No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Survey on traffic prediction in smart cities
... (2) Neural network-based models In recent years, neural networks (NNs, including deep learning) have been extensively used for traffic flow prediction due to their ability to model nonlinear and non-stationary behavior, as well as their high extensibility [4]. Goudarzi et al. ...
... The most popular NN models are convolutional neural networks (CNNs) and long short-term memory (LSTM), which can integrate spatiotemporal properties or environmental data sources. Specifically, the spatial dependencies of networkwide traffic could be captured by CNNs, and the temporal dynamics could be learned by LSTM [4]. Ma et al. [7] applied CNN to traffic image analysis following two consecutive steps include abstract traffic feature extraction and network-wide traffic speed prediction. ...
... (1) ARIMA model and other mathematical equation models For example, ARIMA linear forecasting models or their hybrid variants as well as some nonlinear models combined with ARIMA. Wang [1] demonstrated a model ARIMA (4,1,5) in urban traffic flow prediction, while Lin and Huang [2] combined the ARIMA (3) Other machine learning models Other machine learning models, such as support vector machine (SVM), Bayesian models, k-nearest neighbor, etc. SVM aims to find the optimal hyperplane that separates data into classes, making it effective in capturing the nonlinear relationships observed in traffic flow data [12], Additionally, the SVM regression model (SVR) has been employed to predict future traffic flow based on historical data, with the ability to handle data contain noises [13]. Decision trees (DTs) are another widely used machine learning model for traffic forecasting, where the model splits the dataset into subgroups based on traffic feature values such as time, day of the week, and weather conditions [14][15][16]. ...
Short-term traffic flow prediction plays a crucial role in transportation systems by describing the time evolution of traffic flow over short periods, such as seconds, minutes, or hours. It helps people make informed decisions about their routes to avoid congested areas and enables traffic management departments to quickly adjust road capacities and implement effective traffic management strategies. In recent years, numerous studies have been conducted in this area. However, there is a significant gap in research regarding the uncertainty of short-term traffic flow, which negatively impacts the accuracy and robustness of traffic flow prediction models. In this paper, we propose a novel comprehensive entropy-cloud model that includes two algorithms: the Fused Cloud Model Inference based on DS Evidence Theory (FCMI-DS) and the Cloud Model Inference and Prediction based on Compensation Mechanism (CMICM). These algorithms are designed to address the short-term traffic flow prediction problem. By utilizing the cloud model of historical flow data to guide future short-term predictions, our approach improves prediction accuracy and stability. Additionally, we provide relevant mathematical proofs to support our methodology.
... A link count is defined as the cumulative traffic in the lane between two consecutive intersections of a road, usually over a short period of time [1,2]. Since link counts provide information on traffic flow and density in a road network, the availability of link count data is critical for solving important problems in traffic management, planning, and engineering [3,4]. ...
... Link count data are broadly divided into two categories: (i) scalar data (Eulerian counts at points) and (ii) vector data (Lagrangian trajectory data). Scalar data are collected from static sensors (e.g., loop detectors, magnetic sensors) installed at an intersection or anywhere along a link of a road network [1,5,6]. Vector data are trajectories, where the identity of a vehicle is trackable through time, as opposed to scalar data, which are numbers on a node without the ability to ascribe a particular vehicle. ...
... Two pivotal PIs, Mean Absolute Error (MAE) and Mean Squared Error (MSE), are computed to quantify the precision of the estimated follower velocity. MAE is defined as 1 ...
Traffic count (or link count) data represents the cumulative traffic in the lanes between two consecutive signalised intersections. Typically, dedicated infrastructure-based sensors are required for link count data collection. The lack of adequate data collection infrastructure leads to lack of link count data for numerous cities, particularly those in low- and middle-income countries. Here, we address the research problem of link count estimation using crowd-sourced trajectory data to reduce the reliance on any dedicated infrastructure. A stochastic queue discharge model is developed to estimate link counts at signalised intersections taking into account the sparsity and low penetration rate (i.e., the percentage of vehicles with known trajectory) brought on by crowdsourcing. The issue of poor penetration rate is tackled by constructing synthetic trajectories entirely from known trajectories. The proposed model further provides a methodology for estimating the delay resulting from the start-up loss time of the vehicles in the queue under unknown traffic conditions. The proposed model is implemented and validated with real-world data at a signalised intersection in Kolkata, India. Validation results demonstrate that the model can estimate link count with an average accuracy score of 82% with a very low penetration rate (not in the city, but at the intersection) of 5.09% in unknown traffic conditions, which is yet to be accomplished in the current state-of-the-art.
... Another research [5] also prioritized short-term traffic prediction and provided an overview of prediction techniques and research recommendations. In addition, the paper [6] supplied information on obtaining road traffic data. To estimate traffic congestion, a predictive model was presented that used real-time data from a traffic monitoring system and multiple machine learning techniques [7], such as Gradient Boosting (GB), Support Vector Machine (SVM), and Random Forest (RF). ...
... Additionally, MAE and MAPE are computed using Eqs. (6) and (7), respectively, elucidating the model's predictive accuracy. Furthermore, MSE is calculated using Eq. ...
Effective traffic prediction is crucial for optimizing urban transportation systems, minimizing congestion, and enhancing overall efficiency. Traffic congestion results in prolonged travel durations, higher fuel consumption, economic setbacks, and increased environmental pollution. To tackle these issues, we introduce a Hybrid CNN-GRU-LSTM model—an advanced deep learning framework that combines convolutional neural networks (CNN), gated recurrent units (GRU), and long short-term memory (LSTM) networks. This integrated model is specifically designed to capture both spatial and temporal patterns of traffic flow, making it highly effective for predicting vehicle volumes at intersections. The Hybrid CNN-GRU-LSTM leverages CNN to model spatial dependencies between road segments, while GRU and LSTM layers handle short-term and long-term temporal patterns in traffic data. This combination allows for more accurate predictions by incorporating spatial relationships and temporal dynamics simultaneously. The model was tested using publicly available datasets, including PeMS, the England dataset, the P/Castellano dataset, and the Fedesoriano dataset, and results demonstrate that Hybrid CNN-GRU-LSTM significantly outperforms several state-of-the-art models, achieving a reduction of up to 30–35% in error values. This study highlights the effectiveness of combining CNN, GRU, and LSTM architectures for traffic prediction, offering a robust solution for transportation management. The proposed model’s significant improvement in prediction accuracy can help mitigate the adverse effects of traffic congestion and enhance the overall performance of transportation networks.
... The timeframe for this type of prediction can vary from minutes to days (Zang et al. 2019). For example, long-term traffic prediction in hours can aid in comprehending traffic patterns and supporting transportation planning, because it enables the capture of significant congestion factors in specific scenarios (Nagy and Simon 2018). Furthermore, vehicle speed prediction focuses on predicting micro-speed patterns in the future, which can be valuable for trajectory planning, collision risk reduction (Liu et al. 2019), and eco-driving (Liu et al. 2021). ...
Intelligent transportation systems increasingly require accurate speed predictions to guide, manage, and control traffic. Traffic speed prediction is a significant challenge in traffic systems and is critical for efficient traffic management. However, speed data are often incomplete, noisy, and subject to measurement errors, which affect prediction accuracy. This paper introduces an innovative method for predicting speed. This approach involves utilizing the genetic algorithm to optimize the hyperparameters of a convolutional neural network model. By doing so, we aim to enhance the model’s performance and mitigate the issue of overfitting. The genetic algorithm simulates the natural selection process, in which the fittest individuals are selected and combined to create the next generation. We employ a long short-term memory network to determine the window size parameter through the genetic algorithm. Additionally, we utilize a convolutional neural network with three convolutional layers, three pooling layers, and fully connected layers to extract the temporal variations in speed. We obtain the prediction output by performing result-level fusion with the optimal hyperparameters. The proposed approach demonstrated superior performance to the other benchmarks, achieving the highest prediction accuracy and the lowest prediction error. The Nash-Sutcliff index value of the proposed model is 90.3% in the best-case scenario. Here, this approach helps improve efficiency and reduce computational burdens.
... Some of the algorithms that have already discovered its path to ITS are: Convolutional Neural Network (CNN) (Li et al., 2017), Graph Convolutional Neural Networks (GCNs) (Zheng et al., 2023), Long Short-Term Memory (LSTM) (Zhao et al., 2017), Gated Recurrent Units (GRUs) (Fu et al., 2016), Bidirectional-Long Short-Term Memory (Bi-LSTM) (Redhu et al., 2023), K-Means Clustering (Rao et al., 2019;Pustokhina et al., 2020), Autoencoders (Li et al., 2022), and Generative Adversarial Network (GAN) (Shi et al., 2024). AI algorithms can provide traffic prediction (Nagy and Simon, 2018;Lee et al., 2021), and traffic congestion detection and prediction services (Djenouri et al., 2019;Akhtar and Moridpour, 2021). This paper sheds light on automatizing the process of monitoring traffic status by employing the most common machine learning-based anomaly detection methods to detect unusual behavior in road traffic. ...
In recent years, urban roads have suffered from substantial traffic congestion due to the rapidly increasing number of road users and vehicles. Some traffic congestion patterns on specific roadways, such as the recurring congestion during morning and evening rush hours, can be foreseen. However, unexpected events, such as incidents, may also cause traffic congestion. Monitoring traffic status poses vital importance for city traffic operators. They can leverage the monitoring system for resource allocation, traffic lights adjusting, and adapting the public transport schedules to alleviate traffic congestion. Machine learning-based methods for anomaly detection are valuable tools for monitoring traffic status and promptly detecting congestion on city roads. In this paper, we comprehensively study the performance of the common machine learning methods for anomaly detection in the traffic congestion detection use case. In addition, we provide methods usage insights based on the study fin dings by examining the accuracy, detection speed, and computation overhead of the methods to guide the researchers and city operators toward a suitable method based on their needs.
... The rise of new data sources and the rapid growth of machine learning (ML) make it more precisely than ever possible to analyze and forecast traffic situations in smart cities. In an automated city of the future, this can aid in optimizing the planning and administration of transport services (Nagy & Simon, 2018). We have genuinely reached the era of big data for transport, as seen by the explosive growth of traffic data in recent years. ...
... lives through environmental monitoring [1], traffic control [2], healthcare [3], etc. Urban data is essential for smart city applications, and how to obtain and use data effectively is a key issue. Fortunately, mobile crowdsensing (MCS) [4] provides a useful way to collect data, making use of users' mobile devices (or users themselves) as sensing units to complete large-scale and complex social sensing tasks. ...
Advances in artificial intelligence (AI) including foundation models (FMs), are increasingly transforming human society, with smart city driving the evolution of urban living.Meanwhile, vehicle crowdsensing (VCS) has emerged as a key enabler, leveraging vehicles' mobility and sensor-equipped capabilities. In particular, ride-hailing vehicles can effectively facilitate flexible data collection and contribute towards urban intelligence, despite resource limitations. Therefore, this work explores a promising scenario, where edge-assisted vehicles perform joint tasks of order serving and the emerging foundation model fine-tuning using various urban data. However, integrating the VCS AI task with the conventional order serving task is challenging, due to their inconsistent spatio-temporal characteristics: (i) The distributions of ride orders and data point-of-interests (PoIs) may not coincide in geography, both following a priori unknown patterns; (ii) they have distinct forms of temporal effects, i.e., prolonged waiting makes orders become instantly invalid while data with increased staleness gradually reduces its utility for model fine-tuning.To overcome these obstacles, we propose an online framework based on multi-agent reinforcement learning (MARL) with careful augmentation. A new quality-of-service (QoS) metric is designed to characterize and balance the utility of the two joint tasks, under the effects of varying data volumes and staleness. We also integrate graph neural networks (GNNs) with MARL to enhance state representations, capturing graph-structured, time-varying dependencies among vehicles and across locations. Extensive experiments on our testbed simulator, utilizing various real-world foundation model fine-tuning tasks and the New York City Taxi ride order dataset, demonstrate the advantage of our proposed method.
... Road traffic injuries present themselves as modifiable causes of morbidity and mortality as well as sources of economic loss in the contemporary world (Abdelati, 2024;Gabr, Shoaeb, & El-Badawy, 2018;Nagy & Simon, 2018). Nevertheless, due to the intensification of innovative vehicle systems and infrastructural developments, the figures of causality and injuries still have not come up in accordance with the innovations, particularly in the congested and the conflict-laden urban areas (Berkowicz, Winther, & Ketzel, 2006;Mohan, Khayesi, Tiwari, & Nafukho, 2006). ...
Road traffic accidents continue to be a problem across the world and according to statistics cause high mortality and economic losses. This research work conceptualizes an idea that will use open traffic data and machine learning models to forecast accidents on roads in order to promote road safety. Based on the presented literature review, the framework incorporates a step-by-step procedure to analyze risk factors for targeted safety interventions, including data pre-processing and feature selection, application of a chosen model for high-risk zones identification, and improving the result by altering related factors. The findings show the applicability of open data and predictive analysis in traffic safety matters, with special emphasis on temporal, spatial, and environmental features. Resources allocation, urban traffic control, and monitoring are cases used to illustrate the framework's applicability. Although this is a conceptual model, the challenges, such as data quality, data privacy issues, and practical issues with implementation, are also included in the framework, along with suggestions for future research, such as the use of stream data and improved modeling techniques. This investigation contributes to the literature as a robust theoretical model from which practical solutions for road traffic safety interventions can be derived to reduce and ultimately eliminate traffic accidents and fatalities worldwide.
... Application Scenarios: Commonly used in continuous tracking scenarios such as vehicle positioning systems [3]. ...
This paper explores the application of multi-sensor fusion technology in in-vehicle navigation systems, focusing on improving positioning accuracy, reliability, and robustness in complex environments. By integrating data from various sensors such as GPS, INS, LiDAR, and cameras, multi-sensor fusion overcomes the limitations of individual sensors, such as signal blockage and cumulative errors. The paper reviews common fusion methods, including Kalman filters, particle filters, and deep learning techniques, and presents the design and implementation of a multi-sensor fusion system. Experimental results demonstrate significant improvements in navigation performance, especially in challenging environments like urban canyons and tunnels. However, challenges remain, including the need for precise sensor calibration, the quality of the sensors, and the computational complexity of real-time data fusion. Future research should focus on optimizing fusion algorithms, improving real-time performance through hardware acceleration, and reducing system costs. The integration of emerging technologies such as V2X (vehicle-to-everything) communication and machine learning could further enhance the system’s accuracy and reliability. These advancements will play a crucial role in the future of autonomous driving, smart transportation systems, and other related fields, pushing the development of intelligent navigation technologies forward.
... In recent years, sensor data have been widely used due to the advantages of high precision, high resolution, and strong self-adaptation. Its applications in the field of traffic safety include road condition monitoring [38,[45][46][47], road user behavior monitoring and prediction [48][49][50], traffic congestion monitoring [51,52], and accident prediction or detection [53,54]. ...
Targeted contingency measures have proven highly effective at reducing the duration and harm caused by incidents. This study utilized the Classification and Regression Trees (CART) data mining technique to predict and quantify the duration of incidents. To achieve this, multisensor data collected from the Hangzhou freeway in China spanning from 2019 to 2021 was utilized to construct a regression tree with eight levels and 14 leaf nodes. By extracting 14 rules from the tree and establishing contingency measures based on these rules, accurate incident assessment and effective implementation of post-incident emergency plans were achieved. In addition, to more accurately apply the research findings to actual incidents, the CART method was compared with XGBoost, Random Forest (RF), and AFT (accelerated failure time) models. The results indicated that the prediction accuracy of the CART model is better than the other three models. Furthermore, the CART method has strong interpretability. Interactions between explanatory variables, up to seven, are captured in the CART method, rather than merely analyzing the effect of individual variables on the incident duration, aligning more closely with actual incidents. This study has important practical implications for advancing the engineering application of machine learning methods and the analysis of sensor data.
... Smart city machinery permits city administrators to relate directly with society substructure, track what is going on within the city and how cities are changing. ICT helps to improve the standards, effectiveness and interactions of urbanised services, minimise expenditure and resource usage, and boost communication between individuals and government (Nagy & Simon, 2018). ...
Smart cities are potent instruments to mitigate climate change and foster urban sustainability. This is because smart cities exploit the benefits of advanced technologies and “Internet of Things” (IoT), which makes varieties of information available on issues, thereby easing strategies for tackling them. It therefore, becomes possible to make informed and precise choices that enact change to sustainably enhance survival. The aim of this article is to provide a conceptual evaluation on the implementation of smart cities in solving global climate change and urban sustainability concerns. This paper concludes that, even though smart cities can take major local intervention on climate change and urban sustainability, they will continue to depend on domestic governments to invest in critical infrastructure including defence, power, water supply, communications, and fast transit. Smart cities also need the private sector, through transparent and responsible taxation, to improve their attractiveness, generate wealth and jobs, and boost municipal finance resources.
... From the viewpoint of a transportation expert, it is evident that road traffic exerts a substantial impact on a multitude of facets in our everyday existence, while also playing a pivotal role in driving economic advancement in our modern society. This impact spans from individual commutes to the broader expanse of the transportation industry [1]. This reiterates the paramount significance of accurate traffic volume prediction across diverse domains. ...
A two‐step framework that integrates real‐time data collection with time series forecasting models for predicting traffic volume is proposed. In the first step, the framework utilizes live highway surveillance video data and YOLO‐v7 object detector to construct accurate traffic volume data. In the second step, an ARIMA–LSTM time series model is applied to forecast future traffic volumes. Experimental results show that YOLO‐v7 achieved a vehicle detection accuracy of over 93.30%, ensuring high precision in traffic volume data construction. The ARIMA–LSTM model demonstrated superior performance in traffic volume prediction, with a mean squared error of 87.97, root mean squared error of 10,388.57, and mean absolute error of 101.39. YOLO‐v7's detection speed of 7.8 ms per frame further validates the feasibility of real‐time data construction. The findings indicate that the combination of YOLO‐v7 for vehicle detection and ARIMA–LSTM for traffic prediction is highly effective, offering a significant reduction in training time compared to more complex deep learning models while maintaining high prediction accuracy. This research presents a unified solution for traffic data collection and prediction, enhancing transportation infrastructure planning and optimizing traffic flow. Future work will focus on extending the prediction intervals and further refining the models to improve performance.
... For instance, there are various reasons why traffic speeds can vary. The weather, the day of the week, the time of day, and unplanned events are just a few of the many factors that affect traffic flow, according to [47]. Thus, it is crucial to include external environmental information to reduce prediction errors. ...
Electrical energy is fundamental for contemporary society since failures directly impact other critical infrastructures such as water and gas distribution, hospitals, or banking services. Consequently, resilience, which is the capability of a system to handle high-impact low probability events, is a crucial aspect of such systems. Efficient resilience assessment methods are essential to achieving high-performance, resilient energy systems. This chapter introduces a multilayer method to address several factors of power distribution systems’ resilience. Reliability regressions model the failures’ instant and duration given a weather scenario, a dynamic Bayesian network (DBN) models how probabilities of failure propagate on the system’s physical connections, and a service restoration through switching operations, and field crew routing is obtained through an optimization algorithm for a given set of failures. Information related to these factors has the potential to be structured in a layered manner for a better understanding of the dynamic interaction among different information like weather, routes, power grid, and historical events logs. The ability to model these relationships enables the inference of the system resilience for different inputs during analysis. Resilience can also be inferred by considering the uncertainties associated with these layers due to DBN’s nature. A case study is presented to show the efficacy of this procedure. The findings showed its ability to evaluate the resilience of power distribution systems in the face of uncertainty and the considered aspects for different weather scenarios.
... However, traffic forecasting faces a series of challenges, especially in leveraging historical traffic flow data to uncover the spatiotemporal dependencies among network nodes. This has led to the continuous development of traffic forecasting research 4,5 . ...
Urban traffic flow prediction is an essential task within intelligent transportation systems, and numerous methodologies have been proposed to address it. However, most existing approaches focus on historical traffic information in terms of spatial and temporal aggregation, neglecting the implied spectral analysis of traffic time series. In this paper, we introduce the traffic flow in the frequency domain and, in conjunction with attention mechanisms, comprehensively learn the hidden correlations between spatial, temporal, and frequential. By deeply learning the spatial graph topological correlations of traffic flow, and using spectral analysis, fusing time series and implied periodic correlations in the temporal and frequential, we have constructed an innovative traffic prediction network model known as the Spatial Temproal-Frequential Attention Network (STFAN). The core of this network is the application of attention mechanisms to project the hidden states of traffic features in the current spatial, temporal, and frequential onto future hidden states, thereby comprehensively learning the hidden relationships of each dimension to future states and achieving the prediction of future traffic flow. To validate the performance of the proposed model, experiments were conducted on two public datasets from the California Department of Transportation (PeMS04 and PeMS08). The results demonstrate that the proposed model outperforms existing baseline models in terms of predictive accuracy, especially in medium and long-term traffic flow forecasting. Additionally, the ablation study confirmed the influence of frequency domain characteristics of traffic flow on future traffic states, thus proving the theoretical and practical effectiveness of the model.
... Traditional traffic forecasting methods are usually based on statistical models and time series analysis techniques. Common methods include autoregressive moving average model (ARIMA), exponential smoothing, and gray system model [7]. These methods are based on the trend and periodicity of historical data, and mathematical or statistical models are used to predict future traffic conditions. ...
With the process of urbanization and the rapid growth of population, the issue of traffic congestion has become an increasingly critical concern. Intelligent transportation systems heavily rely on real-time and precise prediction algorithms to address this problem. While Recurrent Neural Network (RNN) and Graph Convolutional Network (GCN) methods in deep learning have demonstrated high accuracy in predicting road conditions when sufficient data is available, forecasting in road networks with limited data remains a challenging task. This study proposed a novel Spatial-temporal Convolutional Network (TL-GPSTGN) based on graph pruning and transfer learning framework to tackle this issue. Firstly, the essential structure and information of the graph are extracted by analyzing the correlation and information entropy of the road network structure and feature data. By utilizing graph pruning techniques, the adjacency matrix of the graph and the input feature data are processed, resulting in a significant improvement in the model's migration performance. Subsequently, the well-characterized data are inputted into the spatial-temporal graph convolutional network to capture the spatial-temporal relationships and make predictions regarding the road conditions. Furthermore, this study conducts comprehensive testing and validation of the TL-GPSTGN method on real datasets, comparing its prediction performance against other commonly used models under identical conditions. The results demonstrate the exceptional predictive accuracy of TL-GPSTGN on a single dataset, as well as its robust migration performance across different datasets.
... The traditional machine learning methods have good robust generalization capabilities for modeling the traffic network but still have difficulty in generalizing non-linear relationships [16], [17], [18], [19], [20]. The deep learning methods are the most widely used methods for traffic prediction today [21], [1], which achieve quite good prediction performance with complex model structures. There are diverse neural network architectures that can be used to model traffic data, including the recurrent neural network (RNN) [22], the convolutional neural network (CNN) [23], the graph neural network (GNN) [24], [5], [3], [25], [26], [27], and the attention-based neural network [28], [29], [30]. ...
With the increasing traffic congestion problems in metropolises, traffic prediction plays an essential role in intelligent traffic systems. Notably, various deep learning models, especially graph neural networks (GNNs), achieve state-of-the-art performance in traffic prediction tasks but still lack interpretability. To interpret the critical information abstracted by traffic prediction models, we proposed a flexible framework termed Traffexplainer towards GNN-based interpretable traffic prediction. Traffexplainer is applicable to a wide range of GNNs without making any modifications to the original model structure. The framework consists of the GNN-based traffic prediction model and the perturbation-based hierarchical interpretation generator. Specifically, the hierarchical spatial mask and temporal mask are introduced to perturb the prediction model by modulating the values of input data. Then the prediction losses are backward propagated to the masks, which can identify the most critical features for traffic prediction, and further improve the prediction performance. We deploy the framework with five representative GNN-based traffic prediction models and analyse their prediction and interpretation performance on three real-world traffic flow datasets. The experiment results demonstrate that our framework can generate effective and faithful interpretations for GNN-based traffic prediction models, and also improve the prediction performance. The code will be publicly available at
https://github.com/lingbai-kong/Traffexplainer
.
... Most previous works have focused on increasing accuracy, time efficiency, and usability of prediction outcomes. Earlier approaches focused on building regression models to fit traffic flows, while deep learning-based techniques, as the most representative approaches in recent years, have strong forecasting performance on traffic flows [3]. ...
Recent achievements in deep learning (DL) have demonstrated its potential in predicting traffic flows. Such predictions are beneficial for understanding the situation and making traffic control decisions. However, most state-of-the-art DL models are considered "black boxes" with little to no transparency of the underlying mechanisms for end users. Some previous studies attempted to "open the black box" and increase the interpretability of generated predictions. However, handling complex models on large-scale spatiotemporal data and discovering salient spatial and temporal patterns that significantly influence traffic flow remain challenging.~To overcome these challenges, we present TrafPS, a visual analytics approach for interpreting traffic prediction outcomes to support decision-making in traffic management and urban planning. The measurements region SHAP and trajectory SHAP are proposed to quantify the impact of flow patterns on urban traffic at different levels. Based on the task requirements from domain experts, we employed an interactive visual interface for the multi-aspect exploration and analysis of significant flow patterns. Two real-world case studies demonstrate the effectiveness of TrafPS in identifying key routes and providing decision-making support for urban planning.
... In many papers, VR is only used for simulating road traffic. All aspects of ML analyses, such as predicting or analysing historical data, are often performed using different systems, running in parallel and external to VR [22]. For example, Wang et al., [23] use a separate ML system parallel to the VR system. ...
The main contribution of this paper is to introduce a framework for integrating Machine Learning (ML), Human, and Virtual Reality (VR) into one platform to promote a collaborative visualisation environment that can assist in better analysis and improve the human-machine teaming capability. This platform was demonstrated using a case study in ana-lysing road traffic conditions. The ‘Ab-normal Machine Learning Road Traffic Detection in VR (AbnMLRTD-VR)’ prototype system was developed to assist the human analyst. The proposed system has two main integrative components: a data-driven ML model and a 3D real-time visualisation in a VR environment. An unsupervised ML model was built using real traffic data. The AbnMLRTD-VR system highlights the outliers in the road sections in actual road contexts of a road traffic network. This gives the human analyst a 3D real-time immersive visualisation in a VR environment to evaluate road conditions. The AbnMLRTD-VR system demonstrated that it could help minimise the need for human pre-labelling of the data. It enables the visualisation of the road traffic conditions more meaningfully and to understand the context of the road traffic conditions of road sections at any given time.
... Traffic flow prediction is a core component of an Intelligent Transportation System (ITS) [1]. An ITS integrates people, vehicles, and roads into a comprehensive consideration so that they can work closely together to achieve a synergistic effect [2]. Traffic flow prediction is a rather complex process with multiperiod features, which is affected by a variety of factors such as traffic patterns, abnormal events, bad weather, and data collection [3]. ...
Traffic flow prediction is one of the challenges in the development of an Intelligent Transportation System (ITS). Accurate traffic flow prediction helps to alleviate urban traffic congestion and improve urban traffic efficiency, which is crucial for promoting the synergistic development of smart transportation and smart cities. With the development of deep learning, many deep neural networks have been proposed to address this problem. However, due to the complexity of traffic maps and external factors, such as sports events, these models cannot perform well in long-term prediction. In order to enhance the accuracy and robustness of the model on long-term time series prediction, a Graph Attention Informer (GAT-Informer) structure is proposed by combining the graph attention layer and informer layer to capture the intrinsic features and external factors in spatial–temporal correlation. The external factors are represented as sports events impact factors. The GAT-Informer model was tested on real-world data collected in London, and the experimental results showed that our model has better performance in long-term traffic flow prediction compared to other baseline models.
... Furthermore, both older ( [21,22]) and recent ( [23,24]) research works described and investigated a variety of alternate methods to traffic congestion study. For example, as revealed in studies [25] and [26], the primary criteria used to predict traffic congestion levels in various locations around the globe include time, speed, size, quality of service, and the traffic signal duration that vehicles must stop for, while [27] differentiates the data into spatiotemporal sequence and external data. ...
Traffic congestion in major cities is becoming increasingly severe. Numerous academic and commercial initiatives were conducted over the past decades to address this challenge, often delivering accurate and timely traffic condition predictions. Furthermore, traffic congestion forecasting has recently become, more than ever, an expanding study field, particularly due to growth of machine learning and artificial intelligence. This paper proposes a low-cost methodology to predict and fill current traffic congestion values for road parts having insufficient or missing historical data for timestamps with missing information, while reviewing several machine learning algorithms to select the most suitable ones. The methodology was evaluated on several open source data originated from one of the most challenging, regarding traffic, streets in Thessaloniki, the second largest city in Greece and was further validated over a second time period. Through experimentation with various cases, result comparison indicated that utilizing road segments proximate to those with missing data, in conjunction with a Multi-layer Perceptron, facilitates the accurate filling of missing values. Result evaluation revealed that dealing with data imbalance issues and importing weather features increased algorithmic accuracy for almost all classifiers, with Multi-layer Perceptron being the most accurate one.
... Machine learning methods have found widespread application in traffic prediction due to their proficiency in handling nonlinear correlations within traffic flow data (Nagy and Simon 2018). However, their effectiveness is constrained when extracting complex spatiotemporal features from real-world traffic behaviors (Zantalis et al. 2019). ...
Understanding commuter traffic in transportation networks is crucial for sustainable urban planning with commuting generation forecasts operating as a pivotal stage in commuter traffic modeling. Overcoming challenges posed by the intricacy of commuting networks and the uncertainty of commuter behaviors, we propose MetroGCN, a metropolis-informed graph convolutional network designed for commuting forecasts in metropolitan areas. MetroGCN introduces dimensions of metropolitan indicators to comprehensively construct commuting networks with diverse socioeconomic features. This model also innovatively embeds topological commuter portraits in spatial interaction through a multi-graph representation approach capturing the semantic spatial correlations based on individual characteristics. By incorporating graph convolution and temporal convolution with a spatial–temporal attention module, MetroGCN adeptly handles high-dimensional dependencies in large commuting networks. Quantitative experiments on the Shenzhen metropolitan area datasets validate the superior performance of MetroGCN compared to state-of-the-art methods. Notably, the results highlight the significance of commuter age and income in forecasting commuting generations. Statistical significance analysis further underscores the importance of anthropic indicators for commuting production forecasts and environmental indicators for commuting attraction forecasts. This research contributes to technical advancement and valuable insights into the critical factors influencing commuting generation forecasts.
... From personalizing content recommendations on entertainment platforms (Lapan 2018) to enabling predictive diagnostics in healthcare (Miotto et al. 2018), and optimizing traffic flow in urban planning (Nagy and Simon 2018), machine learning (ML) has transformed industries and our daily experiences. Despite these advances, the practical implementation of an ML-driven solution in a real-world scenario is fraught with challenges arising from the complexities of dynamic environments, such as changing data distributions, evolving requirements, and the need to constantly update and optimize models to maintain performance and accuracy. ...
Context
In machine learning (ML) applications, assets include not only the ML models themselves, but also the datasets, algorithms, and deployment tools that are essential in the development, training, and implementation of these models. Efficient management of ML assets is critical to ensure optimal resource utilization, consistent model performance, and a streamlined ML development lifecycle. This practice contributes to faster iterations, adaptability, reduced time from model development to deployment, and the delivery of reliable and timely outputs.
Objective
Despite research on ML asset management, there is still a significant knowledge gap on operational challenges, such as model versioning, data traceability, and collaboration issues, faced by asset management tool users. These challenges are crucial because they could directly impact the efficiency, reproducibility, and overall success of machine learning projects. Our study aims to bridge this empirical gap by analyzing user experience, feedback, and needs from Q &A posts, shedding light on the real-world challenges they face and the solutions they have found.
Method
We examine 15, 065 Q &A posts from multiple developer discussion platforms, including Stack Overflow, tool-specific forums, and GitHub/GitLab. Using a mixed-method approach, we classify the posts into knowledge inquiries and problem inquiries. We then apply BERTopic to extract challenge topics and compare their prevalence. Finally, we use the open card sorting approach to summarize solutions from solved inquiries, then cluster them with BERTopic, and analyze the relationship between challenges and solutions.
Results
We identify 133 distinct topics in ML asset management-related inquiries, grouped into 16 macro-topics, with software environment and dependency, model deployment and service, and model creation and training emerging as the most discussed. Additionally, we identify 79 distinct solution topics, classified under 18 macro-topics, with software environment and dependency, feature and component development, and file and directory management as the most proposed.
Conclusions
This study highlights critical areas within ML asset management that need further exploration, particularly around prevalent macro-topics identified as pain points for ML practitioners, emphasizing the need for collaborative efforts between academia, industry, and the broader research community.
... The most recent and popular AI domain is ML, whose popularity can be justified by its ability to autonomously learn patterns in individuals, businesses, processes, transactions and events without explicit programming (Sarker, 2021). Currently ML is seamlessly integrated into various local government sectors, including traffic and parking predictions (Awan et al., 2020;Dogru & Subasi, 2018;Nagy & Simon, 2018), municipal waste consumption (Dubey et al., 2020), yield and crop estimations (Pott et al., 2021;Schwalbert et al., 2020), smart energy systems for sustainable cities (O'Dwyer et al., 2019;Pieroni et al., 2018), and urban green infrastructure (Shaamala et al., 2024;Wu et al., 2022;Yang et al., 2021). For instance, local governments are using realtime data leveraging ML algorithms to create prediction tools for parking occupancy rates (Zheng et al., 2015). ...
In recent years, the rapid advancement of artificial intelligence (AI) technologies has significantly impacted various sectors, including public governance at the local level. However, there exists a limited understanding of the overarching narrative surrounding the adoption of AI in local governments and its future. Therefore, this study aims to provide a comprehensive overview of the evolution, current state-of-the-art, and emerging trends in the adoption of AI in local government. A comprehensive scientometric analysis was conducted on a dataset comprising 7112 relevant literature records retrieved from the Scopus database in October 2023, spanning over the last five decades. The study findings revealed the following key insights: (a) exponential technological advancements over the last decades ushered in an era of AI adoption by local governments; (b) the primary purposes of AI adoption in local governments include decision support, automation, prediction, and service delivery; (c) the main areas of AI adoption in local governments encompass planning, analytics, security, surveillance, energy, and modelling; and (d) under-researched but critical research areas include ethics of and public participation in AI adoption in local governments. This study informs research, policy, and practice by offering a comprehensive understanding of the literature on AI applications in local governments, providing valuable insights for stakeholders and decision-makers.
... Mobility data analytics encompasses various application domains and involves analyzing data from diferent ś sources such as urban [267], maritime [61], aviation [59], animal movement [172], and indoor movement [115]. Among these diferent themes urban mobility stands out with a fairly large body of research including green routing [10], traic anomaly detection [174], hot spot and hot path analysis [167], road traic prediction [162], and travel time estimation [242]. Trajectories of moving objects have been used as means to create and continuously update the road network [160]. ...
Mobility data captures the locations of moving objects such as humans, animals, and cars. With the availability of GPS-equipped mobile devices and other inexpensive location-tracking technologies, mobility data is collected ubiquitously. In recent years, the use of mobility data has demonstrated significant impact in various domains including traffic management, urban planning, and health sciences. In this paper, we present the domain of mobility data science. Towards a unified approach to mobility data science, we present a pipeline having the following components: mobility data collection, cleaning, analysis, management, and privacy. For each of these components, we explain how mobility data science differs from general data science, we survey the current state of the art, and describe open challenges for the research community in the coming years.
... A survey was conducted to explore behavior, Intelligent Transportation System needs, level of service expectations for student parking at a university campus [5]. A detailed presentation of the traffic prediction methods for intelligent cities has been provided, also overviewed the existing data sources and prediction models [9]. The distribution of standstill distance and time headway was investigated and incorporated this distribution into freeway car-following models and an application to estimating freeway travel time reliability [8]. ...
Poor socioeconomic conditions and environmental needs, inadequate and inefficient public transport infrastructure, heterogeneous traffic, poor lane behavior and traffic conditions, increasing transport vehicles and migration are everlasting problems in Delhi, the second-highest populated metropolitan city of India. This leads to deterioration in air quality, increase in noise levels and severe traffic congestion and stop-and-go traffic, which ultimately results in economic, social and environmental losses. In this paper, a Synergistic Metropolitan Transport-System Dynamics Simulation (SMT-SDS) model is designed and developed to evaluate the advantages of implementing an Intelligent Transportation System (ITS) in Delhi. Intelligent Transportation System refers to the interconnection of an adaptive and intelligent integration of vehicles, drivers, and the transport system. On the basis of simulation and analysis, the SMT-SDS model generates the scenarios and policy recommendations, to predict the impact of implementing Intelligent Transportation System (ITS) strategies at a war effort to reduce the traffic congestion and stop-and-go traffic and improve the existing transportation system of Delhi. Scenario generation will help in interpreting the futuristic conditions by analysing the simulated projections. The implementation of ITS strategies in Delhi will have a positive outcome on traffic and transport conditions of Delhi and will improve the saturation of Delhi roads/highways drastically.
Automation and connectivity have increased as autonomous vehicle technology has advanced. CAN is one of numerous serial protocols used in a variety of cars. Contemporary automobiles are increasingly susceptible to attacks targeting automotive networks due to their increased functionality and connectivity. The CAN bus protocol is not secure by design, it is open to many assaults. To identify attacks on the CAN bus, system for intrusion detection (IDS) must be highly accurate in their design. The deep learning based model can achieve higher accuracy in detection and classification of network messages. In this research, we provide an efficient IDS strategy for binary classification based on deep learning. The best model is determined using the keras tuner library and the implementation results suggest that proposed strategy achieves the accuracy of 99% in classification of messages in CAN bus network. The results outperform the classification accuracy with respect to the supervised machine learning based approaches.
Due to advancements in computing systems, intelligent transportation networks, and communication technologies, the development of VANETs is possible. Traffic prediction in high mobility and dynamic network topologies is highly challenging in VANETs. It is very important for high accuracy in traffic prediction as it reduces congestion, provides smooth flow of traffic, and ensures road safety. This chapter is going to discuss how deep learning technologies can be used in solving the problems above. It gives importance to different methodologies, architectures, and practical implementations. Deep learning is one of the subsets of machine learning that has gained much popularity due to its excellent ability to predict complex relationships and patterns in large datasets. It checks applicability to the problem of traffic-patterns forecasting and solving time series problems using CNN, RNN, and LSTM. This chapter refers to strategies on data preparation-data cleaning, normalization, and augmentation. Data preparation with its key demands-normalization, augmentation makes the input data correct for the deep learning model application; thus, correctness also relies on relevance. A list of relevant topics incorporates federated learning and transfer learning. Even though it doesn’t solve the problem of sparse data, transfer learning helps with the adaptation and adoption of models developed from big sets of data to new and related settings. In federated learning, instead, diversity within centralised collections of data is increased because these models decentralised work to perform computations but differ by their levels of privacy. This comprehensive review of deep learning algorithms for traffic prediction in VANETs can provide much-needed insight to academics, practitioners, and stakeholders interested in intelligent transportation systems.
Emerging technologies such as Artificial Intelligence (AI) and the Internet of Things (IoT) have transformed intelligent transportation systems, providing novel solutions to the increasing complexity of managing traffic on roads as cities grow and traffic density rises, particularly in developing countries. Smart road traffic management systems seek to alleviate traffic-related issues, benefiting citizens and society. This study reports on a Preferred Reporting Item for Systematic Reviews and Meta-Analyses (PRISMA)-based systematic literature review (SLR) of 75 papers published between 2014 and 2023. Like prior reviews, this SLR focuses on recent advances in the Internet of Things and Artificial Intelligence for traffic management, covering crucial practical issues such as scalability, data privacy, and resource constraints in growing regions. The study covers significant topics thoroughly, including AI approaches such as machine learning and deep learning, the incorporation of the Internet of Things, and artificial intelligence in traffic management, including the evaluation methods utilized in the examined studies. By synthesizing insights from various studies, recognizing research gaps, and proposing recommendations for future research, this work provides a comprehensive understanding of how recent advances in smart road traffic monitoring will lead to more effective models that improve urban mobility and benefit the community.
Traffic data quality and secure computation are critical for intelligent transportation systems (ITS). To address low-quality traffic data, we propose STAP, a spatio-temporal correlation model using an improved Random Forest for anomaly detection and XGBoost for data estimation, effectively enhancing data quality as validated by experiments on real-world data from Changsha, China. For secure outsourcing computing in vehicular fog environments, we introduce SE-VFC, which combines lightweight BLS and group signatures for anonymous batch authentication and privacy protection of fog vehicles, ensuring correctness and traceability of computations while maintaining low overhead. Compared to traditional anomaly detection methods, STAP improves average accuracy by 5%, while SE-VFC proves effective and practical in vehicular fog computing, offering low communication and computation overhead.
Intelligent transportation systems heavily rely on forecasting urban traffic flow, and a variety of approaches have been developed for this purpose. However, most current methods focus on exploring spatial and temporal dependencies in historical traffic data, while often overlooking the inherent spectral characteristics hidden in traffic time series. In this paper, we introduce an approach to analyzing traffic flow in the frequency domain. By integrating attention mechanisms, we comprehensively capture the hidden correlations among space, time, and frequency dimensions. By leveraging deep learning to capture spatial correlations in traffic flow and applying spectral analysis to fuse time series data with underlying periodic correlations in both the time and frequency domains, we develop an innovative traffic prediction model called the Space-Time-Frequency Attention Network (STFAN). The core of this network lies in the application of attention mechanisms, which project the hidden states of current traffic features across the space, time, and frequency domains onto future hidden states. This approach enables a comprehensive learning of the relationships between each dimension and the future states, ultimately allowing for accurate predictions of future traffic flow. We carry out experiments on two publicly available datasets from the California Department of Transportation, PeMS04 and PeMS08, to assess the performance of the proposed model. The results demonstrate that the proposed model outperforms existing baseline models in terms of predictive accuracy, particularly for mid- and long-term traffic flow forecasting. Finally, the ablation study confirmed that the frequency domain characteristics of traffic flow significantly influence future traffic conditions, demonstrating the practical effectiveness of the model.
From the perspective of telecommunications, next-generation networks or beyond 5G will inevitably face the challenge of a growing number of users and devices. Such growth results in high-traffic generation with limited network resources. Thus, the analysis of the traffic and the precise forecast of user demands is essential for developing an intelligent network. In this line, Machine Learning (ML) and especially Deep Learning (DL) models can further benefit from the huge amount of network data. They can act in the background to analyze and predict traffic conditions more accurately than ever, and help to optimize the design and management of network services. Recently, a significant amount of research effort has been devoted to this area, greatly advancing network traffic prediction (NTP) abilities. In this paper, we bring together NTP and DL-based models and present recent advances in DL for NTP. We provide a detailed explanation of popular approaches and categorize the literature based on these approaches. Moreover, as a technical study, we conduct different data analyses and experiments with several DL-based models for traffic prediction. Finally, discussions regarding the challenges and future directions are provided.
Urban traffic control is a multifaceted and demanding task that necessitates extensive decision-making to ensure the safety and efficiency of urban transportation systems. Traditional approaches require traffic signal professionals to manually intervene on traffic control devices at the intersection level, utilizing their knowledge and expertise. However, this process is cumbersome, labor-intensive, and cannot be applied on a large network scale. Recent studies have begun to explore the applicability of recommendation system for urban traffic control, which offer increased control efficiency and scalability. Such a decision recommendation system is complex, with various interdependent components, but a systematic literature review has not yet been conducted. In this work, we present an up-to-date survey that elucidates all the detailed components of a recommendation system for urban traffic control, demonstrates the utility and efficacy of such a system in the real world using data and knowledge-driven approaches, and discusses the current challenges and potential future directions of this field.
Forecasting counts data in transportation areas can enrich passenger information for public transport passengers, who may thus better plan their trips. Moreover, forecasting with uncertainty is particularly important in the transportation domain, where the risk of poorly managed high demand is to be avoided. In this paper, we propose a new probabilistic prediction model well-suited for multivariate, overdispersed, and possibly correlated count data. This model combines the strength of the deep learning framework with the modeling of counts data allowed by “sums and shares” distributions. Indeed, deep learning models can handle uncertainty by relying on an abstraction of contextual data and by assuming output distributions. Our model learns a latent representation of the input data with the help of a recurrent neural network and then translates it into multivariate count predictions with a sums and shares distribution, well suited to tackle multivariate overdispersed and correlated count data. An extensive benchmark of the proposed model is carried out. We compare this model with seven others from the state-of-the-art probabilistic forecasting models using five open-source data (bikes, taxis, railways, traffic, wikipedia) and a specific use case on pedestrian counts within a multimodal transport hub in the Paris Region. Our model outperforms other models in situations where the data present temporal regularities. The results also highlight the potential of our model in the specific use case. Moreover, this forecasting represents an interesting way to predict short-term pedestrian counts in response to different events, such as concerts or transport disruptions.
Traffic flow prediction is a challenging spatio-temporal prediction task due to its spatio-temporal dynamics and uncertainty. In recent years, Graph Convolutional Neural Networks (GCNs) have been applied to traffic flow prediction due to their exceptional capabilities in processing network information. However, current GCN-based methods fail to capture long-range spatial dependencies between regions with similar functionalities, and the modeling of spatial dependencies within adjacent time steps is neglected. To solve these problems, this work proposes a new Dynamic Graph Convolutional Recurrent Network (CDGCRN) with spatiotemporal category information embedding. An embedding layer with spatiotemporal category information is introduced to enhance the model’s capability in capturing long-range spatial dependencies. This layer aids the model in rapidly identifying traffic patterns with distinct features. Morever, Dynamic spatial information fusion graph convolutional recurrent unit is proposed to capture spatial dependencies within all time steps and adjacent time steps. It incorporates a learnable graph structure with a time window for capturing spatial dependency relationships within adjacent time steps and a globally learnable graph structure for learning spatial dependency relationships across all time steps. Experimental results on four real-world datasets demonstrate that CDGCRN is not only computationally efficient but also outperforms 17 state-of-the-art baseline methods.
Graph-based deep learning models are becoming prevalent for data-driven traffic prediction in the past years, due to their competence in exploiting the non-Euclidean spatial-temporal traffic data. Nonetheless, these models are approaching a limit where drastically increasing model complexity in terms of trainable parameters cannot notably improve the prediction accuracy. Furthermore, the diversity of transportation networks requires traffic predictors to be scalable to various data sizes and quantities, and ever-changing traffic dynamics also call for capacity sustainability. To this end, we propose a novel adaptive deep learning scheme for boosting graph-based traffic predictor performance. The proposed scheme utilizes domain knowledge to decompose the traffic prediction task into sub-tasks, each of which is handled by deep models with low complexity and training difficulty. Further, a stream learning algorithm based on the empirical Fisher information loss is devised to enable predictors to incrementally learn from new data without re-training from scratch. Comprehensive case studies on five real-world traffic datasets indicate outstanding performance improvement of the proposed scheme when equipped to six state-of-the-art predictors. Additionally, the scheme also provides impressive autoregressive long-term predictions and incremental learning efficacy with traffic data streams.
Traffic-state forecasting is crucial for traffic management and control strategies, as well as user- and system-level decision-making in the transportation network. While traffic forecasting has been approached with a variety of techniques over the last couple of decades, most approaches simply rely on endogenous traffic variables for state prediction, despite the evidence that exogenous factors can significantly affect traffic conditions. This paper proposes a multidimensional spatiotemporal graph attention-based traffic-prediction approach (M-STGAT), which predicts traffic based on past observations of speed, along with lane-closure events, temperature, and visibility across a large transportation network. The approach is based on a graph attention network architecture, which learns based on the structure of the transportation network on which these variables are observed. Numerical experiments are performed using traffic-speed and lane-closure data from the Caltrans Performance Measurement System (PeMS) and corresponding weather data from the National Oceanic and Atmospheric Administration (NOOA) Automated Surface Observing Systems (ASOS). The numerical experiments implement three alternative models which do not allow for multidimensional input, along with two alternative multidimensional models, based on the literature. The M-STGAT outperforms the five alternative models in validation and testing with the primary data set, as well as for one transfer data set across all three prediction horizons for all error measures. However, the model’s transferability varies for the remaining two transfer data sets, which may require further investigation. The results demonstrate that M-STGAT has the most consistently low error values across all transfer data sets and prediction horizons.
This paper systematically reviews studies that forecast short-term traffic conditions using spatial dependence between links. We extract and synthesise 130 research papers, considering two perspectives: (1) methodological framework and (2) methods for capturing spatial information. Spatial information boosts the accuracy of prediction, particularly in congested traffic regimes and for longer horizons. Machine learning methods, which have attracted more attention in recent years, outperform the naïve statistical methods such as historical average and exponential smoothing. However, there is no guarantee of superiority when machine learning methods are compared with advanced statistical methods such as spatiotemporal autoregressive integrated moving average. As for the spatial dependency detection, a large gulf exists between the realistic spatial dependence of traffic links on a real network and the studied networks as follows: (1) studies capture spatial dependency of either adjacent or distant upstream and downstream links with the study link, (2) the spatially relevant links are selected either by prejudgment or by correlation-coefficient analysis, and (3) studies develop forecasting methods in a corridor test sample, where all links are connected sequentially together, assume a similarity between the behaviour of both parallel and adjacent links, and overlook the competitive nature of traffic links.
Smart card data is increasingly used to investigate passenger behavior and the demand characteristics of public transport. The destination estimation of public transport is one of the major concerns for the implementation of smart card data. In recent years, numerous studies concerning destination estimation have been carried out—most automatic fare collection (AFC) systems only record boarding information but not passenger alighting information. This study provides a comprehensive review of the practice of using smart card data for destination estimation. The results show that the land use factor is not discussed in more than three quarters of papers and sensitivity analysis is not applied in two thirds of papers. In addition, the results are not validated in half the relevant studies. In the future, more research should be done to improve the current model, such as considering additional factors or making sensitivity analysis of parameters as well as validating the results with multi-source data and new methods.
Spatiotemporal forecasting has significant implications in sustainability, transportation and health-care domain. Traffic forecasting is one canonical example of such learning task. This task is challenging due to (1) non-linear temporal dynamics with changing road conditions, (2) complex spatial dependencies on road networks topology and (3) inherent difficulty of long-term time series forecasting. To address these challenges, we propose Graph Convolutional Recurrent Neural Network to incorporate both spatial and temporal dependency in traffic flow. We further integrate the encoder-decoder framework and scheduled sampling to improve long-term forecasting. When evaluated on real-world road network traffic data, our approach can accurately capture spatiotemporal correlations and consistently outperforms state-of-the-art baselines by 12%- 15%.
Robust and accurate traffic prediction is critical in
modern intelligent transportation systems (ITS). One widely used
method for short-term traffic prediction is k-nearest neighbours
(kNN). However, choosing the right parameter values for kNN
is problematic. Although many studies have investigated this
problem, they did not consider all parameters of kNN at the
same time. This paper aims to improve kNN prediction accuracy
by tuning all parameters simultaneously concerning dynamic
traffic characteristics. We propose weighted parameter tuples
(WPT) to calculate weighted average dynamically according to
flow rate. Comprehensive experiments are conducted on one-year
real-world data. The results show that flow-aware WPT kNN
performs better than manually tuned kNN as well as benchmark
methods such as extreme gradient boosting (XGB) and seasonal
autoregressive integrated moving average (SARIMA). Thus, it is
recommended to use dynamic parameters regarding traffic flow
and to consider all parameters at the same time.
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.
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.
In big-data-driven traffic flow prediction systems, the robustness of prediction performance depends on accuracy and timeliness. This paper presents a new MapReduce-based nearest neighbor (NN) approach for
traffic flow prediction
using
correlation
analysis (TFPC) on a Hadoop platform. In particular, we develop a real-time prediction system including two key modules, i.e., offline distributed training (ODT) and online parallel prediction (OPP). Moreover, we build a parallel
k
-nearest neighbor optimization classifier, which incorporates correlation information among traffic flows into the classification process. Finally, we propose a novel prediction calculation method, combining the current data observed in OPP and the classification results obtained from large-scale historical data in ODT, to generate traffic flow prediction in real time. The empirical study on real-world traffic flow big data using the leave-one-out cross validation method shows that TFPC significantly outperforms four state-of-the-art prediction approaches, i.e., autoregressive integrated moving average, Naïve Bayes, multilayer perceptron neural networks, and NN regression, in terms of accuracy, which can be improved 90.07% in the best case, with an average mean absolute percent error of 5.53%. In addition, it displays excellent speedup, scaleup, and sizeup.
Methods and systems for traffic-based media selection are provided. A first media file may be played on a media player. Traffic around the media player is monitored, and various traffic conditions, including speed and density of the traffic, may be detected. A second media file is selected based on a detected traffic condition and provided to the media player. In some embodiments of the present invention, the traffic may occur in a virtual environment.
To discover regularities in human mobility is of fundamental importance to our understanding of urban dynamics, and essential to city and transport planning, urban management and policymaking. Previous research has revealed universal regularities at mainly aggregated spatio-temporal scales but when we zoom into finer scales, considerable heterogeneity and diversity is observed instead. The fundamental question we address in this paper is at what scales are the regularities we detect stable, explicable, and sustainable. This paper thus proposes a basic measure of variability to assess the stability of such regularities focusing mainly on changes over a range of temporal scales. We demonstrate this by comparing regularities in the urban mobility patterns in three world cities, namely London, Singapore and Beijing using one-week of smart-card data. The results show that variations in regularity scale as non-linear functions of the temporal resolution, which we measure over a scale from 1 minute to 24 hours thus reflecting the diurnal cycle of human mobility. A particularly dramatic increase in variability occurs up to the temporal scale of about 15 minutes in all three cities and this implies that limits exist when we look forward or backward with respect to making short-term predictions. The degree of regularity varies in fact from city to city with Beijing and Singapore showing higher regularity in comparison to London across all temporal scales. A detailed discussion is provided, which relates the analysis to various characteristics of the three cities. In summary, this work contributes to a deeper understanding of regularities in patterns of transit use from variations in volumes of travellers entering subway stations, it establishes a generic analytical framework for comparative studies using urban mobility data, and it provides key points for the management of variability by policy-makers intent on for making the travel experience more amenable.
Having access to the future traffic state information is crucial in maintaining successful intelligent transportation systems (ITS). However, predicting the future traffic state is a challenging research subject involving prediction reliability issues. Predictive performance measures, including the accuracy, efficiency, and stability, are generally considered as the most important priorities in the evaluation of prediction modules. Researchers have developed various K-nearest-neighbors-based searching algorithms that find the future state from the historical traffic patterns. Interestingly, there has not been sufficient effort made for improving the performance. For the emerging big data era, incorporating an efficient search strategy has become increasingly important since the applicability of the prediction module in ITS heavily relies on the efficiency of the searching method used. This paper develops a novel sequential search strategy for traffic state predictions. The proposed sequential strategy is found to be outperforming the conventional single-level search approach in terms of prediction measures, which are prediction accuracy, efficiency, and stability. Compared with the conventional approach, the proposed sequential method yields significantly more accurate results via internal hierarchical improvements across sublevels while maintaining excellent efficiency and stability.
The k-nearest neighbor (KNN) model is an effective statistical model applied in short-term traffic forecasting that can provide reliable data to guide travelers. This study proposes an improved KNN model to enhance forecasting accuracy based on spatiotemporal correlation and to achieve multistep forecasting. The physical distances among road segments are replaced with equivalent distances, which are defined by the static and dynamic data collected from real road networks. The traffic state of a road segment is described by a spatiotemporal state matrix instead of only a time series as in the original KNN model. The nearest neighbors are selected according to the Gaussian weighted Euclidean distance, which adjusts the influences of time and space factors on spatiotemporal state matrices. The forecasting accuracies of the improved KNN and of four other models are compared, and experimental results indicate that the improved KNN model is more appropriate for short-term traffic multistep forecasting than the other models are. This study also discusses the application of the improved KNN model in a time-varying traffic state.
The goal of the paper is to give an overview of the most relevant aspects of mobile crowdsensing that are already utilized by the society. The paper focuses on best practices applied in smart cities today, how these applications can be motivated (incentives), and how they rely on technology enablers of today’s vertical silos and future’s horizontal approaches. We introduce a path for transforming the vertical silos of today containing separated solutions in various domains into a horizontal, unified ecosystem, giving a way to novel technology and business opportunities.
The study of socio-technical systems has been revolutionized by the unprecedented amount of digital records that are constantly being produced by human activities such as accessing Internet services, using mobile devices, and consuming energy and knowledge. In this paper, we describe the richest open multi-source dataset ever released on two geographical areas. The dataset is composed of telecommunications, weather, news, social networks and electricity data from the city of Milan and the Province of Trentino. The unique multi-source composition of the dataset makes it an ideal testbed for methodologies and approaches aimed at tackling a wide range of problems including energy consumption, mobility planning, tourist and migrant flows, urban structures and interactions, event detection, urban well-being and many others.
Traffic flow prediction is a fundamental functionality of intelligent transportation systems. After presenting the state of the art, we focus on nearest neighbor regression methods, which are data-driven algorithms that are effective yet simple to implement. We try to strengthen their efficacy in two ways that are little explored in literature, i.e., by adopting a multivariate approach and by adding awareness of the time of the day. The combination of these two refinements, which represents a novelty, leads to the definition of a new class of methods that we call time-aware multivariate nearest neighbor regression (TaM-NNR) algorithms. To assess this class, we have used publicly available traffic data from a California highway. Computational results show the effectiveness of such algorithms in comparison with state-of-the-art parametric and non-parametric methods. In particular, they consistently perform better than their corresponding standard univariate versions. These facts highlight the importance of context elements in traffic prediction. The ideas presented here may be further investigated considering more context elements (e.g., weather conditions), more complex road topologies (e.g., urban networks), and different types of prediction methods.
Capturing highly disaggregate details about traffic dynamics, microscopic traffic simulation has long proved to be a valuable tool for the evaluation of development plans and operations/control strategies. However, the applications of microscopic models still face a number of methodological and practical challenges in large-scale networks. This paper develops a 24-hour large-scale microscopic traffic simulation model for north Washington, DC metropolitan area. The model consists of over 7,000 links, 3500 nodes, 400 signalized intersections, and over 40,000 origin-destination (OD) pairs. Careful calibration and validation have been applied in order for the robustness and reliability of the model. Based upon the calibrated model, a case study on a newly built toll road in Maryland has been conducted. Along with various network-level, corridor-level, and freeway-level performance measures, Environmental Protection Agency (EPA)’s Motor Vehicle Emission Simulator (MOVES) is linked with the model for the estimation of environmental impacts. The case study demonstrates the capability of the large-scale microscopic simulation in planning/policy applications. Several methodological and practical challenges for developing multi-period large-scale microscopic traffic simulation models have also been discussed in this paper.
Road traffic density has always been a concern in large cities around the world, and many approaches were developed to assist in solving congestions related to slow traffic flow. This work proposes a congestion rate estimation approach that relies on real-time video scenes of road traffic, and was implemented and evaluated on eight different hotspots covering 33 different urban roads. The approach relies on road scene morphology for estimation of vehicles average speed along with measuring the overall video scenes randomness acting as a frame texture analysis indicator. Experimental results shows the feasibility of the proposed approach in reliably estimating traffic density and in providing an early warning to drivers on road conditions, thereby mitigating the negative effect of slow traffic flow on their daily lives.
The D4D-Senegal challenge is an open innovation data challenge on anonymous
call patterns of Orange's mobile phone users in Senegal. The goal of the
challenge is to help address society development questions in novel ways by
contributing to the socio-economic development and well-being of the Senegalese
population. Participants to the challenge are given access to four mobile phone
datasets. This paper describes the three datasets. The datasets are based on
Call Detail Records (CDR) of phone calls and text exchanges between more than 9
million of Orange's customers in Senegal between January 1, 2013 to December
31, 2013. The datasets are: (1) antenna-to-antenna traffic for 1666 antennas on
an hourly basis, (2) fine-grained mobility data on a rolling 2-week basis for a
year with bandicoot behavioral indicators at individual level for about 300,000
randomly sampled users, (3) one year of coarse-grained mobility data at
arrondissement level with bandicoot behavioral indicators at individual level
for about 150,000 randomly sampled users
Mitigating traffic congestion on urban roads, with paramount importance in
urban development and reduction of energy consumption and air pollution,
depends on our ability to foresee road usage and traffic conditions pertaining
to the collective behavior of drivers, raising a significant question: to what
degree is road traffic predictable in urban areas? Here we rely on the precise
records of daily vehicle mobility based on GPS positioning device installed in
taxis to uncover the potential daily predictability of urban traffic patterns.
Using the mapping from the degree of congestion on roads into a time series of
symbols and measuring its entropy, we find a relatively high daily
predictability of traffic conditions despite the absence of any a priori
knowledge of drivers' origins and destinations and quite different travel
patterns between weekdays and weekends. Moreover, we find a counterintuitive
dependence of the predictability on travel speed: the road segment associated
with intermediate average travel speed is most difficult to be predicted. We
also explore the possibility of recovering the traffic condition of an
inaccessible segment from its adjacent segments with respect to limited
observability. The highly predictable traffic patterns in spite of the
heterogeneity of drivers' behaviors and the variability of their origins and
destinations enables development of accurate predictive models for eventually
devising practical strategies to mitigate urban road congestion.
This research proposes a short-term highway traffic state prediction method based on a structural state space model, with the intention to provide a robust approach for obtaining accurate forecasts of traffic state under both recurring and non-recurring conditions. True traffic state is decomposed to three components, namely, regular traffic pattern, structural deviation, and random fluctuation. Particularly, the structural deviation term reflects the change of true traffic state from regular (historical) pattern, due to unexpected capacity reduction and/or demand variations. A polynomial trend is adopted to describe the temporal evolution of structural deviations across different time intervals. We derive an analytical form of structural deviations in a single bottleneck case based on cumulative flow count diagrams. The proposed model is incorporated in a Kalman filtering-based algorithmic framework, together with an adaptive scheme for determining the variances of random errors. A set of numerical experiments was conducted on two test beds in the northern Taiwan highway network. Experimental results show that the proposed approach is particularly superior to an ordinary Kalman filtering method presented in the literature under non-recurring conditions, highlighting the advantage of combining both the polynomial trend model and historical pattern into the proposed short-term traffic state prediction approach.
Climate change is a prevalent issue facing the world today. Unexpected increase in rainfall intensity and events is one of the major signatures of climate change. Rainfall influences traffic conditions and, in turn, traffic volume in urban arterials. For improved traffic management under adverse weather conditions, it is important to develop a traffic prediction algorithm considering the effect of rainfall. This inclusion is not intuitive as the effect is not immediate, and the influence of rainfall on traffic volume is often unrecognizable in a direct correlation analysis between the two time-series data sets; it can only be observed at certain frequency levels. Accordingly, it is useful to employ a multiresolution prediction framework to develop a weather adaptive traffic forecasting algorithm. Discrete wavelet transform (DWT) is a well-known multiresolution data analysis methodology. However, DWT imparts time variance in the transformed signal and makes it unsuitable for further time-series analysis. Therefore, the stationary form of DWT known as stationary wavelet transform (SWT) has been used in this paper to develop a neurowavelet prediction algorithm to forecast hourly traffic flow considering the effect of rainfall. The proposed prediction algorithm has been evaluated at two urban arterial locations in Dublin, Ireland. This paper shows that the rainfall data successfully augments the traffic flow data as an exogenous variable in periods of inclement weather, resulting in accurate predictions of future traffic flow at the two chosen locations. The forecasts from the neurowavelet model outperform the forecasts from the standard artificial neural network (ANN) model.
The k-nearest neighbor (k-NN) nonparametric regression is a classic model for single point short-term traffic flow forecasting. The traffic flows of the same clock time of the days are viewed as neighbors to each other, and the neighbors with the most similar values are regarded as nearest neighbors and are used for the prediction. In this method, only the information of the neighbors is considered. However, it is observed that the “trends” in the traffic flows are useful for the prediction. Taking a sequence of consecutive time periods and viewing the a sequence of “increasing”, “equal” or “decreasing” of the traffic flows of two consecutive periods as a pattern, it is observed that the patterns can be used for prediction, despite the patterns are not from the same clock time period of the days. Based on this observation, a pattern recognition algorithm is proposed. Moreover, empirically, we find that the patterns from different clock time of the days can have different contributions to the prediction. For example, if both to predict the traffic flow in the morning, the pattern from the morning can lead to better prediction than same patterns from afternoon or evening. In one sentence, we argue that both the pattern and the clock time of the pattern contain useful information for the prediction and we propose the weighted pattern recognition algorithm (WPRA). We give different weights to the same patterns of different clock time for the prediction. In this way, we take both virtues of the k-NN method and the PRA method. We use the root mean square error (RMSE) between the actual traffic flows and the predicted traffic flows as the measurement. By applying the results to actual data and the simulated data, about 20% improvement compare with the PRA is obtained.
The Seasonal Autoregressive Integrated Moving Average (SARIMA) model is one of the popular univariate time-series models in the field of short-term traffic flow forecasting. The parameters of the SARIMA model are commonly estimated using classical (maximum likelihood estimate and/or least square estimate) methods. In this paper, instead of using classical inference the Bayesian method is employed to estimate the parameters of the SARIMA model considered for modelling. In Bayesian analysis, Markov chain Monte Carlo method is used to solve the posterior integration problem in high dimension. Each of the estimated parameters from the Bayesian method has a probability density function conditional to the observed traffic volumes. The forecasts from the Bayesian model can better match the traffic behavior of extreme peaks and rapid fluctuation. Similar to the estimated parameters, each forecast has a probability density curve with the maximum probable value as the point forecast. Individual probability density curves provide a time-varying prediction interval unlike the constant prediction interval from classical inference. The time-series data used for fitting the SARIMA model are obtained from
Short-term freeway traffic flow forecasting efforts to date have focused on predictions based solely on previous observations at the location of interest. This univariate prediction is useful for certain types of intelligent transportation system (ITS) forecasts, such as operational demand forecasts at system entry points. In addition, data from upstream sensors should improve forecasts at downstream locations. This motivates investigation of multivariate forecast models that include upstream sensor data. A candidate model is transfer functions with autoregressive integrated moving average errors, otherwise known as the ARIMAX model. The ARIMAX model was applied to motorway data from France that had been the subject of previous traffic flow forecasting research. The results indicate that ARIMAX models provide improved forecast performance over univariate forecast models. However, several issues must be addressed before widespread use of ARIMAX models for ITS forecasts is feasible. These issues include the increased complexity of model specification, estimation, and maintenance; model consistency; model robustness in the face of interruptions in the upstream data series; and variability in the cross-correlation between upstream and downstream observations. The last issue is critical because ARIMAX models assume constant transfer function parameters, whereas the correlation between upstream and downstream observations vary with prevailing traffic conditions, especially traffic stream speed. Therefore, further research is needed to investigate model extensions and refinements to provide a generalizable, self-tuning multivariate forecasting
model that is easily implemented and that effectively models varying upstream to downstream correlations.
The application of seasonal time series models to the single-interval traffic flow forecasting problem for urban freeways is addressed. Seasonal time series approaches have not been used in previous forecasting research. However, time series of traffic flow data are characterized by definite periodic cycles. Seasonal autoregressive integrated moving average (ARIMA) and Winters exponential smoothing models were developed and tested on data sets belonging to two sites: Telegraph Road and the Woodrow Wilson Bridge on the inner and outer loops of the Capital Beltway in northern Virginia. Data were 15-min flow rates and were the same as used in prior forecasting research by B. Smith. Direct comparisons with the Smith report findings were made and it was found that ARIMA (2, 0, 1)(0, 1, 1)96 and ARIMA (1, 0, 1)(0, 1, 1)96 were the bestfit models for the Telegraph Road and Wilson Bridge sites, respectively. Best-fit Winters exponential smoothing models were also developed for each site. The single-step forecasting results indicate that seasonal ARIMA models outperform the nearest-neighbor, neural network, and historical average models as reported by Smith.
This article presents the theoretical basis for modeling univariate traffic condition data streams as seasonal autoregressive integrated moving average processes. This foundation rests on the Wold decomposition theorem and on the assertion that a one-week lagged first seasonal difference applied to discrete interval traffic condition data will yield a weakly stationary transformation. Moreover, empirical results using actual intelligent transportation system data are presented and found to be consistent with the theoretical hypothesis. Conclusions are given on the implications of these assertions and findings relative to ongoing intelligent transportation systems research, deployment, and operations.
We've all been there. You're driving down the highway, just as your navigation app instructed, when Siri tells you to "proceed east for one-half mile, then merge onto the highway." But you're already on the highway. After a moment of confusion and perhaps some rude words about Siri and her extended AI family, you realize the problem: Your GPS isn't accurate enough for your navigation app to tell if you're on the highway or on the road beside it. Those days are nearly at an end. At the Institute of Navigation GNSS+ conference in Portland, Ore., in September, Broadcom announced that it is providing customers samples of the first mass-market chip to take advantage of a new breed of global navigation satellite signals. This new chip will give the next generation of smartphones 30‑centimeter accuracy as opposed to today’s 5 meters. Even better, it works in a city's concrete canyons, and it consumes half the power of today's generation of chips. The chip, the BCM47755, has been included in the design of some smartphones slated for release in 2018, but Broadcom would not reveal which.
We develop a deep learning model to predict traffic flows. The main contribution is development of an architecture that combines a linear model that is fitted using regularization and a sequence of layers. The challenge of predicting traffic flows are the sharp nonlinearities due to transitions between free flow, breakdown, recovery and congestion. We show that deep learning architectures can capture these nonlinear spatio-temporal effects. The first layer identifies spatio-temporal relations among predictors and other layers model nonlinear relations. We illustrate our methodology on road sensor data from Interstate I-55 and predict traffic flows during two special events; a Chicago Bears football game and an extreme snowstorm event. Both cases have sharp traffic flow regime changes, occurring very suddenly, and we show how deep learning provides precise short term traffic flow predictions.
Advances in location-acquisition and wireless communication technologies have led to wider availability of spatio-temporal (ST) data, which has unique spatial properties (i.e. geographical hierarchy and distance) and temporal properties (i.e. closeness, period and trend). In this paper, we propose a Deep -learning-based prediction model for S patio- T emporal data (DeepST). We leverage ST domain knowledge to design the architecture of DeepST, which is comprised of two components: spatio-temporal and global. The spatio-temporal component employs the framework of convolutional neural networks to simultaneously model spatial near and distant dependencies, and temporal closeness, period and trend. The global component is used to capture global factors, such as day of the week, weekday or weekend. Using DeepST, we build a real-time crowd flow forecasting system called UrbanFlow¹. Experiment results on diverse ST datasets verify DeepST's ability to capture ST data's spatio-temporal properties, showing the advantages of DeepST beyond four baseline methods.
Smart card data gathered by Automated Fare Collection (AFC) systems are a valuable resource for studying urban mobility. In this paper, we propose two approaches to clustering smart card data that can be used to extract mobility patterns in a public transportation system. Two complementary standpoints are considered: a station-oriented, operational point of view and a passenger-focused one. The first approach clusters stations based on when their activity occurs, i.e. how trips made at the stations are distributed over time. The second approach makes it possible to identify groups of passengers that have similar boarding times aggregated into weekly profiles. By applying our approaches to a real dataset issued from the metropolitan area of Rennes (France) we illustrate how they can help reveal valuable insights about urban mobility like the presence of different station key-roles such as residential stations used mostly in the mornings, work stations used only in the evening and almost exclusively during weekdays, etc. as well as different passenger behaviors ranging
from the sporadic and diffuse usage to typical commute practices. By cross-comparing passenger clusters with fare types, we also highlight how certain usages are more specific to particular types of passengers.
The authors' approach to intelligent marketing in smart cities uses large-scale crowdsourcing based on mobile user behavior for market planning. Unlike most work, which focuses on the client side, the proposed method is intended for market planning from a marketer perspective. The proposed system tracks and examines user trails via mobile devices. To support distributed analysis for big data, computer clusters project user trajectories onto blocks in a layout map of an actual geo-location. The trajectory falling into a block is independently handled by a distributed computer that detects crowd flows. Because traffic flows are highly relevant to geo-conquesting marketing, two methods are proposed: local longest common subsequence-based (LLCS-based) clustering, and the computation of transition flows and heat maps. Both help marketers analyze crowd flows for geo-conquesting. Simulation of the proposed methods revealed that the system facilitated computation of large-scale crowdsourced data.
This paper investigates how recurrent parking demand can be managed by dynamic parking pricing and information provision in the morning commute. Travelers are aware of time-varying pricing information and time-varying expected occupancy, through either their day-to-day experience or online information provision, to make their recurrent parking choices. We first formulate the parking choices under the User Equilibrium (UE) conditions using the Variational Inequality (VI) approach. More importantly, the System Optimal (SO) parking flow pattern and SO parking prices are also derived and solved efficiently using Linear Programming. Under SO, any two parking clusters cannot be used at the same time by travelers between more than one Origin–Destination (O–D) pairs. The SO parking flow pattern is not unique, which offers sufficient flexibility for operators to achieve different management objectives while keeping the flow pattern optimal. We show that any optimal flow pattern can be achieved by charging parking prices in each area that only depend on the time or occupancy, regardless of origins and destinations of users of this area. In the two numerical experiments, the best system performance is usually achieved by pricing the more preferred (convenient) area such that it is used up to a terminal occupancy of around 85–95%. Optimal pricing essentially balances the parking congestion (namely cruising time) and the level of convenience.
In the last two decades, the growing need for short-term prediction of traffic parameters embedded in a real-time intelligent transportation systems environment has led to the development of a vast number of forecasting algorithms. Despite this, there is still not a clear view about the various requirements involved in modelling. This field of research was examined by disaggregating the process of developing short-term traffic forecasting algorithms into three essential clusters: the determination of the scope, the conceptual process of specifying the output and the process of modelling, which includes several decisions concerning the selection of the proper methodological approach, the type of input and output data used, and the quality of the data. A critical discussion clarifies several interactions between the above and results in a logical flow that can be used as a framework for developing short-term traffic forecasting models.
Obtaining accurate information about current and near-term future traffic flows of all links in a traffic network has a wide range of applications, including traffic forecasting, vehicle navigation devices, vehicle routing, and congestion management. A major problem in getting traffic flow information in real time is that the vast majority of links is not equipped with traffic sensors. Another problem is that factors affecting traffic flows, such as accidents, public events, and road closures, are often unforeseen, suggesting that traffic flow forecasting is a challenging task. In this paper, we first use a dynamic traffic simulator to generate flows in all links using available traffic information, estimated demand, and historical traffic data available from links equipped with sensors. We implement an optimization methodology to adjust the origin-to-destination matrices driving the simulator. We then use the real-time and estimated traffic data to predict the traffic flows on each link up to 30 min ahead. The prediction algorithm is based on an autoregressive model that adapts itself to unpredictable events. As a case study, we predict the flows of a traffic network in San Francisco, CA, USA, using a macroscopic traffic flow simulator. We use Monte Carlo simulations to evaluate our methodology. Our simulations demonstrate the accuracy of the proposed approach. The traffic flow prediction errors vary from an average of 2% for 5-min prediction windows to 12% for 30-min windows even in the presence of unpredictable events.
Accurate and timely traffic flow information is important for the successful deployment of intelligent transportation systems. Over the last few years, traffic data have been exploding, and we have truly entered the era of big data for transportation. Existing traffic flow prediction methods mainly use shallow traffic prediction models and are still unsatisfying for many real-world applications. This situation inspires us to rethink the traffic flow prediction problem based on deep architecture models with big traffic data. In this paper, a novel deep-learning-based traffic flow prediction method is proposed, which considers the spatial and temporal correlations inherently. A stacked autoencoder model is used to learn generic traffic flow features, and it is trained in a greedy layerwise fashion. To the best of our knowledge, this is the first time that a deep architecture model is applied using autoencoders as building blocks to represent traffic flow features for prediction. Moreover, experiments demonstrate that the proposed method for traffic flow prediction has superior performance.
Traffic congestion has become a major challenge in recent years in many countries of the world. One way to alleviate congestion is to manage the traffic efficiently by applying intelligent transportation systems (ITS). One set of ITS technologies helps in diverting vehicles from congested parts of the network to alternate routes having less congestion. Congestion is often measured by traffic density, which is the number of vehicles per unit stretch of the roadway. Density, being a spatial characteristic, is difficult to measure in the field. Also, the general approach of estimating density from location-based measures may not capture the spatial variation in density. To capture the spatial variation better, density can be estimated using both location-based and spatial data sources using a data fusion approach. The present study uses a Kalman filter to fuse spatial and location-based data for the estimation of traffic density. Subsequently, the estimated data are utilized for predicting density to future time intervals using a time-series regression model. The models were estimated and validated using both field and simulated data. Both estimation and prediction models performed well, despite the challenges arising from heterogeneous traffic flow conditions prevalent in India.
: The artificial neural network (ANN) is one advance approach to freeway travel time prediction. Various studies using different inputs have come to no consensus on the effects of input selections. In addition, very little discussion has been made on the temporal–spatial aspect of the ANN travel time prediction process. In this study, we employ an ANN ensemble technique to analyze the effects of various input settings on the ANN prediction performances. Volume, occupancy, and speed are used as inputs to predict travel times. The predictions are then compared against the travel times collected from the toll collection system in Houston. The results show speed or occupancy measured at the segment of interest may be used as sole input to produce acceptable predictions, but all three variables together tend to yield the best prediction results. The inclusion of inputs from both upstream and downstream segments is statistically better than using only the inputs from current segment. It also appears that the magnitude of prevailing segment travel time can be used as a guideline to set up temporal input delays for better prediction accuracies. The evaluation of spatiotemporal input interactions reveals that past information on downstream and current segments is useful in improving prediction accuracy whereas past inputs from the upstream location do not provide as much constructive information. Finally, a variant of the state-space model (SSNN), namely time-delayed state-space neural network (TDSSNN), is proposed and compared against other popular ANN models. The comparison shows that the TDSSNN outperforms other networks and remains very comparable with the SSNN. Future research is needed to analyze TDSSNN's ability in corridor prediction settings.
Ride-sharing systems should combine environmental protection (through a reduction of fossil fuel usage), socialization, and security. Encouraging people to use ridesharing systems by satisfying their demands for safety, privacy and convenience is challenging. Most previous works on this topic have focused on finding a fixed path between the driver and the riders either based solely on their locations or using social information. The drivers’ and riders’ lack of options to change or compute the path according to their own preferences and requirements is problematic. With the advancement of mobile social networking technologies, it is necessary to reconsider the principles and desired characteristics of ride-sharing systems. In this paper, we formalized the ride-sharing problem as a multi source-destination path planning problem. An objective function that models different objectives in a unified framework was developed. Moreover, we provide a similarity model, which can reflect the personal preferences of the rides and utilize social media to obtain the current interests of the riders and drivers. The model also allows each driver to generate sub-optimal paths according to his own requirements by suitably adjusting the weights. Two case studies have shown that our system has the potential to find the best possible match and computes the multiple optimal paths against different user-defined objective functions.
Multianticipative driving behavior, where a vehicle reacts to many vehicles in front, has been extensively studied and modeled using a car-following (i.e., microscopic) approach. A lot of effort has been undertaken to model such multianticipative driving behavior using a macroscopic approach, which is useful for real-time prediction and control applications due to its fast computational demand. However, these macroscopic models have increasingly failed with an increased number of anticipations. To this end, this article puts forward derivation of an improved macroscopic model for multianticipative driving behavior using a modified gas-kinetic approach. First, the basic (microscopic) generalized force model, which has been claimed to fit well with real traffic data, is chosen for the derivation. Second, the derivation method relaxes the condition that deceleration happens instantaneously. Theoretical analysis and numerical simulations of the model are carried out to show the improved performance of the derived model over the existing (multianticipative) macroscopic models.
The use of recently developed time series techniques for short-term traffic volume forecasting is examined. A data set containing monthly volumes on a freeway segment for 1968-76 is used to fit a time series model. The resultant model is used to forecast volumes for 1977. The forecast volumes are then compared with actual volumes in 1977. Time series techniques can be used to develop highly accurate and inexpensive short-term forecasts. The feasibility of using these models to evaluate the effects of policy changes or other outside impacts is considered. (1 diagram, 1 map, 14 references,2 tables)
Region-based analysis is fundamental and crucial in many geospatial-related applications and research themes, such as trajectory analysis, human mobility study and urban planning. In this paper, we report on an image-processing-based approach to segment urban areas into regions by road networks. Here, each segmented region is bounded by the high-level road segments, covering some neighborhoods and low-level streets. Typically, road segments are classified into different levels (e.g., highways and expressways are usually high-level roads), providing us with a more natural and semantic segmentation of urban spaces than the grid-based partition method. We show that through simple morphological operators, an urban road network can be efficiently segmented into regions. In addition, we present a case study in trajectory mining to demonstrate the usability of the proposed segmentation method. Please cite the following papers when using this segmentation tool: [1] Yu Zheng, Yanchi Liu, Jing Yuan, and Xing Xie. Urban Computing with Taxicabs, ACM Ubicomp, 16 September 2011. [2] Nicholas Jing Yuan, Yu Zheng and Xing Xie, Segmentation of Urban Areas Using Road Networks, MSR-TR-2012-65, 2012.
Weather affects many aspects of transportation, but three dimensions of weather impact on highway traffic are predominant and measureable. Inclement weather affects traffic demand, traffic safety, and traffic flow relationships. Understanding these relationships will help highway agencies select better management strategies and create more efficient operating policies. For example, it was found that severe winter storms bring a higher risk of being involved in a crash by as much as 25 times - much higher than the increased risk brought by behaviors that state governments already have placed sanctions against, such as speeding or drunk driving. Given the heightened risk of drivers' involvement in a crash, highway agencies might wish to manage better and restrict use of highways during times of extreme weather, to reduce safety costs and costs associated with rescuing stranded and injured motorists in the worst weather conditions. However, the first step in managing the transportation systems to minimize the weather impact is to quantify its impact on traffic. This paper reviews the literature and recent research work conducted by the Center for Transportation Research and Education on the impact of weather on traffic demand, traffic safety, and traffic flow relationships. Included are new estimates of capacity and speed reduction due to rain, snow, fog, cold, and wind by weather intensity levels (e.g., snowfall rate per hour).
Short-term traffic prediction is of great importance to real-time traveler information and route guidance systems. Various methodologies have been developed for dynamic traffic prediction. However, many existing parametric studies focus on fixed-size data and presume time-invariant models. A proposed online adaptive model takes into account historical off-line data. A recursive algorithm is used to obtain computational efficiency and reduced storage. The algorithm is extended to a more general and flexible state-space model, and the predictions are computed recursively with a Kalman filter. A maximum likelihood off-line estimate of the noise covariance matrix and transition coefficients matrix is provided, as well as a recursive calculation of the optimal time-variant parameters on line. The result proves that the state-space model with the nonzero noise covariance matrix outperforms the other algorithms with loop detector data on Interstate 405 near Irvine, California.
Traffic volume is one of the fundamental types of data that have been used for the traffic control and planning process. Forecasting needs and efforts for various applications will be increased with the deployment of advanced traffic management systems. With the importance of the short-term traffic forecasting task, numerous techniques have been utilized to improve its accuracy. The use of the subset autoregressive integrated moving average (ARIMA) model for short-term traffic volume forecasting is investigated. A typical time-series modeling procedure was employed for this study. Model identification was carried out with Akaike's information criterion. The conditional maximum likelihood method was used for the parameter estimation process. Two white noise tests were applied for model verification. From the analysis results, four time-series models in different categories were identified and used for the one-step-ahead forecasting task. The performance of each model was evaluated using two statistical error estimates. Results showed that all time-series models performed well with reasonable accuracy. However, it was observed that the subset ARIMA model gave more stable and accurate results than other time-series models, especially a full ARIMA model. It is believed that the use of a subset ARIMA model could increase the accuracy of the short-term forecasting task within time-series models.
Real-time road traffic prediction is a fundamental capability needed to make use of advanced, smart transportation technologies. Both from the point of view of network operators as well as from the point of view of travelers wishing real-time route guidance, accurate short-term traffic prediction is a necessary first step. While techniques for short-term traffic prediction have existed for some time, emerging smart transportation technologies require the traffic prediction capability to be both fast and scalable to full urban networks. We present a method that has proven to be able to meet this challenge. The method presented provides predictions of speed and volume over 5-min intervals for up to 1h in advance.