Search

Quick and reliable automatic detection of traffic accidents is of paramount importance to save human lives in transportation systems. However, automatically detecting when accidents occur has proven challenging, and minimizing the time to detect accidents (TTDA) by using traditional features in machine learning (ML) classifiers has plateaued. We hypothesize that accidents affect traffic farther from the accident location than previously reported. Therefore, leveraging traffic signatures from neighboring sensors that are adjacent to accidents should help improve their detection. We confirm this hypothesis by using verified ground-truth accident data, traffic data from radar detection system sensors, and light and weather conditions and show that we can minimize the TTDA while maximizing classification performance by considering spatiotemporal features of traffic. Specifically, we compare the performance of different ML classifiers (i.e, logistic regression, random forest, and XGBoost) when controlling for different numbers of neighboring sensors and TTDA horizons. We use data from interstates 75 and 24 in the metropolitan area that surrounds Chattanooga, TN. Our results show that the XGBoost classifier produces the best results by detecting accidents as quickly as 1.0 min after their occurrence with an area under the receiver operating characteristic curve of up to 83% and an average precision of up to 49%. We describe limitations, open challenges, and how the proposed framework can be used for quicker operational accident detection. • We show the utility of using adjacent traffic sensor measurements to reduce TTDA. • XGBoost classifier produces the best results as quickly as 1.0 min. • XGBoost classifier produces the best results (AUC-ROC 83% and AUC-PR 49%). • We show that features of both upstream and downstream sensors are the most important. • We report results on two different datasets (i.e., I-75 and I-24) in Chattanooga, TN. [ABSTRACT FROM AUTHOR]

Maritime Traffic Situation (MTS) is crucial for enhancing the efficiency and safety of maritime navigation, with the complexity of ship traffic serving as a key aspect of identifying and managing maritime risks. This paper proposes a ship traffic complexity measurement model for MTS. The proposed model quantifies the complexity between ships by evaluating factors such as density, proximity factor, and mitigation index, and then integrates these evaluations into multiple ships interaction scenarios to enable a comprehensive assessment of ship traffic complexity. The results of case study have fully demonstrated the model's capability in identifying potential collision risks during the interaction processes of multiple ships situation within a navigational sea. Research on the ship traffic complexity measurement model can provide strong theoretical and practical support for maritime traffic management and navigational safety. • Building on an in-depth exploration of the intrinsic attributes of ships, this paper introduces ship domain (SD) into the calculation of proximity factor within the ship traffic complexity measurement model. By constructing different SDs for different ships, the model can more accurately assess collision risks. • Incorporating the mitigation index, based on ships' speed and course parameters, into the ship traffic complexity measurement model enables the quantification of ships' movement trends within specified temporal and spatial domains. This integration allows the model to accurately assess the mitigation complexity of ships in various MTS, taking into account their dynamics. • Expanding the ship traffic complexity measurement model from focusing on pairs of ships to encompassing the quantification of traffic complexity in multi-ship interactive scenarios has significantly enhanced the model's practical application value. [ABSTRACT FROM AUTHOR]

• A systematic review of recently published articles on traffic conflict measurements based on PRISMA. • Validation of SSMs is needed in heterogeneous traffic environment, especially in the context of developing countries. • Importance of SSMs in new technologies like driver warning systems in driverless cars. • SSMs are not adequately capturing the likelihood and severity of accidents. • Study on the driver behavior at the unsignalized intersections. In recent times, the assessment of unsignalized intersection safety has received significant research attention because of the complex and diverse traffic movements and driving behaviour at such locations. However, priority traffic regulations are not well followed in comparison to the unsignalized junctions, which leads to more conflicts. Additionally, the severity of conflicts increases with continuous traffic manoeuvres, including right-turns and through traffic, combined with different driving behaviours. Several studies have compared crash-based analysis to proactive traffic safety measures. Current research outcomes imply that surrogate safety measures (SSMs) have the potential to elucidate the sequence of events that result in collisions, their underlying causes, and their outcomes. Therefore, to further understand the appropriateness of SSMs, further study is required based on heterogeneity in traffic along with driver behaviour that incorporates turning vehicle factors. This study presents an all-inclusive evaluation of the recent advancements in SSMs and their practical implementation, with a particular emphasis on unsignalized intersections in developing nations. The findings of this investigation would be helpful in identifying the appropriate safety indicators for evaluating traffic safety at unsignalized intersections. [ABSTRACT FROM AUTHOR]