SHRP 2 Safety Implementation: Driver Behavior in the Vicinity of Closely Spaced Interchange Ramps on Urban Freeways - Phase 2

Project Description State departments of transportation (DOTs) frequently evaluate requests for new access or modifications to existing access on freeways. Transportation agency concerns regarding new or modified freeway access include ramp spacing on the mainline and its impact to freeway operations and safety. Very little information existed on the safety impacts of ramp spacing until recently, when the University of Utah created both crash modification factors (CMFs) and severity distribution functions (SDFs) for entrance-exit (ENEX) ramp spacing and auxiliary lane presence (Le Porter, 2012; Shea et al., 2015). While these CMFs can be used by state DOTs to quantify the freeway mainline safety effects of ramp spacing and auxiliary lane presence, they are purely empirical findings that represent statistical associations. Underlying, causative reasons for the crash patterns can be inferred, but are not fully understood. A more thorough understanding of the “cause-effect” relationships leading to observed crash patterns is needed so that countermeasures can be identified and installed at current and future locations with closely spaced interchange ramps. This data set was intended to analyze the scenario of an entrance ramp followed by a downstream exit ramp (EN-EX), with or without an auxiliary lane between entrance and exit ramps. Drivers included in the dataset encompass “merging” drivers (i.e., merging onto the freeway), “exiting” drivers (i.e., exiting the freeway), and through moving drivers (i.e., traveling on the freeway mainline prior to and after the ramp combinations). Data Request Scope This data set encompasses 40-60 freeway roadway segments located in Washington and North Carolina, with approximately 300 trip epochs included for each segment. The trip epochs are evenly divided between "Entering" vehicles, "Exiting" vehicles, and "Through" vehicles scenarios. Traversals in the data set range from 10 seconds before the initial “buffer” cross (or Link ID entrance) until 10 seconds after the final “buffer” cross (or Link ID exit). Data Specification The data set includes a sample of de-identified time-series variables as well as indicators noting whether a driver was talking on a cell phone (both with hand-held phone and hands-free phone), manually manipulating (dialing/texting) a cell phone, or if the driver was not distracted while traveling through the study location. Forward and rearward video views are also included. Time series data files are in a CSV (text) file format, with timestamps reported on a 10 Hz frequency clock and measures aligned to the closest timestamp. Video files are H.264 encoded with a nominal collection rate of 15 frames per second.