Predicting the Unconfined Compressive Strength of Rice Husk Ash – Treated Fine-grained Soils

This study aims to develop novel and accurate data-driven predictive models to replace labor-intensive laboratory testing for estimating the unconfined compressive strength (UCS) of problematic soils treated with rice husk ash (RHA) Full Quadratic, Interaction, M5P-tree, and Artificial Neural Netwo...

Full description

Saved in:
Bibliographic Details
Main Authors: Rizgar A. Blayi, Jamal I. Kakrasul, Samir M. Hamad
Format: Article
Language:English
Published: Koya University 2025-06-01
Series:ARO-The Scientific Journal of Koya University
Subjects:
Online Access:https://88.198.206.215/index.php/aro/article/view/1967
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850103388177432576
author Rizgar A. Blayi
Jamal I. Kakrasul
Samir M. Hamad
author_facet Rizgar A. Blayi
Jamal I. Kakrasul
Samir M. Hamad
author_sort Rizgar A. Blayi
collection DOAJ
description This study aims to develop novel and accurate data-driven predictive models to replace labor-intensive laboratory testing for estimating the unconfined compressive strength (UCS) of problematic soils treated with rice husk ash (RHA) Full Quadratic, Interaction, M5P-tree, and Artificial Neural Network (ANN) were trained and evaluated using a dataset of 211 samples that involved seven key geotechnical parameters, including RHA content (0–30%), liquid limit (22–108%), plasticity index (1.3–82%), maximum dry density (1.2–1.9 g/cm3), optimum moisture content (10.5–42.6%), and curing time (CT) (0–112 days). Among all these models, the ANN model demonstrated superior performance (R2 = 0.97, RMSE = 24 kPa, MAE = 17 kPa, SI = 0.10). Sensitivity analysis revealed CT as the most influence factor (21.9%), followed by moisture content (16.1%) and RHA content (15.3%). The findings present that these predictive models provide a hybrid empirical–machine learning approach, and an accurate alternative to traditional UCS testing, significantly reducing the need for laboratory experiments. They also emphasize enhanced geotechnical performance and the sustainable reuse of agricultural waste. Furthermore, the models can offer a time-efficient solution with practical applications in areas such as highway development and foundation engineering.
format Article
id doaj-art-030d00f65b3a4cd4be48760ff23fb70c
institution DOAJ
issn 2410-9355
2307-549X
language English
publishDate 2025-06-01
publisher Koya University
record_format Article
series ARO-The Scientific Journal of Koya University
spelling doaj-art-030d00f65b3a4cd4be48760ff23fb70c2025-08-20T02:39:33ZengKoya UniversityARO-The Scientific Journal of Koya University2410-93552307-549X2025-06-0113110.14500/aro.11967Predicting the Unconfined Compressive Strength of Rice Husk Ash – Treated Fine-grained SoilsRizgar A. Blayi0https://orcid.org/0000-0002-9527-4578Jamal I. Kakrasul1https://orcid.org/0000-0002-0094-9529Samir M. Hamad 2https://orcid.org/0000-0001-5100-5520Department of Civil and Environmental Engineering, Faculty of Engineering, Soran University, Soran 44008, Kurdistan Region – F.R. IraqDepartment of Civil and Environmental Engineering, Faculty of Engineering, Soran University, Soran 44008, Kurdistan Region – F.R. IraqScientific Research Centre, Soran University, Soran, Kurdistan Region – F.R. Iraq This study aims to develop novel and accurate data-driven predictive models to replace labor-intensive laboratory testing for estimating the unconfined compressive strength (UCS) of problematic soils treated with rice husk ash (RHA) Full Quadratic, Interaction, M5P-tree, and Artificial Neural Network (ANN) were trained and evaluated using a dataset of 211 samples that involved seven key geotechnical parameters, including RHA content (0–30%), liquid limit (22–108%), plasticity index (1.3–82%), maximum dry density (1.2–1.9 g/cm3), optimum moisture content (10.5–42.6%), and curing time (CT) (0–112 days). Among all these models, the ANN model demonstrated superior performance (R2 = 0.97, RMSE = 24 kPa, MAE = 17 kPa, SI = 0.10). Sensitivity analysis revealed CT as the most influence factor (21.9%), followed by moisture content (16.1%) and RHA content (15.3%). The findings present that these predictive models provide a hybrid empirical–machine learning approach, and an accurate alternative to traditional UCS testing, significantly reducing the need for laboratory experiments. They also emphasize enhanced geotechnical performance and the sustainable reuse of agricultural waste. Furthermore, the models can offer a time-efficient solution with practical applications in areas such as highway development and foundation engineering. https://88.198.206.215/index.php/aro/article/view/1967Modeling techniques Rice husk ash stabilizationSoil propertiesUnconfined compressive strength prediction
spellingShingle Rizgar A. Blayi
Jamal I. Kakrasul
Samir M. Hamad
Predicting the Unconfined Compressive Strength of Rice Husk Ash – Treated Fine-grained Soils
ARO-The Scientific Journal of Koya University
Modeling techniques
Rice husk ash stabilization
Soil properties
Unconfined compressive strength prediction
title Predicting the Unconfined Compressive Strength of Rice Husk Ash – Treated Fine-grained Soils
title_full Predicting the Unconfined Compressive Strength of Rice Husk Ash – Treated Fine-grained Soils
title_fullStr Predicting the Unconfined Compressive Strength of Rice Husk Ash – Treated Fine-grained Soils
title_full_unstemmed Predicting the Unconfined Compressive Strength of Rice Husk Ash – Treated Fine-grained Soils
title_short Predicting the Unconfined Compressive Strength of Rice Husk Ash – Treated Fine-grained Soils
title_sort predicting the unconfined compressive strength of rice husk ash treated fine grained soils
topic Modeling techniques
Rice husk ash stabilization
Soil properties
Unconfined compressive strength prediction
url https://88.198.206.215/index.php/aro/article/view/1967
work_keys_str_mv AT rizgarablayi predictingtheunconfinedcompressivestrengthofricehuskashtreatedfinegrainedsoils
AT jamalikakrasul predictingtheunconfinedcompressivestrengthofricehuskashtreatedfinegrainedsoils
AT samirmhamad predictingtheunconfinedcompressivestrengthofricehuskashtreatedfinegrainedsoils