Multi-Objective Performance Tuning of Concrete Incorporating Glycerin via Central Composite Design and Bio-inspired Metaheuristics

Abstract The thermal and mechanical properties of cementitious composites using glycerin-modified concrete are studied in this research. The effects of the water-binder ratio (0.40, 0.45 and 0.50), glycerin dose (0%, 2.5 %, 5.0 %, 7.5 % and 10%), and Grade of concrete (M20, M30 and M40) on compressive strength, specific heat capacity, thermal conductivity, and thermal diffusivity were statistically examined using Central Composite Design (CCD) under Response Surface Methodology (RSM). With an ideal anticipated strength of 42.6 MPa, the RSM model for compressive strength produced a high coefficient of determination (R² = 0.987). Strong fits were also obtained by the specific heat capacity and thermal diffusivity models, with R² values of 0.973 and 0.968, respectively. Four metaheuristic algorithms—Differential Evolution (DE), Genetic Algorithm (GA), Harris Hawk Optimizer (HHO), and Whale Optimization Algorithm (WOA)—were used for predictive optimization and performance improvement. With a minimal objective function value of 0.00083, DE showed the fastest convergence and the highest optimization efficiency among them. HHO and WOA followed in second and third, respectively, with values of 0.00105 and 0.00118. GA attained moderate performance with a value of 0.00132. The accurate identification of the ideal binder compositions for increased durability and energy efficiency was made possible by the cooperation of optimization approaches and experimental modeling.