Using SHRP2-Nds Data to Investigate Freeway Operations, Human Factors, and Safety: Final Report

2018PDFTech ReportATLAS-2018-23DTRT13-G-UTC54Freeway operationsHighway operationsDriver performanceDriversReaction timeCar followingTraffic speedOperating speedFreeway OperationsDriver PerformanceDriver AgeReaction TimeSpeed ofChoiceCar-following BehaviorRamp SpeedMichiganUnited StatesATLAS Center (Mich.)Avelar, Raul EduardoDadashova, BaharHiggins, LauraGillette, GeorgeTexas A&M Transportation InstituteUniversity of MichiganATLAS Center (Mich.)University Transportation Centers Program (U.S.)US Transportation CollectionThis project identified and quantified relationships between traffic operations variables (such as operating speeds), human factors characteristics (e.g., driver demographics), and safety variables (crash or near-crash outcome) making use of the recently available SHRP2 databases. Researchers studied driver performance and investigated how the operational characteristics may explain safety outcomes. Using nearly 800 events from freeway trips in Washington and Florida at non-curve and under uncongested conditions, researchers fitted one set of models to free-flowing driving situations to investigate variability in speed of choice (SOC). Driver characteristics such as years of driving experience, driver age, vision conditions, and history of traffic violations correlate with the SOC. However, facility type, traffic density, and especially the posted speed limit are most influential to free-flow driving. For another analysis, researchers calibrated a dynamic mixed-effects model to car-following situations and found that the estimated reaction time for a driver increases with increasing age (from 1.1 s for drivers younger than 20 years, up to 2.2 s for drivers older than 69 years). Compared to other age groups, younger drivers were found over-sensitive to relative speed and following gap as the magnitude of their speed adjustment due to these factors was found largest. Drivers 70 years of age or older performed speed adjustments similar to drivers ages 20\u201339, though these adjustments lagged more for the older group (i.e., reaction time of 2.2 s, compared to 1.2 s for ages 20\u201339). Additionally, a consistent trend in adjusting car-following speed toward the predicted free-flow SOC was found for all age groups except for younger drivers. Future research should focus on studying car-following behavior in situations that resulted in crashes or near crashes. The second study compiled a set of speed data at freeway ramps in Pennsylvania. The results showed that the traffic conditions and the geographic location characteristics are the main predictors of a driver\u2019s speed choice on ramps. Based on the influence of these factors, drivers were found to adjust their speed more uniformly. Driver characteristics were found to influence the SOC to a certain degree. Drivers in the group of speeds with larger variability were found to be easily distracted based on their Barkley Attention Deficit Hyperactivity Disorder scores