The planning of production systems involves decisions on layout, capacities, material flows, and process sequences. To evaluate such planning alternatives, simulation models are used to represent system behavior under realistic conditions. Simulation optimization approaches further enable the automated derivation of improved planning decisions, rather than relying solely on scenario-based testing.
Recent research increasingly integrates machine learning techniques into simulation optimization methods, for example by using surrogate models to accelerate the evaluation process. These developments promise powerful and data-driven production planning, but their implementation and application scope are not yet consistently established.
As part of this thesis, a structured literature review on simulation optimization in production planning will be conducted. The objective is to identify relevant scientific contributions, classify them according to suitable criteria, and derive current research focuses as well as open questions. A particular emphasis is placed on approaches in which machine learning is used to support simulation optimization.
If you are interested, please contact Judith Schulze.