Door-to-door Routing Analyst - DORA

Door-to-door Routing Analyst or DORA for short is an open source multi modal routing tool, developed in 2018. It uses the door-to-door approach when retrieving travel times where the whole travel chain is taken into account. It can be used to route car, cycling and walking routes and is able to read any road network setup in a database with pgRouting (v2.3.2) extension.

For privatecar travel times DORA makes use of travel times that are based on Digiroad data set speed limits that have been modified more realistic in the Helsinki region. Times of the road segments are made more realistic by adding crossroad impedances for different road classes. Impedances are based on floating car measurements conducted in the Capital Region of Helsinki (data from the Helsinki City Planning Department and Helsinki Regional Transport Authority HRT).

For cycling travel times DORA is using road network data from Helsinki that has been modified to be suitable for bicycle routing. The road network has been MetropAccess-CyclingNetwork, which is based on the Digiroad data (Digiroad K), developed by the Finnish Transport Agency, but where the speeds along the road segments have been modified more realistic. To make the speeds more realistic, we have used Strava-Metro sports application data from the Helsinki region from 2016 as well as the bike sharing data from Helsinki from 2017. The routing tool has not used segments where cycling is prohibited.

DORA is implemented with the open source database software PostgreSQL and its spatial extension PostGIS and it's making use of the routing library pgRouting (tool: pgr_dijkstraCost). This makes DORA completely open source tool being also available openly.

Installation

    $ git clone git@github.com:DigitalGeographyLab/DORA.git
    $ cd DORA

Add the system path where the project is located in the module main.py.

sys.path.append('/dgl/codes/DORA/')

Libraries required

• OWSLib

    $ cd /opt/anaconda3/bin/
    $ ./conda install -c conda-forge owslib
    $ source ./activate
    $ pip install --upgrade owslib

For windows, also required: Microsoft Visual C++ 9.0 is required. Get it from here.

• nvector (not available in conda repositories)

    $ pip install nvector

• pyproj

    $ conda install -c conda-forge pyproj

• psycopg2

    $ conda install -c conda-forge psycopg2

• geopandas

    $ conda install -c conda-forge geopandas

• joblib

    $ conda install -c anaconda joblib

• psutil

    $ conda install -c anaconda psutil

Run

startPointsFolder: Folder containing a set of geojson files with the origin/start Point geometries. endPointsFolder: Folder containing a set of geojson files with the destination/target Point geometries.

    user@/dgl/codes/DORA$$ python -m src.main -s <../startPointsFolder> -e <../endPointsFolder> -o <../outputFolder> -t BICYCLE -c BICYCLE_FAST_TIME --summary --is_entry_list

-s: Path to the Geojson file containing the set of origin points (or the directory containing a set of Geojsons).

-e: Path to the Geojson file containing the set of target points (or the directory containing a set of Geojsons).

-o: Path where store the output data.

-t: Flag to choose the transport mode for the data analysis [PRIVATE_CAR, BICYCLE].

-c: The impedance/cost attribute to calculate the shortest path.

--route: Store in the output folder the geojson files with the fastest route LineString features.

--summary: Store in the output folder the csv files containing the fastest travel time summary per each pair of entry points.

--is_entry_list: Define if the -s and -e are folders paths and not file paths.

Impedance/Cost -c attribute accepted values:

• DISTANCE (Both PRIVATE_CAR and BICYCLE)
• RUSH_HOUR_DELAY (PRIVATE_CAR only)
• MIDDAY_DELAY_TIME (PRIVATE_CAR only)
• DAY_AVG_DELAY_TIME (PRIVATE_CAR only)
• SPEED_LIMIT_TIME (PRIVATE_CAR only)
• BICYCLE_FAST_TIME (BICYCLE only)
• BICYCLE_SLOW_TIME (BICYCLE only)

Input testPoints.geojson is in the format:

{
  "type": "FeatureCollection",
  "features": [
    {
      "type": "Feature",
      "geometry": {
        "type": "MultiPoint",
        "coordinates": [
          [8443150.541380882, 2770625.5047027366],
          [8436436.62334174, 2776131.3593964037]
        ]
      },
      "properties": {
        "title": "Pair 1",
        "icon": "monument"
      }
    },
    .
    .
    .
    {
      "type": "Feature",
      "geometry": {
        "type": "MultiPoint",
        "coordinates": [
          [8452765.483509162, 2783396.1614870545],
          [8445529.046721973, 2778867.5661432995]
        ]
      },
      "properties": {
        "title": "Pair 2",
        "icon": "monument"
      }
    }
  ]
}

Additonal Layers

You are allowed to add new attributes coming from a polygon layer and attach them to the selected points (start and end point to calculate the shortpath).

See: Additional Layers Operations

Modifying or extending transport modes

The tool uses a hierarchical structure to support different travel modes. By default, the supported travel modes are Private Car and Bicycle. To extend the functionality it is needed either to modify the existing transport modes specifications or by adding a new specification of the abstraction AbstractTransportMode and call it from doraInit.py.