Visitors security analysis has historically relied on police-reported crash statistics, usually thought of the “gold commonplace” as a result of they straight correlate with fatalities, accidents, and property injury. Nonetheless, counting on historic crash knowledge for predictive modeling presents vital challenges, as a result of such knowledge is inherently a “lagging” indicator. Additionally, crashes are statistically uncommon occasions on arterial and native roads, so it could possibly take years to build up enough knowledge to ascertain a sound security profile for a particular highway phase. This sparsity paired with inconsistent reporting requirements throughout areas complicates the event of sturdy threat prediction fashions. Proactive security evaluation requires “main” measures: proxies for crash threat that correlate with security outcomes however happen extra incessantly than crashes.
In “From Lagging to Main: Validating Onerous Braking Occasions as Excessive-Density Indicators of Phase Crash Danger“, we consider the efficacy of hard-braking occasions (HBEs) as a scalable surrogate for crash threat. An HBE is an occasion the place a car’s ahead deceleration exceeds a particular threshold (-3m/s²), which we interpret as an evasive maneuver. HBEs facilitate network-wide evaluation as a result of they’re sourced from linked car knowledge, in contrast to proximity-based surrogates like time-to-collision that incessantly necessitate using mounted sensors. We established a statistically vital constructive correlation between the charges of crashes (of any severity stage) and HBE frequency by combining public crash knowledge from Virginia and California with anonymized, aggregated HBE info from the Android Auto platform.
