Uncertainty and Reliability Appraisal of CPT-Based Methods for Axial Pile Bearing Capacity

CPT-based methods considered as complementary tools for static analyses outperform other axial pile bearing capacity predictive methods. Considering soil spatial variability and minor errors associated with procedure and operator, CPT-base methods mitigate the effect of latent uncertainties in pile design. These methods were developed based on different databases and purposes. Hence, several statistical and probabilistic attributes were defined and implemented to assess the predictive CPT-based methods’ performance via CPT and pile databases in terms of the model parameter. These features make a comparison among methods in terms of precision, accuracy, and efficiency. Furthermore, CPT-based methods can be subjected to scrutiny over their backgrounds and how they were developed via reliability analysis. In this regard, a database of sixty-two driven piles and their adjacent CPT soundings were gathered, and six commonly applied CPT-based methods were examined according to criteria other than prevailing statistical and probabilistic ones, including; (a) input variables, (b) soil behavior classification (SBC) (c) model assumption (d) Influence zone, and (e) axial loading direction. According to these introduced criteria, CPTbased methods were classified and appraised via efficiency ratio. Eventually, these reliability-based, statistical, and probabilistic criteria evaluate the understudied methods performance for the investigated compiled database and help designers select consciously appropriate predictive methods according to the site conditions.