Transit directions at global scale

TL;DR

This paper presents a novel, scalable journey planning system for integrated ground and air public transport at a continental scale, capable of handling irregular networks and flexible date searches, with a prototype comparable to Google Transit.

Contribution

It introduces a new algorithm and system for large-scale, integrated transit planning that effectively manages irregular networks and long-distance trips.

Findings

System performs comparably to Google Transit in the US.

Handles irregular and poorly structured networks effectively.

Supports flexible date searches for long-distance trips.

Abstract

A novel approach to integrated ground and air public transport journey planning, operating at continent scale. Flexible date search, prerequisite for long distance trips given their typical low and irregular service frequencies, is core functionality. The algorithm is especially suited for irregular and poorly structured networks. Almost all of the described functionality is implemented in a working prototype. Using ground transport only, the system is on par with Google Transit on random country-wide trips in the US.