Abstract. The federation of models is a pillar of the BIM (Building Information Modelling) approach: it allows to keep the contents of each discipline separate during the modelling-creation phase, and to merge them together later during the project management phase, from its feasibility to the construction site, to the management of the entire life-cycle of the building.Generally, these models refer to specific disciplines and the architectural, plant and structural model are always identified.When the asset belongs to Cultural Heritage, more generically an existing building, the BIM approach (at this point HBIM - Historic Building Information Modelling) is faced with an additional level of complexity since it is necessary to model something existing (the building in fact) and of which not much information is known. The geometric complexity of the asset often aggravates this situation because if the parametric modelling is preferred, it is difficult to represent such irregular morphologies, and if the surface modelling and a more geometrical detail is preferred, the model becomes very heavy.In many cases the choice is to approximate reality as best as possible through specific and tailor-made modelling approaches, often complex and with some borderline methods, if compared to BIM logic. In other cases, however, it makes sense to define when the geometric complexity and the reliability of the model are necessary, and when, instead, a simplification is required in order to effectively manage the information.The case study presented in this paper is the Arch of Augustus, in Aosta, for the HBIM approach it has been chosen to separate the two approaches,placing side by side the federated models referred to the classical disciplines with two models of the current state: one very accurate with the purpose of maintaining all the quality of acquired 3D geometric information, while the other absolutely schematic, necessary as a 3D index for the information.The approach described here, however, requires a preliminary reflection in order to define the BIM granularity - the smallest object in the model - and to define the methodological procedures that allow the bidirectional relationship between survey model and conceptual model. This paper provides insigth oif the importance of the relation between survey model and conseptual model. Future works will try to make this relation more stronger and efficient.

Abstract. Geomatics' interest in Building Information Modelling (BIM) processes applied to Cultural Heritage is evolving in many directions. The traditional fields of Geomatics -data acquisition and processing- have been largely dealt with about HBIM (Historical Building Information Modelling) systems, but there are still some issues that need to be explored in more detail. In addition to modelling, it is essential to consider the information content of the model, how it is generated, recorded and managed.If the objective of the HBIM model is the conservation project, it becomes essential to understand what types of data can be inserted in the model, how to record them and, above all, what their purpose is. The HBIM approach did not start out as a simple data repository, but as an information tool with the aim of helping the designer at all stages of the construction process. Moving to the field of preservation, much information about the building is represented by thematic maps. They allow to have a graphical image of the state of conservation of a façade or to understand the structural situation of a building.This research, tested on the Arch of Augustus in Aosta, starts from the big amount of data acquired by the RAVA Laboratory of the Superintendency of Aosta during a long period. These data allowed to test different approaches to thematic mapping, according to the specific themes to be represented (previous restoration interventions, diagnostics, decay mapping, etc). Anyway, this experiment also required a theoretical reasoning that preceded the operational phases. Faced with a new system, in fact, it is always advisable to reason about the method applied, to avoid the error of simply translating a method that could instead be developed in new directions. The question, to which this article wants only to begin answering, concerns the role of thematic mappings in the preservation design made by a HBIM approach, their necessity and their implementation towards a truly three-dimensional data, which thus maintains all the information that is acquired directly in three dimensions and which instead, today, are reprojected in 2D or used as a placeholder in the 3D space.