Development of surrogate safety measures based on connected vehicles' speed data

Keywords: Surrogate safety measures, Crash analysis, Connected vehicle data, Big data analytics.
Abstract Type: Paper Abstract

Abstract

Road fatalities continue to be a global public health crisis. Studies have proven that driving speed variations are tightly related to traffic risks, which can be used to generate surrogate safety measures (SSMs) to identify high-risk road segments before crashes occur. However, large-scale assessment of driving speeds' relations with crash risks is still underexplored due to data acquisition difficulties. With connected vehicles (CVs) becoming a reality, massive volumes of driving speed data can be directly collected from vehicular sensors and become accessible to researchers. This emerging transport big data provides unrivaled opportunities for road safety assessment and crash risk modeling. This study innovatively explores the potential and effectiveness of market-available CV data in road safety studies. We generated a list of variables based on the CV data to quantify driving speed variations and modeled their relations with crash risks in Waco, Texas. This study seeks to understand to what extent and how effective and reliable SSMs can be generated from commercially available CV-based driving speed data for identifying high-risk traffic locations.