DATA MINING AND MACHINE LEARNING APPLICATIONS IN STEEL MATERIALS PURCHASING

Seray Mirasçı; Aslı Aksoy

Article Details

DATA MINING AND MACHINE LEARNING APPLICATIONS IN STEEL MATERIALS PURCHASING

Authors:

Seray Mirasçı | Aslı Aksoy

Publisher:

Süleyman Demirel University

Main Topics:

Endüstri Mühendisliği

Journal:

Mühendislik Bilimleri ve Tasarım Dergisi Vol.11 Issue 3, pp.1174-1189

Document Type:

Article

Publication Date:

9/2023

Document Language:

Turkish

Document ID (Org No.):

2001299471

pISSN:

1308-6693

Keywords:

Abstract:

 Companies have to meet certain profitability targets to ensure their survival. Purchasing is one of the key activities that directly affect company profitability. Purchasing processes for critical material groups need to be agile and strategic in a changing world. This study presents data mining and machine learning methods for making strategic purchasing decisions and increasing profitability in the steel materials product group. The dataset used in the study includes steel materials features such as consumption, width, length, cut type, thickness, quality and product group. Data mining techniques were used to identify and clean up noisy data in the dataset. Hierarchical clustering and K-means clustering methods were used to analyze the cleaned data set. The cleaned dataset was analyzed using hierarchical clustering and k-means clustering methods and the ideal number of clusters was determined and validated to identify the most strategically important project. The material details of the selected project have been technically analyzed and the purchasing strategies that are expected to bring profitability have been proposed, considering the consumption, thickness and rolling type of the steel material. This study provides an industry-specific perspective on the focused topic and sector in the steel materials product group. By using data mining and machine learning techniques to identify relationships in the steel materials product dataset, the study provides a different perspective that emphasizes the strategic dimension of purchasing decisions. The analyses proposed in this study have reduced the impact of human error in purchasing processes on the development of purchasing strategies, and time-consuming analysis that would normally be carried out by purchasing staff has been replaced by methods such as data mining and machine learning.

 Companies have to meet certain profitability targets to ensure their survival. Purchasing is one of the key activities that directly affect company profitability. Purchasing processes for critical material groups need to be agile and strategic in a changing world. This study presents data mining and machine learning methods for making strategic purchasing decisions and increasing profitability in the steel materials product group. The dataset used in the study includes steel materials features such as consumption, width, length, cut type, thickness, quality and product group. Data mining techniques were used to identify and clean up noisy data in the dataset. Hierarchical clustering and K-means clustering methods were used to analyze the cleaned data set. The cleaned dataset was analyzed using hierarchical clustering and k-means clustering methods and the ideal number of clusters was determined and validated to identify the most strategically important project. The material details of the selected project have been technically analyzed and the purchasing strategies that are expected to bring profitability have been proposed, considering the consumption, thickness and rolling type of the steel material. This study provides an industry-specific perspective on the focused topic and sector in the steel materials product group. By using data mining and machine learning techniques to identify relationships in the steel materials product dataset, the study provides a different perspective that emphasizes the strategic dimension of purchasing decisions. The analyses proposed in this study have reduced the impact of human error in purchasing processes on the development of purchasing strategies, and time-consuming analysis that would normally be carried out by purchasing staff has been replaced by methods such as data mining and machine learning.

 Companies have to meet certain profitability targets to ensure their survival. Purchasing is one of the key activities that directly affect company profitability. Purchasing processes for critical material groups need to be agile and strategic in a changing world. This study presents data mining and machine learning methods for making strategic purchasing decisions and increasing profitability in the steel materials product group. The dataset used in the study includes steel materials features such as consumption, width, length, cut type, thickness, quality and product group. Data mining techniques were used to identify and clean up noisy data in the dataset. Hierarchical clustering and K-means clustering methods were used to analyze the cleaned data set. The cleaned dataset was analyzed using hierarchical clustering and k-means clustering methods and the ideal number of clusters was determined and validated to identify the most strategically important project. The material details of the selected project have been technically analyzed and the purchasing strategies that are expected to bring profitability have been proposed, considering the consumption, thickness and rolling type of the steel material. This study provides an industry-specific perspective on the focused topic and sector in the steel materials product group. By using data mining and machine learning techniques to identify relationships in the steel materials product dataset, the study provides a different perspective that emphasizes the strategic dimension of purchasing decisions. The analyses proposed in this study have reduced the impact of human error in purchasing processes on the development of purchasing strategies, and time-consuming analysis that would normally be carried out by purchasing staff has been replaced by methods such as data mining and machine learning.

 Firmaların varlıklarını sürdürebilmeleri için, belli karlılık hedeflerini tutturmaları gerekmektedir. Firmalarda karlılık hedeflerine doğrudan etki eden faaliyetlerden biri de satın almadır. Değişen dünya koşullarında satın alma süreçlerinin kritik malzeme grupları için çevik ve stratejik olması gerekmektedir. Bu çalışmada, çelik malzeme ürün grubunda stratejik satın alma kararlarının verilmesi ve karlılığın arttırılması için veri madenciliği ve makine öğrenmesi yöntemleri ortaya konmuştur. Veri setinde bulunan gürültülü veriler tespit edilerek veri madenciliği teknikleri ile temizlenmiştir. Temizlenen veri seti makine öğrenmesi tekniklerinden kümeleme analizlerinden hiyerarşik kümeleme ve K-ortalamalar yöntemleri kullanılarak analiz edilmiştir. Bu analizde ideal küme sayısı ve bulunan ideal küme sayısının doğrulaması yapılmış olup stratejik açıdan en önemli proje ortaya konmuştur. Seçilen projede yer alan malzeme detayları teknik olarak incelenip, tüketim, kalınlık ve çelik malzemenin haddeleme tipi dikkate alınarak karlılık getirmesi beklenen satın alma stratejileri ortaya konmuştur. Bu çalışmada önerilen analizler ile satın alma süreçlerinde, çalışan kaynaklı hataların satın alma stratejileri geliştirme süreçlerindeki etkileri azaltılmış, satın alma çalışanlarının uzun zaman harcayarak yapacağı analizler, veri analizi ve makine öğrenmesi gibi endüstri mühendisliği yöntemleri ile gerçekleştirilmiştir.

Article Details

DATA MINING AND MACHINE LEARNING APPLICATIONS IN STEEL MATERIALS PURCHASING

Authors:

Seray Mirasçı | Aslı Aksoy

Publisher:

Süleyman Demirel University

Main Topics:

Endüstri Mühendisliği

Journal:

Mühendislik Bilimleri ve Tasarım Dergisi Vol.11 Issue 3, pp.1174-1189

Document Type:

Article

Publication Date:

9/2023

Document Language:

Turkish

Document ID (Org No.):

2001299471

pISSN:

1308-6693

Data Mining

Machine Learning

Strategic Purchasing

Clustering AnalysesVeri Madenciliği

Makine Öğrenmesi

Stratejik Satın alma

Kümeleme Analizi

Issues

Article Details

Authors: Seray Mirasçı | Aslı Aksoy

Publisher: Süleyman Demirel University

Main Topics: Endüstri Mühendisliği

Journal: Mühendislik Bilimleri ve Tasarım Dergisi Vol.11 Issue 3, pp.1174-1189

Document Type: Article

Publication Date: 9/2023

Document Language: Turkish

Document ID (Org No.): 2001299471

pISSN: 1308-6693

Issues

Authors:

Seray Mirasçı | Aslı Aksoy

Publisher:

Süleyman Demirel University

Main Topics:

Endüstri Mühendisliği

Journal:

Mühendislik Bilimleri ve Tasarım Dergisi Vol.11 Issue 3, pp.1174-1189

Document Type:

Article

Publication Date:

9/2023

Document Language:

Turkish

Document ID (Org No.):

2001299471

pISSN:

1308-6693