Potentials of Autonomous Vehicles in a Changing Private Transportation System – a Case Study in the Stuttgart Region



Potentials of Autonomous Vehicles in a Changing Private Transportation System – a Case Study in the Stuttgart Region

Authors

Michael Heilig, Karlsruhe Institute of Technology (KIT), Tim Hilgert, Karlsruhe Institute of Technology (KIT), Martin Kagerbauer, Karlsruhe Institute of Technology (KIT)

Description

In this work, we analyzed the potential of automated vehicles in combination with carpooling regarding potentials for savings in car journeys, travel times and vehicle mileages.

Abstract

The breakthrough of autonomous vehicles (AV) is no question of ‘if’ – it´s a question of ‘when’. Stating this expression is correct the big research topics are the impacts of AVs on mobility considering level 5 of automation (=automated driving system undertakes all aspects of the dynamic driving tasks in all roadway and environmental conditions; Society of Automotive Engineers). Autonomous vehicles have obviously advantages such as a higher safety levels (due to less driver errors), improved emissions (optimized speed and car-to-car/car-to-infrastructure communication) as well as more equity (more mobility options of disabled, younger or older people) and time savings (using riding time for other activities, e.g. working or reading). Especially in large urban areas, these advantages of AVs in addition with information and communication technologies may change private transport from private cars towards carpooling. Vehicles can be used more efficiently regarding use of space and waiting times. At the same time, cost efficiency will increase since no driver is needed and the reorganization of vehicles may be automated as well. These advantages offer a lot of new scenarios for carpooling in combination with AV.
In this case study, we examine a future scenario with 100% AVs and a well-working carpooling system. Therefore, we use the results of the agent-based microscopic travel demand model mobiTopp of the Greater Stuttgart region with 2.3 million agents. The first step is an analysis of the time-dependent O-D-demand for the mode car (as driver and as passenger) on workdays (Monday to Friday) resulting from the model. We further increase the rate of car occupation and recalculate the O-D-demand in a sensitive analysis. Based on this analysis, we determine the number of AVs which would be necessary in order to cover the mobility needs of the Stuttgart population without adopting travel behavior. In the second step, we rerun the traffic assignment just considering the AVs to analyze how traffic demands and the usage of vehicles may affect travel times and vehicle mileage.
The results state large impacts on the infrastructure as well on the car usage. Exemplary results show that 44% of car journeys in the Stuttgart area are redundant in a ‘perfect autonomous world’. Further, with only carpooling AVs in use, up to 36% of vehicle miles travelled can be saved. Further results point out the changes in people’s average travel times as well as the lower demand for parking areas.
The paper provides insight in the methodology of the case study und shows some feasibilities how to reorganize future’s transportation systems with automated vehicles.

Publisher

Association for European Transport