In a groundbreaking development for sustainable construction, researchers have harnessed the power of machine learning to predict the compressive strength of concrete incorporating waste powders from construction materials, even when exposed to elevated temperatures. This innovative approach, detailed in a study led by Islam N. Fathy of the Construction and Building Engineering Department at the October High Institute for Engineering & Technology, opens new avenues for green construction practices and could significantly impact the energy sector.
The study, published in the journal Scientific Reports (translated to English as “Nature Scientific Reports”), focuses on the use of waste marble and granite powders as partial cement replacements in concrete. These powders, often discarded as waste, can now be repurposed, reducing the environmental footprint of concrete production. “The addition of these waste powders not only promotes sustainability but also enhances the mechanical properties of concrete,” Fathy explains.
The research team employed three machine learning models—extreme gradient boosting (XGBoost), random forest (RF), and M5P—to predict the impact of elevated temperatures on the compressive strength of concrete modified with these waste powders. The models were trained on a dataset of 324 tested cubic specimens, with variables including the dose of waste granite powder (GWP), waste marble powder (MWP), temperature, and duration of exposure.
The XGBoost model emerged as the most accurate, achieving an impressive R2 value of 0.9989 and demonstrating the lowest prediction errors. “The XGB model’s performance was exceptional, outperforming the other models in predicting the compressive strength of concrete,” Fathy notes. The study also revealed that while GWP and MWP positively influence compressive strength, temperature has the most negative impact.
This research has profound implications for the construction and energy sectors. By optimizing the use of waste powders in concrete, the industry can reduce its reliance on traditional cement production, which is energy-intensive and a significant source of carbon emissions. “This technology can help us build more sustainable structures while maintaining the integrity and strength of the materials,” Fathy adds.
The development of a graphical user interface (GUI) for predicting the compressive strength of concrete containing GWP and MWP subjected to elevated temperatures further enhances the practical application of this research. This tool can assist researchers and industry professionals in making informed decisions, ultimately leading to more efficient and sustainable construction practices.
As the world grapples with the challenges of climate change and resource depletion, innovations like this are crucial. The integration of machine learning in construction materials science not only advances our understanding of material behavior but also paves the way for more sustainable and resilient infrastructure. This research is a testament to the power of interdisciplinary collaboration and the potential of technology to drive positive change in the built environment.