SavedAdd your journal

Journals / Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi / 2026 / Cilt: 16

Stabilization of sandy soils using olive ash and blast furnace slag: Experimental and machine learning approach for unconfined compressive strength prediction

Kumlu zemin stabilizasyonunda zeytin külü ve yüksek fırın cürufu kullanımı: Serbest basınç dayanımının deneysel ve makine öğrenmesi ile tahmini

Ömer Yeşiltepe(Kilis 7 Aralık Üniversitesi)0000-0002-9337-9157
Pages
—
Publication Date
2026-03-16
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

Unconfined Compressive Strength (UCS) test results showed that the maximum UCS value of 10.4 MPa was achieved at 30% OA content after 28 days of curing.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:The utilization of waste materials for soil improvement presents a sustainable alternative to conventional stabilization techniques, providing both economic benefits and a reduction in environmental impact through material recycling.
📊
Key Findings & Data:The sandy soil content was kept constant, while the OA content was varied from 0% to 50% by weight of BFS across six different levels, and specimens were cured for 1, 7, and 28 days.
💡
Conclusion & Impact:Unconfined Compressive Strength (UCS) test results showed that the maximum UCS value of 10.4 MPa was achieved at 30% OA content after 28 days of curing.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Zemin stabilizasyonu · Zeytin Külü · Yüksek Fırın Cürufu

Abstract

The utilization of waste materials for soil improvement presents a sustainable alternative to conventional stabilization techniques, providing both economic benefits and a reduction in environmental impact through material recycling. This study investigates the potential of olive ash (OA) as an alkaline activator for the pozzolanic activation of ground granulated blast furnace slag (BFS) to enhance the mechanical strength of sandy soils. The sandy soil content was kept constant, while the OA content was varied from 0% to 50% by weight of BFS across six different levels, and specimens were cured for 1, 7, and 28 days. Standard compaction tests indicated that increasing OA content led to higher optimum moisture content (w_opt) values and lower maximum dry unit weights (γ_kmax). Unconfined Compressive Strength (UCS) test results showed that the maximum UCS value of 10.4 MPa was achieved at 30% OA content after 28 days of curing. To improve the quantitative interpretation of experimental results and address the limitations associated with a limited dataset, a novel hybrid experimental–Machine Learning (ML) framework incorporating Support Vector Regression (SVR), Random Forest (RF), and Multi-Layer Perceptron (MLP) models was employed for UCS prediction and mixture optimization. Among the evaluated models, SVR exhibited excellent predictive accuracy, achieving a coefficient of determination R² = 0.949 and a root mean square error RMSE = 0.703 MPa. Feature importance analysis performed using the RF model revealed that curing time (78%) was the most influential parameter governing strength development, significantly outweighing the contribution of OA content (22%), highlighting the dominant role of chemical reactions. Furthermore, SVR-based optimization identified the optimum OA content corresponding to maximum predicted UCS as 30.7% at 28 days of curing, thereby mathematically validating the experimental findings. Overall, the results demonstrate that OA–BFS mixtures constitute an environmentally sustainable and technically reliable binder alternative, offering strong potential for practical sandy soil stabilization applications.

Özet

Atık olarak ortaya çıkan malzemelerin zeminlerin iyileştirilmesinde kullanılması hem ekonomik olarak fayda sağlaması hem de atık malzemenin geri dönüştürülerek çevreye verdiği etkinin azaltılması bakımından, günümüzde kullanılan iyileştirme yöntemlerine iyi bir alternatif oluşturmaktadır. Bu çalışmanın temel amacı, zeytin külünü (ZK) sürdürülebilir bir alkali aktivatör olarak kullanarak, kumlu zeminlerin yüksek fırın cürufu (YFC) ile stabilizasyon potansiyelini deneysel ve makine öğrenmesi yöntemleriyle kapsamlı bir şekilde incelemektir. Kum zemin miktarı sabit tutularak, (ZK) 'nin (YFC) ağırlığına oranı %0, %10, %20, %30, %40 ve %50’ye kadar altı farklı düzeyde seçilmiş ve numuneler 1, 7 ve 28 günlük kür sürelerine tabi tutulmuştur. Standart kompaksiyon deneyleri sonucunda, ZK içeriğindeki artışın optimum su muhtevası (w_opt) değerlerini yükselttiği ve maksimum kuru birim hacim ağırlık (γ_kmax) değerlerini düşürdüğü belirlenmiştir. Serbest Basınç Dayanımı (UCS) deneyleri, 28 günlük kür sonunda %30 ZK oranıyla 10.4 MPa ile maksimum UCS değerine ulaşıldığını göstermiştir. Çalışmanın nicel derinliğini artırmak amacıyla, kısıtlı deneysel verilere dayalı UCS tahmin modellemesi ve karışım optimizasyonu için Makine Öğrenmesi (ML) algoritmaları (SVR, RF, MLP) uygulanmıştır. Destek Vektör Regresyonu (SVR) modeli, R^2= 0.949 belirleyicilik katsayısı ve RMSE = 0.703 MPa hata payı ile en üstün tahmin performansını sergilemiştir. Özellik önem analizi için Rastgele Orman (RF) modeli kullanılmış ve kür süresinin (%78 etki) dayanım üzerindeki etkisinin, ZK oranından (%22 etki) daha fazla olduğunu ortaya koyarak, dayanım gelişiminin kimyasal reaksiyonlara bağımlılığını teyit etmiştir. SVR tabanlı optimizasyon sonuçları, maksimum tahmini dayanımı veren optimum ZK oranının 28 günlük kür için %30.7 olduğunu göstererek deneysel bulguyu matematiksel olarak doğrulamıştır. Elde edilen bu hibrit sonuçlar, ZK ve YFC karışımının kumlu zemin iyileştirmesi için yüksek potansiyele sahip çevreci ve uygulanabilir bir alternatif olduğunu göstermektedir.

🔖

Cite This Article

Yeşiltepe, Ö. (2026). Stabilization of sandy soils using olive ash and blast furnace slag: Experimental and machine learning approach for unconfined compressive strength prediction. Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi, Cilt: 16. https://doi.org/10.28948/ngumuh.1833059

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (30)

  1. A. Kordjazi, F. Pooya Nejad, and M. B. Jaksa, Prediction of ultimate axial load-carrying capacity of piles using a support vector machine based on CPT data. Computers and Geotechnics, 55, 91–102, 2014. https://doi.org/10.1016/j.compgeo.2013.08.001
  2. H. Bilici, D. V. Okur, M. Türköz ve H. Savaş, Kil zeminin dayanımı üzerinde uçucu kül ve yüksek fırın cürufu katkılarının etkisi ve karşılaştırmalı analizi. Bitlis Eren Üniversitesi Fen Bilimleri Dergisi, 9(2), 910-919, 2020. https://doi.org/10.17798/bitlisfen.598525
  3. T. Geçkil, T. Sarıcı ve Ö. Çetkin, Killi kum zeminlerin dayanımı üzerinde yüksek fırın cürufunun etkileri. Osmaniye Korkut Ata Üniversitesi Fen Bilimleri Enstitüsü Dergisi, 6(2), 1159-1174, 2023.
  4. F. Maximiliano, C. Costanza, Z. Azzurra, S. Paola, P. B. Elia, P. Leonardo, C. Franco, C. Gianluca, and C. Paola, Use of biomass ash in reinforced clayey soil: a multiscale analysis of solid-state reactions. Recycling, 8(1), 5, 2023. https://doi.org/10.3390/recycling8010005
  5. M. Tanyıldızı, S. Nasıroğlu, A. Zeybek, and İ. Gökalp, Sustainable use of waste hazelnut shell ash in road construction to stabilize expansive subgrade. Construction and Building Materials, 432, 1-15, 2024. https://doi.org/10.1016/j.conbuildmat.2024.136663
  6. G. Bilgen, A. Kavak ve Ö. F. Çapar, Düşük plastisiteli bir kilde katkı olarak çelikhane curufunun kullanılması ve kireç ile etkileşimi. Karaelmas Fen ve Mühendislik Dergisi, 2(2), 30-38, 2012.
  7. T. Geçkil, M. M. Tanyıldızı ve E. S. Yıldıran, Yüksek fırın cürufu ile stabilize edilmiş killi bir zeminin yol esnek üstyapı tabaka kalınlıklarına ve maliyetine etkileri. Fırat Üniversitesi Mühendislik Bilimleri Dergisi, 32(2), 509-520, 2020. https://doi.org/10.31590/ejosat.932204
  8. S. M. M. Pinheiro, A. Font, L. Soriano, M. M. Tashima, J. Monzo, M. V. Borrachero, and J. Paya, Olive-stone biomass ash (OBA): An alternative alkaline source for the blast furnace slag activation. Construction and Building Materials, 178, 327–338, 2018. https://doi.org/10.1016/j.conbuildmat.2018.05.157
  9. H. K. Preetham, and S. Nayak, Geotechnical investigations on marine clay stabilized using granulated blast furnace slag and cement. International Journal of Geosynthetics and Ground Engineering, 5(4), 1-12, 2019. https://doi.org/10.1007/s40891-019-0179-5
  10. B. Lakkimsetti, and S. Nayak, Experimental investigation and performance evaluation of lithomargic clay stabilized with granulated blast furnace slag and calcium chloride. International Journal of Geosynthetics and Ground Engineering, 8(1), 1-16, 2022. https://doi.org/10.1007/s40891-022-00355-5
  11. H. Mujtaba, T. Aziz, K. Farooq, N. Sivakugan, and B. M. Das, Improvement in engineering properties of expansive soils using ground granulated blast furnace slag. Journal of the Geological Society of India, 92(3), 357-362, 2018. https://doi.org/10.1007/s12594-018-1019-2
  12. N. M. Al-Akhras, K. M. Al-Akhras, and M. F. Attom, Performance of olive waste ash concrete exposed to elevated temperatures. Fire Safety Journal, 4, 370–375, 2009. https://doi.org/10.1016/j.firesaf.2008.08.006
  13. A. Font, L. Soriano, J. C. B. Moraes, M. M. Tashima, J. Monzo, M. V. Borrachero, J. Paya, A 100% waste-based alkali-activated material by using olive-stone biomass ash (OBA) and blast furnace slag (BFS). Materials Letters, 203, 46–49, 2017. https://doi.org/10.1016/j.matlet.2017.05.129
  14. N. S. Arabi, A. Qadi, S. K. Taisir, M. A. Haddad, and O. A. Al-Rababah, Investigating the Effect of Olive Husk Ash on the Properties of Asphalt Concrete Mixture. Annales de Chimie - Science des Matériaux, 45(1), 2021. https://doi.org/10.18280/acsm.450102
  15. B. Bagriacik, E. Kahraman, and A. M. Kılıç, Large‑scale experimental model study: soil improvement with olive mill solid residue ash. International Journal of Environmental Science and Technology, 19, 12581–12592, 2022. https://doi.org/10.1007/s13762-022-04150-7
  16. B. Belaidi, A. Messai, C. Belebchouche, M. Boutlikht, K. Hebbache, A. Douadi, and L. Moretti, Physical, mechanical, and durability performance of olive pomace ash in eco-friendly mortars. Materials, 18, 1-17, 2025. https://doi.org/10.3390/ma18112667
  17. R. Sharma, and R. P. Tiwari, Random forest algorithm for the strength prediction of geopolymer stabilized clayey soil. Sustainability, 15(2), 1408, 2023. https://doi.org/10.3390/su15021408
  18. B. T. Pham, T. Trung Nguyen-Thoi, L. Hai-Bang, M. D. Nguyen, N. Al-Ansari, V. Q. Tran, and T. T. Le, Extreme learning machine based prediction of soil shear strength: a sensitivity analysis using monte carlo simulations and feature backward elimination. Sustainability, 12(6), 2339, 2020. https://doi.org/10.3390/su12062339
  19. A. B. Ekmen, H. M. Algin, and M. Özen, Strength and stiffness optimisation of fly ash-admixed DCM columns constructed in clayey silty sand. Transportation Geotechnics, 24, 100364, 2020. https://doi.org/10.1016/j.trgeo.2020.100364
  20. ASTM D698-12. Standart test methods for laboratory compaction characteristic of soil using standart effort. ASTM International. West Conshohocken, PA., 2012. https://doi.org/10.1520/D0698-12E02
  21. ASTM D2166. Standard test method for unconfined compressive strength of cohesive soil. ASTM International. West Conshohocken, PA., 2016. https://doi.org/10.1520/D2166_D2166M-13
  22. Y. B. Dibike, S. Velickov, D. P. Solomatine, and M. B. Abbott, Model induction with support vector machines: Introduction and applications. Journal of Computing in Civil Engineering, 15(3), 208–216, 2001. https://doi.org/10.1061/(ASCE)0887-3801(2001)15:3(208)
  23. P. Samui, Prediction of friction capacity of driven piles in clay using the support vector machine. Canadian Geotechnical Journal, 45(2), 288–295, 2008. https://doi.org/10.1139/T07-072
  24. P. Samui, Prediction of pile bearing capacity using support vector machine. International Journal of Geotechnical Engineering, 5(1), 95–102, 2011. https://doi.org/10.3328/IJGE.2011.05.01.95-102
  25. A. Kordjazi, F. Pooya Nejad, and M. B. Jaksa, Prediction of ultimate axial load‑carrying capacity of piles using a support vector machine based on CPT data. Computers and Geotechnics, 55, 91–102, 2014. https://doi.org/10.1016/j.compgeo.2013.08.001
  26. H. A. Zeini, D. Al-Jeznawi, H. Imran, L. F. A. Bernardo, Z. Al-Khafaji, and K. A. Ostrowski, Random Forest Algorithm for the Strength Prediction of Geopolymer Stabilized Clayey Soil. Sustainability, 15(2), 2023. https://doi.org/10.3390/su15021408
  27. A. U. Uzer, Efficient prediction of compressive strength in geotechnical engineering using artificial neural networks. Turkish Journal of Engineering, 8(3), 2024. https://doi.org/10.31127/tuje.1415931
  28. G. Bilgiç, and A. E. Gök, Comparison of Random Forest and Support Vector Regression Models in Predicting Hydrogen Production Process from Biomass. Cukurova University, Journal of the Faculty of Engineering, 39(2), 475-488, 2024. https://doi.org/10.21605/cukurovaumfd.1514518
  29. W. Zhang, C. Wu, H. Zhong, Y. Li, and L. Wang, Prediction of undrained shear strength using extreme gradient boosting and random forest based on Bayesian optimization. Geoscience Frontiers 12, 469–477, 2021. https://doi.org/10.1016/j.gsf.2020.03.007
  30. A. H. Saad, H. Nahazanan, B. Yusuf, S. F. Toha, A. Alnuaim, A. El-Mouchi, M. Elseknidy, and A. A. Mohammed, A Systematic Review of Machine Learning Techniques and Applications in Soil Improvement Using Green Materials. Sustainability, 15, 9738, 2023. https://doi.org/10.3390/su15129738