The Impact of Multitasking on Transport Mode Choice in Autonomous Vehicle Age

Travelers on board of transport modes conduct active and passive activities to mitigate the negative impact of traveling. Multitasking on board of conventional transport modes is studied by several researchers, but limited efforts are focusing on multitasking in case of autonomous vehicles (AVs) and shared autonomous vehicles (SAVs). In this paper, the effects of multitasking on behavior of travelers on board of conventional cars (cars), AVs, and SAVs are analyzed. Furthermore, finding the impact of several factors on the travel time, the acceptability of SAVs, and conducting onboard activities are assessed. The study considers solely main trip purposes within urban areas. A stated preference (SP) survey is distributed in Budapest, Hungary, and 276 participants are collected. An SP includes discrete choice experiment (DCE) that is designed to mimic the realistic situation when AV is on the market. The DCE considers the attributes and attributes levels of the alternatives where rationality is maintained in the design. A transport mode choice model, which includes several variables influencing the choice of a transport mode, is developed. In addition, an SP includes Likert scale and sociodemographic sections. Likert scale and exploratory factor analysis (EPA) are used to understand the impact of some factors on the travel time, the acceptability of SAVs as transport modes, and conducting onboard activities. The multinomial logit (MNL) model is applied, where a transport mode choice model of cars, AVs, and SAVs is developed. The results of the developed model show that travelers are willing to choose AVs over cars, and SAVs over AVs. Moreover, travelers with high income are more willing to use AVs over SAVs and more likely to use cars than SAVs. Besides, people from the older age group prefer using SAVs more than other age groups. The results demonstrate the probability of selecting a transport mode with active activities on board is larger than the probability of choosing a transport mode with passive activities. Besides, the findings of EFA and Likert scale analyses demonstrate that the waiting time has the largest negative effect on the travel time, seat availability affects the conduction of onboard activities, and the internal design of SAVs influences the use of SAVs as transport modes. The results of current research can be beneficial to transport planners, transport operators, and vehicle manufacturers.