A Survey on Predictive Maintenance for Industry 4.0

Krupitzer, Christian; Wagenhals, Tim; Züfle, Marwin; Lesch, Veronika; Schäfer, Dominik; Mozaffarin, Amin; Edinger, Janick; Becker, Christian; Kounev, Samuel

doi:10.48550/arxiv.2002.08224

Cited by 10 publications

(13 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A proactive approach to maintenance in line with the Maintenance 4.0 concept is currently the most technologically advanced form of maintenance for technical systems. In the literature and practice of the issue under study, it is often emphasized that the Maintenance 4.0 concept is a practical implementation of PdM and SMART PdM solutions [ 80 , 81 ]. Therefore, references can be found to the so-called Predictive Maintenance 4.0 concept.…”

Section: Theoretical Backgroundmentioning

confidence: 99%

“…The identification of the main problems and issues raised in the context of Maintenance 4.0 was based on an extensive review of the available literature. The prepared literature analysis was also supplemented by review publications in the area of Industry 4.0 or Maintenance 4.0 (e.g., works [ 17 , 38 , 42 , 80 ]) or reports in the field of Maintenance 4.0 and PdM 4.0 (e.g., [ 10 , 53 , 108 ]). As a result of the research carried out, five primary research areas were defined, which have been most extensively developed over the past few years.…”

Section: Systematic Literature Review Of the Selected Papers Within T...mentioning

confidence: 99%

See 1 more Smart Citation

Maintenance Performance in the Age of Industry 4.0: A Bibliometric Performance Analysis and a Systematic Literature Review

Werbińska-Wojciechowska

Winiarska

2023

Sensors

View full text Add to dashboard Cite

Recently, there has been a growing interest in issues related to maintenance performance management, which is confirmed by a significant number of publications and reports devoted to these problems. However, theoretical and application studies indicate a lack of research on the systematic literature reviews and surveys of studies that would focus on the evolution of Industry 4.0 technologies used in the maintenance area in a cross-sectional manner. Therefore, the paper reviews the existing literature to present an up-to-date and content-relevant analysis in this field. The proposed methodology includes bibliometric performance analysis and a review of the systematic literature. First, the general bibliometric analysis was conducted based on the literature in Scopus and Web of Science databases. Later, the systematic search was performed using the Primo multi-search tool following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The main inclusion criteria included the publication dates (studies published from 2012–2022), studies published in English, and studies found in the selected databases. In addition, the authors focused on research work within the scope of the Maintenance 4.0 study. Therefore, papers within the following research fields were selected: (a) augmented reality, (b) virtual reality, (c) system architecture, (d) data-driven decision, (e) Operator 4.0, and (f) cybersecurity. This resulted in the selection of the 214 most relevant papers in the investigated area. Finally, the selected articles in this review were categorized into five groups: (1) Data-driven decision-making in Maintenance 4.0, (2) Operator 4.0, (3) Virtual and Augmented reality in maintenance, 4) Maintenance system architecture, and 5) Cybersecurity in maintenance. The obtained results have led the authors to specify the main research problems and trends related to the analyzed area and to identify the main research gaps for future investigation from academic and engineering perspectives.

show abstract

Section: Theoretical Backgroundmentioning

confidence: 99%

Section: Systematic Literature Review Of the Selected Papers Within T...mentioning

confidence: 99%

Maintenance Performance in the Age of Industry 4.0: A Bibliometric Performance Analysis and a Systematic Literature Review

Werbińska-Wojciechowska

Winiarska

2023

Sensors

View full text Add to dashboard Cite

show abstract

“…The most widely adopted methodologies are predictive and preventive maintenance. While the first places emphasis on estimating the time of failure to enable scheduled maintenance—and therefore production line down-times—the other aims to establish a regular, periodic maintenance process for preventing failures and keeping machinery operational for as long as possible by expanding its life-time [ 131 , 132 , 133 ], as shown in Figure 6 .…”

Section: Maintenancementioning

confidence: 99%

Applied Machine Learning for IIoT and Smart Production—Methods to Improve Production Quality, Safety and Sustainability

Frankó

Hollósi

Ficzere

et al. 2022

Sensors

View full text Add to dashboard Cite

Industrial IoT (IIoT) has revolutionized production by making data available to stakeholders at many levels much faster, with much greater granularity than ever before. When it comes to smart production, the aim of analyzing the collected data is usually to achieve greater efficiency in general, which includes increasing production but decreasing waste and using less energy. Furthermore, the boost in communication provided by IIoT requires special attention to increased levels of safety and security. The growth in machine learning (ML) capabilities in the last few years has affected smart production in many ways. The current paper provides an overview of applying various machine learning techniques for IIoT, smart production, and maintenance, especially in terms of safety, security, asset localization, quality assurance and sustainability aspects. The approach of the paper is to provide a comprehensive overview on the ML methods from an application point of view, hence each domain—namely security and safety, asset localization, quality control, maintenance—has a dedicated chapter, with a concluding table on the typical ML techniques and the related references. The paper summarizes lessons learned, and identifies research gaps and directions for future work.

show abstract

“…These measurements are typically captured by monitoring systems as multi-variate time series data sets. Examples can be found in the Internet-of-Things (IoT) or in the DevOps/AIOps space (Lu et al, 2009;Lohrmann and Kao, 2011;Nedelkoski et al, 2019;Li et al, 2020;Krupitzer et al, 2020). Our running example is that of a wind turbine; see Figure 1.…”

Section: Introductionmentioning

confidence: 99%

Intrinsic Anomaly Detection for Multi-Variate Time Series

Rabanser¹,

Januschowski²,

Rasul³

et al. 2022

Preprint

View full text Add to dashboard Cite

We introduce a novel, practically relevant variation of the anomaly detection problem in multi-variate time series: intrinsic anomaly detection. It appears in diverse practical scenarios ranging from DevOps to IoT, where we want to recognize failures of a system that operates under the influence of a surrounding environment. Intrinsic anomalies are changes in the functional dependency structure between time series that represent an environment and time series that represent the internal state of a system that is placed in said environment. We formalize this problem, provide under-studied public and new purpose-built data sets for it, and present methods that handle intrinsic anomaly detection. These address the shortcoming of existing anomaly detection methods that cannot differentiate between expected changes in the system's state and unexpected ones, i.e., changes in the system that deviate from the environment's influence. Our most promising approach is fully unsupervised and combines adversarial learning and time series representation learning, thereby addressing problems such as label sparsity and subjectivity, while allowing to navigate and improve notoriously problematic anomaly detection data sets.

show abstract

A Survey on Predictive Maintenance for Industry 4.0

Cited by 10 publications

References 49 publications

Maintenance Performance in the Age of Industry 4.0: A Bibliometric Performance Analysis and a Systematic Literature Review

Maintenance Performance in the Age of Industry 4.0: A Bibliometric Performance Analysis and a Systematic Literature Review

Applied Machine Learning for IIoT and Smart Production—Methods to Improve Production Quality, Safety and Sustainability

Intrinsic Anomaly Detection for Multi-Variate Time Series

Contact Info

Product

Resources

About