Applying Logistic Regression Analysis in Modeling Settlement Analysis with Ground Improvement

DOI

10.2991/978-94-6463-258-3_30How to use a DOI?

Keywords

Settlement analysis; Ground improvement; Machine Learning; Logistic regression; KNN; SVM; Sequence modelling

Abstract

Settlement analysis plays an essential role in providing a safe measure for the stability of the foundation and other structural components. The settlement analysis with numerical methods such as Finite Element method (FEM), Finite Difference Method, or other sophisticated constitutive models are continuing to grow with the advancement of technology around the world. Machine Learning (ML) has been gaining a high interest in recent years in many different fields of engineering for its ability to predict an outcome given input parameters with certain patterns. Although the theories in ML existed before, there has been a significant growth in recent years due to overflowing information that is available online. There are numerous papers that introduce ML application in attempt to connect these two complex models together to predict a soil property and the behavior by comparing the results with the existing numerical analysis methods or field results. Also, ML algorithm has been used in research for predicting fragility curve analysis with given set of training data to determine the risk factor of structural buildings subjected to earthquake loads. There are many areas of focus on achieving accuracy with some complex models. Also, there are many resources available which discuss the modelling aspect of the analysis that can incorporate the ML algorithms in simple models. The paper introduces the modelling aspect of analyzing settlement with selective hyper parameters to achieve the target settlement. The paper introduces a logistic regression modelling using ground improvement result as a target variable and settlement and soil properties as the parameters to train. The study provides a model using ML algorithms to predict whether the soil model has ground improvement or not, with the selected hyper-parameters using settlement analysis software, Settle3. The paper provides the prediction accuracy with different methods of ML algorithms including logistic regression, K-nearest neighbor (KNN), stochastic gradient decent, Support Vector Machines (SVM) etc. The accuracy shows very high confidence with given parameters and more details on the future studies of the paper is discussed.

Copyright

© 2023 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Seok Hyeon Chai
AU  - Tara Stojimirovic
AU  - Thamer Yacoub
PY  - 2023
DA  - 2023/11/08
TI  - Applying Logistic Regression Analysis in Modeling Settlement Analysis with Ground Improvement
BT  - Proceedings of the Rocscience International Conference  (RIC 2023)
PB  - Atlantis Press
SP  - 287
EP  - 297
SN  - 2589-4943
UR  - https://doi.org/10.2991/978-94-6463-258-3_30
DO  - 10.2991/978-94-6463-258-3_30
ID  - Chai2023
ER  -