Distracted driving is known to be one of the main causes of crashes in the United States, accounting for about 40% of all crashes. Drivers’ situational awareness, decision-making, and driving performance are impaired as a result of temporarily diverting their attention from the primary task of driving to other unrelated tasks. Detecting driver distraction would help in adapting the most effective countermeasures. To tackle this problem, we employed a random forest (RF) classifier, one of the best classifiers that has attained promising results for a wide range of problems. Here, we trained RF using the data collected from a driving simulator, in which 92 participants drove under six different distraction scenarios of handheld calling, hands-free calling, texting, voice command, clothing, and eating/drinking on four different road classes (rural collector, freeway, urban arterial, and local road in a school zone). Various driving performance measures such as speed, acceleration, throttle, lane changing, brake, collision, and offset from the lane center were investigated. Using the RF method, we achieved 76.5% prediction accuracy on the independent test set, which is over 8.2% better than results reported in previous studies. We also obtained a 76.6% true positive rate, which is 14% better than those reported in previous studies. Such results demonstrate the preference of RF over other machine learning methods to identify driving distractions.

To provide an equitable transportation system for the public, it is vital to fully understand underrepresented populations’ active transportation needs and their challenges. This study is designed to identify the active transportation studies that focus on underrepresented populations, evaluate the quantification approaches used in the literature, and uncover barriers to the use of active transportation among underrepresented populations in the United States. We used PRISMA guidelines to systematically review relevant publications. After screening and reviewing the literature, some 60 articles were included in the review. The most frequent results when reviewing the literature include that active travel and daily transport trips decreased from younger to older age groups; bicycling was more popular among men, White, and individuals with higher education, whereas walking was more popular among women; bike usage was found to be low among minorities, and those with lower incomes. Furthermore, communities with poor accessibility to bike infrastructure had a larger concentration of African Americans, Hispanics, Asians, low-wage employees, individuals with lower education, and older adults.

Objective: With the advent of connected and automated vehicle (CAV) technology, there is an increasing need to evaluate driver behavior with the advent of connected and automated vehicle (CAV) technology. This study introduced a red-light violation warning (RLVW) system using CAV technology in a driving simulator environment, to evaluate driver braking behavior when the light changes suddenly from green to yellow. Methods: Researchers recruited a total of 93 participants from diverse socio-economic backgrounds for this study and created a virtual network of downtown Baltimore. An eye tracking device was used to observe distractions and head movements. A Lognormal accelerated failure time (AFT) distribution model was used for this analysis, to calculate speed reduction times from the moment the traffic light changes from green to yellow, to the point where a minimum speed was reached. Results: It was observed that speed reduction times were significantly higher in the presence of a RLVW system, requiring a longer period of time to come to a complete stop at the red light. Inferences can be drawn from the jerk analysis that, the RLVW system results in a highly unsafe jerk at the onset of the warning. Without the RLVW system though, a highly uncomfortable positive jerk occurs closer to the signal, which is due to sudden acceleration, as the participants possibly slowed down a lot initially. Gaze analysis showed that the system was able to attract the attention of the drivers, as the majority of the drivers noticed the displayed warning. Conclusions: The findings suggest that the presence of an RLVW system sends a clear message to the driver about the change in traffic light and gives the driver ample time to adapt their initial approach speed to stop at the signal, avoiding potential intersection crashes.

A single unit of information is an answer to any anticipated question a motorist may ask. Such questions include: “What happened? Where? What do I do?” This study, a first of its kind, analyzes the optimum number of units of information Dynamic Message Signs (DMSs) should display to influence driver speeding behavior. A 155-mi² virtual road network of the Baltimore-Washington Parkway (MD-295) in Maryland was developed for use with a medium-fidelity driving simulator, and 65 participants took part in the study. Six scenarios featured DMSs displaying 2–7 units of information, interchangeably, and a total of 296 simulation sessions were conducted. Mean speeds are calculated over five different phases: the initial speed area, visible area, readable area, lost legibility area and post DMS area. Analysis of variance (ANOVA) and post-hoc analysis showed that participants tend to accelerate as soon as they lose sight of the DMS displaying 2–3 units of information and continue to do so after they pass the DMS. An ordinary least squares (OLS) regression analysis reveals that participants older than 55 slow down the most when they encounter DMSs with 6–7 units of information. Participants in the age group of 26–35 tend to increase speeds, especially when a DMS displays 2–4 units of information. This suggests that the comprehension time is low when there are fewer units of information on a DMS. Too little information may be unclear or ambiguous whereas too much may be hard to comprehend and cause drivers to slow down.

Speeding in a work zone needs to be tackled effectively to enhance work zone safety. This study evaluates the influence of three different work zone signs—speed photo enforced signs, dynamic speed display signs and reduced speed limit signs—on driver speeding behavior using a medium-fidelity driving simulator. The speed photo enforced and reduced speed limit signs were placed at intervals before the start of the work zone and in the work zone. A virtual work zone network was built to replicate the Baltimore–Washington Parkway (MD-295). A total of 66 participants from different socioeconomic backgrounds participated in 264 driving simulation sessions. The speeds were measured over four phases: initial speed area, sign visible area, sign readable area and post sign area. An analysis of variance and post hoc analysis showed that the speed photo enforced sign was the most effective of the three signs. The second speed photo enforced sign within the work zone was more effective at reducing speed than the sign before the start of the work zone, especially in the post sign area. It was observed that female participants tend to slow down more when encountering a speed photo enforcement sign.