Big data and machine learning in geoscience and geoengineering: Introduction

Summary: In an era of developing large-scale underground infrastructures, the need for better prediction of the load-displacement performance of grouted anchors is increasing. This paper introduces FastICA-MARS as a novel data-driven approach for this purpose, which outperforms the classical MARS method in providing accurate predictions. By integrating the FastICA algorithm with the MARS technique, the developed model considers the importance of parameters such as grouting pressures, which is essential in engineering practices.