Predicting Industry Maturity Index Using Machine Learning Methods

Ayhan Doğan; Cihan Ünal

Makale detayları

Predicting Industry Maturity Index Using Machine Learning Methods

Yazarlar:

Ayhan Doğan | Cihan Ünal

Yayıncı:

Istanbul University

Ana Başlıklar:

Makine Öğrenme (Diğer)

Dergi:

Acta Infologica Vol.9 Issue 1, pp.293-313

Belge Tipi:

Makale

Yayın Tarihi:

6/2025

Belge Dili:

İngilizce

Document ID (Org No.):

2001354880

pISSN:

2602-3563

Anahtar kelimeler:

Özet:

 Industry 4.0 has become a widely adopted concept in recent years. Maturity and readiness models are commonly used to assess the current state of industrial organizations in relation to Industry 4.0. Companies maturity levels and index scores are typically determined through structured surveys. However, due to their complexity, time consumption, and high cost, many enterprises lack formally assessed maturity index MI scores. To address this limitation, this study initially employed survey data to evaluate the accuracy of the proposed machine learning ML framework. A 58-question survey was conducted to calculate the MI scores of the companies. These scores were then used as reference values to be predicted based on five easily accessible enterprise-level variables: company age, industry type, ownership structure, number of employees, and annual turnover. This approach tested whether MI could be accurately predicted without relying on lengthy survey processes, using only a minimal set of key enterprise attributes. The results of this study demonstrate that MI can be estimated successfully using ML techniques without the need for answering long and complex surveys. To reduce the burdens associated with conventional survey-based methods, this study employed multiple ML algorithms, including Support Vector Machines SVM, Gaussian Process Regression GPR, Linear Regression LR, Regression Trees RT, and Ensemble Tree-based models, and advanced boosting-based methods, such as extreme gradient boosting XGB and Light Gradient Boosting Machine LGBM. The findings demonstrate that the proposed model predicts MI with high accuracy and offers a practical and scalable alternative for enterprises seeking to assess their Industry 4.0 readiness.

 Industry 4.0 has become a widely adopted concept in recent years. Maturity and readiness models are commonly used to assess the current state of industrial organizations in relation to Industry 4.0. Companies maturity levels and index scores are typically determined through structured surveys. However, due to their complexity, time consumption, and high cost, many enterprises lack formally assessed maturity index MI scores. To address this limitation, this study initially employed survey data to evaluate the accuracy of the proposed machine learning ML framework. A 58-question survey was conducted to calculate the MI scores of the companies. These scores were then used as reference values to be predicted based on five easily accessible enterprise-level variables: company age, industry type, ownership structure, number of employees, and annual turnover. This approach tested whether MI could be accurately predicted without relying on lengthy survey processes, using only a minimal set of key enterprise attributes. The results of this study demonstrate that MI can be estimated successfully using ML techniques without the need for answering long and complex surveys. To reduce the burdens associated with conventional survey-based methods, this study employed multiple ML algorithms, including Support Vector Machines SVM, Gaussian Process Regression GPR, Linear Regression LR, Regression Trees RT, and Ensemble Tree-based models, and advanced boosting-based methods, such as extreme gradient boosting XGB and Light Gradient Boosting Machine LGBM. The findings demonstrate that the proposed model predicts MI with high accuracy and offers a practical and scalable alternative for enterprises seeking to assess their Industry 4.0 readiness.

 Industry 4.0 has become a widely adopted concept in recent years. Maturity and readiness models are commonly used to assess the current state of industrial organizations in relation to Industry 4.0. Companies maturity levels and index scores are typically determined through structured surveys. However, due to their complexity, time consumption, and high cost, many enterprises lack formally assessed maturity index MI scores. To address this limitation, this study initially employed survey data to evaluate the accuracy of the proposed machine learning ML framework. A 58-question survey was conducted to calculate the MI scores of the companies. These scores were then used as reference values to be predicted based on five easily accessible enterprise-level variables: company age, industry type, ownership structure, number of employees, and annual turnover. This approach tested whether MI could be accurately predicted without relying on lengthy survey processes, using only a minimal set of key enterprise attributes. The results of this study demonstrate that MI can be estimated successfully using ML techniques without the need for answering long and complex surveys. To reduce the burdens associated with conventional survey-based methods, this study employed multiple ML algorithms, including Support Vector Machines SVM, Gaussian Process Regression GPR, Linear Regression LR, Regression Trees RT, and Ensemble Tree-based models, and advanced boosting-based methods, such as extreme gradient boosting XGB and Light Gradient Boosting Machine LGBM. The findings demonstrate that the proposed model predicts MI with high accuracy and offers a practical and scalable alternative for enterprises seeking to assess their Industry 4.0 readiness.

 Industry 4.0 has become a widely adopted concept in recent years. Maturity and readiness models are commonly used to assess the current state of industrial organizations in relation to Industry 4.0. Companies maturity levels and index scores are typically determined through structured surveys. However, due to their complexity, time consumption, and high cost, many enterprises lack formally assessed maturity index MI scores. To address this limitation, this study initially employed survey data to evaluate the accuracy of the proposed machine learning ML framework. A 58-question survey was conducted to calculate the MI scores of the companies. These scores were then used as reference values to be predicted based on five easily accessible enterprise-level variables: company age, industry type, ownership structure, number of employees, and annual turnover. This approach tested whether MI could be accurately predicted without relying on lengthy survey processes, using only a minimal set of key enterprise attributes. The results of this study demonstrate that MI can be estimated successfully using ML techniques without the need for answering long and complex surveys. To reduce the burdens associated with conventional survey-based methods, this study employed multiple ML algorithms, including Support Vector Machines SVM, Gaussian Process Regression GPR, Linear Regression LR, Regression Trees RT, and Ensemble Tree-based models, and advanced boosting-based methods, such as extreme gradient boosting XGB and Light Gradient Boosting Machine LGBM. The findings demonstrate that the proposed model predicts MI with high accuracy and offers a practical and scalable alternative for enterprises seeking to assess their Industry 4.0 readiness.

Makale detayları

Predicting Industry Maturity Index Using Machine Learning Methods

Yazarlar:

Ayhan Doğan | Cihan Ünal

Yayıncı:

Istanbul University

Ana Başlıklar:

Makine Öğrenme (Diğer)

Dergi:

Acta Infologica Vol.9 Issue 1, pp.293-313

Belge Tipi:

Makale

Yayın Tarihi:

6/2025

Belge Dili:

İngilizce

Document ID (Org No.):

2001354880

pISSN:

2602-3563

Machine learning

Industry 4.0

Maturity model

Maturity index

Sayı

Makale detayları

Predicting Industry Maturity Index Using Machine Learning Methods

Yazarlar: Ayhan Doğan | Cihan Ünal

Yayıncı: Istanbul University

Ana Başlıklar: Makine Öğrenme (Diğer)

Dergi: Acta Infologica Vol.9 Issue 1, pp.293-313

Belge Tipi: Makale

Yayın Tarihi: 6/2025

Belge Dili: İngilizce

Document ID (Org No.): 2001354880

pISSN: 2602-3563

Machine learning

Industry 4.0

Maturity model

Maturity index

Sayı

Yazarlar:

Ayhan Doğan | Cihan Ünal

Yayıncı:

Istanbul University

Ana Başlıklar:

Makine Öğrenme (Diğer)

Dergi:

Acta Infologica Vol.9 Issue 1, pp.293-313

Belge Tipi:

Makale

Yayın Tarihi:

6/2025

Belge Dili:

İngilizce

Document ID (Org No.):

2001354880

pISSN:

2602-3563