Reviewing autonomous vehicle simulation literature: policy insights for transportation

As autonomous vehicle (AV) technology progresses, a substantial amount of AV simulation work has studied potential future outcomes that could occur given different assumptions, scopes and technologies. As a revolutionary technology, AVs could have significant impacts on urban economic, environmental and social systems. In many cases, the research is directed towards decisionmakers: transportation authorities, policymakers and fleet operators. However, due to the complexity and uncertainty of AV adoption in interconnected urban systems such as transportation and land use, it is difficult for policymakers to understand the overlapping policy suggestions and results across the many simulation studies. In this paper, we summarise the suggested policies in AV simulation research from 170 studies published during 2016–2024 and provide context for the policies being suggested. Specifically, the key question we try to answer is: what are the recommended AV policies based on simulation studies? Many simulation results have shown increased vehicle miles/kilometers travelled (VMT/VKT) due to induced demand, empty vehicle repositioning, or cruising instead of parking. In response, we found common policy recommendations such as congestion pricing, subsidising car-sharing and ride-pooling, increasing parking costs, investing in public transportation and restricting AV demand as ways to address the increase in VMT/VKT. We also identified common policy bundles, or joint recommendations, such as congestion pricing with increased parking costs. Finally, we compare and contrast simulation policy suggestions with a few qualitative AV policy studies, pointing out key research gaps for the future and possible mixed-methods opportunities.