Predicting the ultimate axial resistance of single driven piles
Brown, Rollins Patrick
MetadataShow full item record
Sixteen existing methods of predicting the ultimate side resistance of single driven piles were evaluated against the uplift tests compiled from the archives of the California Department of Transportation (Caltrans). These methods included methods based on standard penetration tests (SPT), laboratory tests, electric cone penetration tests (CPT), and piezocone penetration tests (CPTU). From this evaluation, the overall variability in the predictions made by these methods was found to be due primarily to the poor performance of the methods in sand. For this particular dataset, of the methods for piles in sand, the method with the lowest variability was Decourt’s (1982) SPT-based method. Decourt’s method, which is applicable to piles in both sand and clay, was also found to perform reasonably well in clay. Based on this evaluation, Decourt’s (1982) method was selected as the basis for the development of an improved side resistance method for Caltrans’ use. This improved side resistance method, modeled after Decourt’s method, was developed using regression analyses to fit the model parameters to a dataset of 97 uplift tests. To develop a companion toe resistance method and to establish an empirical relationship between uplift side resistance and compression side resistance, regression analyses were also performed to fit these additional model parameters to a dataset of 44 tests piles having both uplift and compression tests. Once the improved methods for the prediction of the axial resistance of single piles were developed, an analysis of the reliability of the methods was conducted to provide recommendations for appropriate resistance factors. In this analysis, the reliability of Caltrans’ current methods, to which a resistance factor of 0.50 is applied, was used as a benchmark. The improved methods were found to be considerably more accurate than Caltrans’ current methods, allowing higher resistance factors to be recommended for the improved methods.