Using artificial neural networks to predict the compressive strength of cement and sawdust ash-treated lateritic soil

Sawdust industrial residue could be hazardous to the environment, but it becomes pozzolanic when incinerated. Thus, harnessing sawdust ash (SDA) and lateritic soil as road construction materials for low-cost roads is apt. However, modeling and prediction of the properties of soils treated with SDA h...

Full description

Saved in:
Bibliographic Details
Main Authors: James Ukpai, Ugochukwu Okonkwo
Format: Article
Language:English
Published: Unviversity of Technology- Iraq 2025-05-01
Series:Engineering and Technology Journal
Subjects:
Online Access:https://etj.uotechnology.edu.iq/article_187304_b2d79945d6ae914345a2e7b9cb4687fe.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849704772601380864
author James Ukpai
Ugochukwu Okonkwo
author_facet James Ukpai
Ugochukwu Okonkwo
author_sort James Ukpai
collection DOAJ
description Sawdust industrial residue could be hazardous to the environment, but it becomes pozzolanic when incinerated. Thus, harnessing sawdust ash (SDA) and lateritic soil as road construction materials for low-cost roads is apt. However, modeling and prediction of the properties of soils treated with SDA have received very little attention. This study predicted the compressive strength of lateritic soil treated with cement and SDA using an Artificial Neural Network (ANN). The soil was found to belong to A-2-4(0) in the AASHTO rating and clayey sand (SC) in the Unified Soil Classification System. The soil minerals composition was conducted using an x-ray diffraction technique, which comprises non-clay minerals such as quartz, albite, orthoclase, goethite, and muscovite, as well as clay minerals like clinochlore. The tests carried out on the treated soil were compaction tests, California Bearing Ratio (CBR), unconfined compressive strength (UCS), and durability tests. The maximum dry density of the reinforced soil was reduced, whereas the optimum moisture content increased with an increase in SDA content. At the addition of cement and SDA of up to 8% and 30%, respectively, the strength properties like the CBR and UCS at 7 days of curing increased by up to 1965.46% and 989.25%, respectively. Also, the reinforced soil satisfied the durability requirements. The treated soil could be used as the sub-base of road pavement structures. The ANN modeling dependably predicted the unconfined compressive strength at 7 days of curing of the reinforced soil with coefficients of correlation of 0.989 and 0.996 for training and testing, respectively.
format Article
id doaj-art-bd7bb6e9d11f4224bac5b1236026e48c
institution DOAJ
issn 1681-6900
2412-0758
language English
publishDate 2025-05-01
publisher Unviversity of Technology- Iraq
record_format Article
series Engineering and Technology Journal
spelling doaj-art-bd7bb6e9d11f4224bac5b1236026e48c2025-08-20T03:16:39ZengUnviversity of Technology- IraqEngineering and Technology Journal1681-69002412-07582025-05-0143537438510.30684/etj.2025.156269.1873187304Using artificial neural networks to predict the compressive strength of cement and sawdust ash-treated lateritic soilJames Ukpai0Ugochukwu Okonkwo1Civil Engineering Dept., Michael Okpara University of Agriculture Umudike, Abia State, Nigeria.Civil Engineering Dept., Michael Okpara University of Agriculture Umudike, Abia State, Nigeria.Sawdust industrial residue could be hazardous to the environment, but it becomes pozzolanic when incinerated. Thus, harnessing sawdust ash (SDA) and lateritic soil as road construction materials for low-cost roads is apt. However, modeling and prediction of the properties of soils treated with SDA have received very little attention. This study predicted the compressive strength of lateritic soil treated with cement and SDA using an Artificial Neural Network (ANN). The soil was found to belong to A-2-4(0) in the AASHTO rating and clayey sand (SC) in the Unified Soil Classification System. The soil minerals composition was conducted using an x-ray diffraction technique, which comprises non-clay minerals such as quartz, albite, orthoclase, goethite, and muscovite, as well as clay minerals like clinochlore. The tests carried out on the treated soil were compaction tests, California Bearing Ratio (CBR), unconfined compressive strength (UCS), and durability tests. The maximum dry density of the reinforced soil was reduced, whereas the optimum moisture content increased with an increase in SDA content. At the addition of cement and SDA of up to 8% and 30%, respectively, the strength properties like the CBR and UCS at 7 days of curing increased by up to 1965.46% and 989.25%, respectively. Also, the reinforced soil satisfied the durability requirements. The treated soil could be used as the sub-base of road pavement structures. The ANN modeling dependably predicted the unconfined compressive strength at 7 days of curing of the reinforced soil with coefficients of correlation of 0.989 and 0.996 for training and testing, respectively.https://etj.uotechnology.edu.iq/article_187304_b2d79945d6ae914345a2e7b9cb4687fe.pdfcementcompressive strengthmachine learninglateritic soil and sawdust ash
spellingShingle James Ukpai
Ugochukwu Okonkwo
Using artificial neural networks to predict the compressive strength of cement and sawdust ash-treated lateritic soil
Engineering and Technology Journal
cement
compressive strength
machine learning
lateritic soil and sawdust ash
title Using artificial neural networks to predict the compressive strength of cement and sawdust ash-treated lateritic soil
title_full Using artificial neural networks to predict the compressive strength of cement and sawdust ash-treated lateritic soil
title_fullStr Using artificial neural networks to predict the compressive strength of cement and sawdust ash-treated lateritic soil
title_full_unstemmed Using artificial neural networks to predict the compressive strength of cement and sawdust ash-treated lateritic soil
title_short Using artificial neural networks to predict the compressive strength of cement and sawdust ash-treated lateritic soil
title_sort using artificial neural networks to predict the compressive strength of cement and sawdust ash treated lateritic soil
topic cement
compressive strength
machine learning
lateritic soil and sawdust ash
url https://etj.uotechnology.edu.iq/article_187304_b2d79945d6ae914345a2e7b9cb4687fe.pdf
work_keys_str_mv AT jamesukpai usingartificialneuralnetworkstopredictthecompressivestrengthofcementandsawdustashtreatedlateriticsoil
AT ugochukwuokonkwo usingartificialneuralnetworkstopredictthecompressivestrengthofcementandsawdustashtreatedlateriticsoil