A Prototypical Dashboard for Knowledge-Based Expert Systems used for Real-Time Anomaly Handling in Smart Manufacturing

2023 | HICSS | Citations: 0

Authors: Stahmann, Philip

Abstract: The use of machine learning in digitized production increases potentials for pro ... Expand

Abstract: The use of machine learning in digitized production increases potentials for production automation. A milestone on the path to autonomous production is real-time anomaly detection. However, increasing complexity of production makes autonomous decisions difficult to understand for humans as central stakeholders. In this paper, we create a dashboard that incorporates elements from knowledge-based systems, requirements for real-time anomaly detection, and design guidelines for dashboards. Using design science research, the dashboard is designed, implemented and comprehensively evaluated with 98 participants. After the second design science iteration, the dashboard is approved in terms of usefulness and ease of use. Our research primarily contributes to practice, as our implementation constitutes a starting point for designing the interface between humans and autonomous production. We also contribute to academia as the dashboard is an instantiation in the research field of interface design for knowledge-based systems, which can be further developed in future research. Collapse

Topics: digital dashboard expert system production management industry 4.0 internet technology

Methods: survey anomaly detection qualitative interview survey design design guideline

It's Time to Smarten Up! - A Framework for Building Smart Service Systems

2023 | Americas Conference on Information Systems | Citations: 0

Authors: Wache, Hendrik

Abstract: Smart service systems are a network of people, technologies, as well as process ... Expand

Abstract: Smart service systems are a network of people, technologies, as well as processes and rely on learning, adaptivity, and decision-making to provide smart services to their users. The complexity of designing smart service systems results in a plethora of potential implementation obstacles, often leading to an overly technical focus and resulting in isolated solutions. There is, therefore, a need for a guideline for companies that helps them facilitate the development of complex smart service systems. To this end, we apply action design research used in a multi-year design study for digital twins in the manufacturing industry. The resulting four-quadrant framework (domain, system design, organization, and ecosystem) can help researchers and companies guide the development of smart service systems or evaluate past progress. The evaluation of the framework shows that there might be dominant quadrants and certain paths in the framework, which we would like to investigate in further research. KeywordsSmart service system, digital twin, industry 4.0, action design research. Collapse

Topics: production management digital twin business model systems development mobile system

Methods: action design research group interview personal interview design principle design science

A Taxonomy and Archetypes of Business Analytics in Smart Manufacturing

2023 | ACM SIGMIS Database | Citations: 0

Authors: Wanner, Jonas; Wissuchek, Christopher; Welsch, Giacomo; Janiesch, Christian

Abstract: Fueled by increasing data availability and the rise of technological advances fo ... Expand

Abstract: Fueled by increasing data availability and the rise of technological advances for data processing and communication, business analytics is a key driver for smart manufacturing. However, due to the multitude of different local advances as well as its multidisciplinary complexity, both researchers and practitioners struggle to keep track of the progress and acquire new knowledge within the field, as there is a lack of a holistic conceptualization. To address this issue, we performed an extensive structured literature review, yielding 904 relevant hits, to develop a quadripartite taxonomy as well as to derive archetypes of business analytics in smart manufacturing. The taxonomy comprises the following meta-characteristics: application domain, orientation as the objective of the analysis, data origins, and analysis techniques. Collectively, they comprise eight dimensions with a total of 52 distinct characteristics. Using a cluster analysis, we found six archetypes that represent a synthesis of existing knowledge on planning, maintenance (reactive, offline, and online predictive), monitoring, and quality management. A temporal analysis highlights the push beyond predictive approaches and confirms that deep learning already dominates novel applications. Our results constitute an entry point to the field but can also serve as a reference work and a guide with which to assess the adequacy of one's own instruments. Collapse

Topics: production management smart manufacturing analytical information system industry 4.0 database system

Methods: cluster analysis literature study survey literature sample machine learning

Theories: fuzzy set theory

Digital transformation solutions of entrepreneurial SMEs based on an information error-driven T-spherical fuzzy cloud algorithm

2023 | International Journal of Information Management | Citations: 0

Authors: Yang, Zaoli; Chang, Jinping; Huang, Lucheng; Mardani, Abbas

Abstract: The digital transformation of enterprises has become an inevitable development t ... Expand

Abstract: The digital transformation of enterprises has become an inevitable development trend and one of the key driving forces that promotes the sustainable development of enterprises. However, due to the many obstacles of financial burdens, technical thresholds, and talent shortages, digital transformation has become a challenging task for entrepreneurial Small and Medium-Sized Enterprises (SMEs). Additionally, many competitive digital trans­ formation solutions on the market cause confusion when enterprises must choose one. This study drew a new information error-driven T-spherical fuzzy cloud algorithm to evaluate digital transformation solutions of entrepreneurial SMEs and support its selection. First, an evaluation index system for the digital transformation solution of entrepreneurial SMEs was established from four aspects. Then, a new concept of a T-spherical fuzzy cloud was defined to represent the evaluation information of the indicators. Additionally, a T-spherical Fuzzy Cloud Weighted Heronian Mean (T-SFCWHM) operator was used to aggregate the evaluation information. Af­ terward, an evaluation and selection decision framework for the digital transformation solution of entrepre­ neurial SMEs based on the T-SFCWHM operator was developed. Further, a practical example was given to illustrate the effectiveness of the proposed method. Finally, a discussion of the findings in our study was con­ ducted, and the conclusions were summarized. Collapse

Topics: digital transformation cloud computing IT strategic planning decision making smart manufacturing

Methods: computational algorithm expert evaluation design science theoretical contribution synthesis

Theories: fuzzy set theory

Heterogeneous Energy-aware Load Balancing for Industry 4.0 and IoT Environments

2022 | ACM Transactions on Management Information Systems | Citations: 0

Authors: Ahmed, Usman; Lin, Jerry Chun-Wei; Srivastava, Gautam

Abstract: With the improvement of global infrastructure, Cyber-Physical Systems (CPS) have ... Expand

Abstract: With the improvement of global infrastructure, Cyber-Physical Systems (CPS) have become an important component of Industry 4.0. Both the application as well as the machine work together to improve the task of interdependencies. Machine learning methods in CPS require the monitoring of computational algorithms, including adopting optimizations, fine-tuning cyber systems, improving resource utilization, as well as reducing vulnerability and also computation time. By leveraging the tremendous parallelism provided by General-Purpose Graphics Processing Units (GPGPU) as well as OpenCL, it is possible to dramatically reduce the execution time of data-parallel programs. However, when running an application with tiny amounts of data on a GPU, GPU resources are wasted because the program may not be able to fully utilize the GPU cores. This is because there is no mechanism for kernels to share a GPU due to the lack of OS support for GPUs. Optimal device selection is required to reduce the high power of the GPU. In this paper, we propose an energy reduction method for heterogeneous clustering. This study focuses on load balancing; resource-aware processor selection based on machine learning is performed using code features. The proposed method identifies energy-efficient kernel candidates (from the employment pool). Then, it selects a pair of kernel candidates from all possibilities that lead to a reduction in both energy consumption as well as execution time. Experimental results show that the proposed kernel approach reduces execution time by 2.23 times compared to a baseline scheduling system. Experiments have also shown that the execution time is 1.2 times faster than state-of-the-art approaches. Collapse

Topics: production management computer hardware database system internet of things mobile system

Methods: computational algorithm machine learning experiment random forest classification gradient boosting

Institutionalizing Analytic Data Sharing in SME Ecosystems – A Role-Based Perspective

2022 | HICSS | Citations: 0

Authors: Baars, Henning; Weber, Patrick; Tank, Ann

Abstract: There is a variety of reasons that sharing data among Small and Medium-Sized Ent ... Expand

Abstract: There is a variety of reasons that sharing data among Small and Medium-Sized Enterprises (SMEs) carries business potential, particularly for analytical applications. But outside a few niche domains, the number of success stories for data sharing is rather modest. Based on a qualitative study and first experiences from a research project with pilot implementations, we argue that this is mainly due to a lack of an institutionalized governance structure: Founding a separate legal entity for data sharing and analysis can address core concerns regarding sharing valuable data assets. However, this requires a well-calibrated set of defined roles for the involved partners. Based on our results we propose a first concept on delineating and mapping out those roles. Collapse

Topics: digital ecosystem data governance value creation analytical information system data quality

Methods: qualitative interview design artifact design science simulation deep learning

Smart Factory – Requirements for Exchanging Machine Data

2020 | DESRIST | Citations: 0

Authors: Freichel, Chiara; Fuchs, Anna; Werner, Peter

Abstract: Trends and future concepts like Smart Factory contribute to the competitiveness ... Expand

Abstract: Trends and future concepts like Smart Factory contribute to the competitiveness of manufacturing companies by more flexible, automated and efficient processes. Smart Factories are defined by a complete integration of machines, processes and information flows based on advanced information and communication technology with a database containing information about capacities, delivery times or future demand. Essential for a complete integration is an automated machine data exchange. To achieve this goal and contribute to the competitiveness of producers, we seek to elaborate the technically necessary core components required for a Smart Factory. By using the Design Science Research approach, we aim to build a blueprint for implementing an integrated Smart Factory system by defining requirements for machine data exchange. Our results suggest that manufacturing environments must address various requirements, which can be grouped under the categories data acquisition, data storage and data transmission. A successful integration of these components provides the basis for Smart Factory scenarios. Collapse

Topics: production management smart factory computer data storage enterprise architecture enterprise resource planning

Methods: design science design process design requirement

Smart Waste Collection Processes A Case Study about Smart Device Implementation

2020 | HICSS | Citations: 3

Authors: Kamm, Michael Reiner; Gau, Michael; Schneider, Johannes

Abstract: For decades the core processes of collecting waste have been unchanged. Through ... Expand

Abstract: For decades the core processes of collecting waste have been unchanged. Through new IoT-technologies, advances in sensors, and data transfer technologies, data-driven smart waste collection processes will replace old inefficient collection processes. Causing a shift from fix collection intervals to collection on demand, supported by smart algorithms and innovative web-applications. However, implementing such ideas come along with some almost insurmountable challenges related to wireless data transfer, battery lifetime and IoT infrastructure. Therefore, the question arises of how to implement IoT solutions in such complex and challenging environments. In order to contribute to the existing research about smart cities and autonomous IoT devices, we implemented smart devices in glass containers, measuring filling level over several months. The research study’s outcomes are test results, data analysis and a prototype implementation for a reengineered waste collection process. Furthermore, we identified main challenges and key issues which obstruct the implementation and spread of such smart city applications. Collapse

Topics: smart city internet of things mobile system smart device database system

Methods: design artifact case study qualitative interview

Modeling IT Availability Risks in Smart Factories

2020 | Business & Information Systems Engineering | Citations: 0

Authors: Miehle, Daniel; Häckel, Björn; Pfosser, Stefan; Übelhör, Jochen

Abstract: In the course of the ongoing digitalization of production, production environmen ... Expand

Abstract: In the course of the ongoing digitalization of production, production environments have become increasingly intertwined with information and communication technology. As a consequence, physical production processes depend more and more on the availability of information networks. Threats such as attacks and errors can compromise the components of information networks. Due to the numerous interconnections, these threats can cause cascading failures and even cause entire smart factories to fail due to propagation effects. The resulting complex dependencies between physical production processes and information network components in smart factories complicate the detection and analysis of threats. Based on generalized stochastic Petri nets, the paper presents an approach that enables the modeling, simulation, and analysis of threats in information networks in the area of connected production environments. Different worst-case threat scenarios regarding their impact on the operational capability of a close-to-reality information network are investigated to demonstrate the feasibility and usability of the approach. Furthermore, expert interviews with an academic Petri net expert and two global leading companies from the automation and packaging industry complement the evaluation from a practical perspective. The results indicate that the developed artifact offers a promising approach to better analyze and understand availability risks, cascading failures, and propagation effects in information networks in connected production environments. Collapse

Topics: smart factory production management robotic on-premises software systems analysis

Methods: simulation design science qualitative interview Petri net design artifact

Text mining of industry 4.0 job advertisements

2020 | International Journal of Information Management | Citations: 0

Authors: Pejic-Bach, Mirjana; Bertoncel, Tine; Meško, Maja; Krstić, Živko

Abstract: Since changes in job characteristics in areas such as Industry 4.0 are rapid, fa ... Expand

Abstract: Since changes in job characteristics in areas such as Industry 4.0 are rapid, fast tool for analysis of job advertisements is needed. Current knowledge about competencies required in Industry 4.0 is scarce. The goal of this paper is to develop a profile of Industry 4.0 job advertisements, using text mining on publicly available job advertisements, which are often used as a channel for collecting relevant information about the required knowledge and skills in rapid-changing industries. We searched website, which publishes job advertisements, related to Industry 4.0, and performed text mining analysis on the data collected from those job advertisements. Analysis of the job advertisements revealed that most of them were for full time entry; associate and mid-senior level management positions and mainly came from the United States and Germany. Text mining analysis resulted in two groups of job profiles. The first group of job profiles was focused solely on the knowledge related to Industry 4.0: cyberphysical systems and the Internet of things for robotized production; and smart production design and production control. The second group of job profiles was focused on more general knowledge areas, which are adapted to Industry 4.0: supply change management, customer satisfaction, and enterprise software. Topic mining was conducted on the extracted phrases generating various multidisciplinary job profiles. Higher educational institutions, human resources professionals, as well as experts that are already employed or aspire to be employed in Industry 4.0 organizations, would benefit from the results of our analysis. Collapse

Topics: production management industry 4.0 LinkedIn supply chain management smart factory

Methods: natural language processing cluster analysis machine learning computational algorithm term frequency–inverse document frequency