Ahmed Mohamed Mostafa Amer

Modeling Driver Behavior Within a Signalized Intersection Approach Decision-Dilemma Zone

This study uses data gathered in a field test on 60 test participants to develop models that characterize driver brake perception–reaction times (PRTs), brake times, and stop–go decisions at the onset of a yellow indication at a high-speed signalized intersection approach. The study demonstrates that driver PRTs are influenced only by the drivers time to intersection (TTI) at the onset of the yellow indication. The driver PRT is found to increase linearly with TTI and is not affected by the vehicle speed (in the range of 54 to 88 km/h), driver gender, or driver age. In the case of stop–go behavior, the older driver (≥65 years of age) dilemma zone is wider, ranging from a TTI of 4.81 to 1.66 s versus 4.90 to 2.87 s for the younger age group. Female drivers are more likely to stop than male drivers and tend to have a dilemma zone that is closer to the intersection. Finally, the study demonstrates that dilemma zone control systems should consider a dilemma zone from 5.0 to 1.5 s instead of the current state of practice of 5.5 to 2.5 s to capture all potential driver age and gender groups.

Impact of Driver and Surrounding Traffic on Vehicle Deceleration Behavior at Onset of Yellow Indication

This paper discusses driver deceleration levels in a controlled field environment at the onset of a yellow indication on high-speed signalized intersection approaches using an in-vehicle differential Global Positioning System. The impacts of driver gender, driver age, roadway grade, mean approach speed, platooning scenarios (leading, following, or alone), and time to intersection (TTI) on driver deceleration levels were analyzed. This information is critical for the efficient and safe design of traffic signal clearance timings. The IntelliDrive initiative can gather information about the driver, subject vehicle, and surrounding traffic conditions to execute safe and customizable traffic signal indication change warnings. The results indicate that driver deceleration levels are significantly higher than the 3-m/s2 deceleration level used in the state-of-the-practice traffic signal design guidelines. The mean deceleration level is 3.6 to 4.1 m/s2. The results can be used to enhance the design of yellow timings and may be integrated with the new IntelliDrive initiative to provide customizable driver warnings. The results demonstrate that driver deceleration levels are higher at shorter TTIs at the onset of yellow. Drivers are willing to exert deceleration levels in excess of 7 m/s2 at short TTIs (less than 2.5 s). Furthermore, older drivers (60 years of age or older) employ greater deceleration levels compared with younger (under 40 years old) and middle-aged (between 40 and 59 years old) drivers. A driver following another vehicle that proceeds legally through an intersection without stopping exerts higher deceleration levels than drivers driving alone on a roadway or leading another vehicle, and drivers leading a platoon of vehicles are not affected by vehicles behind them.

Characterization of Driver Perception Reaction Time at the Onset of a Yellow Indication

The research presented in this paper characterizes driver perception-reaction times (PRTs) in a controlled field environment at the onset of a yellow-indication transition in high-speed signalized intersection approaches. The study characterized the impact of driver gender, driver age, roadway grade, mean approach speed, platooning scenarios (leading, following, or alone), and time-to-intersection (TTI) on the driver PRT. This characterization is critical for the efficient and safe design of traffic signal clearance timings. The study demonstrates that the driver PRT is higher for female and older drivers (60 + age group) as compared to male and younger drivers. The PRT is larger when vehicles travel along an upgrade section. Driver PRTs are typically higher if they are following a vehicle that runs a yellow light. Furthermore, driver PRTs decrease when they are followed by another vehicle. Finally, driver PRTs increase as the TTI at the onset of the yellow interval increases.

Driver Stopping Behavior on High-Speed Signalized Intersection Approaches

The research presented in this paper characterizes driver stopping behavior at the onset of a yellow phase on high-speed signalized-intersection approaches through the use of controlled field data gathered from 60 test subjects by means of an in-vehicle differential global positioning system. A total of 745 data records were available for analysis of all drivers who stopped at the onset of the yellow phase. These ranged from a minimum time to the stop bar (TTS) of 1.34 s to a maximum of 6.19 s. Statistical analyses were used to investigate the effects of the TTS bar, grade (uphill and downhill), age (younger than 40 years old, 40 to 59 years old, and 60 years of age or older), and gender on seven dependent measures of driver performance, including perception time, reaction time, perception–reaction time, braking time, stopping time, and stopping accuracy. The study demonstrated that driver perception time was not affected by TTS, while reaction time was dependent on TTS, roadway grade, and driver age. Younger drivers had longer reaction times than the older group, but they were able to stop over a shorter period of time, as they typically applied more aggressive braking rates. A lower perception–reaction time was found for drivers who had their foot lifted off the accelerator at the onset of the yellow phase. Male drivers showed slightly higher braking times than female drivers, with no significant differences between male and female drivers. Furthermore, the results demonstrated that drivers who tried to stop in short TTSs were more likely to stop downstream of the stop line and that older drivers were significantly more accurate when they stopped than other age groups.

Agent-based behavioral modeling framework of driver behavior at the onset of yellow indication at signalized intersections

This paper presents a state-of-the-art behavioral model (BM) that can be used as a tool to simulate driver behavior after the onset of a yellow indication until (s)he reaches the intersection stop line. The paper presents the general framework of the proposed BM, its components, and discusses its ability to track Dilemma Zone (DZ) drivers and update the information available to them every time step until they reach a final decision. The BM framework is ideal for testing dilemma zone mitigation strategies before actual implementation. In addition, the BM framework can be easily implemented in any traffic simulation software. The paper performs system-based and agent-based characterization of the components involved in the BM framework using data collected from a controlled field driving experiment. The BM is validated using Monte Carlo (MC) simulations, and produces high success rates of 72.8% and 87.2% for system-based and agent-based models, respectively.

Agent-Based Stochastic Modeling of Driver Decision at Onset of Yellow Light at Signalized Intersections

This paper introduces two logistic statistical models for the driver stop–run decision at the onset of yellow at signalized intersections to capture the stochastic nature of the driver stop–run decision. One model is a classical frequentist model, whereas the other uses a Bayesian statistics approach. The Bayesian model parameters were calibrated by using the Markov Chain Monte Carlo slice procedure implemented within the MATLAB software. Both models were developed with 3,328 stop–run records, which were collected in a field experiment on the Virginia Smart Road, a limited-access highway between Blacksburg and Interstate 81 in Montgomery County, Virginia. The variables included in each model were driver gender, age, time to intersection, yellow time, approaching speed, and speed limit. Both models were shown to be consistent. For the Bayesian model application, two procedures were illustrated: cascaded regression and Cholesky decomposition. Both procedures produced replications consistent with the Bayesian model realizations, while these procedures captured the parameter correlations without the need to store the set of parameter realizations. The Bayesian model produced valid and transferable behavior by replicating multiple experimental results. The proposed Bayesian approach is ideal for modeling multiagent systems in which each agent has its own unique set of parameters.