In the heart of Riyadh, researchers are revolutionizing the way we think about construction materials, and their work could have profound implications for the energy sector. Aïssa Rezzoug, a leading figure at the College of Engineering, Imam Mohammad Ibn Saud Islamic University, has spearheaded a groundbreaking study that marries machine learning with sustainable construction practices. The result? A user-friendly tool that could transform how engineers design and implement plastic-derived mortar mixes, potentially slashing costs and reducing environmental impact.
At the core of Rezzoug’s research is a quest to enhance the compressive strength of plastic-based mortar mixes (PMMs) by substituting cement with industrial waste materials. This isn’t just about recycling; it’s about creating a more sustainable future for construction. “The goal is to make sustainable construction more accessible and practical,” Rezzoug explains. “By using machine learning, we can predict the performance of these mixes with unprecedented accuracy, making it easier for engineers to adopt these eco-friendly materials.”
The study, published in Buildings, employs sophisticated machine learning models such as support vector machines, AdaBoost regressors, and extreme gradient boosting. These models analyze key mix parameters to predict the compressive strength of PMMs. But here’s where it gets really interesting: Rezzoug and his team didn’t stop at the algorithms. They developed a graphical user interface (GUI) to make these complex models accessible to engineers and practitioners who may not have the technical expertise to use them directly.
Imagine an engineer on a construction site, needing to quickly determine the best mix of materials for a particular job. With Rezzoug’s GUI, they can input the proportions of various materials and instantly get a prediction of the compressive strength. This isn’t just about convenience; it’s about making data-driven decisions that can save time, reduce waste, and lower costs.
The implications for the energy sector are significant. Construction materials account for a substantial portion of energy consumption and carbon emissions. By making sustainable materials more predictable and easier to use, Rezzoug’s work could help reduce the energy sector’s environmental footprint. “This technology has the potential to change the way we build,” Rezzoug says. “It’s not just about making construction more sustainable; it’s about making it smarter.”
The research also delves into the significance of key mix parameters using SHapley Additive exPlanations (SHAPs). This analysis helps interpret the influence of input variables on model predictions, providing deeper insights into the mix design process. The GUI serves as a practical and scalable decision support system, encouraging the adoption of machine learning-based approaches in sustainable construction engineering.
As we look to the future, Rezzoug’s work offers a glimpse into what’s possible. It’s a testament to how machine learning can be harnessed to solve real-world problems, making sustainable practices more accessible and practical. For the energy sector, this means a future where construction is not just about building structures, but about building a more sustainable world.
The study, published in the journal Buildings, which translates to ‘Buildings’ in English, marks a significant step forward in the integration of machine learning and sustainable construction. As more engineers and practitioners adopt these tools, we can expect to see a shift towards more sustainable and efficient construction practices. The future of construction is here, and it’s smarter, greener, and more accessible than ever before.