In collaboration with the Institute for Machine Tools and Industrial Management (iwb) at the Technical University of Munich, PTW is working on the systematic detection, classification and quantitative evaluation of defects in PBF-LB/M in order to ensure the quality of safety-critical components more effectively. The background to this is that the complexity of the process leads to unavoidable errors, the type, distribution and severity of which cannot be adequately or only with great effort recorded using existing testing methods such as in-process samples or CT scans. To this end, the manufacturing process is monitored using several optical and acoustic sensor systems, whose data is merged at the system and evaluation level (sensor data fusion). With the help of domain knowledge and characteristic data patterns, the defects are to be described in terms of their probability and severity. Particular attention is paid to the self-healing effect in the process, in which certain errors can be (partially) compensated for in the further course of the process by remelting.