Abstract

Based on current success stories using AI methods, this paper examines the relationship between the problem-solving methods AI and system simulation. An analysis of the process steps of the two approaches highlights the fundamental difference between the black-box approach of learning methods and the glass-box approach of structure-explaining simulation models. The mutual benefits of the two approaches can then be explained using four use cases. A further result of the analysis is the question of the extent to which simulation and AI methods can lead to the same or different results. To this end, the concept for a structural analysis is presented, which is based on the idea of analysing the intersection between the results of AI and the simulation method.