The advancement of additive manufacturing (AM) has been playing a transformational role in metal manufacturing. For many years, AM has been known to address “economies of scope” in which customization, prototyping and low volume manufacturing have been the main foci. However, in recent years, AM has positioned itself to deploy for “economies of scale”, i.e., mass production, without compromising the economies of scope. This promotion from prototyping to a series and mass production platform has opened up many R&D opportunities. Like all conventional techniques, quality assurance procedures/tools (either online or offline) are of the utmost importance to aid manufacturers in quality management and certification when it comes to serial production.

This talk aims to present the recent advancement in monitoring of laser powder-bed fusion LPBF and investigate the feasibility of applying in-situ monitoring to LPBF for the detection of disturbances in the process. This feasibility study will facilitate identifying correlations between physical defects created in printed parts and signals collected from on-axis and off-axis photodiodes that monitor emitted light from the melt pool. Statistical-based approaches and/or artificial intelligence algorithms can be used to deploy the proposed hardware/software to a reliable quality assurance platform required for series production. This talk will address some of these approaches at high level as well as preliminary results obtained from analysis of data collected by the commercially available EOSTATE MeltPool monitoring system, installed on an EOS M290 machine.