CLASSIFICATION OF DEFECT RATES IN PRODUCTION PROCESSES: AN APPROACH BASED ON DEEP LEARNING WITH INNOVATIVE QR CODE CONVERSION

Yunus Emre Gür; Mesut Toğaçar; Bilal Solak; Cem Ayden

Makale detayları

CLASSIFICATION OF DEFECT RATES IN PRODUCTION PROCESSES: AN APPROACH BASED ON DEEP LEARNING WITH INNOVATIVE QR CODE CONVERSION

Yazarlar:

Yunus Emre Gür | Mesut Toğaçar | Bilal Solak | Cem Ayden

Yayıncı:

Anadolu University

Ana Başlıklar:

Production and Operations Management

Dergi:

Anadolu Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi Vol.26 Issue 1, pp.245-276

Belge Tipi:

Makale

Yayın Tarihi:

3/2025

Belge Dili:

Türkçe

Document ID (Org No.):

2001344132

pISSN:

2687-184X

Anahtar kelimeler:

Özet:

 This study presents an innovative method for accurately classifying defect rates in manufacturing processes and optimizing quality control processes. In the study, numerical data are transformed into two-dimensional QR code images and analyzed with the AlexNet model. This method aims to classify defect rates with high accuracy by utilizing the powerful pattern recognition capabilities of deep learning models. The dataset was labeled as low and high defect rates and divided into 80 training and 20 testing. It is compared with various machine learning models such as Decision Tree, Gradient Boosting, K-Nearest Neighbor, Logistic Regression, Naive Bayes, Random Forest and Support Vector Machine. The results show that the AlexNet model classifies defect rates with 100 accuracy. These findings emphasize that deep learning algorithms can be highly effective for quality control and defect detection in manufacturing processes. Furthermore, limitations of the study and suggestions for future research are presented. This innovative methodology has the potential to be widely used in other industrial processes and different data sets, contributing to the improvement of production efficiency.

 This study presents an innovative method for accurately classifying defect rates in manufacturing processes and optimizing quality control processes. In the study, numerical data are transformed into two-dimensional QR code images and analyzed with the AlexNet model. This method aims to classify defect rates with high accuracy by utilizing the powerful pattern recognition capabilities of deep learning models. The dataset was labeled as low and high defect rates and divided into 80 training and 20 testing. It is compared with various machine learning models such as Decision Tree, Gradient Boosting, K-Nearest Neighbor, Logistic Regression, Naive Bayes, Random Forest and Support Vector Machine. The results show that the AlexNet model classifies defect rates with 100 accuracy. These findings emphasize that deep learning algorithms can be highly effective for quality control and defect detection in manufacturing processes. Furthermore, limitations of the study and suggestions for future research are presented. This innovative methodology has the potential to be widely used in other industrial processes and different data sets, contributing to the improvement of production efficiency.

 This study presents an innovative method for accurately classifying defect rates in manufacturing processes and optimizing quality control processes. In the study, numerical data are transformed into two-dimensional QR code images and analyzed with the AlexNet model. This method aims to classify defect rates with high accuracy by utilizing the powerful pattern recognition capabilities of deep learning models. The dataset was labeled as low and high defect rates and divided into 80 training and 20 testing. It is compared with various machine learning models such as Decision Tree, Gradient Boosting, K-Nearest Neighbor, Logistic Regression, Naive Bayes, Random Forest and Support Vector Machine. The results show that the AlexNet model classifies defect rates with 100 accuracy. These findings emphasize that deep learning algorithms can be highly effective for quality control and defect detection in manufacturing processes. Furthermore, limitations of the study and suggestions for future research are presented. This innovative methodology has the potential to be widely used in other industrial processes and different data sets, contributing to the improvement of production efficiency.

 Bu çalışma, üretim süreçlerindeki kusur oranlarının doğru sınıflandırılması ve kalite kontrol süreçlerinin optimize edilmesi için yenilikçi bir yöntem sunmaktadır. Çalışmada, sayısal veriler iki boyutlu QR kod görüntülerine dönüştürülerek AlexNet modeli ile analiz edilmiştir. Bu yöntem, derin öğrenme modellerinin güçlü desen tanıma yeteneklerinden yararlanarak kusur oranlarını yüksek doğrulukla sınıflandırmayı amaçlamaktadır. Veri seti, düşük ve yüksek kusur oranları olarak etiketlenmiş ve 80 eğitim, 20 test olarak bölünmüştür. Karar Ağacı, Gradient Boosting, K-En Yakın Komşu, Lojistik Regresyon, Saf Bayes, Rastgele Orman ve Destek Vektör Makinesi gibi çeşitli makine öğrenmesi modelleri ile karşılaştırılmıştır. Sonuçlar, AlexNet modelinin kusur oranlarını 100 doğrulukla sınıflandırdığını göstermektedir. Bu bulgular, derin öğrenme algoritmalarının üretim süreçlerindeki kalite kontrol ve kusur tespiti için son derece etkili olabileceğini vurgulamaktadır. Ayrıca, çalışmanın kısıtlılıkları ve gelecekteki araştırmalar için öneriler sunulmuştur. Bu yenilikçi metodoloji, diğer endüstriyel süreçlerde ve farklı veri setlerinde de geniş bir kullanım potansiyeline sahip olup, üretim verimliliğinin artırılmasına katkı sağlayacaktır.

Makale detayları

CLASSIFICATION OF DEFECT RATES IN PRODUCTION PROCESSES: AN APPROACH BASED ON DEEP LEARNING WITH INNOVATIVE QR CODE CONVERSION

Yazarlar:

Yunus Emre Gür | Mesut Toğaçar | Bilal Solak | Cem Ayden

Yayıncı:

Anadolu University

Ana Başlıklar:

Production and Operations Management

Dergi:

Anadolu Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi Vol.26 Issue 1, pp.245-276

Belge Tipi:

Makale

Yayın Tarihi:

3/2025

Belge Dili:

Türkçe

Document ID (Org No.):

2001344132

pISSN:

2687-184X

Defect Rate Classification

Machine Learning

Deep Learning

QR Code Conversion

AlexNetKusur Oranı Sınıflandırma

Makine Öğrenmesi

Derin Öğrenme

Karekod Dönüşümü

AlexNet

Sayı

Makale detayları

Yazarlar: Yunus Emre Gür | Mesut Toğaçar | Bilal Solak | Cem Ayden

Yayıncı: Anadolu University

Ana Başlıklar: Production and Operations Management

Dergi: Anadolu Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi Vol.26 Issue 1, pp.245-276

Belge Tipi: Makale

Yayın Tarihi: 3/2025

Belge Dili: Türkçe

Document ID (Org No.): 2001344132

pISSN: 2687-184X

Sayı

Yazarlar:

Yunus Emre Gür | Mesut Toğaçar | Bilal Solak | Cem Ayden

Yayıncı:

Anadolu University

Ana Başlıklar:

Production and Operations Management

Dergi:

Anadolu Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi Vol.26 Issue 1, pp.245-276

Belge Tipi:

Makale

Yayın Tarihi:

3/2025

Belge Dili:

Türkçe

Document ID (Org No.):

2001344132

pISSN:

2687-184X