A dataset containing the traffic network information in Los Angeles city from March to Jun 2012. It is used in the traffic forecasting task in Graph Neural Networks.

The source from the paper: https://arxiv.org/abs/2206.09113

The METR-LA dataset contains 02 information: - A adj_METR-LA.pkl: is the graph that contains the physical connection of 207 loop detectors in the traffic network. This dictionary contains 03 elements: the real sensor ID, the mapped sensor to node ID, and the adjacency matrix. - A METR-LA.h5: is the time series that is collected from each sensor in the traffic network over time.

We provide a standardized graph dataset for traffic based on the large-scale NuPlan v1.1 dataset, converted to a PyTorch-Geometric dataset using our CommonRoad-Geometric tool. The dataset is collected from 3 megacities: Singapore, Boston and Pittsburgh.

We provide a roads dataset that includes the spatial location of roads, the estimated age of each road, and the predicted traffic volume of each road between 1986 and 2020 in Wyoming, USA. Our annual estimates of traffic volume are available for each road and include estimates for all vehicles and truck only traffic. Moreover, we provide the estimated age of each road, where a minimum value of 1986 indicates that the road existed in 1986, and any later year indicates the most likely year that road was developed. This dataset will be beneficial for any research focused on the mechanistic effects of road traffic on wildlife populations. Our roads dataset is based on a comprehensive inventory of paved and unpaved roads in Wyoming of 2015 National Aerial Imagery Program (NAIP) aerial imagery (Fancher et al. 2023). We developed annual estimates of road age and vehicular traffic volume across 147,108 km of highways, arterials, collectors, local, and gravel/graded roads within the state of Wyoming. To assign road age, we leveraged a suite of ancillary data on surface disturbances (e.g., oil and gas drilling operations, wind turbines, and open pit mines) with known establishment dates. Then, we predicted traffic volume for each year across Wyoming using XGBoost, a novel machine learning method, to relate ongoing traffic monitoring by the Wyoming Department of Transportation with a separate set of spatial covariates hypothesized to explain traffic patterns across large regions.

Here we deposit the datasets we have extracted for ten states in the US. In each zip file, we include each state's accident records, road networks, and network features. For further information about using the dataset and how we extracted the data, check out our GitHub repository for instructions.

This dataset was developed for the project of analyzing the transport network in the Mazowieckie Voivodeship and comprises a wide range of traffic-related information. The data were collected from various sources, including road technical quality and road incident data from the Polish General Directorate for National Roads and Motorways, travel time information from Google Maps, data obtained from reverse geocoding, population density data from the GUS database, and specific weather conditions for roads.

Key Features of the Dataset:

Multidimensional Information: The dataset includes information on the date, days of the week, holidays, time (in minutes), and various temporal parameters (T1 - T24).

Road and Node Identifiers: Each record contains identifiers for the road (roadId), the start node (start_node), and the end node (end_node).

Traffic Factors: It includes key traffic information such as the traffic factor (trafficFactor), midlongitude and midlatitude of the road segment (midLongitude, midLatitude), and details about the number of lanes, road width, presence of two-way traffic (two_ways), and traffic density (density).

Weather Conditions: The dataset accounts for various weather conditions, including heavy rain, partial rain, no rain, partial clouds, heavy clouds, clear sky, storms, and fog.

Prediction Outcomes: Data include results on traffic speed (result_speed) and conditions such as shuttle traffic (result_shuttle), full (result_fullyclosed) and partial (result_partiallyclosed) road closures, and the presence of traffic lights (result_trafficlight).

Data Collection Period:

Traffic data were collected from May 25, 2022, to June 22, 2022, providing a comprehensive view of traffic conditions over a nearly one-month period. Data Preparation Process:

The collected data were unified and processed to create one large CSV file. This file was then divided into 384 smaller files, each representing the state of the transport network at a specific moment. This dataset forms a comprehensive basis for analyzing and forecasting traffic conditions, offering extensive possibilities for use in machine learning models.

The complex spatial-temporal correlations in traffic data make the traffic forecasting problem challenging. The proposed model captures the time-varying spatial correlations by progressively adapting to data used for forecasting tasks.

A. SUMMARY San Francisco International Airport Report on Monthly Passenger Traffic Statistics by Airline. B. HOW THE DATASET IS CREATED Data is self-reported by airlines and is only available at a monthly level C. UPDATE PROCESS Data updated quarterly D. HOW TO USE THIS DATASET Airport data is seasonal in nature, therefore any comparative analyses should be done on a period-over-period basis (i.e. January 2010 vs. January 2009) as opposed to period-to-period (i.e. January 2010 vs. February 2010). It is also important to note that fact and attribute field relationships are not always 1-to-1. For example, Passenger Counts belonging to United Airlines will appear in multiple attribute fields and are additive, which provides flexibility for the user to derive categorical Passenger Counts as desired.

SVR traffic data and transport attributes broken down by arcs of the road graph (dataset updated on 27 March 2020). Longitude and latitude data are expressed in the WGS84 reference system. The attributes of the dataset are: Arc Identification Code (IDNO) Origin Node Identification Code (TAIL) Destination Node Identification Code (HEAD) Arc length (LENG) Arc Function Class (HIER) Number of arc lanes (NLAN) Arch Heritage (PATR: A=motorway, S=state, R=regional, P=provincial, C=municipal) Average daily light-duty vehicle traffic, for the year 2018 (TGML_2018) Average daily heavy-duty vehicle traffic, for the year 2018 (TGMP_2018) Average travel speed for 2018 (ASPD_2018) Average daily light-duty vehicle traffic, for the year 2019 (TGML_2019) Average daily heavy-duty vehicle traffic, for the year 2019 (TGMP_2019) Average travel speed for 2019 (ASPD_2019) Morning rush hour congestion index, on an average weekday in November 2019 (CONG_8_9_NOV19) afternoon rush hour congestion index, on an average weekday in November 2019 (CONG_17_19_NOV19) Morning rush hour traffic status on an average working day in November 2019 (TRAF_8_9_NOV19: free, intense, slowed down, tail) State of afternoon rush hour traffic on an average working day in November 2019 (TRAF_17_19_NOV19: free, intense, slowed down, tail) Elements of the Regional TOC used for the preparation of the Regional Mobility Strategic Plan

A. SUMMARY This dataset consists of San Francisco International Airport (SFO) The aircraft landing dataset contains data about aircraft landings at SFO with monthly landing counts and landed weight by airline, region and aircraft model and type. B. HOW THE DATASET IS CREATED Data is self-reported by airlines and is only available at a monthly level. C. UPDATE PROCESS Data is available starting in July 1999 and will be updated monthly. D. HOW TO USE THIS DATASET Airport data is seasonal in nature; therefore, any comparative analyses should be done on a period-over-period basis (i.e. January 2010 vs. January 2009) as opposed to period-to-period (i.e. January 2010 vs. February 2010). It is also important to note that fact and attribute field relationships are not always 1-to-1. For example, Cargo Statistics belonging to United Airlines will appear in multiple attribute fields and are additive, which provides flexibility for the user to derive categorical Cargo Statistics as desired. E. RELATED DATASETS A summary of monthly comparative air-traffic statistics is also available on SFO’s internet site at https://www.flysfo.com/about/media/facts-statistics/air-traffic-statistics

The data is collected by the inductive loop detectors deployed on freeways in Seattle area. The freeways contain I-5, I-405, I-90, and SR-520. This data set contains spatiotemporal speed information of the freeway system. At each milepost, the speed information collected from main lane loop detectors in the same direction are averaged and integrated into 5 minutes interval speed data. The raw data is provided by Washington Start Department of Transportation (WSDOT) and processed by the STAR Lab in the University of Washington according to data quality control and data imputation procedures [1][2]. The data file is a pickle file that can be easily read using the read_pickle() function in the Pandas package. The data forms as a matrix and each cell of the matrix is speed value for the specific milepost and time period. The horizontal header of the data set denotes the milepost and the vertical header indicates the timestamps. For more information on the definition of milepost, please refer to this website. This data set been used for traffic prediction tasks in several research studies [3][4]. For more detailed information about the data set, you can also refer to this link. References: [1]. Henrickson, K., Zou, Y., & Wang, Y. (2015). Flexible and robust method for missing loop detector data imputation. Transportation Research Record, 2527(1), 29-36. [2]. Wang, Y., Zhang, W., Henrickson, K., Ke, R., & Cui, Z. (2016). Digital roadway interactive visualization and evaluation network applications to WSDOT operational data usage (No. WA-RD 854.1). Washington (State). Dept. of Transportation. [3]. Cui, Z., Ke, R., & Wang, Y. (2018). Deep bidirectional and unidirectional LSTM recurrent neural network for network-wide traffic speed prediction. arXiv preprint arXiv:1801.02143. [4]. Cui, Z., Henrickson, K., Ke, R., & Wang, Y. (2018). Traffic Graph Convolutional Recurrent Neural Network: A Deep Learning Framework for Network-Scale Traffic Learning and Forecasting. arXiv preprint arXiv:1802.07007.

The datasets include traffic speed data, spatial contexts and temporal contexts.

The American motor vehicle fleet traveled about ***** billion vehicle-miles in February 2025. Compared with January 2025, traffic decreased by about **** billion vehicle-miles. Between January and December 2024, traffic volume came to around *** trillion vehicle-miles of travel.

In the second quarter of 2024, the mobile data traffic reached almost *** exabytes worldwide, which is an increase of around ** exabytes compared to the same quarter in the previous year. The global mobile voice traffic has remained the same since the first quarter of 2016, with **** exabytes.

Traffic datasets for ICCPS "Uncertainty Quantification for Physics-Informed Traffic Graph Networks"

Linear network representing the estimated traffic flows for roads and highways managed by the Ministry of Transport and Sustainable Mobility (MTMD). These flows are obtained using a statistical estimation method applied to data from more than 4,500 collection sites spread over the main roads of Quebec. It includes DJMA (annual average daily flow), DJME (summer average daily flow), DJME (summer average daily flow (June, July, August, September) and DJMH (average daily winter flow (December, January, February, March) as well as other traffic data. It is important to note that these values are calculated for total traffic directions. Interactive map: Some files are accessible by querying a section of traffic à la carte with a click (the file links are displayed in the descriptive table that is displayed when clicking): • Historical aggregated data (PDF) • Annual reports for permanent sites (PDF and Excel) • Hourly data (hourly average per weekday per month) (Excel) • Annual reports for permanent sites (PDF and Excel) • Hourly data (hourly average per weekday per month) (Excel)**This third party metadata element was translated using an automated translation tool (Amazon Translate).**

This dataset accompanies the paper "STREETS: A Novel Camera Network Dataset for Traffic Flow" at Neural Information Processing Systems (NeurIPS) 2019. Included are: *Over four million still images form publicly accessible cameras in Lake County, IL. The images were collected across 2.5 months in 2018 and 2019. *Directed graphs describing the camera network structure in two communities in Lake County. *Documented non-recurring traffic incidents in Lake County coinciding with the 2018 data. *Traffic counts for each day of images in the dataset. These counts track the volume of traffic in each community. *Other annotations and files useful for computer vision systems. Refer to the accompanying "readme.txt" or "readme.pdf" for further details.

Wireless data traffic surged in the United States in 2023, with more than *** trillion megabytes of data transferred over mobile networks that year. This was almost ***** the volume consumed two years prior, with demand for data soaring amid the adoption of data intensive mobile activities.

Recently, accurate traffic flow prediction has become a significant part of intelligent transportation systems, which can not only satisfy citizens’ travel need and life satisfaction, but also benefit urban traffic management and control. However, traffic forecasting remains highly challenging because of its complexity in both topology structure and time transformation. Inspired by the propagation idea of graph convolutional networks, we propose ripple-propagation-based attentive graph neural networks for traffic flow prediction (T-RippleGNN). Firstly, we adopt Ripple propagation to capture the topology structure of the traffic spatial model. Then, a GRU-based model is used to explore the traffic model through the timeline. Lastly, those two factors are combined and attention scores are assigned to differentiate their influences on the traffic flow prediction. Furthermore, we evaluate our approach with three real-world traffic datasets. The results show that our approach reduces the prediction errors by approximately 2.24%-62,93% compared with state-of-the-art baselines, and the effectiveness of T-RippleGNN in traffic forecasting is demonstrated.

Video of the interior and exterior information of a driving scenario, with the resulting graph-based description. The video shows a combination of exterior data of a driving scenario in combination with information about the drivers eyegaze. On the right side a resulting graph-based model is depicted which combines the states that are relevant for vehicle safety applications.