AI in Manufacturing
Czech Technical University
The Industry 4.0 paradigm strongly influences industrial manufacturing area and highlights Artificial Intelligence as a driving force for increasing production efficiency and developing new business models. Intelligent networks of virtual twins of all physical units engaged in production enable smart system integration as well as permanent system optimization. The agent-based approaches influence both the system and the SW architectures (typically SOA) and lead to such novel concepts like production as a service, smart services of products and smart services provided to products during the whole life-cycle etc. AI has been deployed in many critical tasks like planning and scheduling, big data analytics, optimization by negotiations, learning from experience, predictive maintenance, and system resilience. Experimental testbeds do play the key role in developing new manufacturing solutions. System integration and human-machine interfaces are supported by virtual and augmented reality tools. Interesting industrial use cases will be presented and discussed.
Keywords. Artificial Intelligence, agent-based solutions, virtual twins, system integration, system optimization, production planning and scheduling, testbeds.
Scientific Director, Czech Institute of Informatics, Robotics, and Cybernetics (CIIRC) at the Czech Technical University, Prague, Czech Republic, since 2018.
Full Professor at CTU, appointed in 1990. Head of the Department of Cybernetics – EU Center of Research Excellence, Czech Technical University, 1999-2013
Founder and Director of the Rockwell Automation Research Center Prague, 1992-2009
Founder and Director of CIIRC in 2013-2018, 30 years experience in leading research activities in industrial automation with the focus to applied Artificial Intelligence
Lead of the team of the Czech Industry 4.0 Initiative. Lead of the EU Project “Research and Innovation Center for Advanced Industrial Production – RECAIP- 2019-24” (50 mil EUR).
Awarded by Austrian “Honorary Cross for Science and Art” (2003) and the “Medal of Merit of the Czech Republic” (2017).
Future Perspectives of Association Rule Mining Based on Partial Conditionalization
Tallinn University of Technology
In today's practice, major chunks of data analytics are still done rather interactively (OLAP, MOLAP, with tools such as Cognos or SAP-BW, and with related techniques such as conjoint analysis); i.e., they lack the exploitation of machine learning (AI) approaches. A middle position with respect to this gap between interactive and automatic data analytics is taken by association rule mining, which explains the current huge success of tools such as Rapidminer. Unfortunately, current association rule mining suffers two categories of shortcomings (both theoretically and practically, i.e., in the tool landscape). First, association rule mining works only for discrete-valued columns, i.e., numerical-valued columns cannot be handled, which is a painful shortcoming in practical scenarios. What is needed, is what we call a 'grand pivot report', i.e., a full combinatorial multi-factor analysis against common aggregate functions. Such grand reports needed to come with appropriate, generalized measures of interestingness. Second, and even more critical, today's analysts are not systematically supported in avoiding all kinds of data misinterpretations such as the Yule-Simpson effect. In this talk we discuss, how partial conditionalization can be exploited to generalize association rule mining from conditional probabilities to conditional expected values, including the necessary theoretical measures of interest that are (i) adequate to support multi-factor impact analysis and are (ii) explicitly robust against data misinterpretation. (The theoretical foundations of this endeavor are laid down in the book "Generalized Jeffrey Conditionalization" that can be downloaded for free from: https://www.ttu.ee/?id=170467 )
Dirk Draheim is full professor of information systems and head of the Information System Group at Tallinn University of Technology (Taltech). The Taltech IS Group is deeply involved into the design and implementation of the Estonian e-Governance ecosystem. Dirk holds a Diploma in computer science from Technische Universität Berlin, a PhD from Freie Universität Berlin and a habilitation from the University of Mannheim. Until 2006 he worked as a Researcher at Freie Universität Berlin. From 2006-2008 he was area manager for database systems at the Software Competence Center Hagenberg, Linz, as well as Adjunct Lecturer in information systems at the Johannes-Kepler-University Linz. From 2008-2016 he was head of the data center of the University of Innsbruck and, in parallel, Adjunct Reader at the Faculty of Information Systems of the University of Mannheim. Dirk is co-author of the Springer book "Form-Oriented Analysis" and author of the Springer books "Business Process Technology", "Semantics of the Probabilistic Typed Lambda Calculus" and "Generalized Jeffrey Conditionalization". His research interest is the design and implementation of large-scale information systems. Dirk Draheim is member of the ACM.
Process Technology as Key Driver for Smart Manufacturing
Faculty of Computer Science, University of Vienna
The keynote reports on the journey of process technology in smart manufacturing: from connecting and controlling machines in the EU project ADVENTURE, over developing a modular LEGO factory demonstrating Industry 4.0 concepts to centurio.work  – the manufacturing orchestration engine applied in real-world settings. On this journey we had to face several challenges that arise from the interplay of machines, data, systems, and humans. Examples comprise bridging the gap between software and the physical world, finding the right granularity for controlling the machines, connecting the machines along the process logic (horizontal integration), and integrating systems across the levels of the automation pyramid. Particularly horizontal and vertical integration bear several benefits, for example, the possibility to collect data not only in an isolated manner, e.g., per machine or per system, but in a connected way, following the process logic. This opens the door for advanced analysis of manufacturing data, employing a mix of techniques from cross-sectional and process mining .
Florian Pauker, Juergen Mangler, Stefanie Rinderle-Ma, Christoph Pollak, centu- rio.work - Modular Secure Manufacturing Orchestration, BPM Industrial Track, pp. 164–171 (2018) http://ceur-ws.org/Vol-2196/BPM_2018_paper_33.pdf
Matthias Ehrendorfer, Juergen-Albrecht Fassmann, Juergen Mangler, Stefanie Rinderle-Ma: Combining Conformance Checking and Classification of XES Log Data for the Manufacturing Domain. CoRR abs/1904.05883 (2019)
Stefanie Rinderle-Ma is a full professor at the Faculty of Computer Science, University of Vienna (Austria) where she leads the Research Group Workflow Systems and Technology (WST). Stefanie received her PhD and habilitation degree in Computer Science from Ulm University (Germany). Her research areas are distributed and flexible process technology, process and data science, as well as digitalized compliance management. Application areas comprise manufacturing and (health) care.
How do Linked Data, Open Data, and Knowledge Graphs interplay?
Institute for Information Business, WU Vienna
"A Knowledge Graphs is a graph of data with the intent to compose knowledge" ... this admittedly vague definition was the result of a recent Dagstuhl Seminar on the trending topic of Knowledge Graphs in industry and academia and its meaning for Knowledge Representation on the (Semantic) Web. In this talk I will tackle the question of what's actually new about Knowledge Graphs and how they can help us to get closer to a more natural understanding of data, particularly Open Data on the Web, from a personal perspective of +15 years of research on Semantic Web standards, Linked Data, and Open Data.
Axel Polleres heads the Institute of Information Business of Vienna University of Economics and Business (WU Wien) which he joined in Sept 2013 as a full professor in the area of "Data and Knowledge Engineering". Since January 2017 he is a member of the Complexity Science Hub Vienna Faculty. Between January and June 2018, he has been appointed as visiting professor at Stanford University under the Distinguished Visiting Austrian Chair Professors program hosted by The Europe Center in the Freeman Spogli Institute for International Studies at Stanford. He obtained his Ph.D. and habilitation from Vienna University of Technology and worked at University of Innsbruck, Austria, Universidad Rey Juan Carlos, Madrid, Spain, the Digital Enterprise Research Institute (DERI) at the National University of Ireland, Galway, and for Siemens AG's Corporate Technology Research division before joining WU Wien. Moreover, he actively contributed to international standardisation efforts within the World Wide Web Consortium (W3C) where he co-chaired the W3C SPARQL working group.