Example dataset for taxi simulatoin

This is an example dataset, containing pre-processed taxi trips from Manhattan, New York City, to be used with the simulation code available at https://github.com/dkondor/taxi_simulation

Trip data originally comes from the NYC Taxi and Limousine Commission:
https://www1.nyc.gov/site/tlc/about/tlc-trip-record-data.page

Road network data comes from OpenStreetMap, available under the Open Data Commons Open Database License

The following files are included in this dataset:

1. Road network data:
    NYC_nodes.csv includes a list nodes with a node ID, longitude and latitude. Node IDs range between 1 and 4091 (inclusive).
    NYC_segments.csv includes a list of directed edges between the above nodes, with distances along the edges given in meters.

2. Trip data:
The files nytrips_{day}.bz2 (with {day} ranging from 14975 to 15339) contain trips happening in each day in 2011. These are CSV files compressed with bzip2, without header. The columns include the taxi ID (arbitrary numeric ID), trip start and end timestamp (UNIX timestamp without time zone), trip start and end node ID (corresponding to one of the nodes in the above network files).

3. Trip start distribution:
trip_start_dist.dat is an example distribution of trip start locations, i.e. node IDs along with the number of trips starting there in a one week period.

4. Travel times and indexes:
nyc_travel_times.zip contains a set of binary matrices containing travel times and an index for these that is used by the simulations; these were generated from the trips according to the methodology described in the following paper:
Santi, P., Resta, G., Szell, M., Sobolevsky, S., Strogatz, S. H., & Ratti, C. (2014). Quantifying the benefits of vehicle pooling with shareability networks. PNAS, 111(37), 13290–13294. https://doi.org/10.1073/pnas.1403657111