Journal of Transportation Engineering, Part A: Systems

Overtaking is a complicated maneuver that has implications for the operation and safety of road users on two-lane undivided roads. Understanding the factors influencing the overtaking behavior and process of human drivers becomes essential in the wake of developments in driver awareness, vehicular capabilities, and roadway design. This review classifies the factors which govern the overtaking process of drivers into five categories: vehicle, road, traffic, environment, and driver. Post classification, the effect of each factor based on empirical evidence is discussed on the overtaking behavior of drivers. Based on the analysis, a potential framework is suggested to understand the overtaking process better. Furthermore, future research directions are discussed in the context of modeling approaches, generalizability, connected and automated vehicles, and data collection efforts.

This study presents a ramp control strategy that builds upon the ALINEA framework to enhance the throughput of expressways equipped with vehicles-to-everything capabilities. The conventional ALINEA control strategy relies on input flow data from the previous cycle, which may not accurately reflect the current traffic conditions. To overcome this limitation, the gate recurrent unit is employed to predict the current traffic volume, serving as an improved input flow. Furthermore, a novel combined ramp control strategy is proposed in consideration of the driver’s tolerance level under the constraints of ramp queuing. This combined strategy selectively employs different ramp control methods based on the varying queuing conditions of vehicles on the ramp. A comparative analysis with the conventional ALINEA control strategy reveals that the improved ALINEA approach can reduce total travel times by up to 9.84% in merging area, concurrently reducing ramp queues length by 23.30%. The research used predicted traffic parameters for ramp control, which is a new framework for achieving active traffic control on ramp. In addition, the ramp control strategy takes into account the balance between the ramp and the main line, which is very helpful for avoiding the influence of ramp vehicles on adjacent urban streets.

The hypothesis that drivers are balancing perceptual speed on both sides of the visual field is proposed through the analysis of the driver’s decision on lateral positioning. Moreover, the temporal frequency of the occurrence of visual stimuli may have an impact on perceptual speed. This research conducts a series of experiments on right-turn and left-turn freeway curves to investigate the driver’s lateral driving behavior. A series of edge lines are installed on the road surface to act as visual stimuli, with temporal frequency ratios of $1:2$ and $1:4$ on both sides. Three observation sections, namely, the spiral-circle point, midpoint, and circle-spiral point of the curves, are selected to gather the driver’s lateral driving behaviors. The results revealed that drivers deviate from the side with a higher temporal frequency compared with that with a lower temporal frequency on both right-turn and left-turn curves. The average deviation distance with a temporal frequency ratio of $1:4$ is significantly greater than that with a ratio of $1:2$. Additionally, there exists a positive correlation between the deviation distance and the disparity in temporal frequency between both sides. Furthermore, the disparity in temporal frequency and the rate of expansion of the visual angle produce interactive impacts on the lateral position of vehicles. Moreover, the original lateral positions of vehicles affect this interaction. The findings of this study suggest a novel approach to enhance lateral driving maneuvers and provide a constructive countermeasure to prevent accidents on freeway curves.

Vehicular movement and driver behavior at unsignalized intersections are responsible for increasing the susceptibility to potential crashes. Among them, unsignalized blackspots need serious attention because they have records of several crashes occurring. At intersections, right-turning vehicles from major and minor roads take risky actions and undergo serious conflicts with high-speed through-moving vehicles. So, the study proposed crash severity levels intending to predict the chances of crashes at unsignalized intersection blackspots for crossing and rear-end conflicts using estimated surrogate safety measures (SSM), i.e., postencroachment time (PET) and modified time to collision (MTTC), along with indicators like critical speed and time to stop (Ts). The study also focused on determining the percentage of critical conflicts and vehicle combinations for crossing and rear-end conflicts. To achieve the outcomes of the study, video data and crash data from four unsignalized intersection blackspots (three-legged and four-legged) from Nagpur, India, were collected. The PET values were determined for right-turning and crossing conflicts, and MTTC values were used for rear-end conflicts. Based on the threshold values of these SSM parameters, a range of crash severity values was proposed using the $k$-mean clustering analysis method. The severity level was divided into four levels, from A being the most severe to D having the least severe values, indicating that drivers have less than 1 s to react to a situation of conflict. From the study, it was observed that the percentage of critical conflicts was between 20.67% and 35.22% for crossing conflicts and 66.67% to 89.47% for rear-end conflicts. The vehicle combinations of two wheelers–two wheelers (2W–2W), two wheelers–light motorized vehicles (2W–LMV), and two wheelers–three wheelers (2W–3W) were found to be critical for crossing conflicts, and for rear-end conflicts, 2W–2W, LMV–2W, and LMV–LMV were critical.

Abstract

When planned metro disruptions occur, the connectivity of the metro network is severely reduced, impacting the travel of a large number of commuters. The duration of planned metro disruptions varies from a few days to several weeks. To ensure the quality of public transportation services during this period, we propose a bus bridging strategy to address the travel issues of disrupted passengers. This paper introduces a mathematical model for the timetable and bus scheduling of bus bridging when a section of a metro line is planned for disruption, considering two types of buses. In this model, regular buses transfer disrupted passengers whose original destinations are on the disrupted metro section, whereas shuttle buses transfer disrupted passengers whose destinations are on the normal metro line. The model takes into account the impact of metro disruptions on the original passengers of regular buses and aims to minimize passenger transfer waiting time and congestion disutility. The model is solved using a commercial solver and is validated with the case study of Beijing Bus No. 16 and a hypothetical metro line. The results demonstrate that metro disruptions have a significant negative impact on the transfer waiting time and congestion disutility of regular buses. However, the timetable generated by the model effectively mitigates this impact. Additionally, the shuttle bus scheduling solutions generated by the model are not unique, and a specific schedule can be determined based on the shuttle bus costs and service quality. Both the crowding threshold of buses and the maximum number of pickups can reduce the congestion disutility of shuttle buses, but the impact of the crowding threshold is more significant.

Practical Applications

This study introduces a model to manage the evacuation of metro commuters during planned service disruptions, such as maintenance closures, announced in advance with a set disruption plan and duration. The model employs a coordinated strategy involving both regular buses and dedicated shuttle buses to evacuate affected passengers efficiently while minimizing disruptions to regular bus passengers. Case study results demonstrate that, although metro disruptions significantly impact regular bus passengers, the optimized timetable developed by the model can alleviate these effects. Passenger waiting time and congestion disutility are developed as key performance indicators. Future applications of this model extend beyond metro disruptions to encompass a wide range of urban transit challenges. For instance, the model can be adapted to manage passenger flow during major city events, natural disasters, or unexpected infrastructure failures. By providing a flexible and scalable framework, this model can support urban planners and transit authorities in enhancing resilience and efficiency in their public transportation systems, ultimately reducing the impact of transit disruptions on daily life.

As new mobility technologies become more widely used, they can transform the way that people travel. Among these technologies are autonomous vehicles (AVs) and micromobility such as bike-sharing and scooters. While the penetration and utilization of these mobility technologies have been extensively studied, little is known about their interaction and association between various age groups. Further, studies have focused on whether respondents would want AVs based upon their age rather than whether they would actually use them. The widespread adoption and use of self-driving vehicles is contingent upon user acceptance and utilization. Therefore, this study evaluated the association between different mobility options for various age groups using text network analysis (TNA) and Bayesian networks (BNs). This study used a data set collected from a survey conducted in Gilbert, Arizona, in 2019. The TNA results facilitated the extraction of additional features that could not be retrieved from the traditional data. The BN results revealed that micromobility devices are associated with higher likelihoods of wanting AVs and using AVs across all age groups, except for young adults who want dockless bikes and older adults who will use docked bikes. Furthermore, for all age groups, electric vehicles are associated with higher likelihoods of wanting AVs and using AVs. Regarding gender, young and middle-aged adult males were more likely to want AVs than their female counterparts. In addition, young and older adult males were more likely to use AVs than females. After combining the variables, the maximum probabilities of wanting AVs and using AVs in the city were 87% and 100% for young adults, 95% and 97% for middle-aged adults, and 59% and 93% for older adults, respectively. Emerging mobility society will have a better understanding of how mobility options are perceived by different age groups.

The occurrence of accidents in expressway weaving areas is significantly influenced by frequent lane change maneuvers. Acquiring the lane change behavior pattern characteristics of vehicles in this area can provide prior knowledge for autonomous vehicles when performing lane change maneuvers, which helps ensure the safety of autonomous vehicles. This study aims to extract lane change behavior patterns of vehicles in weaving areas, to analyze the distribution differences of patterns across different lane change maneuvers, and to explore risk characteristics during the lane change process. First, a lane-changing sequence segmentation method was designed based on the hierarchical Dirichlet process hidden semi-Markov model (HDP-HSMM) algorithm, taking into account the interaction with surrounding vehicles and risk factors. Second, the Gaussian mixture model latent Dirichlet allocation (GMM-LDA) algorithm was employed to cluster the segments and derive patterns of lane-changing behavior that include risk attributes. The trajectory data from the UCF SST dataset were used to validate the method framework and make an in-depth analysis. The results show that the behavior patterns obtained by this method are able to better describe the operational and risk states of the vehicle. Variations exist in the behavioral patterns of different types of lane change maneuvers throughout the entire process. Spatial distribution disparities exist in the behavior patterns of lane change maneuvers across various sections of weaving areas. The findings of this study provide behavioral characteristics of different types of lane change maneuvers in weaving areas, which might contribute to enhancing the accurate recognition of lane change behaviors by autonomous vehicles.

The rapid expansion of urban rail transit necessitates accurate short-term passenger flow forecasts (STPFF) to optimize operation plans and enhance service quality. However, the existing STPFF methods do not fully consider the effect of built environment on passenger flow. In this regard, a convolutional long short-term memory neural network (CNN-LSTM) model incorporating built environment indicators has been proposed for accurately short-term passenger flow predictions. First, a system of built environment indicators (including 11 indicators), anchored in the 5Ds framework, is introduced to depict the characteristics of the built environments surrounding rail transit stations. Then, the random forest model (RF) is utilized to measure and rank the indicator importance. Finally, using historical passenger flow and key built environment indicators as input variables, a CNN-LSTM model for short-term passenger flow forecast is built. Taking Beijing city, China as an example for empirical research, the results show that CNN-LSTM model considering built environment can improve the accuracy of STPFF. Utilizing the top four key built environment indicators (the ratio of commercial land area, density of point of interest (POI) categories, and bus station density) as input variables can effectively reduce model computational complexity while concurrently enhancing predictive accuracy. The highest forecasting accuracy of the model is achieved at a time granularity of 5 min. This study can effectively support the operation and management of urban rail transit.

The rapid advancement of car networking and autonomous driving technologies allows for more sophisticated active traffic management. Several studies have demonstrated variable speed limit (VSL) and lane change (LC) control to have the potential to smooth traffic flow and improve bottleneck throughput and road safety. However, VSL control zone length has lacked further exploration as a control parameter. This paper proposes a novel human driving vehicle (HV) discretionary lane-changing model that considers the human driver’s lane change expectation and incorporates driver types and random behaviors to model mixed traffic flow, which more realistically reflects the behavioral differences between connected and autonomous driving vehicles (CAVs) and HVs. Moreover, a multisection cell transmission model (CTM) VSL strategy is adopted to analyze traffic performance under different VSL control zone lengths and CAV penetration rates in the simulation, which provides a valuable reference for selecting the optimal VSL control zone length. Simulation results show that VSL with LC control improves traffic safety between the VSL control zone and the bottleneck. Still, under high traffic demand, control measures may negatively affect the section further upstream. The research also discovered that when control measures are activated, the increase in CAV penetration rate does not necessarily improve traffic efficiency, but it makes the traffic flow more harmonious.

Traffic flow uncertainty quantification is important for making reliable decisions in transportation operations. Compared with well-studied level prediction or point prediction models, the study of uncertainty quantification that can capture the second-order fluctuations of traffic observations is still in its infancy. Current traffic flow uncertainty quantification approaches can be classified in general into distribution- or nondistribution-based. For the former, generalized autoregressive conditional heteroscedasticity (GARCH) model and stochastic volatility (SV) have been widely applied to quantify traffic flow uncertainty in terms of prediction interval, usually under a parametric Gaussian distribution assumption. However, a parametric model relies on a prespecified model structure and cannot meet the requirement raised by the time-varying traffic condition patterns. Therefore, this paper proposed a real-time traffic condition uncertainty quantification approach based on a nonparametric probability density function (PDF) estimation. For this approach, the real-time nonparametric kernel density estimation method is applied to capture the time-varying probability density of traffic flow data based on which prediction intervals are constructed in real time using the quantiles computed from the estimated time-varying nonparametric PDF. Real-world traffic flow data are applied to validate the proposed approach. The results show that the proposed approach outperforms the comparative models of an online GARCH filter and three lower and upper bound estimation (LUBE) models based on multilayer perceptron (MLP), spiking neural network (SNN), and long short-term memory networks (LSTM). The findings indicate that the quantification of traffic condition uncertainty is complementary to the conventional traffic condition level modeling, and combined, traffic level modeling and traffic uncertainty quantification can support the development of proactive and reliable transportation applications.

Urban transportation significantly contributes to greenhouse gas emissions, with ${\mathrm{CO}}_2$ emissions from urban travelers varying across different regions. Identifying the determinants of residents’ travel-related ${\mathrm{CO}}_2$ emissions (TCE) from a spatial perspective is significant for formulating effective carbon reduction strategies. However, limited research explores the spatial heterogeneous relationship between TCE and its determinants. In this study, we focus on Guiyang, China, as a representative city and adopt Traffic Analysis Zones (TAZ) as geospatial units of analysis to capture regional differences in ${\mathrm{CO}}_2$ emissions. By analyzing data from the 2021 Guiyang resident travel survey, we measure per capita ${\mathrm{CO}}_2$ emissions and reveal that in Guiyang, each person emits 599.42 g of ${\mathrm{CO}}_2$ during their travels. Furthermore, we apply a semiparametric geographically weighted regression model to investigate the spatial nonstationarity of the variables’ impact on TCE. Our findings reveal that local variables have heterogeneous effects on TCE across different TAZs. For example, while the Huaxi district in southern Guiyang shows a negative correlation between TCE and the proportion of low-income individuals, most regions in Guiyang exhibit the opposite trend with a positive correlation. In the northern regions of Guiyang, such as Kaiyang and Xifeng, the proportion of car ownership negatively impacts TCE, contrasting with conclusions drawn from other regions. Moreover, the results of geographical variation tests indicate that sociodemographic attributes exhibit greater spatial heterogeneity in their influence on TCE, in contrast to built environment attributes. These results provide valuable theoretical support for policymakers in designing location-specific, low-carbon transportation policies, with significant implications for advancing sustainable transportation development.

Abstract

Preparing the Manual on Uniform Traffic Control Devices (MUTCD) involves gathering inputs from various public and the National Committee on Uniform Traffic Control Devices (NCUTCD) sponsoring agencies. The NCUTCD assists in developing standards, guidelines, and warrants for traffic control devices and practices employed to regulate, warn, and guide traffic flow on roadways. This organization advises the Federal Highway Administration (FHWA) and other relevant agencies on suggested updates and interpretations to the MUTCD and other nationally recognized standards. Examples of such sponsoring agencies include AASHTO, ASCE, Institute of Transportation Engineers (ITE), American Highway Users Alliance (AHUA), National Association of City Transportation Officials (NACTO), National Association of County Engineers (NACE), International Bridge, Tunnel and Turnpike Association (IBTTA), and American Traffic Safety Services Association (ATSSA). Other national and regional agencies have also been invited to provide feedback on the draft manual. Although all comments are considered valuable, those from federal agencies may hold more weight due to their expertise in specific subject matters. This study utilized text-mining techniques to analyze the comments from the sponsoring agencies, aiming to identify commonalities and distinct features in their feedback. The main question addressed was whether each agency effectively represented its interests and if there were shared interests among them. The study also explored the suggestions, questions, and recommendations raised by each agency. The findings revealed that some concerns were shared among agencies, although in most cases, they represented individual interests. Specifically, some agencies shared a common interest in speed limits, whereas others had similar topics concerning signs and markings. Other shared interests are bike lanes, 85th percentile speed, and pedestrian- and bus lane–related topics. This study’s insights into the interests and concerns of various MUTCD sponsoring agencies can aid in creating more effective, engaging, collaborative, and well-rounded technical documents. As a limitation, this study only used the current MUTCD revision comments, and hence the results reflect only recent concerns from the agencies.

Abstract

The Federal Highway Administration collects comments from both the general public and NCUTCD sponsoring agencies for modifying the MUTCD. This study used the comments from the NCUTCD sponsoring agencies submitted by May 2021 and conducted supervised and unsupervised text mining analysis to understand any available patterns, differences, or commonalities in the agencies’ concerns in shaping the MUTCD. The findings indicate the presence of various shared interests among the agencies and some specific features that were specific to the agencies. In addition, it was observed that the majority of the agencies have a great concern for the safety and well-being of society by showing a common interest in safety influencing factors such as speed, road signs, and the definition and use of the 85th percentile speed. This understanding can foster greater collaboration among the agencies, leading to more well-rounded and comprehensive guidelines.