Study on Energy Efficiency and Maintenance Optimization of Run-Out Table in Hot Rolling Mills Using Long Short-Term Memory-Autoencoders DOI Creative Commons
Jae-Jung Yun, So-Won Choi, Eul‐Bum Lee

и другие.

Energies, Год журнала: 2025, Номер 18(9), С. 2295 - 2295

Опубликована: Апрель 30, 2025

The steel industry, as a large-scale equipment-intensive sector, emphasizes the importance of maintaining and managing equipment without failure. In line with recent Fourth Industrial Revolution, there is growing shift from preventive to predictive maintenance (PdM) strategies for cost-effective management. This study aims develop PdM model Run-Out Table (ROT) in hot rolling mills plants, utilizing artificial intelligence (AI) technology, propose methods contributing energy efficiency through this model. Considering operational data characteristics ROT equipment, an autoencoder (AE), capable detecting anomalies using only normal data, was selected base Furthermore, Long Short-Term Memory (LSTM) networks were chosen address time-series nature data. By integrating technical advantages these two algorithms, based on LSTM-AE algorithm, named Predictive Maintenance Model (ROT-PMM), developed. Additionally, concept anomaly ratio applied identify each coil production. performance evaluation ROT-PMM demonstrated F1-score 91%. differentiates itself by developing optimized that considers specific environment operation enhancing its applicability verification actual failure it efficiency. It expected research will contribute increased productivity industrial settings, including industry.

Язык: Английский

Deep learning-stochastic ensemble for RUL prediction and predictive maintenance with dynamic mission abort policies DOI

A. Faizanbasha,

U. Rizwan

Reliability Engineering & System Safety, Год журнала: 2025, Номер unknown, С. 110919 - 110919

Опубликована: Фев. 1, 2025

Язык: Английский

Процитировано

2
Predictive Maintenance Algorithms, Artificial Intelligence Digital Twin Technologies, and Internet of Robotic Things in Big Data-Driven Industry 4.0 Manufacturing Systems DOI Creative Commons
Marek Nagy,

Marcel Figura,

Katarína Valašková

и другие.

Mathematics, Год журнала: 2025, Номер 13(6), С. 981 - 981

Опубликована: Март 17, 2025

In Industry 4.0, predictive maintenance (PdM) is key to optimising production processes. While its popularity among companies grows, most studies highlight theoretical benefits, with few providing empirical evidence on economic impact. This study aims fill this gap by quantifying the performance of manufacturing in Visegrad Group countries through PdM algorithms. The purpose our research assess whether these generate higher operational profits and lower sales costs. Using descriptive statistics, non-parametric tests, Hodges–Lehmann median difference estimate, linear regression, authors analysed data 1094 enterprises. Results show that significantly improves performance, variations based geographic scope. Regression analysis confirmed as an essential predictor even after considering factors like company size, legal structure, Enterprises more effective cost management net were likely adopt PdM, revealed decision tree analysis. Our findings provide benefits algorithms their potential enhance competitiveness, offering a valuable foundation for business managers make informed investment decisions encouraging further other industries.

Язык: Английский

Процитировано

2
Optimizing replacement times and Total Expected Discounted Costs in coherent systems using Geometric Point Process DOI

A. Faizanbasha,

U. Rizwan

Computers & Industrial Engineering, Год журнала: 2025, Номер unknown, С. 110879 - 110879

Опубликована: Янв. 1, 2025

Язык: Английский

Процитировано

1
Study on Energy Efficiency and Maintenance Optimization of Run-Out Table in Hot Rolling Mills Using Long Short-Term Memory-Autoencoders DOI Creative Commons
Jae-Jung Yun, So-Won Choi, Eul‐Bum Lee

и другие.

Energies, Год журнала: 2025, Номер 18(9), С. 2295 - 2295

Опубликована: Апрель 30, 2025

The steel industry, as a large-scale equipment-intensive sector, emphasizes the importance of maintaining and managing equipment without failure. In line with recent Fourth Industrial Revolution, there is growing shift from preventive to predictive maintenance (PdM) strategies for cost-effective management. This study aims develop PdM model Run-Out Table (ROT) in hot rolling mills plants, utilizing artificial intelligence (AI) technology, propose methods contributing energy efficiency through this model. Considering operational data characteristics ROT equipment, an autoencoder (AE), capable detecting anomalies using only normal data, was selected base Furthermore, Long Short-Term Memory (LSTM) networks were chosen address time-series nature data. By integrating technical advantages these two algorithms, based on LSTM-AE algorithm, named Predictive Maintenance Model (ROT-PMM), developed. Additionally, concept anomaly ratio applied identify each coil production. performance evaluation ROT-PMM demonstrated F1-score 91%. differentiates itself by developing optimized that considers specific environment operation enhancing its applicability verification actual failure it efficiency. It expected research will contribute increased productivity industrial settings, including industry.

Язык: Английский

Процитировано

0