Ultimate Bearing Capacity of Bored Piles Determined Using Finite Element Analysis and Cubic Regression

This study aims to establish a method for evaluating the ultimate bearing capacity of bored piles in sandy soil by integrating finite element analysis (FEM) with nonlinear curve analysis. Using FEM, the research simulated vertical displacements of piles under varying load levels and compared these results with those obtained from static compression tests (SCT). The bored piles, constructed from grade B35 concrete with a diameter of 1 m and a depth of 45.3 m, were analyzed in a geological context characterized by an 80-m-thick sandy soil layer with a groundwater table at − 1.5 m. The findings revealed a high correlation coefficient of 0.97 between FEM and SCT results, indicating significant alignment. The cubic regression model applied to the FEM data achieved an R 2 value of 1, confirming the model’s perfect fit and FEM’s ability to accurately simulate nonlinear variations in vertical displacement. This study concludes that FEM offers a reliable and cost-effective alternative to traditional SCT, providing accurate and sensitive predictions of pile behavior under load, which is critical for foundation design in modern construction projects. Furthermore, the use of cubic regression allowed for the precise identification of critical inflection points, facilitating an accurate determination of the ultimate bearing capacity. This integration of FEM and regression analysis not only enhances the evaluation process but also significantly reduces costs and time associated with pile testing.