•  
  •  
 

Turkish Journal of Electrical Engineering and Computer Sciences

Author ORCID Identifier

OZLEM YUREK: 0000-0003-0919-0149

DERYA BIRANT: 0000-0003-3138-0432

DOI

10.55730/1300-0632.4086

Abstract

Predictive maintenance (PdM), a fundamental element of modern industrial systems, employs machine learning to monitor equipment conditions, estimate failure probabilities, and optimize maintenance schedules. Its core objective is to enhance equipment reliability, extend lifespan, and minimize costs through data-driven insights by enabling efficient maintenance scheduling, reducing downtime, and optimizing resource allocation. In this paper, we propose a novel ordinal predictive maintenance with ensemble binary decomposition (OPMEB) method for the PdM domain, considering the hierarchical nature of class labels reflecting the machine's health status, including categories like healthy, low risk, moderate risk, and high risk. The proposed OPMEB method was validated by executing on the C-MAPSS, AI4I 2020, and a real-world hydraulic system's condition datasets. The experimental outcomes were evaluated with four distinct metrics: accuracy, recall, precision, and F-measure. The findings showed the improvement in the model's predictive capabilities achieved by the proposed approach when compared to the traditional ordinal classification algorithm. Furthermore, the experimental results demonstrated the superior performance of the OPMEB method over other ordinal binary decomposition methods, including OneVsAll, OneVsFollowers, and OneVsNext.

Keywords

Predictive maintenance, ordinal classification, binary decomposition, machine learning, classification, ensemble learning

First Page

534

Last Page

554

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Download

Share

COinS