The project NN-CBR is a joint project of the University of Bremen and the University of Potsdam. It is funded by the German Research Foundation (DFG). The focus of the University of Bremen lies in the research of the manufacturing control with the aid of artificial neural networks (ANN). The University of Potsdam systematizes experiences with the aid of case-based reasoning (CBR).
As a result of a more and more turbulent market situation, short product life cycles, demanded variants and diversity and increasing complexity of products and manufacturing systems conventional methods of manufacturing control are not appropriate in order to accomplish the objectives like fast and on schedule processing of individual customer orders by lean production.
With the aid of ANN in manufacturing control good results can be achieved regarding the reduction of storage and processing time if appropriate network topologies and parameter are chosen. Nevertheless the data preprocessing for the manufacturing control with the aid of ANN and the configuration of these ANN is very complex and can usually just be solved by experts.
The case-based reasoning is based on human problem solving based on collected experiences. Already known problems and their known solution are used in order to solve new problems. The task of the University of Potsdam in the joint project is to prepare and structure the experts knowledge with the aid of case-based reasoning that conclusions can be drawn on new problems of manufacturing system configuration.

Project duration: 2 years

Project partners:


Members:

Gronau, Prof. Dr.-Ing. Norbert
Universitätsprofessor und Lehrstuhlinhaber

Fröming, Jane (ehemalige Wissenschaftliche Mitarbeiterin)


Peters, Kirstin (ehemalige Wissenschaftliche Mitarbeiterin)

Scharff, Christian
Studentischer Mitarbeiter


Publications:

Peters, K.; Scharff, C. (2009): Intelligente Produktionssteuerung mit verspätetem Feedback. In: PPS-Management 14 (2009) 1, S. 33-36.


last modification: 09/16/2013 09:26:49 PM go above
Deutsch