This research aims to develop and present a comprehensive and efficient approach for the analysis and evaluation of pavement structures. In the process of pavement evaluation and maintenance, multiple uncertainties exist regarding material properties, loading conditions, and environmental factors that can significantly affect the performance and durability of the structure throughout its service life. Therefore, it is essential to systematically incorporate these uncertainties into the modeling and inverse analysis of pavements to ensure that the evaluation results more closely reflect real-world conditions.
In the first stage, to achieve more accurate approximations of objective functions and structural responses, quadrature computations were employed, reducing the approximation error in the model and providing appropriate conditions for optimization in subsequent stages. Next, interactive metaheuristic optimization algorithms were applied to search for optimal values within the inverse analysis computational space. These algorithms, with their global search capabilities and avoidance of local optima, enabled the identification of more desirable values for the pavement evaluation parameters.
To investigate the role of uncertainties in the ultimate performance of pavement structures, Monte Carlo simulation was utilized. By generating random data and repeatedly running the model, this approach simulated diverse and probable conditions, leading to a more realistic representation of the structural reliability.
In this study, three pavement problems were evaluated. In the first and second problems, a three-layer pavement was analyzed using two methods. In the first method, three unknowns (the elastic moduli of the layers) were considered, yielding mean percentage errors in the elastic modulus of 0/074 for the first case and 0/027 for the second. In the second method, five unknowns (elastic moduli and layer thicknesses) were analyzed, resulting in mean percentage errors