Loading...

Evaluating strength properties of Eco-friendly Seashell-Containing Concrete: Comparative analysis of hybrid and ensemble boosting methods based on environmental effects of seashell usage

Sadaghat, B ; Sharif University of Technology | 2024

19 Viewed
  1. Type of Document: Article
  2. DOI: 10.1016/j.engappai.2024.108388
  3. Publisher: 2024
  4. Abstract:
  5. In the dynamic field of concrete technology, a discernible shift towards sustainability is evident, prompted by the need to minimize reliance on natural resources and reduce the environmental impact of the concrete industry. This transformation involves exploring alternative materials, specifically recycled waste and by-products like seashells, to create eco-friendly green concrete. Seashell waste, abundant in coastal areas, proves promising as a partial or total substitute for fine and coarse aggregates, while seashell powder acts as a sustainable binder, significantly reducing the carbon footprint of cement in concrete production. This study employs three boosting algorithms (Extreme Gradient Boosting, Histogram Gradient Boosting, and Categorical Boosting) and recent optimization algorithms to develop predictive machine learning (ML) models for seashell-containing concrete's compressive, flexural, and tensile strengths. Two ensemble methods (Weighted Averaging and Dempster-Shafer) enhance model robustness by integrating optimization algorithms, promoting compatibility with diverse datasets. Results show that the Quadratic Interpolation Optimization algorithm notably improves accuracy, particularly in Extreme Gradient Boosting, boosting prediction accuracy by 28%, 40%, and 24% in compressive, flexural, and tensile strength estimation. Beyond technical advancements, the research addresses global concerns, highlighting the environmental benefits of an equal proportion of seashells and cement for optimal strength in practical applications, particularly in compressive and tensile strength. © 2024 Elsevier Ltd
  6. Keywords:
  7. Machine learning ; Strength properties ; Adaptive boosting ; Carbon footprint ; Cements ; Compressive strength ; Environmental impact ; Sustainable development ; Tensile strength ; Boosting method ; Comparative analyzes ; Eco-friendly ; Ensemble learning ; Gradient boosting ; Green concrete ; Machine-learning ; Optimization algorithms ; Seashell-containing concrete ; Strength property
  8. Source: Engineering Applications of Artificial Intelligence ; Volume 133 , 2024 ; 09521976 (ISSN)
  9. URL: https://www.sciencedirect.com/science/article/abs/pii/S0952197624005463