Reaching the European Union’s 2050 climate targets depends on renovating a building stock that accounts for roughly 40% of EU energy use and 36% of energy-related greenhouse-gas emissions. Renovation is a staged, multi-year process that the Energy Performance Certificate, a single-point-in-time snapshot, cannot guide on its own. The recast Energy Performance of Buildings Directive answers this with two complementary instruments. The Digital Building Logbook is a repository for building data across the life cycle, and the Building Renovation Passport (BRP) is a stepwise renovation roadmap that draws on that data. Both depend on a data model that defines what building data is relevant and stores it in a reusable form. The data models proposed by existing BRP and logbook initiatives are mostly closed, or specified only as semi-structured Excel, so their coverage and completeness cannot be assessed and each new implementation starts its data modelling from scratch.

An open, formally specified geospatial data model for urban energy modelling already exists adjacent to this use case, the OGC City Geography Markup Language (CityGML) paired with its Energy Application Domain Extension (Energy ADE). Whether it is sufficient as a BRP data model has not been tested. This thesis tests it. It first synthesises a Minimum Set of Required Data (MSRD) from 14 existing and proposed European initiatives, distilling their recurring fields through an explicit relevance filter into a reference set of attributes a BRP needs. It then assesses how much of that MSRD the CityGML 2.0 + Energy ADE 3.0 (beta 8) data model can carry, mapping the set field by field, and tests the mapping at two scales on real Dutch building-stock data. The first is an audit-depth test on a single owner-occupier dwelling. The second is a city-scale breadth test on an entire municipality populated from open government sources (BAG, 3DBAG, EP-Online, and CBS statistics).

The MSRD comprises 276 fields, organised into 13 modules and 3 layers. The current CityGML + Energy ADE 3.0 (beta 8) data model covers roughly 95% of it, 262 of the 276 fields, with coverage complete for 8 of the 13 modules, including the entire assessment layer. The 14 uncovered fields are localised rather than spread across the model, most of them concentrated in the renovation-advice module, the staged roadmap that is the BRP’s defining feature, where coverage falls to 73%. The two-scale implementation confirms that the mapping holds under real data: the schema produced XSD-valid output from the single audited dwelling up to roughly 94,000 buildings without structural failure. Bringing the Dutch open-data sources together in one model also did more than confirm fit. It made the data-quality problems of that stock visible and tractable, exposing patterns that the source datasets, read apart, do not reveal, and it located the situations the model carries only partially.

The CityGML + Energy ADE pairing is therefore a meaningful, strong, and near-complete starting point for a BRP data model, with the gaps localised rather than structural. The contribution is to show that an open, formally specified data model already covers most of what a BRP needs, to name precisely where it does not, and to consolidate the targeted extensions that would close those gaps as input to the further development of the Energy ADE. The MSRD and the two-scale test pipeline are released openly.

https://doi.org/10.5281/zenodo.20669526

https://github.com/DaanSchlosser/CityGML2.0-EnergyADE3.0_creator

https://doi.org/10.4121/89f8909b-4473-4958-8f93-46b55546764d ... Read more