Estimating and Implementing a Vehicle-Type Model in an Activity-Based Travel Model Framework

Greg Erhardt, David Hensle, Mark Bradley, Michelle Imarah, Joel Freedman, Max Gardner, Ben Stabler

Research output: Contribution to journalArticlepeer-review


Characteristics of household vehicles can influence daily travel behavior and can vary greatly by vehicle type. Vehicle type is defined here as a combination of body type, fuel type, and age. Using data compiled primarily from the National Household Travel Survey, a multinomial logit model was developed to predict vehicle type based on characteristics of the household that owns the vehicle. The model was then implemented for the San Francisco Bay Area in the ActivitySim activity-based modeling framework and validated against observed data. The ActivitySim project’s goal is to create and maintain advanced, open-source, activity-based travel behavior modeling software, based on best software development practices, for distribution to the public, free of charge. A naïve 2030 scenario was analyzed to show model response to changes in the vehicle fleet.

Original languageEnglish
JournalTransportation Research Record
StateAccepted/In press - 2024

Bibliographical note

Publisher Copyright:
© The Author(s) 2024.


  • activity based models
  • choice models
  • forecasts/forecasting
  • general
  • mode choice
  • planning and analysis
  • transportation demand forecasting

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Mechanical Engineering


Dive into the research topics of 'Estimating and Implementing a Vehicle-Type Model in an Activity-Based Travel Model Framework'. Together they form a unique fingerprint.

Cite this