Predicting base resistance of super-long piles using a random forest model: A case study from Ho Chi Minh city

Abstract

The prediction of the base resistance for long piles is usually challenging because of the impact of material characteristics and the influence of the nature of the surrounding soil. Artificial intelligence models have been applied in various geotechnical engineering fields, and significant results have been achieved. Based on a well-instrumented static load test dataset (1131 data points) from various projects in the soft soil area of Ho Chi Minh City, this study established a random forest (RF) model considering five input parameters, including the applied load, load point displacement, axial stiffness, standard penetration test value of the soil beneath the pile toe, and the distance from the load point to the pile toe. Twenty percent of the data was designated as the test set, which was used to make predictions using the established model. The results show that the RF model has good predictive performance in terms of prediction accuracy and reliability. A sensitivity analysis of the input factors provided a deeper understanding of the base resistance mechanism, which is important in pile foundation design practice.