"Improving Crash Prediction Using GRIDSMART Infrastructure" is a collaborative research project between the University of Connecticut and the University of Maine, in partnership with the Connecticut Department of Transportation (CTDOT). The study addresses the critical issue of intersection safety, noting that over 50% of all fatal and injury crashes occur at or near intersections, with left turn maneuvers accounting for a significant proportion of these crashes. The research aims to enhance Safety Performance Functions (SPFs) by integrating temporally varying data, such as continuous turning movement counts, into their development. This approach overcomes the limitations of traditional SPFs that mainly consider static measures like the two-way annual average daily traffic (AADT).
The project will utilize GRIDSMART technology, a system implemented for traffic signal management, to collect detailed turning and through movements of various road users at intersections. This data will be used to develop more precise and dynamic SPFs, considering factors like bicycle and pedestrian counts, road geometry, and surrounding land development intensity. The dependent variables in the study will include annual crash counts by manner of collision, categorized by the direction of travel of the involved vehicles.
The methodology involves collecting raw data using GRIDSMART, identifying relevant variables, and selecting intersections based on data quality for analysis. The team will perform temporal aggregation of data and classify crashes by manner of collision. The resulting analysis aims to determine whether turning movements are significant predictors of crashes and how improvements to GRIDSMART infrastructure could enhance intersection safety.