Experimental Design and Predictive Modeling of Hybrid Jute/Kenaf/Glass Fiber Polymer Composites With SiC Nanoparticles for Enhanced Flexural and Hardness Performance

ABSTRACT The paper examines the mechanical properties of hybrid polymer composites reinforced with jute, kenaf and glass fibers with silicon carbide (SiC) nanoparticles as a filler. Three important processing parameters, including fiber orientation (0°, 45°, 90°), fiber stacking arrangement (1–3 layers), and SiC content (3–5 wt%) were evaluated in a systematic manner. The response surface methodology (RSM) was used to design and analyze experimental trials to determine important interactions between the parameters that affected flexural strength and hardness. Meanwhile, predictive modeling was performed with artificial neural networks (ANN), and predictive accuracy was higher than that of RSM with a high correlation between forecasted and experimental outcomes. Optimized flow (90° fiber orientation, three layers stacked, 5 wt% SiC) led to an 18% flexural strength and 32% hardness enhancement over the starting composite. The results illustrate the possibility of using experimental design and machine learning models in order to make robust predictions and optimization of composite properties, and using them to support their application in structural and load‐bearing engineering sustainable components.