Amr Elfar

Investigating the Effects of Reserved Lanes for Autonomous Vehicles on Congestion and Travel Time Reliability

Autonomous vehicles are expected to influence daily travel significantly. Despite autonomous vehicles’ potential to enhance safety and to reduce congestion, energy consumption, and emissions, many studies suggest that the system-level effects will be minimal at low market penetration rates. Introducing reserved lanes for autonomous vehicles is one potential approach to address this limitation because these lanes increase autonomous vehicles’ density. However, preventing regular vehicles from using specific lanes can significantly increase congestion in other lanes. Accordingly, this study explored the potential effects of reserving one lane for autonomous vehicles on traffic flow dynamics and travel time reliability. A two-lane hypothetical segment with an on-ramp and a four-lane highway segment in Chicago, Illinois, was simulated under different market penetration rates of autonomous vehicles. Three strategies were evaluated: (a) mandatory use of the reserved lane by autonomous vehicles, (b) optional use of the reserved lane by autonomous vehicles, and (c) limiting autonomous vehicles to operate autonomously in the reserved lane. Policies based on combinations of these strategies were simulated. It was found that optional use of the reserved lane without any limitation on the type of operation could improve congestion and could reduce the scatter in a fundamental diagram. Throughput analysis showed the potential benefit of reserving a lane for autonomous vehicles at market penetration rates of more than 50% for the two-lane highway and 30% for the four-lane highway. Travel time reliability analysis revealed that the optional use of the reserved lane was also significantly beneficial.

Traffic Shockwave Detection in a Connected Environment using the Speed Distribution of Individual Vehicles

Traffic shockwaves reflect a transition from the free-flow traffic state to the congested state. They create potentially unsafe situations for drivers, increase travel time, and significantly reduce freeway capacity. Several shockwave detection methods based on loop detector data and other traditional databases have been around for years. However, these methods face certain accuracy and reliability issues, many of which are due to the nature and accuracy of available data. Connected-vehicles technology is expected to provide reliable and accurate data about individual vehicles that can be potentially used for shockwave detection. Accordingly, this paper presents a novel method to identify shockwave formation and track its propagation based on the speed distribution of individual vehicles available through connected-vehicles technology. In addition, this paper analyzes the impact of partial connectivity on shockwave identification and compares the accuracy of the proposed method to a wavelet transformation-based method. Vehicle trajectories from the Next Generation Simulation (NGSIM) US-101 dataset were analyzed. The analysis shows a consistent pattern in which shockwave formation, indicated by a drop in speed propagating over space and time, is associated with a sharp increase in the value of speed standard deviation (SSD). Furthermore, the analysis shows that shockwaves can be accurately identified using vehicle trajectory data from connected vehicles at minimum 30% market penetration rates. Finally, the results show that the SSD of individual vehicles is more responsive to shockwave formation than the mean speed wavelet transformation, which can lead to improved shockwave detection accuracy.

Machine Learning Approach to Short-Term Traffic Congestion Prediction in a Connected Environment

Traffic congestion is a complex phenomenon triggered by a combination of multiple interacting factors. One of the main factors is the disturbances caused by individual vehicles, which cannot be identified in aggregate traffic data. Advances in vehicle wireless communications present new opportunities to measure traffic perturbations at the individual vehicle level. The key question is whether it is possible to find the relationship between these perturbations and shockwave formation and utilize this knowledge to improve the identification and prediction of congestion formation. Accordingly, this paper explores the use of three machine learning techniques, logistic regression, random forests, and neural networks, for short-term traffic congestion prediction using vehicle trajectories available through connected vehicles technology. Vehicle trajectories provided by the Next Generation SIMulation (NGSIM) program were utilized in this study. Two types of predictive models were developed in this study: (1) offline models which are calibrated based on historical data and are updated (re-trained) whenever significant changes occur in the system, such as changes/updates to the infrastructure, and (2) online models which are calibrated using historical data and updated regularly using real-time information on prevailing traffic conditions obtained through V2V/V2I communications. Results show that the accuracy of the models built in this study to predict the congested traffic state can reach 97%. The models presented can be used in various potential applications including improving road safety by warning drivers of upcoming traffic slowdowns and improving mobility through integration with traffic control systems.

Exploring the Role of Social Media Platforms in Informing Trip Planning: Case of Yelp.com

Understanding how travelers make mobility decisions has always been central to transportation studies. The growing availability of information and communication technologies in everyday life and their role in conveying more recent, relevant, and customized information have substantially changed the context within which trip decisions are made. Whether travelers are actively seeking pretrip information or merely surfing the web, they have access through social media to user-generated information that may affect their mobility decisions. This study forms an exploratory step in understanding how one such social media platform, Yelp.com, designed to allow users to review and rate their experiences at any visited business, can serve as an information source for activity and trip planning in the pretrip process. In particular, the study explored (a) the relative depth, (b) the sentiment associated with, and (c) the type of information in transport-related reviews on Yelp.com. This research has implications for the study of travel behavior in a highly connected environment and can inform efforts to design information and communication technology tools aimed at affecting behavior.