<p>The contemporary evolution of the concept of the model in architecture and engineering has shifted the model from a representational artifact to a cognitive infrastructure that supports knowledge organization and decision-making across the life cycle of the built environment. This paper combines a theoretical analysis of the epistemic role of the model with a methodological framework that integrates parametric modeling and Building Information Modeling (BIM) as a semantic and relational environment. The modeling and inspection of existing bridges are used as a reference domain due to the scale of infrastructure inventories, typological recurrence, and the need for continuous updating of inspection data. A geometric–relational scripting approach is presented for the large-scale generation of parametric bridge models, based on the integration of recurring inspection datasets within an openBIM workflow. The method produces interoperable and queryable models in which geometry and information are structured into a coherent system of entities, relationships, and properties, enabling reuse across heterogeneous platforms. The results show that the integration of parametric modeling and openBIM is not only an optimization of modeling and update processes, but also a mechanism for the formalization, management, and transmission of infrastructural knowledge. The digital model therefore becomes a shared environment in which technical, past, and operational information can be continuously constructed, verified, and updated.</p>