Average Travel Speed Research Articles

자율차-일반차 혼재상황에서 돌발상황 대응 교통운영전략 USE CASE 시나리오 효과분석

This study aims to analyze the effects of providing traffic operation strategies to autonomous vehicles in response to incidents in mixed traffic situations involving autonomous vehicles and manual vehicles. The use case scenario of providing detour information for incidents in a living lab where autonomous vehicles are expected to operate is intended to derive the effects of traffic operation strategies through comparison of average travel speeds. In order to analyze the effectiveness of traffic operation strategy in traffic situations where autonomous vehicles and manual vehicles coexist, three scenarios were designed for peak and non-peak traffic conditions, and the average travel time of autonomous vehicles for each autonomous driving service was calculated and analyzed as an indicator for judging the effectiveness of the information provision scenario for detours. The simulation analysis results for scenario 2, where an incident occurs, showed that the average travel speed increased compared to scenario 1, and this was analyzed to reflect the bottleneck caused by the incident (reducing operation from 3 lanes to 1 lane). In scenario 3, which simulates the case where detour information is provided as part of the incident response traffic operation strategy, the average travel time of autonomous vehicles decreases compared to scenario 2 for the occurrence of an incident. Therefore, it is judged that the traffic conditions improves as autonomous vehicles select the detour as a strategy for providing detour information in response to an incident, thereby reducing the demand for vehicles passing through the incident location. This study is significant in that it analyzed the effectiveness of traffic operation strategy by implementing an actual traffic environment as a simulation program for the use case of traffic operation strategy that can be provided to autonomous vehicles when an incident occurs in a mixed situation of autonomous vehicles and manual vehicles, and as an initial study in preparation for the era of autonomous driving, it is a study on mixed situations of autonomous vehicles in urban areas.

Evolutionary mechanism of Y-branches in acoustic Lichtenberg figures just below the water surface

The acoustic Lichtenberg figure (ALF) in an ultrasonic cleaner with a frequency of 28 kHz at different power levels was observed using high-speed photography. The nonlinear response of the cavitation structure was analyzed by the entropy spectrum in the ALF images, which showed the modulation influence of the primary acoustic field, exhibiting the fluctuations of the bubble distribution with time. Typical Y-branches predict the paths by which surrounding bubbles are attracted and converge into the structure, the branches are curved due to bubble-bubble interactions, and the curvature increases as the bubbles are approaching the main chain. The average travelling speed of bubbles along the branches is about 1.1 m/s, almost independent of power level of the ultrasonic cleaner. A theoretical model consisting of free bubbles and a straight bubble chain of finite length was developed to explore the evolutionary mechanism of branching. It was found that the bubble trajectories showed a bending tendency similar to the experimentally observed Y-branches, and the stationary straight bubble chain parallel to the main chain could evolve into a curved chain and eventually become a branch of the main chain. The theoretical predictions agree well with the experimental results, verifying the evolutionary mechanism of Y-branches in ALF.

Asynchronous decentralized traffic signal coordinated control in urban road network

AbstractThis study introduces an asynchronous decentralized coordinated signal control (ADCSC) framework for multi‐agent traffic signal control in the urban road network. The controller at each intersection in the network optimizes its signal control decisions based on a prediction of the future traffic demand as an independent agent. The asynchronous framework decouples the entangled interdependence between decision‐making and state prediction among different agents in decentralized coordinated decision‐making problems, enabling agents to proceed with collaborative decision‐making without waiting for other agents’ decisions. Within the proposed ADCSC framework, each controller dynamically optimizes its signal timing strategy with a unique rolling horizon scheme. The scheme's individualized parameters for each controller are determined based on the vehicle travel time between the adjacent intersections, ensuring that controllers can make informed control decisions with accurate arrival flow information from upstream intersections. The signal optimization problem is formulated as a mixed integer linear program model, which adopts a flexible signal scheme without a fixed phase structure and sequence. Simulation results demonstrate that the proposed ADCSC strategy significantly outperforms the benchmark signal coordination methods in terms of average delay, travel speed, stop numbers, and energy consumption. Experimental analysis on computation time validates the applicability of the proposed optimization model for real‐time implementation. Sensitivity analysis on key parameters in the framework is conducted, offering insights for parameter selection in practice. Furthermore, the ADCSC framework is extended to a road network in Qinzhou City, China, with 45 signalized intersections, demonstrating its effectiveness and scalability in the real‐world road network.

Dead-reckoning facilitates determination of activity and habitat use: a case study with European badgers (Meles meles)

BackgroundStudies describing the movement of free-ranging animals often use remotely collected global positioning system (GPS) data. However, such data typically only include intermittent positional information, with a sampling frequency that is constrained by battery life, producing sub-sampling effects that have the potential to bias interpretation. GPS-enhanced ‘dead-reckoning’ of animal movements is an alternative approach that utilises combined information from GPS devices, tri-axial accelerometers, and tri-axial magnetometers. Continuous detailed information of animal movement, activity and habitat selection can then be inferred from finer-scale GPS-enhanced dead-reckoning. It is also a useful technique to reveal the minutiae of an animal’s movements such as path tortuosity. However, examples of studies using these approaches on terrestrial species are limited.MethodsCollars equipped with GPS, tri-axial accelerometer, and tri-axial magnetometer loggers were deployed on European badgers, Meles meles, to collect data on geo-position, acceleration and magnetic compass heading, respectively. This enabled us to compare GPS data with calculated GPS-enhanced dead-reckoned data. We also examined space use, distances travelled, speed of travel, and path tortuosity in relation to habitat type.ResultsNightly distances travelled were 2.2 times greater when calculated using GPS-enhanced dead-reckoned data than when calculated using GPS data alone. The use of dead-reckoned data reduced Kernel Density Estimates (KDE) of animal ranges to approximately half the size (0.21 km2) estimated using GPS data (0.46 km2). In contrast, Minimum Convex Polygon (MCP) methods showed that use of dead reckoned data yielded larger estimates of animal ranges than use of GPS-only data (0.35 and 0.27 km2, respectively).Analyses indicated that longer periods of activity were associated with greater travel distances and increased activity-related energy expenditure. Badgers also moved greater distances when they travelled at faster speeds and when the routes that they took were less tortuous. Nightly activity-related energy expenditure was not related to average travel speed or average ambient temperature but was positively related to the length of time individuals spent outside the sett (burrow). Badger activity varied with habitat type, with greater distance, speed, track tortuosity, and activity undertaken within woodland areas. Analyses of the effects of varying GPS sampling rate indicate that assessments of distance travelled depend on the sampling interval and the tortuosity of the animal’s track. Where animal paths change direction rapidly, it becomes more important to use dead-reckoned data rather than GPS data alone to determine space use and distances.ConclusionsThis study demonstrates the efficacy of GPS-enhanced dead-reckoning to collect high-resolution data on animal movements, activity, and locations and thereby identify subtle differences amongst individuals. This work also shows how the temporal resolution of position fixes plays a key role in the estimation of various movement metrics, such as travel speed and track tortuosity.

Application of Historical Comprehensive Multimodal Transportation Data for Testing the Commuting Time Paradox: Evidence from the Portland, OR Region

There have been numerous theoretical and empirical transportation studies contesting the stability of commuting time over time. The constant commuting time hypothesis posits that people adjust trip durations, shift across modes, and sort through locations, so that their average commuting time remains within a constant budget. There is a discrepancy between studies applying aggregate analysis and those using disaggregate analysis, and differences in data collection may have contributed to the varying conclusions reported in the literature. This study conducts both aggregate and disaggregate analyses with two travel surveys of the Portland region. We employ descriptive analysis and t-tests to compare the aggregate commuting times of two years and use regression models to explore factors affecting the disaggregate commuting time at the individual trip level to examine whether the stability of the commuting time remains after substantial changes in the transportation and land use systems. Our study indicates that the average commuting time, along with the average commuting distance, increased slightly, as the mode share shifted away from driving during the examined period. The growth in shares of non-driving modes, which are slower than driving, coupled with an increased travel distance, contributed to the small increase in the average commuting time. Our analysis also indicates that the average travel speed improved for transit riders as well as drivers, contradicting earlier research that claims that public transit investment has worsened the congestion in Portland.

ANALISIS BIAYA OPERASIONAL KENDARAAN BERDASARKAN KECEPATAN RATA-RATA PERJALANAN DI JALAN MAJAPAHIT KOTA MATARAM

The city of Mataram, the center of trade, offices, and education, experienced an increase in population and vehicles. The 2023 data recorded 441.147 thousand residents and 385,579 motor vehicles. The increase in traffic volume on road Majapahit has an impact on traffic performance and vehicle operating costs. The study entitled " Analysis Of Vehicle Operational Costs Based On Average Travel Speed On Jalan Majapahit Mataram City" was conducted to further understand the condition of road section performance and the Operating Cost of Vehicles passing through the road. The MKJI method is a calculation standard used to analyze the performance of road sections, assess capacity and service levels, based on traffic volume and road geometry. while to calculate BOK (Vehicle Operating Costs) used an equation method developed by the PCI 1989 method to measure vehicle costs, including fixed costs and non-fixed costs, which are influenced by the average speed of travel. The results of the study on the Majapahit road section show the service level of the Majapahit road section in category C. For the calculation of the free flow speed, namely 52.68 km / h (east) and 49.10 km / h (west). The speed of the eastbound peak clock is 48 km/h and the westward is 42 km/h. While the calculation results of BOK Group I (Eastward) Free Flow Speed Rp1.049,- / km, Peak Hour Speed Rp1.082,- /km. Group I (West) Free Flow Speed Rp1.072,- /km, Peak Hour Speed Rp1.163,- /km. Group IIA (Eastward) Peak Hour Speed Rp2.921,- /km, Peak Hour Speed Rp3.007,- /km. Group IIA (Westward) Free Flow Speed Rp2.980,- /km, Peak hour speed Rp3.235,- /km

Evaluation of road roughness and its influence on operating parameters in open-pit mines

Hauling ore and waste rock is a major cost factor in open-pit mines, encompassing both road maintenance and off-road truck operation. Road maintenance costs are directly influenced by traffic volume and the size of trucks using the roads. Off-road truck operating costs are impacted by various factors, including those affecting their movement, such as the direct interaction between tires and the road surface. Tire-road interaction significantly affects rolling resistance. Rougher surfaces with deeper grooves caused by track sinks lead to higher rolling resistance. This study employed laser profiling to characterize the surface roughness of five mine haul roads in two large Brazilian iron ore mines. This study evaluated the impact of these road irregularities on various operating parameters for 240-ton off-highway trucks, including rolling resistance, average travel speed, travel time, productivity, and unit transport cost. The results indicate that a 10 cm increase in road surface roughness can lead to a significant increase in rolling resistance (up to 5%), a substantial decrease in average speed (25%), a notable increase in travel time (26%), a decrease in productivity (19%), and a corresponding increase in unit transport cost (21%).

EFFECTIVENESS ENHANCEMENT OF TRAFFIC CAMERAS BASED ON VEHICLES' AVERAGE SPEED CALCULATION

The analysis of calculating the vehicle speed between two automatic speed cameras within the framework of traffic violation detection demonstrates that using the average speed metric enhances speed limit compliance accuracy. This method reduces the possibility of evading responsibility for violations through temporary speed reduction before cameras. These systems employ usage of existing road cameras infrastructure to determine the minimal travel path of vehicle passing two cameras placed at a certain distance apart to measure the average speed of travel. Technical aspects of implementing the proposed system include utilizing cloud computing for data processing and storage, which reduces infrastructure costs. The system employs the possibility of usage of machine learning algorithms to predict travel time between cameras, considering various factors such as road conditions and speed limits. This approach ensures high calculation accuracy and reduces the number of false violation alerts. To further reduce the possibility of false violations, it is suggested to only use speed limitations based on the traffic signs on the road intervals of the calculated minimal path between two cameras that the vehicle has passed. Implementing the system requires the development of new software for data analysis and integration with existing traffic control systems. This includes designing algorithms to determine the minimum required travel time between cameras and calculating the average speed based on these data. Additionally, methods must be developed to detect and handle exceptions, such as road stops, or route changes, which can affect calculation accuracy. The feasibility of practical implementation of the system was investigated based on the existing infrastructure of traffic cameras in the city of Lviv. Possible minimal routes between cameras were analyzed and compared to the avarage passing speed of vehicles of different categories. The modulated results indicate that such a system would be an effective addition to the traffic control system, while requiring relatively low financial and engineering efforts.

Analyzing Riders’ Behavioral Adaptation to Driving Patterns of Advanced Autonomous Vehicles: A Virtual Reality Simulation Study

The necessity of human supervision and intervention during autonomous driving has long been a topic of controversial discussion. From a developer’s perspective, it is expected that users will readily adapt to well-calibrated autonomous driving systems (ADS) due to their superior performance in dynamic driving tasks (DDT) compared to conventional human-driven vehicles. However, when passengers experience an autonomous vehicle (AV), there may be an adjustment period during which they modify their behavior to accommodate the driving patterns of the ADS. Additionally, some passengers might not adapt to autonomous driving at all, highlighting potential limitations in the current ADS development strategy. This work studies the dynamics of human-automation interaction and introduces an “objective method”, which employs a Virtual Reality (VR)-enabled simulation approach for in-depth behavioral analysis concerning riders’ behavioral adaptation to autonomous driving. Specifically, we examined how participants interacted with and intervened in Level 4 ADS operating under conservative, moderate, and aggressive driving patterns in a fully autonomous environment. A realistic urban road network was recreated in VR, integrated with traffic microsimulation to generate various driving scenarios. Twenty-seven participants completed driving tasks across different AV modes, with their intervention behaviors analyzed in relation to traffic conditions and AV aggressiveness. Key findings include: (1) Participants showed higher intention to intervene but lower actual intervention rates under aggressive AV modes compared to moderate and conservative modes, suggesting quicker adaptation to more challenging driving scenarios. (2) Interventions generally proved unnecessary and sometimes detrimental to overall traffic performance in a full-AV environment. (3) Aggressive AV modes significantly improved traffic efficiency, with a 40% increase in average travel speed and a 53% reduction in waiting time. However, human interventions posed the greatest challenge to achieving optimal traffic conditions. This research provides insights into the complex dynamics of human-AV interaction and adaptation, offering valuable implications for AV interface design, implementation strategies, and public acceptance of autonomous driving technologies.

Analysis of the use of logistics-telematics systems for cost reduction in a transport company

Purpose: The main purpose of this paper is to investigate the extent to which real-time monitoring of vehicles and the collection of operational data about them through telematic systems can contribute to the reduction of operating costs in a transport company. The study focuses on specific factors such as the driver's actions, which can affect the vehicle's technical condition and diesel consumption levels. Methodology: The research is based on an analysis of operational data obtained from the vehicles of a transport company engaged in international freight transportation. The data were primarily analyzed for an indicator related to total fuel (diesel) consumption while driving. Linear models were created using the least squares method, and the best-fitting models were selected to draw conclusions about the impact of telematic systems on the company's costs. Results: The study found that the introduction of a telematic system can have a positive impact on a transport company's operations and help reduce costs associated with the execution of transport orders. The analysis of the data showed that individual factors, such as driving time, average axle load, and average travel speed, can affect the total driving fuel consumption of all vehicles and contribute to the overall cost reduction. Theoretical contribution: This paper contributes to the understanding of the role of telematic systems in the optimization of fleet planning and management, which is a crucial aspect of the logistics of transporting goods. The research provides insights into the potential of real-time monitoring and data collection for improving the efficiency and cost-effectiveness of transport companies. Practical implications: The findings of this study can be useful for transport companies looking to improve their operations and reduce costs. The research highlights the importance of using telematic systems in the right way to carry out the transportation of goods and suggests that real-time monitoring and data analysis can be key to minimizing costs and maximizing profits.

The Feasibility and Operational Performance of Implementation Median U-turn Intersections: A CRITIC Method

Currently, a growing number of cities are adopting the Median-U-Turn (MUT) intersection design to enhance road capacity and traffic efficiency. The critical question in selecting the right intersection design is how significantly the implementation of MUT can enhance intersection performance, focusing on three key aspects: intersection efficiency, capacity, and the environmental impact of the design. To address this question, an evaluation of operational performance under various prevailing conditions (roadway and control) was conducted using VISSIM, a microscopic simulation platform. This evaluation involved five scenarios: conventional intersections (with increased cycle length, grade separation with a signalized at-grade intersection, grade separation with a roundabout), MUT, and signalized crossover MUT at a dense urban arterial intersection in Amman, Jordan's capital. The performance was compared using several metrics: average control delay, number of stops, average travel speed and time, average stopped delay, Carbon Monoxide (CO) emissions, fuel consumption, and vehicle safety. The Criteria Importance Through Intercriteria Correlation (CRITIC) technique was subsequently used to select the optimal design. The findings indicate that the existing intersection configuration is the least effective, while the MUT with signals at the crossing U-turn points is the most efficient solution.

Assessing school evacuation movement characteristics: Children and adolescents speed and flow over stairs and through exit doorways

AbstractThe available theories of evacuation movements are primarily founded on data gathered from adults, making them potentially unsuitable for children, especially in schools. Consequently, it is necessary to undertake further research to collect data on how children move during evacuations to understand their unique characteristics and disparities compared to adults. In this context, this paper aimed to explore the movement of school children and adolescents as they moved over stairs and through exit doorways during evacuations. The evacuation drill involved 295 school children and adolescents, whose behavior was closely monitored using a series of cameras. During the drill, their movement patterns, including flow and speed, were analyzed over stairs and through doorways. The observations revealed that children exhibited frequent interactions and contact with one another, unlike adults, who tend to maintain personal space. The findings of this study indicated that the average traveling speed over stairs was comparable to previous research, although female adolescents had a lower average speed compared to other groups. The speed and flow of participants passing through doorways were found to vary depending on their age and differed from estimates based on adult data. This study highlights that existing evacuation models fall short of adequately accounting for the dynamics of children, indicating the need for further research to improve the generalizability of evacuation models.

A unified dataset for the city-scale traffic assignment model in 20 U.S. cities

City-scale traffic data, such as traffic flow, speed, and density on every road segment, are the foundation of modern urban research. However, accessing such data on a city scale is challenging due to the limited number of sensors and privacy concerns. Consequently, most of the existing traffic datasets are typically limited to small, specific urban areas with incomplete data types, hindering the research in urban studies, such as transportation, environment, and energy fields. It still lacks a city-scale traffic dataset with comprehensive data types and satisfactory quality that can be publicly available across cities. To address this issue, we propose a unified approach for producing city-scale traffic data using the classic traffic assignment model in transportation studies. Specifically, the inputs of our approach are sourced from open public databases, including road networks, traffic demand, and travel time. Then the approach outputs comprehensive and validated citywide traffic data on the entire road network. In this study, we apply the proposed approach to 20 cities in the United States, achieving an average correlation coefficient of 0.79 in average travel time and an average relative error of 5.16% and 10.47% in average travel speed when compared with the real-world data.

Research on Vehicle Congestion Group Identification for Evaluation of Traffic Flow Parameters

The traffic flow parameters of the road network are most often evaluated through volumes, which are compared with its maximum volume (capacity) or speed and density. Capacity assessment was performed, considering horizontal and vertical orientation and characteristics of the traffic stream. This article presents the results of research on the identification of different states of creating congestion groups and their relationship to road capacity or decrease in speed. The following hypothesis was verified: when the capacity of the road is exceeded or almost reached, there is “always” a significant drop in the flow of traffic compared to when the capacity is not exceeded. The analysis showed that the average travel speed drops by 30% for the condition where groups of 25 or more vehicles are formed with a time interval of up to 4 s. The results make it possible to set traffic models in short time intervals according to real spatial conditions and to use them in the analysis of the environmental and safety impacts of road transport.

Analysis of the Impacts of Students Back to School on the Volatility and Reliability of Travel Speed on Urban Road

How to effectively and accurately evaluate and analyze the volatility and reliability of travel speed on urban road before and after students back to school is a hot and key problem in urban road traffic congestion governance research. The Beijing 3rd Ring Road was taken as the research object and the impacts of the students back to school on the volatility and reliability of the travel speed of road sections were qualitatively and quantitatively analyzed based on the road section travel speed data during the weekday morning peak (7:00–8:59). The results showed that the travel speed of the Beijing 3rd Ring Road had cyclicity, time variability, large-scale volatility, and light congestion during the weekday morning peak, and the volatility and reliability indexes of the travel speed of road sections significantly decreased under the impact of the students back to school. The data showed that after the students back to school, the maximum reduction ratio of average travel speed was larger than 55%, and the maximum travel speed reliability reduction value was larger than 0.85 based on the evaluation model of travel speed reliability of car commuters. The research results provide data and theoretical support for urban road traffic congestion mitigation and governance.

Collision-Free Merging Control Via Trajectory Optimization for Connected and Autonomous Vehicles

On-ramp merging zones have long been accident-prone spots as well as major triggers of traffic congestion in freeway networks. This paper proposes a collision-free merging control for connected and autonomous vehicles based on trajectory optimization to facilitate traffic safety and efficiency. The optimization is embedded in a model predictive control algorithm with the aim of maximizing the average travel speeds of all vehicles in the planning horizon. To guarantee collision-free merging, the merging control allows only one vehicle to pass the merging zone at a time. Safety constraints, including minimum time gaps and minimum distances between vehicles, are also considered. A rolling horizon control framework is established to enable dynamic traffic control; specifically, generating optimal acceleration decisions for each time step. The optimization problem is formulated as mixed-integer linear programming to enable efficient computation. Numerical studies and sensitivity analysis were conducted to validate the effectiveness and applicability of the proposed method. The results show that the introduced safety constraints can guarantee collision-free merging. Compared to the traditional method, Ih is built on the intelligent driver model and the minimizing overall braking induced by lane-change model, the proposed optimization-based approach results in 5.0%–22.2% improvements with respect to average travel speeds during the merging process. The safety and efficiency improvements are more significant when dealing with higher demand levels. Notably, compared to the difference in initial speeds between the mainline and on-ramp vehicles, the length of the acceleration lane has a greater impact on the control performance.

Assessing the Accuracy of Average Travel Speeds from Floating Car Data on Urban Street Facilities

To improve traffic flow on regularly congested roads in a network, the performance of these roads has to be evaluated. The average travel speed is a crucial indicator of traffic quality and is used in several performance evaluation methods. Floating Car Data (FCD) obtained from vehicles equipped with GPS offers continuous and ubiquitous data acquisition, making it a potential alternative to conventional speed detection methods. However, the limited penetration and representativeness of detected vehicles pose challenges. This work investigates the suitability of commercially available FCD for determining accurate space-mean speeds. Ground truth data was collected using Automatic Number Plate Recognition (ANPR), and a comparison analysis was performed with FCD-derived travel times and space-mean speeds. Different statistical measures were applied to evaluate the accuracy of FCD data. The results indicate a generally acceptable correlation between the two data sources, but deviations occur, especially for time intervals with a low number of detected FCD trajectories. Subsequent analyses suggest that a minimum number of five trajectories per hour are sufficient to derive valid space-mean speeds. Based on these findings, it is possible to accurately assess the performance of roads exclusively using commercially available FCD.

ЭКОНОМИЧЕСКОЕ ОБОСНОВАНИЕ ВЫДЕЛЕННЫХ ПОЛОС ДЛЯ ОБЩЕСТВЕННОГО ТРАНСПОРТА НА ПРИМЕРЕ ГОРОДА ЧЕЛЯБИНСКА

This paper investigates the effectiveness of public transport lanes. The study compares the current cost of the transport work and the cost of transport work when increasing the average speed of travel with dedicated public transport lanes. The recommendations of the Russian Ministry of Transport were used as the basis for the feasibility study. Based on changes in speed and travel times, the effect on the budget is assessed in monetary terms. Depending on the increase in speed, the budget effect is associated with a reduction in the required number of vehicles to deliver the same levels of transport services for the population. The methodology allows us to evaluate not only the change in traffic flow parameters when implementing dedicated lanes for public transport, but also the economic effect of increasing the speed of travel. The work analyzes the economic effect of organizing a dedicated lane using specific municipal routes in the city of Chelyabinsk.

Road performance analysis due to Sayung Industrial Park development

This study aims to analyze the impact of Sayung Industrial Park development to the surrounding road network. The study merely relies on available secondary data, including recent transport study in Semarang Raya, trip rate, and previous research related on industrial area occupancy growth. Road performance in the terms of degree of saturation and travel speed are analyzed using a simplified four-step model using CUBE software. The results show that in 2023, assuming 50% occupancy of the industrial park, the average travel speed decreases by 8.5% and the average degree of saturation increases by 18.7%. While in 2030, assuming 100% occupancy, the road performance is worsened, the average travel speed decreases by 46.4% and the average degree of saturation increases by 45.5%. A new additional lane in each direction on Semarang–Demak main corridor is simulated to anticipate the full operational of Sayung Industrial Park in the future. The result shows an improvement on road performance.

Who might encounter hard-braking while speeding? Analysis for regular speeders using low-frequency taxi trajectories on arterial roads and explainable AI

Regular speeders are those who commit speeding recidivism during a period. Among their speeding behaviors, some occurring in specific scenarios may cause more hazards to road users. Therefore, there is a need to evaluate the driving risks if the regular speeders have different speeding propensities. This study considers speeding-related hard-braking events (SHEs) as a safety surrogate measure and recognizes the regular speeders who encounter at least one SHEs during the study period as risky individuals. To identify speeding behaviors and hard-braking events from low-frequency GPS trajectories, we compare the average travel speed between pairwise adjacent GPS points to the posted speed limit and examine the speed curve and the corresponding travel distance between these GPS points, respectively. Thereafter, a logistic model, XGBoost, and three 1D Convolutional Neural Networks (CNNs) including AlexNet CNN, Mini-AlexNet CNN, and Simple CNN are respectively developed to recognize the regular speeders who encountered SHEs based on their speeding propensities. The proposed Mini-AlexNet CNN achieves a global F1-score of 91% and recall of 90% on the testing data, which are superior to other models. Further, the study uses the Shapley Additive exPlanation (SHAP) framework to visually interpret the contribution of speeding propensities on SHE likelihood. It is found that speeding by 50% or greater for no more than 285 m is the most dangerous kind among all the speeding behaviors. Speeding on roads without bicycle lanes or on roads with roadside parking and excessive accesses increases the probability of encountering SHEs. Based on the analyses, we put forward tailored recommendations that aim to restrict hazard-related speeding behaviors rather than speeding behaviors of all kinds.