3D printing parametric optimization using the power of Taguchi design: an expository paradigm
John D. Kechagias
Abstract
The main scope of this work is to exposit good practices of incorporating the power of experimental design in stochastic problems. The filament material extrusion (f-ME) 3D printing process and the acrylonitrile butadiene styrene (ABS) filament were adopted for this scope owing to many processing parameters affecting 3D printed parts quality performance and specifically on flexural strength. Therefore, infill density, raster orientation, printing speed, layer thickness, and nozzle and bed temperatures are optimized using Taguchi orthogonal array and general linear analysis of variances (ANOVA). Linear ANOVA analysis was explained in detail for the first and second runs separately, and all experiment runs, and the residuals were examined for their normality distribution. The contribution of the factors turned out to be similar in each case, showing that the one-run Taguchi design can be safely used at half the cost and time compared to the two-repetition case.