Possibilities of Developing of Metallurgical Data Dumps


Data about the technological production characteristics are sent to the archive, where they will be stored for many years. However, the stored data contains many undisclosed links between technological factors and technical and economic production indicators. The article presents a hypothesis about the possibility of processing data generated during production processes of industrial enterprises by analogy developing mining and physical dumps. The article provides an example of studying the sufficiency of the volume of a data metallurgical dump for constructing mathematical models using the experimental planning method. Samples from real production data dumps can compensate for the difficulties of implementing a modern active experiment in training future specialists in secondary vocational and higher education institutions. It is established that the data accumulated over the year in the production archive contain the necessary combinations of realizations of random variables for the two-factor model. The interval method of varying the levels of variables enables to construct an experimental matrix for a three-factor model as well.

Keywords: data dump, production data, metallurgical data, data analysis in metallurgy, matrix, mathematical planning, production management, model, model parameters identification, interval method.

[1] Adler Yu. P. Vvedenie v planirovanie ehksperimenta [Introduction to experiment planning]. Moscow, Metallurgiya, 1968. 155 p.

[2] Leushin I.O. Modelirovanie processov i ob’ektov v metallurgii: Uchebnik [Modeling of processes and objects in metallurgy: Textbook]. Moscow, Forum, 2013. 208 p.

[3] Kotani M., Ikeda S. Materials inspired by mathematics. Science and technology of advanced materials, 2016, vol.17, no 1, pp. 253-259. DOI: 10.1080/14686996.2016.1180233

[4] Michaud D. Metallurgists need statistics. 911Metallurgist, 2015, 8th November. Available at: https://www. 911metallurgist.com/blog/metallurgists-need-statistics (Accessed 8 November 2019).

[5] Yin Sh., Li X., Gao H., Kaynak O. Data-based techniques focused on modern industry: an overview. IEEE Transactions on Industrial Electronics, 2015, vol. 62, I 1, pp. 657-667. DOI: 10.1109/TIE.2014.2308133.