Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
56 Summary A Reflection Cracking Overlay Design Program was cali- brated using field data for more than 400 pavement test sections in 28 states and the four climatic zones of the United States. The program uses a mechanistic model that predicts the reflection cracking lives of a specified HMA overlay due to bending and shearing traffic stresses and thermal stresses. The relationship between the computed lives and the field conditions, in terms of the extent and severity of reflection cracking, depends upon the characteristics of the pavement structure and overlay and of the weather at the location. This relationship is expressed as sets of calibration coefficients that depend upon the season where the overlay was placed. A Userâs Guide for the Design Program is presented in Appendix O. It uses dialog boxes in the same format as the MEPDG and has the same three optional levels of input detail. The program is designed to run independently or to be incorporated as a subprogram of the MEPDG software. A total of 11 sets of calibration coefficients were developed from the available data. The overlay performances that can be predicted with these different sets of calibration coefficients represent the four climatic zones in the United States. Eleven different pavement structure and overlay combinations were identified in the preliminary surveys. These overlays include HMA overlays over cracked asphalt pavement surfaces, jointed concrete pavement as well as asphalt overlays that use strain- absorbing membrane interlayers and reinforcing geosynthetics. When considering the four climatic zones, there are 36 possible sets of calibration coefficients. The methods of developing sets of calibration coefficients for overlays described in this report make it possible to develop sets of calibration coefficients for a specific type of overlay and to design overlays using these sets of calibration coefficients. A Calibration Program has been provided to facilitate this independent calibration process, and a Userâs Guide is provided in Appendix P. Both the Design Program and the Calibration Program have the same internal 150 location weather databases that can be called up by the user. The calibration coefficients were developed in this project and used in the Reflection Cracking Overlay Design Model to produce reflection crackling predictions that match the field observations as closely as possible of the more commonly used overlays. The program runs quickly enough to allow speedy trials of several overlay thicknesses and asphalt mixes in order to find the design that best matches the project objectives. The computational speed is achieved partially by the use of ANN algorithms to perform the computations that must be done many times in the course of a simulated time period of 10 to 20 years. The ANN algorithms are used to compute the mixture modulus at different temperatures and loading rates and to calculate the stress intensity factors which drive the growth of cracks. Another factor in increasing the computa- tional speed is re-writing all of the routines and subroutines in the same C # computer language. Several examples of predicted reflection cracking perfor- mance of various overlays in different parts of the United States are presented in Chapter 3 and Appendix O. These examples illustrate the reproduction of the field observations of the growth of the extent and severity of reflection cracking. The assumed form of these distress growth curves was chosen because it gives realistic predictions and remains within reason- able bounds. However, extreme designs that are well outside the range of the types of overlays that were used for calibration may not produce realistic results. Suggested Research Another 25 sets of pavement structure-overlay and climatic zone combinations remain to be developed, some of which may not be possible because they are simply not built. However, the process of developing calibration coefficients for a par- ticular type of overlay described in this report can be applied to any overlay in any region or even within a state or region. For example, sets of calibration coefficients for overlays incorporating geosynthetic reinforcing products need to be C H A P T E R 4 Summary and Suggested Research
57 developed (only three were developed in the project). It would also be desirable to develop guidelines or standards for such products to provide estimates of the reinforcing stiffness of these products. The fit between the predicted and the observed reflection cracking distress can be improved by reducing the error in both the predicted and observed performance. The observed field data were fitted with an S-shaped curve with two param- eters: Ï, the scale parameter and β, the shape parameter. Both of these parameters were fitted by linear regression analysis with mechanistically predicted numbers of days for traffic and thermal stresses to cause a reflection crack to grow to the surface of an overlay. There are errors that are inherent in both the observed and the predicted values. The error in the observed field data can be reduced by recording mean values of the observed distress from a sampling survey. The error in the mechanistically predicted numbers of days of crack prop- agation can be reduced by having more accurate recorded pavement structure and materials properties, traffic, and weather and by improving the cracking-predicting process of the mechanistic model. A sensitivity analysis of the mechanistic model of the crack growth process will identify material properties which are the most sensitive predictors of overlay performance. Making improvements in these material properties and how they are measured will extend the service lives of overlays and increase the predictive accuracy of the models that are developed. Performance-based specifications which are focused on these material properties will make construction quality control and quality assurance more effective and will extend the service lives of overlays. The Design Program is a good candidate for incorporation into the MEPDG software framework or being a stand alone. Experience using the Design Program in designing overlays will show which sets of calibration coefficients need to be improved and, in addition, will reveal those features of a design that make the greatest extension of an overlayâs service life. An evaluation of available reflection cracking models was per- formed in this project and a number of promising approaches were in the development stage at that time. Most of these models proved to require extensive computational times that would make them impractical to use in design. However, as computer speeds increase, the mechanics-based methods will require less running time and should be considered for future versions of this overlay design function to reduce the systematic errors and simplify the calibration effort.
