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.

Predicting large-scale transportation network traffic has become an important and challenging topic in recent decades. Inspired by the domain knowledge of motion prediction, in which the future motion of an object can be predicted based on previous scenes, we propose a network grid representation method that can retain the fine-scale structure of a transportation network. Network-wide traffic speeds are converted into a series of static images and input into a novel deep architecture, namely, spatiotemporal recurrent convolutional networks (SRCNs), for traffic forecasting. The proposed SRCNs inherit the advantages of deep convolutional neural networks (DCNNs) and long short-term memory (LSTM) neural networks. The spatial dependencies of network-wide traffic can be captured by DCNNs, and the temporal dynamics can be learned by LSTMs. An experiment on a Beijing transportation network with 278 links demonstrates that SRCNs outperform other deep learning-based algorithms in both short-term and long-term traffic prediction.

Traffic forecasting has attracted considerable attention due to its importance in proactive urban traffic control and management. Scholars and engineers have exerted considerable efforts in improving the performance of traffic forecasting algorithms in terms of accuracy, reliability, and efficiency. Spatial feature representation of traffic flow is a core component that greatly influences traffic forecasting performance. In previous studies, several spatial attributes of traffic flow are ignored due to the following issues: a) traffic flow propagation does not comply with the road network, b) the spatial pattern of traffic flow varies over time, and c) single adjacent matrix cannot handle the complex and hierarchical urban traffic flow. To address the abovementioned issues, this study proposes a novel traffic forecasting algorithm called traffic transformer, which achieves great success in natural language processing. The multihead attention mechanism and stacking layers enable the transformer to learn dynamic and hierarchical features in sequential data. Two components, namely, global encoder and global-local decoder, are proposed to extract and fuse the spatial patterns globally and locally. Experimental results indicate that the proposed traffic transformer outperforms state-of-the-art methods. The learned dynamic and hierarchical features of traffic flow can help achieve a better understanding of spatial dependency of traffic flow for effective and efficient traffic control and management strategies.

Identifying and quantifying the influential factors on incident clearance time can benefit incident management for accident causal analysis and prediction, and consequently mitigate the impact of non-recurrent congestion. Traditional incident clearance time studies rely on either statistical models with rigorous assumptions or artificial intelligence (AI) approaches with poor interpretability. This paper proposes a novel method, gradient boosting decision trees (GBDTs), to predict the nonlinear and imbalanced incident clearance time based on different types of explanatory variables. The GBDT inherits both the advantages of statistical models and AI approaches, and can identify the complex and nonlinear relationship while computing the relative importance among variables. One-year crash data from Washington state, USA, incident tracking system are used to demonstrate the effectiveness of GBDT method. Based on the distribution of incident clearance time, two groups are categorized for prediction with a 15-min threshold. A comparative study confirms that the GBDT method is significantly superior to other algorithms for incidents with both short and long clearance times. In addition, incident response time is found to be the greatest contributor to short clearance time with more than 41% relative importance, while traffic volume generates the second greatest impact on incident clearance time with relative importance of 27.34% and 19.56%, respectively.

Accurate metro ridership prediction can guide passengers in efficiently selecting their departure time and transferring from station to station. An increasing number of deep learning algorithms are being utilized to forecast metro ridership due to the development of computational intelligence. However, limited efforts have been exerted to consider spatiotemporal features, which are important in forecasting ridership through deep learning methods, in large-scale metro networks. To fill this gap, this paper proposes a parallel architecture comprising convolutional neural network (CNN) and bi-directional long short-term memory network (BLSTM) to extract spatial and temporal features, respectively. Metro ridership data are transformed into ridership images and time series. Spatial features can be learned from ridership image data by using CNN, which demonstrates favorable performance in video detection. Time series data are input into the BLSTM which considers the historical and future impacts of ridership in temporal feature extraction. The two networks are concatenated in parallel and prevented from interfering with each other. Joint spatiotemporal features are fed into a fully connected network for metro ridership prediction. The Beijing metro network is used to demonstrate the efficiency of the proposed algorithm. The proposed model outperforms traditional statistical models, deep learning architectures, and sequential structures, and is suitable for ridership prediction in large-scale metro networks. Metro authorities can thus effectively allocate limited resources to overcrowded areas for service improvement.

Understanding the relationship between short-term subway ridership and its influential factors is crucial to improving the accuracy of short-term subway ridership prediction. Although there has been a growing body of studies on short-term ridership prediction approaches, limited effort is made to investigate the short-term subway ridership prediction considering bus transfer activities and temporal features. To fill this gap, a relatively recent data mining approach called gradient boosting decision trees (GBDT) is applied to short-term subway ridership prediction and used to capture the associations with the independent variables. Taking three subway stations in Beijing as the cases, the short-term subway ridership and alighting passengers from its adjacent bus stops are obtained based on transit smart card data. To optimize the model performance with different combinations of regularization parameters, a series of GBDT models are built with various learning rates and tree complexities by fitting a maximum of trees. The optimal model performance confirms that the gradient boosting approach can incorporate different types of predictors, fit complex nonlinear relationships, and automatically handle the multicollinearity effect with high accuracy. In contrast to other machine learning methods—or “black-box” procedures—the GBDT model can identify and rank the relative influences of bus transfer activities and temporal features on short-term subway ridership. These findings suggest that the GBDT model has considerable advantages in improving short-term subway ridership prediction in a multimodal public transportation system.