Starring in Part 13 of our Behind the Scenes segment is the Univesity of Thessaly (UTH). We had a conversation with UTH’s Principal Investigator for OPTIMAI, Professor Elpiniki Papageorgiou. In her interview, Elpiniki details the work and research her organisation undertakes in relation to Artifical Intelligence (AI), Digital Twins, deep learning, and more. We also asked her what advice she would give to anyone interested in working with Horizon Europe – keep reading to find out what she said!
Hi Elpiniki, thanks for speaking to us today. Could you start by introducing yourself and telling us where you are based?
My name is Elpiniki Papageorgiou. I am a Professor in Computer Science at the University of Thessaly (UTH), and Principal Investigator in OPTIMAI for UTH. The University of Thessaly is a vibrant institution with an outstanding track record in EU and national research and educational programmes. My research team belongs to the Department of Energy Systems, Geopolis Campus in Larissa, Greece, and is composed of Dr. Theodosis Theodosiou, Asst. Prof., Dr. Konstantinos Papageorgiou, Post-Doc Researcher, and a couple of PhD students.
What does your typical working day in the OPTIMAI project involve?
First, I need to highlight that OPTMAI is a demanding research project with strict deadlines. The UTH is involved in several technical research activities related to the Design and Development of AI modules for quality control toward zero-defect manufacturing, and the research and design of models for digital twins of individual production phases and manufacturing outcomes. So, the typical day usually begins with checking our progress and compliance with the OPTIMAI schedule. We try to keep our tasks on track with short daily discussions and weekly meetings. During these meetings, we discuss our achievements so far and plan our next steps.
What is your main task in the OPTIMAI project?
UTH is involved in most tasks of OPTIMAI, but our focus is on two key aspects. First, we were responsible for reviewing the state-of-art technologies in Artificial Intelligence that can be incorporated into modern manufacturing. This was a great challenge for us since there has been great progress in this domain during the last decade. Our focus was aligned with the OPTIMAI objectives, so we reviewed the relevant research and applications in the areas of Artificial Intelligence for Metrology, Digital Twins, Internet of Things (IoT) sensors, Computer Vision, Augmented Reality, Quality Control and Zero-defect Manufacturing.
Further to this, our team will significantly contribute to the development of deep generative models and fuzzy cognitive maps in order to model production equipment and sensor behavior, virtualize concrete production steps and simulate production outcomes in consecutive manufacturing stages. Moreover, hybrid learning methods that combine deep learning methodologies and fuzzy cognitive maps will be explored to deal with data uncertainty and enhance the decision framework’s interpretability by providing cause-effect relations between production stages
Now, we are working on the development of a component for the identification of defective products using artificial intelligence. Our tool will gather data from sensors strategically placed along the production line; these will include images, 3D scans, and time-series. Then, it will assess the condition of the product and, in the case of a defective item, it will trigger an alarm so that the operator can take immediate action and restore production or initiate maintenance procedures.
What do you like most about your role?
My role in OPTIMAI offers me the freedom to be innovative, facing all the new and exciting industry trends and challenges in the area of Artificial Intelligence in Zero-Defect Manufacturing. I am also given the chance to contribute my experience and skills in the Smart Manufacturing arena. I share the same vision with distinguished colleagues in the field, and benefit from the exchange of new ideas, as well as from the genuine spirit of cooperation and shared goals with everyone involved in the project.
Moreover, I enjoy building and coaching my team, producing fascinating and compelling experiences, and I love the feeling of establishing professional connections with staff, colleagues, and other participating members of OPTIMAI.
Has working remotely affected your work on OPTIMAI in any way?
Not very much. We were lucky that the COVID-19 pandemic struck during the initial stages of the project, when most operations were conducted at a theoretical level – for example, scheduling and planning. During that period, all information could be exchanged remotely and frequent teleconferences were planned, providing a fertile ground for discussions. We really took advantage of all the technologies we could use in order to keep the project running.
In the current phase, the OPTIMAI schedule involves active participation and demands being hands-on. We are lucky to see the European restrictions on COVID-19 being loosened up (at least for the vaccinated people) and thus, we hope to proceed without any issues.
What makes your organisation ideal for participating in the research/activities of OPTIMAI?
The contribution of UTH, context-wise, can be considered as critical in the research/activities of OPTIMAI, since the UTH team has extensive experience in AI algorithms and tools, intelligent decision support systems as well as in Learning Analytics and Big Data, which constitute significant components in all stages of the project.
The research group behind UTH can offer its valuable expertise in working with dynamical models for decision support systems based on fuzzy cognitive maps, artificial intelligence methods and tools, machine learning, deep learning and data mining, in order to develop certain software components for the identification of defective products using artificial intelligence. Additionally, our organization exhibits long, solid experience in the AI framework in several European projects so far, making it an ideal partner who can help OPTIMAI successfully deliver its outcomes.
What does OPTIMAI have in common with other EU-funded research projects?
OPTIMAI is the successor of a long series of successful EU-funded projects. During our literature survey, we were glad to see that a lot of research has already been done. In fact, we identified at least 28 EU-funded projects relevant to digitization in industry, digital twins, computer vision, quality control and zero-defect manufacturing. All these projects tried to address various issues that we run into in OPTIMAI, so, we didn’t have to start from scratch. On the contrary, we took some successful technologies and attempted to take them a few steps further.
Could you describe the overall expected impact of the OPTIMAI project in three words?
Ambition – Hard-work – Innovation
What would be your advice to anyone interested in getting involved with a Horizon Europe project?
As I said at the beginning, OPTIMAI is a demanding project with strict deadlines, pretty much like all other Horizon projects. Horizon projects offer great opportunities to develop and establish new innovative technologies and ecosystems, by involving research centres and companies with niche expertise. So, anyone who wants to join the Horizon Europe effort must be focused and determined to face all challenges. The results of the project are only the tip of the iceberg.
Thanks for taking the time to speak with us, Elpiniki! We look forward to more updates from the team at the UTH.