What is your personal background?
I am deputy head of the Intelligent Data Analytics Group in the Department of Data Science and AI at Fraunhofer FIT. I am also post-doctoral researcher in the Chair of Information Systems and Databases at RWTH-Aachen University, working on the ChampI4.0ns project – a project that explores AI solutions for a sustainable wood industry. I received my PhD in Computer Science from University of Siegen, where I was a Marie-Curie-ITN Fellow working on feature learning from text for data-intensive personalized job recommendation system. Previously, I conducted research on statistical machine translation at the German Research Center for Artificial Intelligence (DFKI). My research interests include data mining, machine learning, information retrieval, and the application of AI in various problem domains.
What is your organization’s role in knowlEdge?
Fraunhofer FIT is one of the key technical partners working on cross-cutting tasks that offer infrastructure for the project, definition and elicitation of user needs and scenario analysis as well as AI model development. More specifically, Fraunhofer FIT leads WP6 where it also leads tasks contributing to the identification of the security requirements for the project and the design and implementation of the tooling to support software provisioning activities on edge devices. In addition, it leads tasks for the development of edge data management services, AI model generation through automating the process of data analysis and annotation as well as the inclusion of human feedback into the automated AI pipeline.
What fascinates you about Artificial Intelligence for manufacturing?
In my view, AI in manufacturing offers several fascinating aspects and benefits that make it a compelling area of innovation and research. For example, it boosts productivity and efficiency; it offers insights that helps make decision for predictive and preventive maintenance and thereby helps save costs through optimizing operations and reducing waste.
It also helps in the development of customizable and personalized processes and systems. In its recent development, it also enables human-machine collaboration, where the integration of AI and robotics in manufacturing facilitates efficient and safe collaboration between humans and machines in order to harness the strengths of both human and machines. Furthermore, it enhances safety by relieving humans of tedious and dangerous tasks in manufacturing.
What are your expectations in knowlEdge?
My expectation in knowlEdge is that it offers us threefold outcomes. The first is, of course, the successful implementation and integration of all the envisioned knowledge components to form the full-fledged knowlEdge platform – a platform that enables end users to assess whether an AI model is giving the expected output for a given input data as well as data scientists to perform data exploration and model generation automatically. My second expectation is broader scientific impact of the knowlEdge outputs through academic publications. The third is the knowlEdge community developed through the course of the project implementation, i.e., partners and other contacts established during project events and scientific conferences where knowlEdge contributes to.
Which target groups can benefit from knowlEdge?
A wide range of target groups can benefit from the result of the knowlEdge project. I prefer to categorize the target groups into two based on whether the benefit is direct or indirect, namely end users and technical personnel. The technical user groups will also be classified into two: domain experts and data scientists.
Manufacturing companies as end users can enhance production efficiency, reduce operational costs, and improve product quality, leading to increased profitability. In addition, domain experts can benefit from knowlEdge for predictive maintenance, quality control and product development. knowlEdge offers the ability to analyse data and observe the behaviour of machines, a feature that helps maintenance team to take action that reduce downtime and extend the lifespan of equipment. Furthermore, knowlEdge offers demand forecasting and simulation that facilitates innovation for optimization processes and services.
Finally, the direct benefactors of the knowlEdge project are data scientists. knowlEdge provides the pipeline for data collection, data quality assurance, explorative data analysis, AI model generation and marketplace for data and model exchange. On top of all these, there is decision support framework as well as human-AI collaboration. Data scientists can leverage from this pipeline and speed up the process of model generation and adaptation.
What is your vision beyond knowlEdge?
My vision beyond knowlEdge is achieving the possibilities that could shape the future of the manufacturing industry by achieving human-AI collaboration for safer and more productive work environments, where humans and machines work together seamlessly, each contributing their unique strengths for flexible manufacturing operations in fully autonomous manufacturing plants.