Hybrid Machine Learning Modeling of Strength in Sustainable Basalt Fiber-reinforced Concrete

Strategic Insights -

Researchers at King Saud University have made a significant breakthrough in sustainable construction materials by developing a comprehensive machine learning framework to predict the mechanical performance of basalt fiber-reinforced concrete (BFRC). This innovative approach optimizes the use of BFRC by identifying the optimal mix design parameters and fiber characteristics, reducing the reliance on empirical trial-and-error methods and mitigating the environmental impact of conventional cementitious materials. According to the study, the optimal fiber dosage for enhancing both compressive strength and split tensile strength is approximately 0.12% by volume.

Key Takeaways:

  • The study addresses the growing need for sustainable construction materials by applying machine learning to predict the mechanical performance of BFRC.
  • A diverse dataset incorporating critical mix design parameters and fiber characteristics was utilized to develop a comprehensive machine learning framework.
  • The optimal model yielded R2 of 0.920 for compressive strength and 0.973 for split tensile strength, with feature importance analysis identifying W/B and Df as primary determinants of strength.
  • Partial dependence plots highlighted nonlinear interactions between fiber characteristics and concrete performance.
  • A fiber dosage of approximately 0.12% by volume was identified as optimal for enhancing both compressive strength and split tensile strength.
  • The research concluded that hybrid machine learning modeling can reduce reliance on empirical trial-and-error methods and mitigate the environmental impact of conventional cementitious materials.
  • The study has been peer-reviewed and published in Materials Today Communications.
  • King Saud University provided financial support for the research.
  • Yassir M. Abbas, Professor at King Saud University, was involved in the research.

Statistics:

  • The optimal fiber dosage for enhancing both compressive strength and split tensile strength is approximately 0.12% by volume.
  • The R2 value for compressive strength is 0.920, indicating a strong predictive accuracy model.
  • The R2 value for split tensile strength is 0.973, indicating an exceptionally strong predictive accuracy model.
  • The feature importance analysis identified W/B and Df as primary determinants of strength, accounting for 70% of the variability in the dataset.

Sources:

  • NewsRx. Studies from King Saud University Update Current Data on Sustainability Research (Hybrid Machine Learning Modeling of Strength In Sustainable Basalt Fiber-reinforced Concrete). Ecology, Environment & Conservation. June 20, 2025; p 553.
  • [Hybrid Machine Learning Modeling of Strength In Sustainable Basalt Fiber-reinforced Concrete. Materials Today Communications, 2025;46.