Pavement Performance

It is a well established fact that the roughness is manifested as an effect of different individual

pavement deterioration parameters. Several studies have been oriented in the direction of

establishing the models capable of predicting the roughness. However, it was felt essential to

develop a model explaining the dynamics of different pavement deterioration parameters on

the roughness. In view of very limited studies reported in this direction, the present study is

carried out, first by grouping available data into homogeneous clusters and then model them

using Feed Forward Back Propagation Artificial Neural Network algorithm. K- Means

partitional clustering algorithm has been adopted for clustering the data. A new mathematical

algorithm has been proposed and used for optimizing the number of clusters, which is further

verified with the available standard validity indices. The present modeling attempt has

indicated strong correlation between the road roughness and the deterioration parameters viz

cracking, raveling, potholes, patching and rutting. The models developed for all the clusters

have shown decent statistical acceptability.

Study into whether chalk subgrade was a key factor in accelerated pavement deterioration.

Current road roughness deterioration modelling and analysis tends to focus on the prediction of roughness progression in terms of change in the IRI over time. Since the IRI is simply a summary index of the actual roughness, which simulates the response of a specific type of vehicle (quarter-car), it is difficult to identify the factors that contribute to the deterioration of road roughness. Understanding the factors that lead to the deterioration of roads and identifying the actual mode of road roughness deterioration will help road controlling authorities refine their specifications on road roughness requirements for road design, construction and maintenance to reduce their adverse influence on roughness.

This research project looked at an alternative method to analyse and define the roughness deterioration modes of different pavement sections of the New Zealand road network by analysing the characteristics of the longitudinal profile of the road surface using wavelet analysis.

This characterisation process was used to analyse the effects of pavement type, traffic loading, environment and maintenance regime on the deterioration of road roughness and ultimately should lead to the development of a strategy for maintaining road roughness of different pavement types commonly found in New Zealand road networks.

The risk involved in pavement design is that the pavement life will be shorter than the design life. While the literature reveals that statistical methods can be used to estimate the risk (or reliability) of a pavement design, the researchers do not appear to have demonstrated the rigour of their analysis by comparing their results with the performance of actual pavements.

This research project, carried out in 2008 on four state highway networks in New Zealand, studied how the interaction of all the variables relating to pavement life combine to influence pavement performance.

The probabilities of failures were investigated through the available RAAM data. The study examined the rutting and roughness performance of unbound granular pavement and full-depth asphalt pavement.

Based on these findings, it is proposed that thin-surfaced granular pavements have a bimodal distribution of life. The first peak is in the first one to two years, when shallow shear and potholing can occur. After this period, the pavement settles down and the average life will approximate 45 50 years under moderate traffic. It is also concluded that although the pavements have not been failing through rutting or roughness, the Austroads Pavement design guides' proposed risk of a 5% probability of not achieving the pavement's design life appears to be correct.

Forecasting road condition after maintenance can help in better road maintenance planning. As road administrations annually collect and store road-related data, data-driven methods can be used in determining forecasting models that result in improved accuracy. In this paper, we compare the prediction models identi fied by experts and currently used in road administration with simple data-driven prediction models, and parsimonious models based on a input selection algorithm. Furthermore, non-linear prediction using radial basis function networks is performed. We estimate and validate the prediction models with a database containing data of over two million road segments.

PhD Thesis from University of Nottingham.

Pavements represent an important infrastructure to all countries. In Saudi Arabia, huge investments have been made in constructing a large network. This network requires great care through conducting periodic evaluation and timely maintenance to keep the network operating under acceptable level of service.



Pavement distress prediction and pavement condition prediction models can greatly enhance the capabilities of a pavement management system. These models allow pavement authorities to predict the deterioration of the pavements and consequently determine the maintenance needs and activities, predicting the timing of maintenance or rehabilitation, and estimating the long range funding requirements for preserving the performance of the network.

In this study, historical data of pavement distress and pavement condition on the main and secondary road network of Riyadh, Saudi Arabia were collected. These data were categorized, processed, and analyzed. These data have been employed to generate prediction of pavement distress and condition models for the Saudi Arabia Urban Road Network (SAURN).

Throughout the study, the most common types of pavement distress on SAURN have been identified. The behavior of these distress types has been investigated. A sigmoid function was found to be an excellent representation of the data. Seven for urban main pavement distress models (UMPDM) have been developed. In addition, six urban secondary pavement distress models (USPDM) have been developed. Moreover, two pavement condition models have also been developed, one for urban main pavement condition (UMPCM), and the other for urban secondary pavement condition (USPCM). The developed models provide a reasonable prediction of pavement condition. The models were assessed by standard error and residual analysis. A suitable procedure for the implementation of the models has also been proposed.

The pavement analysis in this study considers various combinations of pavement rehabilitation treatments (two-course HMA overlay with or without surface milling, concrete pavement restoration, three-course HMA overlay with or without surface milling, three-course HMA overlay with crack and seat of PCC pavement and 3-R and 4-R overlay or replacement treatments). Six road functional classes (rural and
urban of interstates, non-interstates of the NHS, and non-interstates non-NHS) are considered. This allows for estimation of the performance and service life of the pavement, corresponding to each treatment and road functional class. Main findings;

• More than 95 percent of the data points of the RUT pavement performance indicator were below 0.5 inches indicating that this type of distress has become relatively rare on INDOT highways.
• Data points of the PCR were scattered in a very narrow range (between a PCR value of 70 and 100) compared to the scatter of the IRI and deflection. Consequently, distinct thresholds can be obtained from the wide scatter of the IRI and the deflection. This suggests that IRI and deflection are more reliable performance measures than PCR when programming pavement rehabilitation treatments.
• Two-course hot mix asphalt, HMA, overlay (with or without surface milling) treatments were found to have a forecasted
average annual deterioration in IRI of roughly 6 in/mi.
• Three-course HMA overlay with or without surface milling treatments were found to have a forecasted annual average
deterioration in IRI of about 5 in/mi. Three-course HMA overlay with crack and seat of PCC pavement treatments were found
to have a forecasted average annual deterioration in IRI of roughly 4 in/mi. Pavement projects identified as 3-R and 4-R overlay or replacement treatments were found to have a forecasted average deterioration in IRI in the range of 4 to 5 in/mi. Concrete pavement restoration treatments were found to have a forecasted average annual deterioration in IRI of roughly 7 in/mi.
• Average service life of two-course HMA overlay (with or without surface milling) was found to be roughly 10 years; 12 years for concrete pavement restoration; 12 years for three-course HMA overlay (with or without surface milling); 15 years for threecourse HMA overlay with crack and seat of PCC pavement; and 15 years for 3-R and 4-R overlay or replacement treatments. These numbers match closely with the estimated service lives in the current INDOT design manual.
• Unit cost of pavement rehabilitation treatment was strongly correlated (with a high degree of statistical confidence) to the
service life prediction and consequently was used in the performance prediction models.

This research project looks at an alternate method to analyze and define the roughness deterioration modes of different pavement sections of the New Zealand road network by analyzing the characteristics of the longitudinal profile of the road surface using wavelet analysis.

The prediction of road pavement performance may be facilitated by appropriate models developed by analyzing sets of historical data or data collected from accelerated pavement testing facilities. However, there may be systematic or random errors in these data sets and also the data sets may not be complete or represent the full range of conditions likely to occur in the field. As a result the predictions made by the models may not be fully accurate and include a degree of uncertainty. Therefore, ideally the behavioral modeling of long-term pavement performance should include a procedure for taking into account the uncertainty in the data and quantify it accordingly. This paper presents such a methodology that first defines the reliability of pavement performance predictions and its associated risk using a probabilistic approach. It then demonstrates how the reliability of pavement performance predictions can be estimated by considering the variability of the parameters (such as pavement strength, cumulative equivalent standard axle load and initial pavement roughness) that make up the performance model. A framework is presented that uses the Monte Carlo simulation to evaluate the effect of the model parameters variability on the allowable cumulative equivalent standard axle load applications. The analysis demonstrates that data variability has a significant influence on the reliability of pavement performance prediction.

This study of road deterioration models constitutes one of two parts part of WP5 in the ECRPD project.

This study was undertaken to develop a more reliable methodology, based on road profile variance, for identifying and prioritising for treatment, road sections that promote poor ride quality for heavy commercial vehicles. Profile variance is a measure of the difference between the actual road profile and its moving average over selected moving average lengths.

This study quantifies the impact of some of the parameters on the magnitude of the transverse pavement shear forces or scuffing forces generated during constant low-speed turns.

Report on a study to monitor the performance of gravel roads.

This paper by Henning summarises the highlights and progress made with the LTPP programmes in New Zealand.

This sub-study researched the viability of undertaking a national gravel surface performance study in Vietnam; developed appropriate methodologies for the work and proposed a general framework for the Rural Road Gravel Assessment Programme (RRGAP).

This document concludes the initial investigations on Rural Road Gravel Performance in Vietnam. It provides recommendations on the restriction and use of gravel and other ‘unsealed’ materials as a rural road surfacing in the various environments encountered in Vietnam.

Document providing advice on the construction, assessment and maintenance of pavements that are not expected to experience structural deterioration.

Report describing tests to compare the effect of mass (of 8, 10 and 12 tonne single axle dual tyres) on pavement wear for a range of pavements which are more typical of those used on New Zealand roads.

This paper undertakes a review of falling weight deflectometer (FWD) structural testing of pavements and it’s use in the management of pavements today. Particular focus is made on the possibility of supplementing the structural data to more reliably predict pavement remaining life. The intention of this review is to examine the directions that could be taken to resolve this issue, with subsequent work required to resolve the questions identified.

This report documents the calibration results based on the 04/05 analysis round, formulates recommended calibration procedures and reviews the appropriateness of the local authority data based on the past two years of survey data. The main purpose of this research is to improve the fundamental understanding of pavement performance /deterioration including the regional variation.

Report on National Highway pavement deterioration study in India. Data was collected and used for calibration of the HDM-4 pavement deterioration model.

Paper that describes the ongoing adaptation of a suite of innovative performance models, based on HDM-4, for asphalt pavements for use in mild (Mediterranean) climates and operating conditions.

Study on the performance of reinforcements on the asphalt paved surfaces in urban roads subjected to heavy commercial traffic. [Portugese]

The primary aim of the research was the development of a model, validated for conditions found on the New Zealand State Highway network, that allows reliable prediction of in-service skid resistance performance of chipseal surfaces from readily determined surface and traffic characteristics.

Pavement engineers have used road roughness as an indicator of a road’s functional performance for many years now. Generally, the roughness data has been viewed in terms of a pavement’s current absolute condition in which excessively rough roads will be earmarked for further engineering investigation and possible remedial works. Engineers have also attempted to predict the future roughness of their road network so that appropriate funding can be proactively assigned to maintain an acceptable level of service for road users. To undertake these future predictions a roughness progression model, or deterioration model, is used to increase the roughness of the pavement in proportion to its age and other parameters. Many deterioration models are available, however anecdotal evidence from practitioners indicated that the theoretical roughness progression provided by many models is rarely mimicked in real life. As such, a research project partnership was established between the Queensland Department of Main Roads and the Queensland University of Technology to investigate road roughness progression.

 

This paper outlines the research effort undertaken in 2001 that reviewed the actual time-series roughness progression of approximately 16,000 individual pavement segments, each 1km in length. The research was undertaken using data from the Queensland Department of Main Roads’ database and spans between 4 and 14 years of roughness information. The technical challenges and resulting methodology of analysis of the time-series roughness data is discussed. The research also investigated the effects of pavement maintenance and reviewed many of the common variables such as age, soil type etc that are used in roughness prediction models.

 

In 2003, the analysis was re-performed with three years of additional roughness data and revealed similar results to the 2001 research. Currently the Australian Road Research Board is investigating the use of roughness progression to assist in identifying under-performing pavements. Further improvements on the current methodology and future uses of actual pavement roughness progression are also discussed.

This study examined the tire surface friction on wet pavements with special focus on surface friction testers. Data and information were obtained from publications and private communication with worldwide specialists. In this report, potential relations between wet surface tire-to-ground friction and surface texture measurements are identified and analysed.

The Pavement Technology Project (PTP) in Thailand was a three-year research project undertaken jointly by the Department of Highways and the Danish Road Directorate that commenced in 1999. The project had the overall objective of reducing the maintenance costs on the national road network by implementing Falling Weight Deflectometer Technology and developing and implementing analytical pavement analysis and design methods, based on pavement performance and by determining the structural properties of the road pavement materials tested in existing road pavements and in laboratory investigations.

Austroads, the association of Australian and New Zealand road transport authorities, therefore commissioned a study to look at the setting of minor risk breakpoints and the measurement allowances that should apply to the enforcement of mass limits. This report summarises the results of that study. The study considered Australian and overseas practices and analysed Australian vehicle axle mass statistical data in order to estimate the effects of changes to measurement allowances and breakpoints.

In October 2000, more than 120 transportation and planning officials gathered in California, for the Conference on Performance Measures to Improve Transportation Systems and Agency Operations. The objective of the conference was to bring together a group of government, academic, and business leaders who had experience in transportation systems performance measures and performance-based planning and programming.

The objective of this study was to develop models for predicting the cumulative number of load applications to the initiation of cracking for flexible pavements.

This document sets out a proposed procedure for performing Hit Rate analyses which compare pavement modelling outputs with the maintenance programme which is finally accepted, on a road network.

This paper describes the results of the annual surface measurements in the period from 1990-1999 made in Denmark. The main objective for establishing the test sections was to examine how the noise reduction behaviour of the test sections performs on long term. But also issues, such as general performance and winter maintenance of these types of pavements were studied. These parameters are of vital performance for the road users as friction and evenness can have significant influence on both safety and economic aspects for the road users.

The aim of this Report is to review the current state of knowledge on the use of pavement strength data at a road network planning level.

This report provides a literature review and subsequent list of research needs for investigation into pavement deterioration for Asset Management Modelling purposes. In particular, this report provides a review of the ‘,Roughness’, characteristic of road pavements and explains why this parameter is important in a pavement management context. Current world wide roughness modelling practices are reviewed and evaluated with recommendations given for the use of model types in a pavement management focused research effort, in order to provide realistic benefits to the pavement management industry, including government road agencies.

Report on study undertaken to estimate particle emissions from roads.

Paper on pilot study with the objective of developing improved mechanistic deterioration models for flexible pavements based on an accelerated full scale test on an instrumented pavement in the Danish Road Testing Machine. The paper describes in detail the data analysis and the resulting models for rutting, roughness, and a model for the plastic strain in the subgrade.

It is accepted that the term “structural cracking” incorporates both load and age effects and Greg Morisiuk has undertaken to produce a model form for crack progression that includes both these effects. The author was asked to examine the effects of maintenance surfacings on retarding crack initiation and the effect of previous cracking on the performance of the new surfacing.

Report on project on traffic volume and axle load data.

The Specific Pavement Studies include specially constructed pavements that will help develop better understanding of the effects on performance of a few targeted factors not widely covered in GPS and explore options for construction of new pavements, the application and maintenance treatments to existing pavements, and the rehabilitation of distressed pavements. This report summarizes the status of SPS as at June 1992.

This study deals with the impact of roughness-induced dynamic load on flexible pavement distress and performance.