BIM-based Parametric Building Energy Performance Multi-

Objective Optimization
Mohammad Rahmani Asl1 , Michael Bergin2 , Adam Menter3 , Wei Yan4
1
PhD Candidate, Department of Architecture, Texas A&M University
2
Research Scientist, Autodesk Inc.
3
Sustainability Education Program Manager, Autodesk Inc.
4
Associate Professor, Department of Architecture, Texas A&M University
1
sites.google.com/site/bimsimgroup/people/students/mohammad-rahmani-asl
2
www.autodeskresearch.com/people/michaelbergin 3 www.adammenter.com/
4
faculty.arch.tamu.edu/wyan
1,4
{mrah|wyan}@tamu.edu 2,3 {michael.bergin|adam.menter}@autodesk.com

Building energy performance assessments are complex multi-criteria problems.

Appropriate tools that can help designers explore design alternatives and assess
the energy performance for choosing the most appropriate alternative are in high
demand. In this paper, we present a newly developed integrated parametric
Building Information Modeling (BIM)-based system to interact with cloud-based
whole building energy performance simulation and daylighting tools to optimize
building energy performance using a Multi-Objective Optimization (MOO)
algorithm. This system enables designers to explore design alternatives using a
visual programming interface, while assessing the energy performance of the
design models to search for the most appropriate design. A case study of
minimizing the energy use while maximizing the appropriate daylighting level of
a residential building is provided to showcase the utility of the system and its
workflow.

Keywords: Building Energy Performance Analysis, Building Information Model

(BIM), Parametric Modelling, Parametric Energy Simulation, Multi-objective
Optimization

INTRODUCTION and labor-intensive process. Designers deal with a

Due to the considerable impact of buildings on the complex Multi-Objective Optimization (MOO) prob-
environment, it is essential for designers to recognize lem to minimize capital and operating costs while
the importance of improving or optimizing building maintaining occupants comfort (Wang et al., 2005;
energy performance in the early design stage. En- Wright et al., 2002). This complexity comes from the
ergy performance-based design is a highly complex large number of interrelated parameters involved in

BIM - Volume 2 - eCAADe 32 | 455

 sustainable building design such as building geome- significant amount of time and effort in preparing in-
try, space layout, materials, sites, weather data, user put data for building energy simulation while reduc-
behaviors, etc. There is a lack of easy-to-use and effi- ing errors (Kumar, 2008).
cient tools to help architects explore design alterna- In this paper we investigate a systematic integra-
tives and understand their impacts on building en- tion of BIM, parametric modeling, and building per-
ergy performance. Consequently, design practition- formance analysis to provide a new workflow that
ers either decide not to consider energy performance makes the parametric building energy performance
of their designs and instead follow general rules-of- study more accessible for innovative energy efficient
thumbs, which may result in inefficient building de- building design. The workflow uses a MOO algorithm
signs, or seek help from building energy experts to to explore the design space and provide a set of op-
simulate building design alternatives. Since trans- timal solutions to the designers.
ferring an architectural design model to an energy
model is a time consuming and error-prone process, BACKGROUND
the designers and energy experts have to select a lim- The conventional architectural design methodolo-
ited number of design alternatives for energy analy- gies focus on space and form. With the increasing
sis, which result in unoptimized design solutions. importance of building energy-efficiency, designers
Current building energy modeling tools do not have to consider energy performance of their de-
support comprehensive parametric relations among sign by exploring design alternatives that are more
building objects for simulation in tools such as En- promising to save energy in the conceptual design
ergyPlus. For instance, if a wall is transformed in phase (Azhar et al., 2009). A considerable amount
an energy model, none of the related objects in- of literature has been published on building energy
cluding windows, shading devices, rooms, roofs, and simulation tools. For instance Maile et al. (2007) stud-
floors will be updated automatically. In other words, ied the use of a selection of energy simulation en-
parametric intents that are embedded in parametric gines and their user interfaces over different build-
Building Information Modeling (BIM) are not embed- ing lifecycle phases. Also, Crawley et al. (2008) pro-
ded in the energy models. As a result, a manual up- vided a comparison of the features and capabilities of
date of the model data is needed before running the twenty major building energy simulation tools. The
simulations but this is complex, tedious, and error- literature review of this paper is focused on build-
prone. ing energy simulation in conjunction with paramet-
In order to fulfill the requirements of low en- ric modeling, BIM, multi-objective optimization, and
ergy building design there is a need for an innova- visual programming, which are the techniques that
tive design methodology and integrated design pro- are used in the developed integrated system.
cess. The integration of parametric modeling and
BIM is the new trend of building modeling, which can Parametric Modeling and Building Energy
greatly benefit sustainable building design. Paramet- Performance Analysis
ric modeling enables the creative exploration of a de- One of the major benefits of performing energy simu-
sign space by varying parameters and their relation- lation during the design process is to compare design
ships (Azhar and Brown, 2009). BIM is a model-based alternatives using parameters and rules among ob-
process that provides methods and tools for creat- jects. Parametric modeling enables generative form-
ing and managing building projects faster and more making and form-finding on the basis of aesthetic
economically (Eastman et al., 2011). BIM may contain and performance metrics of buildings. Once the con-
most of the data needed for building energy perfor- texts change in a later design stage, parametric mod-
mance analysis and if used appropriately can save a eling allows objects to automatically update (Aish

456 | eCAADe 32 - BIM - Volume 2

and Woodbury, 2005; Stocking, 2009). Designers can Ecotect and Green Building Studio (Azhar et al., 2009,
integrate parametric modeling into the process of 2011), and Modelica-based tools (Yan et al. 2013).
performance analysis in different fields of building The common approaches in this type of research is
design, including, but not limited to, energy simula- to translate the BIM models to energy input files for
tion (Paoletti et al., 2011; Pratt and Bosworth, 2011), solving interoperability issues using Industry Foun-
structural analysis(Shea et al., 2005), and acoustic dation Classes (IFC) (Bazjanac, 2008; Morrissey et al.,
simulation (Wu and Clayton, 2013). 2004) and to create an automatic link between BIM
Parametric studies show a significant potential authoring tools and building energy simulation en-
contribution to optimize the building energy per- gines (Yan et al. 2013).
formance (Naboni et al., 2013; Pratt and Bosworth, Integration of BIM and parametric modeling pro-
2011). Nonetheless, designers rarely use parametric vides a more effective process for performance-
building energy performance analyses for the sake based design. Welle et al. (2011) created a thermal
ofdue to the difficulty in preparing the energy mod- optimization tool, ThemalOpt, which used BIM for ex-
els as well as the long simulation run time. To solve tracting the necessary information for thermal simu-
this issue, there are two common approaches: to lation and optimization. Rahmani et al. (2013) devel-
develop computational algorithms that reduce the oped Revit2GBSOpt, a plug-in for a BIM platform (Au-
number of runs (Coley and Schukat, 2002; Wetter todesk Revit®), which integrates parametric BIM and
and Wright, 2004), or to increase the computational building energy performance simulation. Due to the
power through cloud-based simulation (Garg et al., complexity of parametric design study, an easy and
2010; Zhang and Korolija, 2010; Zhang, 2009). visual approach for designers to set up building pa-
rameters and the inclusion of advanced, open source
BIM and Building Energy Performance Anal- MOO algorithms are needed to improve the existing
ysis studies, as presented in this paper.
BIM is the process of generating and managing dig-
ital representations of the building's physical and Building Energy Performance Optimization
functional characteristics to facilitate the exchange Optimization studies are being used in building de-
of information (Eastman et al., 2011). BIM represents sign after long being computationally intractable, on
the building as an integrated database of coordi- multi-scale systems in various topics including opti-
nated information that can be used for the analysis of mizing construction costs (Radford and Gero, 1987),
the multiple performance criteria including architec- construction elements (Sambou et al., 2009), build-
tural, structural, energy, acoustical, lighting, etc. (Fis- ing shapes (Wang et al., 2006), building envelopes
cher, 2006). Performance-based design supported (Bouchlaghem, 2000; Radford and Gero, 1987), Heat-
by BIM is increasingly used in the building design ing, Ventilation, and Air Conditioning (HVAC) systems
disciplines, allowing practitioners efficiently gener- (Zhang et al., 2006), etc.
ate and modify building models (Fischer, 2006; Welle There are two common approaches to MOO
et al., 2011). problems: 1) simple aggregation 2) Pareto Optimal.
The existing studies that consider BIM as the In simple aggregation, a composite objective func-
central data model for building energy performance tion is defined by combining all of the individual
analysis are mainly focused on automatic preparation objective functions. The composite objective func-
of the building energy model for various energy sim- tion can be determined with various methods, like
ulation tools such as such as eQUEST (Maile et al., use of weighting factors. Determining the composite
2007), EnergyPlus (Maile et al., 2007; Bazjanac, 2008; objective function needs knowledge of the relation-
Cormier et al., 2011), TRANSYS (Cormier et al., 2011), ships among individual objectives and their weight-

BIM - Volume 2 - eCAADe 32 | 457

 ing factors (Fonseca and Fleming, 1993; Konak et ties. The prototype contains a set of new function
al., 2006). Nevertheless, in building design these re- nodes that can be used to optimize building energy
lationships are unknown in many cases. The sec- performance.
ond approach is to seek a set of promising solutions, We have developed multiple Dynamo nodes to
known as Pareto-optimal set (Fonseca and Fleming, contain essential functions for creating parametric
1993), given multiple objectives. Pareto Optimality BIM models in Revit and run parametric simulations
supports decision making by finding the equally op- in GBS. A MOO algorithm (Non-dominated Sorting
timal solutions such that it is not possible to improve Genetic Algorithm-II or NSGA-II, Deb et al., 2002) is
a single individual objective without causing at least created in Dynamo as a package of nodes that can
one other individual objective to become worse off help designers optimize multiple conflicting objec-
(Hoes et al., 2011). A posteriori set of preferences may tives and approach to a set of optimal solutions. The
be used to evaluate the optimal solutions and find NSGA-II node package is built based on the open
the unique solution later by the designers (Gossard source code [1]. The node "NSGA-II" in Dynamo in-
et al., 2013; Konak et al., 2006). cludes a package of nodes and plays the main loop
role for population generation in MOO to get to the
Visual Programming optimal solution (figure 1). The node "Initial Solu-
While computer programming is often needed for tion Set" generates the initial set of random variables
designers to implement their sophisticated design within the provided range and with the size of popu-
intent (e.g. through the use of for-loop and con- lation defined by user. The output of this node is a list
ditional statements) in parametric BIM, visual pro- of variables and objective. The objective values are
gramming interfaces can replace the conventional null and they are assigned by "Population Evaluate"
elaborate coding with a visual metaphor of con- node which gets objective values as input parame-
necting small blocks of independent functionalities ters.
into a whole system or procedure (Boeykens and
Neuckermans, 2009). Visual programming allows Figure 1
users create computer programs by manipulating Implementation of
program elements graphically rather than textually. NSGA-II in Dynamo
Based on a survey of 50 visual programming lan- to optimize
guages (Myers, 1990), it is clear that a more visual daylighting and
style of programming could be easier to understand energy use
for non-programmers or novice programmers (archi-
tects normally fit into these categories). Examples of
visual programming tools for architectural design are
This workflow enables the Dynamo code to ac-
Grasshopper for McNeel Rhinoceros® and Dynamo
cept objective functions as nodes or packages of
for Autodesk Revit®.
nodes. For instance, in this study the "LEED Daylight-
ing" node is created as a package of nodes to calcu-
METHODOLOGY late the LEED daylight values based on LEED Refer-
In this study an integrated system is developed for ence Guide for Green Building Design and Construc-
enabling designers to optimize multiple objectives in tion (USGBC, 2009) as an objective function.
the early design process. A prototype of the system is The node "gbXMLExport" in Dynamo generates
created in an open-source visual programming appli- energy model data in the Green Building eXtended
cation - Dynamo, which can interact with a BIM tool Markup Language (gbXML, 2014) format, which con-
(Autodesk Revit®) to extend its parametric capabili- tains the necessary information for energy simula-

458 | eCAADe 32 - BIM - Volume 2

 tion, using Revit's Application Programming Inter- tion of the energy use requires building information
face (API). The "GBSProject" node is designed to cre- that BIM can provide, for example geometry informa-
ate a new project in GBS by extracting the project in- tion, physical material information, and location data
formation from a BIM model such as the project loca- embedded within the model. The workflows devel-
tion and the building type using Revit API, GBS API, oped in this project can identify parameters from el-
and the Representational State Transfer (REST) pro- ements within the BIM and explore a set of scenarios
tocol. "GBSRun" is designed to create multiple runs for energy performance and daylighting adequacy.
in the GBS project and upload the exported gbXML
files to GBS for whole building energy analysis. When Climate and Context
the simulations are done, GBSRun retrieves the en- The geographic location of the home is in the city
ergy simulation results for further analysis, optimiza- of Indianapolis, Indiana, USA. The climate is domi-
tion, and visualization (figure 2). nated by heating loads with 5892 Heating Degree
The presented system enables designers to ex- Days (HDD) on a yearly basis. Due to site constraints,
plore design alternatives and at the same time assess the long-axis orientation of the structure is fixed at 15
the building performance to search for the most ap- degrees west of true north (figure 3).
propriate design.
Model and Free Parameters (Decision Vari-
Figure 2 ables)
Parametric BIM and The residential home has six rooms at level one and
whole building two rooms at the second level that are included as
energy simulation part of the daylighting calculation and energy use for
integration in the entire building. The light admitted to the build-
Dynamo ing can enter via two fixed curtain walls that are not
included as free parameters in the design space op-
timization. These two curtain systems light the main
living space in the first floor and the balcony in sec-
Figure 3 ond floor. The rooms separated from the main living
Case study building space by interior partitions are lit naturally by fixed
site and floor plans windows with a visual transmission coefficient of 0.9.
The width and height of the windows are identified
within the Dynamo interface as free parameters. The
domains of the width and height of the glazing area
are set independently from 0.5' to 7.0' with an incre-
ment of 0.1'.

CASE STUDY
The objectives of the optimization routine for this
case study is to maximize the number of rooms of the
residential unit that satisfy the requirements of the
LEED IEQ Credit 8.1 for Daylighting while minimizing
the expected energy use. The simulation and calcula-

BIM - Volume 2 - eCAADe 32 | 459

 Figure 4
General overview of
the designed MOO
system

Figure 5
Scatterplot
showing the Pareto
Frontier with model
thumbnails
superimposed on
the plot to illustrate
the association
between the
calculated optimal
solutions and the
building forms.

460 | eCAADe 32 - BIM - Volume 2

 Optimization Algorithm CONCLUSION
The NSGA-II algorithm is implemented with the in- The investigation shows that the use of a BIM model
put of a population size of 100 for each generation, to generate a multiplicity of parametric design vari-
with the maximum evaluations set at 1000 for a total ations for simulated and procedural analysis is a vi-
of 10 generations. The mutation probability is set at able workflow for designers seeking to understand
0.01. The crossover probability is set at 0.9 and both trade-offs between daylighting and energy use. The
the mutation distribution index and crossover distri- availability of a cloud-based energy analysis tool en-
bution index are set at 20.0. Figure 4 shows the gen- ables the quick evaluation of hundreds of design vari-
eral overview of the MOO system designed for this ations and the connection to a visual, parametric pro-
study and figure 1 shows its implementation in Dy- gramming environment allows the design space to
namo to optimize daylighting and energy use of the be quickly and accurately specified.
building. The Pareto Optimal set from the NSGA-II al- Designers with limited parametric modeling and
gorithm is shown in figure 5. This graph shows the programming experience may use the nodes pro-
result for 1000 runs for this experiment which took duced to perform a broad variety of design space
about 3 hours overall. This graph indicates that the analyses. It is possible to optimize each window's
optimization routine begins to converge on the opti- width and height individually though this method
mal solution for each variable from the third genera- expands the design space considerably. It is also pos-
tion onward. sible to include the angle of the building orienta-
From the graph in figure 6 it can be seen that tion and the overall building footprint in the set of
windows of various Widths from 1' to 7' meet the re- free parameters to be modified. For a broader de-
quirements for more than 80% of the rooms correlat- sign space the number of iterations required may be
ing with about $150 in variation for the yearly energy significantly increased to obtain reliable optimization
cost. In this instance, windows between the sizes results.
of 3' and 4' in Height are evaluated, as this parame- In addition to local variables such as window di-
ter is preferred for the reason of style to fit with im- mensions and material variations this system is ca-
mutable horizontal datum elements. For design vari- pable of producing design options in global building
ations within the bottom 30% of energy cost and the geometries such as the footprint, the form of the roof,
full satisfaction of the daylighting metric, the smallest and the interior layouts. These design options are
glazing Width is specified at 2' 8". considered often by architects and engineers in the
design process. The information embedded within
Figure 6
the BIM can quickly be leveraged to obtain quantifi-
Interactive parallel
able sensitivity of the performative implications to a
coordinates plot for
broad set of possible design decisions.
the constraint and
Through the continued development of similar
analysis of design
projects to enable fast BIM-based simulation and rep-
parameters.
resentation of solution spaces and trade-offs, design-
Visualizing the results in an interactive parallel coor- ers may be able to understand dependencies of de-
dinates plot allows the various iterations to be evalu- sign options on the decision variables at the early
ated by the designer. In figure 7 the chart shows the design stage without substantial expertise in energy
sample of design variations that meet 100% of the modeling and daylighting analysis. For parametric
LEED Daylighting requirements. Of these the lowest analysis, large changes in global building geometry
energy use calculated is $4,265 and the smallest win- can lead to alterations in structural requirements and
dow size is specified as a 5' width and 3.5' height. mechanical systems as well. Incorporating a broader

BIM - Volume 2 - eCAADe 32 | 461

 Figure 7
Illustration of a
bi-directional
association
between parallel
coordinates and 3D
model views

variety of simulations in different domains into the REFERENCES

system will lead to more comprehensive exploration Aish, R and Woodbury, R 2005, 'Multi-level interac-
of the solution space and provide better decision tion in parametric design', Smart Graphics, 3638, p.
support for the stakeholders of building construc- 151–162
Azhar, S and Brown, J 2009, 'BIM for Sustainability Anal-
tion.
yses', International Journal of Construction Education
and Research, 5, pp. 276-292
Acknowledgment Azhar, S, Brown, J and Farooqui, R 2009 'BIM-based sus-
This research is partially supported by the National tainability analysis: An evaluation of building per-
Science Foundation under Grant No. 0967446. We formance analysis software', 45th ASC Annual Confer-
would like to acknowledge and thank Autodesk Inc.'s ence, pp. 1-4
Bazjanac, V 2008 'IFC BIM-based methodology for semi-
Building Performance Analysis and Dynamo teams,
automated building energy performance simula-
and all who contributed to our research for all their tion', Improving the Management of Construction
help and support. Also, we would like to thank Projects through IT Adoption, CIB W78, Universidad de
Alexander Stoupine at Texas A&M University for his Talca, pp. 292-299
contribution in this project. Boeykens, S and Neuckermans, H 2009 'Visual Program-
ming in Architecture: Should Architects Be Trained

462 | eCAADe 32 - BIM - Volume 2

 As Programmers?', CAAD-Futures 992–1007
Bouchlaghem, N 2000, 'Optimizing the design of build- Kumar, S 2008, Interoperability between building informa-
ing envelopes for thermal performance', Automation tion models (BIM) and energy analysis programs, Mas-
in Construction, 10, pp. 101-112 ter's Thesis, University of Southern California
Coley, DA and Schukat, S 2002, 'Low-energy design: Maile, T, Fischer, M and Bazjanac, V 2007, 'Building en-
combining computer-based optimization and hu- ergy performance simulation tools-a life-cycle and
man judgement', Building and Environment, 37, pp. interoperable perspective', Center for Integrated Fa-
1241-1247 cility Engineering (CIFE) Working Paper, 107, pp. 1-49
Cormier, A, Robert, S, Roger, P, Stephan, L and Wurtz, Morrissey, E, O’Donnell, J, Keane, M and Bazjanac, V 2004
E 2011 'Towards a BIM-based service oriented plat- 'Specification and implementation of IFC based
form: application to building energy performance performance metrics to support building life cy-
simulation', Building Simulation 2011 cle assessment of hybrid energy systems', Building
Council, USGB 2009, LEED Reference Guide For Green Sustainability and Performance Through Simulation,
Building Design And Construction, US Green Building Boulder, Colorado, USA
Council (USGBC) Myers, BA 1990, 'Taxonomies of visual programming and
Crawley, DB, Hand, JW, Kummert, M and Griffith, BT 2008, program visualization', Journal of Visual Languages &
'Contrasting the capabilities of building energy per- Computing, 1, pp. 97-123
formance simulation programs', Building and Envi- Naboni, E, Zhang, Y, Maccarini, A, Hirsch, E and Lezzi,
ronment, 43, pp. 661-673 D 2013 'Extending the use of parametric simula-
Deb, K, Pratap, A, Agarwal, S and Meyarivan, T 2002, tion in practice through a cloud based online ser-
'A fast and elitist multiobjective genetic algorithm: vice', BPSA - Conference of International Building Per-
NSGA-II', IEEE Trans. Evol. Comput., 6, p. 182–197 formance Simulation Association, Bozen, Italy
Eastman, C, Teicholz, P, Sacks, R and Liston, K 2011, BIM Paoletti, G, Avesani, S, Exner, D and Lollini, R 2011 'De-
Handbook: A Guide to Building Information Modeling signing low energy buildings: application of a para-
for Owners, Managers, Designers, Engineers and Con- metric tool and case studies', 27th International Con-
tractors, Wiley ference on Passive and Low Energy Architecture, Lou-
Fischer, M 2006 'Formalizing construction knowledge for vain La Neuve, Belgium
concurrent performance-based design', 13th inter- Pratt, KB and Bosworth, DE 2011 'A method for the
national conference on Intelligent Computing in Engi- design and analysis of parametric building energy
neering and Architecture, pp. 186-205 models', International Building Performance Simula-
Fonseca, CM and Fleming, PJ 1993, 'Genetic Algorithms tion Association Conference–Building Simulation
for Multiobjective Optimization: Formulation, Dis- Radford, AD and Gero, JS 1987, Design by optimization in
cussion and Generalization', ICGA, 93, pp. 416-423 architecture, building, and construction, John Wiley &
Garg, V, Chandrasen, K, Tetali, S and Mathur, J 2010 'En- Sons, Inc.
ergyPlus Simulation Speedup Using Data Paralleliza- Rahmani, m, Zarrinmehr, S and Yan, W 2013 'Towards
tion Concept', AMSE 2010 4th International Confer- BIM-based Parametric Building Energy Performance
ence on Energy Suistanibility, pp. 1-6 Optimization', 33rd Annual Conference of the Associ-
Gossard, D, Lartigue, B and Thellier, F 2013, 'Multi- ation for Computer Aided Design in Architecture (ACA-
objective optimization of a building envelope for DIA), Cambridge, Ontario, Canada, pp. 101-108
thermal performance using genetic algorithms and Sambou, V, Lartigue, B, Monchoux, F and Adj, M 2009,
artificial neural network', Energy and Buildings, 67, 'Thermal optimization of multilayered walls using
pp. 253-260 genetic algorithms', Energy and Buildings, 41, pp.
Hoes, P, Trcka, M, Hensen, J and Bonnema, B 2011 'Op- 1031-1036
timizing building designs using a robustness indi- Shea, K, Aish, R and Gourtovaia, M 2005, 'Towards inte-
cator with respect to user behavior', Building Sim- grated performance-driven generative design tools',
ulation Proceedings of the 12th Conference of the In- Automation in Construction, 14, pp. 253-264
ternational Building Performance Simulation Associa- Stocking, AW 2009 'Generative Design Is Changing the
tion, pp. 1710-1717 Face of Architecture', Building Design
Konak, A, Coit, DW and Smith, AE 2006, 'Multi-objective Wang, W, Rivard, H and Zmeureanu, R 2006, 'Floor shape
optimization using genetic algorithms: A tuto- optimization for green building design', Advanced
rial', Reliability Engineering & System Safety, 91, p. Engineering Informatics, 20, pp. 363-378

BIM - Volume 2 - eCAADe 32 | 463

 Wang, W, Zmeureanu, R and Rivard, H 2005, 'Applying
multi-objective genetic algorithms in green build-
ing design optimization', Building and Environment,
40, p. 1512–1525
Welle, B, Haymaker, J and Rogers, Z 2011, 'ThermalOpt:
A methodology for automated BIM-based multidis-
ciplinary thermal simulation for use in optimization
environments', Building Simulation, 4, p. 293–313
Wetter, M and Wright, J 2004, 'A comparison of determin-
istic and probabilistic optimization algorithms for
nonsmooth simulation-based optimization', Build-
ing and Environment, 39, p. 989–999
Wright, JA, Loosemore, HA and Farmani, R 2002, 'Opti-
mization of building thermal design and control by
multi-criterion genetic algorithm', Energy and Build-
ings, 34, pp. 959-972
Wu, C and Clayton, M 2013 'BIM-based acoustic simula-
tion Framework', 30th CIB W78 International Confer-
ence, Beijing, China
Yan, W, Clayton, M, Haberl, J, Jeong, W, Kim, JB, Kota, S,
Alcocer, JLB and Dixit, M 2013 'Interfacing BIM with
Building Thermal and Daylighting Modeling', 13th
International Conference of the International Building
Performance Simulation Association (IBPSA), Cham-
bery, France, p. 3521–3528
Zhang, Y 2009 '“Parallel” EnergyPlus and the develop-
ment of a parametric analysis tool', IBPSA Conference,
p. 1382–1388
Zhang, Y and Korolija, I 2010 'Performing complex para-
metric simulations with jEPlus', Proceedings of the 9th
SET Conference, Shanghai, China
Zhang, Y, Lin, K, Zhang, Q and Di, H 2006, 'Ideal ther-
mophysical properties for free-cooling (or heating)
buildings with constant thermal physical property
material', Energy and Buildings, 38, pp. 1164-1170
[1] http://jmetalnet.sourceforge.net/index.html

464 | eCAADe 32 - BIM - Volume 2