Journal of Transportation Engineering, Part A: Systems

Previous studies on connected autonomous vehicles (CAVs) examined pavement performance and lane widths separately and in isolation, and without consideration for roadworks conditions. Hence, this study presents a holistic, optimal highway design solution for connected autonomous trucks (CATs) by testing pavement failure and traffic performance under different cross-sectional configurations incorporating a dedicated CAT lane for both normal and temporary traffic management (TTM) arrangements. Firstly, a dual three-lane motorway (D3M) was selected as a base case site. Next, previous research on substandard lanes was used to produce five nonstandard cross-section alternatives, which were then modeled using commercially available software. Capital investments to implement the alternatives were calculated by applying established industry construction cost models. Each cross-section was then subjected to different CATs penetration rates (PRs) and wheel wander regimes, and their pavement structural deterioration analyzed using the Texas Mechanistic-Empirical Asphalt Concrete Pavement Design and Analysis System (TxME) software. From this, maintenance frequencies and costs were determined. The study estimated delays and delay costs during TTM over a 20-year design period. Finally, initial investment, rehabilitation and delay costs were combined. It was found that the lowest life-cycle cost (LCC) of £19,091,470 occurred for high (80%) CAT PR operating under Standard D3M, whereas the highest LCC of £152,728,100 was also for high PR, but under Substandard D4M. Optimal LCC was found to change with different PRs. Hence cross-sections should be dynamically modifiable, given the anticipated gradual increase in PRs over time.

Lane management is expected to alleviate traffic congestion and improve mobility on roadways. Previous studies have mainly analyzed the impacts of lane management on the road segment rather than the road network. Because lane management strategies can affect traffic flows in the neighboring traffic regions and the entire road network, it is suitable to assess traffic impacts in the entire road network. This study proposed an analytical framework to evaluate lane management’s impacts and economic effects using location-based data, including road segments, traffic zones near road segments, and the road network. Traffic assignments with estimated origin-destination matrices from location-based data allow spatial and temporal impact analysis of lane management. This study analyzed the contraflow lane with movable median barriers installed at the Alex Fraser Bridge (AFB) in Vancouver, British Columbia, Canada, as a case study for lane management. In terms of traffic characteristics, the results showed that the contraflow lane with movable median barriers contributed significantly to improving the states of traffic flow on the AFB (traffic flow increased about 7.4%, travel speed increased about 48.3%, travel time decreased about 31.8%, and volume/capacity ratio decreased about 19.3% on average). This study showed that the contraflow lane on the AFB improved traffic flow and generated an economic benefit of $1.1 M per year (AFB, $12.7 M; zones near the AFB, $-\$1.4\mathrm{M}$; Vancouver area, $-\$10.1\mathrm{M}$) by estimating the changes in the value of travel time before and after lane management. This study contributes to a better understanding of using location-based data for assessing traffic impact and the economic effect of lane management operations at the network level.

Understanding the influencing factors of subway station outbound ridership provides sights into current subway system operations and future expansion needs. The accessibility of a subway station quantifies the potential opportunities that can be accessed by its outbound riders and can be a key factor that influences its existing ridership. This study captures the impacts of 10 types of subway station accessibility on the spatial and temporal variation of the outbound ridership. The geographically and temporally weighted regression (GTWR) modeling framework was used to quantify the spatiotemporal correlation and the spatiotemporal nonstationarity among subway station outbound ridership using 1-month smart card data of one of the largest subway networks in the world (Shanghai, China) containing over 60 million exits. In addition, four separate GTWR models were estimated to capture the potential differences between regular and irregular subway riders and between weekdays and weekends. The results suggest that the GTWR model outperforms the ordinary least-square models and GWR models in both goodness of model fit and explanatory accuracy. The model estimation results highlight the spatial and temporal varying impacts of four types of subway station accessibility on the outbound ridership, including accessibility to commercial locations, bus stations, healthcare facilities, and recreation locations. The results provide valuable insights for predicting subway outbound ridership as a function of spatially and temporally explicit variables which may have implications on addressing operational, tactical, and strategic challenges related to subway systems.

Abstract

Probabilistic yielding behaviors are often observed at permissive left-turn signalized intersections in some countries, which tend to affect the operational performance of intersections such as queue length of left-turn traffic. Based on queuing theory, this research developed an analytical queue length estimation model that took into account the probabilistic priority phenomenon at permitted left-turn signals. The service time distributions for both queued and nonqueued left-turn vehicles were derived, and the M/G2/1 queuing model was applied to determine the queue length of left-turn traffic. To validate the developed queue length model, stochastic simulations with different combinations of left-turn yielding rates, left-turn traffic saturation ratios, and through traffic flow rates were performed. It was found that the left-turn traffic saturation ratio is the most critical factor for the estimation of left-turn traffic queue length. In addition, for permitted intersections with a single through lane and a single left-turn lane, when the opposite through traffic flow rate is greater than $0.3\mathrm{veh}/\mathrm{s}$, the impact of left-turn yielding rate on left-turn queue length tends to be more significant.

Practical Applications

Left-turn movements are considered the most dangerous maneuverer, and traffic engineers have to balance a trade-off between traffic safety and operational efficiency when determining an appropriate left-turn traffic control strategy at signalized intersections. So, the permissive left-turn control strategy has been commonly applied at intersections with low to medium traffic demands. This paper established the relationship between left-turn traffic queue length and left-turn yielding rate, through traffic flow rate, and left-turn saturation ratio. Among these factors, the left-turn saturation ratio was found to be the primary factor that needs to be considered when determining left-turn queue length. In practice, preemptive left-turn behavior could, to some extent, improve traffic operations of the permissive left-turn traffic, while traffic management authorities should reasonably guide such behavior to minimize its impacts on traffic safety. Findings from this study could be used to evaluate the operational efficiency of permissive left-turn intersections under nonstrict priority conditions, and the estimated left-turn queue lengths have the potential of assisting in the optimal design of the storage of the left-turn lane.

The rapid growth of tram use in urban transport has been receiving attention throughout the world, but poor coordination within the traffic system hinders the full achievement of its potential. To deal with the deficiency of study on green bands concerning trams, an arterial green band model is advanced based on MAXBAND model and the multipath progression model. Moreover, a framework of generalized constraints is proposed to extend the arterial model to a network one, which is capable of creating green bands for many paths that go through different parts of a network. The goal of optimization is maximizing the weighted sum of the bandwidth for general vehicles, with trams given green bands wide enough for operation. A statistical appraisal of the model manifests its effectiveness in regard to that problem. Regarding MAXBAND as the benchmark, experimental data suggest that the proposed method offers improvements of 13.50% in average travel time and 12.10% in average halts for outbound general vehicles, while the numbers for inbound general vehicles are 25.10% and 10.00%. Meanwhile, tram operation has also been smoothened, with the travel time being reduced by 3.7% and 1.2% in outbound and inbound directions, respectively.

In contrast to passenger cars, the lane-changing (LC) behavior of heavy vehicles has a more profound impact on traffic flow. This paper investigates LC duration characteristics for heavy vehicles and passenger cars through comparative univariate and regression survival analysis. LC events are extracted from the HighD dataset. Five commonly-used univariate survival models are introduced to explore the overall survival function of LC duration between passenger cars and heavy vehicles. Three accelerate failure time models are employed to investigate the difference of the influencing factors of LC duration. Our results demonstrate that a significant difference exists in LC characteristic between these two types of vehicles. The survival curve of heavy vehicles is always above that of passenger cars, and these two curves show a trend of moving away from each other and then gradually approaching. The median survival time (MST) of heavy vehicles is 0.57 s higher than passenger cars. Heavy vehicles would maintain a longer time-headway and distance-headway with preceding vehicle when changing lane, and their LC durations are less susceptible to such interactions with the preceding vehicle and more susceptible to their own speed. Finally, the main findings, modeling implications, practical applications, and future work have been discussed. We hope this paper could contribute to our further understanding of LC behaviors of heavy vehicles and passenger cars.

This paper assesses the implications of vehicle-specific operating speed and crash data on the geometric design consistency evaluation of rural highways passing through mountainous terrain. The consistency was evaluated based on the operating speeds of passenger cars, empty trucks, and loaded trucks, and the results reveal the vehicle type has a significant impact on the assessment. Regardless of the vehicle type, Safety Criterion II has shown a consistent agreement with the degree of endangerment. The endangerment of trucks has a better agreement with the overall safety criteria (82%–86%) than passenger cars (58%–61%). These findings suggest that safety criteria, particularly Safety Criterion II, can be used to analyze the geometric design consistency of undivided roadways that traverse mountainous terrain. However, Safety Criterion II also classified several geometric elements as Fair or Good in spite of a significant number of crashes occurring at these elements. A further investigation of the operating speed using continuous speed profile data revealed a significant variability in vehicle speed on the curve, and that the operating speeds estimated using the spot speed data do not reflect this variability. This study highlights the need to consider speed variability when assessing design consistency using operating-speed-based consistency criteria.

The present study aimed to evaluate the behavioral changes of nonmotorized and public collective transport users in six Brazilian state capitals (São Paulo, Rio de Janeiro, Salvador, Brasília, Manaus, and Curitiba), due to the SARS-CoV-2 virus pandemic and their impact on the COVID-19 incidence rate. To this end, the mobility and incidence rate variables were correlated in the months of July 2020, January 2021, and March 2021 by the Pearson’s correlation method, using the Minitab software (version 19.2020.1). São Paulo presented a weak direct correlation followed by an inconclusive correlation and a moderate indirect correlation; Rio de Janeiro presented moderate indirect correlation throughout the entire study; Salvador began with a moderate indirect correlation, which evolved to a weak direct and then to an inconclusive correlation; the city of Brasília started with an inconclusive correlation, which then turned into a weak indirect correlation; Manaus began as inconclusive and then evolved into a moderate indirect and finally weak indirect correlation; Curitiba also began as inconclusive and evolved into a weak indirect and then into a weak direct correlation. From most of the Pearson’s correlation values analysis, it was possible to verify that the implementation of specific regulations aimed at containing the SARS-CoV-2 transmission associated with nonmotorized and public transport use regulation promoted a reduction of incidence rates facing an increase in mobility rates. The indirect correlation force was stronger when the regulations specifically addressed urban mobility, either by restricting the population’s access to nonessential services or specifically regulating the supply or maximum capacity and determining urban public transport vehicle hygiene.