In the heart of Tehran, researchers are revolutionizing the way we think about concrete and industrial waste. Ehsan Mohsennia, a civil engineering expert from The Sharif University of Technology, has developed an advanced machine learning framework that could significantly impact the energy sector and construction industry. His work, published in Case Studies in Construction Materials, focuses on predicting the compressive strength and ultrasonic pulse velocity of concrete incorporating various industrial by-products (IBPs).
Mohsennia’s innovative approach integrates both physical mix parameters and detailed chemical compositions of cement and multiple IBPs, including fly ash, ground granulated blast furnace slag (GGBFS), silica fume, and metakaolin. This dual-stage machine learning framework is a game-changer, as it not only predicts ultrasonic pulse velocity (UPV) but also uses this prediction to enhance the accuracy of compressive strength (CS) predictions. “This configuration is underexplored in previous studies,” Mohsennia explains, “but it aligns perfectly with our goal of maximizing the use of diverse IBPs and reducing waste.”
The implications for the energy sector are profound. Power plants and industrial facilities generate vast amounts of by-products like fly ash and slag. By incorporating these materials into concrete formulations, companies can reduce disposal costs and environmental impact while creating more sustainable building materials. “Our model is designed to optimize material usage,” Mohsennia states, “contributing directly to sustainability goals and improving construction practices.”
The research utilized a robust dataset comprising 162 structured IBP concrete samples and 524 data points from existing literature. Among the models tested, the CatBoost (CB) algorithm, optimized with the Whale Optimization Algorithm (WOA), showed exceptional predictive performance. The CB-WOA model achieved impressive coefficients of determination (R²) values of 0.96 and 0.94 for UPV during training and testing, and 0.99 and 0.98 for CS, respectively. These results highlight the potential of UPV as a non-destructive metric for assessing concrete structural integrity and predicting material strength.
One of the standout features of this study is its comprehensive sensitivity analysis, which underscores the importance of targeted feature selection. This not only enhances prediction accuracy but also supports sustainability efforts by optimizing material usage. To make these findings accessible, Mohsennia and his team have developed a user-friendly web application, available at https://ibp-concrete-upv-and-cs-prediction-mohsennia-javid-toufigh.streamlit.app/.
The commercial impacts of this research are far-reaching. Construction companies can adopt these predictive models to create more durable and sustainable concrete structures, reducing the need for frequent repairs and replacements. For the energy sector, this means lower operational costs and a smaller carbon footprint. As Mohsennia puts it, “This technology has the potential to reshape how we think about industrial waste and concrete production, paving the way for a more sustainable future.”
The future of construction and energy production is looking greener and more efficient, thanks to groundbreaking research from The Sharif University of Technology. As industries strive for sustainability, Mohsennia’s work offers a blueprint for integrating advanced machine learning techniques into practical applications, driving innovation and environmental stewardship.