Search

Understanding travelers' route choices can help policymakers devise optimal operational and planning strategies for both normal and abnormal circumstances. However, existing choice modeling methods often rely on predefined assumptions and struggle to capture the dynamic and adaptive nature of travel behavior. Recently, Large Language Models (LLMs) have emerged as a promising alternative, demonstrating remarkable ability to replicate human-like behaviors across various fields. Despite this potential, their capacity to accurately simulate human route choice behavior in transportation contexts remains doubtful. To satisfy this curiosity, this paper investigates the potential of LLMs for route choice modeling by introducing an LLM-empowered agent, "LLMTraveler." This agent integrates an LLM as its core, equipped with a memory system that learns from past experiences and makes decisions by balancing retrieved data and personality traits. The study systematically evaluates the LLMTraveler's ability to replicate human-like decision-making through two stages: (1) analyzing its route-switching behavior in single origin-destination (OD) pair congestion game scenarios, where it demonstrates patterns align with laboratory data but are not fully explained by traditional models, and (2) testing its capacity to model day-to-day (DTD) adaptive learning behaviors on the Ortuzar and Willumsen (OW) network, producing results comparable to Multinomial Logit (MNL) and Reinforcement Learning (RL) models. These experiments demonstrate that the framework can partially replicate human-like decision-making in route choice while providing natural language explanations for its decisions. This capability offers valuable insights for transportation policymaking, such as simulating traveler responses to new policies or changes in the network.

Artificial intelligence (AI) plays a crucial role in autonomous driving (AD) research, propelling its development towards intelligence and efficiency. Currently, the development of AD technology follows two main technical paths: modularization and end-to-end. Modularization decompose the driving task into modules such as perception, prediction, planning, and control, and train them separately. Due to the inconsistency of training objectives between modules, the integrated effect suffers from bias. End-to-end attempts to address this issue by utilizing a single model that directly maps from sensor data to control signals. This path has limited learning capabilities in a comprehensive set of features and struggles to handle unpredictable long-tail events and complex urban traffic scenarios. In the face of challenges encountered in both paths, many researchers believe that large language models (LLMs) with powerful reasoning capabilities and extensive knowledge understanding may be the solution, expecting LLMs to provide AD systems with deeper levels of understanding and decision-making capabilities. In light of the challenges faced by both paths, many researchers believe that LLMs, with their powerful reasoning abilities and extensive knowledge, could offer a solution. To understand if LLMs could enhance AD, this paper conducts a thorough analysis of the potential applications of LLMs in AD systems, including exploring their optimization strategies in both modular and end-to-end approaches, with a particular focus on how LLMs can tackle the problems and challenges present in current solutions. Furthermore, we discuss an important question: Can LLM-based artificial general intelligence (AGI) be a key to achieve high-level AD? We further analyze the potential limitations and challenges that LLMs may encounter in promoting the development of AD technology.

Lane-changing is a critical maneuver of vehicle driving, and a comprehensive understanding of its impact on traffic is essential for effective traffic management and optimization. Unfortunately, existing studies fail to adequately distinguish the impact of lane changes from those resulting from natural traffic dynamics. Additionally, there is a lack of precise methods for measuring the spatial extent and duration of the impact of a single discretionary lane change, as well as a definitive metric to quantify the overall spatiotemporal impact. To address these gaps, this study introduces a quantitative indicator called the Corrected Travel Distance Bias (CTDB), which accounts for variable speeds due to inherent traffic dynamics, providing a more accurate assessment of lane-changing impacts. A comprehensive methodology is developed to compare vehicle trajectory data before and after lane-changing events, measuring both the magnitude and spatiotemporal extent of the lane-changing impact. The results, based on the Zen traffic data from Japan, indicate that the impact of a lane change in the target lane lasts an average of 23.8 seconds, affecting approximately 5.6 vehicles, with a CTDB value of -10.8 meters. In contrast, in the original lane, the impact lasts 25 seconds, affects 5.3 vehicles, and yields a CTDB value of 4.7 meters.

For on-demand dynamic ride-pooling services, e.g., Uber Pool and Didi Pinche, a well-designed vehicle dispatching strategy is crucial for platform profitability and passenger experience. Most existing dispatching strategies overlook incoming pairing opportunities, therefore suffer from short-sighted limitations. In this paper, we propose a forward-looking vehicle dispatching strategy, which first predicts the expected distance saving that could be brought about by future orders and then solves a bipartite matching problem based on the prediction to match passengers with partially occupied or vacant vehicles or keep passengers waiting for next rounds of matching. To demonstrate the performance of the proposed strategy, a number of simulation experiments and comparisons are conducted based on the real-world road network and historical trip data from Haikou, China. Results show that the proposed strategy outperform the baseline strategies by generating approximately 31\% more distance saving and 18\% less average passenger detour distance. It indicates the significant benefits of considering future pairing opportunities in dispatching, and highlights the effectiveness of our innovative forward-looking vehicle dispatching strategy in improving system efficiency and user experience for dynamic ride-pooling services.

It is commonly expected that drivers maintain a driving speed that is lower than or around the posted speed limit, as failure to obey may result in safety risks and fines. By taking randomly selected road segments as examples, this study compares the percentages of speeding vehicles in five countries worldwide, namely, two European countries (Germany and Italy), two Asian countries (Japan and China), and one North American country (the United States). Contrary to expectations, our results show that more than 80% of drivers violate the posted speed limits in the studied road segments in Italy, Japan, and the United States. In particular, a significant portion (45.3%) of drivers in Italy exceed the posted speed limit by a substantial margin (30 km/h), while few speeding vehicles are observed in the road segment examined in China. Meanwhile, it is found that drivers on low-speed-limit roads are more likely to exceed the posted speed limit, particularly when there are fewer on-road vehicles. The comparison of different countries' speeding fines indicates that for the purpose of preventing speeding, increasing fines (as Italy has done) is less effective than enhancing supervision (as China has done). The findings remind law enforcement agencies and traffic authorities of the importance of the supervision of driver's behavior and the necessity of revisiting the rationale for the current speed limit settings.

To explain day-to-day (DTD) route-choice behaviors and traffic dynamics observed in a series of lab experiments, Part I of this research proposed a discrete choice-based analytical dynamic model (Qi et al., 2023). Although the deterministic model could well reproduce the experimental observations, it converges to a stable equilibrium of route flow while the observed DTD evolution is apparently with random oscillations. To overcome the limitation, the paper proposes a route-dependent attraction-based stochastic process (RDAB-SP) model based on the same behavioral assumptions in Part I of this research. Through careful comparison between the model-based estimation and experimental observations, it is demonstrated that the proposed RDAB-SP model can accurately reproduce the random oscillations both in terms of flow switching and route flow evolution. To the best of our knowledge, this is the first attempt to explain and model experimental observations by using stochastic process DTD models, and it is interesting to find that the seemingly unanticipated phenomena (i.e., random route switching behavior) is actually dominated by simple rules, i.e., independent and probability-based route-choice behavior. Finally, an approximated model is developed to help simulate the stochastic process and evaluate the equilibrium distribution in a simple and efficient manner, making the proposed model a useful and practical tool in transportation policy design.

In the area of urban transportation networks, a growing number of day-to-day (DTD) traffic dynamic theories have been proposed to describe the network flow evolution, and an increasing amount of laboratory experiments have been conducted to observe travelers' behavior regularities. However, the "communication" between theorists and experimentalists has not been made well. This paper devotes to 1) detecting unanticipated behavior regularities by conducting a series of laboratory experiments, and 2) improving existing DTD dynamics theories by embedding the observed behavior regularities into a route choice model. First, 312 subjects participated in one of the eight decision-making scenarios and make route choices repeatedly in congestible parallel-route networks. Second, three route-switching behavior patterns that cannot be fully explained by the classic route-choice models are observed. Third, to enrich the explanation power of a discrete route-choice model, behavioral assumptions of route-dependent attractions, i.e., route-dependent inertia and preference, are introduced. An analytical DTD dynamic model is accordingly proposed and proven to steadily converge to a unique equilibrium state. Finally, the proposed DTD model could satisfactorily reproduce the observations in various datasets. The research results can help transportation science theorists to make the best use of laboratory experimentation and to build network equilibrium or DTD dynamic models with both real behavioral basis and neat mathematical properties.

Large-scale data missing is a challenging problem in Intelligent Transportation Systems (ITS). Many studies have been carried out to impute large-scale traffic data by considering their spatiotemporal correlations at a network level. In existing traffic data imputations, however, rich semantic information of a road network has been largely ignored when capturing network-wide spatiotemporal correlations. This study proposes a Graph Transformer for Traffic Data Imputation (GT-TDI) model to impute large-scale traffic data with spatiotemporal semantic understanding of a road network. Specifically, the proposed model introduces semantic descriptions consisting of network-wide spatial and temporal information of traffic data to help the GT-TDI model capture spatiotemporal correlations at a network level. The proposed model takes incomplete data, the social connectivity of sensors, and semantic descriptions as input to perform imputation tasks with the help of Graph Neural Networks (GNN) and Transformer. On the PeMS freeway dataset, extensive experiments are conducted to compare the proposed GT-TDI model with conventional methods, tensor factorization methods, and deep learning-based methods. The results show that the proposed GT-TDI outperforms existing methods in complex missing patterns and diverse missing rates. The code of the GT-TDI model will be available at https://github.com/KP-Zhang/GT-TDI.

Bike-sharing has received considerable practice and research attention over the past decade. As a manpower-driven transportation mode, it seems more sensitive to trip length, since one could take a shared bike to a destinated place where is too far to walk, or choose it for simply replacing walking when going to a nearby place. However, little research has paid attention to it, i.e., the differentiated effects of built environment on the bike-sharing demand with trip lengths. To fill the gap, this paper identifies a threshold of bike-sharing trip lengths from bike-sharing trace data, and employs a semiparametric geographically weighted Poisson regression (SGWPR) model to investigate the relationship between built environment and bike-sharing demand with different lengths considering the heterogeneity in the relationship. The results show that built environment has heterogeneous effects on the bike-sharing demand in urban areas, and the effects differ across groups with trip lengths. The findings contribute to understanding the relationships between built environment and bike-sharing demand, and providing supports for the placements and dispatchment of shared bikes., Comment: 24 pages, 9 figures, 7 tables

Urban rail transit is the backbone of urban transportation, and thus it is significant to understand its vulnerability, i.e., whether the system can still maintain normal operations when facing operational disturbances with different magnitudes. To this end, this paper proposes a network vulnerability assessment method with the joint consideration of static network topology and dynamic travel demand. The method includes an accessibility-based identification of station importance with time-varying passenger demand and a new dynamic vulnerability evaluation index. An empirical analysis is carried out by taking the rail transit system of Beijing, China as an example. Results show that the distribution of high-importance stations varies with the time of day, affected by both static topology and hourly-changing passenger flow. Under the disturbance of operation delay, the impact of high-importance stations on the network vulnerability changes with the increase of delayed travel demand. It is also found that some stations that serve as bridges (i.e., reasonable paths link the origin station and destination) and are visited by large passenger flow have the greatest impact on network vulnerability. Network performance degradation is obviously segmented and stratified in the case of interval continuous failure. The disruption between different lines is the main reason for system performance degradation, and some important stations within the line will act as catalysts to accelerate the performance degradation. This method provides a reference for measuring dynamic passenger flow-related network vulnerability and supplies the field with a new vulnerability evaluation index., Comment: 29 pages, 12 figures

A time-space traffic (TS) diagram, which presents traffic states in time-space cells with color, is an important traffic analysis and visualization tool. Despite its importance for transportation research and engineering, most TS diagrams that have already existed or are being produced are too coarse to exhibit detailed traffic dynamics due to the limitations of existing information technology and traffic infrastructure investment. To increase the resolution of a TS diagram and enable it to present ample traffic details, this paper introduces the TS diagram refinement problem and proposes a multiple linear regression-based model to solve the problem. Two tests, which attempt to increase the resolution of a TS diagram 4 and 16 times, are carried out to evaluate the performance of the proposed model. Data collected at different times, in different locations and even in different countries are employed to thoroughly evaluate the accuracy and transferability of the proposed model. Strict tests with diverse data show that the proposed model, despite its simplicity, is able to refine a TS diagram with promising accuracy and reliable transferability. The proposed refinement model will "save" widely existing TS diagrams from their blurry "faces" and enable TS diagrams to show more traffic details.

Short-term origin-destination (OD) flow prediction in urban rail transit (URT) plays a crucial role in smart and real-time URT operation and management. Different from other short-term traffic forecasting methods, the short-term OD flow prediction possesses three unique characteristics: (1) data availability: real-time OD flow is not available during the prediction; (2) data dimensionality: the dimension of the OD flow is much higher than the cardinality of transportation networks; (3) data sparsity: URT OD flow is spatiotemporally sparse. There is a great need to develop novel OD flow forecasting method that explicitly considers the unique characteristics of the URT system. To this end, a channel-wise attentive split-convolutional neural network (CAS-CNN) is proposed. The proposed model consists of many novel components such as the channel-wise attention mechanism and split CNN. In particular, an inflow/outflow-gated mechanism is innovatively introduced to address the data availability issue. We further originally propose a masked loss function to solve the data dimensionality and data sparsity issues. The model interpretability is also discussed in detail. The CAS-CNN model is tested on two large-scale real-world datasets from Beijing Subway, and it outperforms the rest of benchmarking methods. The proposed model contributes to the development of short-term OD flow prediction, and it also lays the foundations of real-time URT operation and management., Comment: This paper has been accepted by the Transportation Research Part C: Emerging Technologies as a regular paper

As a newly-emerging travel mode in the era of mobile internet, ride-hailing that connects passengers with private-car drivers via an online platform has been very popular all over the world. Although it attracts much attention in both practice and theory, the understanding of ride-hailing is still very limited largely because of the lack of related data. For the first time, this paper introduces ride-hailing drivers' multi-day trip order data and portrays ride-hailing mobility in Beijing, China, from the regional and driver's perspectives. The analyses from the regional perspective help understand the spatiotemporal flowing of the ride-hailing demand, and those from the driver's perspective characterize the ride-hailing drivers' preferences in providing ride-hailing services. A series of findings are obtained, such as the observation of the spatiotemporal rhythm of a city in using ride-hailing services and two categories of ride-hailing drivers in terms of the correlation between the activity space and working time. Those findings contribute to the understanding of ride-hailing activities, the prediction of ride-hailing demand, the modeling of ride-hailing drivers' preferences, and the management of ride-hailing services.

This study proposes a novel Graph Convolutional Neural Network with Data-driven Graph Filter (GCNN-DDGF) model that can learn hidden heterogeneous pairwise correlations between stations to predict station-level hourly demand in a large-scale bike-sharing network. Two architectures of the GCNN-DDGF model are explored; GCNNreg-DDGF is a regular GCNN-DDGF model which contains the convolution and feedforward blocks, and GCNNrec-DDGF additionally contains a recurrent block from the Long Short-term Memory neural network architecture to capture temporal dependencies in the bike-sharing demand series. Furthermore, four types of GCNN models are proposed whose adjacency matrices are based on various bike-sharing system data, including Spatial Distance matrix (SD), Demand matrix (DE), Average Trip Duration matrix (ATD), and Demand Correlation matrix (DC). These six types of GCNN models and seven other benchmark models are built and compared on a Citi Bike dataset from New York City which includes 272 stations and over 28 million transactions from 2013 to 2016. Results show that the GCNNrec-DDGF performs the best in terms of the Root Mean Square Error, the Mean Absolute Error and the coefficient of determination (R2), followed by the GCNNreg-DDGF. They outperform the other models. Through a more detailed graph network analysis based on the learned DDGF, insights are obtained on the black box of the GCNN-DDGF model. It is found to capture some information similar to details embedded in the SD, DE and DC matrices. More importantly, it also uncovers hidden heterogeneous pairwise correlations between stations that are not revealed by any of those matrices., Comment: 9 figures, 3 tables, accepted by Transportation Research Part C: Emerging Technologies

This paper proposes a convolutional neural network (CNN)-based method that learns traffic as images and predicts large-scale, network-wide traffic speed with a high accuracy. Spatiotemporal traffic dynamics are converted to images describing the time and space relations of traffic flow via a two-dimensional time-space matrix. A CNN is applied to the image following two consecutive steps: abstract traffic feature extraction and network-wide traffic speed prediction. The effectiveness of the proposed method is evaluated by taking two real-world transportation networks, the second ring road and north-east transportation network in Beijing, as examples, and comparing the method with four prevailing algorithms, namely, ordinary least squares, k-nearest neighbors, artificial neural network, and random forest, and three deep learning architectures, namely, stacked autoencoder, recurrent neural network, and long-short-term memory network. The results show that the proposed method outperforms other algorithms by an average accuracy improvement of 42.91% within an acceptable execution time. The CNN can train the model in a reasonable time and, thus, is suitable for large-scale transportation networks.

To alleviate traffic congestion, a variety of route guidance strategies has been proposed for intelligent transportation systems. A number of the strategies are proposed and investigated on a symmetric two-route traffic system over the past decade. To evaluate the strategies in a more general scenario, this paper conducts eight prevalent strategies on a asymmetric two-route traffic network with different slowdown behaviors on alternative routes. The results show that only mean velocity feedback strategy is able to equalize travel time, i.e., approximate user optimality; while the others fail due to incapability of establishing relations between the feedback parameters and travel time. The paper helps better understand these strategies, and suggests mean velocity feedback strategy if the authority intends to achieve user optimality., Comment: 16 pages, 8 figures