Journal of Transportation Engineering, Part A: Systems

Public transport is a critical service in Copenhagen, Denmark, because many residents do not own a car, and in any event, car travel is not practical in the city center due to narrow roads and lack of parking. In response to COVID-19, Danish public health authorities have established a minimum 1-m social distancing policy in public spaces. This study simulates passenger pedestrian flow in three representative stations of the Copenhagen metro to determine if these goals can be attained and if any physical changes should be made. The study is conducted with a microsimulation in commercially available software of the passenger flow in three representative stations, with small, medium, and large traffic flows. The simulation is agent-based, and the individual objective function is minimum cost according to walking distance, comfort, and frustration. The results show that for the majority of stations, the physical infrastructure and the expected traffic flow are compatible with the social distancing goals. However, for a few of the highest demand stations, particularly those that serve as intermodal hubs, there are great difficulties in achieving the desired social distancing measures. In particular, the intermodal hub station of Nørreport does not possess corridors and escalators that are distributed correctly according to the pedestrian flow. This station is underground, and it is unfortunately not easy to change this infrastructure.

Railway tracks need to be monitored to ensure safe operations and cost-effective maintenance. The monitoring is commonly conducted using a track recording car that describes deviations from an ideal track geometry. Over time, the measurements provide time series data that can be used to model the observed track geometry deterioration process. However, without simplification, the modeling results are generally too complex to be utilized to their full extent in track asset management. Therefore, this study aimed to implement visualization techniques for track geometry deterioration modeling results analysis which benefit track asset management. The best practices on track geometry deterioration modeling were studied and applied to the track geometry history of a track section located in Finland. After testing the establishing modeling principles, proposals were made regarding the use of the results in practice. This paper presents visualization techniques that use the modeling results of individual cross-sections to generate information about longer sections of track and even whole rail networks. These visualizations digest the massive amount of information from the modeling and present it in an informative way for practitioners to utilize and benefit from. Thus, this study fills the gap between research and practice in railway track geometry deterioration modeling.

This paper explores the efficiency of a novel merging system based on a cooperative late merge strategy (CLMS) to mitigate the capacity reduction in work zones due to lane closure. Cooperative late merge strategies in connected vehicles (CV) and connected and autonomous vehicles (CAV) environments are formulated to enhance throughput by reducing gaps and increasing the synchronized speed in the work zone. We propose decentralized and centralized systems based on vehicle-to-vehicle and vehicle-to-infrastructure communication. The decentralized CLMS incorporates a modified lane-changing model to reflect the cooperative feature under the CV environment. The centralized CLMS is developed to further optimize the work zone throughput based on gap reduction and speed harmonization features enabled by CAV. The results prove that the decentralized CLMS outperforms other systems by increasing throughput as well as reducing delay and queue length. The centralized CLMS demonstrated substantial improvements compared to other systems. The simulation results prove that the decentralized CLMS improves capacity by 17% and the centralized CLMS by 45%, when compared to a traditional work zone system.

Microscopic traffic simulation is a well-established tool for the analysis of transportation systems, with a wide variety of applications in operations, safety, and planning. An essential component of traffic simulation is the car-following model, which defines how vehicles interact with each other and controls acceleration/deceleration to maintain a desired set of speeds and distances when constrained by a leading vehicle. Car-following models are governed by a set of parameters that define the car’s following behavior and can accommodate a range of values to reproduce desired conditions. Typically, calibration of a simulation scenario is conducted to approach a set of target macroscopic traffic condition indicators, such as speed, travel time, or queue, yet it rarely considers the accuracy of individual vehicle behavior, in part due to lack of detailed field data. In this paper, Naturalistic driving study (NDS) data sets were used to extract driving behavior on freeway segments at the microscopic level and directly characterize parameters in car-following models. The data extraction process is described, and the parameter values are illustrated for the Wiedemann 99 model implemented in commerically available software. Results highlight similarities and differences of these parameter values observed in the field and those by default in the software, and simulation outcomes upon NDS guided adjustment were analyzed. The process introduced can be expanded to similar data sets and other complex traffic conditions and therefore produce more accurate simulation results not only for metrics at a macroscopic level, but also for individual vehicle trajectories that closely mimic real-world driving.

Transportation agencies are implementing traffic management strategies to improve mobility and safety on freeways. Ramp metering is a traffic management strategy deployed to mitigate congestion on freeways by using traffic signals installed at on-ramps to control and regulate vehicle entry onto the freeway mainline. Estimating the mobility benefits of ramp metering is critical not only to determine the strategy’s effectiveness but also to inform the decision-making process regarding its deployment. The before-and-after approach and ramp metering shutdown experiments are conventional methods for estimating the benefits of ramp metering. These methods could overestimate or underestimate the benefits. This study aimed to estimate the expected mobility benefits of ramp metering by leveraging ramp metering downtime due to system breakdowns. Buffer index (BI), a travel time reliability measure, was selected as the performance measure. The study was based on data collected from 2016 to 2018 on a corridor with ramp metering signals (RMSs) along I-95 in Miami-Dade County, Florida. Penalized regression methods were used to identify factors that could predict the buffer indices of the freeway segment with RMSs. Factors evaluated include ramp metering operations (on/off), freeway traffic congestion levels, freeway mainline traffic speed, ramp traffic volume, and density of on-ramps and off-ramps. Results showed a 23% reduction in BIs during moderate congestion and a 28% reduction in BIs during severe congestion. Transportation agencies could use the results when evaluating RMSs operational performance and comparing their mobility impact with other alternatives.

Safety performance functions (SPFs) are used in roadway infrastructure safety management for purposes such as crash contributing factor analysis and hazardous roadway section identification. However, due to the lack of traffic data to establish SPFs particular to their locality, some regions and cities need to transfer and calibrate SPFs from other jurisdictions. In China, Shanghai City has been the only city that has conducted major safety analysis and developed facility-specific SPFs, while other cities have not built their local SPFs for lack of sufficient reliable data. To assist most cities in developing safety analysis models quickly enough to keep pace with the increase in crashes, the transferability of SPFs from Shanghai to another Chinese city of Guangzhou was explored. Considering the potential correlations between crash frequencies and explanatory variables during different times of day, negative binomial models were developed separately during morning peak-hours and off-peak-hours. When the models were evaluated by cumulative residual plot, results suggested that the local models could not be transferred directly without proper calibration. To provide insights into the selection of calibration methods considering the goodness-of-fit and practicality, this study compared various calibration methods that have been used in the literature and recommended a full Bayes method with normal distribution as informative priors to determine their practicality. The performances of calibration factor and calibration function, pooled data, the empirical Bayes method, and the full Bayes method were evaluated by mean absolute error, mean squared error, transfer index, and cumulative residuals plot. Among the tested methods, the calibration function method and the full Bayes method are recommended since they perform well in prediction accuracy and require an adaptable sample size.

This paper investigates a mixed optimal scheduling problem of an electric bus fleet and charging infrastructure based on the plug-in charging mode. Partial recharging and shifting the recharge start time are allowed in this problem. A mixed-integer optimization model is formulated to minimize the total cost of the bus system, considering the constraints of vehicle scheduling, charging event time, state of charge, and recharge scheduling in this process. To solve this problem, a solution framework consisting of two primary components is proposed. The first component is a recharge scheduling procedure that determines the start time and duration of each charging event as well as the minimum number of chargers used simultaneously for a given set of vehicle schedules. The second component is a genetic algorithm procedure that contains customized operators to guide the evolution of vehicle schedules. A real-world case study is conducted to test the model and the solution framework, and the impacts of important variables on the solution are determined.

The advantage of self-relocation of connected and automated vehicles (CAVs) can eliminate heavy searching-for-parking traffic in areas with limited parking availability. However, the floating trips will exacerbate local traffic congestion and parking competition if relocated CAVs are not well distributed in the network. To address these issues, this paper proposes a centralized dispatching-for-parking system to dynamically dispatch CAVs between different regions to optimize parking resource utilization and traffic distribution. A macroscopic modeling approach is presented with the consideration of mixed traffic flows of human-driven vehicles (HDVs) and CAVs. The system dynamics are modeled with the representation of the macroscopic fundamental diagram (MFD) in a multiregion road network. The objective of the system is to minimize the total network delay, which is formulated by the framework of model predictive control (MPC). Results of the numerical experiments in a two-region network show that the approach improves the performance of system operation and alleviates traffic congestion and imbalance between parking supply and demand in downtown areas. The sensitivity analysis on the level of CAV penetration reveals that the total network delay gradually decreases with the penetration increase, and HDVs benefit more from the MPC controller. The study demonstrates the applicability and implication of the dispatching-for-parking system in an era of CAVs.

Travel time estimation plays an important role in advanced traveler information systems (ATIS) for dynamic traffic management. Static travel time estimation (STTE) and dynamic travel time estimation (DTTE) are two of the major methods widely explored for travel time measurement. To analyze their performance on route-level travel time estimation on freeways where congestion may occur, this study developed a framework consisting of four steps: traffic state prediction, travel time estimation, results evaluation, and performance comparison. A METANET-based macroscopic traffic model was developed and employed to predict traffic states based on loop detector data. Then, a novel DTTE method was developed and is proposed herein that combines the piece-wise linear speed-based (PLSB) method and the trajectory assumption algorithm. The indices of the mean absolute relative error (MARE) and the root mean squared error (RMSE) were employed to analyze estimation accuracy by the traditional STTE method and the proposed DTTE method. The comparison results illustrate that during high-demand periods, the proposed DTTE method outperforms the traditional STTE method by producing results that better match reference travel times, which were obtained from video sensors installed along the urban freeway corridor.

The use of channelized islands to optimize the operation of advance right-turn motor vehicles (ARTMVs) is an effective intersection design method. To overcome the limitations of previous models with respect to the characteristics of nonstrict priority crossing behaviors and mixed nonmotor vehicle (NMV) flow, we have constructed a new capacity estimation model of ARTMVs. This novel model is based on the driving force model for nonstrict priority crossing behaviors, and fully describes the mixed NMV flow with perceived density. Based on the driving force model for nonstrict priority crossing, the speed of ARTMVs crossing the NMV lane and the headways of ARTMVs are obtained. Finally, the number of ARTMVs crossing the NMV lane in a saturated traffic state, i.e., the capacity, is estimated. The method does not need to consider the gap probability of NMV flow, which makes up for the influence of NMVs passing ARTMVs side-by-side. The model was verified by data collected at validation sites in Kunming, China, and compared with the gap acceptance model; the accuracy of the proposed model with heterogeneous NMV flow improved by 22.2%. The construction method of the model provides a new idea for the capacity estimation of ARTMVs under mixed traffic conditions.

Urban rail transit has effectively alleviated the pressure on road traffic. To explore the key influence factors that a rail station’s built environment has on passenger flow and its heterogeneity along with temporal and spatial changes, in this paper, a geographically and temporally weighted regression (GTWR) model was constructed to identify. Specifically, an empirical study was conducted in Xi’an, China, using 1 month of smartcard and station-level point-of-interest data. Firstly, we extracted an influence factors set (IFS) for ridership at the station level, and thereby three aspects of characteristics were obtained to establish IFS, including land usage, interchange connection facilities, and attributes for the station. Then, variables were exactly determined to describe each aspect characteristic with the analysis of the multicollinearity and spatial self-correlation. In addition, for models, ordinary least squares (OLS), geographically weighted regression (GWR), and geographically and temporally weighted regression (GTWR) were built to explore variables’ heterogeneity and variation influence in spatiotemporal for station ridership over time and location. Results reveal that GTWR outperforms in effectively capturing the spatiotemporal performance of ridership influences. Moreover, we proposed a mixture Poisson model to cluster stations with typical land-use characteristics in order of GTWRs’ application in different types of stations, for practice. In sum, ridership changes of different stations affected by a specific influences over time were analyzed, which highlighted the importance of temporal features in spatiotemporal data. Using GTWR to explore the relationship between ridership and station environment can provide insightful essential information for policymaking in urban rail transit management.

Previous studies used to estimate count-data models for the investigation of crash factors. In this study, an alternative method based on Tobit regression by simply viewing crash rates as a continuous variable was explored. Three years of data collected from four freeways in China were used to estimate the models. A correlated random parameters Tobit with heterogeneity in means (CRPTHM) model was proposed and thoroughly compared with traditional fixed parameters Tobit (FPT), random parameters Tobit (RPT), and correlated random parameters Tobit (CRPT) models. Results show that the CRPTHM model outperformed its three model counterparts by further grasping the correlation of unobserved heterogeneity. In addition, many crash factors were uncovered, including many appealing and important factors that have seldom been investigated for Chinese freeways previously (e.g., the safety effects of pavement conditions). More importantly, additional insights into the interactive safety effects of crash factors, such as the combined effects of interchange area and traffic volume, were inferred based on the CRPTHM model. In short, this study demonstrates the CRPTHM model to be an effective method to investigate road safety, especially when unobserved heterogeneity exists. Additionally, findings from the current study are believed to provide more knowledge of crash occurrence and be beneficial for the development of effective safety countermeasures.

Queue length information is a critical input for ramp metering management. Based on accurate and reliable queue length, the inflow rate can be optimized to maximize the benefit of ramp metering. This paper proposes a queue length estimation method for metered on-ramps. In the proposed method, multiple data sources including INRIX data, controller event-based data, and loop detector data are used. The proposed method is based on the resilient back-propagation neural network model. In addition, the proposed method is enhanced by two techniques. The first technique is implementing the decision tree to determine whether or not the queue length is larger than zero and the second technique is checking whether or not the queue length reaches the ramp queue capacity by using the loop occupancy rate data. Three ramps along the SR-51 freeway in Phoenix, Arizona, were selected to evaluate the proposed method. The proposed method is compared with the Kalman filter (KF)-based method that has been proposed in previous research. The results show that the average improvements over the KF-based method are 46.82% and 63.08% for the estimated mean absolute error and root-mean-square error, respectively.

The COVID-19 pandemic has caused worldwide lockdowns and similar containment measures aiming to curb the spread of the virus. Lockdown measures have been implemented in cities amid the COVID-19 outbreak. After the pandemic is under control, cities will be gradually reopened. This study aims to investigate the variations in urban travel behavior during the lockdown and reopening phases. On the basis of long-term traffic congestion index data and subway ridership data in eight typical cities of China, this study carried out comparisons on urban travel behaviors with and without the pandemic. Changes in the multimodal travel behaviors in different times of day and days of week are analyzed during the lockdown and reopening phases. Multivariate and one-way analyses of variance are conducted to show the statistical significance of the changes. This study further investigates the relationship between the returned-to-work (RTW) rate and travel behaviors in the reopening phase. A stepwise multiple regression is conducted to quantify the impacts of influencing factors (i.e., population migration index, RTW rate, socioeconomic indices, and pandemic statistical indicators) on vehicular traffic after reopening. Results show that the lockdown measure has a significant impact on reducing the traffic congestion during the peak hours on workdays, and the subway ridership dropped to below 10% of the prepandemic level during the lockdown phase. Travel demands tended to switch from subways to private vehicular travel modes during the reopening phase, leading to a rapid recovery of vehicular traffic and a slow recovery of subway ridership. The recovery of vehicular traffic is proved to be related to the RTW rate, certain city characteristics, and new COVID-19 cases after city reopening.

This study examined variations in travel time through a random-effects model using Wi-Fi and Bluetooth data. Previous studies mainly divided a day into discrete time windows and estimated an individual distribution function for each window. However, this overlooks the interrelated nature of travel time among the different time windows of a day. The present research suggests a conditional random-effects model that portrays time-varying travel time as a continuous phenomenon. It was shown that the lognormal probability distribution is the best-fitted unimodal distribution. It was hypothesized that the conditional model with a lognormal distribution as the best-fitting choice associated with random effects, it better fits real data than do existing models in the literature. The proposed model demonstrated a strong relationship between the mean and the standard deviation of travel time, and between these two factors and time of day. The travel time data obtained from Wi-Fi and Bluetooth detectors on a highway section of Tehran, Iran were used as a case study.