ALTO: a Germany-wide long-distance passenger demand model

The professorship has developed a model for long-distance travel, over 40 kilometers, in Germany.

The model is an agent-based and trip-based travel demand model consisting of trip generation, destination choice, mode choice, and departure time choice. Three trip purposes, namely business, leisure, and other, are modeled in the trip generation module. In the destination choice module, 11717 domestic zones and 162 external zones are included. The 11717 domestic TAZs are at the scale of municiplities (Gemeinde in German) and the largest 14 cities (Kreisfreiestadt in German) are at the scale of city district (Stadtbezirke in German). As to the mode choice model, we included auto, long-distance rail, long-distance bus, and air as a multiniomial logit model. Last but not least, the departure time module is a simulation of the observed distribution.

Trip assignment is not included in this model, but a plan file with activities and legs, which is compatible to MATSim is generated along with the model run. Afterward, the demand can be easily simulated in MATSim. The model can be accessed at https://github.com/msmobility/alto