Current research projects demonstrated the efficiency of neural networks as a method for production control. The ability to learn continuously from experience gained through operation, the resulting flexibility in control and a small effort in calculation and modeling are the main vantages compared to classic methods for production control. This research project aims to the simulation of different net types over a long time period. Based on the gained data about the efficiency in learning, the life cycle and the amount of maintenance a hybrid solution with optimal quality for production control is developed.
Contact persons:
S. Oelker
Funded by:
DFG
Duration:
01.01.2009 - 30.12.2011
See project's publications
List all projects
