Traffic conflict assessment using macroscopic traffic flow variables: A novel framework for real-time applications.

The present study develops a comprehensive traffic conflict assessment framework using macroscopic traffic state variables. To this end, vehicular trajectories extracted for a midblock section of a ten-lane divided Western Urban Expressway in India are used. A macroscopic indicator termed "time spent in conflict (TSC)" is adopted to evaluate traffic conflicts. The proportion of Stopping distance (PSD) is adopted as a suitable traffic conflict indicator. Vehicle-to-vehicle interactions in a traffic stream are two-dimensional, highlighting that the vehicles interact simultaneously in lateral and longitudinal dimensions. Therefore, a two-dimensional framework based on the influence zone of the subject vehicle is proposed and employed to evaluate TSCs. The TSCs are modeled as a function of macroscopic traffic flow variables, namely, traffic density, speed, the standard deviation in speed, and traffic composition, under a two-step modeling framework. In the first step, the TSCs are modeled using a grouped random parameter Tobit (GRP-Tobit) model. In the second step, data-driven machine learning models are employed to model TSCs. The results revealed that intermediately congested traffic flow conditions are critical for traffic safety. Furthermore, macroscopic traffic variables positively influence the value of TSC, highlighting that the TSC increases with an increase in the value of any independent variable. Among different machine learning models, the random forest (RF) model was observed as the best-fitted model to predict TSC based on macroscopic traffic variables. The developed machine learning model facilitates traffic safety monitoring in real-time.

Investigating and modeling the influence of PET-types on crossing conflicts at urban unsignalized intersections in India.

Un-signalized intersections in India witnessed the maximum number of crashes and fatalities in 2019. The nature of the crash investigation is still largely reactive, where the need for accurate and reliable crash data for effective safety diagnosis is pivotal. In India, crash records are unscientific, and critical details are missing. Therefore, a proactive approach using surrogate safety measures is more promising and prudent in analyzing traffic safety. The present study investigates and models crossing conflicts at un-signalized intersections under mixed traffic conditions. Traffic video data for 14 un-signalized intersections (eight un-signalized three-legged intersections and six un-signalized four-legged intersections) were collected under normal weather conditions. The crossing conflicts were identified and characterized as critical and noncritical conflicts based on the values of post-encroachment time (PET). Conflicts with PET values between -1 s and 1 s were identified as critical conflicts. The observation revealed the existence of both positive and negative PET values. The investigation revealed that crossing conflicts with negative PET values are riskier and more unsafe than conflicts with positive ones. Therefore, the crossing conflicts with positive and negative PETs were modeled separately. The positive and negative PET-based critical crossing conflicts are modeled as a function of traffic flow and intersection geometry-related characteristics using truncated negative binomial regression under a full Bayesian modeling framework. K-fold cross-validation with fivefold was employed to calibrate the model, and RMSE was used to find the best model. The modeling results revealed that the volume and traffic composition of the offending and conflicting stream and intersection geometry significantly influence the number of positive and negative PET-based critical crossing conflicts. The developed models can interest engineers and safety experts to analyze traffic safety and identify critical intersections in urban road networks.