Use of Genetic Algorithm and Finite Element Method for Backcalculating Layer Moduli in Asphalt Pavements

The backcalculation program (GAPAVE) using genetic algorithm (GA) and finite element method (FEM) is developed to predict the layer moduli from the falling weight deflectometer (FWD) deflections. A number of backcalculation programs are available for estimating pavement layer moduli from surface deflections. However, they are mostly based on the layered elastic theory in calculating the surface deflections and pavement responses. The use of FEM in the forward calculation incorporating with GA enables to improve the accuracy in backcalculating the pavement layer moduli. The optimum GA parameters were selected from the sensitivity analysis for six different pavement structures. A comparison study with the MODULUS program and other GA parameters was conducted to check the prediction accuracy of GAPAVE program. It is found that the use of optimum GA parameters suggested by author can improve the prediction quality in backcalculating the pavement layer moduli. FWD deflection data and resilient modulus test data for 24 pavement sections obtained from Long-Term Pavement Performance database were used to evaluate the performance of the developed backcalculation program. Backcalculated layer moduli for AC layer and subgrade were compared with the resilient moduli obtained from the laboratory testing. The validation results indicate that the GAPAVE can accurately estimate the actual stiffness characteristics of the pavement materials.