Samer W Katicha

Wavelet Denoising of TSD Deflection Slope Measurements for Improved Pavement Structural Evaluation

Nondestructive testing devices are widely used to assess the structural condition and bearing capacity of pavement systems. Continuous deflection devices (CDDs) can safely measure pavement deflection (or other related properties) while traveling at highway speed, which reduces traffic disruption. However, CDD measurements are contaminated with relatively high noise levels compared to stop-and-go devices such as the Falling Weight Deflectometer (FWD). The objective of this article is to evaluate the structural condition of pavement sections using Traffic Speed Deflectometer (TSD) deflection slope measurements. The authors use wavelet transform denoising to: (1) identify localized weak spots such as those that are caused by pavement reflection cracking; and (2) denoise deflection slope measurements to calculate the effective structural number of the pavement. Results show that failure to denoise deflection slope measurements can lead to calculated values that are highly variable (unstable). Attempting to filter these highly variable measurements can lead to erroneous results.

Assessment of Continuous Pavement Deflection Measuring Technologies

This report documents the results of a study that evaluated current technologies implemented in continuous deflection measuring devices. The research team assessed the demand and the potential value of continuous deflection devices for supporting pavement renewal decisions, and identified the technologies best suited for effectively supporting the most critical decisions identified by the potential users through a survey of state departments of transportation. The study produced a catalogue of existing continuous deflection measuring technologies and their characteristics. The capabilities of two devices, that are or showed the most promise to be ready for production use, were reviewed and practical applications for supporting pavement management decision-making identified and illustrated. The Rolling Wheel Deflectometer (RWD) and Traffic Speed Deflectometer (TSD) were selected for further study as these were identified as being in or close to production mode and met the criteria of measurement speed, applied loads and data collection frequency. However, detailed evaluation was only possible for the TSD because the RWD was not available for testing. The research confirmed that, in general, the technology can provide adequate repeatability for network-level data collection and can be used to support critical network-level pavement management business processes. However, the study also showed that the technology is only just maturing and identified possible improvement to make it even more useful and practical.

Fractional viscoelastic models: master curve construction, interconversion, and numerical approximation

We use fractional viscoelastic models that result from the application of fractional calculus to the linear viscoelastic theory to characterize thermorheologically simple linear viscoelastic materials. Model parameters are obtained through an optimization procedure that simultaneously determines the time–temperature shift factors. We present analytical interconversion based on the fractional viscoelastic model between the main viscoelastic functions (relaxation modulus, creep compliance, storage modulus, and loss modulus) and the analytical forms of the relaxation and retardation spectra. We show that the fractional viscoelastic model can be approximated by a Prony series to any desired level of accuracy. This property allows the efficient determination of the fractional viscoelastic model response to any loading history using the well-known recursive relationships of Prony series models.

Conversion of Testing Frequency to Loading Time Applied to the Mechanistic-Empirical Pavement Design Guide

This paper investigates the issue of converting testing frequency in a dynamic modulus test of hot-mix asphalt (HMA) to loading time for implementation in the proposed mechanistic-empirical pavement design guide. Two methods have been proposed in the literature. The first is to convert the test frequency (f; in hertz) to loading time (t; in seconds) by using t = 1/f. The second method is to convert the test angular frequency (to) to the loading time (in seconds) by using t = 1/ω = 1/2πf. An exact interconversion based on the representation of dynamic modulus results by using a generalized Kelvin model and a generalized Maxwell model is presented. The exact interconversion is compared with the two debated methods of converting the dynamic modulus or the storage modulus to a stiffness (inverse of creep compliance) and relaxation modulus. It is shown that both methods result in error in determining either the relaxation modulus or stiffness. On the other hand, it is shown that the resilient modulus can be adequately approximated by the dynamic modulus taken at a frequency of 1/t.

Developing a Network-Level Structural Capacity Index for Asphalt Pavements

This paper presents a network-level structural capacity indicator for asphalt pavements in the state of Virginia. A literature review revealed that several network-level structural capacity indexes have been proposed, and a number of states use structural capacity measures in their network-level decision processes. Some decision methods and structural indexes are compared in this paper using network-level deflection data collected using the falling weight deflectometer and distress data from tests conducted on Virginia interstates. One index that is based on the structural number concept, the Structural Capacity Index, is found to produce network-level decisions that most closely match project-level work done by the Virginia Department of Transportation during the 2008 construction season. The index was adopted, and its sensitivity to various input parameters was determined. Furthermore, the impact of the structural capacity of the pavement on the service life of a pavement maintenance treatment was clearly established in this paper. Equations to define the service life of a corrective maintenance treatment as a function of the structural condition of the pavement are also presented in this paper.

Limits of agreement method for comparing TSD and FWD measurements

This article uses the limits of agreement (LOA) method to compare the falling weight deflectometer (FWD) and the traffic speed deflectometer (TSD), two pavement structural evaluation devices. The TSD measures deflection slope, whereas the FWD measures deflection. For this reason, measurements were converted to the surface curvature index (SCI) and the base damage index (BDI), which can be obtained from each device. The SCI and BDI agreement between the two devices was then evaluated. Although the relationship between the calculated SCI and BDI using both equipments is reasonably close to the line of equality, there is a significant variation and a bias in this relationship. For example, for an average SCI or BDI value of 300 μm, the bias was 30 μm (FWD values lower than TSD values), and the LOA was 380 μm.

Evaluation of Traffic-Speed Deflectometers

Continuous deflection-measuring devices, or continuous deflectometers, are increasingly being used to support project-level and network-level pavement management decisions. Continuous deflectometers are nondestructive pavement evaluation devices that measure pavement deflections caused by a moving load. Some continuous deflectometers can measure with little to no traffic control; this feature makes them more advantageous to use than stationary devices, such as the falling weight deflectometer. The current technologies implemented in different types of deflectometers are discussed, and the most promising devices for supporting network-level pavement management decisions are identified. In that respect, two devices (the rolling wheel deflectometer and the traffic speed deflectometer) have shown promising results and are being evaluated further under Project R06 (F) of SHRP 2.

Enhancing Network-Level Decision Making Through the Use of a Structural Capacity Index

The objective of this study was to show potential applications of a network-level structural index developed for evaluation of flexible pavements. First, several potential applications for implementation of network-level structural measures were identified, and then data from the state of Virginia were used to modify the proposed applications for the index and show examples of them. Several applications were validated with data from network-level deflection testing with the falling weight deflectometer on Interstate highways in Virginia and data from the Virginia Department of Transportation Pavement Management System. The results of the research indicate that including the structural index in the network-level decision process can facilitate a greater understanding of the behavior of the performance of a pavement. Furthermore, the index that was proposed for network-level evaluation of flexible pavements in Virginia was used to develop enhanced deterioration models for particular pavement treatments and to demonstrate the dependence of pavement performance on its structural capacity. It was shown that the functional characteristics of a pavement alone were not adequate to describe the structural condition of the pavement. Therefore, the structural condition, for example, as based on the results of deflection testing, should be considered when network-level pavement management decisions are made.

Probe Vehicles Used to Measure Road Ride Quality: Pilot Demonstration

New vehicle technology is leading to efficient methods for assessing the condition of the National Highway System. The use of simple sensors such as accelerometers, installed in vehicles, could provide a cost-effective way to assess ride quality for pavement management. A pilot study compared data gathered from accelerometers with the current state-of-the-art practices for measuring ride quality. After a review of relevant previous studies involving probe vehicles, this study assessed the use of probe vehicles’ acceleration measurements to evaluate the pavement profile. The repeatability of acceleration measurements with cross-correlation and standard deviation was obtained. With visual methods and the coherence function, acceleration measurements were compared with profile measurements obtained from inertial profilers. The literature review reinforced the view that using probe vehicles for pavement condition data collection would be promising and that measuring pavement condition with typical onboard sensors could provide a cost-effective way to collect data for pavement management. Probe vehicles are most practically used in pavement management applications to describe ride quality by using vehicle accelerometers and the Global Positioning System. The pilot study confirmed that the acceleration runs were repeatable. Visual inspection of the acceleration and profile plots suggested that the acceleration profiles and smoothness measurements were similar. Analysis with the coherence function also confirmed this strong relationship. The tested methodology provides a practical way to evaluate smoothness while providing a wider base of coverage compared with that of inertial profilers.

Assessing the effectiveness of probe vehicle acceleration measurements in estimating road roughness

In this article, we compare roadway roughness measured using a probe vehicle with roadway roughness calculated from the measured profile using an inertial profiler. Roughness is characterised by vehicle body vertical acceleration and probe vehicle roughness index (PVRI), which approximates the international roughness index (IRI) of a full car (rather than a quarter car). The reason the PVRI is used rather than the IRI is that acceleration measurements obtained from a probe vehicle represent the response of the full car rather than a quarter car. An important aspect of this article is that the same physical quantities are compared rather than obtaining a correlation between two different physical quantities. The results suggest that the roughness calculated from probe vehicle measurements is comparable with the roughness calculated from the measured profile; however, the investigation also revealed that data sampling frequency and quarter car parameters, specifically suspension damping and tyre stiffness, can have a significant effect on the measured PVRI.