Enhancing tribological performance of Kevlar–PLA nanocomposites through sustainable optimization of fused deposition modeling parameters using RSM and machine learning

Abstract

This study optimizes fused filament fabrication (FFF) parameters to enhance the tribological performance of Kevlar fibre–reinforced polylactic acid (PLA) nanocomposites. Filaments containing 0–10 wt-% Kevlar were produced by twin-screw extrusion and characterized for mechanical and thermal behavior. A 30-run central composite design evaluated the effects of layer height (0.1–0.3 mm), build orientation (0–90°), wall width (0.2–0.6 mm), and Kevlar content on specific wear rate. Statistical analysis identified layer height as the dominant factor (F = 2336.47), contributing 83.3% of the total variation. Predictive modeling using ANN and adaptive neuro-fuzzy inference systems ANFIS demonstrated superior accuracy for ANFIS (R² = 0.9985 training, 0.9912 validation, 0.9647 testing). The optimized parameter set yielded a minimum wear rate of 0.0187 mm³/s with 1.06% prediction error, validated using functional spur gear testing.