Xiaolei Ma

Large-scale transportation network congestion evolution prediction using deep learning theory.

Understanding how congestion at one location can cause ripples throughout large-scale transportation network is vital for transportation researchers and practitioners to pinpoint traffic bottlenecks for congestion mitigation. Traditional studies rely on either mathematical equations or simulation techniques to model traffic congestion dynamics. However, most of the approaches have limitations, largely due to unrealistic assumptions and cumbersome parameter calibration process. With the development of Intelligent Transportation Systems (ITS) and Internet of Things (IoT), transportation data become more and more ubiquitous. This triggers a series of data-driven research to investigate transportation phenomena. Among them, deep learning theory is considered one of the most promising techniques to tackle tremendous high-dimensional data. This study attempts to extend deep learning theory into large-scale transportation network analysis. A deep Restricted Boltzmann Machine and Recurrent Neural Network architecture is utilized to model and predict traffic congestion evolution based on Global Positioning System (GPS) data from taxi. A numerical study in Ningbo, China is conducted to validate the effectiveness and efficiency of the proposed method. Results show that the prediction accuracy can achieve as high as 88% within less than 6 minutes when the model is implemented in a Graphic Processing Unit (GPU)-based parallel computing environment. The predicted congestion evolution patterns can be visualized temporally and spatially through a map-based platform to identify the vulnerable links for proactive congestion mitigation.

Learning Traffic as Images: A Deep Convolutional Neural Network for Large-Scale Transportation Network Speed Prediction

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.

Spatiotemporal Recurrent Convolutional Networks for Traffic Prediction in 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.

Exploring the influential factors in incident clearance time: Disentangling causation from self-selection bias

Understanding the relationships between influential factors and incident clearance time is crucial to make effective countermeasures for incident management agencies. Although there have been a certain number of achievements on incident clearance time modeling, limited effort is made to investigate the relative role of incident response time and its self-selection in influencing the clearance time. To fill this gap, this study uses the endogenous switching model to explore the influential factors in incident clearance time, and aims to disentangle causation from self-selection bias caused by response process. Under the joint two-stage model framework, the binary probit model and switching regression model are formulated for both incident response time and clearance time, respectively. Based on the freeway incident data collected in Washington State, full information maximum likelihood (FIML) method is utilized to estimate the endogenous switching model parameters. Significant factors affecting incident response time and clearance time can be identified, including incident, temporal, geographical, environmental, traffic and operational attributes. The estimate results reveal the influential effects of incident, temporal, geographical, environmental, traffic and operational factors on incident response time and clearance time. In addition, the causality of incident response time itself and its self-selection correction on incident clearance time are found to be indispensable. These findings suggest that the causal effect of response time on incident clearance time will be overestimated if the self-selection bias is not considered.

Prioritizing Influential Factors for Freeway Incident Clearance Time Prediction Using the Gradient Boosting Decision Trees Method

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.

A geographically and temporally weighted regression model to explore the spatiotemporal influence of built environment on transit ridership

Abstract Understanding the influence of the built environment on transit ridership can provide transit authorities with insightful information for operation management and policy making, and ultimately, increase the attractiveness of public transportation. Existing studies have resorted to either traditional ordinary least squares (OLS) regression or geographically weighted regression (GWR) to unravel the complex relationship between ridership and the built environment. Time is a critical dimension that traditional GWR cannot recognize well when performing spatiotemporal analysis on transit ridership. This study addressed this issue by introducing temporal variation into traditional GWR and leveraging geographically and temporally weighted regression (GTWR) to explore the spatiotemporal influence of the built environment on transit ridership. An empirical study conducted in Beijing using one-month transit smart card and point-of-interest data at the traffic analysis zone (TAZ) level demonstrated the effectiveness of GTWR. Compared with those of the traditional OLS and GWR models, a significantly better goodness-of-fit was observed for GTWR. Moreover, the spatiotemporal pattern of coefficients was further analyzed in several TAZs with typical land use types, thereby highlighting the importance of temporal features in spatiotemporal data. Transit authorities can develop transit planning and traffic demand management policies with improved accuracy by utilizing the enhanced precision and spatiotemporal modeling of GTWR to alleviate urban traffic problems.

Probabilistic Prediction of Bus Headway Using Relevance Vector Machine Regression

Bus headway regularity heavily affects transit riders’ attitude for choosing public transportation and also serves as an important indicator for transit performance evaluation. Therefore, an accurate estimate of bus headway can benefit both transit riders and transit operators. This paper proposed a relevance vector machine (RVM) algorithm to predict bus headway by incorporating the time series of bus headways, travel time, and passenger demand at previous stops. Different from traditional computational intelligence approaches, RVM can output the probabilistic prediction result, in which the upper and lower bounds of a predicted headway within a certain probability are yielded. An empirical experiment with two bus routes in Beijing, China, is utilized to confirm the high precision and strong robustness of the proposed model. Five algorithms [support vector machine (SVM), genetic algorithm SVM, Kalman filter, k-nearest neighbor, and artificial neural network] are used for comparison with the RVM model and the result indicates that RVM outperforms these algorithms in terms of accuracy and confidence intervals. When the confidence level is set to 95%, more than 95% of actual bus headways fall within the prediction bands. With the probabilistic bus headway prediction information, transit riders can better schedule their trips to avoid late and early arrivals at bus stops, while transit operators can adopt the targeted correction actions to maintain regular headway for bus bunching prevention.

Vehicle trajectory reconstruction from automatic license plate reader data

Using perception data to excavate vehicle travel information has been a popular area of study. In order to learn the vehicle travel characteristics in the city of Ruian, we developed a common methodology for structuring travelers’ complete information using the travel time threshold to recognize a single trip based on the automatic license plate reader data and built a trajectory reconstruction model integrated into the technique for order preference by similarity to an ideal solution and depth-first search to manage the vehicles’ incomplete records phenomenon. In order to increase the practicability of the model, we introduced two speed indicators associated with actual data and verified the model’s reliability through experiments. Our results show that the method would be affected by the number of missing records. The model and results of this work will allow us to further study vehicles’ commuting characteristics and explore hot trajectories.

Analyzing the Spatial and Temporal Characteristics of Subway Passenger Flow Based on Smart Card Data

Passenger flow is a core feature of rail transportation stations, and its station-level fluctuation is strongly influenced by its surrounding land-use types. This study develops a sequential K-means clustering algorithm that utilizes smart card data to categorize Beijing subway stations. The temporal characteristics of daily inbound and outbound subway passenger flows are considered in the clustering. The stations are divided into 10 groups that are classified under three categories: employment-oriented, dual-peak, and residence-oriented stations. We analyze how these categories differ in terms of station-level passenger flow. In addition, a station-level buffer area calculation method is used to estimate the land-use density around each subway station. Considering the spatial nonstationarity of passenger flow, we employ a geographically weighted regression (GWR) model to determine the correlation effect between peak-hour passenger flow and land-use density. We then analyze the spatial distribution of the correlation coefficients. Results demonstrate that most residents commute via rail transportation, and the passenger flows for the different categories of stations exhibit distinct characteristics of residences and workplaces. The findings of this study provide insightful information and theoretical foundation for rail transportation network design and operation management.