News
The CORE news hub
Two weeks of Learning, Growth, and Innovation
22nd July 2025
Author: Markela Kerkezou
At Core, we pride ourselves on continuous growth and learning. That’s why we recently took a step back from our day-to-day tasks to dedicate time and energy on something we deeply value: deepening our knowledge and skills through a series of internal trainings! We paused our usual rhythm for some workshops, diving into LLMs, AI Agents, Multi-Agent Systems, and Proposal Writing.
Workshop 1 | LLM Fundamentals
We kicked off by diving into the foundations of LLMs: how they work, how they’re built, and how they can be applied across real-world tasks. The session covered everything from model architecture to prompt engineering, and Retrieval Augmented Generation (RAG), with hands-on examples using Ollama, and LM Studio. The goal? To build a solid understanding that empowers every team member to explore and experiment.
Workshop 2 | Proposal Writing Workshop
This 2-full-days Workshop (a full 16 hours of content!) took us deep into the anatomy of a successful Horizon Europe proposal.
Day 1 was all about building a strong theoretical foundation: We explored the power of storytelling, how to shape an idea into a concept, and how to select the right topic. We worked through practical exercises to understand what adds real value in a proposal (and what doesn’t), and we picked up tips and tricks for writing and structuring proposals effectively.
Day 2 was Game Day!
We played the Proposal Writing Game, a custom card-based game designed by CORE to simulate the proposal development process. Through the game, we practiced combining technologies, choosing strategic partners, and, of course, adapting to the budget. At the end, our internal evaluators stepped in to assess which proposals would get funded.
Workshop 3 | AI Agents Fundamentals
The next step took us into the evolving world of Agentic AI. We explored how autonomous agents can reason, use tools, and take actions to achieve goals, even in open-ended environments. During this Workshop, we explored AI Agents key terms, agent architecture components and building blocks, and Agentic AI stack and frameworks. The session helped us connect the dots between core concepts and real-world tools, like Google’s ADK, which we used in our hand-on examples.
Workshop 4 | Multi-Agent Systems
Our final workshop zoomed out to look at the big picture; how agents can work together. Things got more technical: we explored Multi-Agent Systems, Agent Communication Protocols (A2A), orchestration patterns (hierarchical vs cooperative) and use cases where distributed AI teams outperform single-agent approaches. The hands-on examples made the complexity of multi-agent systems much easier to grasp, and surprisingly fun to explore!
And we are not stopping here. More workshops are already in the works. At CORE continuous learning isn’t a one-off initiative; it’s a mindset.
We're taking the time to invest in our people, because this is how we grow stronger, together.
Middleware Development: Building a Unified Data Platform
Author: Maria Tassi, Nikos Gkevrekis
3rd July 2025
The CORE Innovation Centre technical team has released the alpha version of middleware platform that provides a unified solution for managing heterogeneous data flows across different data sources.
It is a flexible and scalable platform, which offers a robust foundation for seamless data ingestion, storage, processing, and secure access across diverse systems and demonstration sites. The development of this middleware platform is in line with CORE’s digital transformation mission, helping organisations accelerate their transition through cutting-edge research and technology development which addresses real-world barriers that hinder progress for many manufacturers, regardless their specific industry.
Architecture Overview
The middleware is structured around a layered architecture (see the figure below), which consists of four primary layers – Ingestion, Storage, Processing, and Consumption – all supported by Orchestration and Monitoring layers. These interconnected components ensure that the platform can handle a wide spectrum of data types while maintaining operational coherence and traceability.
Middleware architecture
Key features of the architecture
Multi-Source Data Ingestion: Designed to integrate heterogeneous data streams, the ingestion layer supports:
MQTT for real-time data
REST API for batched real time and historical data
File uploads (e.g. images, GIS) through a fileserver
It also supports ETL – Extract, Transform, Load – processes and performs data validation on entry to maintain data quality and consistency.
Versatile Storage: The storage layer is optimized for various data types:
large files
structured data
time-series data
Features like pagination and sorting enhance performance, especially for large-scale datasets.
Secure and Dynamic Data Access: The consumption layer exposes data via RESTful APIs, featuring:
Token-based authentication
Role-based authorisation
This way, users can query real time, historical records and batched data that we ingested from real time sources, specify time ranges, and retrieve files in original or compressed formats. The system also supports dynamic endpoints tailored to specific organizations or devices.
Interoperability and Integration: The platform is built to work across multiple sources and demonstration sites
Scalability and Extensibility: As an alpha release, the architecture anticipates future enhancements including real-time processing, advanced analytics modules, and tighter integration with external systems, supporting the evolving needs of diverse pilot sites.
Ingestion Layer
The development process began with the ingestion layer, which serves as the gateway for all incoming data. Designed with flexibility, this layer can receive data from real-time sources such as MQTT, scheduled or historical data via APIs, and large files like images and geospatial datasets through a fileserver. This fileserver was developed to support large document handling, enabling users to upload, download, and manage files in their original formats, in order to accommodate diverse data requirements from real-time data to large-scale datasets. In addition to managing data intake, the ingestion layer plays a key role in validating incoming information and preparing it for further use. It supports ETL operations, which ensure that data is harmonised, transformed when necessary, and made ready for further analysis or storage.
Storage Layer
In parallel with the ingestion layer, significant progress was made on the storage layer. This layer is designed to efficiently store the wide variety of data collected by the system. It integrates multiple storage technologies like S3 buckets, PostgreSQL, TimescaleDB etc. for general data storage, for handling files from the fileserver, and for managing time-series data ensuring optimal performance and scalability.
Consumption Layer
Development has also begun on the consumption layer, which is responsible for enabling secure access to the stored data. This layer currently provides REST API that are protected by token-based authentication and role-based authorization, ensuring that only authorized users can access sensitive information. Users can query bached real time data, historical data by requesting specific time ranges, and retrieve files either in their original form or in compressed formats and define pagination and sorting which enhance the speed and efficiency of data retrieval. Additionally, the consumption layer supports dynamic endpoint creation based on organizational structures or specific device IDs, allowing it to adapt easily to the varying needs of different demo sites and stakeholders.
Responding to real-life challenges
The system's complexity presented various challenges during development. Managing a variety of input formats, including real-time IoT data, historical API feeds, and large unstructured files, required the development of a flexible and adaptable ingestion system which can process heterogenous types of data. Ensuring data quality across many formats and sources necessitated the development of robust ETL methods as well as versatile and dynamic schema validations.
Another challenge was securyity, as designing a secure system with token-based authentication and role-based authorisation presented difficulties in multi-site, multi-user scenarios. To balance flexibility with system performance, especially for large-scale time-series data and file management, storage solutions have to be carefully selected and configured.
Furthermore, developing and maintaining dynamic endpoints able to consume data as they are being ingested required a careful and complicated database schema and management. Last but not least, deploying, managing and scaling numerous different data ingestion and consumption services requires the development and usage of complex custom orchestrating and monitoring tools.
Conclusions
The release of alpha version of the middleware platform marks a significant step toward a flexible and robust solution for managing heterogeneous data. With its layered architecture, it supports seamless data ingestion from real-time data flows, APIs, and large files, while ensuring efficient storage, validation, and secure access. Features such as ETL processing, dynamic endpoints, and multilevel authentication enable adaptability, interoperability, and data integrity across diverse sources.
The middleware platform has a substantial market impact, because it enables interoperable data exchange across several sources, boosting collaboration in various fields such as manufacturing, climate resilience, and industrial processes. This interoperability accelerates digital transformation by combining real-time, historical, and large-format data to create a secure, scalable infrastructure that improves decision-making and operational efficiency.
Designed to manage complex, multi-site systems, it provides dynamic endpoints and role-based access while establishing the groundwork for future features such as real-time analytics and AI integration.
Elements of a secure and interoperable middleware approach have been explored and developed within two of our Horizon EU projects; CARDIMED, which focuses on boosting Mediterranean climate resilience, and MASTERMINE, which focuses on building a digitalized copy of real-world mines through an Industrial Metaverse approach.
The newly-released middleware platform is aligned with our CORE mission of accelerating digital transformation through cutting-edge research and technology development, especially in data interoperability, artificial intelligence, and industrial digitization and demonstrates our dedication to developing smart, adaptive, and future-ready solutions that address real-world difficulties across industries.
The alpha version lays a strong foundation for future enhancements, including advanced analytics, real-time processing, and broader system integration, positioning the middleware as a key enabler in modern data ecosystems.
Data-Driven Digital Shadows for Process Manufacturing
Author: Iason Tzanetatos
26th June 2025
Circular TwAIn is a Horizon EU project which aims to research, develop, validate, and exploit a novel AI platform for circular manufacturing value chains. This AI platform can support the development of interoperable circular twins for end-to-end sustainability.
Within the scope of our activities in the Circular TwAIn project, CORE IC is responsible for the design, development and deployment of a data-driven Digital Shadow for an industrial process, such as that of a petrochemical plant (the Circular TwAIN end user, SOCAR).
In process manufacturing, particularly in complex and highly regulated industries like petrochemicals, the ability to monitor, analyse, and optimise operations in real time is essential for maintaining efficiency and competitiveness. The integration of advanced digital technologies is reshaping traditional operations, enhancing performance, and driving innovation. To be able to realise this model, we have utilised historical operational data from our partners at SOCAR, that they have kindly shared with the consortium.
Architectural Design of Process Digital Shadow
Working with our partners at TEKNOPAR and SOCAR, we defined their requirements from the technology and identified the following key objectives:
1. Perform anomaly detection on real-time sensorial data, that depict the current conditions of the plant
2. Predict what the sensor readings will be in a 10-minute horizon (i.e., predict short-term future state of the plant)
3. Perform anomaly detection on the forecasts of the plant.
We started the development of a Digital Shadow for a manufacturing process by identifying the involved assets of the production line. Since we are dealing with a process where the involved assets interact with one another, the outputs of some of the assets are considered as inputs to other assets down the line.
After the involved assets and their interactions had been mapped, we moved to the mapping of the sensorial inputs/outputs of each asset. Relevant information such as SCADA schemes were used to determine the sensors that are considered as inputs and outputs of each machine.
Data Augmentation Techniques – Asphalt Use case
To successfully identify any anomalous conditions for each sensorial signal, irrespective of the input/output characterization of the involved sensors, we monitored each signal individually.
We proceeded with training an Autoencoder-like Deep Learning model, to identify anomalous conditions on a per sensor level, meaning that the model examines the data point of each sensor individually.
Autoencoder Neural Network Architecture
As depicted in the figure above, an Autoencoder model comprises three main components:
1. The Encoder, where the input information is compressed
2. The Bottleneck layer, where a compressed low dimensional representation of the input is determined by the model
3. The Decoder, which reconstructs the input relying only on information retrieved by the Bottleneck layer
By training an Autoencoder model with high quality, normally characterised operational data, we essentially have a model that can identify significant deviations on the operations of the involved assets.
However, since this use-case is particularly complex, niche models such as Reservoir Computing deep learning models. This family of models is best suited for time series data with complex patterns, similarly to the operational data of a manufacturing process.
From: Quantum reservoir computing implementation on coherently coupled quantum oscillators
By adopting the approach of the Autoencoder model, we formulate the problem in the exact manner, only we swap models and use a Reservoir Computing model.
Forecasting
To proceed with forecasting 10-minutes ahead of the manufacturing process, our model needs to follow the same interactions of the assets as in the actual plant.
Production flow on Petrochemical line
The process comprises three main components, a Reactor, an Absorber and a Stripper. Each asset interacts back and forth with each other, and these interactions must be replicated in the digital realm as well.
A Digital Shadow has been developed for each physical asset. We utilised the same family of Deep Learning models to perform forecasting, by switching the objective of the models. As a final step, we connected each model by following the physical domain, as previously mentioned.
Through working with our partners, our technical team managed to successfully implement a Process Digital Shadow that achieves real-time anomaly detection, forecasting 10 minutes ahead, and identifies forecasted data points as normal or abnormal, offering the end-user an early warning system.
Stefanos Kokkorikos invited to participate at Circle the Med
Author: Alexandros Patrikios
25th June 2025
Earlier this month, Stefanos Kokkorikos, CORE Group Co-Founder & Managing Partner, was invited to participate at the 8th edition of the Mediterranean Forum, Circle the Med. It was an event full of workshop activities and insightful panel speeches on how EU funding opportunities can drive positive change for the countries of the Mediterranean.
Circle the MED aims to bring together policymakers, political and business leaders, experts and key stakeholders to engage in an interactive dialogue on the urgent need to accelerate the green transition towards climate neutrality and resilience in the Mediterranean.
Panel discussion with Stefanos Kokkorikos
CORE Group co-organized the session on EU Grants Networking Workshop: Building Consortia for a Resilient, Green, and Inclusive Mediterranean, moderated by our partners from JOIST Innovation Park.
The first part of the session comprised a panel discussion on securing EU funding. Stefanos took the stage to introduce CORE Group and the CORE Innovation Centre to forum participants and answered burning questions from the audience on how organisations can profit from funding opportunities at all levels.
You can watch the panel discussion in full through the link below.
Proposal writing workshop
The panel was followed by a proposal writing workshop, where participants where brought together in separate groups to discuss funding opportunities and set out to plan the drafting process.
It was a highly intriguing session, helping us build collaborative networks and introduce CORE Group and the work we do to potential partners in the Mediterranean region through actionable channels.
A big thank you to forum organisers for a very insightful discussion!
We look forward to attending next year.
The s-X-AIPI project has concluded
Author: Vassia Lazaraki, Athanasia Sakavara, Nikos Makris, Clio Drimala
24th June 2025
The s-X-AIPI project aspired to transform the EU process and manufacturing industries by developing an innovative, open-source toolset of trustworthy self-X AI technologies. These AI systems are designed to operate with minimal human intervention, continuously self-improve to boost agility, resilience and sustainability throughout the product and process lifecycle.
s-X-AIPI supports industrial workers with smarter, faster decision-making, while promotes integration into a circular manufacturing economy. Utilising innovative AI tools enhances design, development, operation and monitoring of plants, products and value chains.
s-X-AIPI demonstrated in four industrial sectors: Asphalt, Steel, Aluminum, and Pharmaceuticals, showcasing a portfolio of trustworthy AI technologies like datasets, AI models, applications, which integrated into an open source toolset. Key components of this toolset include an AI data pipeline with automatic computing capabilities, an autonomic manager based on MAPE-K models, that supports Human In The Loop, as well as several AI systems based on continuous self-optimiσation, self-configuration, self-healing and self-protection.
Launched in May 2022 with 14 partners from 6 different conuntries, the 3-year project concluded in April 2025. An online Final Review will be conducted in the summer of 2025, highlighting its significant achievents during the project’s lifespan.
ADAPT AI-Powered Anomaly Detection – Steel Use Case
CORE IC developed ADAPT – Active Detection and Anomaly Processing with smart Thresholds - a cutting-edge anomaly detection system designed to enhance operational oversight in steel manufacturing, through state-of-the-art machine learning and automation.
Powered by a Conditional Variational Autoencoder (cVAE), ADAPT continuously monitors process data—including scrap input, in-process metrics, and product compositions—to identify deviations and present them to process experts.
The base model was trained in an unsupervised manner on large volumes of historical, unlabeled production data, an ideal approach for industrial environments where labeled anomalies are rare and difficult to obtain. A Bayesian optimisation framework was integrated into the training pipeline, to allow data-driven hyperparameter tuning. Finally, an explainability module was introduced to the inference for transparency and user trust: for each detected anomaly, ADAPT highlights the top contributing features, enabling experts to quickly understand root causes and take informed corrective action.
ADAPT’s strength lies in its ability to continuously evolve. It supports two complementary mechanisms for refining the base model:
semi-supervised learning loop, which incorporates expert feedback on the identified anomalies, through a Human-in-the-Loop (HITL) workflow. ADAPT’s active learning framework minimises the need for constant human intervention, while still leveraging expert input where it adds the most value—driving continuous operational decision support.
unsupervised adaptation strategy on new, unseen data that addresses both data drift and concept drift over time.
These fine-tuning processes, along with performance monitoring, model redeployment, and user feedback tracking, are fully automated within ADAPT’s MLOps pipeline. The system also logs and manages historical anomaly data, for tracking patterns, comparing model behavior over time, and supporting process audits or improvement initiatives. Experiment tracking, version control, and metadata management ensure that every model iteration is traceable, reproducible, and aligned with production requirements.
ADAPT End-to-end pipeline for robust, adaptive Anomaly Detection
Data Augmentation Techniques – Asphalt Use case
Software developments in the Asphalt Use Case (UC) faced a key limitation: the scarcity of high-quality laboratory test data. Although the data spans long operational periods, the overall volume remains low, making it difficult to train effective AI models for analysing asphalt behaviour and predicting performance outcomes. This lack of data particularly impacts supervised learning approaches, leading to class imbalance, reduced model robustness, and constrained generalisation. To address this challenge CORE IC developed a three-stage data augmentation pipeline, illustrated below, that expanded the original dataset (~500 rows) to approximately 103.
Three stages of our Data Augmentation Technique
The first stage focused on imputing missing values using the K-Nearest Neighbors (KNN) algorithm, selecting the five closest data points to estimate missing entries. In the second stage, Gaussian noise was added to the dataset to introduce variability and promote model robustness, while preserving the data's underlying structure. The final stage involved experimentation with three generative AI models—Variational Autoencoder (VAE), Denoising Autoencoder (DAE), and ReaLTabFormer—each used to generate synthetic records that enriched the dataset. These enhanced datasets were then evaluated for their impact on predictive performance. Among them, the dataset generated by the VAE method emerged as the most effective, significantly enhancing the model’s accuracy and predictive performance.
Dissemination, Communication and Exploitation Activities
We are just a few days away since the project wrapped up and cannot omit reflecting back on some valuable dissemination and communication (D&C) achievements. Over the past 3-years, CORE IC devised and led the dissemination and communication strategy, working hand-in-hand with the entire consortium to maximise the project's visibility and impact.
The s-X-AIPI team participated in 24 high-impact events presenting their findings and organised the “Transforming Process Industries with AI” project-dedicated concluding event on 9 April, 2025 in Belgrade. These events offered remarkable opportunities for an extensive audience reach across significant target groups worldwide including professionals from research and academia, industry, IT, software, and technology, business consulting, EU institutions, national/regional, and local authorities, as well as policy making, investors and financial stakeholders, specific end user communities, association representatives and the general public and media.
Beyond that, consortium members generated 8 open-access scientific articles (6 already published and 2 currently in the publication process), an important legacy of the project the full list of which can be found on the project website or the ZENODO repository. The AI4SAM Cluster was also formed with two more EU funded projects – AIDEAS and Circular TwAIn - to expand the project’s impact beyond individual efforts. This cross-project collaboration ranged from joint event participations in major conferences, targeted webinars and the s-X-AIPI final event to collaboration on digital communication activities to amplify each project’s outreach.
The s-X-AIPI website, designed and maintained by CORE IC, will continue as a central hub for useful information and resources. On the site, visitors can learn more about the project’s final results, important research activities performed and landmark research news of each project phase through 7 press-releases, 9 newsletter issues, videos, open-access scientific papers, public deliverables and training courses.
Additional strategic digital communication efforts included the creation of 13 videos - all available on YouTube - to convey the s-X-AIPI concept and remarkable achievements in a more engaging manner than text-based content. The project also shaped valuable online communities on LinkedInX, significantly expanding its reach, another reflection of the overall effectiveness of the D&C strategy.
For the Exploitable Results (ERs) that are closer to the market, CORE IC used its’ Profit Simulation Tool (PST) to forecast the financial revenues during the post-project commercialisation period. CORE’s PST offers a structured approach to support commercialisation planning by combining strategic business insights with market data. The financial forecasts provided by the PST were integrated effectively into the project’s exploitation strategy.
Further support from the CORE IC team
SIDENOR, a steel making facility in Spain, worked in s-X-AIPI with a focus on optimising scrap usage, especially addressing challenges from contaminants such as copper commonly found in lower-quality external scrap. Their main goal is to produce high-quality steel, prevent downstream surface defects, and minimise energy consumption in the Electric Arc Furnace (EAF) melting process. CORE IC contributed to this effort by supporting partners BFI and MSI through the development of anomaly detection software, as part of the overall solution.
Additionally, CORE IC was involved in the Asphalt use case. The aim of this use case was to target circularity of the value chain, from quarry to road, by enhancing quality control of feedstock (aggregates, bitumen, recycled asphalt), improving the overall sustainability of the production process (including asphalt paving) and the quality of final product (asphalt mix). Partners CARTIF, DEUSER, and EIFFAGE, also working on this use case, leveraged the augmented data generated by CORE IC to enhance model performance.
CORE Group at the 2025 International Smart Factory Summit
Author: Alexandros Patrikios
19th June 2025
The 6th International Smart Factory Summit took place earlier this month, bringing together smart factory innovators from all over the world to explore the future of smart manufacturing.
During the Summit, Dr. Nikos Kyriakoulis, CORE Group Co-Founder and Managing Partner, participated in a pitching session for the Greek Smart Factory, our CORE initiative facilitated through the Twin4Twin project.
The Summit
Hosted by our Twin4Twin partners Swiss Smart Factory (SSF), the Summit has served as a global platform and an annual gathering for the smart factory community to discuss how these ecosystems can transform industrial operations.
Under the theme “Deep Tech Smart Factory – Uniting Humans, AI, Robots & Processes”, ISFS25 focused on the economic and societal impacts of emerging deep technologies and their role in reshaping manufacturing. The three-day event was geared toward international decision-makers from both the public and private sectors, and brought together a global community of experts from Europe, Asia, Africa, South and North America.
The Greek Smart Factory pitching session
Dr. Nikos Kyriakoulis took the stage to present the Greek Smart Factory, a test and demo platform that brings together manufacturers, tech providers, and academia to drive innovation in real-world industrial settings. During his talk, he presented the strengths of the GSF, as well as key steps forward towards the bigger picture of bringing the initiative to life.
If you would like to know more about the GSF initiative or express your interest, you can fill the form available here and our team will reach out.
Stefanos Kokkorikos, CORE Group Co-Founder and Managing Partner, was also in attendance. Stefanos Kokkorikos is also the Project Coordinator for the Twin4Twin Project, an EU Horizon Widening project and a big milestone for CORE Group and CORE IC. You can find more information on the Twin4Twin project website.
Our warmest thanks to the Swiss Smart Factory for the ongoing collaboration.
CORE Group at Data Week 2025
Author: Konstantina Tsioli, Athanasia Sakavara
17th June 2025
This year’s Data Week, organised by the Big Data Value Association (BDVA) in collaboration with the CORE Innovation Centre, took place in Athens last month.
This year’s event focused on the fundamental elements of Data Value creation. Artificial Intelligence (AI) serves as a pivotal tool in unlocking data value through various means, hence the event’s name, “Odyssey of AI: Navigating the Data Seas”.
Our team was invited to participate in two sessions.
Manufacturing Data Spaces and the European approach to AI
During this BDVA session, we discussed how AI on Demand (AIoD) and European Digital Infrastructure Consortia (EDICs) can accelerate manufacturing-oriented verticals, aligning with the product lifecycle (design, production, supply chain, and end-of-life). The Digital Transformation department from CORE INNOVATION, highlighted the work done within EU initiatives to pilot AI-driven manufacturing solutions—such as modular and adaptive production (MODUL4R) and dynamic supply-chain optimization in servitised manufacturing frameworks (M4ESTRO).
Despite Europe’s booming AI ecosystem, only 13.5% of companies have adopted AI, underscoring the need for structured frameworks like AIoD (offering ready-made tools for rapid prototyping, generative design, and quality control) and EDICs (providing scalable infrastructure like material science LLMs, digital twin platforms, and circular economy data spaces).
The session emphasised AI Continent’s five pillars (computing, data, skills, regulation, and strategic adoption), positioning AIoD as the front-end for industry-ready tools (e.g., federated datasets for supply chain optimization) and EDICs as the backbone for heavy-duty infrastructure (e.g., cross-border data spaces). TEFs and EDIHs bridge adoption gaps, but open questions remain on open-source accessibility and compliance efficiency. The goal? A cohesive ecosystem where stakeholders—from SMEs to R&D—can leverage standardized, compliant AI solutions to fast-track manufacturing innovation.
MLOps, Continuous Learning, and Resource Management in the Edge-Cloud Continuum
For this session, our contributions included shaping the agenda, participating in a panel discussion, and delivering the presentation “Why Models Fail: Leveraging MLOps and Active Learning for Resilient and Adaptive AI.”
In our talk, we addressed key challenges in the machine learning lifecycle, focussing on data drift and concept drift, which often lead to model degradation—and emphasized the importance of Human-in-the-Loop (HITL) methods to build more trustworthy and adaptive AI systems. We highlighted the value of modern MLOps practices such as CI/CD pipelines, automation, versioning, monitoring, and active learning integration, all of which are essential for maintaining high-performing AI in dynamic, real-world environments.
Going from theory to practice, we presented two solutions we developed in ongoing R&D projects that exemplify scalable, automated, and reliable MLOps systems: in S-X-AIPI, we developed ADAPT, an intelligent anomaly detection system; and in M4ESTRO, we introduced a multi-agent reinforcement learning framework that delivers actionable insights for optimizing adaptive value chains in manufacturing.
A big thanks to our growing community, and especially to the BDVA for organising this top-notch annual gathering.
COREbeat Pilot Installation at Viopol
Author: Vasilis Sagiannos
June 5th 2025
We’re excited to announce the successful installation of COREbeat at the production facility of Viopol S.A., a leading manufacturer of polyurethane systems.
This pilot project marks the beginning of a strategic collaboration between CORE Group and Viopol, focused on enhancing asset reliability and operational efficiency through predictive maintenance.
Installation Overview
Last week, our team deployed COREbeat’s beatBox sensors on three critical machines—Fatboy, Tallboy, and IT1—at Viopol’s state-of-the-art production site. These machines, part of the polyurethane Mixing Stations, include three pumps and three electric motors that are essential to Viopol’s daily operations, making them ideal candidates for continuous monitoring and performance analysis.
The sensors are now live, collecting high-frequency vibration and temperature data, which will be analysed in real-time through the COREbeat platform. This data will help the Viopol technical team monitor equipment health, identify potential failure patterns early, and optimize their maintenance planning.
Why it matters
Unplanned downtime and equipment failure are major pain points in manufacturing. By shifting from reactive to predictive maintenance, companies can significantly reduce maintenance costs, improve production uptime, and extend the life of their assets.
Through this pilot, COREbeat will provide Viopol with:
Real-time condition monitoring
Predictive insights based on vibration and temperature data
Monthly performance reports and health scores
Early warnings for emerging mechanical issues
Over the next 10 months, our teams will collaborate closely to refine alerts, assess performance impact, and explore the potential for scaling the solution across other parts of the facility.
A Shared Vision
This initiative reflects both companies’ commitment to data-driven decision making, operational sustainability, and industrial innovation. By leveraging advanced analytics and AI, we aim to empower maintenance teams with the tools they need to prevent failures and operate more efficiently.
We are proud to partner with Viopol on this journey and look forward to the insights and outcomes the pilot will generate.
Optimising Battery Performance and RUL with CE-DSS
Author: Christina Vlassi
June 3rd 2025
Our tech team designed a Decision Support System (DSS) solution that enhances battery usage decision making, paired with an innovative Remaining Useful Life (RUL) estimation module, designed for smarter analysis and long-term optimisation.
We developed the solution as part of the DaCapo project, for end-user Fairphone. DaCapo aims to create human-centric digital tools and services which improve the adoption of Circular Economy (CE) strategies throughout manufacturing value chains and product lifecycles. The project has been ongoing for 2 and a half years and comprises 15 partners across 10 countries with a budget of 5.99 million euros.
Fairphone is a key DaCapo partner, founded in 2013 to address the “make-use-dispose” trend through its focus on modular smartphones that are durable and easy to repair.
How our solution works
Our aim was to develop a DSS that enhances decision-making around battery usage, through its pairing with a RUL estimation module – all of which is accessible to users through a dedicated web app. When users access the app, they get instant access to a battery report preview for all their devices.
From there, the system splits into its two core functionalities: RUL Estimation and Forecasting & Suggestions.
Remaining Useful Life Estimation
In this module, users can see battery capacity loss based on mathematical models that map out capacity degradation over time. These insights reveal how much useful life remains in the battery — giving users a clear view of their device’s condition.
Where the DSS shines is through the provision of personalised guidance. By analysing app usage patterns, the system identifies the impact of each app on battery degradation. Through data analysis, we offer custom recommendations to help users understand and adjust usage patterns that are draining their battery life faster than necessary.
Forecasting & Suggestions
Harnessing the power of machine learning, the system predicts RUL 10 days ahead — tailored specifically to the user’s habits. With these predictions set, the web app goes on to offer actionable tips for longer battery life, with suggestions on the temperature control, charging patterns and usage patterns. When users follow these tips, the RUL of their device is improved, which is reflected visually on the app, showing the tangible benefits of informed action and encouraging sustainability-oriented behaviour.
Towards a Greener Future
By helping users extend the battery life of their devices, our system supports circular economy principles, reducing electronic waste and promoting sustainability. Users are empowered to optimise their battery usage, minimise capacity loss and make smarter, eco-conscious choices – one device at a time.
Multi‑Level Communication and Computation Middleware in MODUL4R
Author: Maria Tassi
May 30th 2025
The MODUL4R project is transforming industrial manufacturing by creating adaptable, resilient, and reconfigurable production systems. The backbone for this transformation is a distributed control framework for modular "Plug & Produce" (PnP) systems, the development of which was successfully completed in April 2025.
CORE IC led these research activities and developed the Multi‑level Communication and Computation Middleware (MCCM) - a fundamental innovation in MODUL4R - that bridges the gap between Cyber‑Physical Systems (CPS) and modular data exchange frameworks, allowing for smooth interoperability across edge, fog, and cloud layers.
The Role of MCCM in Modular Manufacturing
The MCCM was developed to tackle the challenges of modern manufacturing where agility, real‑time decision‑making, and scalability are essential, optimizing industrial operations through dynamic task distribution. By integrating edge, fog, and cloud computing, the MCCM ensures efficient data processing, low latency communication, and dynamic resource allocation. Its architecture is in line with the RAMI 4.0 reference model, supporting Industry 4.0 standards and enabling vendor‑agnostic communication.
Some of the key features of MCCM include:
Hybrid Computing Platform
Edge Layer: Serves as the entry point to the asset layer, ensuring low latency communication and direct interaction with physical systems, enabling real‑time interaction with shopfloor devices.
Fog Layer: Acts as an intermediary for intensive computation, edge orchestration, and near real‑time execution, minimising the latency for critical decision making while reducing cloud dependency.
Cloud Layer: Provides security, scalability, and robustness while enabling external communication with third‑party services.
Dynamic Reconfiguration
Enables real‑time adjustments to workflows, ensuring adaptability to changing production demands.
Supports Infrastructure‑as‑a‑Service (IaaS), allowing third‑party applications to be deployed via containerization (e.g., using Kubernetes).
Service‑Oriented Architecture
Facilitates modular deployment of microservices, ensuring flexibility and scalability.
Uses MQTT, REST APIs or seamless data exchange between layers.
Orchestration Across Layers
The Orchestration Controller offers dynamic service allocation, across different levels, to optimize resource usage across different levels optimizing resource usage.
Enables multi‑cluster management, ensuring efficient workload distribution.
Orchestration within the MODUL4R hybrid computation environment
In more detail, MCCM establishes a service‑oriented, hybrid computation platform designed to orchestrate communication and computation within Cyber‑Physical Systems of Systems (CPSoS) networks. On the shopfloor, data generated by individual CPS components is acquired through industrial communication protocols such as OPC‑UA at the edge layer. This data is then transmitted via MQTT message brokers to corresponding MODUL4R services, where it is processed.
These services, along with the brokers, are typically deployed across the fog and cloud layers, depending on latency requirements and computational needs. MCCM facilitates seamless deployment of services and brokers across the appropriate layers and physical locations, enabling modularity and adaptability. This architecture ensures that each system component can operate on the most suitable computational resource, enhancing efficiency, reducing latency, and supporting scalable system operation.
By enabling dynamic orchestration and real‑time reconfiguration, MCCM provides a flexible infrastructure that supports near real‑time production optimization. The fog layer plays a critical role in enabling low‑latency data processing and rapid decision‑making, essential for responsive manufacturing systems.
Furthermore, MCCM supports third‑party application deployment in an Infrastructure‑as‑a‑Service (IaaS) model using containerization technologies such as Docker and Kubernetes. This approach not only enhances scalability and flexibility but also allows for real‑time system updates and adjustments without production downtime. To maintain system integrity and continuity, MCCM incorporates version control and traceability mechanisms, allowing manufacturers to roll back or update services and algorithms as needed—ensuring both operational stability and adaptability.
Implementation in MODUL4R Use Cases
The MCCM has been successfully deployed across MODUL4R’s pilot cases, demonstrating its versatility and impact:
FFT Use Case: The Quality Check station transmits capacitor inspection data via MQTT to the fog layer, where it is processed and forwarded to the cloud for analytics.
SSF Use Case: A soldering production system uses the MCCM to synchronize data from multiple stations, ensuring real‑time quality monitoring and control.
EMO Use Case: A CNC milling machine streams sensor data through MCCM to the cloud for predictive maintenance analytics.
NECO Use Case: MCCM orchestrates robotic arm coordination and metrology data.
Benefits for the Industry
The MCCM delivers transformative advantages for manufacturers:
Reduced Latency: Fog computing minimizes delays for critical decision‑making.
Scalability: Containerized services allow easy expansion to meet production needs.
Interoperability: Standardized protocols such as MQTT ensure compatibility with legacy and modern systems.
Resilience: Dynamic reconfiguration enhances system robustness against disruptions.
Security: Frameworks and tools such as cloud‑message‑brokers (Kafka, MQTT) and GAIA‑X policies ensure secure data management & distribution, all the way from edge sensors to cloud services
Conclusions
The MCCM delivers transformative advantages for manufacturers, by improving responsiveness, scalability, interoperability, and robustness in production systems. Using fog computing, MCCM minimizes latency through localized data processing, allowing for quicker, real time decision‑making.
Its support for containerized services enables rapid reconfiguration and seamless deployment, making it simple to grow operations as needed. Standardised protocols such as MQTT and REST‑APIs enable interoperability with both legacy and modern systems, whereas dynamic reconfiguration capabilities improve system resiliency, allowing operations to respond easily to disturbances or changing production requirements.
The Multi‑level Communication & Computation Middleware is a key component of the MODUL4R project, allowing for seamless integration of distributed control systems, real‑time analytics, and modular manufacturing workflows. By connecting edge, fog, and cloud layers, the MCCM enables manufacturers to achieve flexible, efficient, and sustainable operations, paving the way for the factories of the future.
3D Simulation‑driven optimisation for smart manufacturing lines
The Swiss Smart Factory use‑case
Author: Pantelis Papachristou
May 26th 2025
The latest demo by CORE Group’s technical team showcases a groundbreaking approach to drive digital transformation of manufacturing lines, using 3D simulation‑driven optimisation.
In collaboration with the Swiss Smart Factory (SSF) in Switzerland and using software developed by Visual Components, our team created a demo that shows how 3D simulations and data analytics can address critical bottlenecks and enhance overall efficiency and productivity in complex manufacturing operations.
3D simulation‑driven optimisation for smarter manufacturing
3D simulation is a powerful approach that uses advanced simulation software to create detailed virtual 3D models of production systems. These realistic virtual representations (3D scenes) accurately mirror the real world, including not only the physical geometry of objects and structures but also their texture, colour, lighting, and other visual properties.
In the context of manufacturing, 3D simulations are used to replicate entire production lines, individual machines, and workflows in a virtual environment. This method enables teams to identify inefficiencies and bottlenecks in the production line, optimise workflows, and test improvements without interrupting real‑world operations and risking downtime or disruption to actual production.
Key Features of the demo
The 3D simulation‑driven demo was developed in collaboration with the SSF, using simulation software developed by Visual Components - partnerships which our team has secured through the Twin4Twin and Modul4r Horizon EU projects. The demo was initially showcased during CORE Innovation Days, Greece’s first Industry 4.0 conference organised and hosted by CORE Group, with a follow‑up demonstration at CORE Group’s Beyond Expo booth.
In our demo, we focused on the SSF’s production line for the F330 drone, which includes several automated and robotic‑enhanced stations, such as a 3D printing farm for the drone’s blades, assembly and packaging stations, and a warehouse. By analysing the flow of components and monitoring machine utilisation, we identified areas for improvement and tested various optimisation strategies to boost overall throughput and efficiency.
3D simulation of the production line: Using the Visual Components simulation software, a virtual model of SSF’s production line was developed, including individual workstations and the transition of components (e.g. through Automated Guided Vehicles). To obtain a clear view of the production line’s behavior and establish a baseline for its performance, virtual sensors were integrated in each station to measure its cycle time, utilisation and throughput over time.
Data‑driven analysis: Leveraging data from each station, provided the foundation for identifying bottlenecks and inefficiencies in SSF’s production line that slowed down production flow. Through a dedicated data analysis, we pinpointed the root causes of these bottlenecks, highlighting the problematic areas which are keen to potential improvements and refining.
Optimisation scenarios: Based on the data analysis results, we tested four targeted optimisation scenarios, to address the observed bottlenecks. For example, we introduced intermediate buffer storage systems, expanded the capacity of particular workstations and added extra stations. Running the 3D simulations on these different scenarios allowed us to compare the optimisation results with the baseline, quantifying the simulated improvements in terms of overall production rate. Interestingly, the results were not always straightforward, as adding extra stations can also create new bottlenecks elsewhere in the line, leading to a decrease in productivity. Such unexpected results underscore the need for simulations in complex manufacturing environments, where interactions between different workstations can have non‑linear and counterintuitive effects.
Final report: Finally, we created a report that juxtaposes the cost for implementing each optimisation scenario with the induced improvement in production rate. This analysis was crucial in understanding the trade‑offs between implementing these optimisation strategies and the tangible benefits they provided. Moreover, this report serves as the starting point for deriving a more detailed ROI analysis. By inserting specific financial metrics, such as production costs and profit margins, manufacturers can calculate the potential financial impact of each optimisation scenario.
The Future of Manufacturing: Smarter, Faster, More Efficient
Our demonstration underscored the potential of 3D simulation‑driven optimisation to revolutionise manufacturing processes. By combining simulations with data analytics and photorealistic 3D models, manufacturers can gain deep insights into their operations, experiment with different optimisation scenarios, and implement improvements without disrupting production.
This approach not only helps identify and solve bottlenecks but also enables manufacturers to make smarter, data‑driven decisions that lead to improved efficiency, reduced downtime, and increased productivity. In this case, manufacturers can test and refine their processes, ensuring that their production lines are always operating at peak performance.
As industries continue to embrace digitalisation, 3D simulation‑driven optimisation will play an increasingly important role in shaping the future of manufacturing.
The ability to simulate, analyse, and optimise processes before implementing changes – “testing before investing” – offers a significant competitive advantage, allowing manufacturers to stay ahead of the curve and continue innovating in an ever‑evolving market.
The DiG_IT project has reached its conclusion
Author: Valia Iliopoulou
6th May 2025
After 4 and a half years, the DiG_IT project, which aimed at the transition to the Sustainable Digital Mine of the Future, is concluded. The main goal of the project was to address the needs of the mining industry to move forward towards a sustainable use of resources while keeping people and environment at the forefront of their priorities.
To this end, our consortium built an Industrial Internet of Things platform (IIoTp) which collects data from the mining industry (from humans, machines, environment and market) and transforms them into knowledge and actions. The aim of our IIoTp was to improve worker health and safety, making operations more efficient and minimising the environmental impact of mining.
The CORE team had a broad role in the project and contributed to the development of key components of the platform.
Safety Toolbox: Biosignal Analytics and Anomaly Detection
CORE was responsible for designing two out of the three components of the Intelligent safety toolbox that provides insights and supports the prevention of hazardous situations for people’s health in the mining field. The first component is the Biosignal Analytics & Anomaly Detection agent that is part of the Safety monitoring system of the Decision Support System (DSS).
The Biosignal Analytics component is a cloud agent responsible for monitoring and detecting changes in the health state of the individuals that work inside the mines. The system utilises the biometrical data from the smart garments and pairs them with Anomaly Detection models that operate in real-time, to detect any possible alerting states in the health of the miners.
Safety Toolbox: Air-quality smart monitoring and forecasting
The second component of the safety toolbox is the Air-quality smart monitoring and Forecasting agent, that is part of the Environmental and Safety monitoring system of the DSS. The agent is responsible for providing predictions of the air quality KPIs, meaning forecasting how specific air-quality substances are going to progress in the future.
This is useful for safety reasons, like notifying that a section of the mine should be evacuated when a dangerous substance exceeds the accepted limits. The development of the agent is based on Multi-Variate Neural Networks that are trained from data gathered from sensors that are deployed inside the mines.
Examples of actual and forecasted values for NO2
Analysis of target variables per hour
Predictive Operation System I
One of our team’s primary roles was the development of a Predictive Operation System that utilises AI architectures. The system consisted of a forecasting agent that predicts the consumption of an individual asset. The knowledge of anticipated consumption is critical for planning of the field operations and optimising the industrial processes.
Our team relied on COREbeat, our end-to-end predictive maintenance platform, to assist Marini Marmi, a historic marble quarry in the north of italy, with the operation of one of their critical assets. COREbeat was installed on an electrical supplied milling machine utilized for cutting though marble cubes producing marble slices. During the project, the operators received critical warnings from COREbeat. CORE team investigated and indicated the origin of the fault to Marini operators, who decided to halt the machine's operation and placed it in maintenance mode, preventing further damage to the machine.
More information on how COREbeat assisted the staff at Marini can be found here.
Predictive Operation System II
CORE was also primarily responsible for a Predictive Maintenance system that utilises Machine Learning methods and techniques. The system was designed to focus on individual assets, enabling the assessment of their overall health and the prediction of their future states in real-time. The assets selected for predictive maintenance were paired with Anomaly Detection models, specifically designed for the asset type, to maintain input consistency between applications. The goal was to create alerts that can provide early warnings to users about possible failures of the operating equipment.
Anomaly scores and anomaly thresholds of the test set
Commercialisation phase
To ensure successful commercialization, CORE’s dedicated team developed an Innovation Strategy, focusing on clear value propositions and competition mapping. The main findings show that European Economy will need to multiply its production of critical raw materials, due to the increasing digitalisation of the economy and use of renewables. Therefore, AI solutions are obligatory, if mining industry companies want to optimise operations as well as offer environmental protection but also health and safety to their employees.
Additionally, CORE developed business models for two key innovations introduced by the project, the DiG_IT IoT Platform, for mining operations optimisation, online measurements and failure predictions, and the Dig_IT Smart Garment, for improved safety and reduced risk of accidents and injuries, the two innovations with the highest TRL exposed in real conditions.
The InBlanc Project has kicked off
Author: Clio Drimala
April 24th 2025
The INBLANC research project has officially kicked off, bringing together professionals around Europe to Delft, the Netherlands for its official launch this past February.
The project aims to transform data usage in the building and construction sector by creating an open ecosystem that will maximise the value of building data across its lifecycle.
Diving into building data
Buildings generate vast amounts of data, yet much remains scattered and underutilized. To fully capitalize upon this untapped resource, INBLANC aspires to develop an open ecosystem that will transform building lifecycle data into actionable insights. Access to rich building information can support smarter decision-making for building owners and facility managers and drive value across the entire value chain. The project developments will include a cost-efficient data accumulation framework, a Building Digital Logbook consolidating building information, structured databases connected to European data spaces, and a suite of high-added value tools and services.
To help understand and improve building efficiency, INBLANC will focus on five key metrics:
Energy – Efficiency, consumption, and performance
Human – Health, occupant comfort, and well-being
Environment – Emissions, materials, and sustainability
Economy – Cost-effectiveness, value over time and resilience
Resilience – How buildings adapt to unexpected disruptions and stress
Using a nexus approach, INBLANC will model how these indicators influence each other, creating a full-picture view of building performance and identifying areas for improvement. All this data and analysis will feed into a suite of digital tools and services designed to help key stakeholders take action. Whether it is planning a renovation, improving energy consumption, enhancing indoor environmental health, or managing a property more efficiently, INBLANC will turn complex data into clear, practical steps.
CORE Innovation Centre’s Role
CORE IC will bring deep expertise to key aspects of the project:
Sustainable Renovations and Smarter Energy Use: CORE IC will play a key role in INBLANC, leading research on “Service toolset integration for capitalising Building Lifecycle Data” and “Energy Management Services”. This includes the creation of tools designed to optimise renewable energy investments and enable real-time energy management across buildings.
The team will also contribute to map financing instruments for deep renovations and contribute to nexus modeling with bio-physical and cyber-physical models. CORE IC’s efforts will also support tool integration within the project’s data model and focus on AI-driven analytics to improve decision-making.
In addition, CORE IC will be involved in a broad set of tasks, including data collection, sustainability roadmaps, decision support systems, and large-scale performance gap analysis. Through advanced monitoring and optimization strategies, CORE IC will help drive more efficient, sustainable renovations and smarter energy use.
Dissemination, Communication and Innovation Management: CORE IC will also lead all strategic communications to increase project awareness and build INBLANC's reputation across crucial target groups, from key specialists and stakeholders to broader audiences. This will ensure that the project's actionable solutions receive the desired attention and add value to potential users. Moreover, CORE IC will lead Exploitation and Innovation Management, focusing on analyzing, refining, and commercializing Key Exploitable Results (KERs). These efforts will ensure INBLANC's research transitions into market-oriented solutions, fostering innovation and sustainability.
Looking towards a sustainable future
Bringing together 22 partners from 10 countries over the project’s 42-month term, INBLANC will play a significant role in building a more sustainable future. Collecting, organizing, analyzing, and acting on building lifecycle data will help make buildings greener and more efficient.
The consortium partners blend expertise from leading SMEs, renowned universities, RTD institutions, and large-scale industrial organizations, including: DEMO CONSULTANTS , Frederick Research Center, Aalborg University, CORE Innovation Centre, Tampere University, R2M Solution, Z Prime, CYPE, MIWenergía, Colouree, Roelofs & Haase, Comfortica, EPLO European Public Law Organization, Municipality of Thessaloniki, Siemens AG Oesterreich, Zenith, CMB, Estia SA, EPFL, EPIQR Rénovation, Hôpitaux Universitaires de Genève (HUG), ETH Zürich
CORE Group at the 2025 Beyond Expo
Author: Alexandros Patrikios
14th April 2025
Beyond Expo took place last week, and our team couldn’t miss out.
It’s our third time joining the expo, and the first time it takes place in Athens. Our team was there to showcase the latest demos from our Horizon projects, as well as the COREbeat end-to-end predictive maintenance platform.
The CORE Group booth
Throughout the three-day event, we had the pleasure of engaging with many attendees stopping by our booth.
The expo was an interesting mix of industry veterans and AI enthusiasts, all of whom were very excited to see our Predictive Maintenance platform, COREbeat, in action. Our showcase included a motor connected to a beatBox, COREbeat’s hardware component, and visitors were allowed to push a nail in the machine and watch COREbeat’s UI platform spot the malfunction in near-real time. You can watch a short version of our COREbeat demo here.
Visitors also got to learn more about our Research & Innovation initiatives through the CORE Innovation Centre, with two live demonstrations of our most recent demos:
FAIRE, which is a cutting-edge solution that combines AI, edge computing, and federated learning to address critical challenges in industrial operations. FAIRE was developed through our participation in the MODUL4R and RE4DY EU projects. You can watch a short version of our demonstration here.
Smart Manufacturing Lines, a demo of a 3D Simulation-driven optimisation approach, testing different scenarios for smart manufacturing lines. This demo was developed in partnership with SSF, our Twin4Twin project partners, and Visual Components, our visualisation partner in the MODUL4R project.
Apart from our live demonstration, we also showcased the ELEXIA, DACAPO, and TRINEFLEX, all of which focus on energy optimisation across different industries.
Our booth was designed and constructed by the amazing team at Level Up events and exhibitions.
Industry Panel Discussions
Our managing partners, Dr. Nikos Kyriakoulis and Stefanos Kokkorikos, were invited to participate in two separate panel discussions, discussing the latest developments in machine learning and AI.
Stefanos Kokkorikos took part in a panel discussion hosted by HETiA, talking about on the opportunities offered by edge computing technologies, as showcased by our Horizon EU projects participating in the expo.
Nikos Kyriakoulis participated in the main conference of the event, talking about the AI boom and how SIRI can help manufacturers evaluate their Industry 4.0 readiness. You can watch the discussion in the video below.
As part of our participation in the expo, Dr. Nikos Kyriakoulis was also invited for an interview on CORE Group and its initiatives. You can watch the video below.
Stay in touch
A big big thank you to everyone who took the time to stop by our booth and say hi!
We look forward to staying in touch and collaborating.
See you all soon!
CORE Group at the 2025 Smart Factory Conference
Author: Alexandros Patrikios
March 27th 2025
Our team sponsored this year’s Smart Factory Conference, which took place earlier this week.
Dr. Nikos Kyriakoulis, CORE Group Co-Founder and Managing Partner, was invited as a speaker, to talk about the SIRI assessment tool and how it can help manufacturers navigate the Industry 4.0 landscape.
The first step towards Industry 4.0 success
Dr. Nikos Kyriakoulis, Co-Founder and Managing Partner at CORE Group, delivered an intriguing presentation, talking about potential pathways towards Industry 4.0 integration. He presented SIRI, the digital maturity assessment tool that helps prioritise new Industry 4.0 initiatives and systematically re‑validate existing projects.
You can learn more about SIRI here, and you can watch the full presentation below.
The CORE Group showcase
Conference attendees got to take a look at COREbeat, our end-to-end Predictive Maintenance solution. You can find out more information about COREbeat here.
We also got to spotlight some of our key Horizon EU projects, raising awareness about the StreamSTEP and DaCapo projects.
StreamSTEP is a newly-launched project, which aims to streamline the optimisation of Sustainable Thermal Energy Systems and prototype new technologies in process industries.
The DaCapo project, already on its third year, aims at the creation of human-centric digital tools and services for improving the adoption of Circular Economy strategies along both manufacturing value chains and products lifecycle.
Stay in touch
A big thank you to the conference organisers. See you all at the next event!
Two white papers published by our Innovation team
Author: Ioannis Batas
March 20th 2025
At CORE Innovation Days, held earlier this year, our Innovation Department published two white papers: "Our Innovation Management Methodology for EC-funded Projects" and "Factory of the Future: What’s Happening, What’s Evolving, and What’s Next."
The two white papers were distributed to attendees, offering valuable insights into our innovation management methodology and the transformative potential of Industry 4.0.
White Paper #1: "Our Innovation Management Methodology for EC-funded Projects"
In the dynamic landscape of EU-funded research, turning innovative concepts into tangible solutions requires a strategic approach. In the first white paper, we present our Innovation Management methodology, which unfolds through a four-phase exploitation strategy designed to help researchers and consortium partners navigate the entire process. From identifying project Key Exploitable Results (KERs) to developing market roadmaps, our methodology ensures that results are protected through Intellectual Property Rights (IPR) and strategically positioned for market adoption.
Our Innovation Management methodology also features the CORE Exploitation Canvas, a tool we designed to streamline the market entry of KERs from EU-funded projects. The canvas guides teams through 10 key blocks: identifying partners and IP ownership, selecting IPR protection, analyzing the target market, addressing barriers, assessing broader impact, evaluating the State of the Art and Unique Selling Points, outlining exploitation routes, setting actions and milestones, and identifying costs and revenue streams. By simplifying complex innovation processes, the CORE Exploitation Canvas helps teams align outputs with market needs, accelerate adoption, and create sustainable impact.
You can download the white paper here.
White Paper #2: "Factory of the Future: What’s Happening, What’s Evolving, and What’s Next"
Industry 4.0 is revolutionising manufacturing, presenting both opportunities and challenges for companies. Our second paper explores the transformative potential of smart factories enabled by AI, IoT, and other advanced technologies. The global Industry 4.0 market is experiencing explosive growth, projected to reach €511 billion by 2032, driven by the shift towards scalable, automated, and interconnected production systems.
This white paper analyses key technologies shaping the future of manufacturing, from AI and IoT to robotics and additive manufacturing, while addressing the barriers companies should mitigate. It highlights critical issues like workforce upskilling, cybersecurity risks, and the integration of legacy systems, offering practical strategies to navigate these challenges. By bridging current capabilities with future possibilities, this paper serves as a guide for manufacturers looking to embrace digital transformation and secure long-term competitiveness in a rapidly evolving market.
Our insights for this second paper were further enriched by our involvement in leading the exploitation management activities of the MODUL4R and M4ESTRO projects.
You can download the white paper here.
White papers now available online
At CORE Group, we believe that knowledge sharing is essential for driving technological progress and empowering innovators to transform ideas into impactful solutions.
If you missed the event or want to explore further, both white papers are now accessible. Learn how CORE Innovation is shaping the future of research exploitation and innovation management.
For more information, don't hesitate to reach out to our Innovation Department.
OPTIMINER Project Launch: A Collaborative Step Towards Sustainable Mining
Author: Eleni Natsi
February 27th 2025
The OPTIMINER project’s kick-off meeting, held over two days in Athens, Greece, marked a significant milestone in the journey to revolutionise Europe’s mining industry. Organised successfully by I-SENSE Group/ICCS, the project coordinator, the meeting served as a key platform for all project partners to gather, discuss the core aspects of the project, and set the stage for what promises to be a transformative European initiative.
During the meeting, each partner had the opportunity to introduce themselves and engage in a fruitful discussion about the project's goals, the different work packages, and upcoming steps. A thorough presentation of the six use cases in Spain, Greece, Poland, Finland, and Chile, focusing on CRM recovery (specifically magnesium, tungsten, REE, especially neodymium, copper, cobalt, and coking coal), took place.
These detailed presentations offered insights into the planned steps, implementation strategies, and anticipated outcomes. The discussions emphasised the importance of applying cutting-edge technologies to real-world scenarios, which will be key to the project’s overall impact.
The event culminated with a delightful dinner in the heart of Athens, providing a relaxed atmosphere for project members to connect, exchange ideas, and reinforce their shared commitment to the project’s success.
The vision
The OPTIMINER project aims to tackle one of the most pressing challenges in Europe’s mining sector: efficiently and sustainably recovering critical raw materials (CRMs) from complex, low-grade ores. This ambitious initiative blends advanced technologies with sustainability efforts, striving to enhance mining efficiency while minimising environmental impact. At its core, OPTIMINER integrates innovative, AI-driven solutions to address challenges across five key modules:
REMINER: Advanced CRM recovery technologies, such as smart ore sorting, bioleaching, and phytomining, powered by an AI-driven CRM Recovery Selector.
DIGIMINER: A digital platform for smart monitoring and control, featuring a Decision Support System, Virtual Miner assistant, and Digital Twins for process optimisation.
ECOMINER: Tools designed to optimise energy and water use, along with waste valorisation, contributing to enhanced sustainability and resilience.
DEMOMINER: Real-world pilot demonstrations in Spain, Greece, Poland, Finland, and Chile, focused on CRMs like magnesium, tungsten, neodymium, copper, cobalt, and coking coal.
GLOBEMINER: Promoting global awareness and fostering EU-Chile strategic cooperation to accelerate market uptake.
CORE Innovation Centre’s Role
As a key partner in the OPTIMINER project, CORE IC will play a pivotal role across several areas:
Technical plan preparation: Leading the effort for the technical plan of each use case, including technical specifications and technological expertise (AS IS situations, data availability, sensors connectivity, and other operating systems).
CRM Recovery Selector: Responsible for defining the criteria and parameters for the CRM Recovery Selector, developing the technology database, and creating detailed profiles for each use case. This also includes the development of customizable algorithms and UI design, along with integrating a digital assistant based on Natural Language Processing (NLP).
Virtual Miner: Tasked with developing an NLP-based multi-role assistant capable of verbal interaction with miners, operators, and managers in the field. This virtual assistant will integrate with the OPTIMINER DSS to provide real-time verbal insights on predictive analytics and strategic planning.
DIGIMINER Platform: Leading the design of the DIGIMINER platform, built on data-driven Digital Twins. This will leverage sensor, historical, and operational data with proper abstraction and distribution among data sources. The goal is to design the connections and interactions of Digital Twins, DSS, and Virtual Miner, as well as establish a cloud infrastructure, featuring a hybrid data warehouse. The platform will also include a modular AI-augmented market observatory to forecast mining market values.
Dissemination, Communication, and Innovation Management: Taking the lead in raising awareness about the project’s outcomes and its impact on sustainable raw materials production. The focus is on communication, exploitation, and innovation management to ensure that the benefits of OPTIMINER reach broader audiences and translate into actionable solutions within the industry.
Stefanos Kokkorikos, Co-Founder & Managing Partner of CORE Group, presenting CORE IC at the kick-off meeting.
Our Consortium
The OPTIMINER project is a collaborative effort, involving 21 partners from 8 countries, combining academic and industrial expertise. With a total budget of €7.29 million over 48 months, the project aims to integrate state-of-the-art technological developments with practical, on-the-ground applications.
Notable partners in the OPTIMINER project include: I-SENSE Group/ICCS, Tapojarvi, JSW (Jastrzębska Spółka Węglowa SA), SALORO SL, Leonore Development, TERNAMAG (part of TERNA S.A.), EUROCORE CONSULTING, AHK Business Centre SA, University of Natural Resources and Life Sciences, Vienna (BOKU), EcoCastulum, CogniSensus, CORE Innovation Centre (CORE IC), ITA · Instituto Tecnológico de Aragón, Fraunhofer Chile, DigitalTwin Technology GmbH, Łukasiewicz – IMN, Główny Instytut Górnictwa (GIG) - Państwowy Instytut Badawczy, Fraunhofer, Iberian Sustainable Mining Cluster | ISMC, LIBRA AI Technologies and HANNUKAINEN MINING OY.
Looking ahead
With the successful launch of the OPTIMINER project, the partners are now focused on the practical implementation of the project’s use cases, the development of new technologies, and increasing awareness through targeted communication and dissemination activities. The coming months will be crucial as teams work to bring the project’s objectives to life.
The OPTIMINER project holds tremendous promise for the future of mining, offering innovative and sustainable methods for recovering and utilising critical raw materials -essential for Europe’s green transition and the global mining industry.
FAIRE: Federated Artificial Intelligence for Remaining useful life Edge analytics
Revolutionising Industrial Operations with FAIRE: Federated AI for Predictive Maintenance
Author: Konstantina Tsioli, Pavlos Stavrou
February 20th 2025
At CORE Innovation Days in January, CORE unveiled a groundbreaking demonstration of FAIRE (Federated Artificial Intelligence for Remaining Useful Life Edge Analytics), a cutting-edge solution that combines AI, edge computing, and federated learning to address critical challenges in industrial operations.
This innovative approach not only enhances operational efficiency but also ensures data privacy and scalability, making it a game-changer for industries like manufacturing, energy, and pharmaceutical.
What is FAIRE
FAIRE is a ground-breaking solution based on the MODUL4R and RE4DY EU projects. FAIRE is a federated AI solution designed to optimise industrial processes by leveraging edge computing and federated learning.
It enables real-time data processing and predictive analytics, while keeping sensitive data secure and on-premise. FAIRE showcased how it can be applied to predictive maintenance for CNC machines, but its applications extend far beyond this use case.
Key FAIRE Features
Edge Computing: This solution utilises edge devices deployed directly on the shop floor to collect and process data locally. This reduces latency, minimises bandwidth usage, and ensures real-time insights without relying on constant cloud connectivity.
In the demo, two edge devices were connected to CNC machines, collecting data relevant to tool wear and predicting the Remaining Useful Life (RUL) of milling tools.
Remaining Useful Life (RUL): is a predictive tool that estimates the time left before a machine or component fails or requires maintenance, based on real-time data and historical performance patterns. In the context of FAIRE, the RUL model predicts tool wear in CNC machines, enabling proactive maintenance and reducing downtime while ensuring data privacy and security.
Federated Learning: FAIRE employs federated learning to enable collaborative intelligence across multiple machines or factories. Instead of sharing raw data, only model parameters (e.g., insights and updates) are sent to a central server, ensuring data privacy and compliance with regulations like GDPR. This approach allows machines to "learn" from each other, improving prediction accuracy and operational efficiency without compromising sensitive information.
Data Privacy and Security: By keeping data on-premise and sharing only model updates, FAIRE ensures that proprietary information remains secure. This is particularly important for industries with strict data protection requirements.
Scalability and Flexibility: FAIRE’s architecture is designed to scale effortlessly. As new machines or edge devices are added to the network, they can seamlessly integrate into the federated learning ecosystem, enhancing the system’s overall intelligence and resilience.
Predictive Maintenance for CNC Machines
The demonstration of FAIRE solution focuses on a real life application: predictive maintenance for CNC machines. Here’s how it worked:
Data Collection: Two edge devices were connected to two CNC machines, collecting real-time data on tool wear and machine performance using industrial protocols like OPC-UA and MQTT.
Local Processing: The edge devices preprocessed the data locally, running AI models to detect anomalies and predict RUL. Results were displayed on monitors, providing operators with actionable insights.
Federated Learning: Model updates from each edge device were aggregated to a central server to update the global model. The updated model was then sent back to the edge devices, enhancing their predictive accuracy.
Real-Time Insights: Operators then could monitor tool wear and RUL in real time, enabling proactive maintenance and reducing downtime.
The benefits of FAIRE
FAIRE offers numerous benefits for industrial operations:
Smarter Machines: Continuous learning and adaptation improve machine performance and operational efficiency.
Enhanced Data Privacy: Sensitive data remains on-premise, ensuring compliance with data protection regulations and/or requirements.
Cost Optimisation: Reduced data transmission and proactive maintenance minimise operational costs.
Collaborative Intelligence: Federated learning enables machines to learn from each other, improving model accuracy across the network.
Scalability: The solution can easily scale to include additional machines or factories, making it suitable for large industrial networks.
Application areas
While the demonstration of FAIRE solution involved an example of CNC machines, its capabilities extend to various industries:
Pharmaceutical: In a sector where protecting sensitive and production data is paramount, this solution safeguards data privacy and security.
Automotive: Enhance predictive maintenance for automotive production lines.
Aerospace: Improve the performance and reliability of aircraft components.
Energy and Smart Grids: Monitor and optimise power grid equipment like transformers and substations.
Mining: Optimise the operation of heavy machinery like excavators and drilling equipment.
FAIRE represents a significant leap forward in industrial AI, combining the power of edge computing and federated learning to deliver real-time insights, enhance data privacy, and optimise operations. By addressing critical challenges like unexpected downtime, inefficient data handling, and legacy equipment limitations, FAIRE empowers industries to achieve smarter, safer, and more efficient operations.
Solutions like FAIRE will play a critical role in shaping the future of industrial automation and data-driven decision-making.
Introducing Smart Data Management to Mining Operations
Authors: Konstantina Tsioli, Nikolaos Gevrekis, Konstantinos Plessas
February 13th 2025
The CORE Innovation Centre team has developed a key backend platform for the MASTERMINE project, which aspires to become the go-to ecosystem for mines that envision digitalisation, environmental sustainability, productivity monitoring and public acceptance.
A key module of the MASTERMINE project is Cybermine, which serves as the access point to the physical world managing the field data for all components, connecting the physical and digital world through technologies like IIoT, the cloud, edge computing and machine learning ensuring the smart mine design and predictive maintenance of equipment and vehicles.
Our team has developed a backend platform within the Cybermine module, which seamlessly integrates and manages data from various sources across the mining industry. As a smart data management system, it automates data collection, storage, and access, ensuring flexibility, scalability and efficiency in the use of data.
Here’s a look at the ingredients that make this platform innovative.
An overview of the back-end platform developed by the CORE IC team
Collecting Data from Different Sources
Everything begins with data sources, which can include sensors, devices, or systems in the mining industry that generate information. However, not all data is the same - different sources provide data in different formats, structures, and transmission methods. Some data is received in real-time, while other data arrives in scheduled batches. In some cases, end users may even need to manually upload files. The platform uses a multi-layered storage approach, enabling it to secure and organise data types like real-time updates, historical records, and large files. Tools such as S3 buckets, InfluxDB, and PostgreSQL ensure both speed and reliability.
Making the Data Available to Users
Once stored, the data needs to be easily accessible. This is where the Consumption Layer comes in. This layer allows users to retrieve any data they need, whether it’s raw data straight from the source, processed insights, real-time feeds, or historical records. Through this layer, users can access raw or processed data quickly and efficiently, tailored to their specific needs, such as real-time monitoring or historical analysis.
Innovation: Making Data Collection Smarter
Traditionally, integrating data from different devices and sensors required custom-built services for each type of data source, making the process slow and complicated. The platform developed by our team eliminates this challenge by offering an automated, intelligent system that dynamically adapts to any new data source. Consequently, the effort needed to integrate new data sources is significantly reduced.
Significance for the Mining Industry
Mining operations are known for their harsh conditions, making it challenging to collect and manage data effectively. A platform like the one developed by CORE IC is critical because it simplifies data integration and enables the seamless collection of data from diverse sources, even in environments where traditional methods struggle. Its robust architecture ensures data reliability and accessibility, even in remote or extreme locations.
The ability to interpret data ahead of time is crucial for mining operations, particularly regarding heavy machinery, where real-time insights can prevent breakdowns, enhance maintenance schedules, and ensure operational continuity. By transforming raw data into actionable insights, the platform empowers decision-makers — whether managers, operators, or engineers — to make informed decisions, improving safety, productivity, and efficiency. Ultimately, the platform supports innovation, sustainability, and operational resilience in the demanding context of the modern mining industry.
CORE Innovation Days: Shaping Tomorrow’s Industry, Today
The inaugural Industry 4.0 conference, hosted by CORE Group, was an astounding success.
Author: Alexandros Patrikios
January 23rd 2025
CORE Innovation Days, Greece’s first Industry 4.0 conference, took place last week. The two‑day, invitation‑only conference brought together key global leaders from industry, academia, and policy‑making to showcase new trends, challenges and successes in digital transformation and discuss the future of Industry 4.0.
The event was funded through the CORE Innovation Centre’s participation in Twin4Twin, a project supported by the Horizon EU Widening Work Programme.
The main event | Day 1
The conference opened with remarks from Stefanos Kokkorikos and Dr. Nikos Kyriakoulis, Founders and Managing Partners of CORE Group, welcoming conference attendees and introducing the Twin4Twin project.
Their welcome was followed by greetings from distinguished guests, including Stefan Estermann, Ambassador of Switzerland in Greece; Prof. Konstantinos Karantzalos, Secretary General at the Ministry of Digital Governance; and Michail Dritsas from the Greek Ministry of Economy & Finance, who highlighted the STEP Programme’s potential in advancing Greek and European organisations.
Session A: Digital Transformation in Industry
Session A explored the transformative journey industries are taking to embrace digital innovation throughout their production pipeline, in Greece and abroad.
George Panagiotopoulos (EY) focused on the digital transformation journey of Greek industry, suggesting ways for manufacturers to leverage digital tools to create value sustainably and at scale. Raimund Klein (INCIT) introduced the strategic framework offered by the SIRI tool to Greek audiences. Dr. Dominic Gorecky (SSF) shared some real-world insights from the Swiss model for a smart factory. Dr. Nikos Kyriakoulis concluded the session, introducing CORE Group’s Greek Smart Factory initiative – a platform for manufacturers and tech providers to network, innovate and test before they invest.
Session B: The Present and Future of Industry 4.0
Session B provided a forward‑looking perspective on the technologies driving Industry 4.0 and was moderated by Prof. David Romero (World Manufacturing Foundation). Prof. Romero presented the 2024 World Manufacturing Report, which offers insights on future-proofing manufacturing and was edited by a global team of contributors, including CORE Group’s Dr. Nikos Kyriakoulis. You can read the full report here.
Discussions focused on how businesses can prepare for an uncertain and ever‑evolving future through strategic frameworks and advanced technologies such as artificial intelligence and digital modelling. Prof. Dimitris Kyritsis (University of Oslo) introduced foundational frameworks for connecting digital assets. Dr. John Soldatos (Netcompany) highlighted the transformative potential of generative AI and LLMs in industry. Dr. Foivos Psarommatis (Zerofect) discussed upcoming regulations on DPPs and how they are expected to affect industrial operations. Ignacio Montero Castro (AIMEN) explained how AASs can act as bridges between physical and digital worlds. Dr. Jacopo Cassina (Syxis) explored industrial data spaces as tools for cognitive production. The final speaker for this session, Jarkko Soikkeli (Visual Components) examined the evolution and applications of Digital Twin technologies in productivity optimisation.
Session C: Industry 4.0 Innovation Case Studies
The final session of the day showcased real-world applications of Industry 4.0 applications, highlighting groundbreaking Horizon EU Industry 4.0 projects from CORE Group and its partners in three separate panel discussions.
The first panel, Digital and Green Transition of Manufacturing, moderated by CORE Group’s Dr. Pantelis Papachristou, explored how digital and green technologies are reshaping manufacturing ecosystems, identified challenges and opportunities, and discussed the roles of innovation, data, and AI-collaboration in driving this change. Projects discussed included, among others, MODUL4R, M4ESTRO, and Circular TwAIn. Panel participants included Dr. Dimitris Panopoulos (Suite5), Prof. Paolo Pedrazzoli (TTS), Prof. Pedro Malo (Unparallel), Dr. Niki Kousi (EIT Manufacturing) and Dr. Vicky Panagiotopoulou (LMS).
The second panel, Energy and Sustainability of Processes with Industry 4.0, was moderated by CORE Group’s Valia Iliopoulou and focused on energy‑efficient and sustainable industrial processes, such as the ones introduced in TRINEFLEX. Panel participants included George Tsimiklis (ICCS), Dr. Mario Pichler (SCCH), Anna Domènech Abella (CELSA) and Dr. Hussam Jouhara (Brunel University of London).
The third panel, Digital Transformation of Mines with Industry 4.0, moderated by CORE Group’s Konstantina Tsioli, attempted to address inefficiencies, sustainability issues, and resilience demands in the mining industry, driven by high global demand for raw materials, declining ore grades, and environmental imperatives. Milestone mining projects such as MASTERMINE took center stage. Panel participants included Dr. Antonis Peppas (NTUA), Dr. Jose Ramon Valdés (ITA), Kostas Botsialas (AURORA), Dr. Santiago Cuesta Lopez (ISMC) and Emilios Vazoukis (Terna MAG).
Parallel Demo Sessions
Throughout the conference, attendees got to stop by demos set up by our team, showcasing different Industry 4.0 technology trends. One of our demos showcased COREbeat, our end‑to‑end predictive maintenance platform, which included a motor connected to a beatBox, COREbeat’s hardware component. Visitors were allowed to push a nail in the machine and watch COREbeat’s UI platform spot the malfunction in near‑real time. Another demo showcased a solution for AI application deployment which boasts data acquisition on the edge, AI modes for Remaining Useful Life and Federated Learning for increasing data volume while retaining enhanced data privacy – a combination of technologies used in MODUL4R and RE4DY, two of our Horizon EU projects. The final demo, which our team carried out in collaboration with the Swiss Smart Factory, scenario–based optimisation of an actual manufacturing process for drones, using Digital Twins and 3D visual representation for enhanced explainability.
Proposal Writing Workshops | Day 2
For the second day of the event, participants were invited to proposal writing workshops, aimed at fostering collaboration and building consortia for upcoming Research & Innovation calls. Throughout the day, partners met in groups to explore specific funding opportunities, during focused parallel sessions carried out in CORE Group’s central offices.
These workshops were designed to empower participants with practical insights and strategies for crafting competitive proposals, ensuring alignment with the latest Horizon EU priorities. By facilitating direct interaction and knowledge exchange, CORE Innovation Days is paving the way for innovative partnerships and impactful solutions that address the challenges of tomorrow.
