Quantifying the Benefits and Harms of Connected and Automated Vehicle Technologies to Public Health and Equity

Automated Vehicles (AVs) have the potential to improve traffic safety by preventing crashes. The safety implications of AVs can vary across communities with different socioeconomic and demographic characteristics. In this study, we proposed a framework to quantify the potential safety implications of AVs in terms of preventable crashes and fatalities, accounting for some of the safety challenges of AV operation, including AV technologies’ safety effectiveness, system failure risk, and the risk of disengagement from the automated system to manual driving. We further defined an empirical study to examine the proposed framework and investigate inequity in AV potential safety implications. The empirical analysis was conducted using 2017 crash data from the Dallas-Fort Worth, Texas, United States area. The results showed that AVs could potentially prevent up to 50%, 46%, 23%, 6%, and 5% of crashes for automation Levels 5 to 1, respectively. Among advanced driver assistance systems, pedestrian detection, electronic stability control, and lane departure warning showed more significant potential in reducing fatal crashes. We found a U-shaped relationship between the AV-preventable fatalities and household median income and ethnically diverse communities. The findings of this study suggests that low-income and ethnically diverse communities can benefit from AV implementation. The policy recommendations of this research suggest that city and state planning and transportation agencies may consider implementing policies and strategies for making AVs available to low-income and ethnically diverse communities at a lower cost.