A preference based multi response decision tree approach for quality improvement in manufacturing

DOLGUN L. E., Ipekçi A. I., KÖKSAL G.

IADIS European Conference on Data Mining 2011, Part of the IADIS Multi Conference on Computer Science and Information Systems 2011, MCCSIS 2011, Rome, Italy, 24 - 26 July 2011, pp.211-213 identifier

  • Publication Type: Conference Paper / Full Text
  • City: Rome
  • Country: Italy
  • Page Numbers: pp.211-213
  • Keywords: CART algorithm, Decision tree, Multiple responses, Quality improvement
  • Anadolu University Affiliated: Yes


This paper presents a multi response decision tree (DT) approach to find the process conditions that yield products with desirable quality characteristics (QCs). We propose several criteria aggregation functions to convert responses (QCs) into an aggregated value. DT modeling is applied on the resulting aggregated values to decide on the process levels. Decision rules obtained separately for each QC can help in predicting the levels of QC values at these process levels thereafter. © 2011 IADIS.