Traffic incidents cause severe problems on roadways. About 6.3 million highway crashes are reported annually only in the United States, among which more than 32,000 are fatal crashes. Reducing the risk of traffic incidents is key to effective traffic incident management (TIM). Quick detection of unexpected traffic incidents on roadways contribute to quick clearance and hence improve safety. Existing techniques for the detection of freeway incidents are not reliable. This study focuses on exploring the potential of emerging connected vehicles (CV) technology in automated freeway incident detection in the mixed traffic environment. The study aims at developing an automated freeway incident detection algorithm that will take advantage of the CV technology in providing fast and reliable incident detection. Lee Roy Selmon Expressway was chosen for this study because of the THEA CV data availability. The findings of the study show that emerging CV technology generates data that are useful for automated freeway incident detection, although the market penetration rate was low (6.46%). The algorithm performance in terms of detection rate (DR) and false alarm rate (FAR) indicated that CV data resulted into 31.71% DR and zero FAR while HERE yielded a 70.95% DR and 9.02% FAR. Based on Pearson's correlation analysis, the incidents detected by the CV data were found to be similar to the ones detected by the HERE data. The statistical comparison by ANOVA shows that there is a difference in the algorithm's detection time when using CV data and HERE data. 17.07% of all incidents were detected quicker when using CV data compared to HERE data, while 7.32% were detected quicker when using HERE data compared to CV data.

Real-time incident detection on freeways plays an important part in any modern traffic management operation by maximizing road system performance. The US Department of Transportation (US-DOT) estimates that over half of all congestion events are caused by highway incidents rather than by rush-hour traffic in big cities. An effective incident detection and management operation cannot prevent incidents, however, it can diminish the impacts of non-recurring congestion problems. The main purpose of real-time incident detection is to reduce delay and the number of secondary accidents, and to improve safety and travel information during unusual traffic conditions. The majority of automatic incident detection algorithms are focused on identifying traffic incident patterns but do not adequately investigate possible similarities in patterns observed under incident-free conditions. When traffic demand exceeds road capacity, density exceeds critical values and traffic speed decreases, the traffic flow process enters a highly unstable regime, often referred to as “stop-and-go” conditions. The most challenging part of real-time incident detection is the recognition of traffic pattern changes when incidents happen during stop-and-go conditions. Recently, short-term freeway congestion detection algorithms have been proposed as solutions to real-time incident detection, using procedures known as dynamic time warping (DTW) and the support vector machine (SVM). Some studies have shown these procedures to produce higher detection rates than Artificial Intelligence (AI) algorithms with lower false alarm rates. These proposed methods combine data mining and time series classification techniques. Such methods comprise interdisciplinary efforts, with the confluence of a set of disciplines, including statistics, machine learning, Artificial Intelligence, and information science. A literature review of the methodology and application of these two models will be presented in the following chapters. SVM, Naïve Bayes (NB), and Random Forest classifier models incorporating temporal data and an ensemble technique, when compared with the original SVM model, achieve improved detection rates by optimizing the parameter thresholds. The main purpose of this dissertation is to examine the most robust algorithms (DTW, SVM, Naïve Bayes, Decision Tree, SVM Ensemble) and to develop a generalized automatic incident detection algorithm characterized by high detection rates and low false alarm rates during peak hours. In this dissertation, the transferability of the developed incident detection model was tested using the Dallas and Miami field datasets. Even though the primary service of urban traffic control centers includes detecting incidents and facilitating incident clearance, estimating freeway incident durations remains a significant incident management challenge for traffic operations centers. As a next step this study examines the effect of V/C (volume/capacity) ratio, level of service (LOS), weather condition, detection mode, number of involved lanes, and incident type on the time duration of traffic incidents. Results of this effort can benefit traffic control centers improving the accuracy of estimated incident duration, thereby improving the authenticity of traveler guidance information.

This report describes optimization techniques which have been developed and applied for the evaluation of design and operations of freeway incident detection-service systems. The report has four major parts: (1) analysis and design of stationary service systems; (2) analysis and design of incident detection algorithms; (3) analysis and design of incident response systems; and (4) analysis and design of freeway on-ramp traffic-responsive control methodology for normal and incident conditions.

This book combines comprehensive multi-angle discussions on fully connected and automated vehicle highway implementation. It covers the current progress of the works towards autonomous vehicle highway development, which encompasses the discussion on the technical, social, and policy as well as security aspects of Connected and Autonomous Vehicles (CAV) topics. This, in return, will be beneficial to a vast amount of readers who are interested in the topics of CAV, Automated Highway and Smart City, among many others. Topics include, but are not limited to, Autonomous Vehicle in the Smart City, Automated Highway, Smart-Cities Transportation, Mobility as a Service, Intelligent Transportation Systems, Data Management of Connected and Autonomous Vehicle, Autonomous Trucks, and Autonomous Freight Transportation. Brings together contributions discussing the latest research in full automated highway implementation; Discusses topics such as autonomous vehicles, intelligent transportation systems, and smart highways; Features contributions from researchers, academics, and professionals from a broad perspective.

Recently, with the development of connected vehicles and mobile sensing technologies, vehicle-based data become much easier to obtain. However, only few studies have investigated the application of this kind of novel data to real-time traffic safety evaluation. This dissertation aims to conduct a series of real-time traffic safety studies by integrating all kinds of available vehicle-based data sources. First, this dissertation developed a deep learning model for identifying vehicle maneuvers using data from smartphone sensors (i.e., accelerometer and gyroscope). The proposed model was robust and suitable for real-time application as it required less processing of smartphone sensor data compared with the existing studies. Besides, a semi-supervised learning algorithm was proposed to make use of the massive unlabeled sensor data. The proposed algorithm could alleviate the cost of data preparation and improve model transferability. Second, trajectory data from 300 buses were used to develop a real-time crash likelihood prediction model for urban arterials. Results from extensive experiments illustrated the feasibility of using novel vehicle trajectory data to predict real-time crash likelihood. Moreover, to improve the model’s performance, data fusion techniques were proposed to integrated trajectory data from various vehicle types. The proposed data fusion techniques significantly improved the accuracy of crash likelihood prediction in terms of sensitivity and false alarm rate. Third, to improve pedestrian and bicycle safety, different vehicle-based surrogate safety measures, such as hard acceleration, hard deceleration, and long stop, were proposed for evaluating pedestrian and bicycle safety using vehicle trajectory data. In summary, the results from this dissertation can be further applied to real-time safety applications (e.g., real-time crash likelihood prediction and visualization system) in the context of proactive traffic management.

This dissertation proposes a framework on how to process and analyze the data available from the driver, the vehicle and the road infrastructure i.e. data streams in real-time. Particularly, it conceptualize measures of driver, vehicle and road infrastructure performance and process the volatilities in data streams from sensors. It also provides a framework for real-time identification of anomalies, linking them with alerts, warnings and control assists. We explore different measures of driving volatility used to explain crash frequencies at intersections through developing a unique database that integrates intersection crash and inventory data with real-world Basic Safety Messages logged by connected vehicles. We introduce location-based volatility (LBV) as a proactive safety measure, quantifying variability in instantaneous driving decisions at intersections. Such an analysis is fundamental towards proactive intersection safety management. In addition, Markov Decision Process (MDP) framework is used to learn observed behavior by analyzing instantaneous driving decisions of acceleration, deceleration, and maintaining constant speed. Moreover, the developed measures of volatilities are applied to speed profiles from the Strategic Highway Research Program 2 (SHRP2) Naturalistic Driving Study (NDS) to come up with the most accurate crash-prediction model with used for real-time driving assist warning generation. Finally, by incorporating the data from the driver, vehicle and infrastructure into the analysis, the impact of detailed pre-crash driving behavior and recently developed measures of driving volatility on crash and near-crash risks is investigated. The knowledge gained from studying individual driving behaviors can be used to generate alerts and warnings for the driver of the host vehicle and to be passed via connected vehicle technology with the purpose of improving safety. The methods applied in this dissertation can form a foundation for human driver behavior prediction and personally revealed choice extraction. They also can help proactively identify locations with high levels of driving volatility (i.e., hot spots where crashes are waiting to happen) as candidates for safety improvements. Proactive warnings and alerts can be generated about potential hazards and transmitted to drivers via connected vehicle technologies such as road-side equipment, increasing drivers' situational and safety awareness.