Traffic Forecasting Research Articles

TEA-GCN: Transformer-Enhanced Adaptive Graph Convolutional Network for Traffic Flow Forecasting.

Traffic flow forecasting is crucial for improving urban traffic management and reducing resource consumption. Accurate traffic conditions prediction requires capturing the complex spatial-temporal dependencies inherent in traffic data. Traditional spatial-temporal graph modeling methods often rely on fixed road network structures, failing to account for the dynamic spatial correlations that vary over time. To address this, we propose a Transformer-Enhanced Adaptive Graph Convolutional Network (TEA-GCN) that alternately learns temporal and spatial correlations in traffic data layer-by-layer. Specifically, we design an adaptive graph convolutional module to dynamically capture implicit road dependencies at different time levels and a local-global temporal attention module to simultaneously capture long-term and short-term temporal dependencies. Experimental results on two public traffic datasets demonstrate the effectiveness of the proposed model compared to other state-of-the-art traffic flow prediction methods.

Advancing complex urban traffic forecasting: A fully attentional spatial-temporal network enhanced by graph representation

Accurate urban traffic forecasting is essential for intelligent transportation systems (ITS). However, the majority of existing forecasting methodologies predominantly concentrate on point-based forecasts (e.g., traffic detector forecasts). A limited number of them pay attention to the urban bidirectional road segments and the complex road network topology. To advance accurate traffic forecasting in complex urban scenarios, this paper proposes a Graph Representation enhanced Fully Attentional Spatial-Temporal network (GR-FAST). First, we construct a refined bidirectional road network graph (BRG) to depict the urban road network topology more accurately, particularly focusing on the turning patterns at intersections. Then, we adopt the graph representation methodology and introduce spatial information encoding (SIE) to explicitly characterize the significance of roads and network structure from multiple perspectives. Enhanced by SIE, spatial attention can capture spatial dependencies from both road network topologies and traffic pattern similarities, thereby forming a unified urban spatial cognition. Finally, a multi-scale residual perception (MRP) module is designed to balance the interplay of short-term temporal variability and long-term periodicity. Experiments on a real-world urban dataset from Wuhan, China, demonstrate that GR-FAST outperforms the state-of-the-art deep learning methods, achieving an improvement of 9.19%. Furthermore, ablation studies suggest that the explicit incorporation of complex road spatial topologies can significantly enhance forecasting accuracy.

Using Statewide Transportation Planning Model to Forecast Demand for Electric Vehicle Charging at Stations Along Intercity Highways

The availability of Charging Stations (CSs) with adequate capacity is critical for the growth in ownership and usage of Electric Vehicles (EVs). In the United States, the state Departments of Transportation (DOTs) are responsible for planning CSs along the Interstate, national, and state highways. While the state DOTs have their Statewide Transportation Planning Models (STPMs), most of these models do not represent EVs internally. The state DOTs can only rely on the existing STPMs to forecast the demand for EV charging at specific CS locations in a state’s highway network. This research proposes a methodology that utilizes an existing STPM, EV and non-EV vehicle registration data, and hourly traffic volumes at strategic counting stations to predict the hourly demand for EV charging (charging port occupancy) at proposed CSs. This approach first estimates the volume of all passenger cars that will pass the vicinity of a CS along intercity highways, and identifies the Traffic Analysis Zones (TAZs) of their trip origins. Based on the EV adoption rates of the TAZs, EV range, and EV owner’s charging behavior, the methodology forecasts the volume of EVs that pass the vicinity of a CS and the number of EVs that will stop at the CS to charge every hour of the day. An example problem demonstrated the application of the methodology in the State of Texas. It not only forecasted the hourly occupancy at CSs along three major corridors, but also demonstrated how to determine the capacities of the CSs. The challenges of using the existing STPM and their solutions are discussed. This proposed methodology enables the state DOTs to forecast the demand for EV charging and the capacities at the proposed CSs along intercity highways, thus allowing the DOTs to apply for federal funds to accelerate the deployment of the CSs. (298 words)

An AI-driven Approach to Traffic Congestion Prediction in Port Harcourt City Using Naïve Bayes

Four important intersections Location Junction, Rumuokoro Junction, GRA Junction, and Rumuokwuta Junction will be the focus of this study's investigation into the persistent problem of traffic congestion in Port Harcourt, Nigeria. In order to provide insights for improved traffic management, the project also aims to construct a predictive model that forecasts traffic congestion on certain routes using the Naïve Bayes algorithm. The study used both exploratory and descriptive research techniques to gather information about Port Harcourt's traffic congestion. Field surveys were carried out at the four intersections between 7 a.m. and 7 p.m. to collect traffic counts and vehicle footage. The investigation also included data from Rivers State Command, a division of the Federal Road Safety Corps (FRSC). Using this data to find patterns and trends in traffic flow, a Naïve Bayes algorithm was developed to classify and predict the degree of traffic congestion. With a 97% accuracy rate, the algorithm accurately classified 1164 out of 1200 traffic data points, predicting traffic congestion. The predictive algorithm used bar charts to show the traffic conditions; tall, moderate, and tiny bars, respectively, indicated high, medium, and low traffic congestion. The model's efficacy in real-time traffic forecasting was reinforced by the close match between the forecasts and observed traffic patterns. According to the study's findings, the Naïve Bayes algorithm can accurately forecast traffic congestion and provides road users and urban planners with useful information. The model can greatly enhance road efficiency by lowering travel times, relieving congestion, and improving traffic management on Port Harcourt's major arteries by accurately predicting traffic conditions.

NxLFTNet: NARX LSTM Forward Taylor Network for Traffic Forecasting

NxLFTNet: NARX LSTM Forward Taylor Network for Traffic Forecasting

The Correlation of the Smart City Concept with the Costs of Toxic Exhaust Gas Emissions Based on the Analysis of a Selected Population of Motor Vehicles in Urban Traffic

The intensive development of road transport has resulted in a significant increase in air pollution. This phenomenon is particularly noticeable in urban areas. This creates the need for analyses and forecasts of the scale and extent of future emissions of harmful substances into the environment. The aim of this study was to estimate the costs of the emission of toxic components of exhaust gases generated by all users of conventionally propelled vehicles travelling on a section of urban road in the next 25 years. The traffic study was carried out on an urban traffic route, playing a key role for road transport in the dimension of a given urban agglomeration. The traffic forecast for the analysed road section was based on the results of our own measurements carried out in April 2023 and external data from the General Directorate for Roads and Motorways. The results of the observations concerned six categories of vehicles for the morning and afternoon rush hours. Based on the data obtained, the generic structure of the vehicle population on the analysed section and the average daily traffic were determined. Using the methodology contained in the Blue Book of Road Infrastructure, parameters were calculated in the form of annual indicators of traffic growth on the analysed section, travel speed, and annual air pollution costs for selected vehicle categories, remembering at the same time that the Blue Book-based methodology does not distinguish between unit costs in relation to the type of emissions. The results of the study confirmed that there was an increase in the cost of toxic emissions for each vehicle category over the projected 25-year period. The largest increases were seen for trucks with trailers and passenger cars. In total, for all vehicle categories, emission costs nearly doubled from 2024 to 2046, from EUR 3,745,229 to EUR 7,443,384, due to the doubling of the number of vehicles resulting from the traffic forecast. The analyses presented here provide an answer to the question of what pollution costs may be faced by cities in which road transport will continue to be based on conventional types of propulsion. In addition, the research presented can be used to develop urban mobility transformation plans for the coming years, within the scope of the widely promoted smart city concept and the idea of electromobility, by pointing out to local authorities the direct economic benefits of these changes.

AI for Traffic Prediction and Management

Urban traffic congestion is a growing problem in cities across the globe, contributing to long delays, higher fuel consumption, environmental degradation, and economic losses. Conventional traffic management systems often depend on static data and rule-based approaches, which fall short in dealing with the complexity and variability of modern traffic. This paper introduces an AI-based traffic management approach that utilizes machine learning models to provide real-time traffic forecasts. By integrating historical data, live sensor inputs, and machine learning techniques, this system aims to enhance traffic flow, alleviate congestion, and improve travel efficiency. The model is compared against existing systems, demonstrating improved accuracy, flexibility, and scalability. Results indicate that the AI-based system offers significant advantages in managing urban traffic, surpassing traditional methods. The study further elaborates on how AI-powered traffic management can cut travel times by ensuring efficiency and safety, therefore playing a part in environmental sustainability through emission reduction, and the opportunities and challenges it brings to the table in the implementation of AI in traffic systems.

Towards effective, robust and utility-preserving watermarking of GPS trajectories

Personal GPS trajectory is essential for businesses and emerging data markets due to its relevance in various data-driven methods, including traffic forecasting, accident prediction, and profiling driving behavior. Watermarking is a method that facilitates verification of data ownership and authenticity by embedding provenance information into the data. Whereas watermarking is commonly adopted in the image and audio domains, only a few initial watermarking methods exist for GPS trajectory data. GPS trajectory watermarking is particularly challenging due to the spatio-temporal data properties and easiness of data modification. As a result, existing watermarking methods often embed only minimal provenance information, lack robustness, and can fail to preserve data utility for downstream applications. In this work, we propose W-Trace - a novel, effective, robust, and utility-preserving GPS trajectory watermarking method. W-Trace transforms a GPS trajectory into a complex domain and applies the Fourier transformation to decompose the trajectory into the frequency representation. W-Trace embeds watermarks in the frequency representation and verifies them in a spatiotemporally-aware procedure. We demonstrate the effectiveness, robustness, and utility of the proposed W-Trace approach in realistic settings using real-world GPS trajectory datasets. In contrast to the baselines, the proposed W-Trace approach is robust to a wide range of trajectory modifications while preserving the GPS trajectory characteristics required for the downstream applications.

Modeling and Numerical Investigations of Flowing N-Decane Partial Catalytic Steam Reforming at Supercritical Pressure

Steam reforming is an effective method for improving heat sinks of hypersonic aircraft at high flight Mach numbers. However, unlike the industrial process of producing hydrogen with a high water content, the catalytic steam reforming mechanism for the regeneration cooling process of hydrocarbon fuels with a water content below 30% is still unclear. Catalytic steam reforming (CSR) and catalytic thermal cracking (CTC) reactions occur at low temperatures, with the main products being hydrogen and carbon oxides. Thermal cracking (TC) reactions occur at high temperatures, with the main products being alkanes and alkenes. The above reaction exists simultaneously in the regeneration cooling channel, which is referred to as partial catalytic steam reforming (PCSR). Based on the experimental measurement results, an improved neural network correction method was used to establish a four-step global reaction model for the PCSR of n-decane under low water conditions. The reliability of the four-step model was verified by combining the model with a numerical simulation program and comparing it with the experimental results obtained by electric heating hydrocarbon fuels with a pressure of 3 MPa and a water content of 5/10/15%. The experimental and predicted results using the developed kinetic model are consistent with an error of less than 5% in the decane conversion rate. The average absolute error between the fuel outlet temperature and total heat sink is less than 10%. Using the PCSR model to predict the heat transfer characteristics of mixed fuels with different water contents, the convective heat transfer coefficient is basically the same, and the Nu number is affected by the thermal conductivity coefficient, showing different patterns with changes in the water content.

The Use of Smart Sensors to Enhance Road Safety: An Experimental Study to Explore the Usability of Smart Traffic Light Systems with Emergency Vehicles

The evolving landscape of transportation systems is being reshaped by the integration of sensors and Internet of Things (IoT) technology, particularly in the context of traffic light systems. These innovations aim to transform traditional traffic lights into smart devices capable of being controlled by emergency vehicles (EVs) to ensure safe passage, thereby enhancing road safety, and reducing intersection accidents. However, there are challenges in implementing such technologies and gaining the trust of EV drivers navigating busy intersections. The traditional fixed-time traffic signal plans are contrasted with the dynamic adjustment made possible by smart sensors and applications, which collect real-time traffic data and adapt signal timings to changing conditions, reducing congestion, and enhancing traffic flow. However, security concerns arise, necessitating the strengthening of systems with advanced technologies to address potential security and safety issues. The technological advancements have a significant impact on the transportation industry. Smart sensors in streetlights play a crucial role in solving long-standing transportation issues, and machine learning is employed for data analysis and traffic forecasting, benefiting both regular and EV drivers. Further research is needed to advance technologies like IoT and smart sensors and to integrate GSM sim cards and GPS to support EV drivers in finding the quickest route to their final destinations. EVs are becoming more than vehicles, evolving into smart devices equipped with technical tools allow drivers to control traffic lights, ensuring a clear path and more safety.

Spatial–Temporal-Correlation-Constrained Dynamic Graph Convolutional Network for Traffic Flow Forecasting

Accurate traffic flow prediction in road networks is essential for intelligent transportation systems (ITS). Since traffic data are collected from the road network with spatial topological and time series sequences, the traffic flow prediction is regarded as a spatial–temporal prediction task. With the powerful ability to model the non-Euclidean data, the graph convolutional network (GCN)-based models have become the mainstream framework for traffic forecasting. However, existing GCN-based models either use the manually predefined graph structure to capture the spatial features, ignoring the heterogeneity of road networks, or simply perform 1-D convolution with fixed kernel to capture the temporal dependencies of traffic data, resulting in insufficient long-term temporal feature extraction. To solve those issues, a spatial–temporal correlation constrained dynamic graph convolutional network (STC-DGCN) is proposed for traffic flow forecasting. In STC-DGCN, a spatial–temporal embedding encoder module (STEM) is first constructed to encode the dynamic spatial relationships for road networks at different time steps. Then, a temporal feature encoder module with heterogeneous time series correlation modeling (TFE-HCM) and a spatial feature encoder module with dynamic multi-graph modeling (SFE-DCM) are designed to generate dynamic graph structures for effectively capturing the dynamic spatial and temporal correlations. Finally, a spatial–temporal feature fusion module based on a gating fusion mechanism (STM-GM) is proposed to effectively learn and leverage the inherent spatial–temporal relationships for traffic flow forecasting. Experimental results from three real-world traffic flow datasets demonstrate the superior performance of the proposed STC-DGCN compared with state-of-the-art traffic flow forecasting models.

Enhancing Traffic Management in Cyber Physical Systems – A Gradient Based Fuzzy Controller Approach

Traffic forecast is a critical aspect of effective traffic management and planning in cyber-physical systems (CPS). In this study, we present a novel approach to traffic prediction and regulation within cyber-physical systems (CPS), introducing the Gradient Rule based Fuzzy Controller. This innovative methodology utilizes dynamic fuzzy logic control enhanced with gradient-based rules to adapt signal timings in real-time, effectively addressing the variable nature of traffic. Our results demonstrate significant improvements in reducing total queue length and delay at intersections, with reductions of up to 91.23%. Furthermore, extensive simulations and evaluations underscore the superiority of our approach compared to state-of-the-art models, highlighting its flexibility and adaptability to diverse traffic scenarios. This research emphasizes the novelty of integrating gradient-based rules into fuzzy control techniques, offering a promising avenue for advancing traffic management systems in CPS environments.

Discrete Time Series Forecasting of Hive Weight, In-Hive Temperature, and Hive Entrance Traffic in Non-Invasive Monitoring of Managed Honey Bee Colonies: Part I.

From June to October, 2022, we recorded the weight, the internal temperature, and the hive entrance video traffic of ten managed honey bee (Apis mellifera) colonies at a research apiary of the Carl Hayden Bee Research Center in Tucson, AZ, USA. The weight and temperature were recorded every five minutes around the clock. The 30 s videos were recorded every five minutes daily from 7:00 to 20:55. We curated the collected data into a dataset of 758,703 records (280,760-weight; 322,570-temperature; 155,373-video). A principal objective of Part I of our investigation was to use the curated dataset to investigate the discrete univariate time series forecasting of hive weight, in-hive temperature, and hive entrance traffic with shallow artificial, convolutional, and long short-term memory networks and to compare their predictive performance with traditional autoregressive integrated moving average models. We trained and tested all models with a 70/30 train/test split. We varied the intake and the predicted horizon of each model from 6 to 24 hourly means. Each artificial, convolutional, and long short-term memory network was trained for 500 epochs. We evaluated 24,840 trained models on the test data with the mean squared error. The autoregressive integrated moving average models performed on par with their machine learning counterparts, and all model types were able to predict falling, rising, and unchanging trends over all predicted horizons. We made the curated dataset public for replication.

Progressive Supervision via Label Decomposition: An long-term and large-scale wireless traffic forecasting method

Long-term and Large-scale Wireless Traffic Forecasting (LL-WTF) is pivotal for strategic network management and comprehensive planning on a macro scale. However, LL-WTF poses greater challenges than short-term ones due to the pronounced non-stationarity of extended wireless traffic and the vast number of nodes distributed at the city scale. To cope with this, we propose a Progressive Supervision method based on Label Decomposition (PSLD). Specifically, we first introduce a Random Subgraph Sampling (RSS) algorithm designed to sample a tractable subset from large-scale traffic data, thereby enabling efficient network training. Then, PSLD employs label decomposition to obtain multiple easy-to-learn components, which are learned progressively at shallow layers and combined at deep layers to effectively cope with the non-stationary problem raised by LL-WTF tasks. Finally, we compare the proposed method with various state-of-the-art (SOTA) methods on three large-scale WT datasets. Extensive experimental results demonstrate that the proposed PSLD significantly outperforms existing methods, with an average 2%, 4%, and 11% performance improvement on three WT datasets, respectively. In addition, we built an open source library for WT forecasting (WTFlib) to facilitate related research, which contains numerous SOTA methods and provides a strong benchmark. Experiments can be reproduced through https://github.com/Anoise/WTFlib.

Adaptive and Interactive Multi-Level Spatio-Temporal Network for Traffic Forecasting

Adaptive and Interactive Multi-Level Spatio-Temporal Network for Traffic Forecasting

Enhancing Road Traffic Prediction Using Data Preprocessing Optimization

Traffic prediction is essential for transportation planning, resource allocation, congestion management and enhancing travel experiences. This study optimizes data preprocessing techniques to improve machine learning-based traffic prediction models. Data preprocessing is critical in preparing the data for machine learning models. This study proposes an approach that optimizes data preprocessing techniques, focusing on flow-based analysis and optimization, to enhance traffic prediction models. The proposed approach explores fixed and variable orders of data preprocessing using a genetic algorithm across five diverse datasets. Evaluation metrics such as root mean squared error (RMSE), mean absolute error (MAE) and R-squared error assess model performance. The results indicate that the genetic algorithm’s variable order achieves the best performance for the ArcGIS Hub and Frementon Bridge Cycle datasets, fixed order one preprocessing for the Traffic Prediction dataset and variable order using the genetic algorithm for the PeMS08 dataset. Fixed order 2 preprocessing yields the best performance for the XI AN Traffic dataset. These findings highlight the importance of selecting the appropriate data preprocessing flow order for each dataset, improving traffic prediction accuracy and reliability. The proposed approach advances traffic prediction methodologies, enabling more precise and reliable traffic forecasts for transportation planning and management applications.

Hybrid Deep-Learning Approach with Geoffrey E. Havers-Based Loss Function and Evaluation Metric for Multilocation Traffic-Flow Forecasting

Traffic forecasting can enhance the efficiency of traffic control strategies such as routing decisions, variable speed limits, and ramp metering, resulting in a decrease in congestion, pollutants, and expenses, and an improvement in journey time predictability. Traffic forecasting, however, remains challenging because of the complex, heterogeneous, and cyclic nature of traffic data. To address this complexity, this research employs a multi-input hybrid deep self-attention network (MIHDSAN) for multilocation forecasting. The model inputs are selected using correlation analysis. New tunable loss and evaluation metrics formulations are proposed based on the traffic-modeling Geoffrey E. Havers (GEH) statistic. The proposed method was validated on two independent real-world traffic datasets from Stockton and Oakland, California. The weekly periodicity was the more relevant periodic input feature compared with daily variations; however, the daily variation was also significant for the Stockton dataset. The inclusion of weekly traffic periodicity (>95% correlated) improved the performance of the model by 3%. Adding daily periodicity was only beneficial for the Stockton dataset (91% correlated). The proposed GEH metric and its standard acceptance criterion offer both quantitative and qualitative means of evaluating the forecasts produced. The GEH loss function was consistent and outperformed current industry-standard methodologies of mean absolute error (MAE) in 80% and mean squared error (MSE) in 94% of cases. Therefore, this research presents evidence to suggest that the proposed GEH loss and evaluation functions validated in this paper become a standard criterion for traffic forecasting.

Research on Traffic Flow Forecasting of Spatio-temporal Convolutional Networks with Auto-correlation

Research on Traffic Flow Forecasting of Spatio-temporal Convolutional Networks with Auto-correlation

Dominance of Artificial Intelligence and Machine Learning Algorithms in Real-Time Traffic Flow prediction and Route Optimization in Autonomous Vehicles

The evolution of autonomous vehicles has elicited significant interest in understanding how real-time data may be used to provide enhanced driving experiences. This research project explored AI and Machine Learning methodologies applied for traffic forecasting and route optimization, and their implications for autonomous vehicles and urban mobility. For this project, the road traffic flow Dataset was utilized from Kaggle, containing 48,000 records of the flow of traffic, each including the following key features. In our work, we deployed some credible and well-established machine learning models: linear regression and random forest. These algorithms were separately trained by using a part of preprocessed data. The MSE for the Random Forest model was significantly lower, which means that the Random Forest Regressor had much smaller errors in estimating the volume of traffic compared to the Linear Regression model. The Random Forest Model had a high R² score, proving that this model explains a great deal of variance in the volume of traffic. This means that the 'model of random forest regressors' excellently fitted the data, snatching most of the important patterns and relationships between input features and target variables.

Enhancing Deep Learning-Based City-Wide Traffic Prediction Pipelines Through Complexity Analysis

Deep learning models can effectively capture the non-linear spatiotemporal dynamics of city-wide traffic forecasting. Evidence of varying deep learning model performance between different cities, different prediction horizons, different scales, specific city regions, and during particular hours of the day abounds in the literature on deep learning-based traffic prediction, yet a unified metric to quantify the complexity of different prediction tasks does not exist. This paper proposes two metrics—model complexity (MC) and intrinsic complexity (IC). While MC quantifies the effective complexity of deep learning models for city-wide traffic prediction tasks, the IC quantifies the underlying complexity of the prediction task. Being an effective complexity metric, MC depends on the model and the data. The IC depends only on the data and is invariant to the model being used. Both metrics are validated through systematic experimentation using traffic volume data from three cities. Finally, we demonstrate how these metrics can improve the workflows for deep learning-based data-driven traffic prediction pipelines and deployment by reducing the hyperparameter search scope and comparing the effectiveness of different design pathways.