Sustainable Sisal Fiber–Red Mud Hybrid Composites: Mechanical Evaluation and ANN–RSM Optimization

This study addresses the limited availability of predictive and optimization frameworks for sustainable hybrid composites incorporating natural fibers and industrial waste fillers. Eco-friendly composites reinforced with sisal fiber and red mud were fabricated using the hand lay-up method with varying compositions of sisal (3.2–8.2 wt%), red mud (5–15 wt%), and particle size (58–86 μm). Mechanical properties, including flexural strength and interlaminar shear strength (ILSS), were experimentally evaluated. Response surface methodology (RSM) was employed to analyze parametric interactions and optimize composite design, while artificial neural networks (ANN) were used for predictive modeling. The ANN model demonstrated strong prediction capability with correlation coefficients (R 2 ) of 0.9664 for flexural strength and 0.9697 for ILSS, with prediction errors below 3%, indicating high modeling accuracy. Results revealed that the weight fractions of sisal fiber and red mud significantly influence mechanical performance, whereas particle size has a comparatively lower effect. The optimal composite configuration achieved flexural strength of 64.15 MPa and ILSS of 51.97 MPa. The novelty of this work lies in the integration of sustainable material design with a dual RSM–ANN optimization framework for reliable prediction and performance enhancement of hybrid composites for engineering applications. • Red mud waste is sustainably valorized as a functional filler in sisal fiber–reinforced epoxy composites. • ANN and RSM models accurately predict and optimize flexural and ILSS properties (R 2 ≈ 0.97). • Red mud and sisal content significantly enhance mechanical performance, while particle size shows minimal effect. • The study provides a practical waste-to-resource pathway for developing eco-friendly structural composites.