In a recent study published in Scientific Reports, researchers explored the effectiveness of various metaheuristic techniques combined with response surface methodology (RSM) in predicting the mechanical properties of self-healing concrete (SHC). The study focused on optimizing the concentration of Bacillus subtilis in SHC to enhance its healing efficiency, structural integrity, and long-term durability, which are critical for sustainable construction.
The research employed advanced metaheuristic techniques, including gray wolf optimization (GWO), multi-verse optimization (MVO), moth fly optimization (MFO), particle swarm optimization (PSO), and whale optimization algorithm (WOA), alongside RSM. These methods were compared to linear multivariate regression (LMR) to forecast the effects of bacteria concentration on SHC’s mechanical properties, such as compressive strength (CS), flexural strength (FS), slump (Sl), and splitting tensile strength.
The results revealed that metaheuristic techniques outperformed RSM, with each method excelling in different areas: GWO was most effective for predicting Sl, PSO for FS, and MVO for CS. The study also highlighted the significant influence of bacterial concentration on SHC properties, underscoring the importance of optimizing this parameter for effective concrete design.
Overall, the research demonstrated the superior predictive capabilities of metaheuristic techniques over traditional methods like RSM and LMR. This advancement could lead to more accurate and efficient designs of self-healing concrete, contributing to the development of sustainable and durable infrastructure. Future studies may focus on hybrid approaches that integrate metaheuristic algorithms with conventional methodologies to further optimize SHC properties and enhance the field of construction materials engineering.
