User-centric knowledge representations based on ontology for AEC design collaboration

Fostering collaboration in the AEC (Architecture/Engineering/Construction) field is difficult, due to the differing educational and disciplinary backgrounds of the participants. Current approaches to managing such collaboration in the AEC industry often fail to overcome the disciplinary differences among the participants, resulting in cost overruns, missed schedules, and diminished satisfaction of the clients or society. Their failure is due to the lack of understanding of the nature of multi-disciplinary design and the lack of tools that can support them. The primary objective of this research is to establish a suitable model for machine-mediated collaboration. In contrast to the monolithic model, which is insensitive to changes, we propose to develop a distributed and flexible model, where each domain of expertise retains its own data in the form most appropriate for its needs, and where ontology-based, intelligent filters translate neutral design data into domain-specific ones. The filtered data appear semantically-rich to the participant, even when it was generated by another participant. To verify the feasibility of the proposed filter-based communication model, we developed and tested a prototype. The result of the prototype test demonstrates that the proposed model can enable designers from different disciplines participating in an AEC project to better understand the dynamic process of design and achieve a high level of shared understanding.

► We propose a filter-based communication model for AEC collaboration.
► User-centric representation is essential for multi-disciplinary design collaboration.
► Domain-specific ontologies are applied to filter and represent design information.
► The proposed model enables semantically-rich design representations.