Assessing the Long-Term Effects of Autonomous Vehicles on Mode Choice Behavior: A Speculative System Dynamics Approach
Wolfgang Gruel, Massachusetts Institute of Technology, Joseph M. Stanford, Massachusetts Institute of Technology
Using a system dynamics approach, we identify several long-term outcomes of the adoption of autonomous driving. We further explore its impact on transportation systems and investigate under which conditions it will be harmful or beneficial.
In recent years, self-driving cars have generated significant attention and discussion—as well as a fair share of hype. By the start of 2015, several companies had presented concepts for self-driving vehicles, and the numbers of the skeptics have diminished, while the ranks of the proponents have continued to grow.
The public discussion around autonomous cars is currently dominated by the technical and legal challenges facing this new technology, and by the wide range of potential benefits the technology presents. Some of these direct benefits have been discussed extensively, and, if realized, they could be substantial. For example, it is widely expected that autonomous vehicles will: dramatically increase road safety; reduce congestion and improve energy efficiency through more efficient driving; increase the utility of a person’s time spent in the car, when his or her attention is no longer needed for driving; and improve mobility for currently under-served populations, such as the elderly and the handicapped.
While it is recognized that a number of technical and legal issues need to be solved, widespread adoption of autonomous vehicles is increasingly considered to be inevitable. It is, therefore, astonishing that the long-term effects of autonomous vehicles—including a wide-range of indirect effects—are rarely discussed. Among these potential impacts are a number of positive and negative outcomes, and the net-effect in terms of societal benefit or harm is far from clear.
In this paper, we identify the several of these outcomes, and we explore conditions in the broader transportation system under which autonomous vehicles will be harmful or beneficial. Using a system dynamics model, we investigate how autonomous operation could affect the attractiveness of traveling by car, how this in turn would affect mode-choice, and how changes in mode-choice would affect the broader transportation system. The model explores four speculative scenarios, using a generic model of an urban area, based on an existing model of Cambridge, Massachusetts, which was developed with VensimPLE software.
The results affirm the expected benefits of autonomous vehicles—if mobility behavior does not change significantly. However, the model also shows that autonomous vehicles offer new behavioral opportunities that lead to undesired results like more congestion and more emissions. Finally, we discuss potential policy interventions to encourage positive outcomes, and we identify additional questions and avenues for research. The paper extends the view on autonomous vehicles and provides insights for policy-makers, transportation and urban planners, vehicle manufacturers, and public transit operators.
Association for European Transport