Optimization of Mechanical Properties of Eco-Friendly Concrete Containing Wood Ash and Nano silica Using Response Surface Methodology and Artificial Neural Network

This study investigates the mechanical performance of concrete incorporating wood ash (WA) and nanosilica (NS) as partial cement replacements, with a focus on optimizing compressive and flexural strength. A total of 21 mortar mixes were designed and tested at various curing ages. Two modelling approaches—Response Surface Methodology (RSM) and Artificial Neural Networks (ANN)—were employed to analyse the effects and interactions of WA and NS on strength development. RSM provided insight into factor significance, while ANN captured complex nonlinear relationships with higher predictive accuracy (R² = 1.000 for 28-day strength). The results showed that NS significantly improved strength up to an optimal dosage (~2.5 g), beyond which performance declined due to agglomeration or matrix over-refinement. Conversely, high WA content generally reduced strength due to dilution effects. Optimization revealed that blends with low WA (≤30 g) and moderate NS (2.0–2.5 g) achieved the best mechanical performance. ANN-based predictions outperformed RSM and multilinear regression, demonstrating its utility for mix design and performance forecasting in sustainable cementitious systems.