Experimental and data-driven exploration of surface topography and tribological properties of additively manufactured polymers using fused filament fabrication (FFF)

Additive manufacturing (AM) has revolutionized rapid prototyping and manufacturing. However, limited research has been done on the effect of build orientation and surface roughness on AM parts’ frictional and wear characteristics. This study examines how different print and system parameters influence the surface topography and tribological behavior of 3D printed PLA. The samples were printed in three orientations and tested under varying normal loads (50–100 N). The vertically printed samples resulted in the best wear performance compared to the other two build orientations ($\sim$ 26.75% and 18.47%, respectively, at 100 N normal load). The coefficient of friction also showed dependency on the orientation of the print. The effect of surface topography parameters on tribological properties was also investigated. Skewness, $S_{sk}$, and maximum valley depth, $S_v$, exhibited a strong positive correlation with the coefficient of friction, indicating that tribological behaviors are more sensitive to extreme surface topography features than average surface roughness ($S_a$). A data-driven approach was employed to predict wear rate and coefficient of friction using four machine learning models: Support Vector Regression (SVR), Artificial Neural Network (ANN), Random Forest (RF), and Extreme Gradient Boosting (XGBoost), where decision tree-based models outperformed others. The RF model achieved an $R^2$ value of 0.98 in predicting the wear rate and the coefficient of friction, where surface roughness parameters and operational parameters (normal loads, sliding distance) played critical roles.