Optimization of Roughness, Dimensional Conformity, and Porosity of 3D-Printed ASA with MEX: Impact of Critical...

Acrylonitrile-styrene-acrylate (ASA) is a material with high potential, making it suitable for outdoor applications, which warrants further investigation. Its presence in additive manufacturing (AM) shows potential. Herein, ASA three-dimensional (3D) printed samples are examined using a Taguchi L25 experimental design to achieve optimum quality characteristics, through the improvement of the performance across multiple response metrics, including root-mean-square roughness (R _q), average roughness (R _a), actual-to-nominal dimensional deviation (A2N₉₅) and CT scan porosity (P _CT). Six variable control parameters were examined: extrusion width, raster orientation, layer height, deposition velocity, extruder temperature, and substrate temperature. Experimental findings showed that R _a and R _q can be improved by more than 250% (R _a reduced from 17.37 to 6.79 μm, R _q reduced from 21.71 to 9.29 μm), geometrical accuracy can be enhanced by 324% (A2N₉₅ reduced from 437.76 134.95 μm), and porosity can be reduced by 564% (PCT reduced from 9.08 to 1.61%) when selecting a proper set of 3D printing settings. Different regression models were evaluated: the reduced quadratic (RQRM), linear (LRM), and quadratic (QRM) regression models. LRM was inferior, while RQRM and QRM had remarkably close prediction accuracies; thus, the RQRM was proposed for this experimental scenario. The two confirmation runs yielded prediction equations with an error of less than 10% between the predicted and calculated values. The extruder temperature and extrusion width were the two parameters causing the greatest impact on the responses.