A primer on predictive maintenance: Potential benefits and practical challenges

Abstract

We present an overview of predictive maintenance (PdM) in industrial operations, highlighting its evolution, benefits, challenges, and potential economic impact. It is assessed that there are considerable benefits associated with PdM but also its practical implementation can prove difficult due to the uncertain benefits of its application, the need for advanced IT infrastructure as well as expert personnel, and, perhaps most importantly, a lack of failure-related data for the PdM model training. We conclude with three PdM case studies to elaborate on some of the issues discussed throughout the paper.

Keywords:

• Condition monitoring
• maintenance strategies
• power plants
• preventive maintenance
• run-to-failure
• statistical process control
• wind turbines

Acknowledgments

As an employee of Ørsted, the first author expresses gratitude to the company for providing the resources, data, and infrastructure necessary to conduct this research.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The first author’s PhD research is entirely funded by Ørsted, which provided the first two case studies. The funding for the third case study came from the Swedish Strategic Innovation Programme ‘InfraSweden2030’ (Project Number: 2016-04757), Trafikverket (Swedish Transportation Administration, STA), Luleå Railway Research Centre (JVTC), and Predge AB.

Notes on contributors

Henrik Hviid Hansen

Henrik Hviid Hansen is a senior data scientist at the Danish energy company Ørsted, where he has worked on a PhD project on the predictive maintenance of power plants. His research interests are in the fields of fault detection and diagnosis, remaining useful life prediction, statistical process control, and machine learning for the use in optimizing power plant processes.

Murat Kulahci

Murat Kulahci is a professor at the Technical University of Denmark and Luleå University of Technology in Sweden. His research currently focuses primarily on large data analytics for descriptive, inferential, and predictive purposes. Many of his research applications involve high dimensional, high frequency data demanding analysis methods in chemometrics and machine learning. He has been collaborating with various industries in many industrial statistics projects and digital manufacturing.

Bo Friis Nielsen

Bo Friis Nielsen is a professor of applied probability in the engineering sciences at the Technical University of Denmark. Besides theoretical contributions to the field of applied probability, he has worked with researchers in transportation and health science. His main contributions concern the theory of uni- and multivariate matrix-exponential distributions.