The continuous improvement of all steel production processes regarding the avoidance of quality deficiencies and the related improvement of production yield is an essential task of each steel producer. Therefore and within the frame work of the Zero-Defect strategy popular today, several quality assurance techniques are used. The present report explains in this context the method of Data Mining and describes its application in the industrial environment and here especially in the steel industry. Beside a description of the general procedure different approaches for the realization of the necessary software tools are presented. The parts of knowledge discovery in different databases are displayed and construction of necessary data table is shown. Some pre-processing steps including training (70 %) and validation (30 %) are analyzed. Analytical interface for the results of the first data exploration is observed, as well as User interface and the results of DataDiagnose. Principal technical solutions for development of proper software are discussed. Operating technique for using wizard-guided data mining tool is presented. User interface of the Nico Miner software tool is considered. Combination of data mining and knowledge usage is illustrated.

1. Peters, H.; Link, N.; Heckenthaler, T.: stahl u. eisen 120 (2000) Nr. 8, S. 71/77.

2. Peters, H.; Heckenthaler, T.; Holzknecht, N.: stahl u. eisen 125 (2005) Nr. 7, S. 29/36.

3. Nakhaeizadeh, G.; Reinartz, T.; Wirth, R.: Wissensentdeckung in Datenbanken und Data Mining: Ein Überblick, [in:] Nakhaeizadeh, G. [Hrsg.]: Data Mining, Physica Verlag, Heidelberg, 1998, ISBN 3-7908-1053-3.

4. Berthold, M.; Hand, D. J.: Intelligent Data Analysis, Springer Verlag, Berlin, 1999.

5. Cherkassky, V.; Mulier, F.: Learning from Data, John Wiley & Sons, Inc., New York, 1998.

6. Peters, H.; Link, N.: Cause and effect analysis of quality deficiencies at steel production using automatic data mining technologies, Proc. 13. IFAC Symp. on Automation in Mining, Mineral and Metal Processing, 2. – 4. Aug. 2010, Kapstadt, Südafrika, S. 93/98.

7. Beindorf, J.; Anstots, T.; Eberle, A.; Ernenputsch, L.; Holzhauser, J.-F.: stahl u. eisen 124 (2004) Nr. 10, S. 53/60.

8. Peters, H.; Heckenthaler, T.; De Abajo, N.; Murri, M.; Hilliges, F.; Bösler, R. P.; Le Goc, M.: Implementation of an assessment and analysing system for the utilisation of a factory-wide product quality database, EUR 20927 EN (2004), Luxemburg, ISBN 92-894-6826-2.

9. EGKS-Projekt 7215-PA/PB/PC/069: Application of data-based Technologies to demonstrate online Quality Control of Mini Mills, 7/2001 – 3/2005.

10. RFCS-Projekt RFSR-CT-2008-00042: Supporting process and quality engineers by automatic diagnosis of cause-and-effect relationships between process variables and quality deficiencies using Data Mining technologies, 7/2008 – 6/2011.

11. Mierswa, I.; Wurst, M.; Klinkenberg, R.; Scholz, M.; Euler, T.: YALE: Rapid Prototyping for Complex Data Mining Tasks, Proc. 12. ACM SIGKDD Intern. Conf. on Knowledge Discovery and Data Mining (KDD 2006), 22. – 23. Aug. 2006, Philadelphia, USA; ACM Press, New York, USA, 2006.

12. ZIM-Projekt KF2231810LF9: Qualitätsfehlerursachenanalyse bei der Stahlproduktion durch Kombination von Materialverfolgung, Datenarchivierung und Data Mining, 3/2010 – 10/2011.

13. Hackmann, J.; Peters, H.: Defect analysis based on the combination of data mining and material genealogy, InSteelCon/SteelSim2011, 27. Juni – 1. Juli 2011, Düsseldorf.