Preference heterogeneity in mode choice for car-sharing and shared automated vehicles

Membership-based short-term car rental-more popularly referred to in the marketplace and popular media as "car-sharing"-has been expanding rapidly worldwide. The advent of self-driving vehicles is likely to facilitate the growth of car-sharing by addressing a couple of the current barriers, including limited dedicated parking and non-competitive access times. Despite the growing importance of this topic, however, the patterns emerging in the market penetration of automated vehicles in car-sharing programs have not received much attention in the literature because of the relative newness of the automated vehicle technologies enabling them. Towards addressing this research gap in this increasingly important area, this paper presents the results of an Australian survey with a focus on consumer preferences towards car-sharing. A stated preference (SP) methodology was adopted to elicit consumers' valuation of specified mode-choice related factors. In particular, vehicle self-driving capability, a factor rarely examined in the literature, was provided as an option to participants in the SP survey. To increase the realness of the experiment, SP tasks were pivoted from respondents' most recent trips. The travel preferences data were analyzed using a random parameter (mixed) logit model. To explore preference heterogeneity, socio-demographic and other factors were interacted with alternative-specific attributes, and their influences on marginal utilities were extracted using simulation analysis. Preference heterogeneity across individuals towards shared automated vehicles (SAVs) was identified. Consistent with the literature, consumers' experience of using car-sharing appears to have significant influence on household mode choices, increasing the probability of using a diversified mode tools (e.g., TWS and Taxi) and decreasing the likelihood of choosing to use privately owned travel tools, such as private car. Although it has been argued in the literature that females, non-drivers, and the elderly are the most likely to benefit from SAVs, the results of this study reveal that these user groups in fact hold negative opinions about SAVs. The findings highlight the challenges policymakers may encounter in maximizing the social and individual benefits of SAVs.