SMS Siemag AG (Düsseldorf, Germany):

Odenthal H.-J., Mag. Eng., e-mail: juergen.odenthal@sms-siemag.com

Uebber N., Mag. Phys.

Schlüter J., Mag. Eng.

Löpke M., Mag. Math.

Technische Universität Dortmund (Dortmund, Germany):

Morik K., Prof., Dr., Chair of Artificial Intellect

Blom H., Mag. Inform., Chair of Artificial Intellect

Intellectual data analysis allows to optimize the process of steel making due to predicting of its aimed parameters. In the context of “Industry 4.0”, or inte-gration industry, predicting models with data management help steel makers in their work, decreasing operation expenses. The methods of automatic education have been applied for the first time in metallurgical production in the framework of durable collaboration between SMS Siemag (from one side) and Dortmund Tech-nical University and AG der Dillinger Hüttenwerke (from other side). The newly developed predicting model can provide self-education on the base of large data massifs and on-line predicting. As an example, the system manages blowing process with on-line correction recommendations. Software and algorithms use robust features and multi-month stable operation in the conditions of information medium of the steelmaking shop at AG der Dillinger Hüttenwerke is observed. It is shown that usage of both predicting models and former physical-chemical models can be successfully combined practically, in order to increase productivity potential and uni-versal character of technological data analysis.