Explore the vast range of titles available.

Continuous, automatic measurements of road traffic volume allow the obtaining of information on daily, weekly or seasonal fluctuations in road traffic volume. They are the basis for calculating the annual average daily traffic volume, obtaining information about the relevant traffic volume, or calculating indicators for converting traffic volume from short-term measurements to average daily traffic volume. The covid-19 pandemic has contributed to extensive social and economic anomalies worldwide. In addition to the health consequences, the impact on travel behavior on the transport network was also sudden, extensive, and unpredictable. Changes in the transport behavior resulted in different values of traffic volume on the road and street network than before. The article presents road traffic volume analysis in the city before and during the restrictions related to covid-19. Selected traffic characteristics were compared for 2019 and 2020. This analysis made it possible to characterize the daily, weekly and annual variability of traffic volume in 2019 and 2020. Moreover, the article attempts to estimate daily traffic patterns at particular stages of the pandemic. These types of patterns were also constructed for the weeks in 2019 corresponding to these stages of the pandemic. Daily traffic volume distributions in 2020 were compared with the corresponding ones in 2019. The obtained results may be useful in terms of planning operational and strategic activities in the field of traffic management in the city and management in subsequent stages of a pandemic or subsequent pandemics.

Traffic prediction is necessary for management departments to dispatch vehicles and for drivers to avoid congested roads. Many traffic forecasting methods based on deep learning have been proposed in recent years, and their main aim is to solve the problem of spatial dependencies and temporal dynamics. This paper proposes a useful dynamic model to predict the urban traffic volume by combining fully bidirectional LSTM, a complex attention mechanism, and external features, including weather conditions and events. First, we adopt bidirectional LSTM to obtain temporal dependencies of traffic volume dynamically in each layer, which is different from the hybrid methods combining bidirectional and unidirectional approaches. Second, we use a more elaborate attention mechanism to learn short-term and long-term periodic temporal dependencies. Finally, we collect weather condition and event information as external features to further improve the prediction precision. The experimental results show that the proposed model improves the prediction precision by approximately 3-7 percent on the NYC-Taxi and NYC-Bike datasets compared to the most recently developed method and is therefore a useful tool for urban traffic prediction.

As one of the key technologies of intelligent transportation systems, short-term traffic volume prediction plays an increasingly important role in solving urban traffic problems. In the last decade, many approaches were proposed for the traffic volume prediction from different perspectives. However, most of these approaches are based on a large amount of historical data. When there are only finite collected traffic data, they cannot be well trained, so the prediction accuracy of these approaches will be poor. In this paper, a tensor model is proposed to capture the change patterns of continuous traffic volumes. From collected traffic volume data, the element data are extracted to update the corresponding elements of the tensor model. Then, a tucker decomposition and gradient descent based algorithm is employed to impute the missing elements of the tensor model. After missing element imputation, the tensor model can be directly applied to the short-term traffic volume prediction through searching the corresponding elements of the model and the storage cost of the model is low. Our model is evaluated on real traffic volume data from PeMS dataset, which indicates that our model has higher traffic volume prediction accuracy than other approaches in the situation of finite traffic volume data.

This paper addresses the problem of locating vehicle-identification sensors along the arcs of the transportation network. The aim is to estimate the traffic volumes for a given set of routes under the assumption that the available sensors are insufficient to uniquely identify all route flows. We present a novel mixed-integer linear programming (MILP) model to determine the sensor locations so that in the system of linear equations solved in the path reconstruction phase, those routes whose volume cannot be uniquely determined, are linked to each other by equations involving a small number of unknowns. By this approach, experts’ opinions or historical information can be used to give a more precise estimation for those routes whose volumes are not uniquely observable. Since the direct resolution of the model via MILP solvers is time-consuming over moderate- and large-sized instances, by utilizing the problem structure, a genetic algorithm is adopted to find high-quality solutions to the model. Computational experiments over different instances, taken from the literature, confirm the effectiveness of the proposed model and algorithm.

Traffic volume data at fixed detectors are of great importance to track the time-varying states of urban traffic, and the volume of each turning movement (i.e., continuing straight, turning left, turning right) in an intersection plays a significant role in traffic control. It is a difficult issue to determine the complete road network volume due to the limited coverage of detector deployment. Existing traffic volume imputation methods seem to be hardly tractable to address this persistent missing data problem of fixed detectors because they mainly function by utilizing spatiotemporal patterns of historical data to address the problem of intermittent missing traffic data collected from probe vehicles. Considering the unattainable temporal information of the target movement, this paper aims to develop an end-to-end graph neural network-based imputation approach to impute traffic volume data with persistent missing patterns. We regard the road network as a directed graph with each turning movement as a node and employ a meta-learning-based graph attention mechanism with graph embedding in multiple spatial granularities to fully extract the relationship between the target node and its neighbours. In addition, a multitask learning mechanism is designed to improve the accuracy of the imputation results by setting auxiliary tasks according to the volume conservation constraints. The proposed approach has been validated with extensive simulation experiments and applied to field tests for Shenzhen and Shanghai, China, with considerable accuracy and reliability, satisfying the demand for traffic control. To the best of our knowledge, this is the first time that traffic volume has been imputed for persistent missing patterns with a graph neural network-based method.

The primary objective of this study is to evaluate the reduction percentage in the yearly concentrations of sulfur dioxide (SO2), nitrogen dioxide (NO2), and CO before and after COVID-19 in Amman, the capital city of Jordan, which has the highest population and traffic densities, and Zarqa, an industrial area with 55% of different types of industries. Additionally, this study examines the effect of metrological parameters such as temperature, humidity, and wind speed on air pollutant dispersion, particularly particulate matter 10 (PM10), which is considered uncontrollable. Furthermore, this study highlights the critical environmental and health effects of air pollution. The Ministry of Environment measured the yearly concentration of air pollutants (SO2, NO2, CO, and PM10) in three areas (Amman, Zarqa, and Irbid) in 12 stations in nearby industrial, urban, and traffic areas using the nitric oxide (NO) NO2 chemiluminescence analyzer Model 42i, hydrogen sulfide (H2S) and SO2 analyzer model 450iQ, and PM10 Peta Attenuation analyzer. The few air pollution studies in Jordan have primarily focused on average yearly concentrations of SO2, NO2, CO, and PM10 without considering the monthly or daily variations that greatly concern health and the environment. The results of the present study reveal that during the COVID-19 pandemic, there was a significant decrease in the annual concentrations of H2S, SO2, and NO2 as the reduction percentage in Amman 70, 58, 87% respectively, and in Zarqa 36, 62, 72% respectively. However, there is a slight reduction in CO and PM10 with 39 and 18% at Amman and 19% and 40% at Zarqa. This decrease is attributed to the reduction of primary sources of air pollutants, which are linked to the reductions in traffic volume and industrial activities during the lockdown. Furthermore, the results show that the Jordanian government has implemented regulations to address air pollution in residential areas. These regulations aim to prevent the burning of trees and smoking. The government is also adopting new transportation technologies to reduce the impact of CO2 and other pollutants produced by diesel and gasoline vehicles. The use of green fuels like synthetic natural gas, green methanol, or ammonia, as well as the increasing use of electric cars, are being encouraged. Implementing the bus rapid transit system, which started in 2021 and includes linked lines in the east and west areas of Jordan, has reduced the number of cars used and solved the main issues in crowded regions. Overall, the country has taken significant steps to address and control air pollution.

This paper introduces a robust methodology for predicting traffic volume and speed on major strategic routes in Hong Kong by leveraging data from data.gov.hk and utilizing deep learning models. The approach offers predictions from 6 min to 1 h, considering detector reliability. By extracting hidden deep features from historical detector data to establish detector profiles and grouping detectors into clusters based on profile similarities, the method employs a CNN-LSTM prediction model for each cluster. The study demonstrates the model’s resilience to detector failures, with tests conducted across failure rates from 1% to 20%, highlighting its ability to maintain accurate predictions despite random failures. In scenarios without failed detectors, the method achieves favorable performance metrics: MAE, RMSE, and MAPE for traffic volume prediction over the next 6 min stand at 5.17 vehicles/6 min, 7.64 vehicles/6 min, and 14.07%, respectively, while for traffic speed prediction, the values are 3.70 km/h, 6.32 km/h, and 6.33%. Considering a failure rate of approximately 6% in the Hong Kong dataset, in simulated scenarios with 6% failures, the model maintains its predictive accuracy, with average MAE, RMSE, and MAPE for traffic volume prediction at 5.24 vehicles/6 min, 7.81 vehicles/6 min, and 14.21%, and for traffic speed prediction at 3.87 km/h, 6.55 km/h, and 6.68%. However, the limitation of the proposed method is its potential to underperform when predicting rare or unseen scenarios, indicating the need for future research to incorporate additional data sources and methods to enhance predictive performance.

Benefiting from the rapid development of geospatial big data-related technologies, intelligent transportation systems (ITS) have become a part of people’s daily life. Traffic volume forecasting is one of the indispensable tasks in ITS. The spatiotemporal graph neural network has attracted attention from academic and business domains for its powerful spatiotemporal pattern capturing capability. However, the existing work focused on the overall traffic network instead of traffic nodes, and the latter can be useful in learning different patterns among nodes. Moreover, there are few works that captured fine-grained node-specific spatiotemporal feature extraction at multiple scales at the same time. To unfold the node pattern, a node embedding parameter was designed to adaptively learn nodes patterns in adjacency matrix and graph convolution layer. To address this multi-scale problem, we adopted the idea of Res2Net and designed a hierarchical temporal attention layer and hierarchical adaptive graph convolution layer. Based on the above methods, a novel model, called Temporal Residual II Graph Convolutional Network (Tres2GCN), was proposed to capture not only multi-scale spatiotemporal but also fine-grained features. Tres2GCN was validated by comparing it with 10 baseline methods using two public traffic volume datasets. The results show that our model performs good accuracy, outperforming existing methods by up to 9.4%.

Traffic volume estimation is a fundamental task in Intelligent Transportation Systems (ITS). The highly unbalanced and asymmetric spatiotemporal distribution of traffic flow combined with the sparse and uneven deployment of sensors pose significant challenges for accurate estimation. To address these issues, this paper proposes a novel traffic volume estimation framework. It combines a dynamic adjacency matrix Graph Convolutional Network (GCN) with a multi-scale transformer structure to capture spatiotemporal correlation. First, an adaptive speed-flow correlation module captures global road correlations based on historical speed patterns. Second, a dynamic recurrent graph convolution network is used to capture both short- and long-range correlations between roads. Third, a multi-scale transformer module models the short-term fluctuations and long-term trends of traffic volume at multiple scales, capturing temporal correlations. Finally, the output layer fuses spatiotemporal correlations to estimate the global road traffic volume at the current time. Experiments on the PEMS-BAY dataset in California show that the proposed model outperforms the baseline models and achieves good estimation results with only 30% sensor coverage. Ablation and hyperparameter experiments validate the effectiveness of each component of the model.

Traffic congestion is a major problem in the last few decades. Though the Government is taking various measures like installing more traffic signals, constructing fly-overs, one-ways, etc., the problem could not be addressed effectively on the heavily congested areas. The increasing number of vehicles, the dynamic population of traffic, the narrowness of roads and the absence of alternative roads or other routes are the major obstacles to reduce traffic congestion. Researchers of various streams have been working on predicting traffic volume to solve this problem. In this paper, a new method named Intuitionistic Fuzzy Grey-Markov Model(IFGMM) is introduced, which is an extension of Grey-Markov to Intuitionistic fuzzy set. The model is distinctive when compared with other models as it has a high level of accuracy performance. The forecasting result of this model is almost in proximity with the actual result of the traffic volume. In addition, the relative errors shown in this method are lesser than that of other models, due to its prediction accuracy is higher. Since the data collected in this method are taken from a heavily congested road from a metropolitan city, the proposed approach will be suitable for any congested area of the world.