(掃描建模) From - scan to BIM
(掃描建模) From - scan to BIM
Automation in Construction
journal homepage: www.elsevier.com/locate/autcon
A R T I C L E I N F O A B S T R A C T
Keywords: The progress in information technology allows an innovative transformation of practices commonly involved in
Built heritage the engineering and construction field, especially in relation to the existing architectural heritage’s control and
3D integrated survey management activities. The proposed methodology takes advantage of an integrated 3D metric survey as a basis
3D modelling
for an HBIM (Historic Building Information Modelling) model to be exploited for the definition of a Finite El-
Point cloud
HBIM
ements Model (FEM). This paper aims to show the applicability of a digital process, stemmed from the inte-
NURBS interoperability gration in Rhinoceros 3D of a BIM structural model, leading to the dynamic simulation of the analytical FEM
Dynamic simulation through PRO_SAP® (a PROfessional Structural Analysis Program). The described workflow investigates the
Finite elements model (FEM) interoperability issues, along with the difficulties in the Scan-to-HBIM processes, demonstrating how HBIM
Seismic analysis models can anyhow support operations aimed at maintaining and preserving existing historical assets, also from
a structural point of view, even if with still persistent criticalities.
* Corresponding author.
E-mail addresses: alessandro.grazzini@polito.it (A. Grazzini), francesca.matrone@polito.it (F. Matrone), marco.zerbinatti@polito.it (M. Zerbinatti).
https://doi.org/10.1016/j.autcon.2022.104518
Received 22 January 2021; Received in revised form 25 July 2022; Accepted 29 July 2022
Available online 3 August 2022
0926-5805/© 2022 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
A. Ursini et al. Automation in Construction 142 (2022) 104518
one to be used in a finite element structural calculation software such as variety of materials, such as wooden structures, were considered too
PRO_SAP®. By applying this innovative procedure, it is also possible to elaborated. In our case, PRO_SAP® proved to be a valid support for the
evaluate how to guarantee a sufficiently high level of detail, in order to linear dynamic analysis of the case study, complex and composed by
obtain more reliable results in terms of stresses and deformations during masonry, leading to significant results in terms of stability and safety.
the representation and simulation phases with tools external to Auto- Other works, such as [38] tries to exploit the object-oriented models
desk Revit. The whole proposed workflow has also been analysed to for preventive conservation considering damages and environmental
promote an optimisation of the coordination between the various actors parameters, but still no automatic (or semi-automatic) workflow is
and stakeholders participating in the different phases of intervention proposed. As described in [39], there are several works dealing with the
(maintenance, conservation, restoration) by implementing a replicable use of BIM models to monitor buildings for structural reinforcement,
and agile workflow that guarantees the correct interoperability and however, the authors point out that the main gaps remain: i) a low level
transfer of geometric information between the several software. of integration of the HBIM models for performance assessment and
Therefore, the research addresses the following issues: i) interopera- structural monitoring, ii) a high number of steps from the point clouds to
bility of HBIM models to be used as the basis for performance and the parametrised model and iii) difficult recognition and reconstruction
structural assessment, though operations of conversion and exchange of of curved historical geometries. Thus the proposed methodology tries to
information between various software; ii) digital representation of irreg- address these drawbacks, demonstrating a simpler workflow able to step
ular geometries within the HBIM models, tackled using NURBS through from the initial point cloud to the intermediate HBIM model to ensure
Rhinoceros and the GiD software; iii) overlong Scan-to-HBIM process even the final structural analyses. As further contributions of the paper, this
more complicated when it comes to structural purposes. In particular, by procedure also identified the structural state of the analysed building
exploiting the interoperability challenges, it has been possible to build a and additionally demonstrated the still present lack of interoperability
geometric model of multidisciplinary value, capable of interfacing with of the object-oriented software when dealing with historic buildings.
software of specific nature, such as those belonging to the structural
field. 1.2. The case study
1.1. Related works The Sacro Monte of Varallo was built at the end of the 15th century in
Valsesia, an alpine valley in north-western Italy. The research work was
The technological progress, as well as the development of Mobile applied to the Nazareth architectural complex (Fig. 1a) which includes
Mapping Systems (MMS), considerably speeded up the data acquisition, chapels 2 (“Annunciation”), 3 (“Visitation”) and 4 (“First Dream of Jo-
leading to the ever-increasing use of point clouds and to the efficient seph”), a set of structures now part of a knotty building, formed through
application of the BIM methodology also to historical buildings of cul- modifications and additions conducted over a long period. For the his-
tural interest [5–7]. Historic BIM [8–10], as we know it today, is a torical and constructive framework of this nucleus within the entire
fundamental methodology for the study, management and development complex of the Sacro Monte di Varallo, reference can be seen in [3,40].
of CH conservation practices, and its modelling process is strongly In particular, the construction of the Nazareth complex dates back a
supported by 3D metric surveys [11–16]. Indeed, the possibility of little later than 1514 and is a modestly sized system with simple
associating technical and historical information with three-dimensional morphological characters. The main entrance of the building consists of
geometries makes this methodology particularly suitable for applica- a vaulted room that leads to chapel 2, containing wooden statues
tions in the CH domain. depicting the “Annunciation to the Virgin Mary”. On its right, there is a
The growing use of HBIM models is certainly due to their versatility, portico (Fig. 1b), which protects chapel 3 (Fig. 1c) and from which a
they are used for different purposes such as fruition and virtual recon- staircase takes to chapel 4, a lower part of the building.
struction [17–19], as support for conservation, restoration and mainte- The chapels are characterised by separation from the external envi-
nance activities [20–24] or for analyses ranging from those with a high ronment only through perforated wooden or lead-welded gratings or by
level of detail, such as the archaeological ones [25,26], up to the ener- windows; this peculiarity makes them particularly vulnerable to
getic or on a territorial scale [27–29]. Although this methodology is degradation phenomena such as detachment of plasters, gaps in the wall
increasingly used, issues remain in the three-dimensional parametric faces, mould due to the high presence of humidity and biodeteriogenic
modelling of complex architectural elements [30] and, consequently, agents. In the specific case of the Nazareth complex, given the evident
this complicates its use for structural analyses. state of deterioration of some structures, extraordinary maintenance and
Nowadays the market offers the possibility to integrate BIM software restoration works of the roofing coverings were necessary (Fig. 1d), as
with special packages precisely designed for finite element analysis well as operations for the water treatment and its removal from the
(FEA), but the main limitation lies in the reduced ability of these plugins basement of the walls.
to interpret geometric complexity, usually typical of buildings of his- These structural weakenings irreparably cause a reduction in the
torical interest. This is mainly due to the selection of BIM technology performance level of the materials constituting the walls. This weak-
mostly aimed at designing and managing new buildings, characterised ening can lead to the anticipated crisis of the masonry walls due to the
by standardised technological solutions and simple formal canons, with exceeding resistance limits of the materials, under both the actions of the
repetitive architectural elements, hardly present in historical structure (own weight), the actions resulting from winds or heavy snow,
architecture. and those deriving from earthquakes or landslides. Precisely for this
In this context, the contribution given by point clouds, mainly used reason, an in-depth study and FEM analysis of the walls and vaulted
to support the HBIM modelling phases, has already provided aid for the surfaces are deemed necessary to understand their mechanisms of
semi-automatic generation of finite element models [31–33]. Never- structural resistance.
theless, we considered useful to propose a methodology that exploits the
geometries of HBIM models as a basis for FEAs. This choice is primarily 2. Methodology
due to the definition of the Italian UNI 11337 standard and the issuance
of the Ministerial Decree 560/2017, that will make BIM and HBIM The data exchange between professionals and stakeholders is of
models increasingly used in the near future. It is therefore advantageous crucial importance in the scenarios with interventions for the safety and
to be able to exploit them, even if without any point clouds. In this re- preservation of monumental historical assets, where it is essential to be
gard, some studies [34–37] have developed and tested methods for the able to build a finite element model from a three-dimensional geometric
use of HBIM models for structural analyses; however, the software used survey [1]. In many cases, the Italian technical legislation requires the
to verify the dynamic behaviour of structures consisting of a wide preliminary assessment of seismic safety, as an indispensable starting
2
A. Ursini et al.
Sample Automation in Construction 142 (2022) 104518
Fig. 1. The complex of Nazareth (a) with, in detail, the protective portico with starry vaults system (b), an interior view of chapel 3 “the Visitation” (c) and
conservation intervention with a new angular supporting beam beneath the original secondary beams (d).
point for planning the subsequent safety measures and seismic Standards for Italian Construction (NTC 2018) [44–46].
improvement of historic buildings, and this is precisely one of the main The exchange of information between the BIM model and the anal-
reasons behind the methodology proposed. ysis software was possible exclusively by using a Drawing Interchange
Starting from an accurate metric survey and through a scan-to-BIM Format (.dxf) extension since the Industry Foundation Classes (.ifc)
process, it was, therefore, possible to obtain an analytical simulation standard was not suitable for this case study due to an incorrect trans-
for structural purposes. mission and coding of data.
For the investigations (diagnostic and modelling) [1], the workflow
(Fig. 2) began with the integrated metric survey, namely the acquisition 2.1. Data acquisition: integrated metric survey
and processing of terrestrial laser scanner (TLS) and aerial photogram-
metric data [3]. Two different software was used to process the data As previously mentioned, to obtain a detailed integrated 3D metric
obtained, commercial and open-source, respectively FARO SCENE and survey of the Nazareth complex (Fig. 3), it was decided to use a com-
MicMac. By exploiting versatility and information interchange [41], it bination of several tools, intended for the realisation of different types of
was possible to generate a dense point cloud, which then constituted the acquisitions. First of all, the measurement of a topographic network was
real reference base of the entire modelling process in the object-oriented carried out for georeferencing the acquired data. The traditional topo-
environment: Autodesk Revit Architecture. The choice of this software graphic survey was realised with a Leica Nova Scan total station,
was due to different factors: through which a framing network consisting of six external vertices was
defined, with the further identification of two internal vertices essential
• the extensive use of this software among professionals and in uni- for the next side shots acquisition. Geomax GPS/GNSS receivers were
versity courses. In fact, although there are other several object- also used to acquire vertexes’ geographical position and to allow RTK
oriented software, Autodesk Revit is the one currently most used; (Real Time Kinematic) measurements for the other GCPs (Ground
• the case study proposed is part of a cross-border Italy-Switzerland Control Points).
inter regional project (Interreg "MAIN10ANCE"1), which involves the The geodetic network compensation was performed using the
creation of HBIM models for the whole system of the Sacri Monti and, STAR*NET software, and it was computed with error ellipses at 95% and
within this project, Revit was chosen as the modelling software; semiaxes of a maximum 2 cm (standard deviation ≤1 cm) (Fig. 4a).
• even by modelling with other software, such as Rhinoceros - or in any Afterwards, were acquired:
case in an external CAD - there is the possibility of creating a direct
link (bridge) of the models; therefore, the use of Revit, especially • 24 laser scans (Fig. 4b) with FARO Focus 3D by CAM2 (operating
under the informative aspect of BIM, allows to stratify information at range 0,6–130 m, ranging error ± 2 mm and vertical/horizontal
multiple levels of analysis. It should be noted that the parallel use Field of View 305◦ /360◦ )
and integration of Rhinoceros with Revit has been under develop- • 131 images (59 nadiral and 72 obliques) via Unmanned Aerial Ve-
ment for some years now. This has led to the creation of a new hicles (UAVs), precisely the DJI Phantom 4 Pro (with sensor size 1”
interchange system recently released (Rhino.Inside.Revit), which CMOS, 20 MP camera and focal length 8.8 mm) for the integration of
permits the direct connection between the two platforms. So the the external roofing data.
concept of modelling is no longer to be separated into a BIM model or
CAD model, but it goes in the direction of integrating these two 3D 2.2. Data processing
modelling paradigms.
Laser scans were registered with an average accuracy of ±5 mm for a
Having encountered some critical issues during the modelling phase, representation scale of 1:100. The same range of accuracy was chosen
the 3D digital object was subsequently integrated at a geometric level for the GCPs of the photogrammetric point cloud, where the maximum
through Rhinoceros 3D and the GiD 14.1.2d tool in order to convert the error on the GCPs was lower than 1 cm. The point cloud thus obtained
mesh into triangular and quadrangular meshes of a suitable size for the (Fig. 5a) turns out to be incomplete, especially in those portions of the
phase of FEA [42]. The FEA was performed with a linear dynamic building for which the laser scanner was not able to acquire the data: the
analysis using the PRO_SAP® software [43] in seismic combination and roofing parts. For this reason, the aero-photogrammetric RGB images
with the verification according to the requirements of the Technical were processed using the open-source MicMac software (Fig. 5b).
The two point clouds were easily integrated thanks to the use of
identical GCPs and targets within the same topographic network, which
1
https://main10ance.eu/ allowed to avoid manual or forced roto-translation operations. After
3
A. Ursini et al. Automation in Construction 142 (2022) 104518
Fig. 2. Representative diagram of the workflow and related software used in the different processing phases.
oi
that, cleaning, filtering and merging operations were carried out, 2.3. Conversion of geometry for HBIM
resulting in a point cloud of about 8 million total points (Fig. 5c). The
filtering operation to 1 cm (space-based method) was primarily neces- After the processing of point clouds, the creation of the HBIM model
sary to facilitate their management in the subsequent phases of model- aimed at the study and analysis of structural elements, started from the
ling in the object-oriented software, maintaining at the same time the choice of a “structural” modelling template. The modelling strategy did
accuracy required by the final representation scale. not give priority to a detailed representation of architectural elements to
The last step, before the geometrical definition, was a suitable dis- obtain a high level of fidelity but to the reproduction, as faithfully as
cretisation of the entire point cloud into topological sub-regions having possible, of only the strictly functional structural elements. For this
an extension readable by the Revit software. For this purpose, ReCap360 reason, the geometries of terrain (Fig. 6a), walls, wooden beams
PRO was used, through which it was possible to divide the point cloud (Fig. 6b), stone columns and separation walls have been reproduced,
into seven groups (surrounding context, chapels 2, 3 and 4, connecting leaving out the detailed representation of capitals, mouldings and
corridor, columns and a further indoor space anciently depicting the wooden or glassy diaphragms. Specifically, the terrain has been created
Jerusalem cave) and 26 sub-regions. They were then exported in .rcs through the “topographic surface” based on the point cloud .txt file; the
format, easily decoded by the BIM software and used as a reference for walls, roofs, columns, pillars, beams, stairs, floors and floors in contact
modelling single elements of the building. with the ground have been modelled using the predefined “system
families”; while the vaults through the “local model - Mass” from im-
ported CAD.
One of the main issues encountered in the modelling phase was to
4
A. Ursini et al. Automation in Construction 142 (2022) 104518
Fig. 4. Top views with the compensation of the geodetic network using STAR*NET software (highlighted the error ellipses enlarged by a scale factor of 30) (a) and
with the location of the laser scans (b).
Fig. 5. Results from laser scans (a) and UAV images (b) processing. From the integration of these two point clouds, we obtained an overall point cloud (c) sub-
sequently cleaned and subsampled.
Fig. 6. Overall isometric views of the HBIM model with highlighted the modelling of the topographic surface (a) and the structural components as beams and
columns (b) or the whole vaulted system imported from Rhinoceros (c).
5
A. Ursini et al. Automation in Construction 142 (2022) 104518
represent, only through the Revit software, the structural elements (such model. This structural analytical model consists of components that
as vaults) without oversimplifying them, with the consequent loss of contain all the properties of materials, loads and geometries, consti-
detail of the irregular parts. As already stated in the literature [47], the tuting the semantic part of the architectural model, and it usually ap-
modelling tools included in the object-oriented software were found to pears as a set of two-dimensional planar surfaces. Objects such as beams,
be inadequate when there is a need to represent geometrically complex pillars and columns are represented by one-dimensional elements,
elements. In our case, it was possible to test three different strategies for perfectly aligned with the central axis of the architectural reference
these elements: i) creating adaptive parametric families, not the best component (Fig. 8).
solution due to the high geometric irregularity of the elements; ii) using Nevertheless, the analytical model thus obtained is highly incom-
the “wall/floor by surface” command based on a specially modelled plete (i.e. lack of description of the vaulted systems or inaccuracy of the
“mass” element; iii) taking advantage from a third party software spe- nodes) and not sufficiently representative of the real architecture. This
cifically aimed at modelling any kind of geometries. The second case was significant gap demonstrates how the Revit software, mainly focused on
the closest to the intended purpose, but the issue was the inability of the the design of modern structures mostly based on bearing frame, cannot
Revit software to manage a vault as a stand-alone typological element, autonomously discretise geometrically complex structures, typical of the
namely a component not attributable to a geometric element with a CH domain, other than those consisting of linear or planar elements.
simple planar development but constituted by a three-dimensional Hence, an intrinsic limitation of the software is pointed out, not allowing
extension and having at the same time a structural function. static or dynamic analyses, impossible through plugins or specifically
To overcome this issue, the potential offered by Rhinoceros 5 soft- dedicated internal tools. The absence of these architectural elements in
ware was exploited, making it possible to delineate the profiles of the analytical model makes the latter useless within the BIM software.
complex geometries with vectors (Fig. 6c). Considered the case study and assuming the use of the PRO_SAP®
This software has proved to be very useful as it can be used both for structural software, there would be the possibility of using an IFC 2 × 3
Boolean modelling (operating with simple extrusions or rotations standard file import command. However, it was unfeasible: in fact, a
around generative axes) and, above all, for the generation of NURBS decoding error of the IFC 2 × 3 file was found by the structural software,
(Non-Uniform Rational Basis-Splines) [48,49]. With NURBS surface, a despite the fact that the exported file contained all the characteristics
mathematical-based representation, it is possible to accurately define all and information of the HBIM model created in Revit. This further
those geometrical entities with a complex shape, using lines and control operational limitation inevitably implies an adaptation of the simplified
points defined directly from the point clouds. It was, therefore, possible structural analytical model by exporting the geometry itself, in .dxf
to outline and reproduce all the useful profiles for the geometric defi- format, to implement it for the use in FEAs.
nition of the intrados of the vaults, with the relative edges generated by As shown in Fig. 9a, the .dxf file exported from Revit consists of mesh
the intersection of the individual ribs. From these profiles, it is possible surfaces, which is fully compatible with the PRO_SAP® software.
to generate NURBS surfaces passing perfectly through these polylines, Nevertheless, to obtain the final model (Fig. 9b) it was necessary to carry
obtaining a continuous geometric surface to be easily used for our out three main corrective operations:
purposes (Fig. 7).
Using this technique, we could faithfully and effectively reproduce • the first one was the geometric conversion of the vaults, previously
the state of the surfaces of all the vaulted portions, which are therefore modelled with NURBS, to obtain components consisting of a trian-
congruent with the reference point cloud. In order to import the geo- gular mesh with a side of at least 15 cm as the likely average size of
metric objects from the .sat file format to the Revit software, it was the vault ashlars. For this operation, the use of GiD 14.1.2d software,
necessary to add thickness to the surfaces by operating a vertical specifically dedicated to pre and post-production for finite element
extrusion of the entire group of NURBS, obtaining a geometric element numerical simulations in the engineering field, resulted beneficial
entirely representative of the vaulted elements. (Fig. 10). In particular, adjusting the mesh texture is essential to
correct the automation process provided by Rhinoceros software for
2.4. The analytical model for structural purposes converting NURBS surface to mesh. Rhinoceros has intrinsic criti-
calities at the mesh level: in the case of conversion from NURBS to
Selecting the structural model template in the initial steps, Revit al- triangular mesh, there is the risk of obtaining a mesh texture with
lows to automatically view, adjust and manage a simplified analytical non-homologous and coincident vertices, as happened here. For this
Fig. 7. Modelling phase of the vaults in Rhinoceros 5 through the identification of notable profiles (a) and the consequent generation of NURBS surfaces (b).
Representative detail of the performance of NURBS surfaces in relation to vault point cloud (c).
6
A. Ursini et al. Automation in Construction 142 (2022) 104518
Fig. 8. Isometric view of the analytical model automatically generated from the architectural one within Revit software.
Fig. 9. View of the analytical model obtained by Revit after importing into Rhinoceros (a) and implemented with the correction of the mesh (b).
Fig. 10. Comparison between the mesh obtained automatically by Rhinoceros (a) and the one obtained through the use of the GiD software (b). From the image it is
possible to see how the nodes of the single mesh faces have been made homologous in the second option and the trend itself has been corrected in order to become
denser in the points of maximum curvature and junction of the starting surfaces. This construction of the mesh obtained is thus sufficiently detailed for representative
purposes and sufficiently large not to generate possible errors in the FEM analysis phase.
reason, GiD software (widely used in the engineering field for the • in the next step, the intersection of the individual nodes, which must
mesh control process according to FEA analyses) has been chosen; have a unique coincidence to allow an appropriate load transfer in
the FEA analysis, was checked;
7
A. Ursini et al. Automation in Construction 142 (2022) 104518
8
A. Ursini et al. Automation in Construction 142 (2022) 104518
legislation, interpolating them with the physical characteristics of the dynamic limit state analysis considering 18 ways of vibrating, since
modelled elements by referring to homogeneous type load files deriving religious buildings do not have a regular geometry to allow us to
from geometry and materials. In this regard, through these automated hypothesise their perfect box-like behaviour. Below (Fig. 12) are pre-
procedures, it was possible to define a structure factor q, which takes sented the local deformation values, expressed in a chromatic scale from
into account the ductile response, equal to 1.875 for the X and Y di- blue (minimum value) to red (maximum value). ULS (Ultimate Limit
rections, and q = 1.5 in the Z direction, while a value of 5 was used for State) is here intended for the heavier combination of the sole static
the damping coefficient of 5%. actions prescribed by NTC18 (Fig. 12a). For the most significant ways of
Concerning the case study, it was decided to carry out a linear modal vibrating the seismic combinations proposed in the figure, however, it is
Fig. 12. Examples of results obtained in terms of deformation: maximum deformation corresponding to the static combination at Ultimate Limit State (ULS) (a). Most
significant deformations deriving from the modes of vibrations for the seismic combinations (b, c, d, e).
9
A. Ursini et al. Automation in Construction 142 (2022) 104518
possible to notice that there are cases of maximum deformation in the The result can be considered plausible as the wood has greater elastic
different modes of vibration that alternatively affect both the roof characteristics than the existing masonry; moreover, in the same point,
structure (Fig. 12d) and the upper part of the wall on which it leans during the survey campaign, an assembly error of the roof beams
(Fig. 12b, c and e). The maximum stress on the roof, obtained both from (committed during an extraordinary maintenance work carried out in
the ULS combination and from the seismic combinations, is consistent previous years) was identified. Therefore, this result indicates that, in
with the anomalous configuration of the main roof structure. This this portion of the building, it is necessary to go through a timely
covering system has been the subject of an extraordinary maintenance decisive intervention. By globally analysing the results obtained in terms
intervention, which changed the load-bearing elements according to a of deformation, it is possible to clearly and accurately identify the por-
non-optimal scheme, not respecting the previous disposition (Fig. 13d). tions of the building requiring consolidation operations (Fig. 13), to
This uncorrected structural intervention could lead to anomalous local avoid possible crisis phenomena in the vaults or the overturning
deformations, consistent with the results highlighted by the analysis of mechanisms found in some masonry walls.
the current state.
Regarding the constraint conditions, since it is an existing building 3.3. Structural verification
without foundation subsidence, starting from the assumption that the
soil-foundation interaction has been balanced for a long time, the base is The performance of the analytical model, simulated relatively to the
sufficient for the translations only, leaving the rotations free, as also characteristics of resistance and efficiency required by the Italian
required and recommended by the manual of the software. legislation, have been afterwards verified. For existing masonry build-
ings, as in our case, the assessment of the overall safety must be carried
3.2. Discussions on the results of the FE analyses out in relation to the collapse mechanisms, both in local and global
terms when the latter are significant. In this case, the verification was
As it can be seen from the previous images, the analytical model is conducted by referring to the data obtained through linear dynamic
aligned with the effective structural situation of the building, as it re- analysis. A reliable attribution of materials to the elements of the model
produces the accumulation of tension in the identical points where the plays a fundamental role for the success of the simulation, as well as the
real structure shows signs of failure. As regards the tensions, the most correct definition of the mechanical performances of the walls or their
stressed areas correspond to the supports of the floors and roofs, but also related specifications (e.g. materials composition).
the base of the principal masonry walls, as well as the points of junction Since from the survey phases, it was not possible to obtain suffi-
with the vaults: these results allow to deduce that the constraint con- ciently accurate data for the definition of the performance characteris-
ditions imposed at the base of the walls and, more generally, in the tics of the masonry, in the structural model the selection of materials
whole model return a likely behaviour of the entire complex structure. suitable for the real architecture was carried out through the PRO_SAP®
On the other hand, analysing the results in terms of deformation, it is default parameters: i.e. stone and mortar masonry was assumed for the
evident that in none of the reported cases, in combination of seismic load walls. Once the calculation was completed, it was possible to obtain a
both at the LLS and ULS, the maximum deformation value exceeds the series of results related to the various structural elements. The results of
order of magnitude of the centimetre. The maximum local deformation the assessments are presented (Fig. 14) in a very simple and intuitive
value of 0.95 cm is shown at the roof beams of chapel 4 with the com- way, based on a colour scale for every single D3 element: blue for ele-
bination at ULS A1 3 (Fig. 12a). The latter is the name of the heavier ments verified with another criterion; yellow for not designed parts;
combination given by PRO_SAP® when performing the analysis; it is a cyan for verified elements and red for not verified components.
numerical notation of the load combinations dictated by the Technical Analysing the images it is possible to notice that the structure has
Standards. widespread criticalities, showing numerous inefficiencies of the walls
Fig. 13. Comparison between analytical models and damages on wall structures on Chapel 2 (fractures on stone masonries and vaults).
10
A. Ursini et al. Automation in Construction 142 (2022) 104518
Fig. 14. Results in chromatic scale of the verifications performed only on the masonry components by analysing out-of-plane, in-plane and shear bending (chapter
7.8.2.2.3 NTC 2018) at 10% of Peak Ground Acceleration (PGA). The portions of mesh marked in red represent the parts of the structure that have not passed the
resistance test and for which it is necessary to carry out analyses and local improvement interventions. (For interpretation of the references to colour in this figure
legend, the reader is referred to the web version of this article.)
even with a ground acceleration value equal to 10% of the design value with local analyses and punctual investigations, to define more accu-
(no vault has been verified for the reasons described in paragraph 3.2. rately the constitutive and performance characteristics of the materials
Furthermore, considering the results, every single chapel would prob- and their mechanical behaviour, obtaining a better overall confidence
ably suffer serious damages, as it can be expected for a structure of this factor of the masonry members.
historical period and construction technique. There is, in fact, a con-
centration of unverified elements, especially in the portions of the 4. Conclusions
building with strong geometric discontinuity and in the upper and lower
bands of the openings. Through operations of selection, conversion and exchange of infor-
Broadly speaking, the positive outcome of this assessment confirms mation between the various software, it has been possible to generate a
and consolidates the usefulness of an operating procedure through 3D parametric model, exploiting the HBIM modelling technologies with
which it is possible to: the Autodesk Revit Architecture software. In this way, it was also
possible to combine all the information of geometric nature, with all the
• identify and precisely define all the architectural and structural el- material and typological features detected visually and rapidly during
ements that must be verified, the survey campaign. The intertwining of these aspects has allowed the
• establish which ones may be subject to structural improvement in reproduction in a virtual environment of the building, obtaining an
future maintenance plans and seismic safety assessments. adequate, versatile and functional result for all future operations.
Through precise conversion steps, the three-dimensional object pro-
However, it must be pointed out that, before any intervention, it is cessed with the BIM methodology was converted into a simplified
necessary to increase the level of knowledge, operating on the building analytical model, in order to exploit it with structural purposes, using
11
A. Ursini et al. Automation in Construction 142 (2022) 104518
the PRO_SAP® software. Through these analyses, it was possible to Data availability
deduce the current performance characteristics, useful for setting the
basis of any consolidation or structural improvement interventions in Data will be made available on request.
support of a more suitable safeguard of the architectural asset (with a
consequent step towards higher levels of safety). In addition, the ana- Acknowledgements
lyses were not performed on a dedicated workstation, but on a personal
notebook (Asus N751JX-T4023H with Intel® Core ™ i7-4720HQ pro- The authors would like to acknowledge the MAIN10ANCE project
cessor and 16GB DDR3L at 1600 MHz RAM), taking approximately 45/ (ID 473472) fundend by the European Regional Development Fund
60 min of processing by PRO_SAP® software alone. This element con- (ERDF), Axis 2 - Enhancement of the natural and cultural heritage. In
firms the potential of the proposed methodology, given that even with a addition, they would like to thank Dr. Elena De Filippis, Past Director of
non-performing PC the calculation times are reduced and easily the Management Body of Sacri Monti, as well as the personnel of the site
manageable. and the professor Andrea Lingua that allowed and supervised the
Despite the strong technological innovation and considering the surveying activities.
growing need for data sharing between the various professional figures
and stakeholders involved, some issues still remain in the proposed References
workflow:
[1] A. Grazzini, F. Chiabrando, S. Foti, G. Sammartano, A. Spanò, A multidisciplinary
study on the seismic vulnerability of St. Agostino church in Amatrice following the
• very long post-processing times compared to relatively short data 2016 seismic sequence, Int J. Architect. Heritage (2019) 1–18, https://doi.org/
acquisition times; 10.1080/15583058.2019.1575929.
• lack of precision in some of the modelling operations in the BIM [2] A. Grazzini, S. Fasana, M. Zerbinatti, G. Lacidogna, Non-destructive tests for
damage evaluation of stone columns: the case study of Sacro Monte in Ghiffa
environment; (Italy), Appl. Sci. 10 (2020) 1–15, https://doi.org/10.3390/app10082673.
• inconsistences and inaccuracies in the phases of interchange between [3] M. Zerbinatti, F. Matrone, A. Lingua, Planned maintenance for architectural
BIM software and FEA software, with no interchange via IFC format; heritage. Experiences in progress from 3D survey to intervention programmes
through HBIM, TEMA – Technol. Eng. Mater. Architect. 7 (1) (2021) 32–42,
• gaps of Autodesk Revit in the representation of historical masonries. https://doi.org/10.30682/tema0701d.
The software is unable to autonomously discretise geometrically [4] A. Grazzini, G. Lacidogna, S. Valente, F. Accornero, Delamination of plasters
complex structures except for those constituted by linear or planar applied to historical masonry walls: analysis by acoustic emission technique and
numerical model, in: IOP Conf. Series: Material Science and Engineering 372,
elements.
2018, pp. 1–7, https://doi.org/10.1088/1757-899X/372/1/012022.
[5] F. Chiabrando, G. Sammartano, A. Spanò, A. Spreafico, Hybrid 3D models: when
The lack of recognition of complex elements is a strong limitation for geomatics innovations meet extensive built heritage complexes, ISPRS Int. J. Geo
the HBIM models, especially for structural safety assessments, as any Inf. 8 (3) (2019) 124, https://doi.org/10.3390/ijgi8030124.
[6] G. Tucci, V. Bonora, A. Conti, L. Fiorini, Digital workflow for the acquisition and
kind of vault in historical architecture is, at the same time, both a spatial elaboration of 3D data in a monumental complex: the fortress of Saint John the
architectural element and a structural element with function of con- Baptist in Florence, in: ISPRS - International Archives of Photogrammetry Remote
necting or transferring of loads from horizontal surfaces to walls or Sensing and Spatial Information Sciences XLII-2/W5, 2017, pp. 679–686.
[7] G. Bitelli, M. Dellapasqua, V.A. Girelli, E. Sanchini, M.A. Tini, 3D geomatics
columns. techniques for an integrated approach to cultural heritage knowledge: the case of
In spite of these issues and the involvement of numerous software, it san Michele in Acerboli’s Church in Santarcangelo di Romagna, in: ISPRS
was demonstrated how to recreate from the point cloud not only an -International Archives of Photogrammetry Remote Sensing and Spatial
Information Sciences XLII-5/W1, 2017, pp. 291–296, https://doi.org/10.5194/
HBIM model but also the subsequent structural analytical model, isprs-archives-XLII-5-W1-291-2017.
representative of the current state of the Nazareth complex, exploiting [8] M. Murphy, E. McGovern, S. Pavia, Historic building information modelling
the HBIM itself. With this process, it was thus possible to faithfully (HBIM), Struct. Surv. 27 (4) (2009) 311–327, https://doi.org/10.1108/
02630800910985108.
reproduce the complex geometries (in particular vaults and roofs) [9] R. Volk, J. Stengel, F. Schultmann, Building information models (BIM) for existing
constituting the main criticality of today’s three-dimensional represen- buildings – literature review and future needs, Autom. Constr. 38 (2014) 109–127,
tation techniques. Furthermore, by exploiting the software great data https://doi.org/10.1016/j.autcon.2013.10.023.
[10] C. Dore, M. Murphy, Current state of the art historic building information
processing capabilities, the attribution to the geometric model and
modelling, in: ISPRS - International Archives of Photogrammetry Remote Sensing
evaluation (as a hypothesis) of the mechanical and dynamic nature of and Spatial Information Sciences XLII-2/W5, 2017, pp. 185–192, https://doi.org/
the materials was also carried out, obtaining very useful and reliable 10.5194/isprs-archives-XLII-2-W5-185-2017.
results based on the state of the building and the boundary conditions. [11] R. Brumana, A. Georgopoulos, D. Oreni, A. Raimondi, A. Bregianni, HBIM for
documentation, dissemination and management of built heritage. The case study of
In conclusion, it is therefore possible to state that the proposed St. Maria in Scaria d’Intelvi, Int. J. Heritage Digital Era 2 (3) (2013) 433–451,
workflow offers a starting point for interesting applications, not only https://doi.org/10.1260/2047-4970.2.3.433.
strictly related to the structural context but also, and above all, in [12] R. Quattrini, E.S. Malinverni, P. Clini, R. Nespeca, Orlietti, From TLS to HBIM.
High quality semantically-aware 3D modeling of complex architecture, in: ISPRS -
relation to the management, conservation and maintenance of the International Archives of Photogrammetry Remote Sensing and Spatial Information
existing architectural heritage, through workflows deriving from HBIM Sciences, Avila, Spain, XL-5/W4, 2015, pp. 367–374, https://doi.org/10.5194/
methodologies. In fact, with these resources, the research group has isprsarchives-XL-5-W4-367-2015.
[13] F. Chiabrando, G. Sammartano, A.T. Spanó, Historical building models and their
defined a programmed maintenance plan (in synergy with the Sacri handling via 3D survey: From points clouds to user-oriented HBIM, in: ISPRS -
Monti Management Body) within the Italy-Switzerland Interreg International Archives of Photogrammetry Remote Sensing and Spatial Information
“MAIN10ANCE” Project and it is currently implementing it. Sciences, XLI-B5, 2016, pp. 633–640, https://doi.org/10.5194/ISPRS-ARCHIVES-
XLI-B5-633-2016.
A future perspective of this work is the implementation of a dynamic [14] L. Inzerillo, M. Lo Turco, S. Parrinello, C. Santagati, G.M. Valenti, BIM e beni
operational analysis of the building to confirm the results of our modal architettonici: verso una metodologia operativa per la conoscenza e gestione del
analysis. patrimonio culturale (BIM and architectural heritage: towards an operational
methodology for the knowledge and management of cultural heritage), Disegnare
Con, Le dimensioni del BIM 9 (16) (2016) 161–169 (ISSN 1828-5961).
Declaration of Competing Interest [15] L. Fregonese, L. Taffurelli, A. Adami, S. Chiarini, S. Cremonesi, J. Helder,
A. Spezzoni, Survey and modelling for the BIM of Basilica of San Marco in Venice,
The authors declare that they have no known competing financial 2017, in: ISPRS - International Archives of Photogrammetry Remote Sensing and
Spatial Information Sciences, XLII-2/W3, 2017, pp. 303–310, https://doi.org/
interests or personal relationships that could have appeared to influence 10.5194/isprs-archives-XLII-2-W3–303-2017.
the work reported in this paper. [16] F. Banfi, L. Chow, M. Reina Ortiz, C. Ouimet, S. Fai, Building information modeling
for cultural heritage: The management of generative process for complex historical
buildings, in: M. Ioannides (Ed.), Digital Cultural Heritage. Lecture Notes in
12
A. Ursini et al. Automation in Construction 142 (2022) 104518
Computer Science vol. 10605, Springer, 2018, pp. 119–130, https://doi.org/ [34] L. Barazzetti, F. Banfi, R. Brumana, G. Gusmeroli, D. Oreni, M. Previtali,
10.1007/978-3-319-75826-8_10. F. Roncoroni, G. Schiantarelli, BIM from laser clouds and finite element analysis:
[17] P. Martínez-Carricondo, F. Carvajal-Ramírez, L. Yero-Paneque, F. Agüera-Vega, Combining structural analysis and geometric complexity, in: ISPRS - International
Combination of nadiral and oblique UAV photogrammetry and HBIM for the Archives of Photogrammetry Remote Sensing and Spatial Information Sciences, XL-
virtual reconstruction of cultural heritage. Case study of Cortijo del Fraile in Níjar, 5/W4, 2015, pp. 345–350, https://doi.org/10.5194/isprsarchives-XL-5-W4-345-
Almería (Spain), Build. Res. Inf. 48 (2) (2020) 140–159, https://doi.org/10.1080/ 2015.
09613218.2019.1626213. [35] L. Barazzetti, F. Banfi, R. Brumana, G. Gusmeroli, M. Previtali, G. Schiantarelli,
[18] A. Osello, G. Lucibello, F. Morgagni, HBIM and virtual tools: a new chance to Cloud-to-BIM-to-FEM: structural simulation with accurate historic BIM from laser
preserve architectural heritage, Buildings 8 (1) (2018) 12, https://doi.org/ scans, Simul. Model. Pract. Theory 57 (2015) 71–87, https://doi.org/10.1016/j.
10.3390/buildings8010012. simpat.2015.06.004.
[19] F. Banfi, R. Brumana, C. Stanga, A., Content-based immersive experience of [36] C. Dore, M. Murphy, S. McCarthy, F. Brechin, C. Casidy, E. Dirix, Structural
Basilica of Sant’Ambrogio in Milan: From 3D survey to virtual reality, in: ISPRS - simulations and conservation analysis -historic building information model
International Archives of Photogrammetry Remote Sensing and Spatial Information (HBIM), in: ISPRS - International Archives of Photogrammetry Remote Sensing and
Sciences, XLII-2/W11, 2019, pp. 159–166, https://doi.org/10.5194/isprs-archives- Spatial Information Sciences, XL-5/W4, 2015, pp. 351–357, https://doi.org/
XLII-2-W11-159-2019. 10.5194/isprs-archives-XLII-5-W1-337-2017.
[20] N. Bruno, R. Roncella, A restoration oriented HBIM system for cultural heritage [37] M. Pepe, D. Costantino, A. Restuccia Garofalo, An efficient pipeline to obtain 3D
documentation: the case study of Parma cathedral, in: ISPRS - International model for HBIM and structural analysis purposes from 3D point clouds, Appl. Sci.
Archives of Photogrammetry Remote Sensing and Spatial Information Sciences, 10 (2020) 1235, https://doi.org/10.3390/app10041235.
XLII-2, 2018, pp. 171–178, https://doi.org/10.5194/isprs-archives-XLII-2-171- [38] S. Bruno, M. De Fino, F. Fatiguso, Historic building information modelling:
2018. performance assessment for diagnosis-aided information modelling and
[21] D. Oreni, R. Brumana, B. Cuca, A. Georgopoulos, HBIM for conservation and management, Autom. Constr. 86 (2018) 256–276. ISSN 0926-5805, https://doi.org
management of built heritage: Towards a library of vaults and wooden beam floors, /10.1016/j.autcon.2017.11.009.
in: CIPA 2013 XXV International Symposium, ISPRS - Annals of Photogrammetry [39] R. Mora, L.J. Sánchez-Aparicio, M.Á. Maté-González, J. García-Álvarez,
Remote Sensing and Spatial Information Sciences vol. 164, 2013, pp. 1–6, https:// M. Sánchez-Aparicio, D. González-Aguilera, An historical building information
doi.org/10.5194/isprsannals-II-5-W1-215-2013. modelling approach for the preventive conservation of historical constructions:
[22] D. Oreni, R. Brumana, S. Della Torre, F. Banfi, L. Barazzetti, M. Previtali, Survey application to the historical library of Salamanca, Autom. Constr. 121 (2021) 1–22,
turned into HBIM: the restoration and the work involved concerning the Basilica di 103449. ISSN 0926-5805, https://doi.org/10.1016/j.autcon.2020.103449.
Collemaggio after the earthquake (L’Aquila), ISPRS - Annal Photogram. Remote [40] E. De Filippis, Guida del Sacro Monte di Varallo (Guide to the Sacro Monte of
Sens. Spat. Inform. Sci. II-5 (2014) 267–273, https://doi.org/10.5194/isprsannals- Varallo), Tipolitografia di Borgosesia, Borgosesia, 2009.
II-5-267-2014. [41] D. Oreni, R. Brumana, F. Banfi, L. Bertola, L. Barazzetti, B. Cuca, M. Previtali,
[23] F. Chiabrando, M. Lo Turco, F. Rinaudo, Modeling the decay in an HBIM starting F. Roncoroni, Beyond crude 3D models: from point clouds to historical building
from 3D point clouds. A followed approach for cultural heritage knowledge, in: information modeling via NURBS, in: Digital Heritage. Progress in Cultural
ISPRS - International Archives of Photogrammetry Remote Sensing and Spatial Heritage: Documentation, Preservation, and Protection vol. 8740, Springer
Information Sciences, XLII-2/W5, 2017, pp. 605–612, https://doi.org/10.5194/ International Publishing, 2014, pp. 166–175, https://doi.org/10.1007/978-3-319-
isprs-archives-XLII-2-W5-605-2017. 13695-0_16.
[24] A. Baik, R. Yaagoubi, J. Boehm, Integration of Jeddah Historical BIM and 3D GIS [42] O.C. Zienkiewicz, The Finite Element Method, Engineering Science, McGraw-Hill,
for documentation and restoration of historical monument, in: ISPRS - London, 1971. ISBN-13: 978–0070941380.
International Archives of Photogrammetry Remote Sensing and Spatial Information [43] PRO_SAP® Instruction Manual, 2Si, Ferrara (Italy). Last access: 7 December 2020,
Sciences, XL-5/W7, 2015, pp. 29–34, https://doi.org/10.5194/isprsarchives-XL-5- Available at, https://www.2si.it/it/download-manuali-ed-esempi/, 2018. Last
W7-29-2015. Access: 20th November 2020.
[25] I. Trizio, F. Savini, A. Giannangeli, R. Boccabella, G. Petrucci, The archaeological [44] D.M. 17/01/2018, Norme Tecniche per le Costruzioni (Technical Standards for
analysis of masonry for the restoration project in HBIM, in: ISPRS - International Construction) (G.U. n. 42 del 20-2-2018), Italian Ministry of Infrastructure and
Archives of Photogrammetry Remote Sensing and Spatial Information Sciences, Transport, Rome, Italy, 2018. Available at, https://www.gazzettaufficiale.it/eli
XLII-2/W9, 2019, pp. 715–722, https://doi.org/10.5194/isprs-archives-XLII-2-W9- /gu/2018/02/20/42/so/8/sg/pdf. Last access: 20th November 2020.
715-2019. [45] Circular 21st Janaury 2019 n. 7 “Istruzioni per l’applicazione dell’’Aggiornamento
[26] V. Bagnolo, R. Argiolas, A. Cuccu, HBIM for archaeological sites: from SfM based delle «Nuove norme tecniche per le costruzioni»” (Instructions for the application
survey to algorithmic modeling, in: ISPRS - International Archives of of the Update of the «New technical standards for constructions»), Ministerial
Photogrammetry Remote Sensing and Spatial Information Sciences, XLII-2/W9, Decree 17th January 2018, (G.U. n. 35 of 11-2-2019 - Suppl. Ordinario n.5), Italian
2019, pp. 57–63, https://doi.org/10.5194/isprs-archives-XLII-2-W9-57-2019. Ministry of Infrastructure and Transport, Rome, Italy, 2019. Available at, https://
[27] L. Diaz-Vilariño, P. Boguslawski, M. Azenha, L. Mahdjoubi, P.B. Lourenco, P. Arias, www.cngeologi.it/wp-content/uploads/2019/02/Circolare-21-gennaio-2019-n.-7.
From LiDAR data towards HBIM for energy analysis, in: Y. Arayici, J. Counsell, pdf. Last access: 20th November 2020.
L. Mahdjoubi, G.A. Nagy, S. Hawas, K. Dweidar (Eds.), Heritage Building [46] DPCM (Decree of the President of the Council of Ministers) 9th February 2011
Information Modelling, Routledge Taylor & Francis Group, 2017, pp. 218–223. “Valutazione e riduzione del rischio sismico del patrimonio culturale con
ISBN 9781138645684. riferimento alle norme tecniche per le costruzioni di cui al decreto ministeriale 14
[28] F. Matrone, E. Colucci, V. De Ruvo, A. Lingua, A. Spanò, HBIM in a semantic 3D gennaio 2008” (Assessment and reduction of the seismic risk of cultural heritage
GIS database, in: ISPRS - International Archives of Photogrammetry Remote with reference to the technical standards for buildings referred to in the Ministerial
Sensing and Spatial Information Sciences, XLII-2/W11, 2019, pp. 857–865, Decree January 14, 2008), Rome, Italy, Available at, https://www.veneto.benicult
https://doi.org/10.5194/isprs-archives-XLII-2-W11-857-2019. urali.it/sites/default/files/Valutazione%20rischio%20sismico%20-%20Direttiva%
[29] E. Colucci, V. De Ruvo, A. Lingua, F. Matrone, G. Rizzo, HBIM-GIS integration: 20PCM%209%20febbraio%202011.pdf, 2011. Last access: 22 November 2020.
from IFC to CityGML standard for damaged cultural heritage in a multiscale 3D [47] F. Banfi, The integration of a scan-to-HBIM process in BIM application: The
GIS, Appl. Sci. 10 (4) (2020) 1356, https://doi.org/10.3390/app10041356. development of an add-in to guide users in Autodesk Revit, in: ISPRS -
[30] S. Fai, K. Graham, T. Duckworth, N. Wood, R. Attar, Building Information International Archives of Photogrammetry Remote Sensing and Spatial Information
Modeling and Heritage Documentation. CIPA 2011 Conference Proceedings: Sciences, XLII-2/W11, 2019, pp. 141–148, https://doi.org/10.5194/isprs-archives-
XXIIIrd International CIPA Symposium, 2011, p. 8. Available at, https://www. XLII-2-W11-141-2019.
autodesk.com/research/publications/building-information-modeling-and. Last [48] F. Banfi, HBIM generation: Extending geometric primitives and BIM; modelling
access: 22nd July 2022. tools for heritage structures and complex vaulted systems, in: ISPRS - International
[31] G. Castellazzi, A.M. D’Altri, G. Bitelli, I. Selvaggi, A. Lambertini, From laser Archives of Photogrammetry Remote Sensing and Spatial Information Sciences,
scanning to finite element analysis of complex buildings by using a semi-automatic XLII-2/W15, 2019, pp. 139–148, https://doi.org/10.5194/isprs-archives-XLII-2-
procedure, Sensors 15 (2015) 18360–18380, https://doi.org/10.3390/ W15–139-2019.
s150818360. [49] L. Barazzetti, F. Banfi, R. Brumana, M. Previtali, Creation of parametric BIM objects
[32] G. Castellazzi, A.M. D’Altri, S. de Miranda, F. Ubertini, An innovative numerical from point clouds using NURBS, Photogramm. Rec. 30 (152) (2015) 339–362,
modeling strategy for the structural analysis of historical monumental buildings, https://doi.org/10.1111/phor.12122.
Eng. Struct. 132 (2017) 229–248, https://doi.org/10.1016/j. [50] A. Grazzini, M. Zerbinatti, Static and seismic behaviour of different dry stone wall
engstruct.2016.11.032. textures, in: G. Milani, A. Taliercio, S. Garrity (Eds.), Proc. of 10th International
[33] G. Bitelli, G. Castellazzi, A.M. D’Altri, S. de Miranda, A. Lambertini, I. Selvaggi, On Masonry Conference, Milan, Italy, The International Masonry Society (IMS),
the generation of numerical models from point clouds for the analysis of damaged Whyteleafe (UK), 9–11 July 2018, pp. 1088–1102 (ISSN 2523-532X).
cultural heritage, IOP Conf. Ser. Mater. Sci. Eng. 364 (2018) 1–8, https://doi.org/
10.1088/1757-899X/364/1/012083.
13