Article
Engineering, Industrial
Yasir Ali, Simon Washington, Md Mazharul Haque
Summary: The current reactive road safety assessment lacks a proper methodological framework to assess real-time crash risk at signalized intersections, leading to a deficiency in developing real-time risk mitigation strategies. This study proposes a traffic conflict-based crash estimation technique using a Bayesian modeling framework to estimate crash risk in real-time. The proposed framework is tested using 96 hours of traffic movement video data from a signalized intersection in Queensland, Australia, and demonstrates the suitability of the model for crash risk estimations. The study concludes that the proposed real-time framework effectively estimates rear-end crash risk at the micro-level, enabling proactive safety management and the development of risk mitigation strategies.
Article
Engineering, Civil
Emmanuel Kidando, Angela E. Kitali, Boniphace Kutela, Alican Karaer, Mahyar Ghorbanzadeh, Mohammadreza Koloushani, Eren E. Ozguven
Summary: This study found that real-time traffic events and signal timing data can influence the injury severity of vehicle occupants at intersections, with approach delay and platoon ratio significantly affecting the severity. Additionally, collision manner, occupant seat position, number of vehicles involved, gender, age, lighting conditions, and day of the week also significantly impact vehicle occupant injury.
TRANSPORTATION RESEARCH RECORD
(2022)
Article
Engineering, Civil
Passant Reyad, Tarek Sayed, Mohamed Essa, Lai Zheng
Summary: This paper presents a novel ATSC algorithm that combines reinforcement learning with extreme value theory to achieve real-time traffic safety optimization. Results from real-world traffic video data validation show that the developed algorithm can significantly reduce crash risk at intersection approaches even with low CV market penetration rates.
TRANSPORTATION RESEARCH RECORD
(2022)
Article
Engineering, Civil
Hao Yang, Fawaz Almutairi, Hesham Rakha
Summary: The study focuses on an eco-driving system that optimizes vehicle fuel consumption while traversing consecutive signalized intersections. By implementing the system in large networks and conducting a comprehensive analysis of various variables, optimal demand levels and traffic signal spacings were identified to maximize the algorithm's effectiveness.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2021)
Article
Engineering, Civil
Jinghui Yuan, Mohamed A. Abdel-Aty, Lishengsa Yue, Qing Cai
Summary: In this study, data preparation and modeling were conducted for real-time crash risk prediction at signalized intersections, and factors such as higher cycle volume and overall average flow ratio across lanes were found to increase the odds of crash occurrence. Random undersampling showed better performance in the model results.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2021)
Article
Engineering, Civil
Thanushan Rajeswaran, Bhagwant Persaud, Alireza Jafari Anarkooli
Summary: This study investigates key issues related to the development and application of crash-conflict models, including model specification, conflict definition, model transferability, and application for estimating crash modification factors (CMFs). A case study is conducted using conflicts and the speed of conflicting vehicles generated from the microsimulation of signalized intersections. The results indicate that including the speed variable improves the statistical relationships between crash frequency and surrogate measures, and the models can be reasonably applied to other jurisdictions.
TRANSPORTATION RESEARCH RECORD
(2023)
Article
Computer Science, Artificial Intelligence
Ziran Wang, Kyungtae Han, Prashant Tiwari
Summary: Digital Twin is an emerging technology that is related to Cyber-Physical Systems (CPS) and Internet of Things (IoT). It is used in this study to design a cooperative driving system at non-signalized intersections, allowing connected vehicles to cooperate with each other to cross intersections without full stops.
IEEE TRANSACTIONS ON INTELLIGENT VEHICLES
(2022)
Article
Computer Science, Information Systems
Sakib Mahmud Khan, Mashrur Chowdhury
Summary: This article explores the interaction between CAVs and non-CAVs in mixed traffic and addresses issues that may arise during CAV left-turn operations through the development of a situation-awareness module. The study shows that considering the intent of following vehicles significantly reduces abrupt braking incidents and can lead to notable reductions in travel time for opposing through traffic volumes.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Article
Transportation Science & Technology
Yangang Ren, Guojian Zhan, Liye Tang, Shengbo Eben Li, Jianhua Jiang, Keqiang Li, Jingliang Duan
Summary: This paper proposes an adversarial learning paradigm to enhance the intelligence and robustness of driving policy for signalized intersections with dense traffic flow. It introduces a static path planner to generate trackable candidate paths for different intersection topologies and builds a constrained optimal control problem considering the uncertainty of the driving environment. Adversarial policy gradient is used to solve the problem by introducing disturbances while the driving policy learns to handle the situation through competition. Simulation results show that the trained policy can handle signal lights flexibly and achieve smooth and efficient passing with a humanoid paradigm, while APG greatly improves the resistance to abnormal behaviors, ensuring a high level of safety for autonomous vehicles.
TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES
(2023)
Article
Public, Environmental & Occupational Health
Mercy Lorlonyo Amegah, Emmanuel Kofi Adanu, Thomas Kolawole Ojo, Shaibu Bukari, Filiberto Asare-Akuffo
Summary: This study fills a research gap on motorcyclists' red-light running and helmet use in low and middle-income countries. The study found that approximately 33.1% of motorcyclists ran red lights and 45.4% did not wear a helmet. Helmet use was low and significantly associated with the presence of a pillion passenger and whether the pillion passenger wore a helmet.
TRAFFIC INJURY PREVENTION
(2023)
Article
Engineering, Civil
Xuejian Chen, Juyuan Yin, Keshuang Tang, Ye Tian, Jian Sun
Summary: This study proposes a novel framework for estimating and fusing vehicle trajectory data, aiming to improve the accuracy and smoothness of traffic flow estimation and signal control optimization.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Economics
Georgios Grigoropoulos, Axel Leonhardt, Heather Kaths, Marek Junghans, Michael M. Baier, Fritz Busch
Summary: The popularity of utilitarian bicycling is increasing in urban areas, impacting traffic flow and capacity at intersections. This study quantifies the impact of bicycle traffic on signalized intersections and proposes factors for the reduction in vehicular capacity. Empirical studies and traffic simulation models are used to assess the effects of bicycle infrastructure on traffic efficiency.
TRANSPORTATION RESEARCH PART A-POLICY AND PRACTICE
(2022)
Article
Green & Sustainable Science & Technology
Mostafa Sharafeldin, Ahmed Farid, Khaled Ksaibati
Summary: Signalized intersections are common locations for severe rear-end crashes. This study examined intersection crash data in Wyoming and identified factors such as motorcycle involvement, improper seat belt use, driver's condition and age, road condition, and pavement friction that impact the severity of rear-end crashes at signalized intersections.
Article
Transportation Science & Technology
Nada Mahmoud, Mohamed Abdel-Aty, Qing Cai, Jinghui Yuan
Summary: This study focuses on predicting signal cycle-level through and left-turn movements in real-time at signalized intersections. The GRU model outperformed others when using variables from six previous cycles. The modeling approach enables accurate prediction of traffic movements for different time horizons.
TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES
(2021)
Article
Economics
Roger Lloret-Batlle, Jianfeng Zheng
Summary: Jam density is a difficult parameter to estimate, but it is crucial for traffic signal control and traffic state estimation. This paper proposes several estimators based on trajectory data to estimate jam density and its variance, and evaluates their performance using synthetic and real-world data, with satisfactory results.
TRANSPORTATION RESEARCH PART B-METHODOLOGICAL
(2023)
Review
Otorhinolaryngology
Marie-Helene Uwents, Catherine Jorissen, Angelique van Ombergen, Bieke Dobbels, Raymond van de Berg, Sebastien Janssens de Varebeke, Marc Lammers, Veerle Ross, Olivier Vanderveken, Tom Brijs, Vincent van Rompaey
Summary: The systematic review aimed to identify and evaluate studies on driving performance of dizzy patients or patients with a vestibular disorder. Most included studies reported a negative impact of dizziness or vestibular disorders on self-reported driving ability and car accidents, but some studies did not find any impairment of driving ability. Further research is needed to establish a causal relationship between dizziness and driving ability.
EUROPEAN ARCHIVES OF OTO-RHINO-LARYNGOLOGY
(2022)
Review
Transportation
Pavlos Tafidis, Haneen Farah, Tom Brijs, Ali Pirdavani
Summary: This study investigated and synthesized existing literature on higher levels of AVs' safety implications using a scoping review approach. The focus was on changes in road safety levels after the deployment of AVs in transport networks. Findings suggest that while AVs hold the potential to improve overall safety on roads, the existing evidence is mainly based on assumptions rather than real data.
Article
Computer Science, Artificial Intelligence
Zhenzhen Yang, Feng Liu, Ziyou Gao, Huijun Sun, Jiandong Zhao, Davy Janssens, Geert Wets
Summary: This paper proposes a novel approach for monitoring and evaluating traffic incidents using vehicle GPS data, showing the potential and effectiveness of the technique through tests on two real-world events. With the widespread adoption of GPS devices globally, the application of this method can be more easily transferable, paving the way for a more spatial-temporal sensitive highway network disruption analysis method.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Engineering, Civil
Iftikhar Hussain, Luk Knapen, Tom Bellemans, Davy Janssens, Geert Wets
Summary: The closed group framework is designed to match employees in large companies for carpooling, providing optimal solutions while minimizing user burden and considering dynamic database changes. The system allows users to negotiate within small groups and provide feedback, resulting in a dynamic, user-centered carpooling system.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Engineering, Civil
Wim Ectors, Bruno Kochan, Davy Janssens, Tom Bellemans, Geert Wets
Summary: This study focuses on analyzing the mechanism that leads to the power law distribution, demonstrating that accurate sets of activity sequences can be generated with minimal information required. The lightweight activity sequence generation model can be implemented in transportation models to save time and cost in data collection and model development.
TRANSPORTATION RESEARCH RECORD
(2022)
Article
Green & Sustainable Science & Technology
Kumar Sumit, Veerle Ross, Robert A. C. Ruiter, Evelien Polders, Geert Wets, Kris Brijs
Summary: This study examines the characteristics of fatal road crashes in Manipal from 2008 to 2018 and predicts future trends using time series analysis. The results indicate that speeding is the primary cause of fatal crashes and provide essential leads for intervention programs.
Article
Ergonomics
Nathalie Moreau, Heike Martensen, Stijn Daniels
Summary: The legal Blood Alcohol Concentration (BAC) limit for general drivers in Belgium has been 0.5 g/L since 1994, while no specific limit has been adopted for novice drivers. A study showed that implementing a zero limit for all drivers could lead to an annual reduction of 10 to 17 fatalities, while applying a zero limit only to novice drivers could result in a decrease of 2 to 4 fatalities per year.
ACCIDENT ANALYSIS AND PREVENTION
(2022)
Article
Public, Environmental & Occupational Health
Tooba Batool, An Neven, Christophe Jp. Smeets, Martijn Scherrenberg, Paul Dendale, Yves Vanrompay, Muhammad Adnan, Veerle Ross, Kris Brijs, Geert Wets, Davy Janssens
Summary: This study combines objective monitoring and intervention to improve physical activity in cardiac patients by changing routine travel behavior. The results show a significant increase in active travel score after the intervention, with a particularly significant short-term effect.
JOURNAL OF TRANSPORT & HEALTH
(2022)
Article
Transportation
Lucy Joseph, An Neven, Karel Martens, Opportuna Kweka, Geert Wets, Davy Janssens
Summary: The use of Informal Public Transport (IPT) in Dar es Salaam is associated with mobility difficulties, but the introduction of the Bus Rapid Transit (BRT) system has improved public transportation options along fixed routes. However, low-income individuals still rely on IPT for commuting to multiple destinations outside the BRT corridor. The study also found that individuals are now incorporating BRT into their travel patterns, enabling them to access more amenities and travel greater distances.
TRAVEL BEHAVIOUR AND SOCIETY
(2022)
Article
Ergonomics
Annelies Schoeters, Maxime Large, Martin Koning, Laurent Carnis, Stijn Daniels, Dominique Mignot, Raschid Urmeew, Wim Wijnen, Frits Bijleveld, Martijn van der Horst
Summary: This paper presents the results of a stated choice study that estimated the Willingness-To-Pay of respondents in four European countries to reduce the risk of fatal and serious injuries in road crashes. The study revealed the values of Statistical Life, Statistical Serious Injury, and Value of Time in these countries, as well as differences between countries. The results can be useful for various socioeconomic studies related to cost-benefit analysis and assessments of socioeconomic costs of road crashes.
ACCIDENT ANALYSIS AND PREVENTION
(2022)
Article
Psychology, Developmental
Helene Dirix, Veerle Ross, Kris Brijs, Laura Bertels, Wael Alhajyaseen, Tom Brijs, Geert Wets, Annemie Spooren
Summary: Community participation and the formation of social networks are important for a qualitative life. Transportation plays a vital role, especially for autistic individuals who rely on public transportation for mobility. However, the challenges they face with public bus transport have not been thoroughly studied. This study aimed to give autistic people a platform to express their difficulties and suggest improvements. The findings revealed three main themes: creating predictability, limiting stimuli, and open and accessible communication. The study may lead to a more autism-friendly public transportation environment.
Article
Green & Sustainable Science & Technology
Ehitayhu Hagos, Tom Brijs, Kris Brijs, Geert Wets, Bikila Teklu
Summary: The safety culture and safety climate of transport companies in Ethiopia are insufficient, leading to a higher rate of road traffic crashes involving heavy vehicles. This study aims to assess the existing safety culture and identify intervention methods to improve it. The results show that most transport companies have limited safety culture and implement inconsistent safety measures.
Article
Green & Sustainable Science & Technology
Bart De Vos, Ariane Cuenen, Veerle Ross, Helene Dirix, Kris Brijs, Tom Brijs
Summary: Speeding is a major risk factor in road safety, leading to accidents and extensive consequences. This study aims to investigate the effectiveness of an intelligent speed assistance system for truck drivers. The results show a significant reduction in relevant parameters when drivers received information and warnings about speeding while driving on a rural road with a specific speed limit.
Article
Transportation
Caroline Arien, Giovanni Vanroelen, Veerle Ross, Yanchao Song, Tom Brijs, Geert Wets, Ellen M. M. Jongen, Joris Cornu, Kristof Mollu, Stijn Daniels, Kris Brijs
Summary: Driving simulator data can be sampled either by distance (equally spaced) or time (constant frequency). However, the chosen sampling method may lead to issues when analyzing the data based on point location or zones. This paper demonstrates these issues using five driving simulator datasets. In point-based analysis, the nearest sampled parameter value near the specific point accurately represents its driving parameter value. On the other hand, the analysis of driving parameters in zones requires a different approach. Therefore, the preferred method is to use interpolation techniques rather than raw sampled data to calculate mean parameter values. Moreover, this paper introduces an equivalent time integral formula to compute the mean value of a driving parameter based on distance.
TRANSPORTMETRICA A-TRANSPORT SCIENCE
(2023)
Proceedings Paper
Automation & Control Systems
Muhammad Wisal Khattak, Hans De Backer, Pieter De Winne, Tom Brijs, Ali Pirdavani
Summary: Transportation safety researchers use crash prediction models to examine roadway safety performance. The traditional negative binomial model is popular but performs poorly with highly overdispersed data. The Poissoninverse Gaussian regression model is a potential alternative for modeling highly dispersed data effectively.
INTERNATIONAL CONFERENCE ON TRANSPORTATION AND DEVELOPMENT 2022: TRANSPORTATION SAFETY
(2022)