Large-scale traffic congestion often develops from local long queue in single intersections, accompanied by spillover. Current traffic control methods neglect the spillover effect, which may be one reason for the low efficiency during peak hours. A novel method for isolated intersection traffic control considering spillover is proposed here. This method distinguishes itself in three aspects: first, channelized section spillover is explicitly taken into account; second, the phase plans are enumerated, therefore flexibility is assured; third, the phase sequences are optimized as well. These features improve the effectiveness of the proposed method. Because of the nature of the established optimization model, the genetic algorithm is used to derive optimal signal settings. Numerical comparisons prove the usefulness of the proposed method.

This study evaluated the in-service performance of the InSync adaptive signal control (ASC) system utilizing field data collected from the private sector travel time vendor INRIX. A total of six corridors in Virginia were tested, with extensive data collected over 6 months. The measurements of effectiveness (MOEs) adopted are mainline delay savings and travel time reliability measurements (95th percentile travel time and buffer index). All the before-and-after MOEs were compared using t-tests, and the Pearson correlation of the MOEs with respect to various quantitative site-specific factors was assessed to determine if effectiveness could be tied to specific site characteristics. The results indicate that the adaptive signal control system generally reduces delay by about 25% and improves travel time reliability by about 16%. The performance of ASC system is affected by site-specific factors, like annual average daily traffic, signal density, and access point density.

Research on lane-changing behavior has been active for decades because of its significant impact on traffic operations. Most of the existing studies focus on generic weaving sections (e.g., on-ramps), and very few of them have investigated the lane-changing behavior along high-occupancy vehicle (HOV) facilities, especially in the context of comparing between different types of HOV-lane configuration or access control (i.e., continuous-access versus limited-access). In this study, extensive statistical analyses on lane-changing maneuvers have been conducted using a unique set of aerial photo data, which were collected from the same roadway segment before and after a conversion of HOV-lane configuration. To ensure the validity of this study, a Kalman filter smoothing algorithm was developed to clean the raw data. Results from this study indicate that the HOV-lane configuration has statistically significant impacts on lane-changing behaviors. For example, the lane-changing intensity (by location) along the continuous-access HOV facility is much more spread-out than the limited-access one. For merging into HOV-lane maneuvers, the time gap statistics (upon lane changing) are significantly different between the limited-access and continuous-access HOV-lane configuration. This study would be beneficial not only to the design of HOV lanes or other types of managed lanes, but also to the calibration of lane-changing behavior parameters in traffic microsimulation models.

Red light running (RLR) is one of the most common violations drivers commit at signalized intersections. To avoid RLR violations, some drivers may decide to stop abruptly, even though they had the opportunity to cross the stop line before the onset of the red light. This action happens more frequently at intersections with a red-light camera (RLC). The consequence of this change in drivers’ stopping behavior is the potential reduction of the usable clearance interval and the slight decline in the intersection capacity. However, different agencies’ guidelines take different approaches to estimate the clearance lost time (CLT) for capacity analysis of signalized intersections; there is not an adjustment factor for considering the impact of RLCs. In an attempt to quantify the effect of RLCs on the capacity of signalized intersections, field data were collected at eight intersections: four with RLCs and four without, in the cities of Opelika and Auburn, Alabama. A total of 1,191 cycles and a total of 1,863 drivers’ responses to clearance intervals were used to estimate the CLT. It was found that the estimated CLT at the approach with a RLC is approximately 2.7 s longer than the default value presented by one set of guidelines and about 1.1 s longer than those in another. On average, the unused yellow time was a half-second longer in RLC intersections than the intersections without RLCs.

The effectiveness of a traffic sign is collectively influenced by the sign’s understandability, legibility distance, glance legibility, and learnability; however, understandability has been repeatedly identified as one of the most important effectiveness measures. This study contributes to best practices for evaluating traffic sign understandability by demonstrating and comparing a variety of online survey questions and immersive driving simulation tasks. These techniques were applied to assess the understandability of five alternative tourist information (TI) signs in Oregon. Several TI sign alternatives were first tested in an online survey with 142 participants, followed by more authentic testing of 42 participants in the Oregon State University Driving Simulator. Sign INFO was correctly understood by 95.7% of the driving simulator subjects. Similar results were obtained for the other testing modalities. Notably, the understandability of TI sign alternatives, including versions with (75.4%) and without (72.8%) a circular border, did not score as high despite their prevalent usage in other contexts. However, the Sign i comprehension rates increased dramatically over those from an earlier study, indicating that periodic review of sign comprehension rates may be needed to reflect changes in understandability with time.

This paper presents a schedule-based passenger’s path-choice estimation model for a multioperator rail transit network, using automatic fare collection (AFC) data from both entry and exit stations. By introducing the train schedule connection network (TSCN), the path-choice estimation is converted into a set generation and weighted assignment problem for feasible TSCN paths (i.e., passenger trajectories). A major factor in path choice, the fail-to-board (FtB) phenomenon because of overcrowding in peak periods, is explicitly modeled. A method for estimating the FtB parameters is described and a weighted assignment function based on FtB parameters is provided. The case of a typical commuting origin-destination pair in the Beijing Subway is analyzed to demonstrate the capability of the proposed model. Results show that: (1) there can be multiple feasible trajectories for one pair of entry and exit records because of the FtB probability, especially in peak periods; and (2) the transfer penalty is more influential during offpeak periods when the in-vehicle times are similar for different paths. The model has potential for locating network capacity bottlenecks at the train run level and for evaluating the effects of adjusting train timetables on the passengers’ path choice.

This paper identifies and characterizes existing bridge formulas from international regions and presents the results of an analysis of the allowable gross vehicle weights and bridge load stress effects on single-span bridges resulting from these formulas. This is done to provide insight into this method of regulating truck size and weight to identify influencing factors and considerations for future decisions regarding the creation of new bridge formulas, or modification of existing ones. It is found that bridge formulas vary significantly in terms of the level of restrictiveness of allowable loads and imposed load effects due to the design criteria used in their development including truck configurations, bridge design methods, design loads, and allowable load rating. Bridge formulas should be designed to limit the imposed stresses on bridges based on criteria suitable to a jurisdiction’s truck fleet and infrastructure characteristics in order to adequately regulate truck sizes and weights. Many issues may result from the implementation of an unsuitable bridge formula for the infrastructure and transportation characteristics of a jurisdiction in terms of the design overstress criteria and additional axle spacing and weight limits. The unintended, and possibly undesirable, outcomes of implementation of a bridge formula must be monitored and resolved for safety, dynamic performance, and infrastructure impacts. With the continuously changing infrastructure and truck transportation characteristics, bridge formulas must be reevaluated and updated to ensure the adequacy of limit weights.

Signage plays an important role in work zones to provide guidance to drivers under changed conditions. This study investigated the safety effect of an alternative merge sign configuration in a freeway work zone. In this alternative configuration, the graphical-only lane closed sign from the Manual on Uniform Traffic Control Devices (MUTCD) was compared against a MERGE/arrow sign on one side and a Right Lane Closed sign on the other side. Although the graphical-only MUTCD signage for work zones has been in use for several years, it is not known if the signage recommended by the MUTCD offers the highest safety for all jurisdictions. The study measured driver behavior characteristics including speeds and open lane occupancies. The measurements were taken at a work zone on Interstate 70 in Missouri. The study found that the open lane occupancy upstream of the merge sign was higher for the test sign in comparison to the MUTCD sign. The occupancy values at different distances between the merge sign and the taper were similar for both signs. The test sign had 11% more traffic in the open lane upstream of the merge sign. In terms of safety, it is desirable for vehicles to occupy the open lane as far upstream from the taper as possible to avoid conflicts due to the lane drop. The analysis of speed characteristics did not reveal substantial differences between the two sign configurations. The 85th percentile speeds with the MUTCD sign were only 1.6 and 3.2 kph (1 and 2 mph) lower than the test sign at the merge sign and taper locations, respectively. In considering all the aforementioned performance measures, the alternative sign configuration was not superior, but performed equal to the MUTCD sign configuration.