Syed Waqar Haider

Estimating Optimum Timing for Preventive Maintenance Treatment to Mitigate Pavement Roughness

The evaluation of pavement preservation interventions is the most important component of a pavement management system. A rational methodology is needed to evaluate pavement preservation alternatives to maximize project- and network-level benefits. A lack of adequate guidance on the timing of preventive maintenance treatments was the main motivation for this study. On the basis of different evaluation criteria, several mathematical models to estimate optimum timing (OT) of various maintenance treatments were developed. On the basis of long-term benefits, the Area2T model was found to be a rational and practical tool to estimate OT for various maintenance treatments by adjusting the models parameters. The sensitivity and practical use of the models are demonstrated in the paper. Although the modeling approach uses the existing understanding, the model can be used to estimate OT of a fix on the basis of the pretreatment pavement performance, jump in pavement condition just after treatment, and expected rate of deterioration after treatment application. The required data for the model can be obtained for various treatments from the existing historical performance data collected for pavement management purposes. The model can be calibrated to local needs by matching actual practice and calculated OT for different fix types.

Development of traffic inputs for Mechanistic-Empirical Pavement Design Guide in Michigan

Characterizing traffic and developing accurate and desirable traffic inputs for the new Mechanistic–Empirical Pavement Design Guide (MEPDG) are critical but challenging activities. The purpose of this study was to develop a process for characterizing traffic inputs in support of the new MEPDG for the state of Michigan. These traffic characteristics include monthly distribution factors, hourly distribution factors, truck traffic classifications, axle groups per vehicle, and axle load distributions for different axle configurations. Axle weight and vehicle classification data were obtained from 44 weigh-in-motion and classification stations located throughout the state of Michigan to develop Level 1 (site-specific) traffic inputs. Cluster analyses were conducted to group sites with similar characteristics for development of Level 2 (regional) inputs. Finally, data from all sites were averaged to establish the statewide Level 3 inputs. The effects of the developed hierarchical traffic inputs on the predicted performance of rigid pavements were investigated with the MEPDG models. An algorithm based on discriminant analysis was developed to acquire the appropriate Level 2 traffic characteristic inputs for pavement design. For pavement analysis and design, it is recognized that site-specific data should be used wherever available. For projects in which site-specific data are not available, it is necessary to know whether Level 2 or Level 3 data are acceptable at a minimum for design. The MEPDG was used to investigate the impact of traffic input levels on predicted pavement performance for rigid pavements. The results of the analysis showed that for pavement design in Michigan, statewide averages should be used instead of MEPDG Level 3 data.

Backcalculated and Laboratory-Measured Resilient Modulus Values

The resilient modulus (MR) of roadbed soils is a necessary parameter in pavement design. The MR of roadbed soils is dependent on the soil type, water content, dry density, particle gradation, Atterberg limits, and stress states. Several procedures can be used for the determination of the MR: laboratory testing, backcalculation with nondestructive deflection testing (NDT) data, and correlations to other soil parameters such as the California bearing ratio, density, and water content. The first procedure is time-consuming and expensive and requires substantial resources to cover the various roadbed soils under the road network. The second is relatively inexpensive and fast and can be designed to cover representative soils under the pavement network. The third procedure is approximate and may be used for the Level 2 design of the Mechanistic–Empirical Pavement Design Guide. The Michigan Department of Transportation (MDOT) sponsored a research project for the development of a systematic procedure to determine the MR values for all soil types encountered under the pavement network in the state of Michigan. The procedure includes laboratory testing of representative disturbed and undisturbed soil samples, NDT with the MDOT falling weight deflectometer, and developing correlations between the MR values obtained from the two tests and between the MR values and other soil parameters that can be obtained using simple tests. The developed correlation equations between the MR measured values and the other soil parameters are presented elsewhere, whereas the correlations between the laboratory-obtained and the backcalculated MR values are discussed in this paper.

Effect of Axle Load Spectrum Characteristics on Flexible Pavement Performance

The Mechanistic–Empirical Pavement Design Guide (MEPDG) uses performance models to predict cracking and rutting in flexible pavements. A unique mechanism controls the initiation and accumulation of each distress, but each mechanism can have several causes. Axle repetitions and loads are the main causes of all load-related distress types. MEPDG incorporates axle load spectra to characterize axle loading for a site and uses them to calculate pavement response and damage accumulation. These load distributions have a bimodal shape, and a mixture of two continuous distributions can be used to model them. In this paper, closed-form solutions are developed to estimate the characteristics of a mixture of bimodal axle load distributions. The observed axle load spectra from 14 sites in different states were used to relate load distribution characteristics to predicted flexible pavement performance. The overall mean and other characteristics of a bimodal axle load distribution explained the variations in expected flexible pavement performance. Cracking, surface rutting, and ride quality are related to the fourth root of the fourth moment of axle load distributions. Rutting in the hot-mix asphalt layer is strongly associated with the overall mean, but in base and subbase layers it is related to the 95th percentile load of axle load spectra. These findings imply that cracking, rutting, and roughness growth in flexible pavements are caused mainly by axle load distributions having heavier tails with infrequent extreme loads. Heavier loads appear to cause more cracking; a higher number of load repetitions is more critical in developing additional surface rutting in flexible pavements.

Characterizing Temperature Susceptibility of Asphalt Binders Using Activation Energy for Flow

The temperature-viscosity relationships for twenty two neat and modified asphalt binders were evaluated using conventional tests (CT) such as penetration, kinematic and absolute viscosities, and ring and ball softening point; rotational viscometer (RV); and dynamic shear rheometer (DSR). The temperature ranged from 25 to 135 °C, 100 to 185 °C, and 7 to 82 °C for conventional, RV, and DSR tests, respectively. Results from several binders studies have revealed that binders having similar penetration, viscosity and performance grade (PG) could show dissimilar rheological behaviors. Hence, there is a need to fully characterize asphalt binders to capture their rheology over a wide range of temperatures and loading frequencies. The activation energy (AE) concept was utilized in this study to characterize the temperature susceptibility of the asphalt binders. Correlations between AE and binder rheological properties (e.g., G*/sinδ and G* sinδ) were evaluated. In addition, a relationship between AE and useful temperature range (UTR) determined from the performance grades (PG) was developed. The advantages of characterizing the relative temperature susceptibility of the binders using AE are discussed. Finally, discussion on mixing and compaction temperatures, especially for polymer modified binders, by using AE is also included in this paper.

Effect of axle load measurement errors on pavement performance and design reliability

In traffic characterization, axle load spectra (ALS) are one of the most critical inputs in the new Mechanistic–Empirical Pavement Design Guide (MEPDG). ALS have a significant impact on the predicted pavement performance. At the design stage, it is typically assumed that ALS as measured by weigh-in-motion (WIM) systems have adequate data quality and accuracy. In fact, the quality of WIM-based data has inherent uncertainties because of inaccuracy and systematic bias. While WIM data accuracy depends on the sensor technology, calibration errors and drift over time may introduce a systematic bias. Several studies have investigated the impact of traffic data collection technologies, data coverage, accuracy, and calibration errors on pavement loading and performance prediction. However, these studies were limited to a few distress measures and did not address design reliability aspects as considered in the MEPDG. This study investigated the impact of probable WIM errors on the ALS and quantified the effects of these errors on the performance of both flexible and rigid pavements. Furthermore, the impact of uncertainties in ALS on design reliabilities is discussed in this paper. Although most findings reinforce existing concepts, the study provides a systematic overview of WIM data accuracy and calibration requirements, and the effect of associated uncertainties in the pavement design process. The results show that cracking in both flexible and rigid pavements is the distress most affected by ALS variations, while rutting in flexible pavements is moderately affected.

Evaluation of New Mechanistic—Empirical Pavement Design Guide Rutting Models for Multiple-Axle Loads

The new FHWA Mechanistic–Empirical Pavement Design Guide (M-E PDG) does away with the AASHO-derived concept of the equivalent single-axle load and calculates damage caused by various axle configurations directly. Because multiple axles represent about half the axle configurations that the pavement will experience, there is a need to evaluate the M-E PDG procedure for predicting rutting due to multiple-axle configurations. Axle factors (AFs) based on rutting were calculated for different axle configurations by using three procedures: the M-E PDG, accounting for the effect of each individual axle within a group, and integrating the entire strain pulse. The AFs from these procedures were compared with laboratory-derived values for asphalt concrete. Also, layer rutting contributions were predicted for six in-service SPS-1 experiment sections from the Long-Term Pavement Performance Program with the M-E PDG software and were compared with the analysis of their transverse surface profiles. The results show that the M-E PDG procedure underestimates rutting prediction due to multiple axles. Calibration of the M-E PDG rut models with field data seems to improve their prediction, although it is still lower than expected for multiple axles. The best method for calculating rut depths under multiple axles appears to be integration of the entire strain pulse. This method shows that rutting damage is proportional to the number of axles within an axle group. This theory was confirmed in the laboratory for asphalt concrete. Also, although the M-E PDG rut models are superior to previous rut models, in that they are able to dissect the total surface rutting between all pavement layers, their prediction of the individual layer rutting contributions does not always agree with results from the analysis of measured transverse profiles.

Estimating optimum timings for treatments on flexible pavements with surface rutting

The evaluation of the pavement preservation interventions is the most important component of a pavement management system. There is a need for a rational methodology to evaluate pavement preservation alternatives to maximize both project- and network-level benefits. Lack of adequate guidance on the timing of preventive maintenance treatments was the main motivation for this study. On the basis of different effectiveness criteria, mathematical models to estimate optimum timing (OT) of maintenance treatments are developed to alleviate surface rutting. The developed model based on the long-term benefits was found to be a rational and a practical tool to estimate OT for various maintenance treatments by adjusting its parameters. The sensitivity and practical usage of the models are demonstrated by using the SPS-3 pavement sections in the paper. Although the modeling approach uses the existing understanding, the model can estimate OT of a fix on the basis of the pretreatment pavement rutting performance, jump, and slope adjustment factors for various treatment types. The required data for various treatments can be obtained from historical performance data collected for pavement management purposes. The results show that slurry and chip seals can be applied on younger pavements when rutting is minimal. However, thin overlay applications on older pavements with higher rutting levels will maximize the long-term benefits.

Impact of Systematic Axle Load Measurement Error on Pavement Design Using Mechanistic-Empirical Pavement Design Guide

In traffic characterization, axle load spectra (ALS) are one of the most critical inputs in the new Mechanistic-Empirical Pavement Design Guide (MEPDG). Axle load spectra have a significant effect on predicted pavement performance and, thus, the design life. Typically, axle load spectra as measured by weigh-in-motion (WIM) systems are assumed to have adequate data quality and accuracy. In fact, the quality of WIM-based data has inherent uncertainties attributable to inaccuracy and systematic bias. Whereas WIM data accuracy depends on the sensor technology, calibration errors and drift over time may introduce a systematic bias. This technical note investigates the effect of axle load measurement bias on pavement design for flexible and rigid pavements. The results show that negative bias in axle load measurements significantly affects cracking performance for both pavement types. The bias is more critical for rigid pavements with thinner slabs. Therefore, a measurement bias limit of less than ±5% should be...

Closed-Form Solutions for Bimodal Axle Load Spectra and Relative Pavement Damage Estimation

The mechanistic-empirical (ME) design procedures utilize axle load spectra to characterize the individual traffic loadings for a site. These loading characteristics are employed to calculate pavement response and for subsequent damage computations. Generally, these axle load distributions exhibit a bimodal shape and a combination of two continuous statistical distributions can be used to model them. In this paper, closed-form solutions are developed to estimate the parameters of the bimodal distribution from data. A combination of two normal distributions is shown to reasonably fit observed axle spectra. Since it is anticipated that the AASHTO equivalent single-axle load (ESAL) concept will continue to be used by pavement engineers even after the full adoption of ME design methods, a closed-form statistical relationship between ESALs and axle load spectra is proposed. Such a relationship will be useful in estimating a traffic level index from an axle distribution. In addition, the relationship can provide an estimate of the relative pavement damage caused by axle distributions, and be used to rank axle load spectra within a geographical region, or between regions in order to identify heavier traffic loading corridors.