Table 7.

The use of an ANN for process optimization of 3D processes.

Process	Purpose	ANN Method	Input Parameters	References
Binder Jetting	Predicting surface roughness, shrinkage rate in y- and z-directions	3-layer BP-ANN	Layer thickness, printing saturation, heater power ratio, drying time	[53]
	Compressive strength, open porosity	Aggregated ANN	Orientation, layer thickness, delay time	[106]
	To characterize defects evolution	Gaussian mixture model	To reduce pore decomposition, shrinkage and smoothing during post-processing	[107]
Selective laser sintering	Dimension	Radial basic function ANN, fuzzy C-means, and pseudoinverse method, k-means	Laser power, scan speed, scan spacing, layer thickness	[48]
	Material analysis	ANN	Structural characterization	[108]
	Shrinkage ratio	ANN	Laser power, scan speed, hatch spacing, layer thickness, scan mode, temperature, interval time	[55]
	Tensile strength	ANN	Laser power, scan speed, hatch spacing, layer thickness, powder temperature	[109]
	Density	ANN	Laser power, scan speed, hatch spacing, layer thickness, scan mode, temperature, interval time	[110]
Selective laser melting	Keyhole porosity	K-means clustering	Energy density	[111]
Stereolithography	Dimensional accuracy	ANN	Layer thickness, border overcure, hatch overcure, fill cure depth, fill spacing, and hatch spacing	[112]
	Printability	Ensemble method, Siamese network	Printability	[113]
Laser-melting deposition	Geometrical accuracy	ANN	Laser power, scanning speed, powder feeding rate	[114]
	Melt-pool width	ANN	Laser power, powder feed rate, laser speed, focal length, contact tip to workpiece distance	[115]
Electron beam melting	Volume, roughness	ANN	Spreader translation speed, rotation speed	[116]