Skip to content

Travel Model Two

Travel Model Two, currently under development, is an important extension of Travel Model One. Fundamental to the foundation of Travel Model Two is the use of a wide ranging set of on-board transit rider surveys as well as the California Household travel survey, which obtained a statistical larger sample in the nine-county Bay Area.

Changes from Travel Model One

Important travel behavior enhancements in Travel Model Two include:

  • A much more detailed spatial representation of transportation system supply including an accurate all-streets network for entire 9-county Bay Area, pedestrian paths\sidewalks from OpenStreetMap, bicycle facilities from MTC’s BikeMapper, and transit networks from MTC’s RTD network
  • Land-use and demographic forecast integration with Bay Area UrbanSim Two represented at a 40,000 micro-analysis zone (MAZ) level
  • Detailed transit access/egress based on actual origin/destinations at the MAZ level considering boarding and alighting at specific transit stops allowing for a more accurate representation of walk times
  • More detailed temporal resolution using half-hourly time windows compared to hourly time windows in Travel Model One
  • The effects of transit capacity and crowding
  • More detailed auto assignments, most notably with the loading of short trips to local streets
  • The inclusion of Taxis and Transportation Network Companies (TNCs) such as Uber and Lyft as a mode choice option
  • Representation of Automated Vehicles

Versions

  • TM2.0: Initial TM2 with Cube, CTRAMP core, 3-zone system. In use by TAM (Marin)
  • TM2.1: TM2 with transit CCR implemented in Emme (uses Cube and Emme), CTRAMP core. Anticipated release: Summer 2022. This work is being performed in the branch transit-ccr -- these docs are for this version
  • TM2.2: TM2 with Emme only (No Cube), CTRAMP core. This work is being performed in the tm2py repository
  • TM2.3: TM2 with Emme only and ActivitySim core.

References