Enhancing flexibility of production systems by self-optimization
For establishing self-optimization in production technology, an extensive knowledge about the effects of process parameters on the product during the manufacturing processes is mandatory for the flexibility needed in complex production systems. In this regard, a controlling software that can autonomously detect this knowledge and use it for optimization is needed. Specific methods for the implementation of cognitive information processing are required. This paper focuses on the development of a self-optimizing controlling software architecture. In order to evaluate the optimization steps and simultaneously validate the rules used for it, an implementation for real production data of an automotive rear-axle-drive is described.