Skip to main content
Thesis defences

PhD Oral Exam - Matin Giahi Foomani, Civil Engineering

Assessing the impact of active signage systems on driving behavior and traffic safety

DATE & TIME
Wednesday, August 10, 2022 (all day)
COST

This event is free

ORGANIZATION

School of Graduate Studies

CONTACT

Daniela Ferrer

WHERE

Online

When studying for a doctoral degree (PhD), candidates submit a thesis that provides a critical review of the current state of knowledge of the thesis subject as well as the student’s own contributions to the subject. The distinguishing criterion of doctoral graduate research is a significant and original contribution to knowledge.

Once accepted, the candidate presents the thesis orally. This oral exam is open to the public.

Abstract

Unsignalized Stop-Controlled Intersections (SCI) are widely used in North America, and accounting for one out of every ten collisions. Understanding how drivers and pedestrians behave at unsignalized intersections is critical for public safety. Drivers who do not obey the stop-sign's indication by not coming to a complete stop or miss or fail to stop at SCI create a substantial safety risk. For decades, visibility and placement of road alignments and signage at intersections have been a concern among transportation safety specialists. Deployment of backlit Light-Emitting Diode (LED) or other illuminated signs (also known as active road signs) has been increased especially at hot-spots and locations with known safety problems, or potential collision risks. While these signs are expected to improve safety measures by regulating safe travelers' passage, their performance is not yet fully understood. Although environmental factors such as intersection type, location, and road design are playing a major role, compositional variables such as driver behaviour, which can be explained in terms of carelessness, lack of attention, or overconfidence, is resulting in a failure to comply with the law of making a complete stop at SCI.

Previous empirical research demonstrated some correlation between several variables such as traveller compliance with road signs and alignments, direct and indirect road safety measures, collision/conflict frequency, and road/traffic characteristics. These studies commonly employ before-after or cross-reference analyses to determine the long-term effects of various countermeasures at SCI. A few studies were also utilizing calibrated micro-simulations models to evaluate the surrogate safety measures at SCI.
This thesis defines a methodology to evaluate the safety performance of a new and untested signage without putting traffic at long risk. To evaluate the performance of the signs, the suggested methodology investigates multiple parameters and identifies influencing variables in a conflict-based collision-prediction model at SCI. The proposed methodology is applied to a real-world network in the city of Montreal, with several three-leg SCI equipped with different countermeasures. The experiment was designed in a fashion which isolates the influence of several variables, allowing the focus to be on the impact of the target variable (signage type). Field experiments have been performed to study the driver's behavior in terms of approaching speed as well as quantitative analysis on reaction to various signs, using different sample groups from the same population. This research sets up a microsimulation model that captures drivers' behaviour with respect to signage according to the observed data. A genetic algorithm was deployed to calibrate the microsimulation model in terms of turning movement counts and the critical conflicts were calculated at each intersection using vehicle trajectories. Collision-prediction regression model was then developed for the intersections under investigation, using traffic volume and conflict.

The results demonstrated a high correlation among countermeasures and drivers' speed and compliance. The relationship between critical conflicts computed in microsimulation models and actual collisions was found to be statistically significant. The model which includes drivers' compliance in collision-prediction regression was also found to fit the collision data better. However, the results of this study do not support the previous assumption that the conflict-based collision-prediction models fit the collision data better than the volume-based collision-prediction models at SCI, especially with drivers' compliance supplementary data. Finally, while the backlit signs' performance was marginally better than that of a normal LED active sign, the difference was not statistically significant.

The methodology suggested in this thesis has the potential to be implemented in safety performance evaluation of a countermeasure without placing traffic at danger for an extended period. For instance, when there is apprehension about an adverse effect. Future research could investigate leveraging drivers' behaviour to countermeasures, to improve the performance of collision-prediction regression models like the one proposed in this thesis. Finally, the results from the performance assessment of the LED active signs can assist transportation specialists in deciding whether or not deploy these countermeasures.

Back to top

© Concordia University