Truck/Pavement/Economic Modeling and In-Situ Field Test Data Analysis Applications ― Volume 2: Verification and Validation of Finite Element Models for Rigid Pavement Using In-Situ Data ― Selection of Joint Spacing

This work is conducted to provide a complete verification and validation of four different finite element programs (ISLAB2000, JSLAB, EVREFE, and OU3D) for rigid pavement. The experimental data used in this process was collected from the Ohio SHRP Test Road (including four core sections with various geometry and pavement layers) and the Ohio University Accelerated Pavement Load Facility (three different loading conditions). The variety of the pavement sections and loading conditions makes this verification a complete and unique study. The verification outcomes are applied to optimize the joint spacing for least critical stresses within the pavement design life by using three concrete fatigue models (PCA, Huang, and Domenichini). The validation results show that the finite element models follow the general trend of the experimental data in strain, deflection, and vertical pressure. However, two issues are pointed out: stress reversals between the truck axle loads, and slab rocking. Experimental results show that moisture loss that occurred after placing concrete slabs produces a residual negative temperature gradient of -2.2oF/inch (-0.48C°/cm). This leads to a permanent loss of support (LOS) under the pavement corners. Thus, positive gradient-based curling will only reduce the amount of LOS without eliminating it. When tested under different loading conditions, ISLAB2000 shows some inaccuracy in modeling the joints under combined traffic and environmental loadings. With the presence of built-in negative temperature gradient due to the curing of concrete, the critical tensile stresses are located at the top of the slab, and are maximized when the two truck axle loads are positioned on the two edges of the slab. This fact is confirmed experimentally; the slab cracks are observed to initiate at the top and develop towards the bottom of the pavement. The review of the three fatigue models shows that the level of tensile stress overcomes the frequency of load application. This is due to the tensile stresses resulting from the built-in negative temperature gradient. For the four pavement sections in the study, 13 ft (4.0 m) slabs had the longest design life.