Selective laser melting (SLM) is a method for producing complex part geometries which cannot be made with traditional manufacturing methods. It is also a very flexible method for short-run production and producing single custom parts of high added value. A major current drawback of the method is the difficulty in adjusting process parameters in such a way that the overall quality of the component is acceptable and complex designs can be manufactured with consistency. Quality parameters being typically surface quality, internal porosity, residual stresses and part distortions, to name a few. Quite often the production of a component and rampup of production requires several attempts to get undistorted and sound parts due to the intrinsic complexity of the manufacturing process and a large array of available process parameters.

Design methods accounting for specifics of SLM are not available to the same degree as for conventional manufacturing, and traditional computer-aided engineering tools have demonstrated limited applicability. One way to address these key issues is to develop a physically accurate numerical model of the process where the effect of individual parameters can be easily assessed one by one, enabling trialling and development of process utilizing modelling and simulation.

The most influential parameters are scanning speed, laser power and hatch clearance but also laser spot size, shape, laser intensity pattern and their variation throughout the part have a role in the manufacturing outcome. Parametric sensitivity studies with this type of model are very cheap compared to the more traditional design of experiments trial-and-error exercises. Eventually, this type of component analysis, or even a library comprising a set of pre-calculated cases, could be incorporated into the 3D-printer as a cyber-physical system with in-situ process analytics to allow the adjustment of process parameters on-the-fly as process anomalies are detected, exploiting novel computation and data analytics methodologies.