In a significant stride towards sustainable construction, researchers have developed advanced artificial intelligence models to predict the compressive strength of concrete made with waste glass powder. This innovation, led by Sushant Poudel from the Department of Civil and Environmental Engineering at Lamar University, Texas, promises to revolutionize the construction industry by reducing cement consumption and landfill waste.
The study, published in the journal ‘Green Technologies and Sustainability’ (which translates to ‘Зелене Технології та Стійкість’ in Ukrainian), addresses a critical gap in previous research. “Most prior studies relied on limited datasets and single machine-learning algorithms, which restricted their generalization,” Poudel explains. To overcome this, the team compiled a comprehensive dataset of 337 experimental results, incorporating key variables such as waste glass powder size, replacement level, water-to-cement ratio, and chemical composition.
The researchers employed five advanced ensemble algorithms, including Gradient Boosting Regressor and Extreme Gradient Boosting Regressor. After rigorous training and optimization using Bayesian hyperparameter tuning and 10-fold cross-validation, the models demonstrated exceptional predictive ability. Notably, the CatBoost Regressor achieved the highest testing accuracy, with an R2 value of 0.96 and a root mean square error (RMSE) of just 2.34 MPa.
The implications for the construction industry are profound. By accurately predicting the compressive strength of waste glass concrete, the models enable engineers to design more sustainable structures with reduced environmental impact. “This research establishes a practically oriented framework that advances sustainable concrete design,” Poudel states.
The team also developed a user-friendly graphical user interface (GUI) based on the CatBoost model, allowing practitioners to easily predict compressive strength for various mix designs. This tool could significantly enhance the adoption of waste glass concrete in commercial projects, contributing to a circular economy and reducing the carbon footprint of the construction sector.
As the world grapples with the challenges of climate change and resource depletion, innovations like these are crucial. Poudel’s research not only showcases the potential of machine learning in construction but also highlights the importance of interdisciplinary collaboration in driving sustainable development.
The study’s findings could shape future developments in the field by encouraging further research into alternative materials and advanced predictive models. As the construction industry continues to evolve, the integration of AI and sustainable practices will likely become a cornerstone of modern engineering, paving the way for a greener, more efficient future.