2014, Beetz, Jakob, Lindlar, Michelle, Dietze, Stefan, Gadiraju, Ujwal, Edvardsen, Dag Field

In this deliverable, a framework for a Semantic Digital Archive (SDA) for buildings and their components is suggested and described: The organizational framework for the Semantic Digital Archive as well as its methodological and technological enablers are outlined and specified. Conceptual approaches for the creation, maintenance and use of mappings between the building-specific Industry Foundation Classes (IFC) schematic and instance models with established datasets and vocabularies (e.g. the ones offered by the Linked Open Data cloud) are provided. It is shown how such mappings can be used to semantically enrich Building Information Model instances and how such enrichments can be preserved independently of the continuous evolvement of the referenced data sets. In order to track such changes in external data sets and reduce the amount of storage needed in the SDA, the concept of a Semantic Digital Observatory (SDO) is introduced and discussed. Using concrete data sets with a high relevance for the building industry, and in particular the reference vocabulary of then bSDD examples are provided that demonstrate the mechanisms introduced here. During future activities of the DURAARK project, the conceptual approaches introduced here are implemented in software prototypes as a proof of concept solution for the long-term preservation of semantically enriched Building Information Models.

2014, Beetz, Jakob, Lindlar, Michelle, Dietze, Stefan, Gadiraju, Ujwal, Long, Qinqin

In this report strategies are outlined how relevant metadata for the digital preservation of architectural data and related external vocabularies can be structured and stored. Related requirements are identified in WP7, further specified in WP2 and documented in the deliverables D 2.2.1, D.2.2.2, D.2.2.3 (Requirements documents) and D.7.7.1. Metadata for data identified relevant within the scope of the DURAARK project will be captured in OWL/RDF metadata schemas that re-use existing and recognized metadata standards and only extend them where deemed necessary. The proposed domain-specific metadata structures cover three main aspects: Descriptive metadata that stores domain-specific summaries and classification information of buildings as an intellectual entity type in a LTP system. Technical metadata that captures information about Building Information Models (BIM’s) and point cloud models which are the two main representation forms of buildings the DURAARK project focuses on. Linked Data metadata that captures information about the vocabularies and datasets that are used for the semantic enrichment of building models. These are stored separately alongside the individual building model Information Packages (IP) in the archive in a Semantic Digital Archive (SDA) for the storage and a Semantic Digital Observatory (SDO) for the processing of such volatile, evolving data. The use of this metadata from IFC models and inference methods are described in generic vendor-neutral terms and exemplified as an extension implementation to the existing archival system. The suggested reuse of existing schemas and vocabularies is illustrated in a number of use cases in section 4. In the appendices of this document complete listings of the respective vocabularies and schemas are provided. The results of this document have been created in collaboration with the work and deliverables in other work packages and specifically with work packages 2 and 7.