Multi-criteria decision-based optimization and multivariable regression analysis of date palm fiber reinforced concrete modified with silica fume under normal and elevated temperatures

Considering its affordability and accessibility, the deployment of date palm fiber (DPF) in cement composites is progressively escalating. However, the primary downside of DPF in cementitious composite is its deleterious effect on the properties of the composite. To fully utilize DPF in concrete, measures to mitigate its undesirable effects must be used. To address this issue, silica fume was employed as a supplementary cementitious material (SCM) in the DPF-reinforced concrete. In this study, the DPF was appended in several measures between 0 and 3% by weight of cement, and silica fume was employed in dosages between 0 and 15% replacement by volume of cement. Testing was performed on the concrete at normal and high temperature exposure ranging between 200 and 800 °C. Multivariable regression analysis (MRA) was performed to develop and predict the relationship between the variables (silica fume and DPF) and the properties of the concrete. Numerous Multi-criteria decision-making (MCDM) procedures were used to systematically assess and select optimal concrete mixes based on a comprehensive range of performance criteria. The experimental results revealed that DPF instigated a depreciation in the mechanical performance of the concrete, the addition of silica fume mitigated these undesirable effects, with 10% silica fume being the optimum. The study employed mathematical and statistical models to predict the performance of Date Palm Fiber (DPF)-reinforced concrete, addressing the lack of suitable prediction models for DPF. The models were validated through residual compressive strength tests at 200 °C and 800 °C. ANOVA analysis confirmed the models’ reliability and applicability in construction, demonstrating high predictability for both normal and elevated temperature conditions. Further, the MCDM analysis indicated that the mix M7 containing 1% DPF and 10% silica fume ranks highest across all methods, demonstrating superior thermo-mechanical and thermal performance despite a lower slump value due to its composition of 1% DPF and 10% silica fume. © The Author(s) 2025.