Defective products management in a furniture production company: A data mining approach

Basit Öğe Kaydı
dc.authoridErsoz, Taner/0000-0001-5523-5120
dc.authoridGuven, Ilker/0000-0002-2754-6893
dc.authoridErsoz, Filiz/0000-0002-4964-8487
dc.contributor.authorErsoz, Taner
dc.contributor.authorGuven, Ilker
dc.contributor.authorErsoz, Filiz
dc.date.accessioned2024-09-29T15:50:38Z
dc.date.available2024-09-29T15:50:38Z
dc.date.issued2022
dc.departmentKarabük Üniversitesien_US
dc.description.abstractQuality is one of the main focuses of the manufacturing companies. Therefore, this issue takes attention of many researchers from both academic and professional environment. In industries like furniture where company types are most likely workshop or small-medium enterprise and production method is traditional, effective methods to prevent faulty production must be considered. Traditional or statistical methods are good to track defective products and keep the within desired levels, however not as good to prevent them from occurring. These methods come with an acceptance to some level of defective production. In this study, it is aimed to evaluate the in the furniture production sector and to reveal the source of the defective production and the factors that cause the defect in terms of which department. In addition, finding the most appropriate methods to accurately analyze the data coming from the company is another research topic of this study. In the research, artificial neural networks and decision tree models were established and inferences were made from the data sets. The model established in the application revealed the causes of the problems experienced in the production process of the company, thus root cause of defective products can be found and prevented. By using data mining techniques, this study developed an effective approach to improve the quality of the production process and to predict and prevent errors before they occur. According to results of the study classification and regression tree algorithm outperformed other methods by yielding 90.12% correct prediction rate. 87.5% of the defects caused by cover and seat cushion problems account for defects in the textile department.en_US
dc.identifier.doi10.1002/asmb.2685
dc.identifier.endpage914en_US
dc.identifier.issn1524-1904
dc.identifier.issn1526-4025
dc.identifier.issue5en_US
dc.identifier.scopus2-s2.0-85131318674en_US
dc.identifier.scopusqualityQ2en_US
dc.identifier.startpage901en_US
dc.identifier.urihttps://doi.org/10.1002/asmb.2685
dc.identifier.urihttps://hdl.handle.net/20.500.14619/3644
dc.identifier.volume38en_US
dc.identifier.wosWOS:000804087200001en_US
dc.identifier.wosqualityQ3en_US
dc.indekslendigikaynakWeb of Scienceen_US
dc.indekslendigikaynakScopusen_US
dc.language.isoenen_US
dc.publisherWileyen_US
dc.relation.ispartofApplied Stochastic Models in Business and Industryen_US
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanıen_US
dc.rightsinfo:eu-repo/semantics/closedAccessen_US
dc.subjectartificial neural networksen_US
dc.subjectdata miningen_US
dc.subjectdecision treesen_US
dc.subjectdefect detectionen_US
dc.subjectquality improvementen_US
dc.titleDefective products management in a furniture production company: A data mining approachen_US
dc.typeArticleen_US

Dosyalar

Koleksiyon

WoS İndeksli Yayınlar Koleksiyonu
Scopus İndeksli Yayınlar Koleksiyonu