Journal of Transportation Engineering, Part A: Systems

On a highway, it is generally observed that as traffic flow increases, vehicles tend to travel closely and form platoons. However, sometimes platoons are formed at even a low traffic density and therefore a high traffic speed, and phantom traffic has been believed to be one of the causes. It is necessary to understand the process and characteristics of platoon formation and dispersion for efficient traffic operation and safety, and to understand why phantom traffic is formed even when traffic density is low. This study assumes that the random variation of speed (involuntary speed change) can be a critical factor that is involved in platoon formation and dispersion. A single-lane platoon-generation model based on a simple vehicle movement rule incorporating the random speed variation is developed and the characteristics of vehicle platoons are investigated through a simulation. The results show that random speed variation can be one of the factors related to platoon generation and dispersion. Initial speed and initial spacing also are found to affect the variation of the number and size of platoons.

This paper investigates the randomness in incident-induced capacity reductions and discusses the further impacts on delay calculation and modeling. For this purpose, incident and traffic count data sets from four important freeways in California, i.e., I-80, I-280, I-580, and, I-880, dating from February 1 to June 20, 2017, are utilized to analyze the incident capacity reductions. Accordingly, the capacity reduction distributions are identified and compared with the findings from the literature. In addition, the impact of variation in capacity reduction on incident delay is derived analytically for a deterministic queuing model. Through further analysis of the data, it is shown that capacity reduction values vary with respect to traffic conditions (e.g., volume). Accordingly, it is discussed that capacity reduction tables that do not incorporate traffic flow conditions and variance may provide incorrect estimations of delay. Considering the widely used capacity reduction tables for delay calculation, a regression tree approach is utilized to provide similar tables that provide traffic-dependent capacity reduction and variance for practitioner use.

Adding lanes is a common method to increase the capacity at intersections. The additional turning lanes are often limited by the road configuration. The paper presents a probabilistic model to calculate the total capacity at signalized intersections for a full through lane and an additional left-turn lane. The proposed model has an advantage over current estimation methodologies by considering not only stochastic arrivals of vehicles but also the signal timing plan, especially signal phasing sequence. Two blockage situations under this channelization form are taken into account. The proposed capacity model shows that the capacities of the through and left-turn movements are related to the proportion of the through traffic volume, the length of the additional turning lane, and the effective green time of both through and left-turn signal phases. It is also found that the through capacity increases with the increase of the proportion of the through traffic and the length of the additional lane, whereas the left-turn capacity declines with an increase of the proportion of the through traffic but increases with an increase of the length of the additional lane. Both capacities decrease with the increase of the green time when the $g/C$ ratio is fixed. The proposed model is compared with design recommendations and verified using the microscopic traffic software VISSIM as well as field data. The results show that the capacities obtained by theoretical models are consistent with the simulated results, and have an advantage over the design recommendations. The capacity models can provide theoretical suggestions for the design of additional turning lanes and signal timing plans.

Computer modeling of road networks requires a detailed and up-to-date data set. This paper proposes a method of combining authoritative databases with the OpenStreetMap (OSM) system. The complete route is established by finding paths in the graph constructed from partial data obtained from OSM. In order to correlate data from both sources, a method of coordinate conversion is proposed. The algorithm queries road data from OSM and provides a means of locating any point on the route in both data sets. A method of calculating the distance of any route point from the origin, and conversion between the distance and geographic coordinates, is described. Next, the location of any route point in the authoritative database is converted to the calculated route distance, which establishes a relation between the two data sources. Additionally, a method of estimating road curvature is proposed. The algorithm was validated in series of experiments. The proposed algorithm may be beneficial for researchers who collect data sets needed for computer simulations, e.g., for evaluation of optimal speed limits, and it shows the usefulness of OSM in transportation-related research.

Railway incidents continue to be a safety concern for transportation agencies throughout the United States. In particular, trespasser incidents, which are the most frequent cause of railway fatalities in the United States, are those that involve a person whose presence is prohibited or actions are unlawful involving railway property. By analyzing past data on trespassing incidents, recommendations for future improvement through countermeasures can be made. This research investigates historical trespasser incidents throughout national Amtrak data from 2011–2017. The data were analyzed at the national as well as megaregional levels in order to determine trends using 14 factors, including both incident (such as time of day, precrash activity, and gender) as well as geographic (such as population density and average income based on the census level in which the incident occurred) factors. A case study on a segment of Amtrak’s northeast corridor alignment is provided in order to serve as an example of connecting to countermeasure recommendations.

At a metered on-ramp, a vehicle queue will be formed in front of the ramp meter when on-ramp demand exceeds the ramp-metering capacity. It is therefore important to provide adequate queue storage space to prevent ramp queue from spilling to the upstream feeding facility, particularly when a metered on-ramp is retrofitted to an existing unmetered ramp. This paper investigates the impact of on-ramp traffic flow arrival profile on queue length at four metered on-ramp types. Based on field investigations at 13 ramp-metering locations, it was found that a higher demand-to-capacity ratio might not automatically lead to more queue storage needs; this nonlinear relationship indicated that queue length tends to be affected by the stochastic nature of on-ramp traffic flow. An analytical approach was then employed to model the on-ramp traffic flow arrival profile at each metered on-ramp type; results revealed that the on-ramp traffic flow arrival profile affected the queuing process at metered on-ramps. Finally, the microsimulation method was used to simulate queue lengths under various demand-to-capacity ratio scenarios at each metered on-ramp type. It was found that the queue generated by the random on-ramp traffic flow arrival profile at Type 4 metered on-ramps was up to 30% shorter than the queue generated by the nonrandom on-ramp traffic flow arrival profiles at Types 1, 2, and 3 metered on-ramps. The platoon dispersion effect at Type 3 metered on-ramps reduced queue length by up to 19.6% in comparison with Type 1 metered on-ramps; simulated queue length at Type 2 metered on-ramps was found to be 5.3% shorter than at Type 1 metered on-ramps.

This study develops a model that combines pavement and safety management system to utilize the resources more efficiently. Sometimes decision makers and road engineers need to estimate critical budgets to maintain and improve safety on roads most efficiently. The critical budget can be defined as the budget below which the rate of improvement (per million dollars) is higher compared with that above the budget. This paper develops optimization models to determine critical budgets for maintaining pavements and improving safety. The developed methodology is used to conduct a sensitivity analysis using an optimization technique (genetic algorithm) to determine the critical budget. Two critical budgets are determined from two models: pavement and safety. The primary factors in the model include pavement serviceability index, traffic volume, cost factor, roadway inventory, crash data, crash modification factor, and safety improvement countermeasures. The developed methodology uses linear integer programming to identify the list of preservation projects that provides the greatest benefit to society. The findings will help lawmakers to make funding decisions to preserve roadway networks.

Adaptive traffic control systems (ATCS) continuously adapt to changing traffic in order to improve traffic performance at signalized intersections. Typical before–after studies evaluate the success of ATCS deployments by assessing only the postimplementation traffic performance. Fully assessing the feasibility of ATCS implementation, however, requires evaluating the changes in long-term ATCS performance with changing traffic demands. This paper illustrates the assessment of long-term ATCS performance of two study corridors. The study uses $\mathrm{volume}/\mathrm{capacity}$ ($v/c$) ratio to evaluate the effects of changing corridor flow on corridor-wide delay benefits. The results show that the Sydney Coordinated Adaptive Traffic System (SCATS) improves corridor flow and decreases corridor-wide delays up to a point of ineffectiveness, beyond which the ATCS performance begins to decrease. The ineffectiveness point helps in deriving a reasonable estimate for the magnitude and duration of potential ATCS deployment benefits.

Transportation infrastructure continues to be at risk to climate change impacts including sea-level rise (SLR) inundation. Past studies have primarily focused on motorized facilities; however, nonmotorized systems also are vulnerable, especially in urban areas where bicycle mobility is essential to commuting and accessibility. This research investigates the SLR vulnerability of bicycle facilities using geographic information systems (GIS) and related economic impacts. GIS is used to map the facilities and inundation layers based on the National Oceanic and Atmospheric Administration’s (NOAA) SLR data. A vulnerability classification hierarchy is developed to showcase the level of impact for network links. A case study on Philadelphia, Pennsylvania, is provided as an example for how agencies can investigate bicycle facilities in their jurisdiction. Results show that at the 1.83 m (6 ft) inundation level, 57% of the offstreet trail mileage is part of a SLR affected link (at least one portion of the link is inundated) with 17% directly inundated, while 12% of the onstreet bike network mileage is considered a SLR affected link or a nonfunctional connector link.