In a scenario of scarce resources, with increasingly tight deadlines and costs, it is imperative to
replace and complement traditional material removal processes with more efficient ones such
as additive processes.

However, there are still many factors preventing the adoption of such technology, especially the
appearance of defects in the manufactured parts, as defects are generally unpredictable,
random and difficult to detect.

Machine manufacturers are developing increasingly sophisticated process control methods aimed primarily at detecting defects in real time, so that production can be aborted when
potential errors are detected. Many of these systems are based on scanning the molten surface
with thermography and have been in development for years; there are currently very few
companies offering them as a complement to additive machines from other manufacturers and
their effectiveness in detecting defects is limited to certain sizes and types. This technology has
not been validated by harmonisation bodies and its use is currently marginal.

In this context, the SCALDATA project, developed by the companies CiTD and Elliot Cloud with
the support of the Smart City Cluster and approved within the line of Digital Technologies
Projects of the Call for Aid to Innovative Business Groups 2022 (AEI) promoted by the Ministry of
Industry, Tourism and Trade of the Government of Spain; Its main objective is to exploit data
related to additive manufacturing processes of powder metal bedding and to use artificial
intelligence in order to generate predictions for the typology, location and effect of defects in
the parts produced. The knowledge of this defectology, both in location and size, is of vital
importance for the validation of the part, especially for applications with fatigue requirements.