Periodic Reporting for period 1 - DATAWiSE (Intelligent and Sustainable Building Management powered by Cross-Sectoral Lifecycle)
Okres sprawozdawczy: 2024-06-01 do 2025-11-30
DATAWiSE will develop and test building and building portfolio management tools, leveraging cross-sectoral lifecycle data within an open, secure, interoperable, and scalable framework. Utilizing cutting-edge artificial intelligence and advanced analytics, the methodology will integrate data from multiple sources to provide a holistic understanding of building operations. DATAWiSE will design and implement a scalable and interoperable data architecture and a Data Sharing Platform that prioritizes data sovereignty to secure data handling and preserve ownership control. DATAWiSE will develop two main toolkits, namely the Data-driven Building Performance Management (DBPM) harnessing Building Information Modeling (BIM) data in conjunction with advanced data mining techniques through a digital twin and the Lifecycle Data-driven Decision Support (LD2S) toolkit, serving as a versatile solution for informed decision-making across planning, renovation, and sustainability domains. Through these tools, the project will offer added value services aimed at optimizing building management. The project also incorporates a supportive market and policy framework to provide evidence-based pathways for widespread adoption and commercialization. The consortium ensures that the technical solutions created during the project are socially acceptable, ethically responsible, and contribute positively to human well-being. During WP3, identified user requirements ensure these social aspects are addressed.
During RP1, the project vision, standards, and user requirements were defined. The reference architecture of DATAWiSE was successfully designed, defining and structuring all system components. Data connectors and ETL pipelines were developed to integrate heterogeneous data sources from the pilots and ensure data harmonization. The first version of the Data Sharing Platform (DSP) was implemented and is currently operational, enabling secure and structured data exchange, dataset storage, cataloguing, and metadata management through a unified user interface.
Data analytics backend services were developed utilising AI techniques, with functionalities like data collection, storage and processing, The data storage component was developed and successfully deployed utilising open-source technologies, enabling storage and management of static and dynamic data as well as the communication with the DSP and the data bus. An intermediate data fusion component was developed serving as a processing pipeline for DATAWiSE services, providing homogenised data from raw sensor streams. The first version of the fusion tool is currently operational and able to handle processing requests from energy forecasting services. Also, the methodology for the Explainable AI framework was developed along with a model registry used by AI-based services to manage and control trained models.
The DBPM toolkit includes now a functional user interface with role-based access control. The BIM-Integrated Digital Twin was deployed as a hybrid system combining a node-based data processor with a web-based 3D viewer synchronizing building models and live sensor data. Electrical and thermal flexibility and forecasting services reached an operational level, with models trained and validated on pilot data. Initial versions of the Comfort Balancing and Climate Resilience services were also integrated.
Moreover, the LD2S toolkit, now features a functional user interface and secure access control, with the following components integrated . A key achievement was the delivery of the Circular Renovation Tool, which features a specialized lifecycle assessment logic to calculate circularity indicators and aggregate them into a Global Circularity Score for building materials. In the domain of risk assessment, the Predictive Maintenance service reached an operational state following the successful deployment of AI-based algorithms designed to determine optimal maintenance thresholds and automate the optimization of repair schedules. Furthermore, the Sustainability reporting and Smart Readiness Assessment (SRI) modules reached a functional level, with the sustainability module now employing a hybrid scoring system that combines automated data with qualitative assessments, and with the SRI tool being updated to enable baseline technological evaluations across all pilot sites.
Data analytics backend services were developed utilising AI techniques, with functionalities like data collection, storage and processing, The data storage component was developed and successfully deployed utilising open-source technologies, enabling storage and management of static and dynamic data as well as the communication with the DSP and the data bus. An intermediate data fusion component was developed serving as a processing pipeline for DATAWiSE services, providing homogenised data from raw sensor streams. The first version of the fusion tool is currently operational and able to handle processing requests from energy forecasting services. Also, the methodology for the Explainable AI framework was developed along with a model registry used by AI-based services to manage and control trained models.
The DBPM toolkit includes now a functional user interface with role-based access control. The BIM-Integrated Digital Twin was deployed as a hybrid system combining a node-based data processor with a web-based 3D viewer synchronizing building models and live sensor data. Electrical and thermal flexibility and forecasting services reached an operational level, with models trained and validated on pilot data. Initial versions of the Comfort Balancing and Climate Resilience services were also integrated.
Moreover, the LD2S toolkit, now features a functional user interface and secure access control, with the following components integrated . A key achievement was the delivery of the Circular Renovation Tool, which features a specialized lifecycle assessment logic to calculate circularity indicators and aggregate them into a Global Circularity Score for building materials. In the domain of risk assessment, the Predictive Maintenance service reached an operational state following the successful deployment of AI-based algorithms designed to determine optimal maintenance thresholds and automate the optimization of repair schedules. Furthermore, the Sustainability reporting and Smart Readiness Assessment (SRI) modules reached a functional level, with the sustainability module now employing a hybrid scoring system that combines automated data with qualitative assessments, and with the SRI tool being updated to enable baseline technological evaluations across all pilot sites.
"The DATAWiSE project has delivered results that advance the state of the art by transitioning from isolated building management toward a unified, lifecycle-oriented ecosystem. Some of the key results include:
• Integrated Data Management and Interoperability Framework: DATAWiSE has established a unified communication framework designed to facilitate seamless interaction between diverse hardware and software systems, including heterogeneous IoT sensors, meters, and building networks. By utilizing standardized data models and homogenization methods governed by the FAIR principles, the architecture ensures that disparate data streams are transformed into a common, high-quality format, eliminating ambiguity and enabling future re-use.
• Holistic, Inter-disciplinary Lifecycle Approach: Moving beyond tools that address isolated building phases, the project provides a comprehensive management framework. By integrating inter-disciplinary data throughout the building’s life, DATAWiSE enables evidence-based decision-making that considers the economic, ecological, and social impact of building assets.
• Human-Centric Design and Transparent AI: The project utilizes Explainable AI techniques integrated into user-friendly service dashboards. These systems provide transparent reasoning for energy forecasts and maintenance alerts, ensuring that complex AI insights are credible, intuitive, and easily actionable for both facility managers and residents.
The integration of these standardized and inter-disciplinary tools is expected to significantly enhance energy efficiency and material circularity within the EU building stock. "
• Integrated Data Management and Interoperability Framework: DATAWiSE has established a unified communication framework designed to facilitate seamless interaction between diverse hardware and software systems, including heterogeneous IoT sensors, meters, and building networks. By utilizing standardized data models and homogenization methods governed by the FAIR principles, the architecture ensures that disparate data streams are transformed into a common, high-quality format, eliminating ambiguity and enabling future re-use.
• Holistic, Inter-disciplinary Lifecycle Approach: Moving beyond tools that address isolated building phases, the project provides a comprehensive management framework. By integrating inter-disciplinary data throughout the building’s life, DATAWiSE enables evidence-based decision-making that considers the economic, ecological, and social impact of building assets.
• Human-Centric Design and Transparent AI: The project utilizes Explainable AI techniques integrated into user-friendly service dashboards. These systems provide transparent reasoning for energy forecasts and maintenance alerts, ensuring that complex AI insights are credible, intuitive, and easily actionable for both facility managers and residents.
The integration of these standardized and inter-disciplinary tools is expected to significantly enhance energy efficiency and material circularity within the EU building stock. "