How can science and technology support the industry in the digital transformation process, contributing to the creation of more sustainable manufacturing systems and supply chains? A group of INESC TEC researchers participated in the 32nd edition of the FAIM2023 International Conference to present the work developed by the institution in flexible automation and intelligent manufacturing – focusing on Deep Reinforcement Learning, Augmented Reality and Digital Twin solutions.
Over the course of a week, FAIM 2023 – hosted by the Polytechnic Institute of Porto – brought together academia and industry representatives to discuss several themes: automation and robotics, manufacturing and optimisation processes, sustainability and green manufacturing, Artificial Intelligence in industry, IoT and Cloud Computing, and Industry 4.0 and 5.0.
Romão Santos, Filipa Ramalho, Ricardo Zimmermann and António Almeida, researchers at INESC TEC Centre for Enterprise Systems Engineering (CESE), attended the event and presented INESC TEC’s work in this domain, also showcasing the results of research initiatives and ongoing projects.
Romão Santos presented an approach based on Deep Reinforcement Learning (DRL) that integrates a Digital Twin architecture, providing a dynamic allocation of resources to tasks, considering a simulated training environment of a production system – in the context of multi-model assembly lines. This methodology has already been tested in industrial environments, with promising preliminary results.
Since Augmented Reality (AR) is one of the emerging technologies associated with the European Industry 5.0 initiative, contributing to increase human-machine interactions, promote resilience through decision-making and flexibility to address variability and unexpected events, Filipa Ramalho presented the results of a research work with major state-of-the-art results in terms of AR application in industrial processes.
António Almeida presented a paper that proposes a framework for the development of a Cognitive Digital Twin. This framework aims to enable modular and semantic interoperability of resources through the digital representation of machines and other types of equipment, including relevant information about static and dynamic operational data.
Ricardo Zimmermann presented the results of a study that analysed how the uncertainty stemming from supply and demand, and the level of supply chain response, influence business performance. The study considered data from almost 180 companies operating in Portugal, and the results showed that adjusting supply chain management strategies to the uncertainties of the companies’ environment is quite positive in terms of business performance.
“The participation in the FAIM conference allowed us not only to disseminate relevant scientific results – achieved by a multidisciplinary team exploring innovative approaches -, but also to learn more about other researchers and their scientific contributions”, explained Romão Santos. “The fact that this was the 32nd edition of this conference, with a very particular nature – based on flexible and intelligent manufacturing systems – was crucial to our participation, allowing us to advance our knowledge towards improving our future lines of research”, he mentioned.
Filipa Ramalho shared this opinion: “International conferences with a scientific community of excellence are a great opportunity; not only to share and present our work, but also to have privileged contact with the authors of other important works, fostering networking and discussion moments. This networking is vital to the evolution of our work, to establish partnerships and to collaborate with other institutions and researchers”.
Besides the scientific programme, which included a set of thematic sessions, the 2023 edition of FAIM also included industrial tours.
Photos: Instituto Superior de Engenharia do Porto (Flickr)
News, current topics, curiosities and so much more about INESC TEC and its community!
