Increasing accessibility and balancing its distribution across space and social groups are two fundamental goals to make transport more sustainable and equitable. In the next decades, autonomous vehicles (AVs) could significantly transform the transport system, influencing accessibility and transport equity. In particular, depending on the assumed features of AVs (e.g., private or collective) and the considered spatial, social, and regulative context (e.g., rural or urban areas), impacts may be very different. Nevertheless, research in this field is still limited, and the relationship between AV assumptions and accessibility impacts is still partially unclear. This paper aims to provide a framework of the key and emerging aspects related to the implications of AVs for accessibility and transport equity. To set this framework, we perform an analysis of the scientific literature based on a conceptual model describing the implications of AVs for the distribution of accessibility across space and social groups. We recognize four main expected impacts of AVs on accessibility: (1) accessibility polarization, (2) accessibility sprawl, (3) exacerbation of social accessibility inequities, and (4) alleviation of social accessibility inequities. These impacts are described and analyzed in relation to the main AV assumptions expected to trigger them through different mechanisms. Based on the results, some recommendations for future studies intending to focus on the relation between AVs, accessibility, and transport equity are provided.
The Space–Time Accessibility (STA) model is broadly used to measure person-based accessibility based on the space, time, and transport constraints experienced at the individual level in connection to the actual modal choices of observed individuals. In this paper, we propose to adjust the STA model (by introducing a so-called PT-STA model) with three core purposes: (1) focusing on public transport accessibility specifically, (2) measuring accessibility to both fixed activities and discretionary opportunities, and (3) integrating travel-time thresholds in the accessibility measurement. These three elements aim to make the PT-STA model a suitable tool to evaluate the impacts of public transport interventions on person-based accessibility and extend the evaluation of public transport accessibility to its fixed and discretionary dimensions. The PT-STA model is tested with a sample of 118 residents of the rural municipality of Mühlwald (South Tyrol, Italy). Results show that the accessibility to fixed activities and discretionary opportunities are limitedly correlated. It is possible to identify people with high accessibility to fixed activities and poor access to discretionary opportunities (typically pensioners and homemakers living in remote locations) and vice versa (e.g. people with articulated rural–urban commutes but daily visiting locations with several amenities). These results preliminarily confirm the importance of combining both accessibility dimensions in the PT-STA model since they tell complementary and not overlapped stories that are relevant for policymakers to evaluate the pros and cons of alternative public transport interventions.
Rural public transport is typically limited in its coverage, frequency and service period. This is often linked to a low and dispersed demand, which makes the provision of competitive transport services often financially unsustainable. Autonomous vehicles (AVs) might change this condition, allowing for an upgrade of public transport in rural areas. Nevertheless, current studies are mostly focused on urban areas, while the potential for rural applications remains underexplored. This paper contributes to this research line by developing a set of so-called AV scenarios for a potential upgrade of public transport in the rural study area of Mühlwald (South Tyrol, Italy). These scenarios are designed following three core principles. First, line-based and on-demand applications of AVs are not only individually tested, but even combined over space and time in different manners to exploit their synergy. Second, the performances of the scenarios are quantified by assuming the today´s agency cost budget, public-transport demand and peak-hour system capacity to be fixed parameters to comply with. Third, the uncertainties regarding the impacts of AVs on the agency costs are taken into account by defining optimistic, neutral and pessimistic variants of each scenario. Results indicate that the full replacement of current bus lines with a system of on-demand shared taxis might provide the highest performance improvements with the same budget as today, but only with a much bigger fleet. At the same time, the combination of bus lines and on-demand services over time (and space) might provide similarly competitive results while keeping the needed fleet size, service distance and service time relatively low. With this study, policy makers may get insights into the potential improvements of rural public transport that might be initially obtained with different uses of AVs, given a fixed agency cost budget and demand.
1 January 2020 in Transportation Research Procedia