A routing engine for public transit implementing Scalable transfer patterns. Not (yet) meant to be a fully featured routing engine, but rather an exploration of the algorithms out there.

Goals

• Performance & Efficiency
  • Preprocessing on Baden-Württemberg:
    • < 3h on laptop (all cores)
    • < 2GB peak RAM
    • < 5GB disk in addition to dataset
  • Queries on Baden-Württemberg:
    • < 50ms
    • < 500MB auxilary RAM per query (in addition to precomputed data)
    • no extra disk space