Skip to content

Python module to implement BNA logic in a PostGIS database

License

Notifications You must be signed in to change notification settings

tooledesign/pybna

Repository files navigation

pyBNA

Introduction

Python module to implement BNA logic on a PostGIS database. Uses a configuration file for customizability. The current implementation requires a user to run the old-style BNA on the command line to import datasets. This library can build the routeable network, process the connectivity routing, and calculate scores.

Bicycle Network Analysis (BNA) is a way to measure how effectively people can connect to destinations via bike on comfortable, "low-stress" routes. More info on the methodology at the What is the BNA? page.

Example of BNA results

Requirements

pyBNA is tested with Python 3.6. The following libraries are required:

  • psycopg2
  • tqdm
  • pyyaml
  • geopandas
  • munch
  • overpass
  • osmnx
  • xlrd

A requirements.txt file is provided for convenience. You can install these via pip:

pip install -r requirements.txt

*Special note for Windows users: Installing geopandas can be a real pain. If you don't already have a working version of Geopandas, we suggest following this excellent guide from Geoff Boeing.

The osmium package is also required for parsing OSM destinations from an .osm extract, but if you're downloading destinations directly from OSM you can safely skip this dependency.

tl;dr (Simple run)

The most simple BNA run, using stock datasets and no customization, can be completed in a few easy steps. The following assumes you already have a database running named "bna" on the local machine.

import pybna

# imports
i = pybna.Importer()
i.import_boundary('/path/to/your/boundary/file')
i.import_census_blocks(fips=16)
i.import_census_jobs("received.neighborhood_census_block_jobs",state="ID")
i.import_osm_network()
i.import_osm_destinations()

# stress
s = pybna.Stress()
s.segment_stress()
s.crossing_stress()

# connectivity
bna = pybna.pyBNA()
bna.calculate_connectivity()

# scores
bna.score("myschema.my_scores_table")
bna.aggregate("myschema.my_aggregate_score_table")

Importing data

pyBNA includes a workflow to import data from publicly available sources (for the United States, at least). Automatic import of demographic data relies on US Census 2010 data. Street network and bicycle facility data is imported from OpenStreetMap.

For more guidance on the import process, see our import instructions.

Traffic Stress

pyBNA has a module that can calculate traffic stress based on roadway characteristics in your roadway data. There's also the ability to apply assumptions for locations where data is not available.

For more information about the traffic stress module, see the traffic stress instructions.

Getting started

First, import pybna and create a pyBNA object by pointing it to the config file.

import pybna
bna = pybna.pyBNA(config="/path/to/my/config.yaml")

Next, you can calculate the connectivity with

bna.calculate_connectivity()

Lastly, you can generate block-level scores with

bna.score("myschema.my_scores_table")

and aggregate scores for the entire study area with

bna.aggregate("myschema.my_aggregate_score_table")

Configuration file

Most options in pyBNA are managed using a configuration file. This file is passed as an argument when creating the pyBNA object and tells pyBNA important things about your data and the assumptions you want to make in the analysis. The configuration file is written using YAML.

There's more information about the configuration file here

Travel sheds

Once you've completed the connectivity analysis, you can develop a low/high stress travel shed for any census block with

bna.travel_sheds([list, of, block, ids, here], my_travel_shed_table)

Scenarios

pyBNA includes the capability to run scenarios as a way to visualize the connectivity impacts of a project or group of projects. More information on scenarios is available here.