Search

The ongoing digitization of the industry leads to new opportunities in industrial maintenance. Several concepts for smart maintenance have been published in literature, however, their application remains limited in industrial practice. To close the existing gap between research and practice, a novel practice-oriented approach is presented. A modified Weibull model is developed and embodied into a digital assistance system to visualize and predict remaining service life and to forecast the required number of spare parts. The developed model has been implemented in the use case in the heat treatment industry, and the results are presented to demonstrate the effectiveness of the proposed approach.

In the era of Industry 4.0, predictive maintenance is regarded as a key factor for reaching business objectives. Cost-effective deployment of Cyber Physical Production Systems raises the question whether data-driven and knowledge-based maintenance affects profitability of smart factories. Several studies reveal that an appropriate data-driven maintenance strategy should not only focus on increasing availability but also should consider economic parameters. This paper presents a novel approach and a proof-of-concept demonstrator using Dynamic Bayesian Networks (DBN). The proposed DBN model enables identifying and predicting the economic impact of maintenance on profitability as well as support planning and monitoring of maintenance activities.

Recent empirical studies reveal that predictive maintenance is essential for accomplishing business objectives of manufacturing enterprises. Knowledge-based maintenance strategies for optimal operation of industrial machines and physical assets reasonably require explaining and predicting long term economic impacts, based on exploring historical data. This paper examines how supervised machine learning (ML) techniques may enhance anticipating the economic impact of maintenance on profitability (IMP). Planning and monitoring of maintenance activities supported by various statistical learning and supervised ML algorithms have been investigated in the literature of production management. However, data-driven prediction of IMP has not been largely addressed. A novel data-driven framework is proposed comprising cause-and-effect dependencies between maintenance and profitability, which constructs a set of appropriate features as independent variables.

Knowledge Management in the era of Industry 4.0 (KM 4.0) in both human- and technology-oriented perspectives is a strategic and operational function comprising exploration and exploitation processes. It is responsible to accomplish two major tasks. First, KM 4.0 should continuously support value generation through enhancing and balancing need- or opportunity-driven knowledge generation and knowledge utilization capacities. Second, KM 4.0 should persistently facilitate developing and protecting human-machine collective intelligence across manufacturing enterprises and in particular smart factories. Hence, KM 4.0 is an enabler to maximize competitive advantages and derive business values in the manufacturing enterprises. The revival of AI and emergence of autonomous and learnable technologies challenge the unique role of human as a knowledge actor, decision-maker, problem-solver and learner. What are the considerations on rethinking KM approaches in relation to the march of technological enhancements? This paper proposes a definition and discusses the theoretical foundation of KM 4.0 as well as related practical aspects that should be taken into consideration, especially in dynamic, data-driven and hybrid human-machine working environments in smart factories.

The digital transformation of manufacturing industries currently re-invents conventional production paradigms through the creation of cyber-physical production systems (CPPS) and smart manufacturing networks. Novel knowledge-based maintenance (KBM) strategies and models are regarded as a key enabler in order to manage the increasing complexity and automatization of CPPS. Thus, securing and improving machine availability and process stability are accomplished. Although industrial decision makers are willing to invest in renovating and enhancing their companies’ maintenance strategy, they lack knowledge regarding their readiness levels in realizing and deploying KBM. In particular, they doubt whether their companies hold fundamental competence and capacity (i.e. appropriate methods for data analytics and systematic guidance) towards realizing KBM. In this paper, the authors present a holistic procedure model that assesses a company´s individual status quo in KBM and enables the identification of strengths and weaknesses on operative, tactical and strategic level following a multidimensional analytical approach. The model thereby builds on the assessment of more than 35 quality indicators assigned to a maintenance execution and data management dimension. The indicators feed into a mathematical calculation of data, information, knowledge and maintenance quality factors. The authors present the model’s development and content. Applying the model to an Austrian manufacturing company, its KBM readiness and maturity level for the creation of a predictive maintenance strategy under the premise of supporting prompt and efficient decision making is assessed. Feedback and validation of the results reveals both scientifically valuable, systematic and transparent applicability in real production environments.

Autonomous production control (APC) is able to deal with challenges, inter alia, high delivery accuracy, shorter planning horizons, increasing product and process complexity, and frequent changes. However, several state-of-the-art approaches do not consider maintenance factors contributing to operational and tactical decisions in production planning and control. The incomprehensiveness of the decision models and related decision support tools cause inefficiency in production planning and thus lead to a low acceptance in the manufacturing enterprises. To overcome this challenge, this paper presents a conceptual cost-based model for integrating different maintenance strategies in autonomous production control. The model provides relevant decision aspects and a cost function for different maintenance strategies using on a market-based approach. The present work thus makes a positive contribution to cope with the high demands on flexibility and response times in planning while at the same time ensuring high plant productivity.

Recent advances in autonomous production control (APC) enhance to cope with growing challenges of manufacturing industry regarding high delivery accuracy and short delivery dates, i.e. concurrently dealing with shorter planning horizons, increasing planning complexity and an increasing rate of changes in planning. However, the majority of APC approaches are based on idealized assumptions, lack integrative planning, and control mechanisms. In particular, the state-of-the-art APC approaches mainly ignore maintenance planning strategies, related decision factors, and maintenance measures. The ultimate practical implications are a reduced decision quality, and a resulting low acceptance of APC systems. In most cases, therefore, implementation of APC systems in industrial applications are not yet affordable. Pursuing this line of research, this paper presents a conceptual model for integrating maintenance strategies in APC. The focus is on evaluation of economic plausibility of the integrated approach, which deploys a cost function for a market-oriented decision-making. Using an agent-based simulation environment, the model is evaluated in an industrial use case in the automotive sector.

Successful implementation of smart manufacturing systems requires tight connectivity and intensive information and knowledge exchange. There is still a need for achieving certain level of smartness and beyond (i.e. cognitive manufacturing) by identifying semantic relations between heterogeneous data sources, extract actionable information and subsequently automate the process of knowledge generation and target-oriented recommendation in industrial context. The missing linkage is the quality of knowledge generation and protection process, which requires further investigation along with standard planning and controlling measures. This can be achieved through application of ontologies as a mean to support interoperability, reasoning and decision making. Considering various alternative approaches, this paper describes a research framework for ongoing study on ontology-based planning and controlling in smart manufacturing systems..

The high equipment intensity and complexity of production processes in semiconductor manufacturing leads to challenging requirements regarding plant availability in this competitive market. In the present paper, we address these challenges by proposing a conceptual model to enable prescriptive maintenance in semiconductor manufacturing. Different Machine Learning Algorithms are used to predict time-to-failure intervals for unplanned downtimes. Furthermore, the concept uses Bayesian Networks to predict the root cause of a failure and ultimately leads to recommendations, which are integrated into maintenance planning routines, in order to increase the system availability by initiating specific maintenance measures. Finally, the benefit of prescriptive maintenance is demonstrated in an industrial use case for etching equipment in semiconductor manufacturing.

Original equipment manufacturers (OEMs), machine operators and suppliers are three main stakeholders in today’s industrial asset management lifecycle. Considering high rate of investment and prices to purchase and operationalize secured IoT-based production systems, insurance companies explore contributing into industrial asset management lifecycle by providing new products to transparently monitor online condition and operation status of industrial machines, and accordingly adjust the insurance prices and offers. Hence, joint efforts and cross-disciplinary research is vital to introduce a new insurance product, i.e. failure-based insurance, gaining benefits from AI technologies and knowledge-based maintenance approaches. This paper provides insights into the aforementioned problem, and presents the work-in-progress research on failure-based insurance model aided by Digital Twin technology. Finally, the practical implications and future research directions are outlined.

The digital transformation already has a strong impact on manufacturing techniques and processes and requires novel data-driven maintenance strategies and models, which support prompt and effective decision-making. This poses new requirements, challenges and opportunities for securing and improving machine availability and process stability. This paper builds on the concept of prescriptive maintenance and proposes a reference model that (i) supports the implementation of a prescriptive maintenance strategy and the assessment of its maturity level, (ii) facilitates the integration of data-science methods for predicting future events, and (iii) identifies action fields to reach an enhanced target maturity state and thus higher prediction accuracy.

Purpose The purpose of this paper is to propose a framework of an automatic bidirectional matching system that measures the degree of semantic similarity of job-seeker qualifications and skills, against the vacancy provided by employers or job-agents. Design/methodology/approach The paper presents a framework of bidirectional jobseeker-to-vacancy matching system. Using occupational data from various sources such as the WageIndicator web survey, International Standard Classification of Occupations, European Skills, Competences, Qualifications, and Occupations as well as vacancy data from various open access internet sources and job seekers information from social networking sites, the authors apply machine learning techniques for bidirectional matching of job vacancies and occupational standards to enhance the contents of job vacancies and job seekers profiles. The authors also apply bidirectional matching of job seeker profiles and vacancies, i.e., semantic matching vacancies to job seekers and vice versa in the individual level. Moreover, data from occupational standards and social networks were utilized to enhance the relevance (i.e. degree of similarity) of job vacancies and job seekers, respectively. Findings The paper provides empirical insights of increase in job vacancy advertisements on the selected jobs – Internet of Things – with respect to other job vacancies, and identifies the evolution of job profiles and its effect on job vacancies announcements in the era of Industry 4.0. In addition, the paper shows the gap between job seeker interests and available jobs in the selected job area. Research limitations/implications Due to limited data about jobseekers, the research results may not guarantee high quality of recommendation and maturity of matching results. Therefore, further research is required to test if the proposed system works for other domains as well as more diverse data sets. Originality/value The paper demonstrates how online jobseeker-to-vacancy matching can be improved by use of semantic technology and the integration of occupational standards, web survey data, and social networking data into user profile collection and matching.

The accurate prediction of manufacturing lead times (LT) significantly influences the quality and efficiency of production planning and scheduling (PPS). Traditional planning and control methods mostly calculate average lead times, derived from historical data. This often results in the deficiency of PPS, as production planners cannot consider the variability of LT, affected by multiple criteria in today’s complex manufacturing environment. In case of semiconductor manufacturing, sophisticated LT prediction methods are needed, due to complex operations, mass production, multiple routings and demands to high process resource efficiency. To overcome these challenges, supervised machine learning (ML) approaches can be employed for LT prediction, relying on historical production data obtained from manufacturing execution systems (MES). The paper examines the use of state-of-the-art regression algorithms and their effect on increasing accuracy of LT prediction. Through a real industrial case study, a multi-criteria comparison of the methods is provided, and conclusions are drawn about the selection of features and applicability of the methods in the semiconductor industry.

Autonomous production control models facilitate coping with co-occurrent and ever-growing production system complexity. The choice of the appropriate control model and objective function in manufacturing industry depends on several interrelated factors such as machine and process flexibility, plant availability, process stability, resource utilization, environmental and economic sustainability. Numerous theoretical models and approaches for autonomous production control have been proposed. Yet, it is hard to determine the effectiveness of efforts in real-life applications, due to neglecting or incomplete consideration of the aforementioned factors. This paper concentrates on the correlative relation between production planning and control (PPC) and maintenance planning. An integrated approach is presented that establishes the link between autonomous production control and prescriptive maintenance model (PriMa). Traditionally, PPC and maintenance planning are carried out apart from one another. In the light of PriMa, various aspects of data-driven maintenance planning and the missing link with autonomous production control are discussed. To this end, this paper presents a conceptual model, which encompasses three sub-systems, namely i) maintenance system, ii) autonomous production control system, and iii) production planning. The proposed model is intended to reduce planning complexity and at the same time achieve an improved plant availability and resource efficiency.

Cyber-physical production systems (CPPS), as an emerging Industry 4.0's technology, trigger a paradigm shift from descriptive to prescriptive maintenance. In particular, maintenance management approaches nowadays are more and more transformed to (semi-) automated knowledge-based decision support systems. This paper is intended to examine existing approaches and challenges towards rethinking maintenance in the context of Industry 4.0 and thus contributes to the literature of production management and planning, by introducing a novel prescriptive maintenance model (PriMa). PriMa is comprising of four layers (i.e. data management, predictive data analytic toolbox, recommender and decision support dashboard as well as an overarching layer for semantic-based learning and reasoning). The integrated approach of PriMa enhances two functional capabilities, namely i) efficiently processing large amount of multi-modal and heterogeneous data collected from multidimensional data sources and ii) effectively generating decision support measures and recommendations for improving and optimising forthcoming maintenance plans correlated with production planning and control (PPC) systems. An industry-oriented proof-of-concept study has been conducted to explore the feasibility of applying PriMa in real production systems by implementing a decision support solution and achieving a significant reduction of downtime. Finally, future research directions in this area are outlined.

The essential and significant components of one’s job performance, such as facts, principles, and concepts are considered as job knowledge. This paper provides a framework for forging links between the knowledge, skills, and abilities taught in vocational education and training (VET) and competence prerequisites of jobs. Specifically, the study is aimed at creating an ontology for the semantic representation of that which is taught in the VET, that which is required on the job, and how the two are related. In particular, the creation of a job knowledge (Job-Know) ontology, which represents task and knowledge domains, and the relation between these two domains is discussed. Deploying the Job-Know ontology facilitates bridging job and knowledge elements collected from various sources (such as job descriptions), the identification of knowledge shortages and the determination of mismatches between the task and the knowledge domains that, in a broader perspective, facilitate the bridging requirements of labor market and education systems.

Improving maintenance knowledge intelligence using text data has not been largely explored in the literature of production and engineering management. The state-of-the-art approaches and solutions mainly focus on either clustering and classification of maintenance logs, or extracting additional (meta-)data e.g. failure time data from maintenance text reports, operators’ workbooks and digital logbook. Knowledge Discovery from Text (KDT) enables finding undetected causalities, hidden patterns, frequencies, associative relations, and sentiments in maintenance text repositories. Applying KDT may enhance understanding the content of text data syntactically and semantically. However, advanced KDT approaches do not significantly provide meaningful and explainable outcomes, due to certain barriers in manufacturing enterprises, namely availability and quality of (longitudinal) maintenance text data. To overcome these barriers in real world industrial maintenance, generate added value in industrial maintenance, and lay the ground for autonomous maintenance decision-support in the context of Industry 4.0, the first step is to adopt KDT methods and accordingly provide maintenance-specific solutions considering practical challenges and possibilities. This paper discusses the lack of understanding maintenance text data and examines its effect on maintenance knowledge intelligence in manufacturing enterprises. A compositional framework for text understanding (TextPlan) is introduced. TextPlan explores quantification of text data in both syntax and semantic levels, i.e. how to vectorize an annotated maintenance report into numeric values, which represent cost data, hidden associations and sentiments. A prominent feature of TextPlan is cost-based text analysis, which decomposes a maintenance text report into separate cost items, and then (re-)composes the findings to estimate the total maintenance cost associated with the given report. Finally yet importantly, TextPlan consolidates the findings into a Text Understanding Map for assisting maintenance planner, based on three proposed measures of text comprehension, namely Association Measuring Index (AMI), Opinion Index (OI) and Cost Vector (CV).

The paradigm shift in production system known as Industry 4.0 imposes changes on work division between human and machine. A human labor on the one side is assisted by smart devices and machines (human-machine cooperation) and on the other should interact and exchange information with intelligent machines (human-machine collaboration). This paper addresses the challenges of mutual human-machine learning in factories of the future. The ultimate goal is to identify new learning patterns in highly digitalized industrial work scenarios. To this end, we give a definition of mutual human-machine learning in digitalized work scenarios; provide exemplary scenarios in the TU Wien Pilot Factory Industry 4.0, and finally identify future research potentials.

This paper presents the development and realization of the “Integrated system for analyzing Maintenance records in Product improvement” (IMaPro). IMaPro has been designed based on research and need analysis through literature survey and consultation with industry partners in Germany. The primary goal is to analyze structured feedback data, such as condition monitoring, service, and customer data, especially to discover improvement potentials in maintenance management and ultimately product improvement. In this context, Bayesian network utilizes the knowledge-based analysis of feedback data. The Bayesian network is used in combination of a mathematical cost model. The model supports cost-based monitoring and controlling of maintenance activities, and consequently leads to the identification of the lack and applying the lessons learned for improving maintenance costing. The concept has been validated and pilot-tested through the use of a sample product, namely block heater system. Furthermore, IMaPro incorporates an in-house development of mobile application for acquisition of service data.

A new method for detecting multitopic structures in text documents, called Associative Gravity, is based on a text-mining method entitled CIMAWA, which imitates the human ability of word association. Specifically, Associative Gravity utilizes word association to detect different topics in a text. The authors named it Associative Gravity because of its resemblance to the physical law of gravitation, that is, mass and attraction. The mass corresponds to the importance of words in a text and the attraction to the asymmetrical associative word space. The innovative characteristic of the described topic detection method is supplied with asymmetrical associative word space provided by CIMAWA. A comparative case study proves the capability of Associative Gravity to separate different topics at very high accuracy.

Capturing data in web-based system can be done using open-ended and/or close-ended user interface elements. In open-ended format, users can enter anything, whereas in close-ended formats users can choose from predefined options to enter data. While allowing users to freely give response without limiting options, open-ended user interface components pose high pressure on the respondent-side and have higher processing costs. On the other hand, close-ended data capturing components pose pressure on the part of designers of user interfaces to predict the respondents' choices and still limit the options on the user side while they store data that are less challenging to process. This study sets out to contribute to the current research by designing and implementing a context-aware dynamic text field for data collection in web-based occupational information systems.

Digital transformation and evolution of integrated computational and visualisation technologies lead to new opportunities for reinforcing knowledge-based maintenance through collection, processing and provision of actionable information and recommendations for maintenance operators. Providing actionable information regarding both corrective and preventive maintenance activities at the right time may lead to reduce human failure and improve overall efficiency within maintenance processes. Selecting appropriate digital assistance systems (DAS), however, highly depends on hardware and IT infrastructure, software and interfaces as well as information provision methods such as visualization. The selection procedures can be challenging due to the wide range of services and products available on the market. In particular, underlying machine learning algorithms deployed by each product could provide certain level of intelligence and ultimately could transform diagnostic maintenance capabilities into predictive and prescriptive maintenance. This paper proposes a process-based model to facilitate the selection of suitable DAS for supporting maintenance operations in manufacturing industries. This solution is employed for a structured requirement elicitation from various application domains and ultimately mapping the requirements to existing digital assistance solutions. Using the proposed approach, a (combination of) digital assistance system is selected and linked to maintenance activities. For this purpose, we gain benefit from an in-house process modeling tool utilized for identifying and relating sequence of maintenance activities. Finally, we collect feedback through employing the selected digital assistance system to improve the quality of recommendations and to identify the strengths and weaknesses of each system in association to practical usecases from TU Wien Pilot-Factory Industry 4.0.