Gene Expression Programming for Predicting Compressive and Flexural Strength of Concrete with Rice Husk Ash and Sawdust Ash

• Author(s): Oyesanya, O. G.
• Paper ID: 1718351
• Page: 4767-4778
• Published Date: 29-05-2026
• Published In: Iconic Research And Engineering Journals
• Publisher: IRE Journals
• e-ISSN: 2456-8880
• Volume/Issue: Volume 9 Issue 11 May-2026

Abstract

This study presents the development and validation of a Gene Expression Programming (GEP) model for predicting the compressive and flexural strength of sustainable concrete incorporating Rice Husk Ash (RHA) and Sawdust Ash (SDA) as supplementary cementitious materials. A dataset comprising 123 concrete mix samples was assembled, with 80% used for model training and 20% reserved for independent testing. The GEP model was carefully configured using multiple chromosomes, optimized genetic operators, and a function set consisting of arithmetic and nonlinear operators to effectively capture the complex interactions governing cement hydration and pozzolanic reactions. Model performance evaluation demonstrated strong predictive capability. For compressive strength, the model achieved a test coefficient of determination (R²) of 0.775, RMSE of 8.15 N/mm², and MAE of 6.32 N/mm². Flexural strength prediction showed improved accuracy, with an R² of 0.841, RMSE of 0.71 N/mm², and MAE of 0.52 N/mm². Five-fold cross-validation further confirmed model robustness and generalization. The evolved expression trees generated explicit mathematical relationships, providing valuable interpretability and insight into the nonlinear influence of mix constituents on mechanical properties. Sensitivity analysis using SHAP identified cement and fine aggregate as the most influential variables for compressive strength, while fine aggregate, water, and coarse aggregate were dominant in flexural strength prediction. RHA exhibited consistently greater influence than SDA, indicating its higher pozzolanic reactivity. The results demonstrate that Gene Expression Programming is an effective and interpretable tool for modeling the mechanical performance of sustainable concrete, offering both reliable prediction accuracy and transparent analytical expressions for engineering applications.

Keywords

Gene Expression Programming, Sustainable Concrete, Rice Husk Ash, Sawdust Ash, Compressive Strength, Flexural Strength, Symbolic Regression

Citations

IRE Journals:
Oyesanya, O. G. "Gene Expression Programming for Predicting Compressive and Flexural Strength of Concrete with Rice Husk Ash and Sawdust Ash" Iconic Research And Engineering Journals Volume 9 Issue 11 2026 Page 4767-4778 https://doi.org/10.64388/IREV9I11-1718351

IEEE:
Oyesanya, O. G. "Gene Expression Programming for Predicting Compressive and Flexural Strength of Concrete with Rice Husk Ash and Sawdust Ash" Iconic Research And Engineering Journals, 9(11) https://doi.org/10.64388/IREV9I11-1718351