In the road geometric design process, speed variation along the road segment is one of the most popular criteria to evaluate road consistency, hence the importance of estimating operating speed and its variations. Most of the estimation models are based on speed spot data collection, usually assuming some hypotheses like operating speed remains constant at curves and speed variations take place entirely at preceding tangent. This paper presents the results of the study of the deceleration phenomenon using a new methodology based on data collection by global positioning system devices. By means of this new methodology, not only new and more accurate models can be developed, but also the previous hypotheses can be checked and new studies can be carried out. Tangent-to-curve speed variations have been evaluated, comparing the 85th percentile speed differential (${\mathrm{\Delta }}_{85}V$) and the differential of the 85th percentile operating speed ($\mathrm{\Delta }{V}_{85}$), analyzing the percentage of deceleration length that takes place at curves, and developing two deceleration models with radius of horizontal curve and parameter of transition curve as explanatory variables.

This paper presents a Bayesian inference-based dynamic linear model (DLM) with switching based on three-phase traffic flow theory to predict online short-term travel time with plate recognition data. The proposed method combines the DLM model with a Hidden Markov Model (HMM) to capture the probability of flow breakdown and delays associated with congestion. By viewing travel time fluctuations as a time-varying stochastic process due to unforeseen events (e.g., incidents, accidents, or bad weather), the proposed dynamic linear model with Markov switching (SDLM) employs the HMM to determine the optimal traffic state sequence corresponding to a given travel time and flow rate observation sequence. The experimental results based on automatic license plate recognition data of a Jingtong Expressway stretch in Beijing City suggest that the proposed method can provide accurate and reliable travel time prediction under various traffic conditions.

This paper quantitatively assesses the performance of manual traffic-signal control by comparing its performance with two optimized traffic-signal controls during oversaturated conditions. To simulate manual traffic-signal control, the VisSim-based hardware-in-the-loop simulation (HILS), which provides similar manual controls at a field traffic-signal, was implemented. Whereas the normal-day traffic signal-timing plan was obtained from the city of Charlottesville, optimized signal-timing plans were obtained from Synchro. The results obtained from the VisSim-based simulation indicate that benefits can be achieved by utilizing manual traffic-signal control and optimized signal-controls relative to normal-day traffic-signal control during oversaturated conditions. In particular, manual traffic-signal control reduced delays to a greater extent than the two optimized signal-timing plans. Although the reduction in delays achieved by the manual signal-controls was not statistically significant when compared with the performance of the two optimized signal-controls, the results of the third and fourth manual signal-control attempts by the same operators reflecting the performance of trained manual controls showed statistically significant improvements in reducing delays.

Transportation professionals in China have started using an unconventional left-turn waiting area design as an innovative approach to mitigate traffic congestion at signalized intersections. The waiting area is set up beyond the stop bar at single or dual left-turn lanes at signalized intersections with lagging protected left-turn phases. This paper evaluated the effects of left-turn waiting areas on capacity and level of service of exclusive left-turn lanes at signalized intersections using empirical data. Capacity and delay models were developed by analyzing the arrival and discharge patterns of left-turning vehicles at single and dual left-turn lanes with differently sized left-turn waiting areas. The saturation headway, start-up lost time, and clearance time for left-turning passenger cars were estimated with field data measured from 22 sites. The capacity and delay models were calibrated and validated against both field-measured and simulated data. The results of sensitivity analyses showed that left-turn waiting areas increase the capacity of left-turn lanes at signalized intersections, and the capacity gains will increase with an increase in the storage capacity of the left-turn waiting areas. A procedure was proposed to determine the level of service of left-turn lanes with left-turn waiting areas at signalized intersections.

This work proposed a model for transit schedule design by integrating the spatial-equity considerations into the slack-time setting problem. The proposed model can help to design a transit service for more equitable reliability of all stops distributed along the route. The transit operators’ risk taking attitudes are also taken into consideration by introducing the travel-time budget (TTB) and mean-excess travel time (METT) as two risk measures. The proposed model can be formulated as a stochastic optimization problem and solved by a subsequently designed Monte Carlo (MC) simulation based genetic algorithm (GA) procedure. Scenario-based numerical examples from a real bus route in Beijing city of China are presented to demonstrate the characteristics of the model. The sensitivity of the parameters setting are also taken into consideration as well, which reflects the impact on the schedule setting of various factors including actual travel-time variability, risk-taking attitudes of transit operators, the bus drivers’ adjustment, and bus-frequency setting.

Freeway service patrol (FSP) is a low-cost but effective method to identify and clear incidents quickly on freeways, and it has been widely implemented in United States. Most of the related literature focuses on evaluating the reduction of travel delay and secondary collision because of a quicker incident clearance. Although such measures are major concerns in urban areas, they may be of less concern in low-traffic areas. A major function of FSP in low-traffic areas is to assist stalled vehicles, protect stranded drivers, and provide free road assistance services, but its major benefit is to bring direct savings to stranded drivers. This paper presents a new benchmark to evaluate the benefit-cost ratio of FSP in low-traffic areas based on a customized event-driven simulation model. The event-driven simulation model is better than the analytical models and microscopic simulation in this context in that it can accommodate more details than an analytical model while achieve a faster simulating speed than microscopic simulations when the road network is large and the simulation period is long. Lastly, the Safety Assistance for Freeway Emergencies (SAFE) Patrol in Kentucky is evaluated by using the method in this paper to evaluate and estimate the benefit-cost ratios of SAFE Patrol under different scenarios.

This paper formulates the cordon pricing design problem with elastic demand as a mathematical program with complementarity constraints (MPCC) to simultaneously optimize the cordon locations and cordon-specific toll levels, thus maximizing total social welfare. The formulation is flexible so that various charging requirements, such as those on cordon numbers, cordon size, and cordon types, can be easily satisfied by slightly modifying the formulation. A dual-based heuristic algorithm is proposed to handle the problem by sequentially solving a relaxed cordon pricing design problem and an updating problem. To avoid directly dealing with the complementarity constraints contained in the two problems, the paper adopts alternative approaches by solving a series of subproblems. These subproblems can be easily handled by using available commercial solvers. Numerical tests are performed to generate different cordon designs for one single-layered cordon, multilayered cordons, and multicentered cordons. The results demonstrate that the proposed model and solution algorithm are able to efficiently produce optimal cordon pricing schemes on real-sized transportation networks.

Mixed traffic flow is often converted to equivalent flow in passenger car units (PCUs) by multiplying the number of each category of vehicles in the traffic stream by their respective PCU factors and then adding them up. PCU is a complex parameter and depends on multitude of factors. This paper presents a methodology to convert a mixed traffic stream into a homogeneous equivalent without making use of PCU factors. Traffic volume and speed data collected on different six-lane urban arterial roads in India are analyzed to determine PCU values of five different categories of vehicles found on these roads. These PCU values are used to convert heterogeneous traffic volume in vehicles per hour to homogeneous volume in PCU per hour. A new term stream equivalency factor (SEF) is introduced in this study and denoted by $K$. It is the ratio of traffic volume in PCU per hour and volume in vehicles per hour. The $K$ factor is related to traffic composition and volume on a road through the regression analysis method. A microsimulation program is used to generate the traffic flow data for two, three, four, and five categories of vehicles in the traffic stream, and these data were utilized to present a generalized solution for determining the value of $K$ for any combination of traffic composition and volume on a road. The proposed method is simple to use and expected to eliminate the problem of estimation of PCU factors required in many studies.

Despite the increasing use of continuous-flow intersections (CFIs) to contend with the congestion caused by heavy through and left-turn traffic flows, a reliable and convenient tool for the traffic community to identify potential deficiencies of a CFI’s design is not yet available. This is due to the unique geometric feature of CFI, which comprises one primary intersection and several crossover intersections. The interdependent relationship between traffic delays and queues at a CFI with five closely spaced intersections cannot be fully captured with the existing analysis models, which were developed primarily for conventional intersections. In response to such a need, this study presents a comprehensive analysis for the overall CFI delay, identifies the potential queue spillback locations, and develops a set of planning-stage models for the CFI design geometry. To facilitate the application of these proposed models, this paper also includes a case study of a CFI at the intersection of MD 4 and MD 235 constructed by the Maryland State Highway Administration.

Providing real-time bus arrival information can help to improve the service quality of a transit system and enhance its competitiveness among other transportation modes. Taking the city of Jinan, China, as an example, this study proposes two artificial neural network (ANN) models to predict the real-time bus arrivals, based on historical global positioning system (GPS) data and automatic fare collection (AFC) system data. Also, to contend with the difficulty in capturing the traffic fluctuations over different time periods and account for the impact of signalized intersections, this study also subdivides the collected dataset into a bunch of clusters. Sub-ANN models are then developed for each cluster and further integrated into a hierarchical ANN model. To validate the proposed models, six scenarios with respect to different time periods and route lengths are tested. The results reveal that both proposed ANN models can outperform the Kalman filter model. Particularly, with several selected performance indices, it has been found that the hierarchical ANN model clearly outperforms the other two models in most scenarios.

Several state DOTs utilize the statistics presented in the Federal Highway Administration (FHWA)’s version of the traffic monitoring guide (TMG) to conduct weigh-in-motion (WIM) data quality checks for DARWin-ME traffic load spectra data preparation. Because the data check statistics in the guide were originally developed in the early 1990s, there is a need to reexamine its continuing use in up-to-date truck traffic streams. This paper aims to validate the TMG data check algorithms using WIM data from the long-term pavement performance (LTPP) special pavement study (SPS) traffic pooled-fund study. Cluster analysis is performed to identify similar traffic patterns, and the statistical summary for each data check criterion is obtained. Subsequently, the statistics based on the pooled fund data are compared to those recommended in the TMG. The results of the paper conclude that the data check criteria in the TMG guide are still valid for most of the sites; however, users should be alert that the guide also acknowledges WIM stations with exceptional patterns. This study presents a framework to check the suitableness of using the guide for such cases.