In forming processes, ductile damage develops as a result of nucleation, growth and coalescence of voids. Ductile damage affects the product properties, and therefore, needs to be considered in process design. The quality of simulation results highly depends on the identification of material parameters. Thus, they need to be determined accurately. Different strategies for the identification of a Gurson-Tvergaard-Needleman (GTN) type ductile damage model are presented for the dual phase steel DP800. Inverse identification processes are performed based on the experimental results of experiments with inhomogeneous deformation such as notched tensile tests. The parameters are identified in terms of load-displacement data. In addition, the nucleation-related parameters of the GTN model are identified based on microscopical investigations of the void volume fraction. To this end, SEM images are analyzed in order to measure the evolution of voids for different deformation states. The predictive capability of the GTN model with respect to the local void volume fraction is analyzed.