Typical tasks for process and technology simulation are to describe an etching process, material deposition, chemo-mechanical polishing, ion implantation, lithography as well as electro-thermal and mechanical investigations of the derived structures. These process steps deal with the patterning and structuring of matter, which requires highly complex models to describe the physical properties as well as the chemical and physical interaction between the materials. However, since the temperature cannot be neglected in many process steps, the model becomes even more complex. Hence, optimization of such problems requires deep knowledge of the important ongoing process inside the current simulation task to estimated the possible bias points of operation. These points can be used as initial values for optimization . If, for instance, an appropriate approximation can be found for a certain bias point, a specially tuned (local) optimization strategy can be applied. For the other cases, heuristic, genetic, or random search methods have to be used to obtain a certain ``feeling'' of the current problem class.