Figure 5 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Source publication
The increasing interconnection of machines in industrial production on one hand, and the improved capabilities to store, retrieve, and analyze large amounts of data on the other, offer promising perspectives for maintaining production machines. Recently, predictive maintenance has gained increasing attention in the context of equipment maintenance...
Contexts in source publication
Context 1
... simplify the categorization, we summarized Web of Science categories to sections. Figure 5 illustrates all sections (In addition, the left side of Figure 6 illustrates the distribution of sections for the search string predictive maintenance.) on one hand and the ones identified as being relevant on the other. ...
Context 2
... on the relevant Web of Science categories, we filter articles utilizing the white list principle, i.e., we keep an article in the data set, if it matches at least with one of the relevant categories. As can be obtained from Figure 5, we excluded a lot of Web of Science categories for the search string predictive maintenance. First, we calculated an article density for sections (cf. Figure 6, right side). ...
Context 3
... 2000Period -2008 For the second period, the research fronts are shown in a 3D visualization (cf. Figure 15). Figure 15 is based on the same calculated data as Figure 10 (cf. ...
Similar publications
The issue of "value creation" has been discussed for a long time, and this paper takes 1203 papers on value creation research published in CNKI database from 2017 to 2022 as the research sample, and conducts bibliometric and visualization analysis with the help of CiteSpace software. The paper focuses on the hotspots and trends of domestic value cr...
Citations
... It effectively explains how a research domain has emerged and grown over a certain period. It helps remarkably in the visualization of the research articles based on various lenses such as key co-occurrence, author affiliations, origin countries and universities (Hoppenstedt et al., 2018). The present study considers KS in the COVID-19 context to review the state of KS and add more value to the gamut of knowledge management (KM) literature, which, as a growing research domain, would be a source of "sustainable competitive advantage" aiding in building resilient organizations (Farooq and Vij, 2019). ...
Purpose
This study aims to conduct a bibliometric analysis of knowledge sharing during COVID-19 and highlight prominent contributors, diverse trends and themes followed with provisions of future research avenues.
Design/methodology/approach
The study through scientific procedures and rationales for systematic literature reviews framework analyses 148 peer-reviewed journal publications and conference proceedings indexed in Scopus and WoS databases from 2020 to 2022. It uses general statistics and diverse bibliometric techniques, including co-occurrence analysis for trend and cluster identification in the literature.
Findings
The findings reveal an exponential annual growth rate of 150% in the domain, highlighting the global research focus. With regards to domain contribution, the Journal of Knowledge Management and China leads with ten publications in their respective categories. The co-occurrence analysis further highlights four diverse clusters in the domain, which are further discussed in detail. The study highlights significant contributions from developed economies, thus providing scope for future research from developing or transitioning economies in the Middle East, Central Asia or Africa. The study concludes by presenting the elementary role of knowledge sharing in response to external crises.
Originality/value
The interest in the knowledge-sharing domain has grown exponentially during the COVID-19 pandemic. This research is the first bibliometric analysis with comprehensive and rigorous analytic techniques to unearth critical developments and insights for a holistic understanding.
... 7 Many other valuable references depict the PdM system from varied perspectives. [8][9][10][11][12][13] In summary, the current PdM development is still looking for a system-level approach to address the challenges from multiple components and their faults in a complex system. In addition, the heterogeneity of the data sources within the system makes a difficult task for the PdM system development. ...
... Since both computer-aided engineering [1] and time series-based machine learning (ML) have shown significant advancements in recent years, an increasing number of manufacturers rely on ML-based solutions to create value out of sensor data. In particular, fields like predictive maintenance [2], condition monitoring [3], and anomaly detection [4] have achieved immense progress and have become well-known terms in the academic and industrial world. ...
Industrial data scarcity is one of the largest factors holding back the widespread use of machine learning in manufacturing. To overcome this problem, the concept of transfer learning was developed and has received much attention in recent industrial research. This paper focuses on the problem of time series segmentation and presents the first in-depth research on transfer learning for deep learning-based time series segmentation on the industrial use case of end-of-line pump testing. In particular, we investigate whether the performance of deep learning models can be increased by pretraining the network with data from other domains. Three different scenarios are analyzed: source and target data being closely related, source and target data being distantly related, and source and target data being non-related. The results demonstrate that transfer learning can enhance the performance of time series segmentation models with respect to accuracy and training speed. The benefit can be most clearly seen in scenarios where source and training data are closely related and the number of target training data samples is lowest. However, in the scenario of non-related datasets, cases of negative transfer learning were observed as well. Thus, the research emphasizes the potential, but also the challenges, of industrial transfer learning.
... Since both computer-aided engineering [1] and time series-based machine learning (ML) showed significant advancements in recent years, an increasing number of manufacturers rely on ML-based solutions to create value out of sensor data. Especially fields like predictive maintenance [2], condition monitoring [3], and anomaly detection [4] have achieved immense progress and have become well-known terms in the academic and industrial world. ...
Industrial data scarcity is one of the largest factors holding back the widespread use of machine learning in manufacturing. To overcome this problem, the concept of transfer learning was developed and it achieved high attention in recent industrial research. Our paper focuses on the problem of time series segmentation and presents the first in-depth research about transfer learning for deep-learning based time series segmentation on the example of industrial end-of-line pump testing. In particularly, we investigate if the performance of deep learning models can be increased by pretraining the network with data from other domains. Three different scenarios are analyzed: source and target data being closely related, source and target data being distantly related, and source and target data being non-related. The results demonstrate that transfer learning can enhance the performance of time series segmentation models in respect to accuracy and training speed. The benefit is most clearly seen in scenarios where source and training data are closely related and the number of target training data samples is lowest. However, in the scenario of non-related datasets, cases of negative Transfer Learning were observed as well. Thus, the research emphasizes the potential, but also the challenges of industrial Transfer Learning.
... The BPM-IoT Manifesto discusses the benefits and challenges of bringing the worlds of BPM and IoT together [1]. Sensors and actuators are new sources of real-time data that provide opportunities to develop advanced monitoring approaches, e.g., for condition monitoring, predictive maintenance [35] or verification of process outcomes [36]. With our work we aim at contributing a solution to the challenge Bridging the Gap Between Event-Based and Process-Based Systems as we raise the abstraction from IoT sensor and actuator events to the level of business processes. ...
The increasing number of IoT devices equipped with sensors and actuators pervading every domain of everyday life allows for improved automated monitoring and analysis of processes executed in IoT-enabled environments. While sophisticated analysis methods exist to detect specific types of activities from low-level IoT data, a general approach for detecting activity executions that are part of more complex business processes does not exist. Moreover, dedicated information systems to orchestrate or monitor process executions are not available in typical IoT environments. As a consequence, the large corpus of existing process analysis and mining techniques to check and improve process executions cannot be applied. In this work, we develop an interactive method guiding the analysis of low-level IoT data with the goal of detecting higher-level process activity executions. The method is derived following the exploratory data analysis of an IoT data set from a smart factory. We propose analysis steps, sensor-actuator-activity patterns, and the novel concept of activity signatures that are applicable in many IoT domains. The method shows to be valuable for the early stages of IoT data analyses to build a ground truth based on domain knowledge and decisions of the process analyst, which can be used for automated activity detection in later stages.
... Another subdomain of informetrics that analyses internet data is webometrics [1]. The subdomain of bibliometrics that defines a quantitative approach for literature research is called bibliometric analysis [13]. Within this work, the term bibliometric analysis (or bibliometric analyses) is used whenever statistical methods to analyze bibliographic information of academic publications are discussed. ...
... A co-citation of two publications is given if both publications are cited by a third publication [1] (see Figure 3). The two publications are then considered a knowledge base [13]. In a co-citation network, the similarity matrix S cc is calculated by S cc = A T pc A pc with size n × n [1]. ...
... If one or more citations of two publications are similar, the two publications are considered bibliographically coupled [1] (see Figure 3). Bibliographically coupled publications form a research front [13]. To compute a bibliographic coupling network, the same bipartite graph and, thus, matrix A pc as for co-citation networks is used; however, the similarity matrix S bc is given by S bc = A pc A T pc with size m × m [1]. ...
A deep understanding about a field of research is valuable for academic researchers. In addition to technical knowledge, this includes knowledge about subareas, open research questions, and social communities (networks) of individuals and organizations within a given field. With bibliometric analyses, researchers can acquire quantitatively valuable knowledge about a research area by using bibliographic information on academic publications provided by bibliographic data providers. Bibliometric analyses include the calculation of bibliometric networks to describe affiliations or similarities of bibliometric entities (e.g., authors) and group them into clusters representing subareas or communities. Calculating and visualizing bibliometric networks is a nontrivial and time-consuming data science task that requires highly skilled individuals. In addition to domain knowledge, researchers must often provide statistical knowledge and programming skills or use software tools having limited functionality and usability. In this paper, we present the ambalytics bibliometric platform, which reduces the complexity of bibliometric network analysis and the visualization of results. It accompanies users through the process of bibliometric analysis and eliminates the need for individuals to have programming skills and statistical knowledge, while preserving advanced functionality, such as algorithm parameterization, for experts. As a proof-of-concept, and as an example of bibliometric analyses outcomes, the calculation of research fronts networks based on a hybrid similarity approach is shown. Being designed to scale, ambalytics makes use of distributed systems concepts and technologies. It is based on the microservice architecture concept and uses the Kubernetes framework for orchestration. This paper presents the initial building block of a comprehensive bibliometric analysis platform called ambalytics, which aims at a high usability for users as well as scalability.
... anks to the development of IIoT, machine learning algorithms are available to utilize as the predictive model, including survival analysis, extremely randomized trees, k-nearest neighbors, and convolutional neural network (CNN) [5][6][7][8]. Take the research from Susto et al. as typical example [1]. ...
Recently, the development of the Industrial Internet of Things (IIoT) has led enterprises to re-examine the research of the equipment-state-prediction models and intelligent manufacturing applications. Take industrial robots as typical example. Under the effect of scale, robot maintenance decision seriously affects the cost of spare parts and labor deployment. In this paper, an evaluation method is proposed to predict the state of robot lubricating oil based on support vector regression (SVR). It would be the proper model to avoid the structural risks and minimize the effect of small sample volume. IIoT technology is used to collect and store the valuable robot running data. The key features of the running state of the robot are extracted, and the machine learning model is applied according to the measured element contents of the lubricating oil. As a result, the cost of spare parts consumption can be saved for more than two million CNY per year.
... To reach this vision and the expected effects, PdM has gained exponentially increasing attention in the context of prognostic and health management (PHM) and has been started to be adopted by many manufacturing companies in industry. Beside potential effects in industrial applications of PdM, it also holds a great interest in academia with rapidly increasing journal and conference contributions [13], [26]. In principle, PdM predicts faults and failures by real-time evaluation of the system state in production and thereby enable for efficient maintenance operations [44]. ...
In this paper, we propose a coherent framework for multi-machine analysis, using a group clustering model, which can be utilized for predictive maintenance (PdM). The framework benefits from the repetitive structure posed by multiple machines and enables for assessment of health condition, degradation modeling and comparison of machines. It is based on a hierarchical probabilistic model, denoted Gaussian topic model (GTM), where cluster patterns are shared over machines and therefore it allows one to directly obtain proportions of patterns over the machines. This is then used as a basis for cross comparison between machines where identified similarities and differences can lead to important insights about their degradation behavior. The framework is based on aggregation of data over multiple streams by a predefined set of features extracted over a time window. Moreover, the framework contains a clustering schema which takes uncertainty of cluster assignments into account and where one can specify a desirable degree of reliability of the assignments. By using a multi-machine simulation example, we highlight how the framework can be utilized in order to obtain cluster patterns and inherent variations of such patterns over machines. Furthermore, a comparative study with the commonly used Gaussian mixture model (GMM) demonstrates that GTM is able to identify inherent patterns in the data while the GMM fails. Such result is a consequence of the group level being modeled by the GTM while being absent in the GMM. Hence, the GTM are trained with a view on the data that is not available to the GMM with the consequence that the GMM can miss important, possibly even key, cluster patterns. Therefore, we argue that more advanced cluster models, like the GTM, can be key for interpreting and understanding degradation behavior across machines and ultimately for obtaining more efficient and reliable PdM systems.
... Internet-of-Things (IoT) technology has opened up many opportunities to obtain data and improve processes in supply chain management ( Ben-Daya, Hassini, & Bahroun, 2019 ). Applications include temperature sensors that track the functionality of reefer containers ( Badia-Melis, McCarthy, Ruiz-Garcia, Garcia-Hierro, & Robla Villalba, 2018 ), gyroscopes that monitor machine conditions and trigger predictive maintenance ( Hoppenstedt et al., 2018 ), and GPS sensors that allow to trace and optimize the routing of delivery trucks ( Braekers, Ramaekers, & Van Nieuwenhuyse, 2016 ). Another interesting application area emerges as manufacturers aim to establish new IoT-based business models ( Porter & Heppelmann, 2014;2015 ). ...
Motivated by recent advances in Internet-of-Things (IoT) technology for household appliances, we analyze a Smart Replenishment system that leverages point-of-consumption (POC) information at end consumers to decide on deliveries of consumables. As such, we extend the classic Vendor-Managed Inventory (VMI) concept to end consumers. We model the system for a single manufacturer who directly serves N end consumers with uncertain demand. End consumers partially adopt the new Smart Replenishment mode, which results in a mix of VMI and non-VMI customers. We assume that unfulfilled demand is lost and that the manufacturer’s dispatch capacity is constrained. Customers compete for the same capacity while featuring different out-of-stock risks and service-level expectations, both of which are costly to the manufacturer. Considering various adoption levels, we decide on the design of such a system and focus on (i) inventory control, (ii) customer prioritization, and (iii) degree of smart, integrated decision-making. Using discrete-event simulation and a full-factorial experiment, we show that replenishment decisions can be significantly enhanced with POC information. It leads to substantial improvements in service levels and capacity utilization without loading customers with inventories. This improvement potential is highest for a low demand coverage of the replenishment quantity, a high gap in the ordering behavior of manufacturer and end consumers, and a long lead time. To realize this improvement potential, we propose a flexible reorder corridor to manage inventories at VMI customers that balances the trade-off between out-of-stock risk and service-level expectation inherent in the system.
... In recent years, there has been an evolution of research into the thermal method of non-destructive testing [7]. The increase in the demand for the thermal method is due to the fact that any mechanical energy is always accompanied by the release of heat energy. ...
An improved method of passive thermal control has been proposed in order to operatively assess the technical condition of rolling bearings, based on studying the heating rate of a bearing. The values of the heating rate of ball, roller, and conical rolling bearings, as well as ball separators, have been determined empirically. It has been shown that the discrete limit heating rate values derived under the regular heating mode of mechanical reducers during stand tests are suitable for use as a diagnostic criterion for rolling bearings. Based on the heating speed criterion for mechanical reducers, it is possible to perform an operative assessment of the technical condition of rolling bearings of different types during the operation of equipment for different purposes. It has been established that for a rolling bearing in a working technical condition the heating rate under a regular heating mode does not exceed 1 °C/min. The resulting value corresponds to the diagnostic criterion for the heating rate of mechanical reducers in a working technical condition of ϑн≤1.1 °C/min. Using a diagnostic parameter of the heating rate under a regular heating mode makes it possible to resolve the issue related to the duration of control over the technical condition of a rolling bearing using a thermal method. Reducing the control procedure duration, when using the improved thermal non-destructive testing, by 4 times, would yield a reduction in operating costs due to the possibility of ongoing control over rolling bearings at the beginning of the equipment operation. It has also been confirmed that the heating rate of rolling bearings under a regular heating mode directly depends on their technical condition and does not depend on the load transferred and the mode of operation of the bearing nodes of the technological equipment. The proposed method could be applied in the maintenance system based on the actual condition of the equipment for the operative control over rolling bearings
