Traffic Time Series Research Articles

Traffic Flow Prediction Method Based on Seasonal Characteristics and SARIMA-NAR Model

Traffic flow is used as an essential indicator to measure the performance of the road network and a pivotal basis for road classification. However, the combined prediction model of traffic flow based on seasonal characteristics has been given little attention at present. Because the seasonal autoregressive integrated moving average model (SARIMA) has superior linear fitting characteristics, it is often used to process seasonal time series. In contrast, the non-autoregressive dynamic neural network (NAR) has a vital memory function and nonlinear interpretation capabilities. They are suitable for constructing combined forecasting models. The traffic flow time series of a highway in southwest China is taken as the research object in this paper. Combining the SARIMA (0,1,2) (0,1,2)12 model and the NAR model with 15 hidden layer neurons and fourth-order delay, two combined models are constructed: the linear and nonlinear component combination method is realized by the SARIMA-NAR combination model 1, and the MSE weight combination method is used by the SARIMA-NAR combination model 2. We calculated that the prediction accuracy of SARIMA-NAR combined model 1 is as high as 0.92, and the prediction accuracy of SARIMA-NAR combined model 2 is 0.90. In addition, the traffic flow forecast under the influence of the epidemic is also discussed. Through a comprehensive comparison of multiple indicators, the results show that the SARIMA-NAR combined model 1 has better road traffic flow fitting and prediction effects and is suitable for the greater volatility of traffic flow during the epidemic. This model improves the effectiveness and reliability of traffic flow forecasting, and the forecasting process is more convenient and efficient.

Spatial-Temporal Semantic Neural Network for Time Series Forecasting

Time series forecasting play an important role in many applications, and combining neural networks to forecast time series features can uncover many potential application situations. For example, forecasting traffic flow time series data is important for urban traffic planning and traffic management. Accurate forecasting of traffic flow provides the basis for road traffic control, which in turn improves traffic efficiency and reduces congestion. In this paper, STSeNN is proposed, which combines graph neural network and TCN to dynamically capture the spatial correlation, temporal correlation and semantic correlation of data. Experiments show that the prediction performance of STSeNN on the traffic flow datasets is more accurate compared with existing methods.

A novel network traffic combination prediction model

SummaryNetwork has become an indispensable part of public life. To improve network utilization, network performance, network quality, and enhance network security, precise prediction of network traffic is an indispensable method and basis for solving the above problems. In order to accurately predict the network traffic, a novel combination prediction model for network traffic is proposed. In this model, local mean decomposition (LMD), bidirectional long short‐term memory (BiLSTM), and Bayesian optimization algorithm are combined. First, the LMD method decomposes the network traffic time series to obtain several product function (PF) components and a residual by LMD. Then, each PF component and residual is predicted with BiLSTM model. Meanwhile, the Bayesian optimization algorithm is introduced to optimize the hyperparameters of BiLSTM. Finally, the predicted value of each PF component and residual is linearly superimposed to obtain the final predicted value. Through the study of two groups of actual network traffic datasets and compared with a variety of state‐of‐the‐art prediction models, the proposed model has a preferable prediction results by comparison of the results.

A Dual-Staged Attention Based Conversion-Gated Long Short Term Memory for Multivariable Time Series Prediction

In multivariate time series modeling, it is necessary to capture short-term mutation and long-term dependence information simultaneously. However, mechanism which can capture short-term change is difficult to be used to grasp long-term dependence information, and vice versa. In order to capture both short-term mutation and long-term dependence information in the same model, this paper proposed a dual-staged attention mechanism based on conversion-gated Long Short Term Memory network(DA-CG-LSTM). Hyperbolic tangent function is introduced into the input-gate and the forget-gate of Long Short Term Memory network(LSTM), which improves the ability of the network to extract the short-term mutation information. Further, dual-staged attention mechanism is added in the network, which includes input attention and temporal attention. Input attention adaptively extracts the feature relations of exogenous sequences, and temporal attention selects the relevant hidden layer states across all the time steps. Experiments on air quality and traffic flow time series data show that the proposed network has lower average absolute error, average absolute percentage error and root mean square error by more than 50% compared with Dual-staged Attention Recurrent Neural Network(DA-RNN) and Transformation-gated LSTM(TG-LSTM).

Risk models based on uncertainty quantification for illicit traffic time series in customs context

Due to the increase of the worldwide trade, the flow of goods crossing borders is increasing; at the same time, the threats faced by society are growing. Customs are encountering many challenges and looking for new tools of decision making for countering such risks. This paper develops models based on uncertainty quantification to analyse the behaviour of the risk time series in customs. We start by introducing the Hilbertian approach related to the representation of random variables and addressing these approximations and their applications in UQ. Then we discuss an extension where these models are applied to handle the seasonal components of risks. The models are fitted to the seized quantities of the illicit traffic on five sites using moment matching method. The results provide a good description of important properties of the data and a tool of decisions making on risk analysis in cases of threats in global supply chain.

Risk models based on uncertainty quantification for illicit traffic time series in customs context

Risk models based on uncertainty quantification for illicit traffic time series in customs context

Nonstationary Temporal Matrix Factorization for Sparse Traffic Time Series Forecasting

Nonstationary Temporal Matrix Factorization for Sparse Traffic Time Series Forecasting

STNN: A Spatio-Temporal Neural Network for Traffic Predictions

Traffic is very important to route planning and people’s daily lives. Traffic prediction is still very challenging as it is affected by many complex factors including dynamic spatio-temporal dependencies and external factors (e.g., road types and nearby points of interest) in the road network. Dynamic spatio-temporal dependencies simultaneously contain spatial and temporal dependencies. Existing models for predicting traffic of links only consider the spatial dependencies from the perspective of links or the whole road network by ignoring the spatial dependencies among regions. To this end, this paper proposes a new Spatio-Temporal Neural Network (STNN) with the encoder-decoder architecture to improve the accuracy of traffic predictions by additionally taking into account the region-based spatial dependencies and external factors. Specifically, STNN learns dynamic spatio-temporal dependencies from historical traffic time series via an encoder in the perspective of the road network, with two spatial models, i.e., <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">region-based spatial model</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">link-based spatial attention model</i> in the perspectives of regions and links, respectively. Further, STNN decodes the output from the encoder via a decoder with a temporal attention model for recording long-term dependencies and fuses external factors in the road network, to improve network-wide traffic predictions. We conduct extensive experiments to evaluate the performance of STNN on three real-world traffic datasets, which shows that STNN is significantly better than the state-of-the-art models.

Wavelet‐attention‐based traffic prediction for smart cities

Traffic congestion is a problem facing today's world, especially in smart cities where the economy is booming. Solving this issue by upgrading the traffic infrastructure of the city might be very cost-inefficient as well as time-consuming. With the help of recent technologies, traffic can be predicted to give the authorities the time to react before congestion evolves. As traffic is affected by several external factors, such as weather and anomalies (accidents, not expected road closures etc.), understanding the relationship between traffic and these factors can improve the prediction even further. In this study, a new method, the weather-based traffic analysis (hereafter WBTA), is utilised to investigate the temporal correlations between the traffic flow and the exogenous weather factors at different frequencies and time intervals. In addition, a novel method, the wavelet-attention-based calculation (hereafter WABC) is introduced to help to understand the importance of each external factor, compared with the others. Five weather factors (temperature, wind speed, rain, visibility, and humidity) are analysed, weighted, and merged with each other as one auxiliary input to improve traffic prediction accuracy. Based on that, the wavelet-attention-based prediction model is introduced, where the mean squared error is reduced by 32.3% and 24.52% for one future time step prediction, and 14.9% and 18.22% for five, compared with using the traffic time series alone, and with external factors without weights, respectively.

Weighted dynamic time warping for traffic flow clustering

This paper presents a novel similarity measure to identify interesting traffic patterns on a large traffic flow time series data for the central suburbs of Melbourne city in Australia. This new measure is a weighted Dynamic Time Warping (DTW) method based on Gaussian probability function, named GWDTW, that reflects the relative importance of peak hours. We have shown its superior performance over two benchmark similarity measures, the Euclidean distance and conventional DTW measure, on the intersection clustering task using k-medoids clustering algorithm, with respect to both internal and external evaluation measures. With intensive evaluation, the results show that GWDTW is a very effective similarity measure for modelling traffic behaviours, which can provide policy makers with more valuable information for infrastructure design, and smart city development.

A complex network analysis approach for estimation and detection of traffic incidents based on independent component analysis

Traffic incidents due to non-recurring congestion frequently occur in urban environments. In this study, we propose the estimation and detection of traffic incidents based on independent component analysis (ICA) and hybrid observer (HO)-generalized likelihood ratio (GLR) techniques. First, we develop the traffic time series to obtain insight into the traffic flow and to detect traffic incidents. Then, we use time series analysis to construct complex networks. Next, we propose the ICA technique to monitor traffic flow. Then, we introduce a piecewise switched linear model based observer to estimate the possible occurrence of traffic incidents. Finally, we propose a new incident detection method that combines HO and GLR techniques. The combined HO-GLR method can produce better incident detection, improve traffic safety, and enhance traffic management systems. We have validated the effectiveness of the proposed method using simulated traffic data generated from the Ayer Rajah Expressway in Singapore and a real-world dataset from the I-880 freeway of California. The performance metrics used to evaluate the performance of the proposed method includes detection rate, false alarm rate, classification rate, mean time to detection and the area under receiving operating characteristics curve. The experimental results show that the proposed method has obtained better performance in all of the criteria when compared with other well-known methods.

Using Noise Pollution Data for Traffic Prediction in Smart Cities: Experiments Based on LSTM Recurrent Neural Networks

Traffic prediction is one of the most important use cases for smart cities. Accurate traffic information is key to managing traffic issues. Many approaches that use traffic time series data to predict traffic flow have been proposed. In addition to traffic- specific parameters, some other features (called signatures) may be associated with road traffic, i.e., air and noise pollution. In this paper, we show how noise pollution and traffic time-series data were used to train Long-Short Term Memory (LSTM) Recurrent Neural Networks (RNNs), which led to better traffic prediction on major roads in Madrid. This approach has already been used with pollution signatures. This work addresses a new potential investigation path closely related to the use of signature profiles and Artificial Intelligent techniques as a way to reduce the specialization of sensing infrastructure.

A hybrid autoregressive fractionally integrated moving average and nonlinear autoregressive neural network model for short-term traffic flow prediction

Intelligent traffic control and guidance system is an effective way to solve urban traffic congestion, improve road capacity and guarantee drivers' travel safety, while short-term traffic flow prediction is the core of intelligent traffic control and guidance system. To investigate the long-term memory and the dynamic feature of short-time traffic flow time series, a hybrid model was proposed by integrating autoregressive fractionally integrated moving average (ARFIMA) model and nonlinear autoregressive (NAR) neural network model to predict short-time traffic flow, in which ARFIMA model can address the long-term memory of linear component and NAR neural network can accommodate the dynamic feature of nonlinear residual component. First, the ARFIMA model was employed to predict the linear component of traffic flow, and the results were compared with those of autoregressive integrated moving average (ARIMA) model. Next, the NAR neural network model was adopted to forecast the nonlinear residual components, and the weighted results were considered as the predicted flow of the hybrid model. The proposed hybrid model was validated by using the cross-sectional traffic flow data in California freeways obtained from the open-access PeMS database. The results showed that the ARFIMA model considering the long-term memory can effectively predict the short-term traffic flow, and the prediction accuracy of the hybrid model is better than that of the singular models. The findings provide an alternative for the short-term traffic flow prediction with lower error and higher accuracy.

Traffic Flow Characteristics mining of Time Series Data Based on Phase Space Reconstruction Using Complex Networks

Traffic flow time series is vital for mining the traditional statistical characteristics by using the theory of statistics and machine learning when its identity is a special time series. The network analysis of the traffic flow time series, who uses the complex network of time series analysis method, is designed to inquire into the special law of traffic flow time series which uses its visualization characteristics. Through the network analysis of traffic data flow, the connotation of traffic data flow can be revealed, and the relationship between all data and some data can be further studied. Therefore, it is constructed by combination with the phase space reconstruction theory. The phase space trajectory may be squeezed and the structure of attractor may change. We need to use C-C method to estimate the time delay according to the characteristics of integral parameters, and use G-P algorithm to estimate the embedding dimension to avoid it. This study can effectively reveal the motion law of the system. After constructing the complex network of traffic flow time series with various traffic parameters, the degree distribution, clustering coefficient and modularization of the representative critical threshold corresponding network are statistically analysed. The analysis results show that the new networked structure of traffic flow time series proposed in this study has strong advantages, and its core is phase space reconstruction, which can well reflect the information space of traffic dynamic fluctuation. The time series networking method based on phase space reconstruction has become a new approach to inquire into the characteristics of traffic flow time series. The degree distribution of the actual multi-traffic parameter time series construction network satisfies the characteristics of a Gaussian distribution. Their average clustering coefficients have attenuation characteristics, and their modularization degree is obvious.

Patterns of urban foot traffic dynamics

Using publicly available traffic camera data in New York City, we quantify time-dependent patterns in aggregate pedestrian foot traffic. These patterns exhibit repeatable diurnal behaviors that differ for weekdays and weekends but are broadly consistent across neighborhoods in the borough of Manhattan. Weekday patterns contain a characteristic 3-peak structure with increased foot traffic around 9:00 am, 12:00–1:00 pm, and 5:00 pm aligned with the “9-to-5” work day in which pedestrians are on the street during their morning commute, during lunch hour, and then during their evening commute. Weekend days do not show a peaked structure, but rather increase steadily until sunset. Our study period of June 28, 2017 to September 11, 2017 contains two American holidays, the 4th of July and Labor Day, and their foot traffic patterns are quantitatively similar to weekend days despite the fact that they fell on weekdays. Projecting all days in our study period onto the weekday/weekend phase space (by regressing against the average weekday and weekend day) we find that Friday foot traffic can be represented as a mixture of both the 3-peak weekday structure and non-peaked weekend structure. We also show that anomalies in the foot traffic patterns can be used for detection of events and network-level disruptions. Finally, we show that clustering of foot traffic time series generates associations between cameras that are spatially aligned with distributions of commercial, residential, and unbuilt space proximal to the camera locations, indicating that foot traffic dynamics encode information about the surrounding built environment.

Traffic Flow Prediction in Urban Area Using Inverse Approach of Chaos Theory

Traffic congestions problem could affect everyday life especially in urban area. In order to solve the issue, an excellent traffic flow prediction needs to be developed for a better traffic management. Hence, this study was conducted in order to predict traffic flow by using the data of total volume of vehicles per hour at two main roads located in urban areas namely Selangor and Kuala Lumpur, Malaysia by using application of chaos theory. Phase space reconstruction was used to determine the chaotic behaviour of the total volume of vehicles per hour data. The reconstruction of phase space involves a single variable of the total volume of vehicles per hour data to m-dimensional phase space. Meanwhile, the inverse approach as well as local linear approximation method was used to develop prediction model of the traffic flow time series data. This study found that (i) the time series data were chaotic behaviour based on the phase space plot and (ii) inverse approach can provide prediction on the traffic flow time series data besides give excellent prediction with the value of correlation coefficient more than 0.7500. Hence, inverse approach of chaos theory can develop to prediction model towards the traffic flow in urban area; thus may help the local authorities to provide good traffic management.

Clustering framework based on multi-scale analysis of traffic flow time series

Clustering analysis of traffic flow time series can improve the accuracy of traffic flow prediction, which is the basis of traffic planning. The correlations between traffic flow series imply that there are some potential patterns of traffic flow movements. In this work, a clustering framework based on multi-scale analysis of traffic flow time series is proposed to seek these potential patterns. The framework includes the selection of the optimum algorithm, the construction of quantitative indicators to evaluate the clustering effect, and the visual inspection. The experimental results of 24-hour long-term traffic flow time series clustering on large-scale road network in Shenzhen indicate that the model can clearly distinguish different classes of traffic flow time series.

Network traffic prediction method based on autoregressive integrated moving average and adaptive Volterra filter

SummaryIn the process of controlling network congestion and network management, how to accurately predict network traffic is very important. A network traffic prediction method based on autoregressive integrated moving average (ARIMA) model and adaptive Volterra filter model is proposed in this study. Firstly, ARIMA model is utilized to capture and predict the linear features of network traffic time series. The residual sequence is generated by network traffic sequence minus ARIMA predictive value sequence. The residual sequence only contains the nonlinear characteristics of original network traffic. Then an adaptive Volterra filter is used to train and predict the nonlinear residual sequence of original network traffic. The optimal parameters of adaptive Volterra filter are obtained by improved particle swarm optimization algorithm. Add the linear value predicted by ARIMA and the nonlinear value predicted by adaptive Volterra filter to get the final predicted value. The simulation results show that the proposed prediction method has high accuracy, small prediction error, and high reliability.

Multi-Step Traffic Speed Prediction Based on Ensemble Learning on an Urban Road Network

Short-term traffic speed prediction plays an important role in the field of Intelligent Transportation Systems (ITS). Usually, traffic speed forecasting can be divided into single-step-ahead and multi-step-ahead. Compared with the single-step method, multi-step prediction can provide more future traffic condition to road traffic participants for guidance decision-making. This paper proposes a multi-step traffic speed forecasting by using ensemble learning model with traffic speed detrending algorithm. Firstly, the correlation analysis is conducted to determine the representative features by considering the spatial and temporal characteristics of traffic speed. Then, the traffic speed time series is split into a trend set and a residual set via a detrending algorithm. Thirdly, a multi-step residual prediction with direct strategy is formulated by the ensemble learning model of stacking integrating support vector machine (SVM), CATBOOST, and K-nearest neighbor (KNN). Finally, the forecasting traffic speed can be reached by adding predicted residual part to the trend one. In tests that used field data from Zhongshan, China, the experimental results indicate that the proposed model outperforms the benchmark ones like SVM, CATBOOST, KNN, and BAGGING.

Unidirectional and Bidirectional LSTM Models for Short-Term Traffic Prediction

This paper presents the development and evaluation of short-term traffic prediction models using unidirectional and bidirectional deep learning long short-term memory (LSTM) neural networks. The unidirectional LSTM (Uni-LSTM) model provides high performance through its ability to recognize longer sequences of traffic time series data. In this work, Uni-LSTM is extended to bidirectional LSTM (BiLSTM) networks which train the input data twice through forward and backward directions. The paper presents a comparative evaluation of the two models for short-term speed and traffic flow prediction using a common dataset of field observations collected from multiple freeways in Australia. The results showed BiLSTM performed better for variable prediction horizons for both speed and flow. Stacked and mixed Uni-LSTM and BiLSTM models were also investigated for 15-minute prediction horizons resulting in improved accuracy when using 4-layer BiLSTM networks. The optimized 4-layer BiLSTM model was then calibrated and validated for multiple prediction horizons using data from three different freeways. The validation results showed a high degree of prediction accuracy exceeding 90% for speeds up to 60-minute prediction horizons. For flow, the model achieved accuracies above 90% for 5- and 10-minute prediction horizons and more than 80% accuracy for 15- and 30-minute prediction horizons. These findings extend the set of AI models available for road operators and provide them with confidence in applying robust models that have been tested and evaluated on different freeways in Australia.