Innovative assessment of cement chemistry and the impact of additives on compressive strength across various specimen sizes and extended curing conditions

Understanding key material properties such as compressive strength (CS) is crucial for improving cement production. Its chemical composition significantly influences the CS of cement. This study investigates the impact of the primary cement constituents such as SiO₂, Al₂O₃, CaO, and Fe₂O₃ along with varying dosages of fly ash (FA) and silica fume (SF) on the compressive strength of cement slurries. Four predictive models with varying complexities were employed: linear regression, pure-quadratic, interaction (IA), and full quadratic (FQ). These models, previously validated for concrete and mortar, were applied to a dataset of 317 records from prior research. Model performance was assessed using correlation coefficient (R²), root mean squared error, mean absolute error, and scatter index. Key input parameters included cement chemical composition (SiO₂: 13.57–22.62%; Al₂O₃: 2.05–6.2%; CaO: 60.76–65.72%; Fe₂O₃: 1.9–6.68%), water-to-binder ratio (0.22–1), FA content (0–70%), SF content (0–40%), curing time (0.333–400 days), and curing temperature (3–260 °C). The study revealed that the FQ model provided the most accurate predictions, while the IA model had the lowest residual error. Sensitivity analysis indicated that Fe₂O₃ most significantly affects CS, with the water-to-binder ratio and curing time being crucial for accurate predictions. Optimal replacement percentages for FA and SF were determined to be up to 12% and 22%, respectively.