In the heart of Riyadh, a groundbreaking study led by Yassir M. Abbas from the Department of Civil Engineering at King Saud University is revolutionizing the way we think about concrete. Abbas and his team have tackled a long-standing challenge in the construction industry: balancing the mechanical performance and cost of sustainable concrete. Their innovative approach, published in the journal ‘Buildings’ (which translates to ‘المباني’ in Arabic), is set to reshape the future of construction materials, with significant implications for the energy sector.
The study introduces an interpretable artificial intelligence (AI)-driven approach that combines the Category Boosting (CatBoost) algorithm with the Non-Dominated Sorting Genetic Algorithm II (NSGA-II). This powerful combination models and optimizes the compressive strength and total cost of quaternary-blended concretes, which incorporate multiple supplementary cementitious materials. “This is the first time such a complex system has been systematically optimized,” Abbas explains, highlighting the novelty of their approach.
The team curated a database of 810 experimentally documented mixtures to train and validate their model. The CatBoost algorithm achieved superior predictive performance, with an impressive R² value of 0.987 and a mean absolute error of just 1.574 MPa. Shapley additive explanations revealed that curing age, water-to-binder ratio, and Portland cement content were the dominant parameters governing compressive strength.
The multi-objective optimization produced Pareto-optimal elite mixtures, achieving compressive strengths ranging from 51 to 80 MPa. Notably, a representative 60 MPa mix required approximately 62% less cement than conventional designs. This significant reduction in cement usage not only lowers costs but also reduces the carbon footprint of concrete production, a critical factor for the energy sector.
The implications of this research are far-reaching. “Our framework offers both methodological innovation and practical guidance for implementing sustainable construction materials,” Abbas states. By providing a scientifically grounded, interpretable methodology for data-driven design, this study paves the way for low-carbon, high-performance concretes. The AI-assisted multi-criteria optimization demonstrated in this research could become a standard tool for engineers and architects seeking to balance performance, cost, and sustainability in their projects.
As the construction industry continues to evolve, the need for sustainable and cost-effective materials will only grow. This research by Abbas and his team at King Saud University is a significant step forward, offering a blueprint for the future of concrete design. The energy sector, in particular, stands to benefit from these advancements, as the reduced carbon footprint of these new concrete formulations aligns with global efforts to mitigate climate change.
In the words of Abbas, “This is not just about optimizing concrete; it’s about optimizing our future.” With this groundbreaking research, the future of sustainable construction looks brighter than ever.