In a groundbreaking study published in the journal *Engineering Reports* (translated from Bengali as “Engineering Reports”), researchers have demonstrated a novel approach to predicting the strength of concrete made from textile effluent sludge (TES), a byproduct of the textile industry. The research, led by Usmi Akter of the Department of Civil Engineering at Khulna University of Engineering & Technology in Bangladesh, explores the potential of TES as a supplementary cementitious material (SCM) and employs advanced machine learning (ML) and deep learning (DL) techniques to optimize its use in concrete applications.
The study examines the predictive capabilities of various ML and DL models, including random forest (RF), support vector machine (SVM), extreme gradient boost (XGBoost), and K-nearest neighbors (KNN), as well as convolutional neural networks (CNN) and long short-term memory networks (LSTM). The research also introduces hybrid models, combining RF with SVM and CNN with LSTM, to enhance predictive accuracy.
“Our goal was to develop a robust framework for predicting the compressive strength (CS) and tensile strength (TS) of concrete utilizing TES,” Akter explained. “By leveraging advanced machine learning techniques, we aimed to demonstrate the viability of TES as a sustainable and effective SCM.”
The findings reveal that the standalone LSTM model achieved the highest accuracy in CS prediction, with an R2 value of 0.90, outperforming all other models. Meanwhile, the standalone SVM model exhibited the highest accuracy for TS prediction, with an R2 value of 0.89. Notably, the hybrid models also showed competitive performance, with the hybrid CNN-LSTM model achieving the second-highest accuracy for CS prediction (R2=0.89) and the hybrid RF-SVM model surpassing the CNN-LSTM model in TS prediction (R2=0.88).
“This research not only promotes the high-scale utilization of TES as SCM in concrete but also contributes to environmental pollution control,” Akter added. “By providing a reliable method for predicting the strength of TES-based concrete, we hope to encourage further adoption of this sustainable material in the construction industry.”
The implications of this research are significant for the construction and energy sectors. As the demand for sustainable and eco-friendly building materials continues to grow, the use of TES as an SCM offers a promising solution for reducing waste and minimizing environmental impact. The advanced predictive models developed in this study can help engineers and researchers optimize the use of TES in concrete applications, ensuring both strength and durability.
Moreover, the integration of machine learning and deep learning techniques in this research highlights the potential for these technologies to revolutionize the construction industry. By providing accurate and efficient predictive models, these techniques can enhance the design and construction processes, leading to more sustainable and cost-effective building practices.
As the construction industry continues to evolve, the findings of this study offer valuable insights into the potential of TES as a sustainable SCM and the role of advanced predictive models in optimizing its use. By embracing these innovations, the industry can move towards a more sustainable and environmentally responsible future.