Search

Abstract This paper, for the first time, presents comparable projections of travel behaviour impacts of the introduction of autonomous vehicles (AVs) into the private car fleet for two countries, namely the USA and Germany. The focus is on fully autonomous vehicles (AVs) which allow drivers to engage in other activities enroute. Two 2035 scenarios – a trend scenario and an extreme scenario – are presented for both study countries. For these projections, we combine a vehicle technology diffusion model and an aspatial travel demand model. Factors that influence AV impact in the behavioural model are mainly new automobile user groups, e.g. travellers with mobility impairments, and altered generalized costs of travel, e.g. due to a lower value of travel time savings for car travel. The results indicate that AV penetrations rates might be higher in Germany (10% or 38% respectively) than in the USA (8% or 29% respectively) due to a higher share of luxury cars and quicker fleet turnover. On the contrary, the increase of vehicle mileage induced by AVs is not higher in Germany (+2.4% or +8.6% respectively) than in the USA (+3.4% or +8.6% respectively). This is mainly due to the lack of mode alternatives and lower fuel costs resulting in a higher share of travel times among the total generalized costs of travel in the USA. These results clearly indicate that context factors shaped by national policy will influence AV adoption and impact on travel demand changes. Based on these results the paper draws policy recommendations which will help to harness the advantages of AVs while avoiding their negative consequences. [ABSTRACT FROM AUTHOR]

Highlights • We evaluated fuel consumption impacts of autonomous vehicles in United States. • The proposed data-rich analytical framework considers dynamics in fleet turnover and fuel efficiency. • We utilized a network activity database contains detailed information of national road links. • Impacts of automation on fuel consumption are quite wide-ranging, −45% to +30%. • Automation on urban road has larger potential on fuel savings than highway automation. Abstract Autonomous vehicles are drawing significant attention from governments, manufacturers and consumers. Experts predict them to be the primary means of transportation by the middle of this century. Recent literature shows that vehicle automation has the potential to alter traffic patterns, vehicle ownership, and land use, which may affect fuel consumption from the transportation sector. In this paper, we developed a data-rich analytical framework to quantify system-wide fuel impacts of automation in the United States by integrating (1) a dynamic vehicle sales, stock, and usage model, (2) an historical transportation network-level vehicle miles traveled (VMT)/vehicle activity database, and (3) estimates of automation's impacts on fuel efficiency and travel demand. The vehicle model considers dynamics in vehicle fleet turnover and fuel efficiency improvements of conventional and advanced vehicle fleet. The network activity database contains VMT, free-flow speeds, and historical speeds of road links that can help us accurately identify fuel-savings opportunities of automation. Based on the model setup and assumptions, we found that the impacts of automation on fuel consumption are quite wide-ranging—with the potential to reduce fuel consumption by 45% in our "Optimistic" case or increase it by 30% in our "Pessimistic" case. Second, implementing automation on urban roads could potentially result in larger fuel savings compared with highway automation because of the driving features of urban roads. Through scenario analysis, we showed that the proposed framework can be used for refined assessments as better data on vehicle-level fuel efficiency and travel demand impacts of automation become available. [ABSTRACT FROM AUTHOR]

• This study explores the 2017 National Household Travel Survey data. • This study focuses on the correlates of vehicle ownership among young Americans. • Spatial approach is used to reveal the vehicle ownership correlates that vary across the country. • Young adults in the Northeast states have extra-lower likelihoods of owing a vehicle. • Vehicle ownership among young American is less sensitive to household income. This study uses the National Household Travel Survey (NHTS) data to investigate the most recent correlates of vehicle ownership among young Americans. This study performs a spatial analysis to examine the potentially non-stationary relationships between sociodemographic factors and vehicle ownership. Consistent with previous studies, modeling results from this study showed that young Americans are more likely to be carless than older adults. The spatial analysis answers the research question – in which regions(s) young Americans are even less likely to have a car. The results highlighted the Northeast states for the young American's extra-lower vehicle ownership if the influences of all other factors are held constant. The cost of living and availability of transportation alternatives are possible reasons. Further, this study built separate models for young adults (25–34 years old) and three older age groups. The vehicle ownership correlates within the young adults are found to be generally consistent with the correlates among all adults. Among young adults, vehicle ownership is still significantly related to their gender, educational attainment, employment status, household characteristics, and travel demand. However, young adults' vehicle ownership seems to be less sensitive to household income than mid-age adults' (35–44 years old), perhaps because young people may not perceive financial stress such as child support and mortgage. This study contributes by using a spatial analysis approach to reveal the non-stationary correlates of vehicle ownership. This approach is useful for future travel behavior research and transportation policy considering the spatial heterogeneity. [ABSTRACT FROM AUTHOR]

The article presents a comment on an article by L. Evans related to motor vehicle injury in the United States that appeared in a previous issue of the American Journal of Public Health. According to the author of this article, Evans' article on driver behavior includes mistaken assertions. He says that programs that change driver behavior are more effective than those that change vehicles and environment. In the opinion of the author, his study of vehicle modifications, belt use laws, and alcohol reductions indicates otherwise. When the coefficients on model year--specific changes in deaths per mile are multiplied by the relevant numbers of model years, and when the coefficients for changes in belt use and alcohol are multiplied by changes in those factors, the results indicate that far more reductions are due to vehicle modifications. Evans' claim that the differences in death rates among model years is attributable to vehicle age. Anti-regulation economists, to counter the author's original studies of regulation, according to him, created this myth. In his study, vehicle age was controlled in the regression.