Content - Session VI: AI Applications in Connected Industries and Productivity

AI Applications in Connected Industries and Poductivity

Session abstract

In this session, we present insights into AI-based applications in industrial use cases. In the light of Industry 4.0, these applications become more and more crucial to gain usage of the vast amounts of data produced by manufacturing and engineering facilities. In this context, many questions arise, for example: How can data be exchanged between partners in a secure and privacy preserving way?, How can we deal with situations, where the facilities - and the data they produce - are decentralised? or How can the data be utilised to improve productivity and efficiency?

Experts from NEC, Schaeffler and Safran Electronics & Defense will present their strategies and current applications which tackle the challenges in employing AI-based methods in an engineering context. The compilation of electronics and computer hardware manufacturers (NEC), automotive suppliers and mechanical engineering (Schaeffler) and aerospace and rocket engine manufacturers (Safran) in this session will provide a broad cross-industry overview of the current state of AI applications.

The session will be chaired by Prof. Junichi Tsuji, Director of the National Institute of Advanced Industrial Science and Technology AIST, Japan and Dr. Gunar Ernis, Head of Business Unit Industrial Analytics of the Fraunhofer-Institute for Intelligent Information and Analysis Systems IAIS, Germany. Both Institutes share the goals to integrate scientific and engineering knowledge to address socioeconomic needs. Through our research, we contribute to the sustainable development of an ecologically sound environment, and an economically successful and socially balanced world.

Points that can be addressed:

  • What will be discussed in the session? Why is it relevant? (political relevance, practical applications, timing)

  • Who are the experts (chairs/speakers – their technical background/what kind of institutions – policy advising/research/private sector – strategy development/practical applications)?

  • What is the added value of a trilateral discussion on that topic? How will the audience be involved/how can the audience engage?

  • What is the objective and expected output/outcome? What can session contributors and audience gain from engaging in the session?)

Challenges in Connected Factories – The Next Phase of the AI revolution
Chair: Prof. Dr. Junichi Tsujii

Autonomy is the hallmark of AI, but it also poses challenges in connected factories. In connected factories, separate organizations which have operated autonomously till now have to exchange not only their data but also their sovereign processes of decision making. How to share data, coordinate independent autonomous agents and, at the same time, to protect their sovereign rights on their data and their decisions will be major challenges in the next phase of the AI revolution. In this talk, I will give some of research activities at the AIRC (Artificial Intelligence Research Centre) to illustrate challenges related to connected factories.

Industry 4.0 in Practice - Federated Learning & Data Sovereignty
Chair: Dr. Gunar Ernis

Two main problems regarding AI applications which arise in industrial contexts are decentralized data acquisition and the exchange of data while maintaining sovereignty over one’s own data-sets. I will give a brief overview of the current state of research regarding federated learning approaches and present the International Data Spaces Association (IDS) which provides a reference architecture that forms the basis for data ecosystems and market places based on European values, i.e., data privacy and security, and offers equal opportunities through a federated design.

Four Waves of AI Business in Connected Industry: NEC the WISE & NEXT
Dr. Satoshi Morinaga

Up to now, we have been experiencing three waves of AI utilization in the real business world. The first one was the utilization of Recognition by AI, the second one was the utilization of Prediction by AI, and the third one is the utilization of Control by AI. NEC the WISE is a portfolio of these AI technologies developed by NEC for enriching human intellect and creativity. After a quick overview of “NEC the WISE”, I will introduce one of our machine learning engines, “Heterogeneous Mixture Learning”, which outputs a “White-Box” type predictor based on training data. In real business, interpretability often becomes crucial to satisfy customers, and this engine has been a main solution in our business. Also, I will mention the approaching fourth wave, Negotiation. Coordination among AI systems will be necessary in the near future, and automatic negotiation technology seems to be one of the keys to that.

A Self-Learning AI through an Ontology-Driven Architecture
Cedric Oette

Time plays a major role in industrial manufacturing, and machine tools are constantly being optimized to maximize value-adding activities. An intelligent manufacturing assistant for machine tools has been developed within Schaeffler's digitization department, which allows an individual OEE increase in machines. Through an ontology-driven architecture, a self-learning AI concept with high transferability could be implemented, which allows analysis of dynamically wearing components based on sensor data and to give recommendations for action at the machine.

Some of the Challenges of AI in an International High-Technology Group
Dr. Daniel Duclos

Machine learning and artificial intelligence technologies are delivering outstanding performances in many applications today. Their potential impacts on an international high-tech group are numerous. I will begin with a short overview of the main technical challenges facing Safran, including enhanced products, new engineering and design tools, Industry 4.0, predictive maintenance and new customer services. However, although they are promising, ML and AI methods are currently far from fulfilling the basic requirements – in terms of accuracy, performance, robustness and security – for them to be used in critical systems. I will therefore focus on the main priorities of our ML and AI research program (and international research collaborations).

Redefining Operator Work Environment with Socio-Technical AI Assistant
Dr. Fabian Schreiber

A socio-technical assistance system is currently being developed in the BMBF research group SozioTex at Institut für Textiltechnik of RWTH Aachen University in cooperation with industry partners. A special focus is on designing the working environment of machine operators with digital solutions. By connecting assistance systems with the shop-floor and collecting user data, conclusions can be drawn on what the participatory design of new AI systems could look like. The collected data is further used to visualize relationships in the worker’s social network within the company, to create a blended learning context and to increase the communication between workers and the product quality.

 

Footer