Researchers have developed a novel approach to transform agricultural waste, such as groundnut shells, into sustainable building materials. By mixing this waste with cement, fly ash, and hydrated lime, they have created a new type of “Bio-Brick” that not only reduces environmental impact but also meets construction standards. The study utilized advanced machine learning techniques like Response Surface Methodology (RSM) and Artificial Neural Networks (ANN) to optimize the properties of these Bio-Bricks, achieving exceptional accuracy in predicting their compressive strength, dry density, and water absorption. This innovative solution paves the way for a more sustainable future in the construction industry.
Tackling the Environmental Impact of Construction
The construction industry has long been a major contributor to environmental pollution, with its heavy reliance on limited materials and release of greenhouse gases, particularly carbon dioxide (CO2). In Asia, the open burning of agricultural waste, such as crop residues, accounts for a significant portion of overall biomass burning, further exacerbating the problem. However, researchers are now exploring ways to transform these agricultural wastes into eco-friendly building materials, aligning with the United Nations Sustainable Development Goals and the objectives of the Paris Agreement.
Transforming Agricultural Waste into Sustainable Bricks
In this study, researchers focused on utilizing groundnut shells as an alternative fine aggregate in the production of a new type of building material called “Bio-Bricks.” By mixing groundnut shells with cement, fly ash, and hydrated lime, they were able to create a sustainable and cost-effective brick that meets the requirements for first-class bricks under Indian construction standards.
Optimizing Brick Properties with Machine Learning
To further enhance the performance of these Bio-Bricks, the researchers employed advanced machine learning techniques, such as Response Surface Methodology (RSM) and Artificial Neural Networks (ANN). These methods allowed them to optimize and predict the key properties of the Bio-Bricks, including compressive strength, dry density, and water absorption, with exceptional accuracy.
The RSM models demonstrated high degrees of accuracy, with R-squared values exceeding 0.88 for the critical properties. The ANN models further improved the predictive power, with R-squared values exceeding 0.99, showcasing the effectiveness of these machine learning approaches in material engineering and optimization.
Balancing Performance and Cost-Effectiveness
In addition to the technical advancements, the researchers also conducted a thorough cost analysis of the Bio-Bricks. They found that the production of these bricks resulted in a lower cost compared to conventional clay bricks, making them a more economically viable option for the construction industry.
The study highlights the potential of agricultural waste, such as groundnut shells, to be transformed into sustainable building materials that not only reduce environmental impact but also offer cost-effective alternatives to traditional construction methods. By leveraging advanced machine learning techniques, researchers were able to optimize the properties of these Bio-Bricks, paving the way for a greener and more efficient construction industry.
Author credit: This article is based on research by G. Nakkeeran, L. Krishnaraj, Pshtiwan Shakor, George Uwadiegwu Alaneme, Obeten Nicholas Otu.
For More Related Articles Click Here
