•  
  •  
 

Abstract

Recycling of ceramic waste in the construction sector is a promising option for reducing the environmental impacts of the concrete industry. The main objective of this study is the elaboration of a solid predictive model for ceramic waste based self-compacting (SCC) concrete properties, using artificial neural networks (ANN). This further, reduces its environmental impact without compromising satisfactory performance. An experimental approach was adopted with different substitution ratios: 0 to 30 wt% and particle sizes ranging from 3-8 mm and 8-15 mm. Tested properties of modified SCC as compressive strength at 7 and 28 days (fc7, fc28), spread (L), sieve stability segregation (ST), porosity (n), density (ρ) and absorption water (A) were investigated. Addition of ceramic waste affects the properties of the obtained material (SCC). The compressive strengths of all mixes composed were larger than that of the control sample. Despite the results workability property, which remain unsatisfactory; this shows that this parameter was within controlled limits. Moreover, predictive modeling of SCC properties was implemented through the artificial neural network (ANN). These models can predict the physico-mechanical properties according to the substitution percentages. As per the ANN, strong correlation between the experimental and predicted value proved the efficacy of the ANN for optimising the multi-response.This research highlights the potential of ceramic waste to produce durable and efficient SCC while leveraging advanced modeling tools to optimize formulations. It opens promising prospects for sustainable waste management in the construction industry.

Download

Share

COinS