Abstract
Autonomous Vehicles (AVs) could have diverse impacts. On the one hand, autonomous demand-responsive transport (DRT) or fixed-route autonomous services may reduce operating costs and increase the share of trips by collective modes of transport. On the other hand, private AVs may increase car attractiveness by decreasing travel-time disutility, improving road capacity, and widening car access. In many countries, rural areas exhibit low accessibility and car-dependent travel patterns. Despite such challenges and the potential of AVs, their impacts in such contexts have received limited attention, leaving gaps in our understanding of future rural mobility scenarios. This work aims to evaluate the combined effects of automation on accessibility and mobility patterns in rural areas. To this end, we implement an agent-based model (MATSim) simulating daily activities and travels in the remote valley of Mühlwald (South Tyrol, Italy). We test scenarios where the existing public transport system is upgraded, replaced or complemented by autonomous DRT. In addition, we assume a progressive (and slow) diffusion of private AVs, by considering variable car travel-time disutility, road capacity and access to private AVs for people without a driving licence. Results suggest that, under a fixed budget constraint, coupling the existing bus line with autonomous DRT could improve accessibility, decrease vehicle kilometres travelled, and car modal share. Nevertheless, these benefits could be undermined by the progressive diffusion of private AVs. Moreover, a complete replacement of traditional public transport with autonomous DRT could be ineffective under the same operating costs, calling for higher investments to implement such solutions.