Heavy Braking Patterns and their Applicability to Assess Traffic Conflicts

Traffic safety has been a significant concern for ever-increasing injuries and fatalities. The advancement in technology and ITS infrastructure implementation has made traffic data collection and storage easier. This data can be further utilized to identify and suggest measures in preventing potential crashes. Despite the availability of huge vehicle trajectory data sets, assessing potential conflicts by segregating all the leader-follower (LF) pairs often becomes a tedious task. Heavy braking (HB) is a variable that is narrowly explored as a crash surrogate for safety assessment. In this study, the empirical vehicular trajectories are utilized to analyze HB events. HB events are based on the deceleration values of vehicles in the stream exceeding a chosen threshold. The vehicle trajectories are also used to estimate Deceleration Rate to Avoid Collision (DRAC) as a Surrogate Safety Measure (SSM). The empirical microscopic trajectory data for three cities in India, namely, Delhi, Surat, and Chennai, are utilized in this study. This study hypothesized that DRAC events are highly correlated with the heavy braking events within the stream. Thus, this paper attempts to understand the correlation between DRAC and HB events. The R2 value at each site is greater than 0.80, thus suggesting a strong correlation between the heavy braking events and the conflicting DRAC events. HB events can be used as a surrogate safety measure in identifying potential crashes or near misses.