Using machine learning to support quality management

Abstract

PurposeTo demonstrate the applicability of machine‐learning tools in quality management.Design/methodology/approachTwo popular machine‐learning approaches, decision tree induction and association rules mining, were applied on a set of 960 production case records. The accuracy of results was investigated using randomized experimentation and comprehensibility of rules was assessed by experts in the field.FindingsBoth machine‐learning approaches exhibited very good accuracy of results (average error was about 9 percent); however, association rules mining outperformed decision tree induction in comprehensibility and correctness of learned rules.Research limitations/implicationsThe proposed methodology is limited with respect to case representation. Production cases are described via attribute‐value sets and the relation between attribute values cannot be determined by the selected machine‐learning methods.Practical implicationsResults demonstrate that machine‐learning techniques may be effectively used to enhance quality management procedures and modeling of cause‐effect relationships, associated with faulty products.Originality/valueThe article proposes a general methodology on how to use machine‐learning techniques to support quality management. The application of the technique in ISDN modem manufacturing demonstrates the effectiveness of the proposed general methodology.

Quality improvement calls data mining: the case of the seven new quality tools

“…Tsironis (2015), Tsironis et al (2005) and Papazoglou (2008) reported several pitfalls of the use of the seven new QTs. Their application is governed by several bureaucratic obstacles.…”

“…This reason has to do with the compatibility of the results of DMT (inference rules and patterns) with the type of information that is needed as an input in order to create them (Köksal et al , 2011; Kahraman and Yanik, 2016). Particularly, the qualitative orientation of the QT achieves a compatible combination with the DMT (Tsironis, 2015; Tsironis et al , 2005; Papazoglou, 2008).…”

“…On the other hand, association rule mining is suggested when some knowledge about the domain is available a priori. We used these two methods because they are a fit for the exploration of the data sets of the cases and they have been successfully used in other similar inquiries, such as the study of Tsironis et al (2005).…”

“…The work recommends the use of machine learning to induce knowledge from data. Tsironis et al (2005), Tsironis (2015) and Papazoglou (2008) demonstrated the applicability of machine learning tools in QM by applying two popular machine learning tools (decision trees (DT) and associations rules) on a set of production case records concerning the manufacture of an ISDN modem. Sung and Sang (2006) used conventional statistical and DM techniques to identify customer voice patterns from data that have been collected through a web-based voice of customer process.…”

Εμπλουτισμός Στατιστικού Ελέγχου Ποιότητας Με Τεχνικές Μηχανικής Μάθησης

“…Πλεονεκτήματα του Στατιστικού Ελέγχου Ποιότητας είναι η στατιστική διερεύνηση της πορείας της διαδικασίας παραγωγής, η αποφυγή απόρριψης και επιδιόρθωσης, και η εξασφάλιση ομοιογενούς ποιότητας.Καταλυτικό παράγοντα τα επόμενα χρόνια στην εξέλιξη του Στατιστικού Ελέγχου Ποιότητας αναμένεται να παίξουν οι τεχνικές Μηχανικής Μάθησης, εμπλουτίζοντας τις ήδη υπάρχουσες μεθόδους και συμβάλλοντας στην ανάπτυξη ακόμα ισχυρότερων εργαλείων για τον έλεγχο ποιότητας (βλ. για παράδειγμα,Tsironis et al, 2005).Η και μάθηση χωρίς επίβλεψη (ή μάθηση από παρατήρηση) (Hastie et al, 2001) (Πίνακας 3.1).…”