Initial International Roughness Index Research Articles

Machine Learning Modeling of Wheel and Non-Wheel Path Longitudinal Cracking

Roads degrade over time due to various factors such as traffic loads, environmental conditions, and the quality of materials used. Significant investments have been poured into road construction globally, necessitating regular evaluations and the implementation of maintenance and rehabilitation (M&R) strategies to keep the infrastructure performing at a satisfactory level. The development and refinement of performance prediction models are essential for forecasting the condition of pavements, especially to address longitudinal cracking distress, a major issue in thick asphalt pavements. This research leverages multiple machine learning methods to create models predicting non-wheel path (NWP) and wheel path (WP) longitudinal cracking using data from the Long-Term Pavement Performance (LTPP) program. This study highlights the marked differences in distress conditions between WP and NWP, underscoring the importance of precise models that cater to their unique features. Aging trends for both types of cracking were identified through correlation analysis, showing an increase in WP cracking with age and a higher initial International Roughness Index (IRI) linked to NWP cracking. Factors such as material characteristics, kinematic viscosity, pavement thickness, air voids, particle size distribution, temperature, KESAL, and asphalt properties were found to significantly influence both WP and NWP cracking. The Exponential Gaussian Process Regression (GPR) emerged as the best model for NWP cracking, showcasing exceptional accuracy with the lowest RMSE of 89.11, MSE of 7940.72, and an impressive R-Squared of 0.63. For WP cracking, the Squared Exponential GPR model was most effective, with the lowest RMSE of 12.00, MSE of 143.93, and a high R-Squared of 0.62. The GPR models, with specific kernels for each cracking type, proved their adaptability and efficiency in various pavement scenarios. A comparative analysis highlighted the superiority of our new machine learning model, which achieved an R2 of 0.767, outperforming previous empirical models, demonstrating the strength and precision of our machine learning approach in predicting longitudinal cracking.

Modeling of road damage prediction due to overload using HDM-4 method

The road use load should essentially conform to the plan load as stated in the design guidelines. But in reality, the load received by the road is more than standard, so it has the potential to accelerate road damage. This study aims to predict the increase in the value of the International Roughness Index (IRI) and model the increase in value (IRI) due to overload. The approach method used in the analysis is the prediction of an increase in the IRI value due to excess load. The data collection method is carried out on road sections which include IRI values, vehicle load testing, CBR values and environmental conditions. The location of data collection is carried out on BTS Provincial Road. Surabaya City – Legundi. The analysis method for the prediction of the IRI value is the HDM-4 method, while to model the prediction of the increase in the IRI value is used exponential method with various parameters related to the results. The prediction results show that overload affects the increase in the IRI value. From the initial IRI value of 2.37, if the road is traversed by vehicles with standard loads for 20 years, the IRI value increases to 10.42. While the road traveled by overloaded vehicles within 20 years from an IRI value of 2.37 will increase to 16.88. With overload, the IRI value increases very sharply with the exponential model y=2.3583e0.0982x. While the standard load by generating an exponential model y=2.3825e0.0739x. From the results of the exponential model, load overload resulted in a significant increase in the IRI value with a growth rate of 0.0982. When compared to normal expenses which only experienced a growth rate of 0.0739. The modeling results show that overload will increase the growth of very high IRI values with the exponential model increase graph

International Roughness Index Prediction for Jointed Plain Concrete Pavements Using Regression and Machine Learning Techniques

Rigid pavements are recently gaining vast interest for their durability and large traffic volume endurance. This raises the need for a robust method to predict their long-term performance to support decisions on their future maintenance and rehabilitation requirements. One of the most descriptive parameters of pavement ride quality is the International Roughness Index (IRI). IRI can be used as a performance indicator for the level of serviceability of rigid pavement. This study presents the evaluation of 21 variables covering climate, traffic, and structural parameters in predicting IRI for jointed plain concrete pavement (JPCP) sections using 1,414 data points acquired from the Long-Term Pavement Performance (LTPP) database. It was found that only 10 variables are significant in the IRI prediction for JPCP. Moreover, four modeling techniques were applied, namely, linear regression, multivariate adaptive regression splines, Gaussian process regression, and artificial neural network. It was found that a deep ANN model with a structure of one input layer, three hidden layers, and one output layer [10-36-18-9-1] gives the highest IRI prediction accuracy with root mean square error and coefficient of determination (R2) of 0.117 and 0.92, respectively. Finally, a sensitivity analysis was performed to evaluate the impact of each input variable on the accuracy of IRI prediction for JPCP sections, and it was found that the most influential variables are initial IRI, pavement age, mean and standard deviation of humidity, standard deviation of evaporation, mean and standard deviation of freezing index, pavement compressive strength, ratio between pavement and base layers thickness, and finally an estimated parameter called site factor which is a function of (age, freezing index and soil percentage passing sieve no. 200).

Novel Soft-Computing Approach to Better Predict Flexible Pavement Roughness

Road infrastructures are fundamental parts of peoples’ lives, allowing them to access various destinations and activities. Accordingly, infrastructure should be in an appropriate condition. A pavement maintenance plan should be optimized, and pavement condition should be predicted accurately to obtain optimal pavement maintenance solutions. Therefore, the prediction of pavement conditions with high accuracy has been an immense concern. This study aims to introduce a new approach to accurately predict pavement international roughness index (IRI) over the long term. To this end, all the vital parameters, including initial IRI, pavement age, lane width, traffic loadings, structural characteristics, climatic features, and pavement distresses, are considered. With all the vital parameters, the prediction problem includes 58 variables. Thus, the application of a proper feature-selection technique is vital. To this end, a novel hybrid feature-selection method is introduced by a combination of arithmetic optimization algorithm and stochastic gradient descent regression (AOA-SGDR). Moreover, the performance of the proposed feature-selection method is compared with Lasso and all features. Five machine-learning algorithms, including random forest regression (RFR), support vector machine, multi-layer perceptron, decision-tree regression, and multiple linear regression, are employed for the prediction process. By employing AOA-SGDR, the average testing-data mean absolute error (MAE) reduces by at least 7.92%. Meanwhile, RFR provides the highest accuracy, with average testing-data MAE of 0.095 m/km. Moreover, analyzing the parameters indicates that initial IRI, pavement age, equivalent single axle load (ESAL), and structural number (SN) have the most significant relative influence on IRI.

International Roughness Index prediction for flexible pavements using novel machine learning techniques

International Roughness Index (IRI) is an important pavement performance indicator that is widely used to reflect existing pavement condition and ride quality. Due to the importance of this significant index, the current research aims to develop a precise IRI prediction model using the Gaussian Process Regression (GPR) and Locally Weighted Polynomials (LWP). The long-term pavement performance (LTPP) datasets of pavement age, initial IRI, alligator, longitudinal and transverse cracks, standard deviation of rutting, and subgrade plasticity index variables are employed in predicting IRI. These datasets are collected from 126 different flexible pavement sections of the LTPP specific pavement studies (SPS-1) located in different climatic zones in the US. The total number of IRI measurements in the collected database is 925 which covers a wide range of IRI values. Multiple linear regression (MLR) is firstly applied to classify the input variables. Then the MLR model is compared with four machine learning techniques which are GPR, LWP, Particle Swarm Optimization-Adaptive Network based Fuzzy Inference System (PSO-ANFIS) and PSO-Artificial Neural Networks (PSO-ANN). The developed models’ performance is validated using different statistical indices, including the coefficient of determination (R2). The results demonstrate that the GPR (R2=0.93) and LWP (R2=0.90) outperformed the PSO-ANFIS (R2=0.65) and PSO-ANN (R2=0.52) in predicting IRI. Thus, the GPR model is found to be more accurate for IRI modeling compared to the hybrid investigated models.

Contribution to the mechanistic-empirical roughness prediction in asphalt pavements

The guarantee of traffic conditions with safety and comfort is one of the primary aims during a road structure design. Thus,this study aims to contribute to the roughness prediction, represented by the International Roughness Index (IRI), bydeveloping models that describe the functional performance of Brazilian pavements. With monitoring actions carried out insites located in Santa Maria/RS and Rio de Janeiro/RJ, the paper sought to evaluate the IRI's evolution. The mechanisticmethodology was inserted to the study's scope by predicting the IRI evolution using the average fatigue damage, which wascalculated with FlexPAVETM software. The procedure enabled the development of a mechanistic-empirical model, basedon the initial IRI obtained after the end of the construction process and the computationally asphalt layers calculateddamage. The developed model was also validated with other sites within the Brazilian territory, which results in significantpractical applicability.

International Roughness Index Modeling For Jointed Plain Concrete Pavement Using Artificial Neural Network

Climate attributes such as precipitation, extreme temperature, and freeze-thaw cycles along with traffic loads cause pavement distresses. The maintenance need for pavements is decided based on the pavement condition rating such as International Roughness Index (IRI). Generally, an IRI rating less than 2.68 m/km is acceptable, and a rating greater than 2.68 m/km is considered unacceptable and classified as “very poor” condition of the pavement. It is imperative to be able to accurately predict pavement conditions to prepare proper Maintenance and Rehabilitation (M&R) programs for the pavements. This study aims to develop IRI models that can successfully estimate the IRI values for Jointed Plain Concrete Pavement (JPCP) considering the M&R history of the pavements using Artificial Neural Networks (ANNs) approach. The study was carried out with the database collected from Long Term Pavement Performance (LTPP) program. The variables used for the ANN model development are initial IRI, pavement age, concrete pavement thickness, equivalent single axle load (ESAL), climatic region (wet-freeze, wet non-freeze, dry-freeze, dry non-freeze), construction number (CN), and several climatological data. After utilizing various ANN model structures, the best performing ANN model resulted in promising statistical measures (i.e. R2 = 0.87). The IRI prediction model can successfully estimate the increase of IRI values with the increase of ESAL value over time. The IRI prediction model can also estimate the decrease of IRI value after maintenance and rehabilitation. The predicted IRI values with good accuracy will help the local and state agencies to prepare for M&R programs for JPCP pavements and allocate a projected budget accordingly.

Development of Roughness Prediction Models for Life Cycle Assessment Studies of Recycled Pavement Projects

The few existing life cycle assessment studies considering pavement recycling techniques usually omit the stages of maintenance and rehabilitation (M&R) and use. The reason for this omission is the lack of information about how the pavement’s performance evolves over time and absence of methods to determine the M&R frequency and service life for completed projects. As a result, the deterioration of pavement recycling projects in the long term is not clearly understood. Few projects have available data, the majority of which are on low volume primary and secondary roads. This paper describes an approach to develop a family of roughness models for recycling projects in Colorado using functional data analysis, and individual models for selected projects in Virginia to support ongoing life cycle assessment (LCA) studies. In the case of Colorado, full depth reclamation (FDR) projects will most likely deteriorate following an average group rate of 1.4 in./mi/year, with an initial international roughness index (IRI) between 52 and 70 in./mi. For the individual roughness models developed for Virginia projects, the initial IRI values and the rate of change for the treatments analyzed were found to range between 49 and 107 in./mi and between 0.7 and 5.2 in./mi/year, respectively, depending on the recycling method and type of stabilization treatment. The results of an LCA case study show that, in addition to recycling, Virginia Department of Transportation can achieve statewide emission reduction goals if focus is placed on achieving smoother roads while measures are taken to keep the annual rates of deterioration low.

Input Parameters for the Mechanistic-Empirical Design of Full-Depth Reclamation Projects

Cold recycling technologies such as full-depth reclamation (FDR) are sustainable and cost-effective techniques for pavement rehabilitation that reduce environmental impacts and construction costs and time. The limited information available on the material properties of FDR mixtures and their characterization in mechanistic-empirical (M-E) pavement design hinders the full deployment of FDR. Previous research has found current M-E default values to be non-representative and overly conservative, leading to an underestimation of the true performance capabilities of FDR materials. To address this gap, this paper analyzes the performance of 11 FDR sites constructed throughout Colorado, U.S., and compares their long-term performance with M-E predictions. The objective of this paper is to recommend input values for the M-E design of FDR base materials that result in reliable predictions of FDR long-term performance. The analysis includes both non-stabilized and emulsion-stabilized FDR projects. Both initial International Roughness Index (IRI) and resilient modulus were found to have a significant impact on M-E predictions and were calibrated in a two-step process. The proposed input parameters lead to a conservative design of FDR projects and result in improved IRI predictions compared with the ones derived from current design criteria. With the current design parameters, IRI predictions were, on average, overestimated by 51 in./mi, whereas the proposed input parameters make it possible to reduce this difference to 17 in./mi. Future research is needed to improve current models in M-E pavement design software to adequately model cold in-place recycled layers such as FDR.

A hybrid wavelet-optimally-pruned extreme learning machine model for the estimation of international roughness index of rigid pavements

ABSTRACT International Roughness Index (IRI) is a key parameter in pavement performance evaluation. This study investigates developing a reliable prediction model that can be used to estimate IRI of rigid pavements using innovative machine learning techniques. Optimally Pruned Extreme Learning Machine (OP-ELM) and Wavelet analysis are integrated to improve the OP-ELM results and design a novel hybrid Wavelet-OPELM (WOPELM) model for the IRI prediction. The proposed model is compared statistically to the OP-ELM and conventional feed-forward Artificial Neural Network (ANN) as well as regression model with respect to their efficiency to predict IRI of jointed plain concrete pavement (JPCP) sections in USA. The relevant data was collected from the Long-Term Pavement Performance (LTPP) database. Eight input variables, initial IRI, pavement age, transverse cracks, percent joints spalled, flexible and rigid patching areas, total joint faulting, freeze index, and percent subgrade passing No. 200 U.S. sieve, are assessed and used to predict the IRI. The results show that the initial IRI, total joint faulting, and freezing index are the most significant parameters for IRI prediction. The WOPELM is found to be a robust and more accurate modelling technique compared to OP-ELM, ANN, and regression for IRI prediction with only a 7% prediction error.

Life-Cycle Cost Analysis of Pay Adjustment for Initial Smoothness of Asphalt Pavement Overlay

Abstract This study aims at developing pay adjustments for initial smoothness of asphalt pavement overlay using life-cycle cost analysis (LCCA). The relationship between the initial International Roughness Index (IRI) and pavement life based on the threshold of surface distress index (SDI) was developed to calculate the pay adjustment using LCCA. It was found that the terminal IRI values when pavement overlay life was reached based on the SDI threshold of poor condition had an average value of 128 in./mile (2.02 m/km) with standard deviation of 15 in./mile (0.24 m/km). The pay adjustments were significantly affected by the assumption of analysis period in LCCA considering single or successive overlay applications. The Bayesian approach with Markov Chain Monte Carlo (MCMC) methods was used to develop the probabilistic model between the expected pavement life and the initial IRI. The probabilistic distribution of pay adjustments was presented when the uncertainty was considered with respect to the effect of initial IRI on pavement life and the assumption of overlay sequence. Compared with the pay adjustment obtained from the deterministic approach, the probabilistic analysis resulted in a similar pay adjustment for the case of single overlay but a greater pay adjustment for the case of successive overlays. The proposed analysis approach can be implemented by agencies in decision making of quality assurance based on actual practice of maintenance strategy.

Use of random forests regression for predicting IRI of asphalt pavements

Random forest is a powerful machine learning algorithm with demonstrated success. In this study, the authors developed a random forests regression (RFR) model to estimate the international roughness index (IRI) of flexible pavements from distress measurements, traffic, climatic, maintenance and structural data. To validate the model, more than 11,000 samples were collected from the database of long-term pavement performance (LTPP) program, with 80% randomly sampled data for training and 20% of them for testing the RFR model. The performance of the RFR model was then compared with that of the regularized linear regression model. The results showed that the RFR model significantly outperformed the linear regression model, with coefficients of determination (R2) greater than 0.95 in both the training and test sets. The variable importance score obtained from the RFR revealed that the initial IRI was the most important factor affecting the development of the IRI. In addition, the transverse cracking, fatigue cracking, rutting, annual average precipitation and service age had important influences on the IRI. Other distresses such as longitudinal cracking, edge cracking, aggregate polishing, and potholes exerted little impact on the evolution of the IRI.

Local Calibration of the MEPDG Distress and Performance Models for Ontario’s Flexible Roads: Overview, Impacts, and Reflection

Built upon a seven-year local calibration study of Ontario’s flexible pavements, this paper provides a summary of the calibration results and design impact and, more importantly, shares the experience and lessons learned from the process. The best results have been achieved on the local calibration of the rutting, bottom-up fatigue cracking, and international roughness index (IRI) distress models minimizing the residual sum of squares (RSS) while maintaining the average bias at zero. Significant efforts have been made to calibrate the other distress models with limited success. A design impact study found that local calibration of the rutting models was very important, whereas the alligator fatigue cracking did not usually govern the design in Ontario, although the global model was found to under-predict the cracking damage. The performance of the calibrated IRI model in the design of heavy traffic freeways for both reconstructed and rehabilitated sections was unsatisfactory and needs further study. The paper also presents several open questions for future research. These include the handling of section-length effects of observed cracking data, the determination of initial IRI, the updating of standard deviation functions and the overall reliability models, and the prioritization of pavement research under the new paradigm of the Mechanistic–Empirical Pavement Design Guide (MEPDG).

International Roughness Index Prediction Model for Thin Hot Mix Asphalt Overlay Treatment of Flexible Pavements

Pavement performance prediction after maintenance and rehabilitation is important to pavement management. A two-parameter exponential international roughness index (IRI) regression model for thin hot mix asphalt overlay was developed based on information from the U.S. Long Term Pavement Performance (LTPP) database. The model influence parameters α and β, which represent the initial IRI as the thin overlay completion and shape factor of IRI deterioration curve, were statistically analyzed. The results suggested that the IRI deterioration trends in high-temperature and low-temperature regions are different. This is because β was mainly affected by the structural strength and equivalent single axle loads in the high and medium temperature region and mainly affected by the average annual precipitation in low temperature region. In-situ data from LTPP database was used to verify the IRI prediction model, and it was found that the predicted IRI and measured IRI exhibited similar trends.

International Roughness Index prediction model for flexible pavements

International Roughness Index (IRI) is a pavement performance indicator which reflects not only the pavement condition but also the ride quality and comfort level of road users. The aim of this paper is to develop an accurate IRI prediction model for flexible pavements using both multiple linear regression analysis and artificial neural networks (ANNs). The models were developed based on the Long-Term Pavement Performance Database. The data were collected for both original and overlaid flexible pavements from the general pavement studies (GPS-1, GPS-2 and GPS-6) and the specific pavement studies (SPS-1, SPS-3 and SPS-5). The final database consisted of 506 sections with 2439 observations. The proposed models (regression and ANNs) predict IRI as a function of pavement age, initial IRI (IRI just after pavement construction), transverse cracks, alligator cracks and standard deviation of the rut depth. The regression model yielded a coefficient of determination (R2) value of 0.57 while the ANNs model resulted in a much higher R2 value of 0.75.

Life-cycle cost analysis of optimal timing of pavement preservation

Optimal application of pavement preservation or preventive maintenance is critical for highway agencies to allocate the limited budget for different treatments. This study developed an integrated life-cycle cost analysis (LCCA) model to quantify the impact of pavement preservation on agency cost and vehicle operation cost (VOC) and analyzed the optimal timing of preservation treatments. The international roughness index (IRI) data were extracted from the long-term pavement performance (LTPP) program specific pavement studies 3 (SPS-3) to determine the long-term effectiveness of preservation treatments on IRI deterioration. The traffic loading and the initial IRI value significantly affects life extension and the benefit of agency cost caused by pavement preservation. The benefit in VOC is one to two orders greater in magnitude as compared to the benefit in agency cost. The optimal timing calculated based on VOC is always earlier than the optimal timing calculated based on agency cost. There are considerable differences among the optimal timing of three preservation treatments.

The Data Mining Applied for the Prediction of Highway Roughness due to Overloaded Trucks

Currently, the quality of the Indonesian national road network is inadequate due to several constraints, including overcapacity and overloaded trucks. The high deterioration rate of the road infrastructure in developing countries along with major budgetary restrictions and high growth in traffic have led to an emerging need for improving the perfo rmance of the highway maintenance system. However, the high number of intervening factors and their complex effects require advanced tools to successfully solve this problem. The high learning capabilities of Data Mining (DM) are a powerful solution to thi s problem. In the past, these tools have been successfully applied to solve complex and multi -dimensional problems in various scientific fields. Therefore, it is expected that DM can be used to analyze the large amount of data regarding the pavement and traffic, identify the relationship between variables, and provide information regarding the prediction of the data. In this paper, we present a new approach to predict the International Roughness Index (IRI) of pavement based on DM techniques. DM was used to analyze the initial IRI data, including age, Equivalent Single Axle Load (ESAL), crack, potholes, rutting, and long cracks. This model was developed and verified using data from an Integrated Indonesia Road Management System (IIRMS) that was measured with the National Association of Australian State Road Authorities ( NAASRA) roughness meter. The results of the proposed approach are compared with the IIRMS analytical model adapted to the IRI, and the advantages of the new approach are highlighted. We show that the novel data -driven model is able to learn (with high accuracy) the complex relationships between the IRI and the contributing factors of overloaded trucks .

Design Factors Affecting the Initial Roughness of Asphalt Pavements

Past studies have shown that initial pavement roughness greatly affects future pavement roughness and roughness progression rate. Highway agencies typically use incentives/disincentives to control the initial roughness of pavements. However, the current blanket specifications do not account for inherent factors that affect initial roughness, such as design factors. In other words, pavements are designed to have different initial roughness. This study analyzes the design factors affecting the initial roughness of asphalt-surfaced pavements. An initial international roughness index (IRI) of 442 asphalt-surfaced pavements constructed from 2000 to 2004 was analyzed based on analysis of covariates. The factors considered in this study include hot mix asphalt (HMA) layer thickness, project location (urban vs. rural), base type, HMA mix classification, and pavement length. The statistically significant factors affecting initial pavement roughness were identified and discussed. These factors can be taken into account to develop specifications for initial roughness of asphalt pavements.

Probabilistic Performance-Related Specifications Methodology Based on Mechanistic–Empirical Pavement Design Guide

In current construction practice, as-built asphalt concrete acceptance is typically based on several key quality assurance–quality control parameters. These parameters have pay factors that are empirically established by judgment of the pavement community. The pay factor has been used to reward or penalize the contractor for construction quality. These properties are known to influence pavement performance, but the specific effect on performance and project life has been impossible to estimate. Because of this, there is an urgent need to tie deviations of the field production from the specified job mix to performance and pavement life. This research focused on the integration of AASHTO's Mechanistic–Empirical Pavement Design Guide (NCHRP 1-37A and NCHRP 1-40D) with the Simple Performance Test (NCHRP 9-19) to develop a probabilistic quality specification for quality assurance of hot-mix asphalt (HMA) construction. This has been the major objective of NCHRP 9-22. The methodology is based on relating the HMA dynamic modulus, through a closed-form solution, to major pavement distress. A Monte Carlo simulation is run by using the project mix design to predict the as-designed distress and its associated variability. Then the remaining service life of the pavement is predicted from the as-designed distress. Similarly, Monte Carlo simulation is run for the as-built material to estimate the remaining service life. The difference between the as-built and the as-design distress provides the predicted difference in quality of construction from the mix design. The difference is used to calculate the pay factor for each distress. The total pay factor is the sum of the individual distress pay factors. Finally, the initial international roughness index, representing the ride quality pay factor, is added.

Evaluation of international roughness index for asphalt overlays placed over cracked and seated concrete pavements

Crack, seat and overlay (CS&O) is a rehabilitation technique that has been used on jointed plain concrete pavements. Only a few studies have evaluated the surface roughness of pavement sections rehabilitated using this technique. The purpose of this paper was to evaluate the roughness of roadway sections rehabilitated employing the CS&O technique. The data extracted from the long-term pavement performance database were separated based on the weather region and analysed. Then, international roughness index (IRI) prediction models were developed. A separate model was developed for pavement sections in California. For sections with bound bases, thick overlays provide a smoother surface (lower IRI). However, the effect of the overlay thickness on the IRI for sections with unbound bases does not appear to be considerable. Prediction models developed in this study are shown to provide adequate predictive capabilities. Sections in California have initial IRI values that are lower than those found for sections in the wet-with-freeze (WF) and wet-with-no-freeze (WNF) regions. However, California sections are predicted to develop higher IRI values over time, when compared to those sections in the WF and WNF regions.