Traffic safety evaluation has traditionally relied on police-reported crash statistics, but researchers say that approach is limited for predictive modeling because crash data is a “lagging” indicator and crashes are rare on arterial and local roads. In a new paper, the authors examine whether hard-braking events can serve as a higher-frequency measure of crash risk.
In “From Lagging to Leading: Validating Hard Braking Events as High-Density Indicators of Segment Crash Risk“, the authors evaluate hard-braking events (HBEs) as a scalable surrogate for crash risk. They define an HBE as an instance where a vehicle’s forward deceleration exceeds -3m/s², which they interpret as an evasive maneuver.
The paper says HBEs can support network-wide analysis because they come from connected vehicle data, unlike proximity-based surrogates such as time-to-collision that often require fixed sensors. The authors say they found a statistically significant positive correlation between crash rates and HBE frequency.
To test the approach, they combined public crash data from Virginia and California with anonymized, aggregated HBE information from the Android Auto platform.
The researchers say the method addresses a practical problem in safety analysis: it can take years to collect enough crash data to establish a valid safety profile for a specific road segment, and inconsistent reporting standards across regions can complicate risk modeling. HBEs are presented as a “leading” measure that occurs more frequently than crashes and may help with proactive safety assessment.
Source: research.google.
Companies can share verified announcements through Newz9’s international press release submission page.
