Periodic Reporting for period 1 - PLIADES (AI-Enabled Data Lifecycles Optimization and Data Spaces Integration for Increased Efficiency and Interoperability)
Berichtszeitraum: 2024-01-01 bis 2025-06-30
Europe’s digital transformation is deeply dependent on the ability to create, manage, share, and reuse data efficiently, securely, and sustainably across sectors. However, the current landscape of data management remains fragmented, with limited interoperability between different sectors, organizations, and technologies. This fragmentation prevents the full realization of the value that data can bring to key technological domains such as AI, robotics, and advanced digital services. Without the capacity to link, share, and analyse data across diverse environments, innovation is slowed, operational efficiency is reduced, and the potential benefits for society, the environment, and the economy remain largely underexploited.
PLIADES addresses these challenges by developing a novel data integration framework that builds upon state-of-art architectures, extending them with advanced elements to solve essential challenges in data creation, management, and sharing. PLIADES focuses on optimizing the entire data lifecycle, including green data creation and storage, data ownership, discovery, sharing, reuse, and responsible disposal. By leveraging AI-driven technologies, the project will enable the seamless linking and interoperability of multiple data spaces, transforming how data is used and shared across sectors. The framework will be deployed and validated through six use cases across five data spaces: mobility, healthcare, green deal, energy, and industrial. PLIADES will extend data spaces to cover full lifecycles, enhanced by AI-based integration. AI brokers will suggest suitable data spaces by understanding personalized needs and context, saving time and costs when accessing distributed, sovereign sources. The project will also design and test training and support schemes to ease adoption of new standards and architectures. By combining data from multiple domains, PLIADES will foster innovation, strengthen the EU economy, and improve quality of life. The aim is to validate data space integration by building meaningful aggregations that deliver insights and support decision-making.
PLIADES addresses these challenges by developing a novel data integration framework that builds upon state-of-art architectures, extending them with advanced elements to solve essential challenges in data creation, management, and sharing. PLIADES focuses on optimizing the entire data lifecycle, including green data creation and storage, data ownership, discovery, sharing, reuse, and responsible disposal. By leveraging AI-driven technologies, the project will enable the seamless linking and interoperability of multiple data spaces, transforming how data is used and shared across sectors. The framework will be deployed and validated through six use cases across five data spaces: mobility, healthcare, green deal, energy, and industrial. PLIADES will extend data spaces to cover full lifecycles, enhanced by AI-based integration. AI brokers will suggest suitable data spaces by understanding personalized needs and context, saving time and costs when accessing distributed, sovereign sources. The project will also design and test training and support schemes to ease adoption of new standards and architectures. By combining data from multiple domains, PLIADES will foster innovation, strengthen the EU economy, and improve quality of life. The aim is to validate data space integration by building meaningful aggregations that deliver insights and support decision-making.
The main results achieved by the PLIADES project so far are summarized as follows:
State-of-art analysis and requirements definition: A comprehensive literature review on the existing landscape of European data spaces, AI data lifecycle methodologies and data management approaches was completed. The results contributed to the identification of user needs, technical requirements, and the system specifications for PLIADES.
Design of the PLIADES architecture: A flexible, scalable reference architecture was designed to enable full data lifecycle management, secure data sharing, and AI-based services, ensuring alignment with European data spaces and regulations such as GDPR.
Development of sustainable data management methods: The design and development of approaches for green data management was initiated. Methods for data creation, compression and filtering have been developed, aiming to minimize energy consumption and support European Green Deal objectives.
Privacy, sovereignty, and security mechanisms: Tools for decentralized identity management, consent-based access control, and data privacy-preserving protocols for data space operations are developed to empower data owners and ensure secure and compliant data sharing.
AI-based metadata broker and data connectors: The first versions of the AI-driven metadata broker and data space connectors were designed to facilitate intelligent, context-aware data discovery and access across multiple, distributed, and sovereign data spaces.
Human-centric tools and training strategies: Initial user support materials, guidelines, and interface designs were created to enable individuals, businesses, and public organizations to adopt and benefit from the PLIADES solutions.
Deployment preparation of six use cases: The preparation for the six real-world use cases in the domains of mobility, healthcare, energy, industry, and the circular economy that will be launched, includes their scenario definitions, initial data collection and detailed planning of their deployment.
Communication, dissemination, stakeholder engagement, and exploitation: The project’s visibility and outreach were strengthened through active participation in events, publications, and the establishment of synergies with related European initiatives. An initial Key Exploitable Results analysis has been conducted by the consortium, forming the bases of (joint) exploitation activities.
State-of-art analysis and requirements definition: A comprehensive literature review on the existing landscape of European data spaces, AI data lifecycle methodologies and data management approaches was completed. The results contributed to the identification of user needs, technical requirements, and the system specifications for PLIADES.
Design of the PLIADES architecture: A flexible, scalable reference architecture was designed to enable full data lifecycle management, secure data sharing, and AI-based services, ensuring alignment with European data spaces and regulations such as GDPR.
Development of sustainable data management methods: The design and development of approaches for green data management was initiated. Methods for data creation, compression and filtering have been developed, aiming to minimize energy consumption and support European Green Deal objectives.
Privacy, sovereignty, and security mechanisms: Tools for decentralized identity management, consent-based access control, and data privacy-preserving protocols for data space operations are developed to empower data owners and ensure secure and compliant data sharing.
AI-based metadata broker and data connectors: The first versions of the AI-driven metadata broker and data space connectors were designed to facilitate intelligent, context-aware data discovery and access across multiple, distributed, and sovereign data spaces.
Human-centric tools and training strategies: Initial user support materials, guidelines, and interface designs were created to enable individuals, businesses, and public organizations to adopt and benefit from the PLIADES solutions.
Deployment preparation of six use cases: The preparation for the six real-world use cases in the domains of mobility, healthcare, energy, industry, and the circular economy that will be launched, includes their scenario definitions, initial data collection and detailed planning of their deployment.
Communication, dissemination, stakeholder engagement, and exploitation: The project’s visibility and outreach were strengthened through active participation in events, publications, and the establishment of synergies with related European initiatives. An initial Key Exploitable Results analysis has been conducted by the consortium, forming the bases of (joint) exploitation activities.
PLIADES introduces multiple key innovations that push beyond the current state-of-the-art in data spaces and data lifecycles. These advancements are summarized below:
• An AI-driven data lifecycle optimization pipeline that streamlines data collection, filtering, annotation, and sharing, improving efficiency, reducing environmental impact, and ensuring compliance for high-quality data products.
• A comprehensive Federated Learning architecture to perform AI training pipelines in a federated manner, preserving security and efficiency.
• A decentralized Clearing House based on blockchain, capable to log data sharing activities and enhance sovereignty, integrity and traceability.
• An MLOps pipeline, capable to connect diverse tools to automatically generate ML models.
• A decentralized Identity Management tool, coupled with access control mechanisms, to be integrated and deployed with data spaces.
• An initial version of the AI-driven metadata broker and data space connector, based on the IDS RAM architecture, which enable interoperability across domains.
• Analytics and decision support systems, including explainable AI and data aggregation mechanisms in federated environments.
• An ontology matching and merging mechanism, coupled with ontology abstractions, which allow the operation of cross-domain data findability and querying.
These innovations are still under development, yet their first versions demonstrate high potential ahead of project completion.
The project will accelerate AI and robotics adoption across key sectors, boost Europe’s digital competitiveness, support climate goals by lowering the data carbon footprint, and reinforce Europe’s leadership in trusted data spaces. To ensure long-term impact, PLIADES engages with standardization bodies, policy makers, and industry stakeholders.
• An AI-driven data lifecycle optimization pipeline that streamlines data collection, filtering, annotation, and sharing, improving efficiency, reducing environmental impact, and ensuring compliance for high-quality data products.
• A comprehensive Federated Learning architecture to perform AI training pipelines in a federated manner, preserving security and efficiency.
• A decentralized Clearing House based on blockchain, capable to log data sharing activities and enhance sovereignty, integrity and traceability.
• An MLOps pipeline, capable to connect diverse tools to automatically generate ML models.
• A decentralized Identity Management tool, coupled with access control mechanisms, to be integrated and deployed with data spaces.
• An initial version of the AI-driven metadata broker and data space connector, based on the IDS RAM architecture, which enable interoperability across domains.
• Analytics and decision support systems, including explainable AI and data aggregation mechanisms in federated environments.
• An ontology matching and merging mechanism, coupled with ontology abstractions, which allow the operation of cross-domain data findability and querying.
These innovations are still under development, yet their first versions demonstrate high potential ahead of project completion.
The project will accelerate AI and robotics adoption across key sectors, boost Europe’s digital competitiveness, support climate goals by lowering the data carbon footprint, and reinforce Europe’s leadership in trusted data spaces. To ensure long-term impact, PLIADES engages with standardization bodies, policy makers, and industry stakeholders.