First International Workshop on

Data Science for Industry 4.0

In conjunction with EDBT 2019

About the Workshop

Industrial enterprises currently address the challenge of transforming the ideas of the Internet of Things, Industry 4.0, Cyber-Physical Systems, and similar concepts into reality. A direct application of the IoT approach to the production chains in manufacturing companies is presently not feasible, as there are many more parameters, but much less available data compared to other big data application domains. Modern production is characterized by vast amounts of data. However, this data is neither easily accessible, interpretable, nor connected to gain knowledge. Digital twins are supposed to provide a digital representation of a production landscape, but the challenges in building, maintaining, optimizing, and evolving digital twins in inter-organizational production chains that cross several boundaries have not been addressed yet in a systematic manner.

In this context, also various data management challenges have to be addressed. Huge amounts of heterogeneous sensor data (numerical, audio, video, etc.) have to be processed in real-time in order to control the production machines. In addition, unstructured data from production reports or external sources have also to be integrated to analyze and optimize the production process. Well established mathematical models for production engineering have to be integrated with data-driven machine learning for cross-domain knowledge generation.

On the other hand, Industry 4.0 or the Industrial Internet of Things are the basis for new applications and business opportunities. By connecting physical objects, systems, machines, and applications, the data produced by these objects may become a valuable resource, i.e., a product in its own right. Thus, data management and analysis operations have to be linked questions about value creation within and across enterprises. These ideas raise new requirements in terms of trust, data security, and data sovereignty, which also have to be considered in data-oriented industrial applications.

The workshop aims at bringing together researchers from different domains and to discuss the challenges for data science in industrial settings. The workshop will provide a forum for the presentation of recent research results, work-in-progress reports, vision papers, and an attractive keynote speaker.

Back to Top


The workshop starts at 14:00h in room "2.1" with a keynote talk by Asterios Katsifodimos, and conclude with a joint poster session for all EDBT/ICDT workshops. See also the detailed program of the conference.
Keynote Talk: Asterios Katsifodimos (TU Delft):
Stream processing v3.0: from real-time analytics to scalable & consistent event-driven applications
14:45 Christian Beecks, Kjeld Willy Schmidt, Fabian Berns and Alexander Graß (University of Münster & Fraunhofer FIT, Germany):
Gaussian Processes for Anomaly Description in Production Environments (short paper)
15:05 Dominic Duxbury, Norman Paton and John Keane (University of Manchester, United Kingdom):
Trusted and Auditable Decision Aids (regular paper)
15:30Coffee Break
16:00 Yorick Spenrath and Marwan Hassani (Eindhoven University of Technology, The Netherlands):
Ensemble-Based Prediction of Business Processes Bottlenecks With Recurrent Concept Drifts (regular paper)
16:25 Hung Nghiep Tran (SOKENDAI, Japan) and Atsuhiro Takasu (National Institute of Informatics, Japan):
Analyzing Knowledge Graph Embedding Methods from a Multi-Embedding Interaction Perspective (regular paper)
16:50 Burkhard Hoppenstedt, Manfred Reichert, Klaus Kammerer (Ulm University, Germany), Myra Spiliopoulou (Otto-von-Guericke-University, Germany) and Rüdiger Pryss (Ulm University, Germany):
Towards a Hierarchical Approach for Outlier Detection in Industrial Production Settings (short paper)
17:10 Jakub Valcik and Wojciech Indyk (Konica Minolta Laboratory Europe, Czech Republic):
Generic Data Imputation and Feature Extraction for Signals from Multifunctional Printers (short paper)
17:30Short break to setup poster sesson
18:00Poster Session, followed by Welcome Reception
Back to Top

Keynote Talk

Asterios Katsifodimos

Stream processing v3.0: from real-time analytics to scalable & consistent event-driven applications

Abstract: In the last decade we are witnessing a widespread adoption of architectural styles such as microservices, for building event-driven software applications and deploying them in cloud infrastructures. Such services favor the separation of a database into independent silos of data, each of which is owned entirely by a single service. As a result, traditional OLTP systems no longer fit the architectural picture and developers often turn to ad-hoc solutions that rarely support strict transactional consistency. At the same time, we are witnessing the gradual maturation of distributed streaming dataflow systems. These systems have departed from the mere analysis of streaming windows and complex-event processing, employing sophisticated methods for managing state, keeping it consistent, and ensuring exactly-once processing guarantees in the presence of failures. In this talk I will first talk about the requirements of modern stateful software services in terms of consistency and scalability and I will identify how well existing solutions meet those requirements. I will then talk about how we can enable general event-driven applications and services to be developed on top of streaming dataflow systems.

Brief Bio: Asterios is an assistant professor at TU Delft. Before joining TU Delft, Asterios worked at the SAP Innovation Center in Berlin, designing and implementing scale-out data management architectures for SAP's Leonardo ML foundation. Asterios spent three years as a senior researcher with the database systems group in TU Berlin working on language models and systems for scalable data processing. He received his PhD from INRIA Saclay. Asterios is one of the receivers of the SIGMOD Research Highlights Award in 2016 and the Best Paper award at EDBT 2019.

Back to Top


The proceedings of the workshop are published online as volume 2322 of the CEUR Workshop Proceedings, ISSN 1613-0073, a well-known website for publishing workshop proceedings. It is indexed by the major publication portals, such as Citeseer, DBLP, and Google Scholar. The proceedings are part of the joint volume with other EDBT workshops. Furthermore, a special issue in an international journal related to the workshop theme is planned for which the authors of the best papers of the workshop will be invited to submit an extended version of their work. Back to Top

Important Dates

  • Submission (extended): January 8, 2019
  • Notification: January 21, 2019
  • Camera-Ready: January 29, 2019
  • Workshop: March 26, 2019
Back to Top

List of Topics

Data Stream Processing for Industrial Data

  • Adaptive systems
  • Data Stream Mining
  • Concept Drift Adaption
  • Machine Learning in Industrial Applications
  • IoT Analytics

Query Processing and Data Integration for Industrial Data

  • Integration of Sensor Data
  • Query Processing in Distributed Streaming Systems
  • Data Integration and Change Propagation

Distributed Architectures for Efficient Management of IoT Data

  • Edge & Fog Computing
  • Blockchain for IoT & Industry 4.0
  • New Hardware Architectures for Industrial Data Management
  • In-Network Data Processing and Analysis
  • Distributed Communication Networks and Data Analysis
  • Dependability
  • Quality of Service

Applications for Industry 4.0 and IoT

  • Data Management for Manucfacturing Engineering
  • Smart Homes, Smart Cities, Smart Facilities
  • Data Analytics in Industrial Internet of Things

Other Emerging Topics for Industrial Applications

  • Modeling & Reasoning for Industry 4.0, IoT, Digital Twins
  • Data Security and Data Sovereignty
  • Human-centered Interfaces
  • Semantic Web and IoT, Web of Things
  • Standardization in Industrial IoT Applications
Back to Top

Submission Guidelines

DSI4 welcomes the full paper submission of original and previously unpublished research.
All submissions will be peer-reviewed and, once accepted, they will be included in the workshop proceedings.
Back to Top


The workshop is supported by the following research projects:

Back to Top

Workshop Organizers

Publicity Chair

  • Chair Public Relations

    Rihan Hai

    RWTH Aachen University, Germany

Back to Top