Using empirical traffic trajectory data for crash risk evaluation under three-phase traffic theory framework

Area of Science:

• Traffic Engineering
• Road Safety Analysis
• Transportation Systems

Background:

• Understanding traffic dynamics is crucial for road safety.
• The three-phase traffic theory categorizes traffic flow into distinct states.
• Microscopic analysis of vehicle trajectories offers detailed insights into crash risks.

Purpose of the Study:

• To evaluate crash risks across different traffic phases and transitions using surrogate safety measures.
• To investigate the influence of traffic flow variables (flow rate, density, speed) on crash risks.
• To develop predictive models for crash risk based on traffic states and parameters.

Main Methods:

• Utilized empirical vehicle trajectory data from highways in California, USA, and Nanjing, China.
• Identified traffic phases based on traffic flow variables and their correlations.
• Applied advanced crash risk indexes derived from vehicle trajectories.
• Developed regression models to quantify the relationship between traffic variables, states, and crash risks.

Main Results:

• Significant variations in safety performance were observed across different traffic states.
• Synchronized flow and wide moving jam phases were identified as the most hazardous.
• High traffic density and low speeds correlate strongly with increased crash risk.
• Integrating traffic phases and parameters improved crash risk prediction accuracy.

Conclusions:

• Traffic phase and its characteristics are critical determinants of road safety.
• Specific traffic states like synchronized flow and wide moving jams require targeted safety interventions.
• Microscopic analysis and advanced risk indexes provide valuable tools for traffic safety assessment and management.