Pavement Deterioration Models Research Articles

Prediction of the fundamental viscoelasticity of asphalt mixtures using ML algorithms

The complex modulus (E*) and phase angle (δ) are fundamental viscoelastic (VE) properties of asphalt material, crucial for pavement performance analysis and modeling. This study presents a novel approach to predict the fundamental VE properties of asphalt mixtures using 20 study mixes with disparate mix design information including different Performance Graded (PG grade) binders, recycled asphalt (RAP) contents, with/without polymer modifications, and mix volumetric properties. By employing the XGBoost model for feature importance analysis, integrating the Bailey method to streamline gradation information, and utilizing master curve shape parameters for modeling, this study has effectively reduced data redundancy and dimensionality. Among four machine learning (ML) models compared, the Multi-Hidden-Layer Backpropagation Neural Network (Multi BP) model demonstrates superior applicability to the asphalt mixture dataset, providing accurate predictions for VE properties at any temperature and frequency combinations. Notably, this predictive model relies solely on the mixture design information as inputs, minimizing the need for laborious laboratory experiments. The developed model can be integrated with Mechanistic-Empirical (ME) design, predicting the variations in rutting and cracking performance of asphalt mixtures over the entire service life. This provides essential foundational input support for long-lasting pavement design.

Pavement Sections’ Reliability Based on Deterioration Model Using Artificial Neural Network (ANN)

Pavement distresses, such as cracks and ruts, reduce pavements’ effectiveness and serviceability and can lead to failure. This underlines the importance of predicting pavements’ deterioration in pavement management systems (PMSs) for effective maintenance and rehabilitation (M&R) strategies. Consequently, it is essential to understand the concept of service life, which represents how long a pavement will remain in service based on how reliable it is. This study introduces a pavement deterioration model using data from the Long-Term Pavement Performance program for the international roughness index (IRI) and other factors. Different machine learning methods were utilized in developing the model to incorporate eight factors that significantly affect pavement roughness; these methods are: linear regression, regression tree, Gaussian Process Regression (GPR), Support Vector Machine (SVM), Ensemble Trees, and Artificial Neural Network (ANN). For comparison, the models’ performances were evaluated using Root Mean Squared Error (RMSE), Mean Squared Error (MSE), and R squared (R2). The weights and biases of the best model and the Federal Highway Administration (FHWA) recommended IRI ranges were utilized to create the limit state function. A reliability analysis using Monte Carlo Simulation (MCS) was determined to calculate the sections’ probability of failure. This study concluded that pavement sections in the US and Canada are reliable and that the mean yearly Kilo Equivalent Single Axle Load (KESAL) significantly contributes to pavement failure. Keywords: Pavement deterioration, IRI, Machine learning, Neural network, Pavement reliability, Probability of failure.

Calibration of Effective Structural Number and Tensile Strain Models Using Traffic Speed Deflectometer (TSD) Data for Enhanced Project-Level Assessment on Flexible and Composite Pavements

Pavement deterioration models provide the basis for predicting future changes in network conditions, estimating future funding needs, and determining the effectiveness and timing of maintenance and rehabilitation activities. Determining the accurate structural condition of pavements helps identify effective maintenance strategies which enhance the sustainability and service life of pavements. This study aimed to use the Traffic Speed Deflectometer (TSD) and Fast Falling Weight Deflectometer (FFWD) for project-level evaluation of pavements and use pavement properties to calibrate current models that help to predict the structural condition of pavements. Model parameters were calibrated to determine the effective structural number and tensile strains at the bottom of asphalt concrete for asphalt and composite pavements from TSD deflections. Tensile strains from the KENLAYER highlighted varied behaviors for composite pavements. A significant improvement in the calibration was observed for asphalt concrete pavements. While the TSD has higher daily operational costs than FWD, its per-mile cost is significantly lower, making it a viable choice for extensive coverage, even though the quantitative results might differ between the two devices.

Considering the Spatial Structure of the Road Network in Pavement Deterioration Modeling

Pavement deterioration modeling is important in providing information with respect to the future state of the road network and in determining the needs of preventive maintenance or rehabilitation treatments. This research incorporated the spatial dependence of the road network into pavement deterioration modeling through a graph neural network (GNN). The key motivation of using a GNN for pavement performance modeling is the ability to easily and directly exploit the rich structural information in the network. This paper explored if considering the spatial structure of the road network will improve the prediction performance of the deterioration models. The data used in this research comprises a large pavement condition dataset with more than a half million observations taken from the Pavement Management Information System maintained by the Texas Department of Transportation. The promising comparison results indicate that pavement deterioration prediction models perform better when the spatial relationship is considered.

Advances in Methodology for Fatigue Assessment of Composite Steel–Concrete Highway Bridges Based on the Vehicle–Bridge Dynamic Interaction and Pavement Deterioration Model

Fatigue cracking is one of the most prominent causes of mechanical failure limiting the service life of existing steel and composite steel–concrete bridges and is among the central concerns of structural and bridge engineers. In this context, the current work presents some recent advancements in an existing methodology for fatigue analysis developed by the authors throughout the years. The methodology is specifically devoted to the fatigue assessment of composite steel–concrete bridges employing the local hot-spot S-N approach and a coupled vehicle–pavement–bridge system considering progressive pavement deterioration with stochastically generated roughness profiles. Two different methodologies were used to solve the dynamic equilibrium equations: the modal superposition method to solve the bridge dynamic equations and a direct integration method to solve the vehicle dynamic equations. From a computational point of view, the present approach is more efficient and detailed than previous versions, as it allows a significant reduction in the analysis time and the use of complex bridge and vehicle finite element models. In this regard, a case study of a highway composite steel–concrete bridge spanning 40 m was selected in order to demonstrate the usefulness of the presented improved methodology by carrying out a fatigue analysis. The results of this investigation (displacements and stresses) are presented, aiming to verify the factors that directly influence the structural response and, consequently, the service life of steel–concrete composite highway bridges.

Effect of Data Imbalance on the Performance of Pavement Deterioration Models

Pavement performance models that can model consistently well across different pavement types, structures, traffic, and environmental conditions as well as throughout pavements’ lives are essential for the efficient management of transportation infrastructures. However, network pavement performance data are inherently imbalanced, and most pavement performance models are built on algorithms that assume balanced data distribution and equal error cost. As a result, the performance of most pavement performance models may vary for different pavement groups. Considering this, the objective of this study was to evaluate the data imbalance of the network pavement performance data and its effect on the performance of pavement deterioration models. Specifically, the analysis result for the longitudinal cracking model is presented in this paper. The data imbalance of the network pavement performance data mainly arises from two sources: the intrinsic pavement characteristics and pavement life expectancies. The intrinsic data imbalance arises either from the design or the management process, whereas the data imbalance introduced by the pavement life expectancies is primarily the result of the pavement management process. The distribution of the data used in this study exhibits substantial skewness owing to the data imbalance caused by the intrinsic pavement characteristics and life expectancies. The existence of the data imbalance leads to the biased performance of the deterioration models toward pavement classes with more data.

Detection of Unreported Treatments in Pavement Management System of Iowa DOT Using Machine Learning Classification Algorithm

Treatment records are among the most frequently underreported data items in pavement management systems (PMSs), which negatively affects various PMS analysis tools, such as pavement performance and deterioration models. Disregarding unreported treatments may lead to inaccurate pavement age and condition estimates, resulting in erroneous and nonoptimal maintenance and rehabilitation decisions. Nevertheless, the unreported and frequently missing pavement treatment data has received limited attention. Hence, this paper contributes to the body of knowledge by introducing a methodology for detecting unreported treatment actions and their occurrence probabilities over pavement age using a machine learning classification algorithm. Logistic regression models were developed using historical pavement condition data and validated on two levels: (1) split validation; and (2) manual validation using video logs of the pavement condition before and after treatment application. The results show that the developed models can detect unreported pavement treatments with accuracy, precision, and F1 scores ranging from 89% to 96%, 82% to 91%, and 70% to 85%, respectively. The presented methodology and developed models will help highway agencies identify unreported and missing pavement treatments, contributing to more cost-effective maintenance and rehabilitation decisions.

Traffic and Climate Impacts on Rutting and Thermal Cracking in Flexible and Composite Pavements

The study presented in this paper analyzed four long-term pavement performance (LTPP) test sections located in the states of New York (NY) and California (CA). Two of them are flexible pavement sections, whereas the other two are composite pavement sections. Two levels of analysis—in-state analysis and cross-state analysis—were performed for these pavement sections to determine the impacts of traffic and climate conditions. The performance of the pavement sections was evaluated in respect of thermal cracking and rutting resistance. The in-state analysis focused on comparing the pavement sections located in the same state. The two pavement sections located in CA exhibited insignificant variation in thermal cracking, although one of them had an additional 1.5” (38 mm) dense-graded asphaltic concrete (AC) layer. On the other hand, the additional 1.5” (38 mm) AC layer resulted in a significant reduction in the rutting depth in one pavement section. The in-state analysis of the two pavement sections located in NY revealed that the 0.8” (20.4 mm) chip seal layer had significantly low resistance to thermal cracking and rutting. The cross-state analysis examined pavement sections of comparable structural capacities—two with low structural capacity, and two with high structural capacity. The performance comparison of the two pavement sections with low structural capacity revealed that the chip seal layer exhibited a significantly high rutting depth, i.e., low rutting resistance under high traffic loads in a freezing climate. On the contrary, the two pavement sections with high structural capacity showed relatively high rutting resistance in both warmer and freezing climates. Furthermore, this paper presents the pavement deterioration models for rutting and thermal cracking in the LTPP test sections. These models were developed using multiple linear regression considering the pavement service life (age), traffic load (average annual daily truck traffic, AADTT), and climate impact (freezing index, FI). The deterioration models had coefficients of determination (r2) in the range of 0.82–0.99 and standard errors varying from 0.01 to 9.92, which indicate that the models are reliable.

Modeling the Effects of Asphalt Overlay Design on Pavement Distress Initiation with Random Parameters Hazard-Based Duration Models

Pavement deterioration model is a key input to a pavement management system that optimizes pavement maintenance and rehabilitation (M&R) planning. As a typical pavement M&R technique, asphalt overlay is commonly implemented to restore pavement performance. However, the impact of its design factors on pavement deterioration is not well documented in existing publications. This study aimed to investigate the effects of asphalt overlay design on the initiation of various types of cracking and severe rutting with consideration of endogenous bias and unobserved heterogeneity. A series of random parameters Weibull duration models were developed based on 317 asphalt overlay projects, which were collected from the Long-Term Pavement Performance program. Endogenous overlay design, overlay thickness, milling indicator, and overlaid pavement deflection were identified to represent their heterogeneity caused by unobserved factors over highway sections. Based on analysis results, the endogenous overlay design was shown to underestimate the service life of overlaid pavement. The occurrence probability of overlaid pavement distress would significantly decrease with an increase of asphalt overlay thickness. The application of 30% reclaimed asphalt pavement materials would retard the occurrence of severe rutting, but accelerate the occurrence of all types of cracking. Relative to asphalt overlay, milling and overlay can significantly reduce the hazard rate of severe rutting and all types of cracking. In addition, existing pavement condition, overlaid pavement deflection, traffic volume, and climate factors were identified to have significant effects on the distress initiation in overlaid pavements.

Assessment of Performance and Maintenance of Flexible Pavement Using KENLAYER and HDM-4

This paper describes the performance of flexible pavement using the KENLAYER and Highway Development and Management (HDM-4) software for urban roads. This study includes the study area of four-lane and two-lane roads at Hyderabad city, such as Gandi Maissama to Bachupally (GB), Bachupally to Nizampet (BN), Miyapur to Bachuapally (MB), JNTU to Pragathi Nagar (JP). The distress data like rutting, fatigue, pothole, patches, and edge drop data of each roads section were taken by adopting the manual methods. The pavement deterioration model was selected and developed based on the available distress using both the software KENLAYER and HDM-4. The current study aims to forecast the performance of specific roads in order to calculate the damage ratio. It also provides the design life of the roads as per the current and future traffic based on the distress model. Further, the maintenance of the particular roads is carried out using the HDM-4 models. It is also noticed that major damage was found in Gandi Maissama to Bachupally than the other road section. This study concluded that the comparative study of different roads section and governing factors for distress along with the allowable limits of distress in each section of the flexible pavement were found. Finally, the output of KENLAYER and HDM-4 has been compared.

Analysis of Overweight Truck Permit Policy in New Jersey

The roadway infrastructure constantly deteriorates because of environmental conditions, but other factors such as exposure to heavy trucks exacerbates the rate of deterioration. Therefore, decision-makers are constantly searching for ways to optimize allocation of the limited funds for repair, maintenance, and rehabilitation of New Jersey’s infrastructure. New Jersey legislation requires operators of overweight (OW) trucks to obtain a permit to use the infrastructure. The New Jersey Department of Transportation (NJDOT) issues a variety of permits based on the types of goods carried. These permits allow OW trucks to use the infrastructure either for a single trip or for multiple trips. Therefore, one major concern is whether the permit revenue of the agency can recoup the actual cost of damage to the infrastructure caused by these OW trucks. This study investigates whether NJDOT’s current permit fee program can collect enough revenue to meet the actual cost of damage to the infrastructure caused by these heavy-weight permit trucks. The infrastructure damage is estimated by using pavement and bridge deterioration models and New Jersey permit data from 2013 to 2018 containing vehicle configuration and vehicle route. The analysis indicates that although the cost of infrastructure damage can be recovered for certain permit types, there is room for improvement in the permit program. Moreover, based on permit rules in other states, the overall rank of the New Jersey permit program is evaluated and possible revisions are recommended for future permit policies.

Use of Time–Temperature Superposition Principle to Create Pavement Performance Master Curves and Relate Pavement Condition Index and International Roughness Index

The international roughness index (IRI) is one of the most popular indices to measure pavement roughness. State agencies and cities with plenty of resources often collect IRI and pavement distresses every year or every other year, but some others with fewer resources will collect this information every 3 to 5 years. Collecting IRI is much more affordable than collecting pavement distresses. With this in mind, the objective of this paper was to establish a relationship between IRI and pavement condition index (PCI) using pavement deterioration models for both PCI and IRI based on the concept of time–deterioration superposition similar to the time–temperature superposition principle, and then combine both models to establish this relationship. Additionally, this study was used to establish threshold limits for IRI measurements that can be used as a general reference for pavement condition. Data from the Long-Term Pavement Performance InfoPave was used to perform the analysis for three network samples from Arizona, California, and Wisconsin. This analysis only included flexible pavements. The results from Arizona, California, and Wisconsin showed a good relationship between IRI and PCI using the proposed approach with a coefficient of determination ranging from 0.71 to 0.85. Furthermore, the analysis showed that the change in IRI over time can be related to the change in PCI over time. The general thresholds developed in this study apply to the sections evaluated but the approach can be used to set limits for other networks.

Using Machine Learning to Examine Impact of Type of Performance Indicator on Flexible Pavement Deterioration Modeling

AbstractLimited research has been conducted on the application of data analytics to the prediction of the Pavement Condition Index (PCI) of asphalt roads. More importantly, studies comparing the pr...

IRI deterioration model for asphalt concrete pavements: capturing performance improvements over time

Pavement deterioration models are an essential component of performance-based management of transportation infrastructure. This study aims to improve existing techniques by developing a deterioration model that captures the effect of pavement performance improvement across time. By using an iterative reweighted least squares approach, the model is applied to publicly available roughness data of asphalt concrete pavements collected as part of FHWA’s Long-Term Pavement Performance (LTPP) program. The results of the case study demonstrate that pavement performance has significantly improved over time. The parametric model suggests that the expected serviceable life of a typical asphalt pavement built in 1970 is only 22 years whereas a section with identical structural capacity and traffic loading built in 2000 achieves a 31-year serviceable life. These discrepancies highlight the importance of incorporating the construction year of a pavement as another explanatory factor within pavement deterioration models. The proposed model accounts for biases in existing approaches and will allow decision-makers to improve their current deterioration models and produce more optimal resource allocation strategies for existing pavement assets.

Modeling the international roughness index performance on semi-rigid pavements in single carriageway roads

Pavement deterioration models are a vital feature in any pavement management system since they are capable of predicting the evolution of pavement characteristics. Pavement roughness is measured by most of the highway administrations due to its relation to comfort and safety, generally by means of the International Roughness Index (IRI). The Regional Government of Biscay (Spain) has collected IRI values since 2000 on its road network. Although many models have been developed for flexible pavements, very few have been proposed for semi-rigid pavements. The paper aims to develop IRI prediction models for semi-rigid pavements in single-carriageway roads. Considering the high quantity of available information in the database, deterministic models were selected. Due to the importance of the pavement structure in IRI evolution observed in flexible models, only segments with completely known pavement details were employed, i.e., a section where the complete structure is known: materials and thickness of existing layers above the subgrade. The pavement age, as precise as practical, and the accumulated total traffic and heavy traffic through the section were identified as roughness accelerating factors. Conversely, the materials used in base and subbase layers, their thickness, and the total thickness of bituminous layers were observed as degradation reducing factors. Possible treated base and subbase materials included in the model were soil–cement, gravel-cement, and gravel and slag. The obtained model achieved a determination coefficient (R2) of 0.569. Additionally, the bituminous material of the surface layer was verified as an affecting factor too, which can be introduced to improve the model’s accuracy. Possible surface layer materials included dense (D) and semi-dense (S) asphalt concrete, with a maximum aggregate diameter of 16 and 22 mm, discontinuous mixing (BBTM 11A) and porous asphalt (PA 11). The additional model achieved a higher determination coefficient (0.645) and, hence, a more accurate IRI prediction resulted.

Upgradation of pavement deterioration models for urban roads by non-hierarchical clustering

Pavements, being a significant component of urban infrastructure, their maintenance and rehabilitation to the desired serviceability level is a challenging problem faced by engineers. The development of a reliable pavement deterioration model is essential to devise proper maintenance policies. This exploratory paper presents the development of network-level pavement performance prediction models for the selected arterial and sub-arterial roads of Tiruchirappalli city, India. Road inventory, traffic volume, maintenance history, pavement condition, and roughness data of the study area are collected periodically for seven years. The Pavement Condition Index (PCI) is determined from the data collected through visual evaluation of the type, severity, and amount of pavement distress. Roughometer is deployed to obtain the International Roughness Index. The parameters which influence pavement deterioration vary widely for different roads within the same network. The pavement sections are assembled into three homogeneous clusters using k-means clustering, which is a nonhierarchical clustering algorithm, so that they can be modeled with better acceptability. Pavement performance prediction models are generated for different clusters using multiple linear regression analysis, and comparison is made with that developed for non-clustered data. The error in prediction is found to be less for clustered models. While the pavement sections in cluster 2, when left unmaintained, deteriorates from a PCI value of 100 to 77 in 5 years, those belonging to cluster 3 are found to deteriorate from 100 to 13. The variation in the deterioration process and the significance of clustering pavement sections for efficient pavement maintenance management is established.

Deterioration Pattern of Flexible Pavement with the Help of Falling Weight Deflectometer

Maintenance and repair of the highway network system are major expenses in the state budget. For this reason, various concerned organizations are pointing out the need for developing an intelligent and efficient pavement performance model that can prioritize pavement maintenance and rehabilitation works. Such models can forecast the remaining pavement service life and pavement rehabilitation needs, and can help in the formulation of pavement maintenance and strengthening programmes which will reduce the road agency and road user costs. The flexible pavement performance or deterioration models involve the complex interaction between vehicles, environment, structure and surface of the pavement. Performance models relating to the pavement distress conditions like, cracking, ravelling, potholing, and roughness are analysed and developed by various researchers. Understanding the deterioration pattern of the flexible pavement is very important in order to take the decision for strengthening the pavement. The remaining life of the pavement depends on various factors such as Traffic, Environment and climatic conditions hence keeping in mind these factors. the thesis presents the pattern of the deterioration of remaining life of pavement. The thesis emphasis on determining the remaining life of pavement by conducting the FWD test. The FWD test is conducted on the same pavement for three time at regular interval to verify the remaining life of the flexible pavement.

A modified pavement condition rating index for flexible pavement evaluation in Egypt

Pavement evaluation is conducted to assess the functional and/or structural condition of existing pavement systems which can be done on project or network levels. Through pavement condition assessment, pavement deterioration models can be established and thus maintenance and rehabilitation alternatives can be proposed. The prediction of pavement performance which is primarily based on the pavement evaluation results is a critical element of a successful pavement management system (PMS). The main objective of this research is to revise and modify the current pavement condition rating (PCR) method for pavement evaluation as a major component of an effective PMS for roads and transport directorates in Egypt. Pavement condition data were collected from literature, General authority of roads, bridges and land transport in Egypt and the long-term pavement performance (LTPP) database. Due to its accuracy, only the LTPP data were used to modify the current PCR method to yield comparable values to the very well-known pavement condition index (PCI). The proposed modified PCR was validated using pavement distress data collected from two rural roads in Egypt, and the data showed reasonable accuracy as compared to the PCI method.

Statistics and Artificial Intelligence-Based Pavement Performance and Remaining Service Life Prediction Models for Flexible and Composite Pavement Systems

In their pavement management decision-making processes, U.S. state highway agencies are required to develop performance-based approaches by the Moving Ahead for Progress in the 21st Century (MAP-21) federal transportation legislation. One of the performance-based approaches to facilitate pavement management decision-making processes is the use of remaining service life (RSL) models. In this study, a detailed step-by-step methodology for the development of pavement performance and RSL prediction models for flexible and composite (asphalt concrete [AC] over jointed plain concrete pavement [JPCP]) pavement systems in Iowa is described. To develop such RSL models, pavement performance models based on statistics and artificial intelligence (AI) techniques were initially developed. While statistically defined pavement performance models were found to be accurate in predicting pavement performance at project level, AI-based pavement performance models were found to be successful in predicting pavement performance in network level analysis. Network level pavement performance models using both statistics and AI-based approaches were also developed to evaluate the relative success of these two models for network level pavement performance modeling. As part of this study, in the development of pavement RSL prediction models, automation tools for future pavement performance predictions were developed and used along with the threshold limits for various pavement performance indicators specified by the Federal Highway Administration. These RSL models will help engineers in decision-making processes at both network and project levels and for different types of pavement management business decisions.

Development of An Artificial Neural Network Model for Asphalt Pavement Deterioration Using LTPP Data

Deterioration models are important and essential part of any Pavement Management System (PMS). These models are used to predict future pavement situation based on existence condition, parameters causing deterioration and implications of various maintenance and rehabilitation policies on pavement. The majority of these models are based on roughness which is one of the most important indices in pavement evaluation. High correlation between International Roughness Index (IRI) and user comfort led to modeling pavement deterioration based on IRI during PMS history. On the other hand, in recent years Artificial Neural Network (ANN) which is a valuable tool of soft computing is used in pavement modeling, widely. This study assessed the development of an ANN pavement deterioration model based on IRI using Back-Propagation Neural Networks (BPNN) technique. The Long-Term Pavement Performance (LTPP) data was extracted from two General Pavement Study (GPS) sections including GPS-1 and GPS-2. After training and testing the developed model, results were compared with a polynomial regression model. Results showed that predicted IRI values with developed ANN model have a good correlation with measured values rather than the polynomial regression model for both GPS-1 and GPS-2 sections.