With the increasing development of intelligent detection devices, a vast amount of traffic flow data can be collected from intelligent transportation systems. However, these data often encounter issues such as missing and abnormal values, which can adversely affect the accuracy of future tasks like traffic flow forecasting. To address this problem, this paper proposes the Attention-based Spatiotemporal Generative Adversarial Imputation Network (ASTGAIN) model, comprising a generator and a discriminator, to conduct traffic volume imputation. The generator incorporates an information fuse module, a spatial attention mechanism, a causal inference module and a temporal attention mechanism, enabling it to capture historical information and extract spatiotemporal relationships from the traffic flow data. The discriminator features a bidirectional gated recurrent unit, which explores the temporal correlation of the imputed data to distinguish between imputed and original values. Additionally, we have devised an imputation filling technique that fully leverages the imputed data to enhance the imputation performance. Comparison experiments with several traditional imputation models demonstrate the superior performance of the ASTGAIN model across diverse missing scenarios.
29 September 2023 in Journal of Advanced Transportation
Expressway transportation is an essential part of regional development. An efficient expressway system can enhance cities connectivity and coordinate long-distance trips between urban areas. Understanding how travel demand affects the flow of expressways is crucial to designing an efficient expressway management system. However, congested expressways are substantial obstacles to unimpeded expressway travel. Here, we explore the relationship between demand origin locations and congested expressways. We extract the time-varying OD demand matrix from empirical tollgate data collected in Shandong province, China. The incremental traffic assignment method is introduced to obtain the traffic flow of expressway road segments. It was found that congested expressways were generated due to only a few origin locations. In addition, expressway congestion during peak hours could be effectively alleviated by controlling the travel demand of these origin locations. Therefore, the proposed method can provide a novel perspective for expressway management.
1 January 2024 in Journal of Advanced Transportation
With the increasing density of the freeway network, frequent traffic incidents on road segments have a significant impact on the operational efficiency of the road network. Therefore, it has become urgent and important to study traffic route guidance strategy on the road network level. The previous traffic route guidance method primarily focused on the congestion on the road segments where incidents occurred, with insufficient attention given to the impact of congestion on the road network level. In this study, a route guidance model with limited overlap is proposed to improve freeway network reliability under traffic incidents. Specifically, in order to explore alternative paths, we conducted a study on the problem of finding k‐short paths with limited overlap. The objective is to identify a set of k‐paths that are both sufficiently dissimilar and as short as possible. Then, we promptly update the route guidance information using a stochastic dynamic traffic assignment model that aligns with travelers’ path choice psychology. Moreover, we use the reliability of the road network to evaluate the network performance. To illustrate the model, the Jinan freeway network is selected as an experimental study. The effectiveness of this method was validated through SUMO simulations, comparing it with alternative route guidance methods, including Yen’s algorithm, A∗ algorithm, and ant colony algorithm. These results show that the proposed method has proven effective in mitigating traffic congestion arising from incidents and performs well in regard to the reliability of the road network under the impact of incidents.
14 August 2023 in IET Intelligent Transport Systems
Studites on traffic management is crucial for the development of intelligent transportation systems and smart cities. However, identifying the development stages of traffic management field based on bibliometric analysis is still lacking. In this study, CiteSpace and VOSviewer software are used to explore “traffic management” field by summarizing development process and predicting future research trend. A total of 3,028 relevant documents over the past 40 years were collected from Web of Science. Results show that (1) studies on traffic management were mainly published by researchers from USA (30.55%), China (20.90%), and some European countries; (2) the key traffic management research contents can be classified into four categories, that is, background requirements, traffic problems, method models, and control strategies; (3) the evolution process can be divided into four stages, that is, budding stage (1990–1994), development stage (1995–2003), calm stage (2004–2010), and maturation stage (2011–); (4) machine learning, deep learning and other intelligent algorithms have played more important roles in recent years, and connected vehicle management is also a potential development trend. Results suggest that cooperative vehicle‐infrastructure systems or machine learning‐based studies might be the hotspots on traffic management studies.
1 January 2024 in Journal of Advanced Transportation
Expressway traffic information is important for guiding driving routes and alleviating traffic congestion. However, the current research on expressway guidance information focuses on existing expressways. In this study, strategies for providing expressway guidance information under reconstruction and expansion scenarios are investigated. Multiple factors of expressway reconstruction and expansion, such as the length of construction areas and the number of lanes occupied by construction areas, are extracted. A panel latent class logit model considering individual heterogeneity is established to fit the survey data collected by 825 respondents. The results show that the proposed panel latent class logit model fits the data best, and the studied drivers could be categorized into three classes, i.e., the information provision time‐sensitive class, the information promotion detour class, and the information suppression detour class. The research results can support expressway operators in designing appropriate traffic information provision strategies, providing personalized guidance to drivers, and ensuring the safe operation of expressways in construction areas.