Pavement Deterioration

Pavements are susceptible to accelerated deterioration due to changing climate conditions, leading to increased maintenance and excess fuel consumption through pavement-vehicle interaction. China's diverse climates raise concerns about the environmental and economic sustainability of flexible pavements amid climate change and the effectiveness of preventative maintenance strategies. This study examines climate change's potential impacts on long-term pavement performance, greenhouse gas (GHG) emissions, and costs, employing the Mechanistic-Empirical Pavement Design Guide method. The calibrated World Bank's Highway Development and Management Model 4 assesses the impacts of surface characteristics on vehicle fuel consumption. Life cycle assessment and life cycle cost analysis quantify GHG emissions and costs. Results reveal significant pavement deterioration in Southeast and Central China. Preventative maintenance strategies reduce fuel consumption, with GHG emissions and cost savings from smoother driving conditions outweighing those from maintenance. These insights stress the importance of proactive maintenance strategies for mitigating climate-induced deterioration and enhancing sustainability.

Louisiana Department of Transportation and Development (DOTD) pavement preservation program consists of various pavement preservation treatments, including single, double, and triple chip seals. Such treatments are applied mainly to collectors and local roads. Louisiana DOTD has spent substantial financial resources on various rehabilitation and maintenance treatments to minimize the pavement distresses and improve the pavement life. However, the effectiveness of any treatment largely depends on the time of treatment and trigger governed by treatment performance models. This study focuses on the performance evaluation of chip seal treatment on flexible pavements using the before and after treatment performance data. The treatment benefits in terms of remaining service life, service life extension, and treatment life were evaluated using the treatment transition matrix. Such analysis yielded effective treatment trigger values for each distress type. In addition, the before and after treatment data were utilized to develop IRI and transverse cracking prediction models for chip seal treatment on flexible pavements.

Unpaved roads consist of considerable portions of the roadway network in many countries. These roads play crucial roles in the development of infrastructure systems, advancements of socio-economic activities and improvements of the agricultural and production sectors. Thereby, unpaved roads benefit the underdeveloped, rural, and remote neighborhoods, and act as lifelines for these geographically disadvantaged communities. Frequent and regular maintenance activities keep the roadway system operational at a desired level of service. Resurfacing is one of the major maintenance treatments for unpaved roads. A gravel loss prediction model (GLPM) can evaluate the impacts of varying magnitude of resurfacing treatments on the roadway performance. Thus, a GLPM can provide valuable insights for roadway maintenance budget scheduling and decision-making tasks. In this paper, the backgrounds, input requirements, and output results of three popular GLPMs were reviewed. These models were (i) Highway Development and Management Model 4 (HDM-4), (ii) South African Technical Recommendation for Highways Model 20 (TRH-20), and (iii) Australian Road Research Board (ARRB) Model. In addition, the practicality of roadway resurfacing frequency charts which were developed based on these models was also evaluated. This study determined that the existing GLPMs and the corresponding roadway resurfacing frequency charts were often unreliable and impractical. In this study, a beta regression (BR) analysis methodology was utilized to develop and calibrate a GLPM for Iowa. Because of its simplified yet effective nature, the BR model outperformed the popular GLPMs and offered a practical approach to quantify annual roadway gravel loss.

A methodology was proposed that can improve pavement smoothness prediction. As an initial attempt, smoothness due to solely transverse cracking was investigated. The new model incorporated amplitudes and wavelengths of transverse crack which are directly related to IRI computation. The crack amplitude and wavelength were measured by automated profiler from the MnRoad test roads and formulized with mathematical form expressed with respect to various crack depths and widths, respectively. Instead of conventional severity criteria, crack severities could be determined by crack depth and width in the mathematical form. Utilizing digital signal processing techniques, cracked surfaces with various severities were successfully prepared. The computed IRI values were found to be significantly related to crack severities expressed by width and depth. Then, an IRI prediction model was derived containing the simulated various crack severities and crack occurrences. Ultimately, the IRI model was validated using transverse crack performance data from MnRoad. The model incorporated spectral information that could fundamentally and accurately predict IRI increment by crack. This confirmed the necessity of incorporation of amplitude and wavelength information in predicting IRI.

Chip Seals are a type of pavement preservation treatment characterised by its ability to reduce pavement deterioration and restore surface condition. Chip seal performance, like other pavement preservation treatments, depends on several factors such as environmental conditions, traffic level and the condition of the existing pavement structure prior to treatment application. In cold regions, the presence of freeze and thaw cycles promotes the generation of thermal cracks and possible tenting distresses, consequently, pavement roughness has been a prominent parameter to assess treatment performance. The objective of this study was to evaluate the roughness progression on several chip seal configurations in a cold climate region, utilising a set of field performance data from the Pavement Preservation Group Study, a broader research study conducted by the National Center for Asphalt Technology and the Minnesota Department of Transportation’s Road Research Facility. The main findings indicate that chip seals provide a negligible effect on immediate roughness improvement; however, on a long-term basis, they provide a reduction in IRI progression over time. The magnitude of the IRI benefit is dependent on the type of chip seal implemented and the traffic level experienced.

Machine learning algorithms are powerful AI tools that have demonstrated strong robustness and reliability in the performance prediction of road infrastructure. In this study, the authors have attempted to develop a prediction model to estimate the transverse cracking in jointed plain cement concrete pavements using three machine learning approaches, namely, decision tree regression (DTR), random forest regression (RFR), and deep neural network (DNN). The results show that the DNN model outperformed DTR and RFR with coefficients of determination (R²) greater than 0.95 in both training and testing data sets. Performance metrics are summarised and presented for all three methods used in this study.

This study presents a machine learning model for predicting representative surface distresses (crack, durability, patching, joint spall) in concrete pavements, focusing on South Korean examples. It thoroughly analyzes specific distress types using time series data to understand their development over time, aiming to surpass traditional regression methods in forecasting pavement conditions. The research fills a gap by applying machine learning algorithms to detailed long-term data, enhancing the accuracy of distress progression predictions, which is crucial for efficient pavement management. A notable aspect of this study is the use of particle filtering, recognized for its effective resampling in analyzing time series data. To validate predictions, we compared the results from particle filtering with those from traditional regression models, long short-term memory (LSTM) networks, and Deep Neural Networks (DNNs). The accuracy varied significantly, with differences ranging from 3.32% to 23.64%, indicating particle filtering’s suitability for time-series-based pavement condition predictions. These findings are especially relevant in the context of current image-based machine learning and AI research in pavement distress detection and prediction. This research offers a comprehensive reference that is especially valuable due to the lack of studies using long-term usage data, thereby making a significant contribution to pavement management research and practice.

Surveying the condition of the pavement is one of the most important processes in managing the road network. The information collected during these surveys allows for the calculation of the Pavement Condition Index, which is a derivative cumulative qualitative indicator that evaluates various pavement characteristics and defects. Deterioration modelling of these measured indicators and calculated indices is a critical element and its most accurate prediction brings the process of pavement management closer to a higher quality process by more efficiently allocating funds and repair work.

Many different models – both extremely complex and simple – are used in the world to simulate the condition of individual pavement indicators. However, these models are developed based on the data of a certain country or region and are not suitable in another country due to different requirements for pavement structures and other reasons. In Lithuania, measurements of the quality indicators of road surfaces with new generation survey equipment have been carried out recently but the information stored in the databases about road sections is minimal, and it becomes difficult to adapt the models applied abroad due to the missing information. The aim of this study is to create pavement condition index prediction models by evaluating such quantitative and qualitative indicators as traffic loads, road surface unevenness, type of repair, pavement age, climatic zones, and pavement construction classes.

Under the present guidelines, the road roughness has been considered as prime indicator for selecting the road stretches for maintenance works and first priority is given to worst road sections. It is found that the existing maintenance policies no longer provide the required results, cause discomfort to the road users, and result in various accidents and frequent traffic jams. Various newly road surfaced stretches from recently declared National Highways of Madhya Pradesh: NH 59, NH59-A, State Highway-18 were selected for the analysis purpose and data was collected as per the requirement by carrying out field studies which includes, evaluation of pavement conditions through field work, crust details, road inventory survey, structural evaluation, functional evaluation, traffic volume survey data, evaluation of pavement material through laboratory work, photographic survey for various distress conditions. The study indicates that roads recently surfaced with thin renewals were prematurely failed and could not serve the designed life. Study also suggest that new renewals provided prior major rehabilitation without framing different distress conditions and avoiding mandatory investigations required to find out root causes of failures would result in complete waste of money and leads to disaster. The result also shows that routine maintenance approach with thin renewals prematurely failed with in two years and starts from rating good with initial roughness 1.8 meters/km and ends in completely road failure and required emergency reconstruction.

India’s growing demand for adequate maintenance and upkeep of the world's second-largest road network of 5.89 million kilometers is a major problem for road administrators and policymakers. To address the issue of proper upkeep of road networks, many countries have implemented a robust pavement management system (PMS) to handle the tasks related to pavement maintenance and management. The Highway Development and Management (HDM-4) system is widely used as a tool for highway investment and maintenance planning and programming. The broad objective of this study is to develop pavement deterioration models for cracking, ravelling, potholes, rut depth, and edge break using non-linear regression techniques in MATLAB and compare them with the models present in HDM-4 and observed deteriorations for their effectiveness. The deterioration models for urban and rural pavement sections have been developed based on the large volume of field data collected in the Jaipur district of Rajasthan State using automated as well as manual methods of field evaluation. These sections were continuously monitored for 5 years for the pavement surface condition data. The validity of these models and calibrated HDM-4 models was assessed by examining the distress predictions generated by the regression models and calibrated deterioration models to the distress observed on the selected pavement sections. The proposed pavement deterioration models and the calibrated HDM-4 models are likely to apply to other developing nations with comparable traffic patterns, soil types, meteorological conditions, terrain kinds, and pavement composition as well.

With improvements in data collection, storage, and processing, machine learning (ML) is gaining momentum as a behavior prediction method in the field of engineering. Several studies have evaluated these algorithms’ potential to predict pavement serviceability, however some challenges limit its use. Training data preprocessing has a great impact on the model’s predictive performance, is highly dependent on the modeler’s experience, and is not typically reported in engineering-related literature. The objective of this study was to assess the effects of data preprocessing, hyperparameter selection, and time series size on the model’s evaluation metrics. Therefore, this paper analyzes the performance of three ML algorithms on maximum deflection (D₀) and international roughness index (IRI) prediction: support vector machine, random forest (RF), and artificial neural network (ANN). An R² and mean square error (MSE) analysis was conducted on 12 training datasets, with two sizes of historical data and five stages of data preprocessing. The results indicated that ANN was the most accurate technique with an R² of 0.99 and MSE of 20 ×10⁻³ mm on the D₀ prediction and an R² of 0.91 and MSE of 0.03 m/km on the IRI prediction. RF was also identified as an effective technique, generating similar results with less data preprocessing. The addition of structural and traffic categorical features to the training dataset resulted in the most significant improvement of the support vector regression and ANN performance metrics; the hyperparameter selection was effective only on IRI prediction, especially with the ANN algorithm.

Asset management of pavement network requires understanding of pavement deterioration rate for cost-effective maintenance and adequate budget allocation. The pavement industry has recognized the challenge of uncertainty or variation in deterioration processes that could not be captured by deterministic deterioration models. This study investigated the stochastic Markov chain theory for modeling deterioration of pavement network. The discrete condition data for the Markov model is obtained by a proposed maintenance-related condition rating scheme (MRCR) that combines three commonly inspected pavement distresses including cracking, rutting and roughness. The Markov model is calibrated by the proven Bayesian Markov chain Monte Carlo simulation method, and the statistical Chi-square test is used for testing model fitness. A case study with time series data of pavement distresses collected from regular inspection of a highway network is used in this study. Various influential factors to pavement deterioration are also investigated in this study to understand their impact on the deterioration rate of highways. The results on the case study show that the Markov model is suitable for modeling deterioration of highway network, and there are significant differences in deterioration rates of highways among influential factors including traffic volume, rainfall amount, demographic location, and prioritized maintenance. The outcomes of this study provide more understanding of pavement deterioration of road networks and demonstrate the forecasting of maintenance budget by the deterioration prediction of Markov model for supporting asset management of pavement network.

This report documents a project designed to update the current Austroads road deterioration (RD) models using the long-term pavement performance and long-term pavement performance maintenance (LTPP/LTPPM) dataset and other data available, such as the traffic speed deflectometer (TSD) datasets, to improve these models’ explanatory power. A proof-of-concept approach based on using the available TSD datasets undertaken early in the project determined that these datasets were not appropriate at this stage for developing an updated rutting RD model.

The rutting RD model was based on a mechanistic-deterministic approach using a multi-variate non-linear regression analysis. The rutting RD model uses cumulative rutting as the dependent variable for thinly surfaced flexible unbound granular pavements using the Austroads LTPP/LTPPM dataset collected from 1994 to 2018. The rutting RD model may need some calibration to suit locally observed rutting conditions using a calibration coefficient, Kr.

The model also has the capacity to be adapted to surface maintenance treatments other than double/double seals to determine their impact on cumulative rutting by using relative performance factors for surface maintenance treatments. The model was validated using an independent dataset.

This report documents a project designed to update the current Austroads road deterioration (RD) models using the long-term pavement performance and long-term pavement performance maintenance (LTPP/LTPPM) dataset and other data available, such as the traffic speed deflectometer (TSD) datasets, to improve these models’ explanatory power. A proof-of-concept approach based on using the available TSD datasets was undertaken early in the project which determined that these datasets were not appropriate at this stage for developing a refined structural RD model.

Development of a structural RD model was therefore based on the extensive time series captured by the LTPP/LTPPM datasets to produce a single structural RD model for flexible sprayed sealed pavements. The structural RD model was based on a mechanistic‑empirical deterministic approach using a multi-variate non‑linear regression (MVNLR) analysis. Seasonal correction of the measured maximum deflection proved to be problematic using measures of the lower soil moisture content, so the corrections were made using the current Austroads approach to seasonal variation.

This report documents a project designed to update the current Austroads road deterioration (RD) models using the long-term pavement performance and long-term pavement performance maintenance (LTPP/LTPPM) dataset and other data available, such as the traffic speed deflectometer (TSD) datasets, to improve these models’ explanatory power. A proof-of-concept approach based on using the available TSD datasets undertaken early in the project determined that these datasets were not appropriate at this stage for developing an updated roughness RD model.

The roughness RD model documented in this report was based on a mechanistic-deterministic approach using a multi-variate non‑linear regression analysis. The roughness RD model uses cumulative roughness as the dependent variable for thinly surfaced flexible unbound granular pavements based on the Austroads LTPP/LTPPM dataset collected from 1994 to 2018. The roughness RD model may need some calibration to suit locally observed roughness conditions.

The model also has the capacity to be adapted to surface maintenance treatments other than double/double seals to determine their impact on cumulative roughness by using relative performance factors for surface maintenance treatments. The model was validated using an independent dataset.

This report documents a project designed to update the current Austroads road deterioration (RD) models using the long-term pavement performance and long-term pavement performance maintenance (LTPP/LTPPM) dataset and other data available, such as the traffic sped deflectometer (TSD) datasets, to improve these models’ explanatory power.

The cracking RD model documented in this report was based on a mechanistic-empirical, deterministic approach using a multi-variate non-linear regression (MVNLR) analysis. The cracking RD model uses cumulative cracking as the dependent variable for thinly surfaced flexible unbound granular pavements. The model was developed using the New South Wales dataset collected by the TSD from 2014 to 2018. The cracking RD model will need calibration to suit locally observed cracking conditions as location has a bearing on the rates of cracking progression.

Highway Authorities in the UK use Surface Condition Assessment for the National Network of Roads (SCANNER) in assessing and managing their road networks. This survey vehicle utilises laser measurements to detect and quantify most of the distress on the road surface, such as rutting, cracking and texture depth. It is however a data intensive and expensive approach since it is conducted annually. This study presents a simple method to predict pavement distress using previous SCANNER measurements. The previous measurements are used to develop Distress Deterioration Master Curves (DDMC) that relate distress deterioration rate with the severity of the distress. These curves can be used to predict future distress severity based on the current state without the need to provide further data such as pavement age or pavement material properties. To demonstrate the application of this method, a significant amount of SCANNER data covering around 400 km of class A roads in Nottinghamshire collected between 2014 and 2020 were analysed, and rutting, crack intensity, and texture depth were modelled in this study. DDMRs of these distress types were built based on data collected between 2014-2018, then 2020 data were used to validate the predictions. The results show that the developed method can be implemented in predicting surface distress of roads using previous measurements, which makes it a valuable addition tool for highway authorities subject to underfunding.

Transverse thermal Cracking is one of the major distress modes manifesting in Uganda’s
bituminous pavements. It occurs when thermal stress builds up due temperature changes as result
of a series of hot and wet climatic conditions, and thus the tensile strength of the pavement is
exceeded (A. Shalaby, 1998). Failure of bituminous roads in Kampala has been partly attributed
towards failure to do adequate timely maintenance on these roads in Kampala reducing the service
life. This is attributed to the inability to detect paved road deterioration in time which starts with
the manifestation of cracks such as transverse cracking that propagates from the bottom to the top
of the pavement surface as a result of temperature changes. This research investigated the
progression of transverse thermal cracking on bituminous roads in Kampala. The Research was
carried out between March and September 2022. Relationships between Transverse Thermal
cracking and pavement surface age, for selected bituminous roads in Kampala, were established
based on field observations, HDM-4 model for initiation and progression of Transverse thermal
Cracking. The Following features were considered; number of transverse thermal cracks(No/Km),
time to initiation of Transverse thermal cracks(years), coefficient of Transverse Thermal cracking,
Construction Defects indicator for bituminous Surfacings, calibration factors for both initiation
and progression of transverse Thermal cracking, pavement surface age, time since crack initiation
to reach maximum number of cracks and fraction of analysis year in which transverse thermal
cracking progression applies. A Kcpt value of 2 was found to best represent the rate of progression
of transverse thermal cracks on Bituminous roads in Uganda. This means that the rate of transverse
thermal crack progression in Uganda is twice that predicted by the HDM-4 equation of prediction
of transverse thermal crack progression.

Thesis looking at predicting reflective cracking for urban pavements.

Study looking at the initiation and progression of cracking.

Pavement degradation prediction is essential for road management systems to predict the most effective maintenance time. The existing degradation prediction models are calibrated for different countries; therefore, they cannot be used directly for the local condition to predict the maintenance requirements. In Sri Lanka, this is causing many difficulties during the maintenance activities of road infrastructure and drainage systems running along the roadways. Therefore, in this study, a pavement degradation model for unpaved road infrastructure in Sri Lanka is proposed using Markov analysis. The analysis consists of identifying performance parameters, condition stages of the road sections, Transition Probability Matrix (TPM), and finally prediction of the road lifetime and level of maintenance requirements. Developed Unpaved Condition Index (DUPCI) is proposed for the unpaved roads as the performance parameter which includes all the possible deterioration types of gravel roads such as potholes, corrugations, overexposed aggregates, erosion, roadside drainage, and rutting. Development of the TPM is proposed based on the data collected at three gravel roads in Dambulla Pradeshiya Sabha. Calibrated HDM3 model is proposed for Sri Lankan paved roads based on the data collected at A9 road. Both models were calibrated and validated using the collected data for Sri Lankan unpaved and paved roads.

The number of potholes in the world has rapidly increased due to the growth of vehicles, temperature changes, and the concentration of the population. Potholes cause danger in driving and reduce passengers' comfort. Therefore, an accurate prediction of number of potholes provides timely maintenance and rehabilitation, and also it enhances safety for drivers. This study aims to improve the accuracy of number of potholes prediction model by considering independent variables such as minimum temperature, relative humidity, precipitation, and traffic volume. The model was established by conducting variable analysis. Various machine learning methods were then employed to develop an optimal model that provides the highest accuracy in predicting pothole occurrence. The study also suggests a computer vision-based system for spotting potholes based on the image segmentation method, followed by calculating the damage ratio. The results confirm that the proposed models have the potential in predicting and detecting pothole occurrence.

Understanding the relationship between pavement raveling and traffic characteristics is important to pavement management and maintenance planning. In this work, we propose a framework to empirically quantify this relationship. It consists of an alignment method to tackle the inconsistent spatial-temporal scales of the raveling and traffic measurements and we propose spatial-temporal maps to qualitatively analyze and compare the data. A non-parametric correlation is done on the aligned raveling and traffic flow data. This framework is applied to five study areas in the Dutch highway network. The correlation analysis of the study areas provides empirical evidence to a commonly held theory that traffic flow has effects on raveling. Categorizing the correlation by lanes indicates that the raveling is homogeneous in the through or auxiliary lanes, and the severe raveled sections are parallel to the road discontinuity, suggesting the potential effect of mandatory lane changing on raveling development. The proposed framework can be employed in empirical raveling models that predict raveling based on traffic and other factors.

In this research, a methodology was developed to optimize the design of Warm Mix Asphalt (WMA) with the inclusion of three recycled materials as partial replacement of natural aggregates (NAs), namely Crumb Rubber (CR), Reclaimed Asphalt Pavement (RAP), and Recycled Concrete Aggregate (RCA). The methodological proposal is composed of 4 sections denominated: (I) environmental module, (II) economic module, (III) decision-support module, and (IV) results report module. Initially, a Life Cycle Assessment (LCA) is carried out to quantify the environmental impacts associated with WMA production. Similarly, in the second module, a Life Cycle Costing (LCC) is performed to estimate the financial investment required by the process under evaluation. Meanwhile, a computational model based on genetic algorithms (GAs) is created in the decision-support module to execute multi-objective optimization (minimization of costs and contaminating potential). In the last module, the more accurate WMA designs are presented employing a Pareto front, ternary plot, composition pie chart, and statistical analysis of the influence of the CR, RAP, and RCA on the validation criteria. This study concludes that even under long hauling distances and huge prices, it is possible to design WMA with CR, RAP, and/or RCA additions that form sustainability benefits compared to conventional WMA.

Nice paper from Bosnia-Herzegovenia on a roughness prediction model that was developed from field data.

The development of a linear mixed model to describe the degradation of friction on flexible road pavements to be included in pavement management systems is the aim of this study. It also aims at showing that, at the network level, factors such as temperature, rainfall, hypsometry, type of layer, and geometric alignment features may influence the degradation of friction throughout time. A dataset from six districts of Portugal with 7204 sections was made available by the Ascendi Concession highway network. Linear mixed models with random effects in the intercept were developed for the two-level and three-level datasets involving time, section and district. While the three-level models are region-specific, the two-level models offer the possibility to be adopted to other areas. For both levels, two approaches were made: One integrating into the model only the variables inherent to traffic and climate conditions and the other including also the factors intrinsic to the highway characteristics. The prediction accuracy of the model was improved when the variables hypsometry, geometrical features, and type of layer were considered. Therefore, accurate predictions for friction evolution throughout time are available to assist the network manager to optimize the overall level of road safety

The gravel road pavement has a lower construction cost but poorer performance than the asphalt surface. It also emits dust and deforms under the impact of vehicle loads and ambient air factors. The resulting ripples and ruts are constantly deepening, increasing vehicle vibrations and fuel consumption, reducing safe driving speed and comfort. In this article, existing pavement quality evaluation indexes are analysed, and a methodology for their adaptation for roads with gravel pavement is proposed. This article reports the measured wave depth and length of the gravel pavement profile by the straightedge method of a 160 m long road section in three road exploitation stages. The measured pavement elevation was processed according to ISO 8608, and vehicle frequency response has been investigated using simulations in MATLAB/Simulink. The applied International Roughness Index (IRI) analysis showed that a speed of 30-45 km/h instead of 80 km/h provides the objective results of IRI calculation on the flexible pavement due to a decreasing velocity of vehicle's unsprung mass on a more deteriorated road pavement state. The influence of the corrugation phenomenon of gravel pavement has been explored, identifying specific driving safety and comfort cases. Finally, an increase in the Dynamic Load Coefficient (DLC) at a low speed of 30 km/h on the most deteriorated pavement and a high speed of 90 km/h on the middle-quality pavement demonstrates the demand for timely gravel pavement maintenance and the complicated prediction of a safe driving speed for drivers.

he International Roughness Index (IRI) has become the reference scale for assessing pavement roughness in many highway agencies worldwide. This research aims to develop two Artificial Neural Network (ANN) models for Double Bituminous Surface Treatment (DBST) and Asphalt Concrete (AC) pavement sections using Laos Pavement Management System (PMS) database for National Road Network (NRN). The final database consisted of 269 and 122 observations covering 1850 km of DBST NRN and 718 km of AC NRN, respectively. The proposed models predict IRI as a function of
pavement age and Cumulative Equivalent Single-Axle Load (CESAL). The obtained data were randomly divided into training (70%), validation (15%), and testing (15%) datasets. The statistical evaluation results of the training dataset reveal that
both ANN models (DBST and AC) have good prediction ability with high values of coefficient of determination (R2 = 0.96 and 0.94) and low values of Mean Absolute Error (MAE = 0.23 and 0.19) and Mean Squared Percentage Error (RMSPE =
7.03 and 9.98). Eventually, the goodness of fit of the proposed ANN models was compared with the Multiple Linear Regression (MLR) models previously developed under the same conditions. The results show that ANN models yielded higher
prediction accuracy than MLR models.

Large quantities of waste generated in the municipal, commercial and industrial and construction and demolition sectors have caused widespread environmental issues. The replacement of virgin materials with recycled in pavement construction is a possible solution for waste management and achieving sustainability goals in the infrastructure sector. There are, however, questions about environmental and economic impacts of waste-derived materials in road construction that need to be answered. Life cycle assessment and life cycle cost analysis are two approaches to quantify and assess the environmental performance and the costs of decisions regarding the selection of materials for pavement construction. While considerable research has been conducted on pavement materials, the impacts of particular materials such as recycled concrete aggregates, lignin, waste plastic, recycled glass, crushed brick and crumb rubber are not currently well understood. This research presents a synthesis of the state of the art of selected recycled materials in pavement construction and limitations of existing environmental and economic analysis. A major interest towards recycling of materials and necessity of their sustainability analysis is highlighted. The results indicate that the sustainability analysis of selected recycled materials is in its infancy with considerable inconsistencies, hindering the meaningful comparison of results. Furthermore, exclusion of impacts of maintenance, usage and end of life phases from sustainability analysis, impose uncertainty on the long-term viability of these materials. Further research is needed to develop better understanding of these impacts so that more informed decisions could be made by policy makers.

With the pressing need to improve the poorly rated transportation infrastructure, asset managers leverage predictive maintenance strategies to lower the life cycle costs while maximizing or maintaining the performance of highways. Hence, the limitations of prediction models can highly impact prioritizing maintenance tasks and allocating budget. This study aims to investigate the potential of different predictive models in reaching an effective and efficient maintenance plan. This paper reviews the literature on predictive analytics for a set of highway assets. It also highlights the gaps and limitations of the current methodologies, such as subjective assumptions and simplifications applied in deterministic and probabilistic approaches. This article additionally discusses how these shortcomings impact the application and accuracy of the methods, and how advanced predictive analytics can mitigate the challenges. In this review, we discuss how advancements in technologies coupled with ever-increasing computing power are creating opportunities for a paradigm shift in predictive analytics. We also propose new research directions including the application of advanced machine learning to develop extensible and scalable prediction models and leveraging emerging sensing technologies for collecting, storing and analyzing the data. Finally, we addressed future directions of predictive analysis associated with the data-rich era that will potentially help transportation agencies to become information-rich.

Paper investigating the implications of utility cuts on pavement roughness.

Excellent PhD dissertation on developing structural deterioration models for flexible pavements for the Queensland road network using Traffic Speed Deflectometer (TSD) data collected over a five-years period by the Department of Transport and Main Roads (TMR), Queensland. 

The research reported in the thesis, considers the experimentation conducted on the road connecting Gonder to Debark were analyzed to evaluate the functional characteristics of pavement and to investigate relationship between IRI and falling weight deflection with pavement distress such as cracking, potholes, rutting and raveling. The data collected with the help of road surface tester (RST) vehicle is obtained with the help of ERA authority, in addition to rutting (RUT), cracking (CRA), pothole (POT) and raveling (RAV).The IRI is evaluated over equally spaced traveled along the road profile, and roughness measurement performed at speed between 80 km/hr., as per standards of AASHTO. Mintab 17 is adopted for developing models for predictions. The first model was roughness model and it had a prediction power of R2 of 0.834. The developed model was adopted for investigation between roughness and easily measurable distress namely cracking, rutting, raveling and pothole. The deflection model of R2 of 0.851. The independent variables of deflection model are the same as roughness model.

Many countries have large unsealed road networks which are essential for business efficiency, social connectedness, and community safety. These roads are maintained at considerable cost mainly through blading and re-gravelling. The latter is a major component of the maintenance and is determined by gravel loss. Gravel loss prediction models can be used to assess the effect of different wearing courses, which provides useful input into the design and management of unsealed roads. This paper reviews the origins, input parameters, and output of four available gravel loss prediction models, namely, the Transport and Road Research, HDM-4, Australian, and South African models. It was found that the predictive accuracy of the models is in general low and they predict very different gravel loss results. There is also a lack of integration between the design and maintenance, leading to the wearing course properties recommended for design not being directly linked to gravel loss. These findings led to further analysis of the models and the development of re-gravelling frequencies based on traffic, climate (annual rainfall), and material property (plastic factor), which can be used in the selection of the appropriate wearing course material and in the determination of re-gravelling budgets. This presents a simplified approach to the use of existing gravel loss prediction models in the design and management of unsealed roads which mitigates some of the shortcomings in the use of uncalibrated models.

Gravel loss is the change in thickness i.e. reduction in thickness of gravel roads surfacing over a period of time. For effective maintenance management of gravel road, a gravel loss prediction model works as a tool in selecting the optimal re-gravelling schedule. As gravel loss can be caused by many factors, for formulating a prediction model here average daily traffic (ADT) representing Traffic, absolute gradient (G) representing geometric design feature, mean monthly precipitation (MMP) representing climatic factors, plasticity index (PI), gradation of the aggregate (P20) representing surface material quality and the duration of observation in terms of day (D) are used as the model independent variable. The dependent variable of the model, gravel loss, are collected by observing the selected gravel roads in six month interval and the independent variables included in the model are gathered using standard procedures and methods. The selected roads for the study are NuwakotAsurkot-Pyuthan Road (Arghakhanchi district), Argha-Dharampani-Maidan Road (Arghakhanchi district), Sahid BasudevMarg,Ambhanjyang Road(Makawanpur district), Kabahigodh-Piparadi-Patarhati Road(Bara district), SonbarsaGadi-Sakhuwa-Parsauni– Mahuwan-Ramnagari Road (Parsa district), Bindawasni-Bairia-Birta Road (Parsa district), Bahurwabhatta-Pokharia, Padam Road (Parsa district). In each road a 60m of longitudinal grid are considered which are further divided into 10 m interval where elevation across the width of the road are observed using auto level. Model is developed using SPSS which will be helpful to find the residual life and appropriate time for re-gravelling.

The World Bank HDM 4 model is adopted in many countries worldwide. It is consisted of the developed models for almost all types of deformation on the pavement structures, but it can’t be used as it is developed everywhere in the world without proper adjustments to local conditions such as traffic load, climate, construction specificities, maintenance level etc. This paper presents the results of the researches carried out in Macedonia for determining calibration coefficient of the rutting model in HDM 4.

Pavement roughness is one of essential performance indicators that are used in road maintenance. A model was developed in this study to obtain roughness value from easily measurable distress values, namely cracking and potholes, for low-volume roads in India. The data collected at 173 in-service flexible pavements were utilised for model development. Using the model developed in this study, a satisfactory roughness value can indirectly be obtained from the cracking and potholing data, even without the use of a roughness measuring device.

This paper presents comparisons of the environmental impacts and life cycle costs of various alternative strategies for a portland cement concrete (PCC) pavement project in Manitoba to demonstrate the opportunity to optimize the cost, pavement performance and environmental impacts. A matrix of 10 different strategies that include alternative PCC mix, pavement design, and maintenance and rehabilitation (M&R) treatments have been used to contrast both the life cycle costs and environmental impacts with MI’s past practice (base case). The presented analysis is expected to assist highway agencies to better understand and weigh not only the economics of alternative strategies, but also the environmental implications of alternative roadway materials, design, construction, and maintenance and rehabilitation practices.

The aim of this project was to clarify the effectiveness of pavement maintenance (preservation) activities, in the form of periodic maintenance and rehabilitation, on pavement condition and distress (roughness, rutting and cracking) deterioration rates. This was addressed by estimating trends in pavement deterioration in three jurisdictions (New South Wales, Queensland and Victoria) from a time series of observational data (supplied by the jurisdictions) using mostly iPave condition and deflection data. In order to assess the effectiveness of the various pavement maintenance treatments, a comparison of these observational trends with historical and predicted rates of deterioration was made.

The main findings of the study were as follows:

• A comparison between the observed deterioration rates derived from the time series of observational data with historically-derived rates and the Austroads RD model estimates suggested that the three approaches produced comparable results in terms of roughness and rutting deterioration, but not for cracking.
• Based on the historical rates of roughness and rutting deterioration representing pre-treatment deterioration, the post-treatment roughness deterioration rates in NSW and Victoria were reduced by a range of between 8% (seals) and 58% (OGA), demonstrating the effectiveness of the surface treatments.

Comparisons of functional condition parameters (roughness, rutting and cracking) against the mean maximum deflection found that their deterioration rates were significantly influenced by pavement strength. The effect of traffic and climate on deterioration was not as strong as pavement strength.

A reliable pavement performance prediction model is needed for road infrastructure asset management systems or pavement management systems. In this study, the data on roughness progression of asphalt pavements in the long-term pavement performance (LTPP) database was analyzed in order to develop such a model. The international roughness index (IRI) is a reasonable measure of the ride comfort
perceived by occupants of passenger cars and hence used as the basis for the pavement performance prediction model developed in this research. A quantitative relationship between roughness progression and accumulative traffic load, structural number, annual precipitation, and freezing index was developed and validated. Five pavement performance levels were developed to express the extent of asphalt
pavement deterioration. This is coupled with a reliability analysis based on the Weibull model to estimate the remaining service life of asphalt pavements. Effective treatments of pavements at the project level for each condition state level were also proposed, which can aid network level optimization of the overall condition and corresponding budget allocations.

This is a very interesting and important report which presents the results of a major study looking into the impact of vehicle loading on pavement deterioration, in particular, the '4th power law' damage factor relevance for lower standard pavements. Highly recommended.

Pavement management at a network level requires reliable accurate performance prediction models to help road agencies make useful complex decisions about highways maintenance and rehabilitating activities. The purpose of this paper is to report the approach adopted for model development and validation for heavy duty flexible pavements representing by seven rural freeways segments. Hierarchical generalized linear modelling approach has been applied to predict multilevel model to capture the effect of variations among time series data, among road sections and among highways with same duty pavements. The estimation of pavement cracking progression has been based on longitudinal dataset contain cracking data (reported as a percent of the affected area) as dependent variable and cumulative traffic loading, pavement strength and environmental conditions as independent variables. The study illustrates how panel data can be nested to predict the probability of crack progression to capture the effect of significant unobserved heterogeneity. The significance of relevant contributing factors in predicting crack progression were presented and elucidated.The validation results indicate that the model replicates the pavement behavior well, and that the inclusion of additional factors in addition to time is improving the model prediction.

In this research paper, prediction model is developed for the progression of roughness, which is the most important performance indicator of flexible pavements. Since many stretches of flexible pavements in village roads in India are not exposed to routine maintenance for years together due to the paucity of the funds and models are not available for predicting the performance of the pavements under such conditions, this research is focused to develop prediction model for the roughness of flexible pavements exposed to least or nil routine maintenance. Roughness data were collected from the selected in- service pavements (171 stretches in Tamil Nadu state in India) and model is developed using stepwise regression analysis. The model has been validated with independent field data. The versatile roughness prediction model developed in this study will be useful for practicing engineers in managing the flexible pavements in low volume roads exposed to least or nil routine maintenance.

Bituminous surfacing seals are used on a high percentage of the southern African road network to protect the mostly granular pavement base layers and to provide wet weather skid resistance, appropriate for the conditions at hand. A study has been designed to empirically model crack initiation and texture loss to assist with the development and calibration of a Finite Element Model (FEM) for seals. Thirty five road sections have been selected throughout South Africa covering different seal and binder types (Single seals, multiple stone seals and Cape seals), age of seal, traffic volume and climatic region for performance investigation. Two samples were taken from each site (In the wheel-path and outside the wheelpath) to also evaluate the effect of traffic on ageing of the binder. In addition to this the performance of different seals on more than six hundred road sections, over a period of fourteen years, has been evaluated to quantify the effect of binder type and film thickness on crack reflection. A synthesis of the key performance variables has led to the development of survivor curves for different seal types, which is a strategic output of the study This paper provides an overview of findings related to the long term performance of seals in the South African environment. Conclusions are drawn regarding the contribution of different factors influencing crack initiation and texture loss.

Impact of local distresses on roughness.

 

Report addressing the impacts dust emissions from unsealed roads have on people, and identifying environmentally sustainable and financially cost effective mitigation measures likely to be effective at reducing those impacts.

 

This report describes a two-year study designed to quantify the immediate and longer-term maintenance impact of grader blading and surface re-sheeting on unsealed roads.  The project assembled and analysed roughness data collected by Cassowary Coast Regional Council in Queensland, Blayney Shire Council in New South Wales and Moorabool Shire Council in central Victoria to expand the current works effects (WE) models to cover a wider range of traffic and climatic conditions and to validate the existing unsealed road roughness deterioration (RD) model. WE models were developed for light blading, medium blading and granular re-sheeting maintenance works and a RD model was developed for roughness progression between maintenance activities.  The suggested modifications to the RD and WE models should assist local government asset managers in their management of unsealed roads. It is expected that the models could be adapted to the varying local conditions of unsealed roads in other locations.

The objectives of this study are to: (1) prediction of pavement distress such as low temperature cracking, (2) estimate different types of user costs incurred by pavement roughness resulting from distresses, (3) compare agency investments for different maintenance and rehabilitation strategies and associated roughness-related user costs, (4) analyze environmental impacts of construction, maintenance, and rehabilitation (CMR) activities used in pavement engineering, (5) estimate and compare agency costs, user costs due to roughness, and emission costs due to CMR activities, and; (6) estimate emission costs associated with pavement roughness. By considering the cost associated with the environmental impact of CMR activities, a more realistic estimate of the ROI associated with maintaining relatively smooth pavement throughout its service life was assessed.

It is quite essential to investigate the causes of pavement deterioration in order to select the proper maintenance technique. The objective of this study was to identify factors cause deterioration of recently constructed roads in Khartoum state. A comprehensive literature concerning the factors of road deterioration, common road defects and their causes were reviewed. Three major road projects with different deterioration reasons were selected for this study. The investigation involved field survey and laboratory testing on those projects to examine the existing pavement conditions. The results revealed that the roads investigated experienced severe failures in the forms of cracks, potholes, and rutting in the wheel path. The causes of those failures were found mainly linked to poor drainage, traffic overloading, expansive subgrade soils, and the use of low quality materials in construction. Based on the results, recommendations were provided to help highway engineers in selecting the most effective repair techniques for specific kinds of distresses.

This paper describes briefly the state-of-the-art in terms of rutting models. Some of the
models are analysed by comparing rutting evolution prediction for a set of representative
Portuguese pavements structures and traffic conditions. HDM-4 deterioration model was
considered to be the most promising to implement in a new Portuguese Maintenance
Optimisation System (MOS).

This paper describes the development of a cracking prediction model for Portuguese conditions which is expected to integrate the Pavement Management System (PMS) of Estradas de Portugal. The World Bank’s highway development and management (versions III and 4) and PARIS models are used as reference for the development of a deterministic (mechanistic-empirical) model, using pavement condition data from sections of the main road network. A two-phase distress evolution model is proposed where the initiation of cracking (1st phase) is ruled by a different equation than the progression of cracking (2nd phase). Cracking initiation is predicted on a traffic basis, from the annual traffic load and the structural capacity of the pavement. An absolute model is presented and recommended for the maintenance and rehabilitation (M&R) programming in the long-term and for the analysis of non-cracked segments. Absolute and relative type models were obtained for cracking progression. The relative model shows better agreement to data and is proposed for short- to medium-term analysis on segments with cracking history, while the absolute model is proposed for the M&R programming in the long-term and the analysis of non-cracked segments. Finally, the recommended model is evaluated based on the application to a set of pavement structures defined in the Portuguese pavement design guide.

Flushing is the process whereby chipseal texture depth is lost over time, resulting in a loss of skid resistance. It is the single most important reason for resealing on New Zealand state highways.

This report details research carried out from 2012 to 2015. In the first part of the work the aim was to identify and investigate the physical mechanisms causing flushing. The aim of the second part of the project, undertaken by researchers at the University of Auckland, was to use pavement condition data to develop a model to predict the rate of flushing progression in chipseals.

 

Factors making a major contribution to flushing are:

• aggregate abrasion and breakdown
• compaction and reorientation of the seal layer under traffic
• water venting and sub-surface stripping in seal layers.

Factors having no or making only a minor contribution to flushing are:

• thermal expansion of the bitumen
• excess bitumen application
• binder viscosity.

 

Further work is needed to quantify the significance of chip embedment into the basecourse.

A two-part model using parameters in the NZ Transport Agency Long-Term Pavement Performance database was developed. The first part uses a logistic model to predict the onset of flushing and an accuracy of 74% when used to predict the initiation of flushing on a separate data set.

The second part uses a linear model to predict the rate of flushing progression. First-coat seals, and second and higher generation seals were modelled separately.

The linear model was statistically strong (R2 of 0.445 for first-coat seals and 0.628 for second and higher generation seals).

Flushing is the process whereby chipseal texture depth is lost over time, resulting in a loss of skid resistance. It is the single most important reason for resealing on New Zealand state highways.  This report details research carried out from 2012 to 2015. In the first part of the work the aim was to identify and investigate the physical mechanisms causing flushing. The aim of the second part of the project, undertaken by researchers at the University of Auckland, was to use pavement condition data to develop a model to predict the rate of flushing progression in chipseals.

Flushing is a defect which has a damaging effect on the functional performance of sprayed seal (chipseal) pavements. Accurate understanding of flushing can have a significant impact when predicting the future performance and maintenance needs of pavements. The reported study was conducted to develop a prediction model to effectively identify, asses and manage flushed pavements. The study also aimed to develop a decision-making tool for treating flushed pavements. This study utilised pavement data from New Zealand’s Long-Term Pavement Performance programme and data analysis was conducted to develop a model to predict the flushing potential of chipseal pavements. Additionally, the study conducted laboratory testing on pavement samples from flushed chipseal pavements. The conducted tests included wheel tracking and rutting measurements, air void volume measurements, as well as computed tomography scanning and image analysis. The laboratory test results were used to supplement the outputs of the performance prediction model in detecting the mechanisms that were causing flushing. The outcomes of this study included a model that was able to predict a) the probability of flushing initiation, and b) the quantity of flushing on a pavement. This model was statistically robust where the flushing initiation model had an accuracy of 76%. The flushing prediction model and the laboratory results were incorporated into an overall pavement assessment guideline for flushed pavements. This assessment guideline will aid pavement practitioners with accurate identification of flushing on a pavement network as well as with selecting the best method of maintenance treatment for flushed chipseal pavements.

The fatigue cracking behaviour of laboratory prepared chipseal beams and beams cut from field samples was studied using a four-point bending test method. Preliminary results indicate that chipseal fatigue lives at 5ºC are up to eight times greater than those of estimated values for asphalt mix under the same loading conditions. The results suggest binder oxidation was not the dominant factor in seal cracking and that cracking in the field may be primarily due to very high, localised deformations. Such deformations may arise through weak basecourse patches formed during construction or more likely, from water damage (to both the basecourse and seal structure itself) arising from leaking seals.

Data from long-term pavement performance sites show that overall, the average number of cracks initiated per site increased approximately linearly from the time of crack initiation. The average annual increase in crack length is approximately half the crack length, so as the crack grows the rate of crack growth in mm/year increases. A brief analysis was carried out for two sites that showed an approximately three-year lag between crack initiation and pothole formation.

The report proposes practice guidelines and the outline of a performance-based specification for the crack repair of chipseals.

This report details research carried out from 2012 to 2015 into chipseal flushing. The physical mechanisms causing flushing were investigated and a model was developed to predict the growth of flushing over the New Zealand state highway network. Factors making a major contribution to flushing are: • aggregate abrasion and breakdown • compaction and reorientation of the seal layer under traffic • water venting and sub-surface stripping in seal layers. Factors having no or making only a minor contribution to flushing are: • thermal expansion of the bitumen • excess bitumen application • binder viscosity. Further work is needed to quantify the significance of chip embedment into the basecourse. A two-part model using parameters in the NZ Transport Agency Long-Term Pavement Performance database was developed. The first part uses a logistic model to predict the onset of flushing and an accuracy of 74% when used to predict the initiation of flushing on a separate data set. The second part uses a linear model to predict the rate of flushing progression. First-coat seals, and second and higher generation seals were modelled separately.

Prediction models for low volume village roads in India are developed to evaluate the progression of different types of distress such as roughness, cracking, and potholes. Even though the Government of India is investing huge quantum of money on road construction every year, poor control over the quality of road construction and its subsequent maintenance is leading to the faster road deterioration. In this regard, it is essential that scientific maintenance procedures are to be evolved on the basis of performance of low volume flexible pavements. Considering the above, an attempt has beenmade in this research endeavor to develop prediction models to understand the progression of roughness, cracking, and potholes in flexible pavements exposed to least or nil routine maintenance. Distress data were collected from the low volume rural roads covering about 173 stretches spread across Tamil Nadu state in India. Based on the above collected data, distress prediction models have been developed using multiple linear regression analysis. Further, the models have been validated using independent field data. It can be concluded that the models developed in this study can serve as useful tools for the practicing engineers maintaining flexible pavements on low volume roads.

India is developing her national highway network through widening and rehabilitation of existing highways along with the construction of expressways in different phases, since 1999. Unprecedented growth of road traffic, high variations in pavement temperature and need of long lasting pavements have increased the use of modified bitumen specifically in wearing courses of many flexible pavement road sections of national highway network in entire country. Crumb rubber-modified bitumen (CRMB) and polymer-modified bitumen (PMB) of different grades are mostly used modified binders under different climatic and environmental conditions. During the design life, bituminous road sections show different rates of initiation and propagation of distresses under varying traffic and climatic conditions. In this study, an effort has been made to calibrate the internationally recognised Highway Development & Management (HDM-4) road deterioration models for the selected flexible pavement sections over time with traffic. The different road distresses are modelled using HDM-4 tool for the newly constructed flexible pavement sections of Indian national highway network having modified binder in bituminous concrete (BC) mixes which are located in different regions of the country. Pavement condition data of 23 in-service flexible pavement sections were collected for three consecutive years starting from 2011 to end of the year 2013. Data collected from the study were analysed for calibration and validation of HDM-4 distress models for similar climatic conditions, pavement compositions and traffic loading characteristics. The results of this study are useful for developing pavement maintenance management strategies for Indian national highway network with similar climatic conditions, pavement compositions and traffic characteristics.

The objectives of this study are to simulate the Thornthwaite moisture index (TMI) for zones within the Atlantic provinces of Canada (APC) during three 30-year periods in the 21st century and to estimate the interactive effect of TMI and simulated freight traffic loads on the deterioration of pavement structure during the same period. Regional Highways 1, 2, 7, 15, 16, 102, and 104 connecting the APC are considered as the case study. Integration of spatial input-output and transportation models simulates freight movements on the selected regional highways during the period of 2012–2100. TMI is estimated using downscaled average monthly precipitation and temperature at 34 stations within the APC. Simulated traffic loads and TMI are applied to mechanistic modeling of roughness progression on the pavement structure. The findings of this study show that an increase in TMI can cause 11–68% increase of roughness progress rate on pavement structure.

Thermal cracking of pavements may occur due to the top of the pavement being exposed to cold atmospheric conditions. The formation of cracks may cause a deterioration of ride quality and ingress of water to the base. Developing of capabilities to model thermal cracking of pavements is beneficial because it will allow the assessment of susceptibility of pavement materials and geometries to thermal cracking. Fracture resistant asphalt mixtures can then be identified. A methodology is proposed by which numerical modelling of thermal cracking of pavements is carried out using the distinct element method and cohesive cracks. The distinct element method efficiently handles the subdivision of the originally intact material. The cohesive crack method allows the inclusion of experimentally determinable fracture properties into the formulation. Results obtained were consistent with an analytical model available in literature. Thermal cracking observed in a field experiment could then be numerically replicated.

Deterioration models allow road managers to assess current condition and to predict future conditions of their road networks. Due to heavy vehicle axle repetitions and the effect of environmental factors, permanent deformation (rutting) develops gradually in the wheel paths and impacts on structural and surface performance of flexible pavements. This paper reports the approach adopted to develop absolute deterministic models for permanent deformation of low volume roads. A representative large sample network (23 highways) of light duty pavements was selected. For each section, time series data from four consecutive condition surveys were collected. Multiple regression analysis was carried out to develop models to predict pavement rutting progression over time as a function of a number of contributing variables. They include traffic loading, pavement strength, climate and drainage condition. For more powerful prediction, family group data-fitting approach was utilised to estimate future rutting progression based on the average rut depth curve for a series of pavements with similar characteristics. This study highlighted that separate family deterioration models are preferred and needed for more realistic results. The paper concludes that the analysis approach used for developing the models confirmed their accuracy and reliability by well-fitting to the validation data with low standard error values. Also, study results show that higher traffic loading, lower pavement strength, poor drainage and climates with high seasonal variation contribute to increasing rutting progression rate.

This report documents research to further develop probabilistic modelling work undertaken in 2012 and demonstrate the impact the variations in predictions of the main dependent variable, roughness, has on road agency costs (RAC) and vehicle operating costs (VOC) in a pavement life-cycle costing analysis.

This phase of work used a pilot application of a three road network with each road having different levels of traffic and pavement performance outcomes. The pilot study showed that the RAC estimates are highly sensitive to the percentile roughness estimates compared to the VOC estimates because during the pavement life-cycle costing analysis the roughness levels are contained to within 4 IRI by the intervention of rehabilitation.

The current Austroads approach to assess the relative damaging effects of different axle groups on road pavements is by comparison of the peak static pavement deflection response under the axle groups. The assumption that deflection is the most appropriate indicator of pavement damage is open to question and is not consistent with the use of strains to calculate the performance of pavement materials.

In response, research conducted has determined that, with regard to the fatigue damage of asphalt and cemented materials, the standard load for an axle group type is dependent upon the thickness and modulus of the asphalt and the underlying pavement structure.

As a result, it is proposed that the mechanistic design procedure for flexible pavements not use the concept of standard loads, but rather that the procedure determines the pavement damage resulting from each axle load and each axle group within a traffic load distribution. An examination of the implications of pavement design outcomes in using this method determined that in general, reductions in both asphalt and cemented material thicknesses of up to 50 mm would result.

The research also determined that the currently used standard loads for tandem, triaxle and quad-axle were appropriate for use with the current empirical procedures for the design of granular pavements with thin bituminous surfacings.

This report details a study conducted using detailed roughness measurements on a panel of unsealed local roads located in Moorabool Shire Council at both the pre- and post-works effect (WE) treatment stages. The aim of the study was to develop interim WE models and validate any existing interim WE models, as well as validate current road deterioration (RD) models for roughness. Simple WE models were developed with the independent variable, IRIb, the roughness before treatment, which explained 59% to 91% of the dependent variable, IRIa, the roughness after treatment, depending on the type of works effect being modelled. No validation of any existing RD roughness model was possible due to the limited material and environmental variations.

A reliable pavement performance prediction model is needed for road infrastructure asset management systems or pavement management systems. In this study, the data on roughness progression of asphalt pavements in the long-term pavement performance (LTPP) database was analyzed in order to develop such a model. The international roughness index (IRI) is a reasonable measure of the ride comfort perceived by occupants of passenger cars and hence used as the basis for the pavement performance prediction model developed in this research. A quantitative relationship between roughness progression and accumulative traffic load, structural number, annual precipitation, and freezing index was developed and validated. Five pavement performance levels were developed to express the extent of asphalt pavement deterioration. This is coupled with a reliability analysis based on the Weibull model to estimate the remaining service life of asphalt pavements. Effective treatments of pavements at the project level for each condition state level were also proposed, which can aid network level optimization of the overall condition and corresponding budget allocations.

Highway agencies worldwide strive to ensure that highway users pay fees that not only recover the costs of pavement damage but also are equitable. In addressing the limitations of past research and quantifying the resulting adverse consequences on their analysis outcomes, this paper presents a comprehensive framework to derive more representative estimates of pavement damage cost. The developed framework incorporates practical pavement repair schedules that include all the key repair categories as a basis for estimating the marginal pavement damage cost (MPDC). The framework was applied to pavements of different surface type, functional class and age. On average, the MPDC was found to range from $0.0032 per ESAL-mile on Interstate highways to $0.1124 per ESAL-mile on non-national highways. It was determined that in each highway functional class, the marginal cost of pavement damage is influenced significantly by the pavement material type, traffic levels and age. Within any specific functional class, it was determined that the marginal cost increases with increasing traffic level and pavement age. The study also determined that non-consideration of at least one key repair category such as reconstruction or routine maintenance leads to significant (27–45%) underestimation of the actual MPDC.

Paper describing statistical analysis of road condition data using machine learning methods.

Over the lifetime of an asphalt concrete (AC) pavement, the roadway requires periodic resurfacing and rehabilitation to provide acceptable performance. The most popular resurfacing method is an asphalt overlay over the existing roadway. In the design of asphalt overlays, the thickness is related to the structural strength of the existing pavement. As the layers are overlaid their stru ctural
characteristics change due to aging of asphalt. However, currently there is no method to determine the effect of aging on the strength of existing pavement layers. 

This study examined structural characterization of six pavement test sections in Kansas using three different test methods: Falling Weight Deflectometer (FWD), Portable Seismic Property Analyzer (PSPA), and Indirect Tensile (IDT) tests. The results were analyzed to determine how the modulus of an AC pavement layer changes over time. The results indicate that as the AC pavement ages, its modulus decreases due to pavement deterioration. The most prominent cause for AC pavement deterioration was observed to be stripping. Two of the test sections on US-169 and K-4 showed little signs of stripping and had a minimal reduction or even an increase in AC moduli. 

The analyzed results from different test methods for moduli were inconclusive as far as any correlation among the methods is concerned. While the correlation between various test methods studied was mostly consistent for a particular roadway, no universal correlation was found for all pavement sections tested.

Fatigue test results show that older pavement layers have a higher propensity for fatigue failure than the newer layers. However, some older pavement layers showed excellent fatigue life. Fatigue results correlated well with the condition of the cores as assessed by visual observation.

The research presented in this thesis investigated the occurrence of flushing of chip seal pavements. The research aimed to determine the effects of volumetric properties of chip seal surfaces on flushing, focusing on the relationship between air voids and flushing initiation. Additionally, the research aimed to develop a pavement deterioration model to predict the initiation and progression of flushing on chip seal pavements. Mechanical testing was conducted on cores obtained from in-service, flushed chip seal pavements from Auckland, Waikato, Christchurch and Dunedin regions of New Zealand. The tests that were performed on the cores included wheel tracking, air voids measurement, bitumen extractions and sieve analyses to determine aggregate grading profiles. Measurements were made of the depth and pattern of deformation that had developed on the cores during wheel tracking. Samples extracted from tested cores were scanned using a computed tomography scanner and the scan images were analysed using image analysis techniques to calculate the volume of air voids within the samples. The reduction in air void volume during wheel tracking was compared to the flushing that had occurred on the samples to establish the relationship between flushing and air voids. Furthermore, data analysis was performed on pavement condition data gathered from New Zealand’s long-term pavement performance database to identify the combination of factors that provided the best prediction of flushing, and regression analysis was performed to develop a model to predict the initiation and progression of flushing. The laboratory testing described above revealed that the thickness of a chip seal surface has a direct correlation to flushing, where thicker surfaces tend to have more severe flushing. Moreover, the reduction in air void volume that occurred in a chip seal structure due to loading was directly related to the amount of flushing likely to occur on that surface. The pattern of deformation of a chip seal structure provided an indication of its state of stability, which in turn indicated the best method of treatment for flushing. From the data analysis it was identified that the combination of factors that provided the best indication of flushing were surface thickness, surface age, rutdepth and grade of aggregates. The flushing initiation model had an accuracy of 76% and the flushing progression model was also statistically strong at predicting the quantity of flushing. Using these research outcomes, a pavement condition assessment guideline was developed to aid with managing flushed pavements. Overall, this research has significantly increased the understanding of the mechanisms that lead to flushing and established ways to better identify and manage flushing.

Pavement deterioration is a serious problem for road and traffic highway sector in Jordan. The allocated cost for construction of new roads, replacement and rehabilitation, and maintenance was 292.1 M JD in the implementation program (2007-2009) and 192. 2 M JD in (2011-2013). The current research aims to describe the most affecting causes for road deterioration in Jordan by a questionnaire designed and directed to contractors in road construction and maintenance. A list of causes was prepared through conducting literature review, consulting and interviewing a group of 15 managers from contractors and experts in the field, they advised to study 51 of expected causes for road deterioration. Then a questionnaire was prepared and directed to 150 of contractors in road construction and maintenance. The mission was involving to give a scale (rank) from 1
(strongly disagree), 2 (disagree), 3 (do not know), 4 (agree), and 5 (strongly agree) to the expected causes. 38 (25.33%) responses were received and analyzed. The criterion (defects caused during construction due to poor construction quality) takes the highest rank of 4.15, while the lowest factor is (Inadequate resistance to polishing of surface aggregate) of 2.73. Also, the causes for deterioration were reorganized into 11 consistent groups of relevant causes. The first group (Effect of Cracks and Structural Failure) ranks 3.96 and the last group (Effect of Pavement Width) ranks 2.93 in group comparison. The research is helpful in highlighting the causes for road deterioration in Jordan, and in avoiding of these causes or mitigating their effect during design, construction, and
maintenance through operation.

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 and the environment, and the structure and surface of the pavement. Performance models relating to the pavement distress conditions like, cracking, raveling, potholing, and roughness are analyzed and developed by various researchers. But most of these models are found applicable to a particular set of traffic or environment conditions, thus highlighting the need of model(s) that can work in varied conditions satisfactorily. The paper presents a detailed review of various pavement performance models to examine the role of factors related to pavement materials, environmental conditions, type of traffic and volume of traffic, and to identify the limitations and gaps in the present knowledge on such models.

Deterioration models allow predicting pavement condition and the development of maintenance programs. Normally, when evaluating pavement performance through model deterioration, the forecast given is a determinist value. However, pavement engineering projects, as any engineering poject, have a degree of uncertainty. This implies that an adequate performance of the engineering solution cannot be absolutely guaranteed.

The aim of this research is to incorporate probability in the output of a structural cracking model. To achieve this objective the model of crack initiation and progression of HDM-4 was used under several scenarios defined based on geographic location, type of traffic and structural capacity of 86 roads located in Chile. The input data for each scenario were obtained from the Ministry of Public Works of Chile and calibration studies of deterioration models to local conditions. To incorporate probability in the structural cracking models, a simulation model that reproduced the deterioration due to cracking for a lifecycle of 25 years was developed, based on random input data sets. With the set of outputs of the simulation, probability density functions that represent the probabilistic response of the deterioration model were developed.

The main output of this research is a set of probability density functions of cracking initiation and cracking progression of all structural cracks and wide cracks of 14 groups of Chilean roads. Although the research was carried out using Chilean data, the methodology presented could be applied to other states or countries.

This study was undertaken to establish whether various pavement deterioration models incorporated into the New Zealand – Deighton’s Total Infrastructure Management System (NZ-dTIMS) and Highway Design and Maintenance Model 4 (HDM-4) pavement management systems could be modified to reliably predict the condition of a pavement after it had been exposed to sudden extreme traffic loading, which can occur with the commencement of forestry logging, mining or enforced detours. Specifically, the deterioration models considered were for roughness progression (both the NZ-dTIMS and HDM-4 models) and rutting progression (NZ-dTIMS model only).

The objective of this research (undertaken 2008–2011) was the development of an improved method of modelling the decision to rehabilitate a typical New Zealand thin-surfaced unbound granular pavement. This was driven by previous research that had found a poor correlation between the data recorded in the road asset maintenance management (RAMM) database and the decision to rehabilitate. It had been hoped that by talking to local engineers and examining pavements proposed for rehabilitation, distress not currently recorded may be identified. This would have then driven the development of better models and may also have expanded the detail collected in the visual surveys. The research found, however, that the drivers are not obvious and that the decision maybe being based on factors such as the engineers’ assessment of the risk of rapid failure.

The conclusions from this research are:

• According to a visual engineering inspection, many pavement sections require rehabilitation.

• In many cases, a significant quantum of deferred maintenance needs to be performed for the do-minimum option. This maintenance is not necessarily obvious from the data in RAMM or visual observations of the high-speed data videos.

• The methods used to determine future maintenance costs vary widely. This ranges from including the deferred maintenance cost into one year and extrapolating from this cost, to ignoring the deferred maintenance cost in the analysis.

• The timeframe for assessing maintenance cost history is variable.

• The net present value (NPV) calculation can be very sensitive to assumptions made on future maintenance and seal lives. This includes assuming that higher priced polymer-modified seals need to be used.

• Rutting and flushing at 88% and 80% of the surveyed pavements are the two most commonly quoted distress mechanisms. These do not appear in a proposed rehabilitation algorithm.

• Digouts are a factor mentioned in 55% of justifications.

• The inspection length associated with the visual pavement inspection did not reflect the treatment section length in 40% of sites.

• The influence of non-engineering factors, such as concerns over ‘consuming the asset’ and fears of rapid pavement failure, need to be investigated.

• The difference in condition between the typical pavements in a network and those chosen for rehabilitation can often be minor and thus very difficult to quantify.

• Better guidelines should be developed to assist and standardise the decision process. These guidelines need to be based on a risk and consequence approach, which, it is believed, will better reflect the engineers’ approach.

Report on the establishment of pavement deterioration models for Norway.

Pavement surface performances have a great influence on road functionality and can affect user’s safety, vehicle operational costs, environmental sustainability. The assessment of evolution of pavement surface performance plays a fundamental role in road pavement management and is useful in order to allocation of maintenance resources.

In the light of the above, Authors introduced the first results carried out from a two-year monitoring of an experimental road section. Four different dense graded wearing courses were designed with different aggregates: limestone, basalt and expanded clay. Several surface performances were measured by different devices (Skid Tester, Sand Patch Test, Laser Profilometer).

Report describing framework for predicting pavement deterioration models for asset management.

In this paper, the authors present a method to automate data collection and mapping of pavement roughness on various functional classification roadways and to quantify fuel consumption for enhanced benefit-to-cost analysis, asset management and infrastructure investment strategies. A new low-cost and efficient technology is presented to measure and contrast vehicles' vertical accelerations over smooth and rough pavement sections. Fuel consumption is then correlated with the vehicle's vertical acceleration over both pavement conditions. The results of the study validate the relationship between acceleration/road roughness and fuel consumption. Finally, using the described technology, a pothole detection algorithm is devised to automatically map road roughness, a step towards a better asset management system.

In the life-cycle prediction of road pavement, it needs the model that should be able to predict the expected change of condition in the future. The model should consider current condition, pavement strength and age characteristics, environment, incremental time and incremental traffic. The aim of this study is to application the traffic simulation model for predicting initiation and progression of crack on road pavement. The aim of the study can be achieved by developing a computer simulation model that can predict road deterioration. The research develop coefficient of each models that agree with local condition based on observed data that collected for 1.5 years. These models are able to predict progression of cracking with R2=0.5925 to 0.8765 more appropriate than the existing model (R2=0.304 to 0.314). The coefficient of crack initiation model has difference with the existing models that are 5.7% to 20% for asphalt mix on asphalt pavement, 2.8% to 14% for asphalt mix on stabilized base, 1.6% to 2.2% for asphalt mix on granular base. While progression of cracking are 5.7% to 20% for asphalt mix on asphalt pavement, 2.8% up to 14% for asphalt mix on stabilized base, 1.6% to 2.2% for asphalt mix on granular base. In addition, the cracking model can be used as guidance for maintenance intervention criteria.

Series of reports from the NordFoU project which developed deterioration models for nordic countries.

Report presenting the results of extended pavement performance trials including various types of bitumen seal, clay and concrete bricks and unreinforced concrete, using TRL and Austroads design standards.

This research project is the second report detailing findings from the NZ Long Term Pavement Performance (LTPP) programme. This programme includes the monitoring of 63 sections on the State Highways and 82 sections on local authority roads. This report details all work related to developing a rutting model for New Zealand conditions. Previous work highlighted some data limitations in the LTTP programme – some of this development work relied on the Transit CAPTIF accelerated pavement testing programme.

This research project also investigated a total new method of predicting rut changes over time including:

• a simplified model proposed for the initial rut/initial densification of the pavement;

• model formats considered for the prediction of the annual change in rutting progression; and

• an additional component to the rutting model added to predict the probable point when the accelerated rut stage starts.

A practical method for predicting the performance of unbound granular materials, including alternative, industrial by-products and recycled materials has been proposed. It is recommended that this method replace the existing laboratory repeated load triaxial (RLT) method adopted in the Transit New Zealand specification TNZ M/22 (2000) to determine the suitability of the alternative road material for use as a base or sub-base material for thin-surfaced granular pavements in the New Zealand context.

Report on how the modelling of roughness progression can supplement pavement design.

Models for predicting pavement deterioration developed in Nordic countries.

Report describing the HDM-4 models for predicting pavement deterioration and mainteance effects.

How the HDM-4 pavement deterioration model was calibrated to Chile (in Spanish)

Road agencies in Australia are adapting HDM-4 technology to the strategic management of their road
networks. There is a need to develop HDM-4 technology road deterioration (RD) models for sealed granular
pavements as they comprise 95% of Australia’s rural arterial roads. Austroads has funded ARRB Transport
Research (ARRB) since 1994 to adapt HDM-4 technology to Australian conditions.
Since 1994 ARRB has monitored long term pavement performance (LTPP) sites to observe road
deterioration with traffic loading, climate and pavement type. From 1998 onwards, ARRB has also
monitored the influence of maintenance on sealed granular pavement performance using long term
pavement performance maintenance (LTPPM) sites on Australia’s arterial roads, specifically varying surface
maintenance treatments at each site. In addition, ARRB, in an independent consulting capacity, has
performed field data-driven RD model calibrations for a number of States.
Accelerated load testing (ALT) of experimental sealed granular pavements commenced in 1999 to quantify
the influence of maintenance on relative pavement performance, under controlled conditions of loading,
climate and maintenance. In 2003 ALT was used to quantify the influence of increased axle mass loading
on pavement deterioration.
Historical Australian seal life and binder hardening data was available to ARRB to develop a refined binder
hardening model using variables for environmental conditions, elapsed time, binder characteristics and the
nominal seal size. This refined model, in conjunction with an existing distress viscosity model, allowed the
development of an explicit seal life model to predict the expected life of different nominal seal sizes in
different climates throughout Australia.
This paper presents the current state-of-the-art characteristics, by means of re-calibrated default coefficients,
for the RD models for roughness and rutting progression which were shown to vary with the environment
and surface maintenance treatments. The RD model re-calibration used the observational data from the
LTPP and LTPPM sites in conjunction with the relative performance factors estimated for various surface
maintenance treatments from the ALT data. The current imitations of these revised models with the above
sealed granular pavement data are stated, including those for the seal life model. As a result, the RD
models are more responsive to changes in maintenance under a range of Australian climatic conditions.
It was not possible to predict the impact of surface maintenance treatments, time and traffic on the structural
deterioration of sealed granular pavements. Consequently, the HDM-4 structural deterioration model for
sealed granular pavements could not be re-calibrated. This outcome suggests that modifications to the
HDM-4 structural deterioration model are needed for sealed granular pavements.

Details on the modelling approach in HDM-4 and the Swedish PMS

The design of new pavements in New Zealand and rehabilitation treatments are currently performed in accordance with the Austroads Pavement Design Guide and its New Zealand supplement. New Zealand is also adopting pavement deterioration modeling based on the World Bank HDM models. This paper demonstrates how the modeling of roughness progression can supplement pavement design. It also demonstrates that the long life >50 years of many NZ pavements without significant roughness is not unexpected. It concludes that a combination of deterioration modeling and mechanistic design can be a powerful tool and that the rehabilitation of most of the network is associated with failure other than the classic roughness and rutting.

Results of a study to calibrate the HDM-4 rut depth model to Japan

Adaptation of HDM-4 models for Mediterranean climate

How the HDM-4 pavement deterioration model was calibrated to South Africa

Paper from LVR Conference on HDM calibration to Chile.

Results of pavement performance study in Indonesia and suitability of modelling with HDM-4 approach

Identifies the sensitive input data for pavement deterioration modelling. Although focused on dTIMS, the models were based on HDM

Report describing various techniques for establishing pavement strength and how they were implemented in New Zealand.

Paper describing the calculation of SNP using FWD data

Paper describing the basis for the new SNP in HDM-4

Draft Final Report. The main objective of this Transfund funded research project is to identify the sensitive parameters which will affect the output of the analysis using NZ dTIMS system. This will help the user on deciding which data item should be given higher priority in terms of accuracy level with which data are to be acquired.