Fused Deposition Modeling (FDM) enables the fabrication of multi-material thermoplastic composites with tailored properties through controlled internal geometries. However, optimizing the trade-off between mechanical properties in PETG–TPU (Polyethylene Terephthalate Glycol and Thermoplastic Polyurethane) composites remains a challenge due to complex parameter interactions. This study investigates the tensile and compressive strength of bioinspired PETG–TPU structures using a surrogate-assisted multi-objective optimization framework. Three architected infill geometries (Octet, Gyroid, and Cross 3D), infill density, and layer thickness were examined. Specimens were fabricated using dual-extrusion 3D printing and characterized via ASTM D638 and ASTM D695 standards. An Artificial Neural Network (ANN) was developed to model nonlinear relationships between printing parameters and mechanical responses, achieving a correlation coefficient (R 2 ) above 0.99 with low prediction errors. The ANN was coupled with a Multi-Objective Genetic Algorithm (MOGA) to identify Pareto-optimal solutions for simultaneous strength maximization. Pareto analysis revealed distinct performance trade-offs relative to architectural configurations. The proposed MOGA–ANN framework provides a computationally efficient approach to optimize architected thermoplastic composites, demonstrating the potential of combining bioinspired design with advanced metaheuristics for high-performance additive manufacturing.
Deshwal et al. (Sat,) studied this question.