2

I'm reading Microsoft's white paper "Round-Based Public Transit Routing" (the RAPTOR algorithm). In this algorithm, there are some definitions that I don't understand:

About trip:

trip represents a sequence of stops a specific vehicle (train, bus, subway, . . . ) visits along a line

About route:

Each route consists of the trips that share the same sequence of stops.

So, if I understand, trip means a "bus route" and a route is collection of trips to make a path? If my guess is true, why does it state:

Typically there are more routes than trips

And if a route is just a collection of trips, how can I build all routes?

Rob Bednark
  • 103
  • 4
Trần Kim Dự
  • 121
  • 3

2 Answers2

5

Trip represents a sequence of stops plus associated arrival and departure times on each stop.

Group all trips by their sequence of stops, ignoring the time information, then each group of trips sharing the same sequence of stops forms a route.

eg. for air traveling, each flight is a trip, while each airline is a route, grouping all flights flying on the same airline.

What I saw in the paper is

Typically, there are many more trips than routes.

Which is obvious since there is an one-to-many mapping between route and trips.

Terence Hang
  • 789
  • 4
  • 7
1

Be careful, as the definition of route given in this paper does not seems to match the definition given in other standards (namely, GTFS). In GTFS a route is a somehow arbitrary collection of trips, but they do not have to share the exact same pattern of stop sequence (and they usually don't).

What the paper describes for a route is sometimes called a route pattern — a set of trips sharing an exact identical stop sequence.

Laurent Grégoire
  • 143
  • 6