": "elevation.tif"
}

Then run the following code to generate the network:

Run the tool with Docker as follows:

# On Linux:
sudo docker run -v $(pwd):/app ghcr.io/urban-analytics-technology-platform/od2net:main /app/config.json
# or in Windows:
sudo docker run -v ${pwd}:/app ghcr.io/urban-analytics-technology-platform/od2net:main /app/config.json

After that you should see something like the following in the output folder:

fs::dir_tree("output")