TR744 In Situ Modululs Measurement using Automated Plate Load Testing for Statewide Mechanistic-Empirical Design Calibration.

The Iowa Department of Transportation (DOT) selected the Automated Plate Load Testing (APLT) to conduct a state-wide field calibration study to develop the AASHTOWare Pavement ME Design™ input data needed for typical Iowa foundation layers. A total of 10 project sites were selected that covered common unbound foundation layer cross-sections used in Iowa highways. The goal at each site was to perform cyclic APLTs to determine stress-dependent composite and layered resilient modulus (Mr) using a 12 in. diameter loading plate and perform static APLTs to determine modulus of subgrade reaction k-value using a 30 in. diameter loading plate. The cyclic APLTs showed that the Mr values on the unbound layers are variable across the state and within each project site. The coefficient of variation (Cv) at each site varied between 7% and 70%. Sites with 2 ft of special backfill consisting of recycled asphalt material to improve subgrade, provided higher Mr values than other project sites. Results demonstrated that using the typical values provided in the AASHTOWare Pavement ME Design™ guide based on soil classification can significantly under or overestimate the Mr values. This emphasizes the importance of field verification testing of design input parameters. The modulus of subgrade reaction k-values obtained across the state varied between 35 pci to 300 pci. 11 out of the 14 tests performed across the state showed k values < 150 pci. Permanent deformation (p) values at the end of static APLTs show the values varied between 0.05 and 0.4 in., with 11 out of the 14 tests performed across the state showed p greater than the critical 0.05 in. limit considered in developing loss of support beneath the pavement. Finite element (FE) analysis conducted using the range of values obtained from APLTs indicated that the two most significant measures that can be taken to reduce the bending stresses developed in the pavement layer are increasing pavement thickness and reducing the p in the unbound foundation layers. Simply changing the foundation stiffness/modulus value without accounting for LOS that can potentially occur due to p under repeated loading, the calculated bending stresses in the pavement can be misleading