Generative and Modular Simulation Models for Supply and Manufacturing Networks

The application of Discrete-Event Simulation (DES) models for purposes of planning and optimization of factories and supply networks is characterized with various abstraction levels and granularities of the model structure. These two aspects are dependent on the complexity of the systems to be simulated, the business goals to be achieved and the project objectives where the simulation models are deployed. This is especially intensified when different product parts and components on different levels withn the supply networks are included into one model, like production lines and work centers within existing and emerging factory shop floors combined with the network of suppliers and additionally flavoured with the ramp up of new products, new work centers or both. Very often the complexity is increased due to the organizational nature of production types and different project groups with own modelling paradigms. This is particularly a characteristic of supply networks that deliver very complex commodity products like whole power plants or respective components. The usual foundation to describe and model such complex systems is the data around the three principal consisting domains (PPR): Products to be delivered (raw-materials, parts, components, finished products), Processes that produce them (from supply chain steps down to operational steps) and Resources necessary to acomplish the work (suppliers, factories, production lines, work centers, machines, etc.). Yet the data is not enough to build the simulation model that, following the paradigm of digital twin, also represents its behaviour as well as the interdependencies between the consisting elements within the PPR-Domains. These interactions, behaviours, and cause-and-effect graphs are usually embodied as a procedural programming, affecting the scope and the depth of the modelled logic and therewith they influence the abstraction levels within the model. The situation is even more complex, in a case when the simulation models represent a workshop-like production and the same simulation model is intended to be deployed for various factories and different products within one big and multifaceted company like Siemens Energy. In opposite of the typical assembly lines like in automotive or electronics industry, here we are talking about product and respective parts and components that are running through different resources in an arbitrary sequence defined by product features and manufactruing technologies available in the considered factories or within the supply network.