Due to the strong demand for reduced cost and time in manufacturing of topology-optimized and light-weight parts, additive manufacturing of metallic materials, e.g., the laser powder bed fusion (PBF-LB/M), gained in importance in industry. However, a major challenge in PBF-LB/M continues to be the achievement of a high part quality. To achieve a high part quality, the identification of process parameters leading to high density is necessary, as first step. Even if the search is basically limited to the three important parameters laser power, scan speed, and hatching distance, wide ranges of these parameters must be investigated to identify suitable process parameters. Exploring these ranges solely by the use of experiments can be difficult and time-consuming, especially if the processing windows are expected to be very narrow, e.g., due to the use of a high layer thickness. Therefore, approaches combining experiments and data-driven modeling are a promising way to identify processing windows in PBF-LB/M. One of those approaches using a space filling experiment design and a multiple regression model is presented within the present study. The approach is used to identify processing windows for the manufacturing of the aluminum alloy AlSi10Mg and the steel 42CrMo4 using a relatively high layer thickness of 60 µm.