143 results found
Sha H, Singh MK, Haouari R, et al., 2024, Network-wide safety impacts of dedicated lanes for connected and autonomous vehicles., Accid Anal Prev, Vol: 195
Cooperative, Connected and Automated Mobility (CCAM) enabled by Connected and Autonomous Vehicles (CAVs) has potential to change future transport systems. The findings from previous studies suggest that these technologies will improve traffic flow, reduce travel time and delays. Furthermore, these CAVs will be safer compared to existing vehicles. As these vehicles may have the ability to travel at a higher speed and with shorter headways, it has been argued that infrastructure-based measures are required to optimise traffic flow and road user comfort. One of these measures is the use of a dedicated lane for CAVs on urban highways and arterials and constitutes the focus of this research. As the potential impact on safety is unclear, the present study aims to evaluate the safety impacts of dedicated lanes for CAVs. A calibrated and validated microsimulation model developed in AIMSUN was used to simulate and produce safety results. These results were analysed with the help of the Surrogate Safety Assessment Model (SSAM). The model includes human-driven vehicles (HDVs), 1st generation and 2nd generation autonomous vehicles (AVs) with different sets of parameters leading to different movement behaviour. The model uses a variety of cases in which a dedicated lane is provided at different type of lanes (inner and outer) of highways to understand the safety effects. The model also tries to understand the minimum required market penetration rate (MPR) of CAVs for a better movement of traffic on dedicated lanes. It was observed in the models that although at low penetration rates of CAVs (around 20%) dedicated lanes might not be advantageous, a reduction of 53% to 58% in traffic conflicts is achieved with the introduction of dedicated lanes in high CAV MPRs. In addition, traffic crashes estimated from traffic conflicts are reduced up to 48% with the CAVs. The simulation results revealed that with dedicated lane, the combination of 40-40-20 (i.e., 40% human-driven - 40% 1st ge
Wang X, Ye C, Quddus M, et al., 2023, Pedestrian safety in an automated driving environment: calibrating and evaluating the responsibility-sensitive safety model, Accident Analysis and Prevention, Vol: 192, ISSN: 0001-4575
The severity of vehicle-pedestrian crashes has prompted authorities worldwide to concentrate on improving pedestrian safety. The situation has only become more urgent with the approach of automated driving scenarios. The Responsibility-Sensitive Safety (RSS) model, introduced by Mobileye®, is a rigorous mathematical model developed to facilitate the safe operation of automated vehicles. The RSS model has been calibrated for several vehicle conflict scenarios; however, it has not yet been tested for pedestrian safety. Therefore, this study calibrates and evaluates the RSS model for pedestrian safety using data from the Shanghai Naturalistic Driving Study. Nearly 400 vehicle-pedestrian conflicts were extracted from 8,000 trips by the threshold and manual check method, and then divided into 16 basic scenarios in three categories. Because crossing conflicts were the most serious and frequent, they were reproduced in MATLAB's Simulink with each vehicle replaced with a virtual automated vehicle loaded with the RSS controller module. With the objectives of maximizing safety and minimizing conservativeness, the non-dominated sorting genetic algorithm II was applied to calibrate the RSS model for vehicle-pedestrian conflicts. The safety performance of the RSS model was then compared with that of the commonly used active safety function, autonomous emergency braking (AEB), and with human driving. Findings verified that the RSS model was safer in vehicle-pedestrian conflicts than both the AEB model and human driving. Its performance also yielded the best test results in producing smooth and stable driving. This study provides a reliable reference for the safe control of automated vehicles with respect to pedestrians.
Singh MK, Haouari R, Papazikou E, et al., 2023, Examining parking choices of connected and autonomous vehicles, Transportation Research Record, Vol: 2677, Pages: 589-601, ISSN: 0361-1981
Raising parking charges is a measure that restricts the use of private vehicles. With the introduction of connected and autonomous vehicles (CAVs), the demand for parking has the potential to reduce as CAVs may not park at ‘pay to park’ areas as they are able to “cruise” or return home. However, it might not be financially feasible for them to return to their origin if the destination region is far away. Therefore, the question is: how could we develop parking policies in the CAVs era? To determine the best parking strategy for CAVs, four scenarios were tested in this paper: (i) enter and park within the destination area, (ii) enter, drop off, and return to the origin, (iii) enter, drop off, and return to outside parking and (iv) enter and drive around. Since real-world parking demand data for CAVs are not available, a simulation model of the road network in Santander (Spain) was employed to collect data on both CAV operations (e.g., conservative versus aggressive behaviors) and parking choices. Multinomial logistic regression model was used to identify the best parking option for CAVs. Performance indicators such as traffic, emissions, and safety were employed to compare the performance of a range of parking alternatives. It was found that the balanced scenario (i.e., combination of all parking choices) performs better with the greatest change in delay (around 32%). With 100% CAV market penetration, traffic crashes were reduced by 67%. This study will help local authorities formulate parking policies so that CAVs can park efficiently.
Schumann H-H, Haitao H, Quddus M, 2023, Passively generated big data for micro-mobility: state-of-the-art and future research directions, Transportation Research Part D: Transport and Environment, Vol: 121, ISSN: 1361-9209
The sharp rise in popularity of micro-mobility poses significant challenges in terms of ensuring its safety, addressing its social impacts, mitigating its environmental effects, and designing its systems. Meanwhile, micro-mobility is characterised by its richness in passively generated big data that has considerable potential to address the challenges. Despite an increase in recent literature utilising passively generated micro-mobility data, knowledge and findings are fragmented, limiting the value of the data collected. To fill this gap, this article provides a timely review of how micro-mobility research and practice have exploited passively generated big data and its applications to address major challenges of micro-mobility. Despite its clear advantages in coverage, resolution, and the removal of human errors, passively generated big data needs to be handled with consideration of bias, inaccuracies, and privacy concerns. The paper also highlights areas requiring further research and provides new insights for safe, efficient, sustainable, and equitable micro-mobility.
Formosa N, Quddus M, Man CK, et al., 2023, Appraising machine and deep learning techniques for traffic conflict prediction with class imbalance, Data Science for Transportation, Vol: 5, ISSN: 2948-135X
Predicting traffic conflicts is pivotal for vehicle-based active safety system to prevent crashes. Yet, conflict prediction is a challenging task as correct prediction depends on the nature of data and techniques employed. Moreover, traffic conflicts data are naturally imbalanced with traffic conflicts being the minority class. Working with imbalanced dataset might result in biased and inaccurate predictions. Therefore, this study aims to appraise machine learning and deep learning techniques systematically, to identify the optimal technique which can reliably predict real-time traffic conflicts by making use of cost-sensitive learning. Five machine learning techniques were optimised and utilised including: logistic regression (LR), Support vector machines (SVM), deep neural networks (DNN), long short-term memory (LSTM) and LSTM convolutional neural network (LSTM-CNN) to appraise their predictability performance using a large, imbalanced, and disaggregated traffic dataset. Unlike existing studies, a wide range of interconnected factors are employed for real-time traffic conflict prediction to provide a more reliable prediction outcome. A large heterodox dataset was gathered from the M1 motorway in the UK to evaluate these techniques. Results suggested that DNN outperform other techniques in predicting conflicts with 0.72 sensitivity at 5% false alarm rate. Such promising results reflect that DNNs can be further applied to deepen our understanding in predicting traffic conflicts design more reliable primary safety systems for intelligent vehicles. Moreover, exploring state-of-the-art classification techniques with class imbalance on big data is significant to the future of big data analytics.
Cai B, Quddus M, Wang X, et al., 2023, New modeling approach for predicting disaggregated time-series traffic crashes, Transportation Research Record, ISSN: 0361-1981
Short-window crash prediction is a fundamental step in proactive traffic safety management that can monitor traffic conditions in real time, identify unsafe traffic dynamics, and implement suitable interventions for traffic conflicts. Short-window (e.g., hourly) traffic-collision count data, however, exhibits excessive zeros and serial autocorrelation. Most of the commonly used regression-based models fail to address excessive zeros and temporal structure simultaneously in hourly traffic-collision prediction. For example, Hurdle models and zero-inflated (ZI) models can address the overdispersion issue caused by excessive zeros but lack power to control the significant spatiotemporal characteristics inherent in the time-based collision data. To overcome these issues simultaneously, this paper develops a novel statistical model termed Zero-Inflated Logarithmic link for count Time series (ZILT) which is based on the framework of ZI models. Covariates (e.g., speed, vehicle type, and traffic volume) were extracted through deep-learning computer vision methods in vehicle detection and tracking on the image space. This new statistical model (i.e., ZILT) performs better at solving the issues of excessive zeros and serial dependencies. The prediction accuracy of the ZILT model improved by around 5% in relation to zero-inflated Poisson (ZIP) and Hurdle models. Results show that traffic crashes happening in the previous hour and other covariates such as truck-to-car ratio, holiday effect, traffic flow, and speed have significant influence on collision occurrence. Findings from this study could be utilized by relevant transport agencies in developing engineering interventions and countermeasures to proactively manage road safety.
Wang X, Azati Y, Quddus M, et al., 2023, Statistical analysis of traffic crashes on mountainous freeway tunnel sections, Transportation Research Record, ISSN: 0361-1981
Tunnels on mountainous freeways are affected by abrupt changes in brightness, complex geometric alignments, heavy traffic flow, bad weather, and other factors, some of which contribute toward tunnels having more traffic crashes than other sections of the freeway. Previous research, however, has given limited attention to tunnel length and heterogeneity in the parts of the tunnels, such as at the tunnels’ entrance and exit zones. Focusing on 36 tunnels on the Guidu Freeway in China’s Guizhou Province, this study collects data on crashes and their influencing factors over 2 years (2020–2021), constructs a negative binomial panel data random effects model, and analyzes single-vehicle crashes, multi-vehicle crashes, and total crashes. The results show that: 1) multi-vehicle crashes occur throughout the tunnel sections, 2) crashes are more likely to occur in long tunnel sections, 3) the crash frequency from the tunnel entrance zone to the mid zone is higher than in other areas of the tunnel, 4) the crash frequency is higher for circular curve/easy curve tunnel sections than for straight tunnel sections, 5) the crash frequency is higher for downhill and concave curve sections than for flat sections, 6) the crash frequency increases with heavy traffic flow and adverse weather conditions, and 7) the crash frequency increases as road surface skidding resistance and ride quality decrease. These findings can provide theoretical support for engineering improvement and the formulation and revision of specifications for designing freeway tunnel sections, especially in mountainous areas.
Yi D, Fang H, Hua Y, et al., 2022, Improving Synthetic to Realistic Semantic Segmentation With Parallel Generative Ensembles for Autonomous Urban Driving, IEEE Transactions on Cognitive and Developmental Systems, Vol: 14, Pages: 1496-1506, ISSN: 2379-8920
Semantic segmentation is paramount for autonomous vehicles to have a deeper understanding of the surrounding traffic environment and enhance safety. Deep neural networks (DNNs) have achieved remarkable performances in semantic segmentation. However, training such a DNN requires a large amount of labeled data at the pixel level. In practice, it is a labor-intensive task to manually annotate dense pixel-level labels. To tackle the problem associated with a small amount of labeled data, deep domain adaptation (DDA) methods have recently been developed to examine the use of synthetic driving scenes so as to significantly reduce the manual annotation cost. Despite remarkable advances, these methods, unfortunately, suffer from the generalizability problem that fails to provide a holistic representation of the mapping from the source image domain to the target image domain. In this article, we, therefore, develop a novel ensembled DDA to train models with different upsampling strategies, discrepancy, and segmentation loss functions. The models are, therefore, complementary with each other to achieve better generalization in the target image domain. Such a design does not only improves the adapted semantic segmentation performance but also strengthens the model reliability and robustness. Extensive experimental results demonstrate the superiorities of our approach over several state-of-the-art methods.
Yang K, Quddus M, Antoniou C, 2022, Developing a new real-time traffic safety management framework for urban expressways utilizing reinforcement learning tree, ACCIDENT ANALYSIS AND PREVENTION, Vol: 178, ISSN: 0001-4575
Man CK, Quddus M, Theofilatos A, et al., 2022, Wasserstein generative adversarial network to address the imbalanced data problem in real-time crash risk prediction, IEEE Transactions on Intelligent Transportation Systems, Vol: 23, Pages: 23002-23013, ISSN: 1524-9050
Real-time crash risk prediction models aim to identify pre-crash conditions as part of active traffic safety management. However, traditional models which were mainly developed through matched case-control sampling have been criticised due to their biased estimations. In this study, the state-of-art class balancing method known as the Wasserstein Generative Adversarial Network (WGAN) was introduced to address the class imbalance problem in the model development. An extremely imbalanced dataset consisted of 257 crashes and over 10 million non-crash cases from M1 Motorway in United Kingdom for 2017 was then utilized to evaluate the proposed method. The real-time crash prediction model was developed by employing Deep Neural Network (DNN) and Logistic Regression (LR). Crash predictions were performed under different crash to non-crash ratios where synthetic crashes were generated by Wasserstein Generative Adversarial Network (WGAN), Synthetic Minority Over-sampling Technique (SMOTE) and Adaptive Synthetic (ADASYN) sampling respectively. Comparisons were then made with algorithmic-level class balancing methods such as cost-sensitive learning and ensemble methods. Our findings suggest that WGAN clearly outperforms other oversampling methods in terms of handling the extremely imbalanced sample and the DNN model subsequently produces a crash prediction sensitivity of about 70% with a 5% false alarm rate. Based on the findings of this study, proactive traffic management strategies including Variable Speed Limit (VSL) and Dynamic Messing Signs (DMS) could be deployed to reduce the probability of crash occurrence.
Formosa N, Quddus M, Ison S, et al., 2022, A new modelling approach for predicting vehicle-based safety threats, IEEE Transactions on Intelligent Transportation Systems, Vol: 23, Pages: 18175-18185, ISSN: 1524-9050
Existing autonomous driving systems of intelligent vehicles such as advanced driver assistant systems (ADAS) assess and quantify the level of potential safety threats. However, they may not be able to plan the best response to unexpected dangerous situations and do not have the ability to cope with uncertainties since not all vehicles can always keep a safe gap from preceding vehicles and drive at a desired velocity. Previous research has not taken such uncertainties into account, it is, therefore, necessary to develop models which are not restricted by the predefined movement patterns of a vehicle. Existing systems are based on a model that estimates the threat level based only on one factor Time-To-Collision (TTC). This approach is limited since it cannot handle all scenarios and ignores all uncertainties. To overcome these limitations, this paper utilised deep learning to develop a range of models that rely on a group of factors to reliably estimate the threat level and predict conflicts under uncertainty using the concept of looming '. Comparative analyses were undertaken by incorporating new varying input factors to each model (e.g., surrogate safety measures, vehicle kinematics, macroscopic traffic data). Real-world experiments demonstrated that adding new factors increases the reliability and sensitivity of the models. Results also indicated that the models that consider looming provide low false alarm rate extending their applications for a wider spectrum of traffic scenarios. This is paramount for ADAS as uncertainties are inherent in the deployment of connected and autonomous vehicles in a mixed traffic stream.
Alotaibi S, Quddus M, Morton C, et al., 2022, Transport investment, railway accessibility and their dynamic impacts on regional economic growth, Research in Transportation Business and Management, Vol: 43, ISSN: 2210-5395
This paper examines the impact of large-scale transport investment and the resulting increase of accessibility on the Gross Domestic Product (GDP) in the Kingdom of Saudi Arabia (KSA). Spatial and temporal economic data for the 13 regions of the country from 1999 to 2018 are analyzed using static and dynamic panel data models. Both first difference GMM (Generalised Method of Moments) models and system GMM models fit the data satisfactorily, with system GMM models generally offering better modelling accuracies. The results show that the elasticity of the one-year lagged GDP variable is positive and statistically significant in all specifications considered, indicating the presence of a dynamic pattern toward economic growth. Generally, the value of transport investment for the one-year lag shows positive and significant statistical relationships with regional GDP among several specifications of the dynamic models. On the other hand, railway accessibility value presents positive and significant impact on GDP in two years lag. Our study finds that the impact of the monetary value of transport investment, in general, was manifested immediately in the following year. However, the railway accessibility improvement requires a period to deliver its benefits.
Enoch M, Monsuur F, Palaiologou G, et al., 2022, When COVID-19 came to town: Measuring the impact of the coronavirus pandemic on footfall on six high streets in England, Environment and Planning B: Urban Analytics and City Science, Vol: 49, Pages: 1091-1111, ISSN: 2399-8083
Town centres in the economically developed world have struggled in recent years to attract sufficient visitors to remain economically sustainable. However, decline has not been uniform, and there is considerable variation in how different town centres have coped with these challenges. The arrival of the coronavirus (COVID-19) pandemic public health emergency in early 2020 has provided an additional reason for people to avoid urban centres for a sustained period. This paper investigates the impact of coronavirus on footfall in six town centres in England that exhibit different characteristics. It presents individual time series intervention model results based on data collected from Wi-fi footfall monitoring equipment and secondary sources over a 2-year period to understand the significance of the pandemic on different types of town centre environment. The data show that footfall levels fell by 57%–75% as a result of the lockdown applied in March 2020 and have subsequently recovered at different rates as the restrictions have been lifted. The results indicate that the smaller centres modelled have tended to be less impacted by the pandemic, with one possible explanation being that they are much less dependent on serving longer-distance commuters and on visitors making much more discretionary trips from further afield. It also suggests that recovery might take longer than previously thought. Overall, this is the first paper to study the interplay between footfall and resilience (as opposed to vitality) within the town centre context and to provide detailed observations on the impact of the first wave of coronavirus on town centres’ activity.
Man CK, Quddus M, Theofilatos A, 2022, Transfer learning for spatio-temporal transferability of real-time crash prediction models., Accid Anal Prev, Vol: 165
Real-time crash prediction is a heavily studied area given their potential applications in proactive traffic safety management in which a plethora of statistical and machine learning (ML) models have been developed to predict traffic crashes in real-time. However, one of the fundamental issues relating to the application of these models is spatio-temporal transferability. The present paper attempts to address this gap of knowledge by combining Generative Adversarial Network (GAN) and transfer learning to examine the transferability of real-time crash prediction models under an extremely imbalanced data setting. Initially, a baseline model was developed using Deep Neural Network (DNN) with crash and microscopic traffic data collected from M1 Motorway in the UK in 2017. The dataset utilised in the baseline model is naturally imbalanced with 257 crash cases and 16,359,163 non-crash cases. To overcome data imbalance issue, Wasserstein GAN (WGAN) was utilised to generate synthetic crash data. Non-crash data were randomly undersampled due to computational limitations. The calibrated model was then applied to predict traffic crashes for five other datasets obtained from M1 (2018), M4 (2017 & 2018 separately) and M6 Motorway (2017 & 2018 separately) by using transfer learning. Model transferability was compared with standalone models and direct transfer from the baseline model. The study revealed that direct transfer is not feasible. However, models become transferable temporally, spatially, and spatio-temporally if transfer learning is applied. The predictability of the transferred models outperformed existing studies by achieving high Area Under Curve (AUC) values ranging between 0.69 and 0.95. The best transferred model can predict nearly 95% crashes with only a 5% false alarm rate by tuning thresholds. Furthermore, the performances of transferred models are on par with or better than the standalone model. The findings of this study proves that transfer learning
Formosa N, Quddus M, Papadoulis A, et al., 2022, Validating a traffic conflict prediction technique for motorways using a simulation approach, Sensors, Vol: 22, Pages: 566-566, ISSN: 1424-8220
With the ever-increasing advancements in the technology of driver assistant systems, there is a need for a comprehensive way to identify traffic conflicts to avoid collisions. Although significant research efforts have been devoted to traffic conflict techniques applied for junctions, there is dearth of research on these methods for motorways. This paper presents the validation of a traffic conflict prediction algorithm applied to a motorway scenario in a simulated environment. An automatic video analysis system was developed to identify lane change and rear-end conflicts as ground truth. Using these conflicts, the prediction ability of the traffic conflict technique was validated in an integrated simulation framework. This framework consisted of a sub-microscopic simulator, which provided an appropriate testbed to accurately simulate the components of an intelligent vehicle, and a microscopic traffic simulator able to generate the surrounding traffic. Results from this framework show that for a 10% false alarm rate, approximately 80% and 73% of rear-end and lane change conflicts were accurately predicted, respectively. Despite the fact that the algorithm was not trained using the virtual data, the sensitivity was high. This highlights the transferability of the algorithm to similar road networks, providing a benchmark for the identification of traffic conflict and a relevant step for developing safety management strategies for autonomous vehicles.
Ye M, Guan L, Quddus M, 2021, TDMP: Reliable Target Driven and Mobility Prediction based routing protocol in complex Vehicular Ad-hoc Network, Vehicular Communications, Vol: 31, ISSN: 2214-2096
Vehicle-to-everything (V2X) communication in vehicular ad hoc network (VANET) has emerged as a crucial component in advanced Intelligent Transport System (ITS) for information transmission and vehicular communication. One of the vital research challenges in VANET is the design and implementation of novel network routing protocols which bring reliable end-to-end connectivity and efficient packet transmission to V2X communication. The organically changing nature of road traffic vehicles poses a significant threat to VANET with respect to the accuracy and reliability of packets delivery. Therefore, position-based routing protocols tend to be the predominant method in VANET as they overcome rapid changes in vehicle movements effectively. However, existing routing protocols have some limitations such as (i) inaccurate in high dynamic network topology, (ii) defective link-state estimation (iii) poor movement prediction in heterogeneous road layouts. Therefore, a novel target-driven and mobility prediction (TDMP) based routing protocol is developed in this paper for high-speed mobility and dynamic topology of vehicles, fluctuant traffic flow and diverse road layouts in VANET. To implement an effective routing protocol, TDMP primarily involves the destination target of a driver for the mobility prediction and Received Signal Strength Indicator (RSSI) for the inter-vehicular link-status estimation. Compared to existing geographic routing protocols which mainly greedily forward the packet to the next-hop based on its current position and partial road layout, the proposed TDMP is able to enhance the packet transmission with the consideration of the estimation of inter-vehicular link status, and the prediction of vehicle positions dynamically in fluctuant mobility and global road layout. Based on the extensive simulations carried out on operational road environments with varying configurations and complexity, the experimental results show better performance in terms of improvin
Monsuur F, Enoch M, Quddus M, et al., 2021, Modelling the impact of rail delays on passenger satisfaction, Transportation Research Part A: Policy and Practice, Vol: 152, Pages: 19-35, ISSN: 0965-8564
Rail use and rail traffic in the UK increased substantially in the 25 years from 1994 to the end of 2019, a situation which led to progressively more delays and increasingly dissatisfied passengers. This study aims to quantify how disruptions to rail services are perceived by passengers to highlight situations that cause the highest rates of dissatisfaction so that they can be more effectively managed by the rail industry. Passenger satisfaction data from 7000 or so responses to the UK National Rail Passenger Survey (NRPS) where passengers had experienced delays were integrated with Network Rail data of the exact operational performance (e.g. train punctuality, service frequency, delay cause, magnitude of delay) that was encountered on each surveyed trip. An ordered logit model was then applied which allows for random taste variation to understand how passenger satisfaction was affected by rail delays. The study found that passengers reacted negatively to delays over 30 min, and dissatisfaction was exacerbated when passengers had to stand during the journey and/or received poor information, and when trains were cancelled. Policy implications for train operators include: (1) only cancel trains as a last resort; (2) prioritise trains approaching the ten minute delay threshold; (3) prioritise minimising delays to trains carrying high numbers of standing passengers; (4) enhance information quality and information delivery mechanisms as far as possible. Government should re-orientate franchise contracts to: (1) incentivise train operating companies to place more emphasis on passenger satisfaction when implementing service recovery strategies; and (2) improve delay information provision. Already the results are helping rail operators and practitioners to develop targeted recovery strategies aimed at minimising passenger dissatisfaction. This is the first academic study to investigate how rail passenger satisfaction is influenced by operational factors such as real-time de
Panagiota Deligianni S, Papadoulis A, Monsuur F, et al., 2021, Improving School Travel Plan effectiveness through enhanced diagnostic tools, Case Studies on Transport Policy, Vol: 9, Pages: 1273-1283, ISSN: 2213-624X
Issues around the journey to school are often in the news, and so are never far removed from the public and political consciousness. School Travel Plans (STPs) have been proven to reduce car use and their impacts around the world but are not aswidespread as they might be because of a perceived lack of knowledge about the potential outcomes of less common packages of measures and the associated lack of skills in delivering those solutions in specific contexts. Accordingly, this study develops and then applies a low-cost and risk-reducing approach which first determines existing travel behaviour within a school context, and then measures the responses of parents to a range of potential solutions (a proxy for their likely effectiveness) before they are adopted. It draws on results from an on-site traffic survey, an online survey of 746 parents and employees and an online stated preference survey of 101 parents from four independent schools in Loughborough, England. The study finds that parents are more influenced by the price of alternatives than the price of car (though this is still important), whilst measures that are relatively low-cost, uncontroversial and easy to implement (i.e. bus app, supervised walking and lockers) can also potentially be effective and should therefore be important components of any future STP package. It is also clear that the approach used could be transferable to other similar pre-project evaluation exercises, subject to minor tweaks to enable context-specific issues to be incorporated.
Wang X, Chen J, Quddus M, et al., 2021, Influence of familiarity with traffic regulations on delivery riders' e-bike crashes and helmet use: Two mediator ordered logit models., Accid Anal Prev, Vol: 159
Micro-mobility vehicles such as electric bicycles, or e-bikes, are becoming one of the essential transportation modes in metropolitan areas, and most deliveries in large cities are dependent on them. Due to the e-bike's popularity and vulnerability, e-bike crash occurrence has become a major traffic safety problem in many cities across the world; finding the most important human factors affecting e-bike safety has thus been an important recent issue in traffic safety analysis. Since delivery riders are a key group of e-bike users, and since helmet use plays a crucial role in reducing the severity of a crash, this study conducted a city-wide online survey to analyze the helmet usage of 6,941 delivery riders in Shanghai, China. To determine the in-depth mechanisms influencing helmet use and e-bike crash occurrence, including the direct and indirect effects of the relevant factors, two mediator ordered logistic regression models were employed. The mediator ordered logistic model was compared with the traditional logistic regression model, and was found to be superior for modeling indirect as well as direct influencing factors. Results indicate that riders' familiarity with traffic regulations (FTR) is an extremely important variable mediating between the independent variables of riders' educational level and age, and the dependent variables of helmet use and e-bike crashes. Improving riders' FTR can consequently increase helmet use and decrease crash occurrence. Authorities can apply these findings to develop appropriate countermeasures, particularly in legislation and rider training, to improve e-bike safety.
Papazikou E, Thomas P, Quddus M, 2021, Developing personalised braking and steering thresholds for driver support systems from SHRP2 NDS data., Accid Anal Prev, Vol: 160
Examining the relationships between the factors associated with the crash development enabled the realisation of driver support systems aiming to proactively avert and control crash causation at various points within the crash sequence. Developing such systems requires new insights in personalised pre-crash driver behaviour with respect to braking and steering to develop crash prevention strategies. Therefore, the current study utilises Strategic Highway Research Program 2 Naturalistic Driving Studies (SHRP2 NDS) data to investigate personalised steering and braking thresholds by examining the last stage of a crash sequence. More specifically, this paper carried out an in-depth examination of braking and steering manoeuvres observed in the final 30 s prior to safety critical events. Two algorithms were developed to extract braking and steering events by examining deceleration and yaw rate and another developed and applied to determine the sequence of the manoeuvres. Based on the analysis, thresholds for detecting emerging situations were recommended. The investigation of driver behaviour before the safety critical events, provides valuable insights into the transition from normal driving to safety critical scenarios. The results indicate that 20% of the drivers did not react to the impending event suggesting that they were not aware of the imminent safety critical situation. Future development of Advanced Driver Assistance Systems (ADAS) can focus on individual drivers' needs with tailored activation thresholds. The developed algorithms can facilitate driver behaviour and safety analysis for NDS while the thresholds recommended could be exploited for the design of new driver support systems.
Hussain RS, Quddus MA, Enoch MP, et al., 2021, Time series analysis of local authority policy interventions on highway works durations, Proceedings of the Institution of Civil Engineers: Transport, Vol: 174, Pages: 283-293, ISSN: 0965-092X
Highway works are highly inconvenient and disruptive for society. Accordingly, four highway policy interventions were investigated in Derby, UK, for potential corresponding reductions in highway works durations. Time series analysis was used to test the durational impacts on works led by Highway Authorities (HAs) and utility industries. The modelling results demonstrated that a highway works management permit scheme (chargeable) reduced utility works durations by 5·4% (727 work days annually). Conversely, three conflated interventions-namely, the permit scheme (cost-free to HAs), JCB Pothole Master deployment and construction direct labour organisation-did not make any statistically significant difference to HA works durations; however, introducing an automated works order management system (Woms) reduced HA works duration by 34% (6519 work days annually). The key finding of this study is that chargeable permit schemes can create the impetus for change, as demonstrated by the utility industry. Furthermore, the Woms revealed that back-office efficiency can lead to on-site efficiency in works execution.
Zhao W, Quddus M, Huang H, et al., 2021, The extended theory of planned behavior considering heterogeneity under a connected vehicle environment: A case of uncontrolled non-signalized intersections., Accid Anal Prev, Vol: 151
With the emergence of connected vehicle (CV) technology, there is a doubt whether CVs can improve driver intentions and behaviors, and thus protect them from accidents with the provision of real-time information. In order to understand the possible impacts of the real-time information provided by CV technology on drivers, this paper aims to develop a model which considers the heterogeneity between drivers with the aid of the extended theory of planned behavior. At the uncontrolled non-signalized intersections, a stated preference (SP) questionnaire survey was conducted to build the dataset consisting of 1001 drivers. Based on the collected dataset, the proposed model examines the relationships between subjective norms, attitudes, risk perceptions, perceived behavioral control and driving intentions, and studies how such driving intentions are simultaneously related to driver characteristics and experiences in the CV environment. Furthermore, driver groups which are homogenous with respect to personality traits are formed, and then are employed to analyze the heterogeneity in responses to driving intentions. Four key findings are obtained when analyzing driver responses to the real-time information provided by CV technology: 1) the proposed H-ETPB model is verified with a good fitness measure; 2) irrespective to driver personality traits, attitudes and perceived behavioral control have a direct and indirect association with driving intentions to accelerate; 3) driving intentions of high-neurotic drivers to accelerate are significantly related to subjective norms, while that of low-neurotic drivers are not; 4) elder high-neurotic drivers, and low-neurotic drivers who have unstable salaries or ever joined in online car hailing service have a strong intention in accelerating. The findings of this study could provide the theoretical framework to optimize traffic performance and information design, as well as provide in-vehicle personalized information service in the CV a
Yu R, Wang Y, Quddus M, et al., 2021, Investigating vehicle roadway usage patterns on the Shanghai urban expressway system and their impacts on traffic safety, International Journal of Sustainable Transportation, Vol: 15, Pages: 217-228, ISSN: 1556-8318
The urban expressway system serves as a key role in the roadway transportation system. It provides an efficient and comfortable approach for long-distance travel within the city. However, the safety status of the urban expressways is becoming a critical issue as the high-frequent traffic crashes have severely influenced the traffic operations. Among the safety influencing factors, including traffic operational parameters (such as traffic speed and volume), geometric features and traffic participants’ characteristics (such as vehicle roadway usage patterns), the traffic operational parameters and geometric features have been widely investigated. However, the impacts of traffic participants’ characteristics on traffic safety have never been examined. This unprecedented study aims to link vehicles’ roadway usage patterns with traffic safety through crash frequency analyses. First, the roadway usage patterns were identified using Latent Class Cluster Analysis (LCCA) based on their traveling rates. Then, the hourly-based crash frequency analysis data were formulated with traffic operational parameters, geometric features and crash data. Finally, crash frequency analysis models were developed to unveil the relationships between the crash occurrence and their influencing factors. The modeling results showed that the Random Effects Hurdle Negative Binomial Model (REHNBM) provided better goodness-of-fit. And it concluded that higher proportions of vehicles with low-level roadway usage pattern would substantially enhance the possibility of crash occurrence; while the proportions of vehicles with the medium-high-level roadway usage pattern had negative impacts on crash occurrence probability. Finally, safety improvement recommendations and strategies based on the modeling results were put forward.
Sharath MN, Velaga NR, Quddus MA, 2020, 2-dimensional human-like driver model for autonomous vehicles in mixed traffic, IET Intelligent Transport Systems, Vol: 14, Pages: 1913-1922, ISSN: 1751-956X
Classical artificial potential approach of motion planning is extended for emulating human driving behaviour in two dimensions. Different stimulus parameters including type of ego-vehicle, type of obstacles, relative velocity, relative acceleration, and lane offset are used. All the surrounding vehicles are considered to influence drivers' decisions. No emphasis is laid on vehicle control; instead, an ego vehicle is assumed to reach the desired state. The study is on human-like driving behaviour modelling. The developed motion planning algorithm formulates repulsive and attractive potentials in a data-driven way in contrast to the classical arbitrary formulation. Interaction between the stimulus parameters is explicitly considered by using multivariate cumulative distribution functions. Comparison of two-dimensional (lateral and longitudinal) performance indicators with a baseline model and generative adversarial networks indicate the effectiveness and suitability of the developed motion planning algorithm in the mixed traffic environment.
Feng M, Wang X, Quddus M, 2020, Developing multivariate time series models to examine the interrelations between police enforcement, traffic violations, and traffic crashes, Analytic Methods in Accident Research, Vol: 28, ISSN: 2213-6657
Safer roads and police enforcement are closely associated since the latter directly encourages road users to improve their behavior by complying with basic traffic rules and laws. Understanding the relationships between police enforcement, driving behavior, and traffic safety is a prerequisite for optimizing enforcement strategies. However, there is a dearth of research on the contemporaneous relationships between these three parameters. Using multivariate time series techniques, this study provides an in-depth analysis of contemporaneous relationships and dynamic interactions among police enforcement, traffic violations, and traffic crashes. The amount of police patrol time per day was used as a surrogate measure for police enforcement intensity. A vector autoregressive (VAR) model was first used to examine the influences of exogenous factors including weather conditions and holidays. Based on the findings of the VAR model, a structural vector autoregressive (SVAR) model was developed to determine contemporaneous effects; the Granger causality test was employed to detect any dynamic interactions between the three parameters. The results indicated that traffic crashes and violations had weekly variation and were significantly impacted by holiday and weather conditions, while police patrol time was not impacted. A contemporaneous negative impact of police patrol time was found in traffic crashes: each 1% increase in police patrol time was associated with a 0.15% decrease in contemporaneous crash frequency. The findings from the Granger causality test demonstrated that police patrol time did not Granger-cause traffic crashes, but crashes Granger-caused police patrol time. The significant bidirectional interactions in conditional variances of police enforcement, traffic violations, and traffic crashes further confirm the necessity to analyze the three simultaneously. The findings of this study are expected to assist the relevant traffic authorities in devising policies
Yu R, Long X, Quddus M, et al., 2020, A Bayesian Tobit quantile regression approach for naturalistic longitudinal driving capability assessment., Accid Anal Prev, Vol: 147
Given the severe traffic safety issue, tremendous efforts have been devoted to identify the crash contributing factors for developing and implementing safety improvement countermeasures. According to the study findings, driving behaviors have attributed to the majority crash occurrence, among which inadequate driving capability is a key factor. Therefore, a number of studies have been conducted for developing techniques associated with the driving capability assessment and its various improvement. However, the conventional assessment approaches, such as driving license exams and vehicle insurance quotes, have only focused on basic driving skill evaluations or aggregated driving style classifications, which failed to quantify driving capability from the safety perspective with respect to the complex driving scenarios. In this study, a novel longitudinal driving capacity assessment and ranking approach was developed with naturalistic driving data. Two Responsibility-Sensitive Safety (RSS) based driving capability indicators from the perspectives of risk exposure and severity were first proposed. Then, Bayesian Tobit quantile regression (BTQR) models were introduced to explore the relationships between driving capability indicators with trip level characteristics from the aspects of travel features, operational conditions, and roadway characteristics. The modeling results concluded that nighttime driving and higher average speed would lead to higher longitudinal collision risk and its severity. Besides, the BTQR models have provided varying factors significances among different quantile levels, for instance, driving duration is only significant at high quantiles for the driving capability indicators, implying that duration only affects drivers with large longitudinal risk exposures and strong close following tendencies. Furthermore, the case studies provided how to deploy the developed model to obtain the relative longitudinal driving capability rankings. Finally, the
Yi D, Su J, Hu L, et al., 2020, Implicit Personalization in Driving Assistance: State-of-the-Art and Open Issues, IEEE Transactions on Intelligent Vehicles, Vol: 5, Pages: 397-413
In recent decades, driving assistance systems have been evolving towards personalization for adapting to different drivers. With the consideration of driving preferences and driver characteristics, these systems become more acceptable and trustworthy. This article presents a survey on recent advances in implicit personalized driving assistance. We classify the collection of work into three main categories: 1) personalized Safe Driving Systems (SDS), 2) personalized Driver Monitoring Systems (DMS), and 3) personalized In-vehicle Information Systems (IVIS). For each category, we provide a comprehensive review of current applications and related techniques along with the discussion of industry status, benefits of personalization, application prospects, and future focal points. Both relevant driving datasets and open issues about personalized driving assistance are discussed to facilitate future research. By creating an organized categorization of the field, we hope that this survey could not only support future research and the development of new technologies for personalized driving assistance but also facilitate the application of these techniques within the driving automation community.
Formosa N, Quddus M, Ison S, et al., 2020, Predicting real-time traffic conflicts using deep learning., Accid Anal Prev, Vol: 136
Recently, technologies for predicting traffic conflicts in real-time have been gaining momentum due to their proactive nature of application and the growing implementation of ADAS technology in intelligent vehicles. In ADAS, machine learning classifiers are utilised to predict potential traffic conflicts by analysing data from in-vehicle sensors. In most cases, a condition is classified as a traffic conflict when a safety surrogate (e.g. time-to-collision, TTC) crosses a pre-defined threshold. This approach, however, largely ignores other factors that influence traffic conflicts such as speed variance, traffic density, speed and weather conditions. Considering all these factors in detecting traffic conflicts is rather complex as it requires an integration and mining of heterodox data, the unavailability of traffic conflicts and conflict prediction models capable of extracting meaningful and accurate information in a timely manner. In addition, the model has to effectively handle large imbalanced data. To overcome these limitations, this paper presents a centralised digital architecture and employs a Deep Learning methodology to predict traffic conflicts. Highly disaggregated traffic data and in-vehicle sensors data from an instrumented vehicle are collected from a section of the UK M1 motorway to build the model. Traffic conflicts are identified by a Regional-Convolution Neural Network (R-CNN) model which detects lane markings and tracks vehicles from images captured by a single front-facing camera. This data is then integrated with traffic variables and calculated safety surrogate measures (SSMs) via a centralised digital architecture to develop a series of Deep Neural Network (DNN) models to predict these traffic conflicts. The results indicate that TTC, as expected, varies by speed, weather and traffic density and the best DNN model provides an accuracy of 94% making it reliable to employ in ADAS technology as proactive safety management strategies. Furthermore
Naqvi NK, Quddus MA, Enoch MP, 2020, Do higher fuel prices help reduce road traffic accidents?, Accid Anal Prev, Vol: 135
Road traffic accidents have decreased in most developed nations over the last decade. This has been attributed to improvements in vehicle and road design, medical technology and care, and driver education and training. Recent evidence however indicates that fuel price changes also have a significant impact on road traffic accidents through other mediating factors such as reductions in driver exposure through less car travel and more fuel-efficient driving e.g. speed reduction on high-speed roads. So far though, no study has examined the effects of changing fuel prices on road traffic accidents in a country such as Great Britain where fuel prices are kept artificially high for public policy reasons. Consequently, this study was designed to quantify the effects of fuel price on road traffic accident frequency through changes and adjustments in travel behaviour. For this purpose, weekly fuel prices (between 2005-2015) have been used to study the effects on road traffic accidents using the Prais-Winsten model of first order autoregressive (AR1) and the Box and Jenkins seasonal autoregressive integrated moving average models (SARIMA). The study found that with every 1% increase in fuel price there is a 0.4% reduction in the number of fatal road traffic accidents. In light of this, one concern raised was that recent UK government plans to phase out petrol and diesel vehicles by 2040 may also risk a rise in fatal road traffic accidents, and hence this will need to be addressed.
Papazikou E, Quddus M, Thomas P, et al., 2019, What came before the crash? An investigation through SHRP2 NDS data, Safety Science, Vol: 119, Pages: 150-161, ISSN: 0925-7535
Investigating crash progression through naturalistic driving studies (NDS) could give valuable insights in crash causation analysis and thus, benefit crash prevention. This study utilises NDS data from the Strategic Highway Research Program 2 (SHRP2 NDS data) to look into the whole crash sequence, from a normal driving situation until a crash or a near-crash event. The objectives are to explore vehicle kinematics before the event, investigate the feasibility of crash risk indicators to detect the early stages of crash development and further examine the factors affecting Time To Collision (TTC) values during the crash sequence. An empirical approach and a multilevel mixed effects modelling technique were followed. The results reveal that longitudinal acceleration, lateral acceleration and yaw rate can be reliable indicators for detecting deviations from normal driving. Moreover, TTC values are affected by vehicle type, speed of the ego vehicle, longitudinal acceleration and time within the crash sequence. The model indicates a timestamp where a detectable reduction in TTC values occurs, which could be a first step towards more effective Advanced Driver Assistance Systems (ADAS) aiming to halt early deviations before they evolve to mishaps.
This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.