An examination of externalities associated with alternative routing criteria and third-party platforms.
Jupyter Notebook Python
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
community_externalities
mapping_platforms
preprocessing
routelevel_externalities
utils
LICENSE
README.md

README.md

route-externalities

An examination of externalities associated with alternative routing criteria and third-party platforms.

Origin-Destination Pair Preprocessing

  1. Choose geographic area and download OSM-based road network from Mapzen
  2. Generate grid covering city area
    • preprocessing/grid_creation.py
  3. Generate od-pairs for a grid
    • preprocessing/generate_od_pairs.py

Scenic Routing Preprocessing

  1. Gather Flickr and Twitter data for study region
  2. Score each grid cell based on Empath analysis of Flickr/Twitter data
    • preprocessing/generate_grid_scores.py

Safety Routing Preprocessing

  1. Generate mapping of grid cells to census tracts (or equivalent) for city
    • utils/gridcell_to_ct_mapping.py
  2. Gather crime data for city
  3. Aggregate grid data to census tract equivalent and set threshold for unsafe areas
    • utils/aggregate_grid_values_to_ct.py

Analysis Steps

  1. Get Google and Mapquest routes for od-pairs
    • mapping_platforms/get_routes.py
  2. Get GraphHopper routes for od-pairs
  3. Match Google and Mapquest routes to GraphHopper network for final metrics
  4. Merge together GraphHopper metrics and original data
    • utils/merge_api_gh_results.py
  5. Route-level analyses between all routes
    • routelevel-externalities/routelevel_externalities.ipynb
  6. Compute signiificant differences with GraphHopper fastest routes
    • utils/significant_diff_segments.py
  7. Run community-level analyses
    • community_externalities/calc_hmi.py

Example Data

  • Load in statsobjs_routeexternalities.pkl in routelevel_externalities/routelevel_externalities.ipynb