Applicability of decision trees in manufacturing industry

Marcin Perzyk , Robert Biernacki, Artur Soroczyński

Warsaw University of Technology, Institute of Materials Processing, Narbutta 85, 02-524 Warszawa.

DOI:

https://doi.org/10.7494/cmms.2008.2.0189

Abstract:

In manufacturing companies large amounts of data are collected and stored, related to designs, products, equipment, materials, manufacturing processes etc. Utilization of that data for improvement of product quality and lowering manufacturing costs requires extraction of knowledge from the data, in the form of appropriate conclusions, rules, relationships and procedures. Data mining provides tools and methodologies for semi-automated extraction of that type of knowledge. It is a multidisciplinary field, rapidly growing in recent years, and used mainly in business, medicine, social sciences. Applications to manufacturing and design on a large scale are relatively seldom. In the present work some important manufacturing-related problems are characterized, from the standpoint of benefits from application of data mining methods. In the second part of the paper some selected results of the authors’ studies and research are presented, showing performance of decision trees in solving important typical problems in manufacturing industry.

Cite as:

Perzyk, M., Biernacki, R., Soroczyński, A., (2008). Applicability of decision trees in manufacturing industry. Computer Methods in Materials Science, 8(2), 70 – 78. https://doi.org/10.7494/cmms.2008.2.0189

Article (PDF):

Keywords:

Data mining, Manufacturing processes, Parameter significance, Statistical methods, Artificial neural networks

References: