Research Areas at TUKE

Supply chain management is an integrated tool that connects planning, data collection, and evaluation from current logistics processes into a cohesive whole. Currently, the supply chain process is confronted with challenges that hold immense potential for its development, aiming to increase efficiency and reliability while transforming traditional management models in relation to digital transformation. Therefore, the supply chain must be capable of implementing new digital technologies that will advance and support its digital transformation processes. 

Research focuses on big data and AI-driven causal research in areas such as expenditure management, supply management, procurement decision-making automation, negotiation processes, supply chain risk management, data-driven supplier management, category management, predictive sourcing, unethical practices in supply chains, supply chain performance optimization, AI-driven procurement innovation, and more.

The implementation of AI and IoT sensors aims to predict equipment failures before they occur, minimizing breakdowns that pose risks to workers and operations, as well as reducing production quality loss, downtime, and maintenance costs. Integration of advanced systems and sensors for object and image recognition, supported by AI, allows for real-time/online monitoring of product quality. This helps in defect identification and improving the consistency of manufacturing processes.

The integration of AI aims to advance smart and sustainable buildings and responsive architecture, including the development of new types of materials within the context of a circular economy and environmental decarbonization. Research currently focuses on decarbonization strategies for the construction sector, including the necessary investments for the mass availability and economic efficiency of proposed low-carbon materials and processes. The goal is to develop new types of structures and materials with low or zero greenhouse gas emissions and high circularity scores, leading to decarbonization and the creation of a carbon-neutral environment. Mitigating climate change requires a multi-level assessment of urban heat islands, climate change resilience, and perceived comfort.

The European Commission has proposed a comprehensive set of measures to ensure the EU's access to secure, diversified, affordable, and sustainable supplies of critical raw materials. These critical materials are essential for a wide range of strategic sectors, including industries working towards climate neutrality, digitalization, space exploration, and defense. Research into the technological parameters of processing ores containing critical metals is an important area focused on optimizing the extraction processes of critical metals from raw materials. These activities aim to improve the efficiency, yield, and environmental sustainability of these processes. The processes involved in critical metals extraction include: mining and optimization of yield and efficiency, hydrometallurgy—dissolving metallic components of ores in solution, pyrometallurgy—high-temperature melting and refining of metals, recycling, separation, and concentration—specifically froth flotation for the recovery of critical metals.

Research focused on new progressive materials, resulting in new possibilities for example for photonic technologies (including high-energy lasers), thermoelectric materials (radioisotope thermoelectric generators and thermoelectric materials for wearable electronics) with applications in semiconductor technologies (approved for use in space), clean technologies efficiently utilizing resources (advanced methods of using hydrogen as an alternative energy source in reduction processes in metal production), battery technologies (electrochemical Li-ion and Na-ion cells and high-density batteries), hydrogen technologies, carbon capture and storage technologies, nuclear technologies (materials for nuclear fission energy), energy efficiency technologies related to energy systems, transformative industrial technologies for decarbonization, manufacturing and recycling technologies, water purification and desalination technologies, circular economy technologies, and nanobiotechnology.

Advanced manufacturing technologies are transforming the manufacturing environment, enabling smarter, more efficient, and sustainable production methods. The integration of interconnected systems and devices across the production level allows for data exchange, collaboration, and real-time decision-making. Digital twins serve as a powerful tool in the field of smart manufacturing, facilitating the integration of various technologies and processes within the manufacturing environment. The integration of digital twin technology with VR/AR/MX reality creates an engaging platform for dynamic visualization, interactive training, and real-time collaboration between humans and machines.

Within electric drive mobility, activities will focus on the analysis and design of the required charging infrastructure, available and innovative battery technologies, and the impact of electromobility on the electrical grid. Strategies for integrating renewable energy sources with electric vehicle charging systems will be examined, along with the analysis of policies and standards related to electromobility. Practical applications will be developed for the design and evaluation of electric charging systems and their sources to achieve the set goals for decarbonizing the energy sector. 

In hydrogen drive mobility, activities will focus, among other things, on the analysis and design of unique hydrogen technologies; the design, production, and testing of new types of materials and their implementation into prototype hydrogen devices; the design, production, and testing of unique devices for hydrogen regulation, compression, and separation; and the implementation of innovative processes and methodologies that will contribute to increasing the safety of hydrogen device operations in practice.

Artificial intelligence in medical imaging has proven to be a transformative force, enhancing diagnostic accuracy and efficiency. A significant challenge in this field is the generalization and adaptation of AI models across various imaging datasets and clinical environments. The goal of domain generalization is to develop AI models that perform consistently under different imaging conditions, while domain adaptation focuses on fine-tuning models to specific datasets, thus improving their robustness and reliability. Prognostic AI-driven models further expand the impact of medical imaging by predicting disease progression and patient outcomes, integrating imaging data with clinical information for personalized medicine. As AI continues to evolve, its integration into medical imaging promises to revolutionize healthcare delivery, improving diagnostic accuracy and patient care.

The research aims to develop an advanced AI-based framework for threat detection and response tailored to smart environments, aligning with the strategic priorities of the European Union. Objectives such as developing AI-driven threat detection models, enhancing privacy and security protocols, automating incident response using machine learning, large language models, and deep learning techniques, aim to create models capable of identifying and predicting cyber threats in real time. These models will analyze data streams from devices to detect anomalies and potential security breaches. By addressing challenges in cybersecurity, we can provide innovative solutions for a safer and more resilient digital ecosystem.