Thanks to the ideas of the semantic web in the late 1990s and the efforts of organizations involved in the development of the IFC format, semantics and ontologies have become some of the key elements of the standardization being discussed in the construction industry by the mid-2020s.
Semantic technologies are unification, standardization and modification of large heterogeneous data sets and implementation of complex search.
To store semantic data we use OWL ontology language (Web Ontology Language), represented in the form of RDF graphs -triplets (Resource Description Framework) (Fig. 6.1-7). OWL refers to graph data models, the types of which we discussed in more detail in the chapter “Data models: data relations and relationships between elements”.
Theoretically, the logical inference of risoners (programs for automatic logical inference) allows new statements to be derived from ontologies. For example, if the building ontology records that “a foundation is a support for a wall” and “a wall is a support for a roof” (Fig. 6.1-7), the risoner is able to automatically infer that “a foundation is a support for a roof”.
Such a mechanism is useful for optimizing data analysis because it avoids explicitly prescribing all dependencies. However, it does not create new knowledge, but only identifies and structures already known facts.
Semantics does not create new meaning or knowledge per se and is not superior to other data storage and processing technologies in this aspect. Representing data from relational databases as triplets does not make them more meaningful. Replacing tables with graph structures may be useful for unifying data models, easy searching and secure editing, but it does not make the data “smarter” – the computer does not begin to understand its content better.
Logical relationships in data can be organized without complex semantic technologies (Fig. 6.1-8). Traditional relational databases (SQL) as well as CSV or XLSX formats allow building similar dependencies. For example, in a columnar database, you can add a “roof support” field and automatically associate the roof with the foundation when creating a wall. This approach is implemented without the use of RDF, OWL, graphs or risoners, remaining a simple and efficient solution for storing and analyzing data.
The decision of a number of large construction companies and the IFC format development organization (А. Boiko, “The post-BIM world. Transition to data and processes and whether the construction industry needs semantics, formats and interoperability,” December 20, 2024) to follow the semantic web concept, which seemed promising in the late 1990s, has had a significant impact on the development of standards in the construction industry.
However, the paradox is that the very concept of the semantic web, originally intended for the Internet, has not been widely adopted even in its native environment. Despite the development of RDF and OWL, the full-fledged semantic web has not appeared in its original conception, and its creation is already unlikely.
