Left turn crash risk analysis: Development of a microsimulation modeling approach

Area of Science:

• Transportation Engineering
• Traffic Safety Analysis
• Microsimulation Modeling

Background:

• Traffic simulation is crucial for analyzing complex transportation issues.
• Advancements in microsimulation and surrogate safety measures enable detailed traffic safety analysis.
• Existing crash prediction models often lack disaggregated risk assessment for specific maneuvers like left turns.

Purpose of the Study:

• To examine how left-turn crash risk varies with changing traffic conditions at signalized intersections.
• To develop statistical models predicting left-turn crash risk based on intersection parameters and traffic conditions.
• To support the implementation of adaptive traffic control strategies, such as time-variable flashing yellow arrow (FYA) phasing.

Main Methods:

• Utilized a calibrated traffic microsimulation model.
• Employed surrogate safety assessment model analysis.
• Simulated 750 unique combinations of intersection geometry, traffic, and signal timing.
• Developed statistical models linking crash risk to left-turn phasing and prevailing conditions.

Main Results:

• Quantified left-turn conflicts per hour across diverse simulated scenarios.
• Established statistical models predicting the risk of left-turn crashes.
• Demonstrated the relationship between left-turn phasing mode and crash risk.
• Validated the model's potential for time-variable safety-based phasing selection.

Conclusions:

• Traffic microsimulation and surrogate safety measures provide a viable approach for detailed traffic safety analysis.
• The developed statistical models offer a disaggregated method for predicting left-turn crash risk.
• The findings support the use of FYA phasing for optimizing traffic operations and safety at signalized intersections.