Stabilization of Soil Using Fly Ash: A Laboratory and Machine Learning Approach

DOI

10.2991/978-94-6463-884-4_42How to use a DOI?

Keywords

Soil stabilization; Fly ash; California bearing ratio (CBR); Unconfined compressive strength (UCS); Random Forest (RF)

Abstract

California Bearing Ratio (CBR) and Unconfined Compressive Strength (UCS) are key soil strength parameters, particularly for geotechnical engineering projects and road pavements. This study investigates the locally available non-cohesive soil strength stabilized with fly ash. The physical characteristics of the collected materials and compaction characteristics, as determined by the modified proctor test and CBR value, were analyzed with various percentages of fly ash (0%, 10%, 20%, 30%, 40%, 50%). The CBR value of the stabilized soil maximized at 30% fly ash for both soaked and unsoaked conditions, and afterward, a decreasing trend was observed. This study also incorporates a machine learning approach, namely the Random Forest (RF) model, to predict the UCS value of cohesive soil, as the laboratory investigation requires time, effort, and cost. A total of 90 datasets were collected from the literature, and data were partitioned into 80–20 for training and testing purposes. Several performance matrices, such as the coefficient of determination (R), root mean square error (RMSE), and mean absolute error (MAE), were used to evaluate the effectiveness of the RF model. For both training and testing data, the R-squared value (R²) is over 0.96, indicating the effectiveness of the proposed model. Furthermore, the Shapley additive explanation (SHAP) and dependency plot were implemented to illustrate the relative influence of various input parameters on the model’s output. The UCS value of cohesive soil increased sharply up to 15% of fly ash content, and other features positively impacted the UCS.

Copyright

© 2025 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  - Sabikun Nahar Nilu
AU  - Kamol Debnath Dip
AU  - Shoma Ghosh
PY  - 2025
DA  - 2025/11/18
TI  - Stabilization of Soil Using Fly Ash: A Laboratory and Machine Learning Approach
BT  - Proceedings of the 8th International Conference on Engineering Research, Innovation, and Education 2025 (ICERIE 2025)
PB  - Atlantis Press
SP  - 350
EP  - 358
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6463-884-4_42
DO  - 10.2991/978-94-6463-884-4_42
ID  - Nilu2025
ER  -