Signal Treatments to Reduce the Likelihood of Heavy Vehicle Crashes at Intersections: Microsimulation Modeling Approach

Abstract

Traffic simulation modeling has been applied to many transport planning and traffic engineering situations. The further development of traffic simulation modeling to study the impact of safety measures in a systematic, rigorous, and transparent fashion is becoming increasingly viable as the models improve and the understanding of driver behavior is improving. This paper presents an application of traffic simulation to the study the safety problem of heavy vehicle red-light running at a vehicle actuated highway intersection in a metropolitan area in Australia. The processes of data collection, model calibration and validation, and evaluation are described. Modeling driver stop-or-go behavior of drivers of heavy and light vehicle in the “dilemma zone” of the vehicle actuated signal is also illustrated, along with the modeling of surrogate safety measures. Five alternative signal treatments intended to reduce heavy vehicle crash risk were considered. The results of a comparison against the existing situation showed that an extension of amber time was the most effective short-term treatment. Over the long term, however, this treatment is likely to be subject to behavioral adaptation. A green extension for heavy vehicles detected in the dilemma zone and an all-red extension for potential red-light runners were found to be treatments likely to provide a sustainable safety improvement with little impact on operational efficiency. This type of safety modeling is considered to have great potential in the future.