The powder bed fusion (PBF) process is increasingly employed by industry to fabricate complex parts with stringent standardcriteria. However, fabricating parts “free of defects” using this process is still a major challenge. As reported in the literature,thermally induced abnormalities form the majority of generated defects, and are mainly the result of thermal evolution.Monitoring & controlling the temperature and its distribution throughout a layer under fabrication is an effective and efficientproxy to controlling such an evolution. In this paper, we introduce a novel online thermography and closed-loop hybrid-control(NOTCH)©, a practical control approach, to modify the scan strategy in metal PBF real-time. This system employs differentmathematical thermophysical -based models, designed to optimize the printing sequence of different zones throughout aprinting layer as well as the islands or stripes within each zone. Moreover, NOTCH uses artificial neural network (ANN) tooptimize the energy density applied on each zone in order to avoid or mitigate some prevalent thermal anomalies. NOTCHstrategy has two aims. First, producing a uniform temperature distribution throughout an entire layer to mitigate the thermallyinduced residual stress and its related distortion. In this step, we optimize the printing sequence of islands or stripes in their respective scan strategies. This paper expands on three potential models, explains pros and cons of these models, and presents preliminary results for a printed prototype. Second, controlling the laser specifications in order to avoid heat affected zones (HAZ) and mitigate thermal abnormalities such as the balling phenomenon. This step enables a smart adjustment of energy density by using ANN to avoid or mitigate HAZs, generate uniform microstructure with minimum porosity, and contribute to a more uniform temperature distribution. The completion of the latter step is part of the ongoing research which should be reported in future publications.