Karim Chatti

Estimating the Effects of Pavement Condition on Vehicle Operating Costs

This report presents models for estimating the effects of pavement condition on vehicle operating costs. These models address fuel consumption, tire wear, and repair and maintenance costs and are presented as computational software on the accompanying CD-ROM, CRP-CD-111, to facilitate use. The material contained in the report should be of immediate interest to state pavement, construction, and maintenance engineers; vehicle fleet managers; and those involved in pavement investment decision processes and financial aspects of highway transportation.

Calibration of HDM-4 Models for Estimating the Effect of Pavement Roughness on Fuel Consumption for U. S. Conditions

Fuel consumption costs are an essential part of life-cycle cost analysis. These costs are influenced by vehicle technology, pavement condition, roadway geometrics, environment, speed, and other factors. Many models for the effects of pavement condition on fuel consumption were developed on the basis of data generated years ago in other countries for vehicles that vary substantially from those used currently in the United States. Therefore, new information is needed to help in refining and developing models that would better apply to U.S. conditions. The mechanistic model developed as part of the Highway Development and Management software (HDM-4) is recommended after calibration for predicting fuel consumption. The results of the calibration exercise for U.S. conditions, with field data collected as part of the NCHRP Project 1-45, are presented. The calibrated HDM-4 fuel consumption model was able to predict very adequately the fuel consumption of five different vehicle classes under different operating, weather, and pavement conditions. The better accuracy achieved after calibration has improved the prediction of the effect of roughness on fuel consumption. The comparison of sensitivity analyses before and after calibration has shown that the effect of roughness on fuel consumption increased by 1.75 for the van, 1.70 for the articulated truck, 1.60 for the medium car, 1.35 for the sport utility vehicle, and 1.15 for the light truck.

Toward an Integrated Smart Sensing System and Data Interpretation Techniques for Pavement Fatigue Monitoring

Abstract: Currently, pavement instrumentation for condition monitoring is done on a localized and short-term basis. Existing technology does not allow for continuous long-term monitoring and network level deployment. Long-term monitoring of mechanical loading for pavement structures could reduce maintenance costs, improve longevity, and enhance safety. In this article, on-going research to develop and validate a smart pavement monitoring system is described. The system mainly consists of a novel self-powered wireless sensor based on the integration of piezoelectric transduction with floating-gate injection capable of detecting, storing, and transmitting strain history for long-term monitoring and a novel passive temperature gauge. A technique for estimating full-field strain distributions using measured data from a limited number of implemented sensors is also described. The ultimate purpose is to incorporate the traffic wander effect in the fatigue prediction algorithms. Preliminary results are shown and limitations are discussed.

Effective layer temperature prediction model and temperature correction via falling weight deflectometer deflections

Surface deflections and backcalculated layer moduli of flexible pavements are significantly affected by the temperature of the asphalt concrete (AC) layer. The correction of these deflections and moduli to a reference temperature requires the determination of an effective temperature of the AC layer. In light of this, a new temperature prediction model for determination of the AC temperature on the basis of a database approach is presented, and temperature correction factors for AC modulus are developed. Temperature datum points (n = 317) and deflection profiles (n = 656) were collected from the six in-service test sites in Michigan. Temperature datum points (n = 197) from three of the test sites were used to develop the temperature prediction model, and data from the remaining sites were used for validation. The temperature prediction model developed has an R2 value greater than 90 percent and an F-statistic significantly greater than 1.0. For further validation of the temperature prediction model, temperature datum points (n = 18,444) from seven Seasonal Monitoring Program sites (Colorado, Connecticut, Georgia, Nebraska, Minnesota, South Dakota, and Texas) were obtained from DATAPAVE, version 2.0 (Long-Term Pavement Performance program database). The validation results suggested that the model could be adapted to all seasons and other climatic and geographic regions. The major improvements over existing models are that (a) the model does not require temperatures for the previous 5 days, (b) it takes into account temperature gradients due to diurnal heating and cooling cycles, and (c) it needs fewer parameters than other published models. The effect of temperature prediction error on the performance prediction was also investigated.

Field Investigation into Effects of Vehicle Speed and Tire Pressure on Asphalt Concrete Pavement Strains

An asphalt concrete section on a test track in the PACCAR Technical Center in Mount Vernon, Washington, was fitted with strain gauges at the surface and in pavement cores and tested using an instrumented truck operated at different speeds and with different tire pressures. The field test results are presented. The results indicate that the effects of both vehicle speed and tire pressure-contact area on pavement strains are significant: increasing vehicle speed from 2.7 Km/hr (1.7 mi/hr) to 64 km/hr (40 mi/hr) caused a decrease of approximately 30 to 40 percent in longitudinal strains at the bottom of the asphalt concrete layer, which was 137 mm (5.4 in.) thick. The speed effect on transverse strains is lower, causing only a 15 to 30 percent decrease. Reducing tire pressure from 620 kPa (90 psi) to 214 kPa (30 psi) caused a decrease of approximately 20 to 45 percent in the horizontal strains at the bottom of the asphalt concrete layer. The pressure effect on surface strains was significantly lower, causing...

Dynamic Time Domain Backcalculation of Layer Moduli, Damping, and Thicknesses in Flexible Pavements

A dynamic time domain method was developed to backcalculate the layer moduli, damping ratios, and thicknesses of asphalt pavements from dynamic falling weight deflectometer test data. The method uses the SAPSI program as a forward routine and a Newton-Raphson method for the backcalculation. The advantage of the time domain approach is that SAPSI can match selected features of the measured time histories directly, and inaccurate measurements at the ends of histories can be ignored. The peak deflections and the time lag between the peak of the load and the peak of the deflections at different sensors are matched. The Newton algorithm in MICHBACK is adopted to perform the backcalculation. The new algorithm is capable of backcalculating the moduli, damping ratios, and thicknesses of three-layer flexible pavements reasonably well. Backcalculations based on synthetic time histories generated using SAPSI show excellent stability and accuracy. Backcalculation using measured time histories yields less accurate results. Possible sources for the discrepancies when using field data are addressed.

Backcalculation of Dynamic Modulus Mastercurve from Falling Weight Deflectometer Surface Deflections

The need to characterize the structural condition of existing pavements accurately has increased with the recent development, release, and ongoing implementation of the new Mechanistic–Empirical Pavement Design Guide (MEPDG). There is a strong need to identify and evaluate the way that falling weight deflectometer (FWD) testing is operated and integrated into the new design procedure. One of the key inputs in the MEPDG for asphalt pavements is the dynamic modulus (|E*|) mastercurve. If the damaged |E*| mastercurve of the asphalt concrete in an existing pavement can be obtained from FWD deflections, a more accurate prediction of its remaining service life can be achieved. A methodology backcalculates the |E*| mastercurve of the asphalt pavement layer by using the time history of FWD surface deflections. The method uses a layered viscoelastic forward algorithm in an iterative backcalculation procedure for linear viscoelastic characteristics of asphalt pavements. With deflection time histories from a typical FWD test, it was possible to backcalculate the relaxation modulus curve, E(t), up to about t ∼ 10−1 s and the complex modulus curve, |E*|, from f = 10−3 Hz and above. Recommendations to improve the accuracy of the backcalculated |E*| mastercurve are provided in the context of enhancing current FWD technology and test procedures.

EFFECT OF DIFFERENT AXLE CONFIGURATIONS ON FATIGUE LIFE OF ASPHALT CONCRETE MIXTURE

The fatigue life of an asphalt mixture under different truck axle configurations was determined directly from the indirect tensile cyclic load test by using load pulses that were equivalent to the passage of an entire axle group or truck. The dissipated energy approach was adopted in analyzing the results and determining the number of repetitions to failure for each case; a unique fatigue curve that can be used for multiaxle configurations was developed. Trucks consisting of up to 11 axles and axle groups of up to 8 axles were studied. The results indicated that the normalized damage per load carried decreased with an increasing number of axles within an axle group. Additionally, the fatigue lives predicted by using single load pulses were compared with the measured ones from the different axle groups and trucks.

DEVELOPMENT OF NEW PROFILE-BASED TRUCK DYNAMIC LOAD INDEX

A new roughness index called the dynamic load index (DLI) is developed for the purpose of identifying pavement profiles that are likely to generate high dynamic truck-axle loads. The DLI is calculated as a weighted index of variances of the profile elevation in the frequency ranges of 1.5 to 4 Hz and 8 to 15 Hz. The first frequency range corresponds to truck body bounce, and the second frequency range corresponds to axle bounce. The analysis showed a very good relationship between the DLI and dynamic load. The DLI was tested on a range of road profiles from in-service pavements, and it was found that for any particular value of ride quality index (RQI), the DLI can cover a wide range of values, and this variation in DLI was found to correlate well with dynamic load, as predicted by a truck simulation program. This was not the case for the international roughness index, which gave a low coefficient of correlation with dynamic load for the same range of profiles. Therefore, the new index can differentiate between profiles that generate high dynamic loads and those having the same RQI but generating low dynamic loads. Most importantly, the use of the DLI negates the need for running a truck simulation program. This makes it possible for a state highway agency to decide whether a particular pavement with a given surface profile needs smoothing (to extend its service life) based on the DLI value.

SAPSI-M: Computer Program for Analyzing Asphalt Concrete Pavements Under Moving Arbitrary Loads

A new solution for the dynamic analysis of asphalt concrete pavements subjected to moving fluctuating loads has been developed. The method builds on an existing model for stationary loads, SAPSI, which uses the complex response method of transient analysis with a continuum solution in the horizontal direction and a finite-element solution in the vertical direction. The structural model is an n-layered damped-elastic medium. The subgrade can be modeled as either a rigid base or a semi-infinite half space. The loads are surface pressure loads, and the analysis is under axisymmetric conditions. The moving loads are modeled as a series of pulses with a duration equal to the time required for the wheel to pass a fixed point in the pavement. The proposed method of analysis in the new version of the program, SAPSI-M, is an improvement over the existing methods because (a) it treats moving fluctuating loads on an n-layered damped-elastic system; (b) it incorporates such important factors as wave propagation, inertia, and damping effects of the medium, as well as frequency-dependent asphalt concrete properties; and (c) it can handle any load configuration, thus making possible the modeling of multiple wheel configurations of truck axles and airplane landing gears. Theoretical results have shown that the effect of vehicle speed is significant, in part due to the frequency-dependent properties of the asphalt concrete. Comparison with field strain data from full-scale pavement tests has shown excellent agreement.