Cold rolling of sheet metal is one of the most important processes in sheet metal processing, which is used to reduce the thickness, equalize the thickness and ensure the appropriate mechanical properties of the material. In order to achieve consistently high product quality and maximum productivity of the equipment, it is important to properly adjust and adapt the process parameters to the current process tasks.
The focus of this thesis is on the development of a recommender system for operator decision support in industrial processes. The main goal of the thesis is to verify if the practice of recommender systems, commonly used in the field of marketing, can be transferred to the field of industrial decision support systems. The goal of such a system is to extract the most effective operator practices, transfer them to less experienced operators, and use them to automatically adapt process settings to current process tasks.
A special tool has been developed to demonstrate the use of the developed decision support system in an industrial process. This tool is intended for operators to adjust recipe parameters according to the characteristics of the individual workpiece, and for technologists to determine more appropriate general recipes for a wider range of products to be manufactured. The developed system can be used to identify alternatives and improvements compared to current practice, taking into account previous knowledge and experience.
The developed operator tool and recommender system approaches are not limited to the specific characteristics of the considered rolling mill, which allows transferability and application of these approaches also in related production processes with appropriate adaptation.
|
