Advancing safety through connected and autonomous vehicles: a meta-analysis of market penetration and safety improvement rates from 2015 to 2024

Abstract

The field of Connected and Autonomous Vehicles (CAVs) promises to improve road safety through advanced communication technologies and data sharing. This study addresses the need for a clearer understanding of the relationship between CAV Market Penetration Rates (MPRs) and Safety Improvement Rates (SIRs) by conducting an updated meta-analysis that includes 49 studies published between 2015 and 2024. Across these studies, the primary focus of safety surrogate measures (SSMs) was on Time-to-Collision (TTC) and Post Encroachment Time (PET), reflecting their dominant role in evaluating traffic safety impacts of CAVs. We applied sensitivity analysis to refine the data and mitigate the effects of variability across simulation parameters and scenarios. A power function was identified as the most accurate model to describe the MPR-SIR relationship, capturing how safety benefits scale with increasing adoption of CAVs. At low MPRs (10%–20%), safety gains are modest, with SIRs of 2.8% and 4.9%, respectively. Mid-range MPRs (30%–60%), while frequently studied, show overestimated safety benefits in raw findings; our corrected estimates yield SIRs of 8.0% at 30% MPR and 9.4% at 40% MPR, highlighting the complexity of mixed traffic environments. At high MPRs (70%–90%), safety benefits become substantial and more consistent, with an SIR of 39.4% at 90% MPR. These findings inform transportation planners and policymakers on the safety potential of CAVs, emphasizing the importance of high market penetration to realize meaningful safety improvements.