Abstract

In geotechnical engineering practices, vertical anchor blocks are considered as a robust structural measure to counter the exerted pullout forces from the retaining structure and to maintain the overall stability of it. Several theoretical models are available in the literature to estimate the pullout capacity of the vertical anchor in cohesionless soil. This paper aims to estimate the accuracy and reliability of those theoretical models by comparing the predicted pullout capacity with the experimental results available in the literature. For this purpose, a rigorous literature review is carried out to compile the experimental studies conducted on vertical anchor blocks. A comparative assessment among the theoretical prediction models to assess the pullout capacity is done based on six statistical parameters: Mean Absolute Percentage Error (MAPE), Root Mean Square Error (RMSE), McFadden’s pseudo R-square, Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and Nash-Sutcliffe model efficiency coefficient. These comparisons reveal the relative strength and weakness of each theoretical model. Thus, it will help the engineers to take an informed decision before adopting any theoretical model for design purposes.