Information extracted from GTFS transit feeds to use with RAPTOR algorithm.
Each folder corresponding to the public transport data of a location, and consists of the following files:
in_hubs.gr.gzout_hubs.gr.gzstop_routes.csv.gzstop_times.csv.gztransfers.csv.gztrips.csv.gzwalking_graph.gr.gz
All the files are in gzip format, each line of the decompressed files can be considered as a row of a table,
consisting of the same number of columns, with the following differences:
*.csv.gzfiles: the elements in each row are separated by a comma, and each file has a header,*.gr.gzfiles: the elements in each row are separated by a space, and there is no header.
Header: stop_id,route_id
This file contains the information about the routes serving a stop in the timetable. The rows s,r indicate that
the stop s is served by the route r.
Header: route_id,trip_id
This file contains the information about the trips belong to a route. The rows r,t indicate that the trip t belongs
to the route r. Every trip of a route has the same stop sequence, which can be found using stop_times.csv.gz.
Moreover, the rows are sorted by the id of the routes. And in each group of trips belong to the same
route, the trips are sorted so that the departure times at each stop of a trip is after that of the
previous trip at the same stop. For example, if we organize the departure times of a route into a table as below,
where s[0], s[1], s[2] are the stops, t[0], t[1], t[2] are the trips after being ordered, dep[i, j] is the
departure time of the trip t[i] at stop s[j], then dep[i+1, j] >= dep[i, j] for all i, j.
s[0] |
s[1] |
s[2] |
|
|---|---|---|---|
t[0] |
dep[0, 0] |
dep[0, 1] |
dep[0, 2] |
t[1] |
dep[1, 0] |
dep[1, 1] |
dep[1, 2] |
t[2] |
dep[2, 0] |
dep[2, 1] |
dep[2, 2] |
Header: trip_id,arrival_time,departure_time,stop_id,stop_sequence
trip_id/stop_id: The id of the trip/stop to which the event represented by a row belongs.arrival_time/departure_time: The arrival/departure times at the stopstop_idfor the triptrip_id. The unit is second, counted from midnight.stop_sequence: Thestop_sequencefield identifies the order of the stops for a particular trip. The values forstop_sequenceare non-negative integers, and increasing along the trip.
Header: from_stop_id,to_stop_id,min_transfer_time
from_stop_id: Thefrom_stop_idfield contains a stop ID that identifies a stop where a connection between routes begins.to_stop_id: Theto_stop_idfield contains a stop ID that identifies a stop where a connection between routes ends.min_transfer_time: Themin_transfer_timefield defines the amount of time in seconds to transfer in a connection.
This file represents the unrestricted walking graph for each location. The graph is directed and weighted. Each arc of the graph is represented by a line in the decompressed file.
The format of each line is s t d, where s/t are the id of the source/target. d is the walking time
from s to t, which is an integer with the unit of 1 second.
This file contains the information about the in-hubs of each stop in the timetable.
The format of each line is h s d, where h is the id of the in-hub, which is a node in the walking graph. h is an
in-hub of the stop s, and s is the shortest-path distance from h to s.
This file contains the information about the out-hubs of each stop in the timetable.
The format of each line is s h d, where h is the id of the out-hub, which is a node in the walking graph. h is an
out-hub of the stop s, and s is the shortest-path distance from s to h.
The in-hubs and out-hubs satisfy the following covering property:
where H⁺(u) is the set of out-hubs of u and H⁻(v) is the set of in-hubs of v.
| London | Paris | Switzerland | |
|---|---|---|---|
| routes | 1622 | 1973 | 13930 |
| trips | 122047 | 78757 | 369744 |
| stops | 19746 | 23519 | 25427 |
| events | 4695285 | 1915253 | 4740869 |
| transfers | 64756 | 338772 | 12838 |
| nodes | 280765 | 531720 | 604230 |
| edges | 857516 | 1666740 | 1876947 |
| avg in hubs | 69.56 | 118.98 | 78.87 |
| avg out hubs | 68.99 | 117.79 | 78.87 |