python-otp4gb

This is a port of the OTP4GB tool written by Tom Forth at Open Innovations.

Preparation and dependencies

Make sure you have a working Java environment, and have Maven installed.

Run the setup script to install dependencies.

python setup.py

This will run maven and download the osmconvert64 executable for Windows.

You will also need to install a series of python libraries, which are captured in requirements files. There are two: one for conda and one for pip.

Populate the assets folder with a copy of the OSM map and any GTFS files you want to include in the analysis.

Running

You will need to create a directory to set up the tool. This needs the following structure:

• config.yml A file with config for the run. A sample is provided in the root of this project.
• input directory containing the latest osm download
• input\gtfs directory containing the GTFS files that are to be included in the OTP routing graph

prepare.py

This script prepares the OTP directory.

Usage:

prepare.py [-F] [-d <date>] [-b <bounds name>] <directory>

command line options

• -b, --bounds Specify bounds
• -d, --date Specify date for filtering
• -F, --force Force overwrite of existing files

Environment config

OSMCONVERT_DOCKER - if set, will use a dockerised version of osmconvert (useful to run on macos, where there appears to be no native osmconvert)

process.py

Having created the OTP directory, the process script runs a batch as defined in the config file.

Usage:

python process.py <directory>

The <directory> is the one prepared by the prepare script.

No command line options as yet.

This creates a CSV file of the Travel Time matrix of all included MSOAs. In addition, it has creates travel isochrone GeoJSON files in the isochrones subdirectory.