Machine Learning-Based Modelling of Soil Properties for Geotechnical Design: Review, Tool Development and Comparison

Machine learning (ML) holds significant potential for predicting soil properties in geotechnical design but at the same time poses challenges, including those of how to easily examine the performance of an algorithm and how to select an optimal algorithm. This study first comprehensively reviewed the application of ML algorithms in modelling soil properties for geotechnical design. The algorithms were categorized into several groups based on their principles, and the main characteristics of these ML algorithms were summarized. After that six representative algorithms are further detailed and selected for the creation of a ML-based tool with which to easily build ML-based models. Interestingly, automatic determination of the optimal configurations of ML algorithms is developed, with an evaluation of model accuracy, application of the developed ML model to the new data and investigation of relationships between the input variables and soil properties. Furthermore, a novel ranking index is proposed for the model comparison and selection, which evaluates a ML-based model from five aspects. Soil maximum dry density is selected as an example to allow examination of the performance of different ML algorithms, the applicability of the tool and the model ranking index to determining an optimal model.

Automated summary

This study comprehensively reviewed the application of machine learning algorithms in modelling soil properties for geotechnical design and proposed a new ranking index for model comparison and selection. Six representative algorithms were chosen to build a tool that can easily construct ML models. Automatic determination of optimal configurations of ML algorithms was achieved through evaluating model accuracy, applying the developed ML model to new data, and investigating relationships between input variables and soil properties.