Vehicle chooser model

The vehicle chooser model in POLARIS determines the number of vehicles in the households based on a random choice on the underlying vehicle type distribution for a given zip code. In addition to that, vehicle reduction models are developed to constrain vehicles in the hosuehold for a future forecast, e,g. in the presence of shared autonomous vehicles. Two vehicle reduction models are developed, for single-vehicle households and multi-vehicle households, using data from a web-based survey for targeted population groups. The models estimate peoples’ willingness to relinquish a household vehicle from the existing vehicle fleet, both in single- and multi-vehicle households, thus reducing their household vehicle ownership level by one, in the presence of shared autonomous vehicles. Their willingness to relinquish a vehicle is collected as ordinal responses, which is modeled using a random parameter ordered probit formulation.

References

Menon, N., Barbour, N., Zhang, Y., Pinjari, A. R., & Mannering, F. (2019). Shared autonomous vehicles and their potential impacts on household vehicle ownership: An exploratory empirical assessment. International Journal of Sustainable Transportation, 13(2), 111-122.