LRFD Pullout Resistance Factor Calibration For Soil Nails Incorporating Survival Analysis And PLAXIS 2D

Abstract

The use of soil nail walls (SNWs) in the United States has increased since their introduction in the mid-1970's, to where currently the analysis, design and construction are commonly performed (Lazarte, 2011). These SNW designs were mostly based on ASD methods and LRFD-based methodologies were lacking until the 1998 FHWA manual on SNW design (Byrne, Cotton, Porterfield, Wolschlag and Ueblacker, 1998), which provided uncalibrated resistance factors developed from ASD safety factors. As a result, little improvement was made toward a more efficient design, until fully calibrated LRFD pullout resistance factors were provided in the NCHRP Report 701 (Lazarte, 2011). These pullout resistance factors were calibrated with a variety of load factors commonly used for retaining structures as part of a bridge substructure. Although fully calibrated resistance factors were calculated, the predicted pullout resistance was not based a specific design procedure, but rather on multiple design procedures (Lazarte, 2011). The main objective of this study was to achieve a greater understanding of the bond strength of soil nails in North Dallas Texas. In an effort to accomplish this objective, pullout resistance factors were calibrated for cohesive soils within the project location. Pullout resistances were determined using creep test data, field observations, and methods commonly used in tension piles. This resulted in 25 cases that met failure criteria out of the 47 verification tests conduction in cohesive soil for the LBJ Express construction project. Statistical analysis was conducted to evaluate the predicted to measured pullout resistance for the failed tests, and Survival Analysis was utilized to incorporate the non-failed tests. In addition, PLAXIS 2D was used to fit a finite element model to testing results and used to predict failure in three cases. Results from analysis test results, Survival Analysis and PLAXIS 2D were combined with the 25 failed cases along with soil nail testing results found in the NCHRP Report 701. Five soil nail databases were established from these results and utilized in the remainder of the study. Then, LRFD reliability analysis using Monte Carlo simulations were performed to calibrate pullout resistance factors at a target reliability index of 2.33 and load factors of 1, 1.35, 1.5, 1.6 and 1.75. The final step involved incorporating SNAILZ to compare the required soil nail length between the existing design method and the calibrated resistance factors for a typical SNW.