Reliability Analysis of Analytical Al Methods for Pile Capacity Prediction at Lahore Ring Road Project

Abstract

This study summarises the analysis of nine Standard Penetration Test (SPT) based methods to predict the ultimate load carrying capacity of drilled shaft piles at Lahore Ring Road. Standard Penetration Test was conducted at the site and result was compiled and bore logs were prepared. Lab test were conducted to identify the soil properties. Nine SPT based methods were used to foretell the Ultimate Pile Load capacity. Eleven SPT boreholes were conducted and analysed for the capacity. To measure ultimate load carrying capacity Eleven Pile load Tests were conducted and results were interpreted using Davisson’s method, 5% Base Settlement criteria. The following methods were analysed using the SPT data: Salgado soil properties based method, Salgado SPT method, Meyerhof method, O’Neil and Reese 1999 method, O’Neil and Hassan 1994 method, Quiros and Reese, Reese and Wright, Touma & Reese 1975 method, FHWA method. Statistical analysis was carried out to analyse the SPT methods based to select best method on their ability to predict ultimate pile capacity. Three different criteria were used to evaluate the ratio of the predicted to measured pile capacities. These criteria are: Best-fit Line, Arithmetic Mean and Standard Deviation and Cumulative Probability. Final ranking was achieved by averaging the ranking obtained from each criterion. As per results, Salgado Soil Properties Based Method showed the best performance in predicting the load carrying capacity of drilled shaft piles in Lahore soil. The worst prediction method was the Reese and Wright Method. Salgado Soil Properties Based Method over predicts the pile capacity. O’ Neil & Hassan Method, Meyerhof Method and Salgado SPT Based Method give conservative results. Conservative method will be more suitable for our Construction industry due to inherent poor quality control during construction phase