Highway Network Research Articles

Utilizing Machine Learning for Predictive Maintenance of Climate-Resilient Highways through Integration of Advanced Asphalt Binders and Permeable Pavement Systems with IoT Technology

This review provides a comprehensive examination of the application of advanced technologies in enhancing the climate resilience of highway infrastructure. The focus is on the integration of machine learning for predictive maintenance, the use of Internet of Things (IoT) devices for real-time monitoring, and the deployment of advanced materials, specifically permeable pavements and modified asphalt binders. The review explores how these technologies work synergistically to create durable, low-maintenance highway systems capable of withstanding extreme environmental conditions. Advanced asphalt binders, such as those modified with styrene-butadiene-styrene (SBS) and nanomaterials, are discussed for their enhanced flexibility, thermal stability, and load-bearing capacity. Additionally, the environmental and structural benefits of permeable pavements are highlighted, particularly their role in stormwater management and reduction of urban heat island effects. Several case studies highlight the successful implementation of these technologies in diverse geographical and climatic conditions, including North Carolina, Australia, and Raipur, India. These cases illustrate the adaptability and environmental benefits of permeable pavements in managing stormwater, recharging groundwater, and mitigating the urban heat island effect. By synthesizing recent advancements and practical implementations, this review emphasizes the importance of integrating predictive technologies and resilient materials to develop sustainable, climate- adaptive highway systems. These insights offer a foundation for future infrastructure policies and technological innovations aimed at enhancing the durability and environmental sustainability of highway networks.

Comparative Analysis of Deep Neural Networks and Graph Convolutional Networks for Road Surface Condition Prediction

Road maintenance is essential for supporting road safety and user comfort. Developing predictive models for road surface conditions enables highway agencies to optimize maintenance planning and strategies. The international roughness index (IRI) is widely used as a standard for evaluating road surface quality. This study compares the performance of deep neural networks (DNNs) and graph convolutional networks (GCNs) in predicting IRI values. A unique aspect of this research is the inclusion of additional predictor features, such as the type and timing of recent roadwork, hypothesized to affect IRI values. Findings indicate that, overall, the DNN model performs similarly to the GCN model across the entire highway network. Given the predominantly linear structure of national highways and their limited connectivity, the dataset exhibits a low beta index, ranging from 0.5 to 0.75. Additionally, gaps in IRI data collection and discontinuities in certain highway segments present challenges for modeling spatial dependencies. The performance of DNN and GCN models was assessed across the network, with results indicating that DNN outperforms GCN when highway networks are sparsely connected. This research underscores the suitability of DNN for low-connectivity networks like highways, while also highlighting the potential of GCNs in more densely connected settings.

Quantitative attribution of industrial agglomeration patterns in Africa: global, local drivers and indirect effects

ABSTRACT Finding appropriate measures to drive industrial agglomeration is particularly urgent for Africa, which has been facing a decline in industrial production activities. Therefore, this study contributed by developing a multi-level spatial analysis framework that explored the global driving mechanisms and local heterogeneity of industrial agglomeration in Africa through various critical factors. The findings are as follows: (1) From 2009 to 2019, the industry in Africa has always been highly concentrated in a few countries, but the degree of agglomeration has been declining, with some countries in the East and West African regions showing a significant increase in the level of industrial agglomeration and a high potential for development. (2) Industrial agglomeration in African countries was driven by several factors, of which GDP per capita, highway network density, and electricity supply were the strongest and most consistent drivers. The impact strength of factors varied considerably across regions. (3) The drivers didn’t act independently and directly on the dependent variable, but were the product of synergy after the interaction between the two factors. Synergies between access to electricity and GDP per capita and other factors dominated the pattern of industrial agglomeration in Africa.

Robust evaluation of big data-driven winter weather traffic models using six weigh-in-motion sites as testbeds in Alberta's highway network

ABSTRACT This study assesses the temporal transferability of winter traffic models using dummy variable regression at six Weigh-in-Motion (WIM) sites on Alberta's highway network. Models for two vehicle classes were developed using five years of traffic and weather data. To evaluate transferability, an additional year of data was collected, and two model structures—dummy variable and naive—were tested. The models’ estimation accuracy was measured using R² and five error metrics. Results indicate successful transferability of the models to a different year, with performance varying by road function and vehicle class. The study highlights that different model types might be needed for each vehicle class to ensure high temporal transferability. Additionally, the quality of the test data is crucial for obtaining reliable transfer results. This research addressed gaps in previous studies by collectively testing the temporal transferability of winter traffic models across an entire highway network.

Modelling Severe Weather Impact on Traffic Mobility and Evaluating Cross-Regional Applicability to Functionally Distinct Segments in Cold Region Highway Networks

This study in Alberta, Canada, focuses on winter traffic modeling, crucial for transportation agencies in cold regions. By gathering data from six Weigh-in-motion (WIM) sites, models were developed and tested for three vehicle types. Notably, the research assessed the transferability of models between different highway segments, indicating successful transfer of coefficients to varied road functionalities. This suggests that agencies can optimize model structures for specific vehicle classes, enhancing usability without additional monitoring sites. Overall, the study highlights the feasibility of improving winter traffic models and their applicability across diverse road segments, benefiting transportation planning and management in cold climates.

An Information Gradient Approach to Optimizing Traffic Sensor Placement in Statewide Networks

Traffic sensors are vital to the development and operation of Intelligent Transportation Systems, providing essential data for traffic monitoring, management, and transportation infrastructure planning. However, optimizing the placement of these sensors, particularly across large and complex statewide highway networks, remains a challenging task. In this research, we presented a novel search algorithm designed to address this challenge by leveraging information gradients from K-nearest neighbors within an embedding space. Our method enabled more informed and strategic sensor placement under budget and resource constraints, enhancing overall network coverage and data quality. Additionally, we incorporated spatial kriging analysis, harnessing spatial correlations of existing sensors to refine and reduce the search space. Our proposed approach was tested against the widely used Genetic Algorithm, demonstrating superior efficiency in terms of convergence time and producing more effective solutions with reduced information loss.

Spatiotemporal statistical features of velocity responses to traffic congestions in a local motorway network

Abstract The causal connection between congestions and velocity changes at different locations induces various statistical features, which we identify and measure in detail. We carry out an empirical analysis of large-scale traffic data on a local motorway network around the Breitscheid intersection in the North Rhine-Westphalia, Germany. We put forward a response function which measures the velocity change at a certain location versus time conditioned on a congestion at another location. We use a novel definition of the corresponding congestion indicator to ensure causality. We find that the response of velocities to the congestion exhibits phase changes in time. A negative response at smaller time lags transforms into positive one at larger time lags, implying a certain traffic mechanism. The response decays as a power law with the distance. We also identify a scaling property leading to a collapse of the response functions on one curve.

Evaluation of carbon sink and photovoltaic system carbon reduction along roadside space

As China's photovoltaic (PV) sector experiences rapid growth, the availability of land resources has become a pivotal policy focus, driving the need for comprehensive research and strategic planning for roadside PV initiatives. Utilizing a fuzzy multi-criteria decision-making approach, combined with GIS spatial analysis and a modular design framework, our study quantitatively compared the carbon reduction capabilities of PV systems against the carbon sequestration potential of various vegetative arrangements along the roadside space. The roadside space analysis modular considers a range of factors including topography, meteorology, and construction costs. We examined the spatial distribution of suitability for PV installation and vegetation establishment along the provincial expressway network in China. The results revealed that Inner Mongolia stood out as the frontrunner in carbon reduction potential within high-suitability zones for PV construction, achieving an impressive 4.845 million tons of carbon reduction—nearly four times greater than that of Shaanxi Province. In contrast, the carbon sequestration attributed to vegetation greening in areas less suited for PV development revealed a higher propensity in the southeastern provinces. Guangdong led the charge with an impressive annual carbon sequestration of 2.89 million tons. This was closely followed by Yunnan, Sichuan, Hebei, Guizhou, and Henan, each achieving carbon sequestration amounts exceeding 2 million tons. These results offer valuable quantitative support and practical recommendations for achieving low-carbon objectives in the construction of China's expressways.

Application of AERMOD dispersion model for assessment PM10 concentrations from mobile sources in Kuala Lumpur Metropolitan City, Malaysia.

Deterioration of air quality in Kuala Lumpur caused by mobile sources and traffic-related activities with interaction with climatic conditions and dispersion in the atmosphere. This study was focused on predicting the averaged 1-h concentration of particulate matter (PM10) that was emitted from private cars in Kuala Lumpur by applying the air dispersion model American Meteorological Society (AMS)/United States Environmental Protection Agency (EPA) Regulatory Model (AERMOD) in 2014. The AERMOD model indicates that private cars in Kuala Lumpur recorded 1-h concentration of PM10 below the Malaysian Ambient Air Quality Guidelines (RMAQG) in 2014. The highest concentration was 59.2µg/m3 in 2014 and mainly concentrated around the city center and highway networks. The highest concentrations are recorded in early morning and late at night, with the highest concentrations at 4 am and 2 am and 48.68µg/m3 and 38.42µg/m3, respectively. Late night to early morning is identified as the dominant time interval for pollutant dispersion. The findings demonstrated that air quality had a significant impact on the 1-h concentration of PM10 emitted from private cars in Kuala Lumpur.

A Python toolkit for integrating geographic information system into regulatory dispersion models for refined pollution modeling

AERMOD is designated as U.S. Environmental Protection Agency (EPA)'s preferred air dispersion model for refined transportation project hot-spot analyses beginning in 2020. One of the key challenges in its modeling process is spatially encoding roadway geometry, especially when simulating highways with complex geometric designs. This research proposed an open-source Python package, GTA, which enables conversion of publicly available roadway Geographic Information System (GIS) layers into defined sources, and source-based emission rates from MOtor Vehicle Emissions Simulator (MOVES) output for AERMOD modeling. The research selected a suburban area in Atlanta, and conducted a comprehensive analysis in terms of annual PM2.5 concentration results and the speed of preparing AERMOD input files for highway network modeling both manually and using software developed based on the proposed methodology. The results prove that the proposed methodology significantly expedites the AERMOD input preparation process, and facilitates convenient testing of multiple modeling configurations for multi-scenario or sensitivity analysis.

Influence from highways on regional economic growth – based on the trade potential in China

The question of whether the construction of the highway network is economical and can produce positive economic benefits has been a hot topic of discussion in recent years. Previous scholars have explored the impact from multiple perspectives. Our paper draws the “trade potential” model proposed by Armstrong, based on the universal gravity model and the principle of space interaction, which is different from the traffic accessibility, market potential, and market access used in most of the literature. We argue that it is more appropriate to consider both the size impact and the time distance or trade cost impact of the two cities. The paper constructs a conceptual framework and theoretical model for the impact of highways on regional economic growth, measures the “minimum transit time” of highways between prefecture-level cities in China, and calculates the trade potential of prefecture-level cities. Through corresponding empirical model testing, we have obtained some meaningful conclusions.

Novel static decoupling algorithm for the multi-axis wheel force sensor based on the Informer network

Ensuring the stability of heavy multi-axle vehicles necessitates the accurate calibration and decoupling of Multi-axis Wheel Force Sensors (MWFS). Traditional methods often neglect the temporal coupling present in MWFS output data, leading to reduced accuracy. This paper introduces an Improved Decoupling Algorithm (IDI) based on the Informer network, designed to temporally decouple MWFS and enhance precision. The Decoupling embedding layer (DE) performs linear decoupling of the MWFS, while the Token embedding layer (TE) and Informer encoder extract timing coupling features. The highway network and linear fully-connected layer then provide nonlinear decoupling compensation. Experimental results demonstrate that the IDI algorithm significantly outperforms traditional methods like Extreme Learning Machine (ELM) and Back Propagation Network (BPNN), achieving at least a 41.12% improvement in accuracy in the highly coupled Mz channel. In conclusion, the IDI algorithm not only achieves high-precision decoupling of MWFS but also presents a robust framework for modeling various sensor types.

Expressway Network Model Construction and Road Section Impedance Estimation Based on Mixed Data

The acquisition of high-speed road network information and traffic environment information is of great significance for traffic management and planning, and at the same time, an accurate and efficient road network model is also the basis for the acquisition and application of traffic parameters. In order to quickly obtain high-speed road network information and road section traffic state information, this paper takes Chongqing Highway as an example, analyzes the traffic condition and characteristics of the highway network, and establishes the topology of the high-speed road network using network data and actual high-speed toll data from the graph theory and complex network theory. At the same time, based on the topology theory, fully exploiting the high-speed toll data, a more accurate estimate of different time periods, vehicles in the high-speed road section of the travel time, for highway traffic analysis to provide data reference, to improve the traffic planning program to enhance the efficiency of traffic management, to promote traffic planning and decision-making, as well as to enrich the theoretical basis of transportation research is of great significance.

Design and Development of a Real-Time Monitoring System for Highway Road Surface Anomalies

The recent unprecedented increase in highway networks in India has increased the need to maintain safe and smooth roads. The increasing number of road surface anomalies such as cracks, surface irregularities, and anomalies can lead to vehicle damage and accidents. This research work presents the design and development of a real-time monitoring system for detecting anomalies in highway road surfaces. The proposed system utilizes the machine vision model integrated with cameras, sensors, and edge computing to provide timely and accurate alerts for the drivers and also to the road maintenance authorities. The proposed solution is designed specially to work under different environmental conditions and enable large-scale deployment. The output generated from the proposed model is visual feedback of the detected anomalies and its severity analysis, enabling quick road maintenance actions.

Data-driven bottleneck detection on Tehran highways

In metropolitan areas, traffic congestion has become a prevalent challenge due to rapid urbanization and increased vehicle usage, adversely impacting mobility, productivity, and quality of life. Identifying and mitigating persistent traffic bottlenecks is crucial for developing efficient transportation systems and guiding infrastructure planning decisions. This research proposes an innovative data-driven methodology to pinpoint recurrent traffic bottlenecks in Tehran's extensive highway network, addressing the limitations of traditional traffic monitoring methods. Through data mining and image processing techniques applied to 16 months of traffic flow maps from Google Maps, diverse information is extracted, including traffic nodes, congestion hotspots, and locations with the longest queue lengths. The image processing approach involves color-based segmentation, pixel-level analysis, and machine learning algorithms to determine congestion levels across the highway network. The identified bottlenecks are validated against ground truth data from CCTV cameras, demonstrating a remarkable 92 % correlation for key identified points. The proposed approach leverages the power of advanced analytics to comprehensively analyze all major highways, including areas lacking CCTV infrastructure. The robust validation process reinforces the reliability of this data-driven solution in capturing real-world traffic dynamics. As urban mobility challenges escalate globally, the integration of diverse data sources and cutting-edge techniques will be instrumental in guiding intelligent transportation planning and policy decisions.

Dynamic Division of Control Subareas for Highway Networks Based on Improved Label Propagation Algorithm

Dynamic Division of Control Subareas for Highway Networks Based on Improved Label Propagation Algorithm

Data Source Importance Evaluation for Highway Networks: A Complex Network-Based Approach

Data collection technologies or data sources are critical for highway network management. However, due to the limitations on available management resources, determining the importance of these data sources is necessary to allocate these resources reasonably. This study proposes a complex network based method for evaluating the importance of multiple data sources in highway networks. This method includes mainly three steps. First, the business-data source relation will be identified and formulated for the highway network. Second, a business data source complex network is built from the previously identified business-data relationship. Third, an entropy weight method is used to compute and rank the importance of data source nodes by combining three indexes of degree centrality (DC), closeness centrality (CC) and structural holes (SC) computed based on the complex network. The proposed method is applied and illustrated using the highway network of Xuzhou City, Jiangsu Province, China. The results show that among the data sources, the most important data source is the continuous traffic survey station, followed by an automatic gantry-based station and vehicle detectors-based system. Discussions on the limitations, applications and future studies are provided for the proposed approach.

Stable weight updating: A key to reliable PDE solutions using deep learning

Deep learning techniques, particularly neural networks, have revolutionized computational physics, offering powerful tools for solving complex partial differential equations (PDEs). However, ensuring stability and efficiency remains a challenge, especially in scenarios involving nonlinear and time-dependent equations. This paper introduces novel residual-based architectures, namely the Simple Highway Network and the Squared Residual Network, designed to enhance stability and accuracy in physics-informed neural networks (PINNs). These architectures augment traditional neural networks by incorporating residual connections, which facilitate smoother weight updates and improve backpropagation efficiency. Through extensive numerical experiments across various examples—including linear and nonlinear, time-dependent and independent PDEs—we demonstrate the efficacy of the proposed architectures. The Squared Residual Network, in particular, exhibits robust performance, achieving enhanced stability and accuracy compared to conventional neural networks. These findings underscore the potential of residual-based architectures in advancing deep learning for PDEs and computational physics applications.

Cycle highway effects: Assessing modal choice to cycling in the Netherlands

Cycle highways are regarded as a promising new type of infrastructure because they promote longer-distance cycling between (sub)urban residential areas and work and study centers. This study examines whether the emerging network of regional cycle highways in the Netherlands has contributed to a modal shift from car to bicycle. More specifically, we investigate the effect of these routes on commuting bicycle mode choice. Our main data sources are a national travel survey covering commuting journeys that were made between 2010 and 2021 and a comprehensive dataset we have compiled to document the exact timing and status of all cycle highways in the Netherlands. We employ a difference-in-differences approach with a binary logit model, comparing bicycle mode choice versus the car for trips that benefited from a new cycle highway, before and after the introduction of the new infrastructure, with a control group of trips that were not affected by the construction of a new route. We present results from a novel routing-based approach to measuring exposure to this new cycling facility, which allows us to establish the extent to which the fastest route to work traverses a newly constructed cycle highway. After controlling for relevant covariates, our main results indicate that the introduction of cycle highways has contributed to a shift in commuting behavior toward cycling, with an increase of approximately 10 % in cycling probability post-intervention for trips highly exposed to cycle highways. The results also indicate some heterogeneity in the effects of cycle highways across different groups of individuals. The findings of this study are especially important in the context of the Netherlands (or similar biking countries, such as Denmark). Although these countries have well-established cycling infrastructure, they can still derive benefits from new cycling routes and can support decision-makers in other countries who want to invest in cycling in the near future.

Location method for emergency rescue node on expressways based on spatio-temporal characteristics of vehicle operation

Expressway networks are continuously developing and emergency rescue demand is increasing proportionately. The location of expressway emergency rescue nodes needs refinement to meet changing requirements. In this study, the expressway was modeled as an expressway network. The differences in the origin destination (OD) distribution matrices for working days and major holidays were used as the bases for determining the need for temporary emergency rescue nodes. Overlapping and non-overlapping community detection algorithms were used to extract the distribution characteristics of OD during both day categories. These distributions were used to determine permanent and temporary emergency rescue sites. In this study, we considered the differences in traffic volume, distance, and impact of four vehicle types on traffic accidents to select the location of emergency rescue nodes, and allocate emergency resources. An emergency rescue node selection model for an expressway network was established based on spatio-temporal characteristics. The results based on a regional example determined that 22 permanent and 25 temporary emergency rescue nodes were appropriate. The average rescue time for traffic accidents during working days and major holidays compared to the P-center location model, was reduced by approximately 27.08% and 6.70%, respectively. The coefficient of variation of emergency rescue time was reduced by approximately 28.22% and 21.41%, respectively. The results indicated that the model satisfied the expressway emergency rescue demand requirements, and improved the rationality of the rescue center node layout.