See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/317901689

Robust IFC Files to Improve Information Exchange: An Application for Thermal

Energy Simulation

Conference Paper · July 2017

DOI: 10.14455/ISEC.res.2017.8

CITATIONS READS
3 233

4 authors, including:

Qian Chen Yunus Emre Harmanci

ETH Zurich ETH Zurich
7 PUBLICATIONS   35 CITATIONS    19 PUBLICATIONS   91 CITATIONS   

SEE PROFILE SEE PROFILE

Borja García de Soto

New York University Abu Dhabi
57 PUBLICATIONS   254 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Frontiers in Construction Management View project

Spatiotemporal Modelling for Construction Project Management View project

All content following this page was uploaded by Borja García de Soto on 05 July 2017.

The user has requested enhancement of the downloaded file.

 Resilient Structures and Sustainable Construction
Edited by Pellicer, E., Adam, J. M., Yepes, V., Singh, A., and Yazdani, S.
Copyright © 2017 ISEC Press
ISBN: 978-0-9960437-4-8

ROBUST IFC FILES TO IMPROVE INFORMATION

EXCHANGE: AN APPLICATION FOR THERMAL
ENERGY SIMULATION
QIAN CHEN1, YUNUS EMRE HARMANCI2, YAOWEN OU2,
and BORJA GARCIA DE SOTO1
1
Dept of Civil, Environmental & Geomatic Engineering, Institute of Construction and
Infrastructure Management, ETH Zürich, Switzerland
2
Dept of Civil, Environmental & Geomatic Engineering, Institute of Structural Engineering,
ETH Zürich, Switzerland

Despite many efforts from software vendors, AEC community and researchers,
interoperability is still one of the main issues regarding reliable and robust transfer of
information among different applications. In most cases, the Industry Foundation
Classes (IFC) files fail to provide proper interoperability between geometric building
models (architects) and thermal simulation software (engineers). This causes time
consuming interactions and manual corrections prompt to errors. This paper evaluated
two approaches for an efficient and robust transfer of IFC models considering space
boundary characteristics to conduct thermal energy simulation (TES). The first
approach was a multi-platform process which IFC files could be used by different TES
tools. The second consisted of a single-platform process in which a single CAD
software with built-in energy simulation capabilities was used. The two processes were
tested with a simple residential building. Results indicated that the first process still
required manual corrections and its performance was influenced by the TES tool used.
The second approach addressed the interoperability problems, but caused “software
dependency”. It was found that geometry data reflecting different levels of space
boundaries significantly influenced energy simulation results, indicating that proper
definition of space boundaries improved the robustness of IFC files. This showed that
IFC files can be enhanced to facilitate TES. This study also showed opportunities for
improvement regarding interoperability and suggested other ways to tackle this
problem.
Keywords: Space boundaries, Interoperability, Industry foundation class, Building
information model, Building energy model.

1 INTRODUCTION
Attempting to optimize the building design from the energy saving point of view can be seen as
the ultimate goal of conducting a thermal energy simulation (TES), which refers to the simulation
of the internal energy present in a building due to temperature variations. The state-of-the-art
approaches to realize TES rely on the integration of parametric Building Information Modeling
(BIM) and energy simulation engines through standard data exchange formats (El Asmi et al.
2015). For instance, the energy model that is imported into Simergy (i.e., an energy simulation
engine) for TES purposes is generated from Autodesk Revit (a BIM authoring tool) in the form of
an IFC file (i.e., an exchange format). Although numerous BIM-based tools and processes have

1
Pellicer, E., Adam, J. M., Yepes, V., Singh, A., and Yazdani, S. (eds.)

been developed for TES to be used during the design phase (Ahn et al. 2014), a robust
information transfer of required information to deliver accurate TES results through standard data
schemes is still proved to be inefficient and not intuitive. Incongruent information and a lack of
rule-based information translation or interoperability also cause the TES processes to encounter
iteratively manual model checks and modifications (Wimmer et al. 2015). Therefore, a robust
and reliable translation process of building information is required for improving the TES results.
Several achievements have been made to refine the information exchange interface in order to
reduce interoperability problems. One major solution is to overcome the mismatches between the
BIM library of architectural design and the TES-required energy modeling information, that is, to
transform the spatial geometry into thermal geometry with a robust exchange format (Eastman et
al. 2011). A set of data requirements for TES has been developed to improve the quality of
geometry models through space boundary surfaces (Maile et al. 2013). Furthermore, a physical
BIM library has been investigated to perform semi-automatic translation from the building design
models to TES engines using BIM authoring tool’s application programming interface and object-
oriented physical models (Kim et al. 2015). However, the results by these attempts imply that
there is still a lack of reliable object relationships and corresponding transferring processes
between BIM and TES. Therefore, this paper aims to provide a robust process of building
information exchange by comparing the multi-platform and single-platform processes for TES,
while finding the missing links embedded in the IFC file based on the effects of space boundary
conditions. A more consistent implementation of a robust IFC file proves itself to improve the
quality of TES results and to facilitate multi-domain collaborations.

2 BUILDING GEOMETRY REQUIREMENTS OF TES

Current practice shows that an architectural design model is usually transformed into a TES view
through an IFC file, which contains geometry information of various building elements, such as
IfcWall, IfcCartesianPoint (BuildingSMART 2016). Within the context of thermal energy
analysis, space boundaries play a vital role, which are virtual objects related to spaces or rooms in
buildings. They are represented with two levels in an IFC file. The first level of space
boundaries is defined by the surfaces of building elements bounding a given surface, which
depend only on virtual boundaries immediately adjacent to the zone of interest without
considering dividing parts. The second level depends on the invisible space behind the boundary,
which is considered as the subdivision of the first level space boundary that represents a divided
space with unique and consistent rate of heat flow or specific thermal performance. In terms of a
floor plan of a building, when comparing the two levels of space boundaries, the second level of
space boundaries require that the floor slab or wall of a space be broken along the centerline of
the slab or wall adjacent to it. This allows some neighboring relations with adjacent spaces to be
defined correctly. This is important for thermal energy analysis because a specific configuration
of the space boundary determines the energy flow or airflow between neighboring surfaces.
The space boundaries include two categories of information to be used for the thermal energy
analysis: the surface area and the material properties (Bazjanac 2010). The surface area, also
called the building envelop, determines the thermal zoning and its corresponding thermal mass.
The more surface area a building has, the more heat exchange will take place. For instance, in
hotter climate, the taller ceiling, which results in more surface area and building volume, is
appreciated because cooling takes place quickly under this condition. The direction of heat
transfer is also important so that the vector information of space or surface should be
distinguished. However, the information of surface area is often missing after exported from an
architectural design software. Building components in a 3D architectural model should be

2
 Resilient Structures and Sustainable Construction

recognized as interior or exterior surface areas in a TES model, and likewise, a room should
automatically indicate a thermal zone. The material properties (i.e., the thermal abilities of a
material), such as its specified thermal conductivity coefficient, also determine the building
energy performance. The material properties influence the effects of solar radiation on window
frames or wall surfaces; therefore, considering the solar radiation to which a building space is
exposed to, is particularly important for reliable TES. Theoretically, the material properties
closely related to thermal energy analysis should be consistent and comprehensive after they are
exported from 3D architectural model, however, current practices are using and transferring
insufficient information and most thermal properties of a material are not available to be assigned
to building components in architectural design software. Obviously, due to a lack of information
of surface area and material property provided by the architectural model, iteratively manual
corrections of the model are required and unavoidable for TES.

3 MULTI AND SINGLE PLATFORM PROCESSES FOR TES

In an attempt to assess the efficiency of proposed processes, three buildings of different sizes and
complexities have been considered. They have been obtained from an open IFC repository (Open
IFC model Repository 2016). No comments will be made on the quality of IFC files and thermal
performance since this work focuses on the process itself. This also implies that the IFC files will
be used as-is and no manual corrections will be made to overcome any importing/simulation
problem to fit the thermal analysis. Buildings considered within this study are presented in
Figure 1. Since this paper only investigates the interoperable process, the impact on the
experiments from the volumes and types of selected buildings will not be considered.

Figure 1. Buildings considered with floor, external wall, and external window information.

3.1 Multi-platform Processes

The current application of standard exchange format can be described as a multi-platform
process, where a 3D architectural model is created and subsequently imported and analyzed in
some dedicated TES software. For this study, Simergy (Simergy 2013), Energy Plus (Energy
Plus) and CYPETHERM Eplus (CYPETHERM 2016) were used. The TES may be conducted
either by an engineer or an architect in the early-design phases (Bambardekar and Poerschke
2009). As described earlier, interoperability problems are common with the IFC-file standard and
require manual corrections on the architectural model for further accurate energy simulation.
Moreover, particular modification of the model may be required back and forth for TES relevant
parameters, such as reassigning of thermal zones and extra addition of material properties.
Lastly, in order to have a comprehensive and an accurate simulation that translates well to the
actual thermal energy performance, a number of simulation parameters (e.g., weather conditions,

3
Pellicer, E., Adam, J. M., Yepes, V., Singh, A., and Yazdani, S. (eds.)

building orientation, internal occupational loads, HVAC attributes, etc.) are required as an input.
A flowchart summarizing the main steps of the multi-platform process is shown in Figure 2.

Export IFC file Import IFC file

Design and
from architectural into energy
modify building Read no errors
design software simulation tools
components
(BIM) (TES engine) TES results

Manual modifications and additions of TES relevant parameters

Figure 2. Multi-platform process explained via a flowchart.

All three buildings used as an example to test this process (see Figure 2) encountered
problems within different phases of simulation and thus it was not possible to come up with a
conclusion on the efficiency of this multi-platform process. Simergy had errors in importing the
geometry correctly; existence of windows and roofs/slabs was the main problem. Energy Plus
(Energy Plus) was able to import the files (converted to IDF) with no warnings or errors, yet it
was revealed after simulation that windows were not recognized. Lastly, CYPETHERM Eplus
was not able to open the IFC files at all. The difficulty in importing the geometry correctly,
especially first try without the need for manual modification was deemed not possible.

3.2 Single-platform Processes

The efficiency of employing a single-platform process, i.e. the use of single software both for
architectural design as well as TES, was investigated as an alternative to the conventional multi-
platform process. For this purpose, the IFC (IFC1) files of the same buildings presented in Figure
1 were imported into the 3D-CAD software Revit. As an initial observation, no difficulties arose
during the importing process. They were then analyzed with the built-in energy simulation tool
of this software. The procedure was quite straightforward and required only a minimal number of
operations. No manual corrections of geometry as well as setting of other parameters (e.g.
glazing material type of a window, conductivity coefficient of a wall etc.) were required. The
results were presented in the end in an electronic report format. This shows that a single-platform
process can potentially eliminate interoperability issues. Additionally, professionals would
require experience in a less amount of tools. However, the software dependency created by this
process should not be neglected.
In order to further optimize this process, Space Boundary Tool (Berkeley Lab 2014) was
incorporated in an attempt of automating the cumbersome task of defining space boundaries,
enhancing the precision of surface areas of a building and exporting an enhanced IFC (IFC2) file
for further analysis. Automatic handling of space-boundaries could drastically reduce manual
labor for energy simulations. The workflow of a single-platform process with the inclusion of
Space Boundary Tool is shown in Figure 3. Results from TES, both with and without the use of
Space Boundary Tool, is compared and presented in the next section and its impact on the
simulation results discussed.

Import IFC2 file

Design in CAD Enhance IFC1 file
Export IFC1 file into CAD
software through Space
from the CAD software
(with energy Boundary Tool
software (with energy
simulation and export IFC2
(BIM) simulation TES results
capability) file
capability)

Figure 3. Single-platform process explained via a flowchart.

4
 Resilient Structures and Sustainable Construction

4 IMPLICATIONS FOR ROBUST IFC FILES: EFFECT OF SPACE BOUNDARY

According to the comparative simulation results of TES with and without Space Boundary Tool
conditions in Table 1, an increase in the values of energy simulation metrics (e.g., electricity use,
fuel use, energy cost, etc.) were observed under the condition that the space boundaries in a
building were optimized using the Space Boundary Tool. It is assumed that the building energy
consumption will increase when more precise room or thermal zone definitions are detected for a
simple building structure. A more correct assignment of surface area to a room and its building
component’s material property yielded an improvement of energy simulation accuracy. For
example, all the three buildings generated increased energy costs due to a reliable validation of
surface areas that determined the specific shape and functionality of each room. However, it is
noted that the variations in the simulation results, to some extent, depended on the building
masses themselves, for example, building #1 experienced a significant change in electricity use
after it was experimented with Space Boundary Tool. Generally, it is revealed that the quality of
building models in terms of the level of space boundary concretization influences the TES
processes and results. However, usually there was missing information of space boundaries and
boundary allocation rules in a raw IFC file directly exported from CAD design software. Also, a
complex building geometry would cause the difficulty of defining proper building space
boundaries, such as a curve-shaped geometry that is hard to define and calculate. Hence, in the
perspective of an architect, his/her architectural model requires various evaluations to improve its
quality due to the customized space boundary data in an IFC file needed for TES. On the other
hand, TES engineers are recommended to consider the effect of space boundary optimization in
order to ensure an acceptable level of energy model quality.

Table 1. Results from TES of the three buildings.

Building No. Without Space Boundary Tool With Space Boundary Tool Change [%]
Electricity use: 195 kWh/sm/yr Electricity use: 370 kWh/sm/yr 89.7
Building #1 Fuel use: 964 MJ/sm/yr Fuel use: 1402 MJ/sm/yr 45.4
Energy Cost: 66,789 CHF Energy Cost: 70,277 CHF 5.2
Electricity use: 136 kWh/sm/yr Electricity use: 137 kWh/sm/yr 0.7
Building #2 Fuel use: 271 MJ/sm/yr Fuel use: 274 MJ/sm/yr 1.1
Energy Cost: 131,310 CHF Energy Cost: 130,560 CHF -0.6
Electricity use: 148 kWh/sm/yr Electricity use: 151 kWh/sm/yr 2.0
Building #3 Fuel use: 514 MJ/sm/yr Fuel use: 502 MJ/sm/yr -2.3
Energy Cost: 540,610 CHF Energy Cost: 647,285 CHF 19.7

5 CONCLUSIONS AND FUTURE RESEARCH

This paper proposes two processes to handle building information exchange between an
architectural model and a thermal energy simulation model in order to address current
interoperability. The proposed processes were compared to evaluate their performance and
efficacy. Three buildings were tested respectively using the multi-platform process and single-
platform process, revealing that the single-platform process was preferred because no manual
corrections of geometry were required to conduct a TES. Additionally, the Space Boundary Tool
was applied to investigate the information reliability and robustness embedded into an IFC file
using the single-platform process. Results show that the accuracy of TES simulation figures (e.g.,
electricity use, fuel use and energy cost) can be improved with the usage of Space Boundary
Tool. Quick adaption of surface areas and material properties will enhance the quality of the

5
 Pellicer, E., Adam, J. M., Yepes, V., Singh, A., and Yazdani, S. (eds.)

energy model and yield reliable energy simulation results. Although the space boundary
considerations can improve the performance of an IFC file in order to facilitate TES, the rules to
conduct the space optimization for TES are still abstract and not well defined, especially for
complex building structures. Future research focuses on the accuracy of geometry information
transfer, especially the accuracy of reading coordinate system of the building geometry.
Meanwhile, the Finite Element Method can be investigated to overcome the geometry translation
problems with its well-defined coordinate system and naturally embedded material properties.

Acknowledgments
This paper resulted from elaboration on the initial work conducted during the Academia Industry
Modelling (AIM) week organized by the Computational Science Zurich (CSZ), a joint initiative of ETHZ
and UZH (http://www.zhcs.ch/education/academia-industry-modeling/). The authors would like to thank
the organization committee and kindly acknowledge Ms. Dantong Yang and Dr. Stephan Huck from
Siemens AG for providing the topic and their support.

References
Ahn, K., Kim, Y., Park, C., Kim, I., and Lee, K., BIM interface for Full vs. Semi-automated Building
Energy Simulation, Energy and Buildings, 68, 671-678, 2014.
Bambardekar, S., and Poerschke, U.. The Architect as Performer of Energy Simulation in the Early Design
Stage, Building Simulation, 1306-1313, 2009.
Bazjanac, V., Space Boundary Requirements for Modeling of Building Geometry for Energy and Other
Performance Simulation, in CIB W78: 27th International Conference, 2010.
Berkeley Lab Simulation Research, U.S. Department of Energy, Space Boundary Tool, 2014. Retrieved
from https://simulationresearch.lbl.gov/projects/space-boundary-tool on November 24, 2016.
BuildingSMART International Ltd. (BuildingSMART), Summary of IFC Releases, IFC Release,
Specifications, BuildingSMART International home of openBIM, 2016. Retrieved from
http://www.buildingsmart-tech.org/specifications/ifc-releases/summary on December 6, 2016.
CYPETHERM, Version 2016, CYPETHERM Suite: EPlus, CYPE Ingenieros, S.A. Retrieved from
http://www.store.cype.com/en/cypetherm-suite/462-cypetherm-eplus.html on November 20, 2016.
Eastman, C. M., et al., BIM Handbook: A Guide to Building Information Modeling for Owners, Managers,
Designers, Engineers, and Contractors, John Wiley & Sons, 2011.
El Asmi, E., Robert, S., Haas, B., an Zreik, K., A Standardized Approach to BIM and Energy Simulation
Connection, International Journal of Design Sciences & Technology, 21(1), 2015.
Energy Plus, Version 8.6.0., U.S. Department of Energy. Retrieved from https://energyplus.net/ on
November 20, 2016.
Kim, J. B., Jeong, W., Clayton, M. J., Haberl, J. S., and Yan, W., Developing a Physical BIM Library for
Building Thermal Energy Simulation, Automation in Construction, 50, 16-28, 2015.
Maile, T., O’Donnell, J., Bazjanac, V., and Rose, C., BIM–Geometry Modelling Guidelines for Building
Energy Performance Simulation, in Building Simulation Conference, 2013.
Open IFC model Repository, Four storey office building, Four sto-001
(http://openifcmodel.cs.auckland.ac.nz/_models/20160414office_model_CV2_fordesign.ifc), ERDC:
Duplex Apartment Model, ERDC - D-001
(http://openifcmodel.cs.auckland.ac.nz/_models/030811DuplexModel-IFC-2011-05-05.zip), and KIT:
AC11-Institute-Var-2-IFC|KIT: AC1-001 (http://openifcmodel.cs.auckland.ac.nz/Model/Details/110 ),
Department of Computer Science, The University of Auckland, 2016. Retrieved from
http://openifcmodel.cs.auckland.ac.nz/Model on November 25, 2016.
Simergy, Simergy: A Free, Comprehensive Graphical User Interface for Energy Modeling with
EnergyPlus, Version 1.3. U.S. Department of Energy, 2013. Retrieved from
https://eetd.lbl.gov/news/article/56613/simergy-a-free-comprehensive-gr on November 20th, 2016.
Wimmer, R., Cao, J., Remmen, P., Maile, T., O'Donnell, J., Frisch, J., Streblow, R., Müller, D. and Van
Treeck, C., Implementation of Advanced BIM-based Mapping Rules for Automated Conversion to
Modelica, in Proceedings of the 14th IBPSA Conference, 2015.

View publication stats