hope-graph-builder (deprecated)

This repository is deprecated as a separate project - all functionality have been integrated into hope-green-path-server.

This repository contains utilities for importing and processing OpenStreetMap based street network graphs created and exported with OpenTripPlanner. The graphs will be used at least in hope-green-path-server which is a route planner for walking and cycling that finds routes with less traffic noise and air pollution.

Features

• otp_graph_import.py
  • Import graph data to igraph from CSV files exported by OTP
  • Remove private, unwalkable and street segments not suitable for biking from a graph
  • Decompose graph and remove unconnected edges & nodes
  • Create a subset of the graph for Helsinki Metropolitan Area
  • Export raw and processed graph features to GeoPackages for debugging
• noise_data_preprocessing.py
  • Preprocess noise data from different sources to common schema
• noise_graph_join.py
  • Join environmental noise data to graph features to enable noise exposure based routing
  • Interpolate noise values for edges missing them (on municipal boundaries)
• green_view_join_v1.py
  • Join street level Green View Index (GVI) values from GVI point data and land cover layers
• graph_export.py
  • Calculate biking impedances ("adjusted lengths") by bike safety factors
  • Finalize graph for Green Paths route planner by exporting only relevant attributes

Materials

• SYKE - Traffic noise modelling data from Helsinki urban region
• Traffic noise zones in Helsinki 2017
• OpenStreetMap
• Green View Index (GVI) point data
• Land cover data

Tech

• Python 3.8
• igraph
• GeoPandas
• Shapely

Installation

$ git clone git@github.com:DigitalGeographyLab/hope-graph-builder.git
$ cd hope-graph-builder/src
$ conda env create -f env_graph_tools.yml

Running the tests

$ conda activate graph-tools
$ cd src/test
$ python -m pytest green_view_join_v1_test.py -v
$ python otp_graph_import_test.py
$ python noise_graph_join_test.py