Origin-Destination (OD) demands are the backbone of traffic management and urban planning, serving as the fundamental input to numerous mobility applications. Existing studies mainly focus on modeling and predicting OD demands in the short term, while studies for long-term OD demand forecasting are limited. In particular, long-term OD demand prediction provides insights into the evolving spatiotemporal distribution of travel demands over extended periods, informing service scheduling and resource allocation that short-term forecasting cannot adequately support. One of the most distinguishing characteristics of OD demand time series is their inherent periodicity, punctuated by intermittent fluctuations. In view of this, we propose a novel interpretable Bayesian framework for long-term OD demand forecasting, which integrates both periodic patterns and transient fluctuations into a unified predictive model. By leveraging stochastic variational inference (SVI) and a modified tensor decomposition approach, the posterior distributions of the periodic and residual components in OD demands are formally derived. This enables the generation of both point-valued predictions and corresponding prediction intervals, which effectively quantify the predictive uncertainty and enhance model reliability. To validate the effectiveness of our proposed framework, we conduct experiments on real-world OD datasets. The results show that our model consistently outperforms state-of-the-art deep learning approaches under diverse scenarios. This underscores the effectiveness and robustness of our model in addressing the challenges of long-term multiple OD demand forecasting.

202602 bcjz

In the geospatial domain, universal representation models are significantly less prevalent than their extensive use in natural language processing and computer vision. This discrepancy arises primarily from the high costs associated with the input of existing representation models, which often require street views and mobility data. To address this, we develop a novel, training-free method that leverages large language models (LLMs) and auxiliary map data from OpenStreetMap to derive geolocation representations (LLMGeovec). LLMGeovec can represent the geographic semantics of city, country, and global scales, which acts as a generic enhancer for spatio-temporal learning. Specifically, by direct feature concatenation, we introduce a simple yet effective paradigm for enhancing multiple spatio-temporal tasks including geographic prediction (GP), long-term time series forecasting (LTSF), and graph-based spatio-temporal forecasting (GSTF). LLMGeovec can seamlessly integrate into a wide spectrum of spatio-temporal learning models, providing immediate enhancements. Experimental results demonstrate that LLMGeovec achieves global coverage and significantly boosts the performance of leading GP, LTSF, and GSTF models.

Networked urban systems facilitate the flow of people, resources, and services, and are essential for economic and social interactions. These systems often involve complex processes with unknown governing rules, observed by sensor-based time series. To aid decision-making in industrial and engineering contexts, data-driven predictive models are used to forecast spatiotemporal dynamics of urban systems. Current models, such as graph neural networks, have shown promise but face a tradeoff between efficacy and efficiency due to computational demands. Hence, their applications in large-scale networks still require further efforts. This article addresses this tradeoff challenge by drawing inspiration from physical laws to inform essential model designs that align with fundamental principles and avoid architectural redundancy. By understanding both micro- and macro-processes, we present a principled interpretable neural diffusion scheme based on transformer-like structures, whose attention layers are induced by low-dimensional embeddings. The proposed scalable spatiotemporal transformer (ScaleSTF), with linear complexity, is validated on large-scale urban systems including traffic flow, solar power, and smart meters, showing state-of-the-art performance and remarkable scalability. Our results constitute a fresh perspective on the dynamics prediction in large-scale urban networks.

Spatiotemporal traffic data (STTD) displays complex correlational structures. Extensive advanced techniques have been designed to capture these structures for effective forecasting. However, because STTD is often massive in scale, practitioners need to strike a balance between effectiveness and efficiency using computationally efficient models. An alternative paradigm based on multilayer perceptron (MLP) called MLP-Mixer has the potential for both simplicity and effectiveness. Taking inspiration from its success in other domains, we propose an adapted version, named NexuSQN, for STTD forecast at scale. We first identify the challenges faced when directly applying MLP-Mixers as series- and window-wise multivaluedness. To distinguish between spatial and temporal patterns, the concept of ST-contextualization is then proposed. Our results surprisingly show that this simple-yet-effective solution can rival SOTA baselines when tested on several traffic benchmarks. Furthermore, NexuSQN has demonstrated its versatility across different domains, including energy and environment data, and has been deployed in a collaborative project with Baidu to predict congestion in megacities like Beijing and Shanghai. Our findings contribute to the exploration of simple-yet-effective models for real-world STTD forecasting.

Spatiotemporal traffic data imputation is a fundamental task in numerous smart mobility applications. Existing studies indicate that accurately estimating missing values from observed data relies on capturing the spatiotemporal dependencies in traffic data. However, such dependencies may exhibit distinct characteristics across varying spatiotemporal areas, such as local short-term traffic fluctuations versus global long-range periodic commuting patterns. The comprehensive multiscale nature of dependencies in traffic data, encompassing more than just local and global levels, has not been well explored in the literature. To address this issue, we propose a physics-regularized multiscale attention network (PRMAN) that hierarchically extracts spatiotemporal features from local dynamics to global trends. Specifically, the proposed PRMAN builds upon the novel Swin Transformer and introduces a hierarchical architecture that performs self-attention in local spatiotemporal windows. By systematically expanding the window size across layers, this hierarchical design explicitly addresses the distinct characteristics between local and global spatiotemporal dependencies at different scales. Meanwhile, a physics-regularized loss function is developed to align learned spatiotemporal dependencies with traffic dynamics described by the fundamental diagram. This improves the model’s generalizability beyond the training data, ensuring robust performance on unseen datasets. Numerical experiments on multiple benchmark datasets demonstrate that our proposed PRMAN achieves state-of-the-art performance in handling diverse and complex missing data patterns. The code and model are publicly available athttps://github.com/2222ad/PRMAN

The balance between model capacity and generalization has been a key focus of recent discussions in long-term time series forecasting. Two representative channel strategies are closely associated with model expressivity and robustness, including channel independence (CI) and channel dependence (CD). The former adopts individual channel treatment and has been shown to be more robust to distribution shifts, but lacks sufficient capacity to model meaningful channel interactions. The latter is more expressive for representing complex cross-channel dependencies, but is prone to overfitting. To balance the two strategies, we present a channel-aware low-rank adaptation method to condition CD models on identity-aware individual components. As a plug-in solution, it is adaptable for a wide range of backbone architectures. Extensive experiments show that it can consistently and significantly improve the performance of both CI and CD models with demonstrated efficiency and flexibility. The code is available at https://github.com/tongnie/C-LoRA.

Missing data is a pervasive issue in both scientific and engineering tasks, especially for the modeling of spatiotemporal data. Existing imputation solutions mainly include low-rank models and deep learning models. The former assumes general structural priors but has limited model capacity. The latter possesses salient expressivity, but lacks prior knowledge of the underlying spatiotemporal structures. Leveraging the strengths of both two paradigms, we demonstrate a low rankness-induced Transformer to achieve a balance between strong inductive bias and high expressivity. The exploitation of the inherent structures of spatiotemporal data enables our model to learn balanced signal-noise representations, making it generalizable for a variety of imputation tasks. We demonstrate its superiority in terms of accuracy, efficiency, and versatility in heterogeneous datasets, including traffic flow, solar energy, smart meters, and air quality. Promising empirical results provide strong conviction that incorporating time series primitives, such as low-rankness, can substantially facilitate the development of a generalizable model to approach a wide range of spatiotemporal imputation problems.

Real-time traffic prediction models play a pivotal role in smart mobility systems and have been widely used in route guidance, emerging mobility services, and advanced traffic management systems. With the availability of massive traffic data, neural network-based deep learning methods, especially graph convolutional networks (GCNs) have demonstrated outstanding performance in mining spatio-temporal information and achieving high prediction accuracy. Recent studies reveal the vulnerability of GCN under adversarial attacks, while there is a lack of studies to understand the vulnerability issues of the GCN-based traffic prediction models. Given this, this article proposes a new task—diffusion attack, to study the robustness of GCN-based traffic prediction models. The diffusion attack aims to select and simulate attacks on a small set of nodes to degrade the performance of the traffic prediction models, and it can be used to examine vulnerabilities of the traffic prediction models. We propose a novel attack algorithm, which consists of two major components: 1) approximating the gradient of the black-box prediction model with simultaneous perturbation stochastic approximation (SPSA) and 2) adapting the knapsack greedy algorithm to select the attack nodes. The proposed algorithm is examined with three GCN-based traffic prediction models: 1) ST-GCN; 2) T-GCN; and 3) A3T-GCN on four cities. The proposed algorithm demonstrates high efficiency in adversarial attack tasks under various scenarios, and it can still generate adversarial samples under the drop regularization, such as DROP OUT, DROP NODE, and DROP EDGE. The research outcomes could help to improve the robustness of the GCN-based traffic prediction models and better protect the smart mobility systems.

Automatic Vehicle Identification (AVI) technology has been widely used for real-time path travel time estimation. For a study path equipped with AVI sensors at both ends, the difference between the timestamps of vehicles entering and leaving the path is AVI data. In urban areas, there can be several alternative routes and vehicle entry/exit points for the study path. Consequently, invalid AVI data occur that fall outside the scope of the travel time of the study path. Some AVI technologies based on identification information of vehicles can match vehicles precisely. However, for cities like Hong Kong with concerns of privacy issues, only commercial vehicle data can be collected. Under this scenario, the resultant AVI data are accurate but with few valid samples in a relatively short time interval due to the unavailability of private car data. The estimation accuracy of path travel times on a real-time basis will then be affected significantly by the existence of invalid AVI data. In this paper, a novel unsupervised algorithm is proposed to filter out real-time invalid AVI data efficiently although there is no ground truth available for training purposes. It is tested and compared with other benchmark algorithms on two selected paths in the Hong Kong urban road network. It is found that the proposed unsupervised algorithm can still filter limited but accurate AVI data with satisfactory performance. Sensitivity tests with ground truth are also conducted with different sampling rates. Some insightful findings are given for filtering AVI data under various scenarios.

City-wide short-term parking vacancy (PV) prediction is essential to transportation management. In many cities, only partial parking lots are equipped with sensors, causing severe data missing issues. Predicting PVs for parking lots without sensors is attractive but challenging. First, PV prediction itself is a nontrivial task since spatial dependencies among different regions in a city are complex and dynamic. By connecting parking lots based on geographical closeness using pre-designed rules, state-of-the-art PV prediction models achieve reasonable performance. However, those connectivities are not able to adapt to capture the changing dependencies in the graph. Second, node-to-node-based spatial and temporal dependencies will no longer be reliable since neighboring parking lot sensors may be intensely missing, which deteriorates the ability of the model to accurately impute PVs, leveraging the complex dependencies. To this end, we propose a novel framework named Collaborative graph Learning for pArking vacancy Prediction (CLaP). Specifically, to overcome the limitation of traditional pre-designed node connections, we propose a collaborative training method incorporating node attributes into graph augmentation, thus enhancing the ability to capture dynamic spatial dependencies through message aggregation. Besides, a recurrent PV recovering module is developed to impute missing embeddings by deeply coupling spatial-temporal dependencies. Experimental results on two real-world datasets demonstrate that CLaP outperforms state-of-the-art PV prediction models on city-wide PV prediction and imputation precision for parking lots without sensors.

To provide real-time parking information, existing studies focus on predicting parking availability, which seems an indirect approach to saving drivers’ cruising time. In this paper, we first time propose an on-street parking recommendation (OPR) task to directly recommend parking spaces for a driver. To this end, a learn-to-rank (LTR) based OPR model called OPR-LTR is built. Specifically, parking recommendation is closely related to the “turnover events” (state switching between occupied and vacant) of each parking space, and hence we design a highly efficient heterogeneous graph called ESGraph to represent historical and real-time meters’ turnover events as well as geographical relations; afterward, a convolution-based event-then-graph network is used to aggregate and update representations of the heterogeneous graph. A ranking model is further utilized to learn a score function that helps recommend a list of ranked parking spots for a specific on-street parking query. The method is verified using the on-street parking meter data in Hong Kong and San Francisco. By comparing with the other two types of methods: prediction-only and prediction-then-recommendation, the proposed direct-recommendation method achieves satisfactory performance in different metrics. Extensive experiments also demonstrate that the proposed ESGraph and the recommendation model are more efficient in terms of computational efficiency as well as saving drivers’ on-street parking time.

This paper presents a Bayesian data fusion framework for estimating journey time variability that uses a mixture distribution model to classify feeding data into different traffic states. Different from most studies, the proposed framework offers a generalized statistical foundation for making full use of multiple traffic data sources to estimate the vehicular journey time variability. Feeding data collected from multiple data sources are classified based on the associated traffic conditions, and the corresponding estimation biases of the individual data sources are determined by arbitrary distributions. The proposed framework is implemented and tested on a Hong Kong corridor with actual data collected from the field. Different statistical distributions of prior and likelihood knowledge are applied and compared. The findings of the case study show significant improvement in the journey time estimations of the proposed method compared with the individual measurements. The results also highlight the benefit of incorporating a traffic state classifier and prior knowledge in the fusion framework. This study contributes to the development of reliability-based intelligent transportation systems based on advanced traffic data analytics.

Sudden-onset hazards like earthquakes often induce cascading secondary hazards (e.g., landslides, liquefaction, debris flows, etc.) and subsequent impacts (e.g., building and infrastructure damage) that cause catastrophic human and economic losses. Rapid and accurate estimates of these hazards and impacts are critical for timely and effective post-disaster responses. Emerging remote sensing techniques provide pre- and post-event satellite images for rapid hazard estimation. However, hazards and damage often co-occur or colocate with underlying complex cascading geophysical processes, making it challenging to directly differentiate multiple hazards and impacts from satellite imagery using existing single-hazard models. We introduce DisasterNet, a novel family of causal Bayesian networks to model processes that a major hazard triggers cascading hazards and impacts and further jointly induces signal changes in remotely sensed observations. We integrate normalizing flows to effectively model the highly complex causal dependencies in this cascading process. A triplet loss is further designed to leverage prior geophysical knowledge to enhance the identifiability of our highly expressive Bayesian networks. Moreover, a novel stochastic variational inference with normalizing flows is derived to jointly approximate posteriors of multiple unobserved hazards and impacts from noisy remote sensing observations. Integrating with the USGS Prompt Assessment of Global Earthquakes for Response (PAGER) system, our framework is evaluated in recent global earthquake events. Evaluation results show that DisasterNet significantly improves multiple hazard and impact estimation compared to existing USGS products.

The rapid advancements of Internet of Things (IoT) and Artificial Intelligence (AI) have catalyzed the development of adaptive traffic control systems (ATCS) for smart cities. In particular, deep reinforcement learning (DRL) models produce state-of-the-art performance and have great potential for practical applications. In the existing DRL-based ATCS, the controlled signals collect traffic state information from nearby vehicles, and then optimal actions (e.g., switching phases) can be determined based on the collected information. The DRL models fully “trust” that vehicles are sending the true information to the traffic signals, making the ATCS vulnerable to adversarial attacks with falsified information. In view of this, this article first time formulates a novel task in which a group of vehicles can cooperatively send falsified information to “cheat” DRL-based ATCS in order to save their total travel time. To solve the proposed task, we develop CollusionVeh , a generic and effective vehicle-colluding framework composed of a road situation encoder, a vehicle interpreter, and a communication mechanism. We employ our framework to attack established DRL-based ATCS and demonstrate that the total travel time for the colluding vehicles can be significantly reduced with a reasonable number of learning episodes, and the colluding effect will decrease if the number of colluding vehicles increases. Additionally, insights and suggestions for the real-world deployment of DRL-based ATCS are provided. The research outcomes could help improve the reliability and robustness of the ATCS and better protect the smart mobility systems.

Accurate short-term traffic prediction plays a pivotal role in various smart mobility operation and management systems. Currently, most of the state-of-the-art prediction models are based on graph neural networks (Gnns), and the required training samples are proportional to the size of the traffic network. In many cities, the available amount of traffic data is substantially below the minimum requirement due to the data collection expense. It is still an open question to develop traffic prediction models with a small size of training data on large-scale networks. We notice that the traffic states of a node for the near future only depend on the traffic states of its localized neighborhoods, which can be represented using the graph relational inductive biases. In view of this, this paper develops a graph network (Gn)-based deep learning model LocaleGn that depicts the traffic dynamics using localized data aggregating and updating functions, as well as the node-wise recurrent neural networks. LocaleGn is a light-weighted model designed for training on few samples without over-fitting, and hence it can solve the problem of few-sample traffic prediction. The proposed model is examined on predicting both traffic speed and flow with six datasets, and the experimental results demonstrate that LocaleGn outperforms existing state-of-the-art baseline models. It is also demonstrated that the learned knowledge from LocaleGn can be transferred across cities. The research outcomes can help to develop light-weighted traffic prediction systems, especially for cities lacking historically archived traffic data.

Accurate real-time traffic forecast is critical for intelligent transportation systems (ITS) and it serves as the cornerstone of various smart mobility applications. Though this research area is dominated by deep learning, recent studies indicate that the accuracy improvement by developing new model structures is becoming marginal. Instead, we envision that the improvement can be achieved by transferring the ''forecasting-related knowledge" across cities with different data distributions and network topologies. To this end, this paper aims to propose a novel transferable traffic forecasting framework: Domain Adversarial Spatial-Temporal Network (DASTNet). DASTNet is pre-trained on multiple source networks and fine-tuned with the target network's traffic data. Specifically, we leverage the graph representation learning and adversarial domain adaptation techniques to learn the domain-invariant node embeddings, which are further incorporated to model the temporal traffic data. To the best of our knowledge, we are the first to employ adversarial multi-domain adaptation for network-wide traffic forecasting problems. DASTNet consistently outperforms all state-of-the-art baseline methods on three benchmark datasets. The trained DASTNet is applied to Hong Kong's new traffic detectors, and accurate traffic predictions can be delivered immediately (within one day) when the detector is available. Overall, this study suggests an alternative to enhance the traffic forecasting methods and provides practical implications for cities lacking historical traffic data. Source codes of DASTNet are available at https://github.com/YihongT/DASTNet.

The provision of lane-level travel time information can enable accurate traffic control and route guidance in urban roads with distinctive traffic conditions among lanes. However, few studies in the literature have been conducted to estimate lane-level travel time distributions. This study proposes a new vehicle re-identification (V-ReID) method for estimating lane-level travel time distributions using video images from widely deployed surveillance cameras. In the proposed method, a lane-based bipartite graph matching is introduced to obtain optimal matches between upstream and downstream vehicles by considering lane-level traffic conditions and vehicles’ lane changing behaviors and visual features. A lane-based travel time estimation technique is introduced to real-time estimate full spectrum of lane-level distribution parameters, including not only the mean but also the standard deviation and the distribution type. A comprehensive case study is carried out on a congested urban road in Hong Kong. Results of case study show that the proposed method outperforms the state-of-the-art link-based V-ReID method and is capable for providing accurate lane-level travel time distribution information on congested urban roads.

Urban rail transit (URT) system plays a dominating role in many megacities like Beijing and Hong Kong. Due to its important role and complex nature, it is always in great need for public agencies to better understand the performance of the URT system. This paper focuses on an essential and hard problem to estimate the network-wide link travel time and station waiting time using the automatic fare collection (AFC) data in the URT system, which is beneficial to better understanding the system-wide real-time operation state. The emerging data-driven techniques, such as the computational graph (CG) method in the machine learning field, provide a new solution for solving this problem. In this study, we first formulate a data-driven estimation optimization framework to estimate the link travel time and station waiting time. Then, we cast the estimation optimization model into a CG-based framework to solve the optimization problem and obtain the estimation results. The methodology is verified on a synthetic URT network and applied to a real-world URT network using the synthetic and real-world AFC data, respectively. Results show the robustness and effectiveness of the CG-based framework. To the best of our knowledge, this is the first time that the CG is applied to the URT. This study can provide critical insights to better understand the operational state of URT.

Small and marginal farmers, who account for over 80% of India's agricultural population, often sell their harvest at low, unfavorable prices before spoilage. These farmers often lack access to either cold storage or market forecasts. In particular, by having access to cold storage, farmers can store their produce for longer and thus have more flexibility as to when they should sell their harvest by. Meanwhile, by having access to market forecasts, farmers can more easily identify which markets to sell at and when. While affordable cold storage solutions have become more widely available, there has been less work on produce price forecasting. A key challenge is that in many regions of India, predominantly in rural and remote areas, we have either very limited or no produce pricing data available from public online sources.