Where do you live and what do you drive: Built-environmental and spatial effects on vehicle type choice and vehicle use

Abstract

The carbon footprint of daily travel for an individual household is based on the types of vehicles that the household owns, their fuel efficiency, and the number of miles traveled (Federal Highway Administration, 2009). Therefore, the types and usage of vehicles a household owns are informational to overall sustainable planning. This study develops a multinomial logit model for vehicle type choice and a spatial autoregressive (SAR) model for vehicle use for Ohio households using the most recent Household Travel Survey (HTS) collected in the Cincinnati metropolitan area. The developed vehicle choice model identifies the effects of the built environment characteristics on vehicle type based on vehicle size, controlling for socio-economic attributes. People residing in neighborhoods characterized with higher densities in population, employment, and intersection and with fewer single detached housing are more likely to own smaller vehicles as compared to the ones living in suburban areas. A good transit accessibility would decrease the probability of owning pickup trucks. The SAR model reveals how the vehicle a household owns is used is still slightly positively associated with his/her neighbour’s vehicle use after controlling for the effects of the built environment and socio-economics. The models will enable decision makers to estimate potential changes in vehicle fleet composition under changing built environment scenarios and neighborhood culture in using private vehicles. Although the model results and estimated coefficients may be different for different urban areas, the methodology is transferable to other communities, locally and nationally.

Keywords:

• Cluster analysis
• multinomial logit
• spatial autoregression
• vehicle type choice
• vehicle use

Acknowledgements

The authors would like to particularly thank the project manager Ms. Rebekah Anderson of ODOT for her time and efforts for providing the datasets.

Additional information

Funding

This research was supported in part by Ohio Department of Transportation’s OPREP grant.