Development of a Spatially Comprehensive Layered Elastic Analysis Tool for Asphalt Pavement Forensic

Abstract: Layered elastic analysis (LEA) is widely used in pavement design and analysis to predict stress, strain, and deflection in flexible pavement systems. LEA is particularly useful when looking for response predictions at a limited number of specific points (e.g., the bottom of the asphalt layer directly beneath a tire load) since the location of calculation must be defined as a priori, and most existing codes limit the number of calculation locations to be computed in a single analysis. This feature becomes problematic from a computational and resource allocation standpoint when the analysis is less prescriptive, and characterization of the entire pavement cross-section is desired. Such cases often arise when performing forensic investigations. In these situations, pavement engineers often turn to finite element or finite difference approaches that mesh the pavement section so complete spatial characterization can be achieved. However, these approaches are much more complicated and yield the same results as LEA if elasticity equations are used as the constitutive models for each pavement layer. The meshed approach just provides a more complete spatial understanding of pavement responses throughout the structure. Given these complications, there is a need to develop an approach that uses LEA but produces a mesh-like solution that provides full spatial characterization. The objective of this research was to develop such a tool by combining an existing LEA tool (WESLEA for Windows) with a macro-recorder to easily generate thousands of pavement simulations that were then fed into MATLAB to render heat maps that resemble the output of more complicated meshing-type solutions.
This paper documents both the development of the tool and its use in a forensic study of a full-scale flexible pavement section at the National Center for Asphalt Technology (NCAT) Pavement Test Track. The pavement section was built with a dry-process rubber-modified asphalt mix, and it had unexpected mid-depth cracking. To investigate the potential causes of the cracking, different percentile moduli with different slip conditions were simulated in the newly developed tool. The meshing was done with quarter-inch increments for the depth and 1-inch increments in the longitudinal direction from the center of loading. For each 50-inch-wide long-layer case, the normal stress and shear stress responses from WESLEA were recorded. All the recorded responses were entered into MATLAB code and heatmaps were generated.
The heatmap showed the overall pavement responses and the critical stress conditions that were found from the simulated results. The location of the peak stress condition would help to identify the critical depth for cracking origination. Another set of simulations was performed with the slip condition at the critical depth and showed how cracking could propagate both at the critical depth and from the bottom of the asphalt. It was found that the critical depth matches the cracking depth of the test track section. This method can help to predict the failure type and the required rehabilitation for prolonging pavement life.