Application of BEM and using BIM database for BEM_ A review

Journal of Building Engineering 23 (2019) 1–17
Contents lists available at ScienceDirect
Journal of Building Engineering

journal homepage: www.elsevier.com/locate/jobe
Application of BEM and using BIM database for BEM: A review T

⁎,1
Z. Pezeshki , A. Soleimani, A. Darabi
Faculty of Electrical and Robotic Engineering, Shahrood University of Technology, Shahrood, Semnan, Iran
A R T I C LE I N FO A B S T R A C T
Keywords: This paper reviews Building Energy Modeling (BEM) development using classification and review of articles for
BEM the recent years (2015–2018) to explore how much BEM methodologies have been used Building Information
Building energy Modeling (BIM) database during this period. Nowadays, BEM and supervision are the most important challenges
BIM of smart cities, and BIM has a good and useful source of information for buildings. Recent innovations have
Optimization
provided the opportunity for construction industry to invest in state-of-the-art technologies and adopt new
New methods
processes. Among these new methodologies is BIM, which has developed for the construction industry over the
last two decades. Integrating graphical and non-graphical information, it enables construction industry stake-
holders to work collaboratively for efficient project delivery throughout the life cycle of construction projects.
The use of BEM in construction industry has come about to respond to the global call for energy conservation and
sustainability. BEM tools and processes can be used to simulate energy performance, evaluate energy needs and
optimize architectural design. This article reviews and classifies BEM applications into eight different categories
such as prediction, estimation, consumption, optimal design, evaluation, efficiency, management, and optimi-
zation. In addition to examination of BEM in different categories and according to BIM that is a coordination
model and includes all information in construction industry as well as electrical, mechanical, and architectural
information, this review paper wants to know how many articles have used BIM such as a database in this
industry.
1. Introduction factors such as productivity characteristics, thermal efficiency, energy

management, and so on. This function is primarily based on energy
The construction sector is the largest consumer of energy, and while consumption efficiency to close the comfort level of building residents
we attend to comfort and feasibility of the construction, measures with a minimum amount of loss. On the other hand, energy require-
should reduce energy consumption. Building energy simulation means ments of buildings are heavily dependent on local weather and con-
using a software to predict energy consumption of a building. struction conditions (such as the envelope of the building, the facade to
Simulation software has the ability to define key parameters for dif- glass/no-glass ratio, and direction). Therefore, due to increasing em-
ferent parts and components of a building. This simulation should be phasis on energy consumption and greenhouse gas emissions, assess-
based on virtual displays, for example, a model of a building and ment of energy performance is often done by energy simulation soft-
models that interact with energy variables, such as energy simulations. wares for productivity and comfort [4–6]. The current review using a
In order to conduct a detailed study, building energy models should be cluster of articles, theses and books examines the evolution of EM from
developed to make more precise simulations. Energy Modeling (EM) 2015 to 2018 for energy simulation and optimization. This review is
means simulating energy models with an energy simulation software. based on selected journals, conferences, theses and books for EM
This simulation software, in accordance with needs of a building, can (Table 1). Basically, the review of these papers with seven types of
predict monthly energy consumption, annual energy costs, annual orientation has been followed to examine the building Energy Modeling
carbon emissions, compare different efficiency measures and make a (BEM): (1) Building Energy Prediction & Estimation, (2) Building En-
decision on cost savings. The most important data required to study an ergy Consumption, (3) Building Energy Optimal Design, (4) Building
EM case study of a building, are: 1) building geometry, 2) building Energy Evaluation, (5) Building Energy Efficiency, (6) Building Energy
characteristics, 3) measured energy consumption, and 4) weather data Management, and (7) Building Energy Optimization. With a review of
[1–5]. The energy performance of the building is affected by complex these studies, the rest of the paper is organized as follows. Sections 2–8
⁎
Corresponding author.
E-mail address: [email protected] (Z. Pezeshki).
1
http://www.shahroodut.ac.ir
https://doi.org/10.1016/j.jobe.2019.01.021
Received 6 August 2018; Received in revised form 22 December 2018; Accepted 14 January 2019
Available online 18 January 2019
2352-7102/ © 2019 Elsevier Ltd. All rights reserved.
Z. Pezeshki et al. Journal of Building Engineering 23 (2019) 1–17
Table 1 such as database for BEM are extracted. Today, researchers believe that
Research work in most cited journal, conferences, thesis, and book in organizations must use a Coherent Structure Model (CSM) in their
2015–2018. collective work especially energy problems that BIM can solve energy
Journal/Conference/Thesis/Book 2015 2016 2017 2018 problems as a powerful database [7]. So, innovations in energy solu-
tions with BIM has been introduced (Fig. 14). Finally, Section 10 con-
Journal clude how many papers has used BIM for BEM such as database and
Ain Shams Engineering Journal 1
why other papers hasn’t used that.
Applied Energy 2 2 1
Archives of Computational Methods in Engineering 1
Automation in Construction 1 2. BEM for building energy prediction and estimation
Building and Environment 1 1
Clean Technologies and Environmental Policy 1
In 2015, Bianchi et al. [8] explored the effects of climate boundary
Energy and Buildings 1
Energy 1 3 2
conditions for energy simulations of 4 types of commercial buildings by
Energy Procedia 4 comparing electricity and heating loads and producing photovoltaic
Energy Research Journal 1 electricity based on 3 locations in the city. In this study, for importing
Energy Technology 1 weather data into the EnergyPlus simulation software, these data were
Engineering, Construction and Architectural 1
first converted to files with “epw” extensions. Then, simulation was
Management
PLoS ONE 1 carried out and the results of this simulation showed that each of these
Journal of Architectural Engineering 1 1 commercial buildings are equally affected by the specific weather
Journal of Building Engineering 1 conditions of the site. At the same time, the comparison of the esti-
Journal of Building Performance Simulation 1 1 1 1
mated solar power production between data and daily data suggests
Journal of Professional Issues in Engineering 1
Education and Practice
that power loads do not affect production changes of this electricity. In
Journal Technologi (Science & Engineering) 1 2016, Yi [9] conducted a study on the important issue of daylight use to
International Journal of Sustainable Development 1 improve the building energy performance. He used the Kriging method
and Planning to predict daylight conditions throughout the year and combined this
Urban Climate 1
method with existing energy simulation tools. Kriging is a local and
Conference
ASHRAE and IBPSA-USA SimBuild 2016 Building 1 global interpolation method based on geo-statistical techniques that
Performance Modeling utilizes all observations of the area. The purpose of this study was to
Conference find a way reducing the actual computing time to be applied in practice.
Construction Research Conference 2016 1 In this study, a method for integrating the Kriging daylight model with
SIBRAGEC ELAGEC 2015 1
Sustainable Build Environment, Regenerative and 1
an energy simulation tool was proposed to trace time distribution of
Resilient Urban light in the interior space. In this research, the proposed method was
Environments 2016 (SBE16) compared with existing models of daylight in simulation tools. For this
Conference purpose, EnergyPlus, Delight, Winkelmann, eQuest and DOE software
International Conference on Sustainable 1
models were tested. The test results showed a better Root-Mean-Square
Infrastructure 2017
Proceedings of the 15th IBPSA Conference 1 Error (RMSE) and Mean Bias Error (MBE) in proposed method. Also, the
Proceedings of the 2017 Winter Simulation 1 simulation time of total radiation of the proposed method decreased
Conference (WSC) from 39.167 to 5.918 h. This study showed a more realistic result
11th International Conference on Countermeasures 1 compared to daylight models of energy simulation softwares to reduce
to Urban Heat
Island
computational costs. In 2015, Mohammadpour et al. [18] found that
2015 IEEE Conference on Technologies for 1 BEM is necessary to estimate energy consumption of buildings. BEM
Sustainability (SusTech) prediction enables owners, designers, and facility managers to measure
2017 IEEE Conference on Energy Conversion 1 building energy consumption and help them determine the energy
(CENCON)
performance of the building during the design phase. In order to esti-
2017 IEEE 41st Annual Computer Software and 1
Applications Conference mate the energy consumption of buildings at three different climate
(COMPSAC) zones, cold, hot-humid, and mixed-humid, this research is focused on
2016 International Conference on Micro- 1 two different commercial buildings with two H and rectangular designs.
electronics, Computing and For this purpose, a DOE building simulation software was used to create
Communications (MicroCom)
2017 National HVAC Simulation Academic Annual 1
and simulate building configurations created using the Monte Carlo
Meeting simulation technology. Ten thousand simulations were made for each
Thesis shape and climatic region to create data set included entire design
Sensitivity Analysis and Bayesian Calibration of 1 parameters. The steps taken in this study are shown in Fig. 1.
Building Energy
In this study, standardized regression coefficients were used to un-
Models
Energy modeling for estimation of indoor 1 derstand the effect of building parameters on annual energy con-
temperature of the building sumption in three different climate zones and in both commercial
on the BIM platform buildings with two H and rectangular designs. The results of models
Book showed an acceptable error rate of less than 5%, and occupancy plan-
Building Energy Performance Assessment in 1
Southern Europe
ning had the greatest effect on building energy consumption for both H
Building Energy Management: Performance 1 and rectangular designs in all three climatic zones. Chaudhary et al. [2],
Verification and System in 2016, introduced an Autotune method to automated production of
Simulation calibrated building energy models using measured data of 2 case stu-
Genetic Algorithm Essentials 1
dies. In the first case study, the building model was de-tuned by a de-
liberate input of the error in more than 60 parameters. Then, this model
was calibrated using Autotune and its accuracy was evaluated by
present review results of BEM based on the above mentioned categories,
Normalized Mean Bias Error (NMBE) and Coefficient of Variation Root
respectively. Section 9 presents some discussions of BEM methodolo-
Mean Squared Errors (CV-RMSE) according to the original model. In
gies, domains, and applications. In this section, the papers used BIM
this study, after energy simulation of building models using EnergyPlus,
2
Fig. 1. The steps taken in parametric analysis of building energy use [18].
calibration of the EM building was done to estimate energy consump- building energy models, Tian et al. [20], using the correlation analysis
tion. In addition to total energy consumption of the building, lighting and hierarchical clustering method, applied this calculation to de-
results, power loading, HVAC energy consumption, district tempera- termination and selection of useful energy data. They used an En-
tures, and other variables were also analyzed. In the second step, Au- ergyPlus energy simulation software to extract 4 uncertain input
totune calibration was directly compared with calibration of manual parameters from a case study (a retail building). This study is focused
simulated occupation of a residential building. At this stage, the results on identification of useful energy data in Bayesian calibration of
of Autotune calibration were obtained at much less time. This study building energy models. The results of this study showed that different
also provides a brief outline of strength and weaknesses of Autotune combinations of energy data can estimate unknown input variables
calibration approaches: (1) improving the adjusted model based on the with different accuracy by calibrating the model. Therefore, when there
comparison of unused data in calibration; and (2) a qualitative and are missing energy data that can be treated as uninformative output
quantitative evaluation of manual calibration accuracy versus Autotune data, Bayesian computation will be appropriate for extracting the
calibration. Martin et al. [19] believe that calibration of the building parameters. This method is also very useful for determining the effec-
energy model should be performed before the evaluation of energy tive energy-saving measurement in building improvement projects. In
saving, so that good strategies can be achieved by improving the 2016, Monari [21] argued that the framework of a system shows precise
system. Simple building energy models are often accurate and detailed mathematical behavior of various types of uncertainties involved in
and slow the calibration process down. Also, the difference between modeling and calibration issues. These uncertainties are observed
measured internal energy consumption and the potential estimated throughout the analysis process from beginning of the uncertainty to
value can be increased by simplifying the building energy model. In- the end of the calibration phase, and delivers information as field data
stead of specifying conditions for a precise urban model in 2016, they for determining the final solution of the studied problem. In this study,
set conditions for a simplified energy model in a city model. They first he used the ESP-r energy simulation software as an environment of
developed their precise and simplified model with all the structural and modeling. In this study, the Python library which was able to generate
thermal characteristics of a real office building in Singapore into En- automated patterns containing executable instructions of ESP-r models,
ergyPlus. Then, using the fitness analysis, the difference in monthly was developed for the hypothesis set of input vectors and aggregates of
cooling load estimates of precise and simplified EnergyPlus models, results. He began his analysis with virtual version of a real building in
which was a function of the number of classes, was shown. The same ESP-R. The ESP-R model was divided into 2 thermal zones, one on the
trend was applied as a percentage difference between simulations of inside of the library, and other in the outer space, which was used to
two models to EnergyPlus, and time and monthly cooling load con- rerun the test. Then, the ESP-r model was calibrated according to sys-
sumption were calculated on continuous and separate models in En- tematic observations generated by the simulation program, which data
ergyPlus. Then, precise and simplified EnergyPlus models were com- were noise-free and the model was able to present the considered
bined with a centralized urban thermal model to analyze the effect of process. To test capabilities of the author's predictive calibration model,
structural and thermal parameters of the building on the difference in calibration was performed by deleting one-third of final data and with
monthly cooling load estimates. In this study, external walls with U- all conditions not used in its training. The main goal of this study was to
value, the ceiling with U-value, the window with U-value and glazing demonstrate how the simulator had a sufficient ability to present be-
ratio were parameters that were obtained from sensitivity analysis. havior in the computer model and the calibration process was actually
Finally, an approximation of the distribution difference between able to estimate undefined parameters of the model, accurately. Two
monthly cooling load estimates and the Monte Carlo analysis was ob- processes of Autoregressive Moving Average (ARMA) and Generalized
tained. The results of this study showed that the difference between Autoregressive Conditional Heteroskedasticity (GARCH) were used to
monthly cooling estimates obtained from precise and simplified En- simulate noise. In the Finite Element Method (FDM) model (ESP-r
ergyPlus models can be explained by an increasing rational function of model), there were local correlations between differential vectors and
the number of floors. As a result, this difference has an upper limit. The boundary conditions. Predictions of the model were sufficiently con-
parameters of external walls with U-value and windows with U-value sistent with considered values, and constraints seemed reasonable.
that were obtained from sensitivity analysis had the greatest effect on Then, the analyzed models were compared with fitness criterion.
the difference between cooling load estimates. Also, in 2016, due to Common measurements of NMBE, CV-RMSE, and Goodness of Fit (GOF)
flexibility and accuracy of the Bayesian computation in calibration of were used to perform simple and easy calculations, but they led to
3
model equifinality. Also, due to multipurpose variables and databases, which affects thermal comfort of inhabitants and energy consumption
these measurements were poorly evaluated and how they are collected in conditioned space. One of the important parameters used in thermal
for these cases are not clear. Hence, the Bayes factors were used to analysis is the heat transfer coefficient of the building envelope. In this
verify that they were performed better. Given the equifinality of the study, they evaluated the effect of a gap reduction between archi-
model, naturally, probabilistic models used to calibrate different vari- tectural information on BIM maps and building conditions on EM of the
ables and databases should be allocated to this kind of the computation. BIM model and analyzed existing buildings. In spite of the large capa-
Despite the presence of noise in measurements, the results of this dif- city of description of conditions for construction plans in three-di-
ferential analysis depend on the internal temperature of the period. It mensional form, the DOE energy simulation software is still required to
was the reason for the difference between the output and measurements define thermal properties of each building element to calculate building
of the model. This virtual experiment has shown that even a relatively energy consumption and improve modeling accuracy. The validity of
simple model can have equifinality and identifiable problems. Shi and EM was obtained from conditions of construction plans, especially for
O’Brien [22] proposed a new method for reducing the number of old buildings of cold regions, and results were satisfactory. This re-
thermal areas in an EM building, and they called it ’model-cluster-research presents a method as improvement of building energy analysis.
duce’. Their model using the Principal Component Analysis (PCA) In another study, Harish and Kumar [11] developed the math Building
clustering technique, determined the initial forms and resolved areas. In Energy System (BES) model, which uses the mapping of energy transfer
this study, an experiment was conducted using a BIM model in En- processes occurred inside the building. In this study, components of
ergyPlus. To automate the process, a library was written in Python to building space and the location of heating and cooling which are re-
read BIM files and perform simulation tutorials. After training, the sponsible for thermal comfort of residents, were modeled. Heat equi-
model was reduced and the BIM file was regenerated. In total, 4 models librium of conduction, convection and radiation of transferred thermal
were produced using this method and compared with original model, energy of building components and equilibrium of thermal convection
random selected models and an expert model. By simulating reduced of the building surface were modeled and the space of the building was
models, annual energy was estimated and the results of parametric si- also modeled for transmission of tangible and hidden thermal energy.
mulations had also margin of error of 5% while they reduced simulation To use an invariant constrained optimization algorithm, the state space
time to 95%. In the same year, Li et al. [23] introduced a new model approach was used to model building components such as walls and to
called City Building Energy Use Model (CityBEUM) to building energy estimate parameters. The HVAC system was modeled on a control
consumption estimates. The CityBEUM model utilizes energy data of valve, a heat emitter, an occupancy driven ventilation controlled
U.S. Energy Information Administrations survey data, monthly con- through a Proportional Integral Derivative (PID) controller. In this re-
sumption data and actual weather data to calibrate EnergyPlus soft- search, the modeling process of a complete BES was presented based on
ware. Then, spatial and temporal patterns of building energy con- physical principles of the building (Fig. 2). The BES model was simu-
sumption were estimated using CityBEUM model. The results of this lated using the Matlab/Simulink, and the results with less computa-
study showed that CityBEUM calculates spatial variations of building tional time displayed the temperature changes inside the building. Si-
energy consumption well. Also, by comparing building energy con- mulations of this model showed that Matlab/Simulink can model
sumption using actual atmospheric and weather data, the importance of building energy systems and thus can be a suitable platform for de-
these data has been shown to improve BEM. veloping control techniques need to develop control strategies to save
energy and maintain comfort and efficiency of building energy systems.
3. BEM for building energy consumption In the same year, Duarte et al. [12] modeled energy of EM, a
28000 m2 building with 20 floors (each of them are 1400 m2 ). They de-
Ng et al. [6] found that a large percentage of building energy losses fined the mass of objects in order to calculate the effects of thermal
occurs due to leakage in envelope of the building, the performance mass in internal regions instead of modeling the interior and other of-
efficiency of the building envelope and equipment of the Heating, fice furniture. The apparent area of the internal mass in the core and
Ventilation and Air-Conditioning (HVAC) systems can greatly reduce periphery was 1378 and 300 square meters, respectively, and the in-
energy consumption of buildings. Hence, only energy simulation is not ternal mass of both areas was 182 kg per square meter. The authors set
enough to calculate impermeability (blocking against air penetration). maximum limits for the conversion of solar radiation energy (shortwave
The impermeability equations of energy simulation softwares and ex- radiation) into longwave radiation of inner regions in EnergyPlus. This
isting instructions for input of existing information are for use of low- allowed the core region to distribute longwave radiation to the per-
floor residential buildings, and the determination of impermeability iphery. This study showed that the Envelope Thermal Transfer Value
and permeability with weather, system performance, and envelope (ETTV) was an effective process of the building, but it should be com-
leakage are essentially associated with the pressure subject that phy- pared with envelope heat of different walls of a warm and humid cli-
sical modeling of such a process is not very convenient in existing si- mate. This study showed that the total cooling load of all glass facades
mulation softwares. To solve this problem and to calculate the im- was approximately equal to the cooling load of the concrete facade of
permeability, authors determined coefficients of the available the building. In general, this difference was due to difference in thermal
experimental equation in EnergyPlus software. But this method re- mass, but the conduction distribution versus radiation distribution
quired data for impermeability rate, such as those that were generated varied greatly. Yuan et al. [13] found in 2017, when EMs of existing
using simulation and measurement software such as CONTAM, that buildings are being developed, they always carry unknown and un-
were not available. To address this limitation, they used the key measured parameters that should be determined in simulation models.
building specifications to calculate these coefficients in EnergyPlus. The These authors state that calibration of these parameters is necessary
results of this study showed that a large percentage of building energy before applying building EMs to achieve consumption value. This study
losses will also occur through leakage in envelope of the building as explores several methods in this regard. First, controlling the determi-
building performance improves. Therefore, although impermeability nation problem of a large amount of data can be addressed by the
effects may be important for building energy consumption, but the si- Gaussian Process (GP). Second, several types of different buildings
mulation software ability and precision is limited to calculate im- should be considered a more comprehensive presentation, and finally,
permeability and to improve effects of blocking against air penetration. the model must be consistent with other simulation models other than
Since in urban environments, building energy consumption highly de- building EMs. In this study, they proposed that a metamodel calibrates
pends on weather conditions around the building. In 2016, Jeon and the building EMs with Bayesian approach. This method can calculate
Ham [10] found that thermal conditioning of interior air is primarily various uncertainties. For further improvement of computations, au-
related to the heat transfer between inside and outside of buildings thors recommended a posterior approximation method for analyzing
4
Fig. 2. Overall structure of Building Energy System (BES) model [11].
the evaluation of posterior distributions using the Bayesian approach. Building Energy Modeling (UBEM). In 2018, Andriamamonjy et al. [17]
They used their proposed method to calibrate an EnergyPlus model that compared Modelica with BIM library. After this comparison, they found
was developed and simulated an office building in Singapore. The nu- that the Modelica library had many limitations on EM and used BIM to
merical results of this study showed that this method is more accurate automate building models for BEM. They used Python to do this and
and effective for calibration of building EMs. In 2017, Mauree et al. named their method Building Energy Performance Simulation (BEPS).
[14] suggested a new method for the relationship between atmospheric
model and energy consumption model of a building. The authors aimed
to improve boundary conditions of both models without increasing 4. BEM for building energy optimal design
computational time. In this study, the Canopy Interface Model (CIM)
was associated with CitySim software. CitySim is an urban EM tool [15] In 2016, Sun and McArthur [24] found that energy simulation has
that can find the amount of the energy demand out about building on a become common to codify and stabilize performance systems of con-
city-scale. CitySim offers geometric features for CIM and calculates struction projects, but this type of simulation due to frequent energy
atmospheric variables on a high accuracy. It uses these variables for modeling and costing, especially when building geometry changes fast,
calculating urban energy flows. First, the measured data of the area can rarely be used in the optimal design process. They saw that the
(data for 2015) were used for CIM and its qualitative assessment to energy simulation works more efficiently if the energy model can pro-
make do with high precision vertical sections. Then, the authors com- duce the BIM model. The main obstacle to use the BIM model as a basis
pared local consumption data with data obtained from a meteorological for energy simulation is the transfer of information from BIM tools to
station located outside the city. In both experiments, they showed the energy simulation softwares. In this study, a simple model was created
importance of the building energy software and urban atmospheric and stored using Autodesk REVIT. Then a three-stage study was con-
variables. They also proved that several buildings and urban forms ducted to the process examination of this research. First, REVIT files
could be more sensitive to the local environment. Chen and Hong [16] were converted to GBS files, because this type of files can be used for
In 2018, evaluated the impacts of 3 zoning methods and the use of floor energy simulation software such as eQuest and EnergyPlus, but when
multipliers on simulated energy consumption of 940 offices and retail they tried to import these files, there was an error in these softwares. In
buildings in 3 climate zones using City Building Energy Saver (Ci- the second stage, error sources were investigated and main influential
tyBES). In this study, exact time of the simulation of the EnergyPlus factors in data integration were identified with focus on geometry and
model depends on the number of areas and their levels, precisely. The properties of materials. In the third stage, the results of the eQuest
first zoning method uses a building surface and creates each floor a energy simulation were compared with results of the EnergyPlus energy
thermal area. The second zoning method is an AutoZone method that simulation. These results reported a total of several unrealistic loading
separates the building surfaces into core and peripheral regions. In this and occupancy density assumptions of different types of buildings, and
method, a novel, pixel-based automatic zoning algorithm was devel- options for the type of HVAC system were limited. In 2017, Gerrish
oped. The third zoning method uses the U.S. Department of Energy's et al. [25] proposed a new plan that uses BIM ability to improve EM on
(UDE) reference building prototype shapes. The results show that the both IES-VE and EnergyPlus softwares (Fig. 3). They believe that BIM
simulated energy consumption of each floor with floor multipliers is can be a clear information source for all members of the project team
2.6% higher than EM of each floor which lasts 2–3 times longer. and optimal design.
Compared to AutoZone method, the first zoning method led to the re- In 2017, Bai et al. [26] found that the design should be consistent
duction of thermal loads and less capacity of equipment, 15.2% less with EM credible approach for optimal design of urban energy, but
than fan capacity, 11.1% less than cooling capacity, 11% less than Urban Energy Modeling (UMI) approaches due to data collection and
heating capacity and 7.5% less than cooling loads. The difference in management problems of large-scale building data use a small amount
energy consumption was from −7.6% to 5.1%. The difference in energy of information to describe buildings. This abstraction of building data
consumption of the original sample method was from −12.1% to 19% creates many uncertainties in the modeling and energy simulation. An
compared to AutoZone method, and this difference was higher for important part of the solution to this challenge is to integrate building
heating loads and equipment. This study showed that zoning methods information systems at a city of urban communities with individual
have a great effect on energy consumption simulation of Urban-scale buildings, which are based on Geographical Information System (GIS)
and BIM, respectively. This research examines this challenge and
5
Fig. 3. BIM/BEM information exchange process without open exchange formats [25].
Fig. 4. An overview of Pacific Institute for Climate Solution (PICS) energy design workflow [26].
provides an approach of semantic web technologies such as Web On- follow predefined standards precisely that they cause difficulties to
tology Language (WOL) and Resource Description Framework (RDF) for retrieve related information. Therefore, an urban planning per-
integrating GIS and BIM data. In this approach, authors used an Extract, spective say that it is very useful to facilitate the use of BIM with the
Transform and Load (ETL) tool to convert GIS and BIM data to RDF. promotion of whole BIM.
Then, to generate required information for energy simulation with
EnergyPlus software, they conduct queries on integrated RDFs. Their In the same year, Yuan et al. [27] ranked input parameters ac-
entire cycle of work consists of 4 processes: (1) data integration, (2) cording to their effect on building energy performance despite the ca-
energy simulation, (3) visualization, and (4) design scenario review libration of them, and in order to determine building improvement
(Fig. 4). priorities in optimal design before applying to the building energy si-
In this study, authors used one of the ETL tools called the Feature mulation model. In this paper, authors calibrated and ranked input
Manipulation Engine (FME). FME can read more than 300 different parameters of building EMs at the same time using their metamodel-
data formats, including BIM and GIS data. This engine includes a set of based Bayesian method. Then, they measured monthly electricity con-
data structure transformers and content manipulation that ultimately sumption in 24 months from 2013 to 2014 and simulated electricity use
loads them onto another form. In this study, authors created repetitive using EnergyPlus. Findings of this study showed that calibrated para-
and automated workflow to integrate BIM and GIS data with FME, and meters are usually different from initial hypotheses.
eventually loaded data into a Template Library File (TLF) (Fig. 5).
In addition to an advantage, this research has 3 main constraints: 5. BEM for building energy evaluation
1. The Level of Detail (LOD) of the BIM software is much more than the In 2016, Rahmani Asl et al. [1] explained that more energy saving
GIS modeling scale. opportunities could be made available through the design, but energy
2. A graphical interface is required to allocate resources other than performance assessments were conducted after design basically due to
Information Technology (IT) to easily conduct data queries. lack of effective and useful tools that can help architects. They in-
3. Existing BIMs are not often complete, and contents of data do not troduced a new model called Form-based Energy Performance
6
Fig. 5. An overview of the data integration process [26].
Regression Model (FEPRM) in this study, which could measure energy an approach was created due to amounts of effort and parametric un-
performance in real-time based on BIM instrumentation changes. Their certainty that prevented some problems from occurring during dynamic
model uses a regression model to produce a set of models and simula- modeling of the existing HVAC system. In this study, results of model
tions based on characteristics of the project, and provides integrated comparison showed the average of 20% Root-Mean-Square (RMS) de-
feedback for the designer during conceptual design process. This model viation for hourly heating and cooling during one day of the month.
uses an easy-to-use algorithm and begins to produce the set of random These results allowed a quick assessment of energy consumption and
geometry that are matched with project conditions. The height of the potential improvements for existing buildings. Ahn and Park [30] in
floor-to-floor, the number of levels and the floor plan of these com- 2017 stated that for a quick and easy building energy performance
pounds are different. This algorithm generates convex geometry for assessment of existing buildings should refer to BEM tools that are
each level, but since layouts of each floor are different, the entire costly and time-consuming and require expertise. Authors in this study
composition of the building can not fit into this convex geometry. After developed BEM with Artificial Neural Network (ANN) models. ANN
each combination, FEPRM generates Energy Analysis Models (EAMs) models were created by an EnergyPlus simulator and were sampled by
and sends them in the form of a gbXML file to the DOE energy simu- Monte Carlo method. In this study, MBE and CV-RMSE of EnergyPlus
lation software. After completing all simulations, results are gathered and ANN models were 1.53% and 7.82%, respectively. Authors con-
and regression models are created. This process takes less than an hour cluded that their system requires minimal inputs and provides a precise
for a medium size building. In this process, BIM creates the EAM model. assessment of the hypothetical building performance. In the same year,
In this study, weka version 3.6 software was used for training, testing Ming-ming et al. [31] recommended a new method called normative
and creating regression models. Weka is a Machine Learning (ML) li- building energy consumption model based on international energy
brary based on Java. This software supports preprocessing, classifica- consumption standards for assessing and analyzing current building
tion, regression, clustering and visualization [28]. The FEPRM model energy consumption. The EM simulation was done in EnergyPlus soft-
features of these authors are summarized below: ware which was more accurate, faster, and was performed in a higher
speed. In this research, an analysis and assessment of energy con-
1. FEPRM improves building energy consumption prediction due to the sumption was performed in an office building at the University of
geometry variation, greatly. Petroleum in China. Elbeltgie et al. [32] found that in order to assess
2. FEPRM will immediately provide energy consumption feedback of the impact of building energy consumption, a reliable energy simula-
the building model in BIM. tion model needs predicting energy consumption. During the con-
3. FEPRM can help designer concentrate on the shape analysis of the ceptual design phase, they recommended a new strategy for visualizing
building by fixing other parameters that can affect energy perfor- parametric energy analysis of buildings. This strategy provides energy-
mance, such as construction building characteristics. efficient workflow in accordance with Egyptian standards. In this study,
4. FEPRM uses more geometry related parameters to produce a more a residential building model was developed by the association of
realistic view of building geometry. parametric analysis tools and building EM to create an energy con-
5. The current version of FEPRM only deals with production of convex sumption database. A coordinate diagram was used to imagine the
forms of floor plans. database and evaluate flexibility of energy performance. The benefits of
6. Since FEPRM uses the project specification to create regression this design are: (1) providing a new strategy for visualizing the fore-
models, it doesn’t need any large database to train regression models casting of building energy consumption data, (2) facilitating the deci-
for producing accurate results. sion-making process for energy design alternatives during initial stages,
(3) integrating parametric modeling and energy simulation engine in
In order to investigate the evaluation of uncertainty in energy con- the framework of cooperation, (4) the development of a comprehensive
sumption modeling and the systematic reduction of uncertainty in model with variant energy simulation parameters, and (5) presenting a
modeling existing buildings, Collinge et al. [29] provided a framework simplified and automated approach to map capabilities to simulate and
for hybrid modeling by developing a combination of dynamic BEM of analyze building energy performance with a timely, accurate, and ef-
EnergyPlus simulation software with the empirical model derived from fective process. In order to implement a thermal analysis, each trans-
existing data. This hybrid framework included a dynamic model of formation of the design parameter was simulated. This simulation was
building conditioned space along with an empirical model of the implemented in EnergyPlus which allowed the software to analyze
building HVAC system. The experimental model was calibrated using multivariate parameters to generate different strategies for hypothetical
linear regression of the existing HVAC system temperature and in order variables and related levels. The parametric energy simulation pro-
to develop the estimation of internal loads and relationships between grams: Rhino, Grasshopper, EnergyPlus, DIVA and customized compo-
envelope heat gains/losses and indoor/outdoor temperatures, the nents were also used to illustrate the possible way to integrate energy
building automation system flow measurements were developed. The design with simulation. In the same year, Suesser and Dogan [33] found
authors showed how the relatively simple EnergyPlus model of an ex- that access to urban-scale models is limited to use, and if they are
isting building without any details about the HVAC system could be available, they can only predict the total energy demand according to
combined with an experimental model of a calibrated HVAC system for old data and statistical models. In this study, they sought to provide a
Building Automation System (BAS) set in hybrid framework work. Such new method using GIS data to generate calibrated multi-zone energy
7
Fig. 6. Building geometry colored by floor type and thermal zone divisions [33].
models for assessing the building energy efficiency. The authors of this amount of air permeability in all openings. Surfaces of sample buildings
study, according to recommendations given in ASHRAE 90.1 Appendix in contact with adjacent ones were similar in thermal insulation and
G (ASHRAE 2013) Standard, divided their areas automatically into thermal conditions. Commercial spaces and parking lots were removed
thermal zones [34]. Then these areas were extruded in the form of from the simulation model and different patterns were used for re-
thermal blocks of each floor. Fig. 6 shows buildings and thermal zone sidential use. The adjacent surfaces were considered as thermal in-
geometry. Further, thermal zones were coupled with sample data that sulation. To create operational conditions, standard values of admin-
described materials, loads and settings for each area. Samples of each istrative documents were used.
building were specified, along with detailed description of areas such as This study showed that energy demand reduction measures could be
basements, ground, intermediate and roof floors. Then, adjacent areas discussed (40% would be required by passive methods), and reducing
and the entire building were automatically displayed using a program carbon emissions requires different strategies, including renewable
that plotted graphs and zones. The Excel software was used to compile energy sources. Electricity production combined with mixed systems,
input parameters for each zone, and then these sample parameters were renewable and non-renewable sources, can be considered as a key to
linked to the generation process of the energy model in the Rhino/ long-term sustainability of historic districts. In 2018, Kamel and
Grasshopper software. When all inputs were collected, Archsim and Memari [36] described the advanced tool that automatically simulated
EnergyPlus and Grasshopper workflow were used to categorize simu- building energy and using BIM and modified source code of energy
lations of all models. These models together with a weather file created simulation tools such as EnergyPlus and OpenStudio provided fine-
from previous weather data of the adjacent building, were divided into gained outputs. This tool called Automated Building Energy Modeling
2 separate EnergyPlus simulations of heating and cooling load cali- and Assessment Tool (ABEMAT) receives gbXML file and provides the
bration. One of the simulations was for a week in September and an- amount of heat transfer through every single building envelope com-
other simulation was for 2 weeks in December. This simulation process ponent such as windows, doors, and walls through an automated pro-
of the building and its results showed the importance of analytical cess for users. The results of this study showed that ABEMAT achieved
behavior of separate buildings on a area scale and the zone. very good results.
Caro-Martnez and Sendra [35] stated in 2018 that 40% of final
energy consumption of the European Union is related to buildings and
residential use. They say that from 2006 to 2016, the share of European 6. BEM for building energy efficiency
residential buildings has been analyzed at the national level to make
energy efficiency processes clear and effective, but at the meta-scale of In 2015, Prada-Hernández et al. [37] entered BIM information into
regions, the BEM case study analysis of regions, cities, or neighbour- one of the EM softwares called eQuest. They wanted to study building
hoods is unfeasible. To overcome this limitation, emerging UBEM energy efficiency. They concluded that geometric models could not be
modeling has shown significant progress in assessing the impact of entered as same as building design maps from BIM software to eQuest.
building energy performance on a city-scale. The authors in this study In 2016, Cui et al. [38] found that existing buildings had different
presented an UBEM alternative approach, especially to historic districts specifications and operating conditions that could alter performance of
of cities where many buildings are under protection. They say that the energy model. The current research is mainly based on the test and
preservation laws can limit the scope of measurements and improve error approaches, development of several models and identification of
energy efficiency and the implementation of renewable energy systems. the best performer for a specific building or assumption of a global
These authors introduced a new parameter called Heritage Asset Den- model and is used in different building environments. In this study, they
sity (HAD) to the classification of building blocks. Then, HAD was proposed a meta-learning framework termed Building Energy Model
transferred to the study area and the sample block was selected as the Recommendation (BEMR) system (Fig. 8). BEMR works based on
most frequent HAD sample. The GIS tool was used to describe spatial building physical properties as well as time-series meta-features ex-
distribution of heritage assets and frequency distribution at study level. tracted from practical data and energy consumption and can determine
The file data included grading of heritage buildings. Each study block the most appropriate load forcasting model of each building. In this
with the number of variables and the degree of preservation could in- study, they examined cooling electricity consumption of buildings in
clude or exclude a particular composition of listed buildings. For each the summer. Simulation data were obtained from the simulation of
block, individual buildings were assigned an integer 1–4, respectively, reference building energy consumption models in July. These data were
according to their grading from A to D (Fig. 7). The Design Builder generated at half-hour granularity using EnergyPlus energy simulation
energy simulation software was used for thermal simulation. Each software, which resulted in 48 data points per day and 1488 data points
building was divided into floor thermal areas and shading and reflec- for one month. Three predicted items were tested as follows: (1) single
tion effects of surrounding buildings were taken into account. Without day and one-week test, (2) an extrapolation test, and (3) a real building
considering thermal bridges, this model assumes that there is a large validation test. Experimental results of this study showed that BEMR is
compatible with various types of building from the restaurant to a large
8
Fig. 7. Sample block. Aerial photography (a); Plan with classification of buildings depending on technical regulation period and heritage listing grade (b) [35].
office and gives excellent performance regarding accurate and efficient deterministic definitions. Both optimization and Bayesian calibration
modeling calculations. are approaches for selecting input parameter space vectors that result in
Zhang et al. [39] believe that one of the main challenges of building the lowest calibration error and minimize the objective function (eg,
performance analysis is to determine the main source of input para- calibration error) by changing input parameters of the model. They
meters that has the greatest impact on energy efficiency of buildings. followed the same procedure for each building by creating an input file
Critical input parameters can be determined by use of sensitivity ana- in EnergyPlus using GIS and tax assessment data. Data replication was
lysis models. In this study, they outlined an invariant probabilistic not performed, because energy data was assumed non-existent. For
sensitivity analysis technique to enable screening of the most important possible implementations, sample parameters were calibrated from
parameters in the building energy models. Their proposed probabilistic distributions and obtained using the training set. This method has 2
analysis method measures the average distance between the uncondi- advantages: (1) simulated energy consumption showed uncertainties
tional and conditional probability distribution of an input. The de- that the map faced creating the model, and (2) this model represents
scription of this distribution is based on the Scale-invariant Heat Kernel more precise energy consumption for building. Similarly, in this year,
Signature (SI-HKS). In this study, they extended the SI-HKS method for Chen et al. [40] found that urban energy consumption accounts for
invariant probabilistic sensitivity analysis. The results of this study 30–70% of primary consumption, hence energy efficiency is one of the
demonstrated the application of the proposed method through the use key strategies for developing a sustainable city. The authors of this
of EnergyPlus building-performance modeling tool. In 2017, Sokol et al. paper, based on the user selections of the city building data collection
[3] argued that Urban Building Energy Modeling (UBEM) is a new way and Energy Conservation Measures (ECM), and using EnergyPlus soft-
for exploring energy efficiency solutions at urban spaces. The most ware, provide the CityBES up-to-date analysis feature in automated
common approach for formulating an UBEM is to distinguish a building production and UBEM simulation. CityBES is a new open web-based
into archetypes, describe each type, and validate the model by com- tool for support of urban plans and strategic energy efficiency programs
paring its results with total energy consumption. In this study, authors at city-scale. In this study, the technical details of the CityBES use, in-
presented a more precise method for setting up UBEMs while faced with cluding workflow, key assumptions and major databases for production
incomplete information about buildings. This procedure, by defining and simulation of UBEM, were introduced (Fig. 9). The results of this
unknown or indeterminate parameters in the initial shape description, study showed that: (1) all measurements could result in local energy
calls into possible distributions and, if available, uses measured energy savings of 23–38% per building, (2) replacing lighting with light-
data to correct these distributions by Bayesian calibration. This proce- emitting diodes and adding air economizers to HVAC systems are most
dure was also used for both monthly and annual energy consumption cost-effective with an average payback form 2.0–4.3 years, and (3) in
data. The results of this study showed that both annual and monthly mild climate due to minimal heating and cooling, upgrading HVAC
calibration lead to much better energy consumption compared with old systems or replacing windows will not be economical.
Fig. 8. Framework of Building Energy Model Recommendation (BEMR) system [38].
9
Fig. 9. CityBES key components, data flow, and use cases [40].
In the same year, Heidarinejad et al. [41] reduced building energy during the BIM-BEM process, the standards for building specications
models by simplifying the building exterior and thermal zones and such as Uniformat, and the complete Industry Foundation Classes
provided a framework for creating urban scale rapidly. This reduced (IFC) sets could be used to define the Model View Definition (MVD).
urban scale depended on contribution of influential variables for in- This would allow easy extraction of the required information from
ternal, external, and system thermal loads. OpenStudio Application the existing architectural model to proceed with energy simulations.
Programming Interface (API) was used for automating the process of 2. The aspects considered in this article are important to enhance the
creating the model and displaying the developed framework. This BIM-BEM process; however, one of the most valuable ideas is to
model was simplified to enter into EnergyPlus and create models based extend the life of the BIM to be used during the building operation
on energy consumption classes in OpenStudio. This research focuses on phase. Thus, the framework could be extended to cover the entire
heat transfer processes and allows the classification of the building EM building life cycle and be validated using a case study from early
critical inputs for (1) internal, (2) external, and (3) system inputs. design until the operation phase.
Automation of building simulations needed to simplify the building
envelope, thermal zones and HVAC systems. This research used the best Nowadays, with increasing awareness of the contribution of build-
practices, including (1) geometry, (2) fenestration, (3) thermal zones, ings to global warming, the construction industry has begun to be more
and (4) HVAC systems to simplify the building and create a model. The responsive to energy-efficient buildings. The emergence of BIM with
results of this study showed that accuracy of reduced building energy BEM provides industry with means to address related issues through
models will change only 10% by choosing different thermal zones and integrated energy analysis of buildings. Nevertheless, connecting BIM
will not have a very high effect on building energy efficiency. Hamid technology to BEM includes issues of information exchange among the
et al. [5], in 2017, determined the highest contributing factors of en- participants in the process. While the prevailing view in the literature is
ergy consumption in buildings and recommended ECM for analyzing that there is a need to redefine the process, it is suggested that beyond
the building energy performance to improve the building energy effi- that the success of BEM includes stakeholder management, which can
ciency. In this study, authors analyzed the energy performance of a be implemented through BIM's social capabilities. Hence, Maskil-Leitan
commercial building located in northern Malaysia. They argue that the and Reychav [56] proposed a social concept for connecting BIM to
poor design of the building performance in the early stages can lead to BEM, which is reflected in the model for implementation of Corporate
useless consumption of energy and thermal comfort reduction. In this Social Responsibility (CSR) through BIM for BEM (CSR-BIM-BEM
study, they used the SketchUp software to develop a building model for model). The examination of the relevance of BIM's social capabilities to
energy analysis. They used energy simulation tools (Sefaira SketchUp the BEM process according to this model was conducted on 3 levels: 1)
Plugin) for EM and energy savings estimate of the building. Heat ana- the implications of integrating BIM technology into the BEM process
lysis was also performed to obtain the best ECM. This study achieved were examined through expert interviews, and the need for social
energy savings of more than 52.82% and 36.44%, respectively, in the perception became apparent. 2) the examination of the model's criteria
building energy index (energy usage intensity) and annual cooling en- by the experts supported its ability to promote the BEM process. 3) an
ergy. In 2018, Farzaneh et al. [55] proposed a comprehensive BIM-BEM examination of 2 case studies proved, using social network analysis and
framework initially created based on theoretical and field investigations participant interviews, the model's feasibility, in terms of BEM process
and then this framework was enhanced by a trial investigation using a parameters. A direct link was found between BIM's management cen-
testing case. The proposed framework is illustrated in an easy-to-follow trality, reflected in BIM involvement in stakeholder management, and
process map that includes work/data flows and specifies data ex- progress in the BEM process. These results illustrate the importance of
changers and information requirements. The framework is validated BIM's social role in BEM.
using a complex case study to demonstrate its applicability for different
IES-VE tools. This paper suggests that an adequate and appropriate LOD 7. BEM for building energy management
is required for BEM that increases energy efficiency in buildings. Stra-
tegies to overcome this issue were proposed: In 2015, Lewis et al. [42] describe the capabilities and limitations of
EM in the form of an educational program as building energy man-
1. An advance on the Information Requirements (IRs) by defining LOD agement. This training focuses on a home and change of the wall type,
10
insulation levels, window glazing, etc.. After training and learning, of energy savings. The author states that the simulation model of system
students can conduct EM analysis of other locations. In this way, stu- components can be programmed by Excel and with the equations
dents interpret the results of EM in detail. Overall, the findings of this written in this software. Then, this simulation model can be im-
study indicate that the implementation of this teaching module can plemented with the LCEM tool to estimate the loading effect of old and
improve students' confidence in construction delivery decisions. In new equipment. Simulating this system using BEM data and consuming
2017, Bottaccioli et al. [43] noted that in recent years, a lot of re- electric power and auxiliary equipment (for example, the percentage of
searches have been done for energy loss and carbon emission in Europe power consumption that is not measured and can not be obtained from
that the main purpose of them is to reduce the effects of greenhouse practical data alone) can predict effects of improvements and impact
emission. They believe that Information and Communication Tech- determination of existing practical improvement methods and the es-
nology (ICT) can play a key role in achieving this goal. In this study, the timation of future effects of practical improvement methods. The LCEM
authors presented a software architecture to behavior of energy man- tool can be used to examining and simulating the effect of fan power
agement and simulation in buildings through integration of BIM, In- consumption reduction by replacing the Constant Air Volume (CAV)
ternet of Things (IoT), and GIS data and meteorological services. This with Variable Air Volume (VAV). A lot of this energy is consumed over
integration consisted of 4 steps: (1) real-time visualization of energy the lifetime of a building for daily use, and it is obvious that LCEM
consumption information in the building context, (2) building perfor- implementation will help reduce such consumption.
mance evaluation through EM, (3) simulation exploiting data from the
field and real weather conditions with IoT devices, and (4) the creation 8. BEM for building energy optimization
of GIS urban energy maps along with the energy demand and energy
consumption of each building. They performed their simulations by Optimization is the method to find optimal solutions which these
extracting exact models of BIM. These simulations were performed by solutions have better quality than other solutions. We often seek a
EnergyPlus software. Fig. 10 shows the main strategy of this study. global optimization solution that achieves the best solution in the whole
In 2018, Allegrini and Carmeliet [44] found that the local climate space. Finding this solution as well as having constraints such as noise,
had a strong impact on the energy demand of buildings and comfort and uncertain fitness conditions, instability, and a lot of local issues, can be
health of their inhabitants in building energy management. In this a tedious task [46]. We explored the optimizations made using EM in
study, the authors simulated effects of existing buildings on local cli- this review. According to this review, So far the BEM has been done for
mate in Zurich. This study was conducted with the relationship between Building Energy Optimization (BEO) by Genetic Algorithm (GA), Arti-
Computational Fluid Dynamics (CFD) and building energy simulations ficial Neural Network (ANN), and MBPC, which are described below.
to determine the increase in air and surface temperatures due to the
presences of new buildings. Local weather temperatures were mon- 8.1. BEO by GA
itored due to the weather conditions and ambient air temperature that
had the greatest effect on thermal comfort. The CitySim simulation tool In 2017, Hong et al. [47] found that a lot of researches done to
was used to model the energy fluxes in a quarter with size ranging from develop the BES model, were for saveing energy, so they also did a
a small neighbourhood to an entire city. The results of this study study for building energy use. This research consists of 5 steps: (1)
showed that the formation of local hot spots strongly depends on the collecting the target facility information; (2) establishment of the BES
building geometries, building materials, the strength of buoyancy, the model; (3) calibration of the BES model in accordance with the CV-
wind directions, and wind speeds. In the same year, Tanaka [45] de- RMSE; (4) setting the design variables and objective functions; (5) de-
scribed energy management in HVAC systems. He says that verification velopment of the automatic calibration model using optimization al-
and improvement of the operation of a HVAC system is carried out gorithm. In this study, EnergyPlus software was selected to simulate
using practical data and system simulation. In case studies, the mod- energy models. In order to use the weather data of EnergyPlus software,
eling and simulation process of a system is described by the LCEM tool Incheon district located close to Seoul was selected. As you know,
for heat source system and the HVAC air-side system in office buildings. building energy consumption includes heating, cooling, lighting, elec-
The author says that the annual energy consumption of buildings is tric equipment and domestic hot water. They calculated these con-
measured by examining the energy consumption of buildings that is sumptions manually using EnergyPlus software. After getting the re-
associated with the total floor area. Therefore, floor area is an indicator sults, CV-RMSE was obtained by comparing simulation based manual
energy consumption with actual energy consumption. Given the energy
consumption of the BES model, they found that heating consumption
are higher than cooling consumption. In this study, data entered into
EnergyPlus software was divided into 3 groups: (1) selective insulation
of the building in accordance with Korean insulation standards; (2)
indoor temperatures in the summer and winter based on the criteria of
the Korea Energy Agency and, (3) the lighting parameter of the BES
model of each zone. Then, using GA, the automatic calibration model
was developed for minimizing CV-RMSE. Hong et al., for the reasons
mentioned here, chose this model: (1) implementation of this model is
convenient for systems that are with Java and Python programming
languages and (2) it ensures that there is a high accuracy when we
search for optimal solutions. For the connection between EnergyPlus
and GA, Python programming software was used. In the proposed
model, with automated repetition of the optimization algorithm, the
computational time was reduced which was superior to previous stu-
dies. Similarly, the objective function was improved by setting CV-
RMSE. In the same year, Lara et al. [48] found that optimization was
essentially related to design, and in order to achieve better goals, dif-
ferent design solutions should be compared. They state that the use of
calibrated energy simulation models is considered as a key for success
Fig. 10. Schema of on-site deployment for the proposed solution [43]. of building improvements or their executive management, and
11
Fig. 11. The workflow of the Nagpal et al.'s proposed method. A trained alternative model by setting for determining the exact simulation results of the energy model
with the optimization algorithm [51].
Fig. 12. BEO by developing the methodology of Gerrish et al. method [57].
optimization techniques can improve the credibility of results. This simulate the building energy performance) and static modeling (by
study compares the results of the Brute-Force approach with evolu- developing mathematical methods using simulation results of an input)
tionary optimization methods applied to calibrate a large educational to overcome the weakness of both modeling techniques with improving
building model in the north of Italy. The total design space consists of energy decision-making. In this study, Matlab software was used for
about 72,000 EnergyPlus building models. Two optimization-based implementation of GA optimization. In the same year, Arida et al. [50]
calibrations have been repeated using a Genetic Algorithm (GA) by investigated the modeling methods of building energy systems using
means of jEPlus+EA on a local computer and through parametric si- non-linear auto-regression ANN. The proposed methodology of the
mulations implemented by jEPlus on a cloud service. Then, the quality authors was able to forecast the whole building energy consumptions
of results of evolutionary optimization tools was compared with the given 4 hypothetical variables: dry-bulb and wet-bulb outdoor air
complete parametric study applied to the calibration. The best results temperatures, hours of day and type of days. For determining the op-
from this approach required 137 generations from a total of 1368 si- timal structure of the model by GA model, this paper also optimized this
mulations to be run that were equivalent to those obtained from run- structure and obtained the minimum error between actual and simu-
ning of full parametric analysis. Jafari and Valentin [49] proposed a lated data during the test. The variables of this problem or the para-
decision-making framework called Life Cycle Energy Management meters of the ANN model were (1) the time delay of input, (2) the time
(LCEM), which (1) calculates the economic benefits of energy retro- delay of feedback, and (3) the number of neurons. Statistical indexes
fitting in terms of reduction of life-cycle cost for a specific building such as RMSE and CV-RMSE were used to measure the accuracy of the
during its service life; (2) determines the optimum retrofitting budget model. Data were collected from existing buildings and simulations.
that minimizes the total life-cycle cost of the building during its service- The first part of the study results depends on a parametric study that
life; and (3) selects the optimum energy retrofitting strategy (among was obtained using Matlab. Collected data was used to test and train
available energy retrofitting measures) to maximize the economic proposed models as well as the optimization process. Various ANN
benefits during service-life of the building based on available invest- structures were examined using various inputs and feedback delays.
ments. The research focuses on 3 approaches: (1) considering a com- The results of this study showed that the proposed model can accurately
prehensive economic objective to do energy efficiency decision-making predict the building energy consumption. In 2018, Nagpal et al. [51]
that involves most of the cost-related components of the building; (2) recommended a new automated methodology for reducing auto-cali-
introducing a new and simplified forecasting method to energy by in- brating calculations that uses a data-driven approximation technique to
tegrating dynamic and static modeling, and (3) combining the un- estimate a few unknown performance characteristics of building per-
certainties of energy efficiency decision-making to achieve more accu- formance. This method uses an optimization algorithm (statistical al-
rate results. The developed model in this study develops a new and ternative models) to estimate the properties of missing parameters of
easy-to-measure energy assessment method that was combined with the building instead of using Brute-Force simulations and precise en-
dynamic modeling (using eQuest energy simulation software to gineering models (Fig. 11). In this method, first, the total monthly
12
Table 2
Areas related to energy simulation and software used for EM.
Paper Year Area Softwares used for EM
Ng et al. [6] 2015 Energy consumption EnergyPlus

Bianchi et al. [8] 2015 Energy prediction EnergyPlus
Mohammadpour et al. [18] 2015 Energy estimation DOE
Prada-Hernndez et al. [37] 2015 Energy efficiency eQuest
Rahmani Asl et al. [1] 2016 Energy evaluation DOE
Chaudhary et al. [2] 2016 Energy estimation EnergyPlus
Yi [9] 2016 Energy prediction EnergyPlus, Delight
Winkelmann, DOE, eQuest
Jeon and Ham [10] 2016 Energy consumption DOE
Harish and Kumar [11] 2016 Energy consumption Matlab Simulink
Martin et al. [19] 2016 Energy estimation EnergyPlus
Tian et al. [20] 2016 Energy estimation EnergyPlus
Monari [21] 2016 Energy estimation ESP-r
Sun and McArthur [24] 2016 Energy optimal design EnergyPlus, eQuest
Collinge et al. [29] 2016 Energy evaluation EnergyPlus
Cui et al. [38] 2016 Energy efficiency EnergyPlus
Zhang et al. [39] 2016 Energy efficiency EnergyPlus
Sokol et al. [3] 2017 Energy efficiency EnergyPlus
Hamid et al. [5] 2017 Energy efficiency Sefaira ShetchUp
Duarte et al. [12] 2017 Energy consumption EnergyPlus
Yuan et al. [13] 2017 Energy consumption EnergyPlus
Mauree et al. [14] 2017 Energy consumption CitySim
Shi and OBrien [22] 2017 Energy estimation EnergyPlus
Gerrish et al. [25] 2017 Energy optimal design EnergyPlus, IES-VE
Bai et al. [26] 2017 Energy optimal design EnergyPlus
Yuan et al. [27] 2017 Energy optimal design EnergyPlus
Ahn and Park [30] 2017 Energy evaluation EnergyPlus
Ming-ming et al. [31] 2017 Energy evaluation EnergyPlus
Elbeltagi et al. [32] 2017 Energy evaluation EnergyPlus
Suesser and Dogan [33] 2017 Energy evaluation EnergyPlus
Chen et al. [40] 2017 Energy efficiency EnergyPlus
Heidarinejad et al. [41] 2017 Energy efficiency EnergyPlus
Bottaccioli et al. [43] 2017 Energy management EnergyPlus
Chen and Hong [16] 2018 Energy consumption eQuest
Andriamamonjy et al. [17] 2018 Energy consumption Autodesk REVIT
Li et al. [23] 2018 Energy estimation EnergyPlus
Caro-Martnez and Sendra [35] 2018 Energy evaluation Design Builder
Kamel and Memari [36] 2018 Energy evaluation LCEM
Allegrini and Carmeliet [44] 2018 Energy management CitySim
Tanaka [45] 2018 Energy management EnergyPlus
Farzaneh et al. [55] 2018 Energy efficiency IES-VE
Table 3 In 2018, the thesis developed the methodology of one of the articles
Areas related to energy optimization and software used for EM. [25] (Fig. 12) by proposing a new design using thermal energy opti-
Paper Year Algorithm used Softwares used for EM
mization and temperature estimates obtained from the COMSOL Mul-
tiphysics software to determine the best location the installation of
Salakij et al. [52] 2016 MBPC EnergyPlus cooling appliances, and predicted the best location for heating and
Hong et al. [47] 2017 GA EnergyPlus cooling appliances. In this study, all information about one of the units
Lara et al. [48] 2017 GA EnergyPlus
Jafari and Valentin [49] 2017 GA eQuest
of the Toos Arman Star Apartment Hotel project located on Navvab
Arida et al. [50] 2017 GA Matlab Safavi Avenue in Mashhad, near Imam Reza Shrine (PBUH), was ob-
Banihashemi et al. [53] 2017 ANN EnergyPlus tained from this project, and was modeled on BIM software. Then, the
Nagpal et al. [51] 2018 GA EnergyPlus key parameters of the building were investigated [58] and all of them
Pezeshki [57] 2018 GA COMSOL Multiphysics
were extracted from the BIM software. In the following, EM was in-
vestigated using the energy simulation software. Validation was also
done by comparing the simulation results and the actual results. For
energy consumption data were collected for determining the unknown
optimization, 2 optimization methods of GA and Taguchi method were
and uncertain properties of the hypothetical building. The whole BEM
used to determine the optimal location of cooling devices. In GA, it tried
in this study is an engineering model that uses the EnergyPlus energy
to maximize the satisfaction of building residents by defining the fitness
simulator engine. Then, this model was trained to set the composition
function as approaching the thermal comfort point of 25 °C. The results
of unknown parameter specifications and determine the GOF error
of GA and Taguchi optimizations were introduced into the energy si-
between monthly energy consumption and simulation results of each
mulation software, and again, EM was performed using determined
sample. Further, the properties of this parameter were estimated using
critical points. So, the EM optimization result of GA was compared with
optimization and developed exact models were used in EnergyPlus.
the EM result of two Signal-to-Noise Ratio analysis (SNR) and Analysis
Finally, the energy consumption results of energy models were com-
of Variance (ANOVA) of the Taguchi method and EM result of the unit
pared with initial energy consumption for calibration efficiency as-
studied. In this comparison, the EM results obtained from GA optimi-
sessment. The results of this study showed that this workflow provides
zation showed the better results than the Taguchi and EM of the studied
accurate estimates for building specifications when the building en-
unit. In GA optimization method, there were 2 sensitive points that one
velope information is available which estimate-speed is about 500 times
point showed a better response than the other point [57].
faster than the old ones.
13
based model order reduction to a lower order system model (RE-BEAM)

to be easily integrated with a control algorithm. Then, using the RE-
BEAM model, the Model-Based Predictive Control (MBPC) method was
developed to combine building critical information with control algo-
rithms, such that the building energy consumption was minimized
while comfort conditions were maintained. In this method, the MBPC
was implemented by formulating and solving the finite horizon optimal
control problem at time intervals of physical model sampling. By ob-
taining the maximum load and energy reduction from the simulation of
a building structure, they show the superiority of this method compared
with the traditional constant control methods and control methods that
use an occupants-varying temperature schedule.
Fig. 13. The number of work performed on BEM in various energy areas from 8.3. BEO by ANN
2015 to 2018.
In 2017, Banihashemi et al. [53] considered the combination of
Table 4 discrete and continuous energy parameters to provide a hybrid objec-
New methods provided for EM. tive function of machine learning algorithms in order to optimize the
Paper Year A new way provided for EM
energy consumption of residential buildings. They used a 4-story 4-unit
building to collect their data, and each floor was simulated in-
Rahmani Asl et al. [1] 2016 FEPRM dividually. Each level area was 400 square meters summing up to the
Chaudhary et al. [2] 2016 Autotune total area of 1600 square meters. This building was modeled on the
Collinge et al. [29] 2016 Hybrid modeling
Rhino software, parameterised in Grasshopper software, and simulated
Cui et al. [38] 2016 BEMR
Salakij et al. [52] 2016 BEAM using EnergyPlus software to estimate annual energy. An ANN classi-
Yuan et al. [13] 2017 Posterior approximation method with the fication algorithm was used for teaching and creating a hybrid function
Bayesian approach and the model was finally validated via the weighted average of the
Mauree et al. [14] 2017 Connection between atmospheric model and
resultant error for the performance. This developed model has been
energy consumption model
Shi and O’Brien [22] 2017 Model-cluster-reduce
able to improve the accuracy of the consumption objective functions
Ming-ming et al. [31] 2017 Normative building energy consumption model effectively upon prediction of building energy and optimization pro-
Elbeltagi et al. [32] 2017 Energy-efficient workflow in accordance with blems. The results of this study showed that this new approach solved
standards the integration issue of continuous and discrete parameters of energy in
Bottaccioli et al. [43] 2017 Architecture for management and simulation of
the objective function without integrity and consistency of the building
energy behaviours
Banihashemi et al. 2017 Integration of continuous and discrete energy database.
[53] parameters of energy in the objective function
Li et al. [23] 2018 CityBEUM 9. Results and discussion
Kamel and Memari 2018 ABEMAT
[36]
Nagpal et al. [51] 2018 Data-driven approximation technique
This paper reviewed the recent efforts at BEM, including the whole
building modeling such as prediction & estimation, consumption, op-
timal design, evaluation, efficiency, management, and optimization.
Different BEM approaches, were discussed. Based on the discussions
about different EM approaches of building operations, all of the fol-
lowing areas associated with the orientation along with simulation
software are listed in Table 2.
Also, considering the BEM for optimization, the type of the follow-
up algorithms associated with this orientation and the simulation
software used are listed in Table 3.
According to this review from 2015 to 2018, a lot of work was done
by EnergyPlus. Also, mostly GA has been used in energy optimization
articles. Fig. 13 shows year wise development of BEM and applications.
Some papers introduced new methods for EM. Table 4 shows the new
methods proposed for EM from 2015 to 2018. According this review a
lot of these papers have used another database instead of BIM and there
Fig. 14. The number of work performed on BIM-BEM in various energy areas are only a few studies that used BIM such as database that some of them
from 2015 to 2018. have presented innovations in this context. Fig. 14 shows year wise
development of BIM-BEM and applications. Table 5 show why these
8.2. BEO by MBPC papers have used BIM such as database.
The review reveals that most of the proposed solutions tackled
In 2016, Salakij et al. [52] found that the heat and moisture transfer specific technical aspects rather than looking at how to integrate the
of the building have an important impact on the climate and overall proposed approaches to the entire design process in construction in-
thermal performance of buildings. In this study, they developed a dustry followed by the professionals. BIM information can help to solve
Building Energy Analysis Model (BEAM) to predict the overall heat and this problem and a few of the existing processes focused on BIM to BEM
moisture transfer of the building. Their model was used to calculate the implementation during the design process in mentioned approaches.
heat transfer and continuous moisture of the main hygrothermal effects Indeed, there is only limited information to support an effective and
on buildings. This coupled model was implemented in Matlab and ap- comprehensive work/data flow embedded in the complete BIM to BEM
proved by EnergyPlus. the BEAM model was reduced via a physically process that covers all the building design phases, but these limited
information achieved important findings and innovations shown in
14
Table 5
Reasons why papers have used BIM such as database.
Paper Year Reason for using BIM such as database
Prada-Hernndez et al. [37] 2015 1. Detailed designs of building projects in BIM.

2. Using a more robust and complex BIM model.
3. Share information in a perfect way.
Rahmani Asl et al. [1] 2016 1. Capable of real time quantification of the building energy performance based on the form variation in a BIM tool.
2. Collecting project information from BIM database using an API and providing instant energy usage feedback of the building model in
a BIM interface.
3. Taking note of the option for the user to set parameters for the project before generating the building forms in BIM.
Jeon and Ham [10] 2016 1. Decreasing the gaps between the architectural information in the as-designed BIM and the as-is building conditions on the BIM-based
energy modeling and analysis for existing buildings.
2. Reliable BIM-inspired energy performance modeling and analysis of existing buildings to support retrofit decision-makings.
Sun and McArthur [24] 2016 1. The energy simulation process will be much more efficient.
2. Facilitating energy modeling to optimize building energy performance.
3. Detailed designs of building projects in BIM.
4. Maximizing benefit of changes while minimizing the design cost.
Shi and OBrien [22] 2017 1. Because BIM are often based on space partition, so the number of zones could be reduced by eliminating those similarities selectively.
Gerrish et al. [25] 2017 1. BIM is useful in contributing to effective design progression and information availability.
2. BIM offers an extensible medium for parametric information storage, and its implementation in design development offers the
capability to include BEM parameter-integrated construction information.
3. The minimum required information for BEM is applied to the LOD specification for input into a BIM-integrated BEM data repository,
so information becomes less likely to change and is therefore more representative of the completed building.
4. BIM enables the monitoring and management of this information to allow its collection in a Common Data Environment (CDE).
Bai et al. [26] 2017 1. Detailed designs of building projects in BIM.
Bottaccioli et al [43] 2017 1. Development of complex and rich energy models for simulation and prediction purposes.
2. Validation of building energy models with real data and evaluation of their accuracy.
3. Moving forward the Smart Building and Smart City views.
4. BIM is the central element for the building life-cycle: i) design, ii) construction, iii) management and iv) deconstruction in smart
buildings.
5. Allowing the interoperability and data exchange among different software.
7. Managers and end-users can visualize in BIM model environmental data provided by the related IoT sensors.
8. Managers can run simulation to assess refurbishment actions.
9. BIM establishes a proper knowledge of the buildings.
10. BIM could be reduced by removing excessive details in the architectural model.
Andriamamonjy et al. [17] 2018 1. Creating a very promising context for an even broader application of BEPS.
2. A well-defined BIM-based workflow.
4. API and providing instant energy usage feedback of the building model in a BIM interface.
5. Automating the creation of BEPS models.
Kamel and Memari [36] 2018 1. Providing fine-grained outputs.
2. Making data handling easier for follow-up applications and facilitates interpretation of outputs.
4. Decreasing the gaps between the architectural information in the as-designed BIM and the as-is building conditions on the BIM-based
5. Automating the creation of BEPS models.
6. Facilitating and improving the energy modeling, simulation, and visualizing process.
7. Make the process faster and more accurate by avoiding reentering all the data in different stages of modeling.
Farzaneh et al. [55] 2018 1. Offering significant timesaving potentials by minimizing efficiency issues, legal disputes, added costs, and delays.
Maskil-Leitan and Reychav [56] 2018 1. Promoting BEM process.
2. Managers and end-users can visualize in BIM model environmental data provided.
3. Managers can run simulation to assess refurbishment actions.
Pezeshki [57] 2018 1. Detailed designs of building projects in BIM.
2. Decreasing the gaps between the architectural information in the as-designed BIM and the as-is building conditions on the BIM-based
5. Providing fine-grained outputs.
6. A well-defined BIM-based workflow.
7. BIM is the central element for the building life-cycle: i) design, ii) construction, iii) management and iv) deconstruction in smart
buildings.
9. Development of complex and rich energy models for simulation and prediction purposes.
10. The minimum required information for BEM is applied to the LOD specification for input into a BIM-integrated BEM data
repository, so information becomes less likely to change and is therefore more representative of the completed building.
11. Maximizing benefit of changes while minimizing the design cost.
12. Reliable BIM-inspired energy performance modeling and analysis of existing buildings to support retrofit decision-makings.
13. Taking note of the option for the user to set parameters for the project before generating the building forms in BIM.
14. Using a more robust and complex BIM model for better building energy optimization by BEM.
15
Table 6
Important findings and innovations for BIM-BEM.
Paper Year Important findings and innovations for BIM-BEM
Rahmani Asl et al. [1] 2016 1. FEPRM improves building energy consumption prediction due to the geometry variation, greatly.
2. FEPRM will immediately provide energy consumption feedback of the building model in BIM.
3. FEPRM can help designer concentrate on the shape analysis of the building by fixing other parameters that can affect energy, performance
such as construction building characteristics.
4. FEPRM uses more geometry related parameters to produce a more realistic view of building geometry.
5. The current version of FEPRM only deals with production of convex forms of floor plans.
6. Since FEPRM uses the project specification to create regression models, it doesn't no need any large database to train regression models for
producing accurate results.
Shi and O'Brien [22] 2017 1. Model-cluster-reduce uses a BIM model in EnergyPlus.
2. To automate Model-cluster-reduce process, a library was written in Python to read BIM files and perform simulation tutorials.
3. After training, Model-cluster-reduce reduces the number of thermal areas in an EM building and BIM file was regenerated.
4. Model-cluster-reduce uses the Principal Component Analysis (PCA) clustering technique, determined the initial forms and resolved areas.
5. Simulation time of reduced models in Model-cluster-reduce, reduces to 95%.
Bottaccioli et al [43] 2017 1. Architecture to behavior of energy management and simulation of energy behaviours in buildings through integration of BIM, Internet of
Things (IoT), and GIS data and meteorological services.
2. Real-time visualisation of energy consumption information in the building context.
3. Building performance evaluation through EM.
4. Simulation exploiting data from the field and real weather conditions with IoT devices.
5. The creation of GIS urban energy maps along with the energy demand and energy consumption of each building.
6. Performing simulations by extracting exact models of BIM in EnergyPlus software.
Kamel and Memari [36] 2018 1. ABEMAT reduces energy consumption in buildings.
2. Data related to energy consumption can be either accumulated such as total energy consumption in a thermal zone (e.g., living room) or it
can be fine-grained by providing the heat transfer through each component separately such as each exterior wall, windows, and doors.
3. The existing energy simulation tools such as BIM provide either only the accumulated data concerning the energy consumption of a thermal
zone or the fine-grained data cannot be obtained through an easy, accurate, and fast process.
4. Application of BIM can facilitate the energy modeling, simulation, and visualizing process and make the process faster and more accurate by
avoiding reentering all the data in different stages of modeling.
5. Multiple shortcomings and issues exist in BIM-to-BEM interoperability process (BBIP), which need to be addressed in studies focused in this
area.
6. ABEMAT is a tool developed by compiling different tools, which can automate the energy modeling and simulation process. It is comprised of
three major tools developed by modifying the source code of existing tools.
7. The first tool in ABEMAT receives the gbXML file and performs some corrective actions on the file to eliminate the issues related to building
envelope, which occur in BIM-to-BEM process.
8. The second tool that uses existing modules within OpenStudio source code receives the corrected gbXML file and converts it to an Input Data
File (IDF) file for EnergyPlus. This file converter uses the existing modules in OpenStudio designed for this purpose.
9. The third tool that is a modified EnergyPlus receives the IDF file, performs the energy modeling, and generates text files containing fine-
grained data on heat transfer through every single building envelope component including the windows, doors, walls, floors, and roof.
10. These tools can use fine-grained data for designing more energy efficient houses, performing better evaluation of existing homes for energy
retrofit, and monitoring purposes in Graphical User Interfaces (GUIs) related to energy smart homes.
Table 6. authoring tools and BEM tools currently relies on open exchange
formats of IFC and gbXML. Both provide means of storing geometry
10. Conclusion with attributed data; however, this information is often not accu-
rately exported by the BIM tool or interpretable by the BEM tool.
This review paper has tried to explore the development of EM for 7. Use of BIM authoring tools to do so is costly for building operators
the period 2015–2018. Also non-English publications which find the who would need to purchase software and invest in training to ex-
efforts of different culture developing BEM are not included in this tract relevant information [17,25,57].
review except one article. This study led to three important findings: 8. Also, according to [2,22,24,25,37,54–56], the main reasons for ig-
noring BIM as a database are: 1) low BIM training, and 2) very low
1. Lots of work was done with EnergyPlus as EM software rather than willingness for owners to use BIM due to the exact calculation of
other EM softwares. construction costs. Owners think that the BIM will add to con-
2. A few papers have used BIM as a database for accurate data entry struction costs, while BIM will reduce the construction costs.
[1,10,17,22,24] to [26,36,37,43], and [55–57] that
[1,22,36,43] have presented new methods (Table 4) for EM. Acknowledgments
3. In large-scale, for integration BIM and GIS, graphic interface is
needed to allow non-IT professionals to easily conduct data queries This review is related to my thesis entitled “Energy modeling for
[26]. estimation of indoor temperature of the building based on the BIM
4. Also, in large scale the quantity and quality of BIM remains a big platform”. The authors wish to express sincere gratitude to the anon-
challenge, because the LOD of the BIM software is much more than ymous reviewers for their constructive comments and helpful sugges-
the GIS modeling scale, existing BIMs are often incomplete, and the tions, which lead to substantial improvements of this paper.
data content often does not strictly follow the standard [26].
5. The lack of interoperability between BIM and BEM causes difficul- References
ties in the development and delivery of projects that are seeking
sustainable, efficient, and good energy performance throughout its [1] M.R. Asl, W. Xu, J. Shang, B. Tsai, I. Molloy, Regression-based building energy
life cycle, for example incomplete or incorrect HVAC system mod- performance assessment using building information model (BIM), in: Proceedings of
the ASHRAE and IBPSA-USA SimBuild 2016 Building Performance Modeling
eling and missing information about controls Conference Salt Lake City, UT, 2016, pp. 357–364.
[2,17,22,24,25,37,55–57]. [2] G. Chaudhary, J. New, J. Sanyal, P. Im, Z. O’Neill, V. Garg, Evaluation of autotune
6. According to [17,25,36,43,55] Information sharing between BIM calibration against manual calibration of building energy models, Appl. Energy 182
16
(2016) 115–134. analysis of a building based on normative building energy consumption model, in:
[3] J. Sokol, C. Cerezo Davila, C.F. Reinhart, Validation of a bayesian-based method for Proceedings of the 2017 National HVAC Simulation Academic Annual Meeting
defining residential archetypes in urban building energy models, Energy Build. 134 China, 2017.
(2017) 11–24. [32] E. Elbeltagi, H. Wefki, S. Abdrabou, M. Dawood, A. Ramzy, Visualized strategy for
[4] S. Ferrari, V. Zanotto, Building Energy Performance Assessment in Southern predicting buildings energy consumption during early design stage using para-
Europe, PoliMI SpringerBriefs Free Preview, 2016. metric analysis, J. Build. Eng. 13 (2017) 127–136.
[5] M.F.A. Hamid, N.A. Ramli, N.M.F. Syawal Nik Mohd Kamal, An analysis of energy [33] T. Suesser, T. Dogan, Campus energy model: using a semi-automated workflow to
performance of a commercial building using energy modeling, in: Proceedings of build spatially resolved campus building energy models for climate change and net-
the 2017 IEEE Conference on Energy Conversion (CENCON) Kuala Lumpur, zero scenario evaluation, in: Proceedings of the 15th IBPSA Conference San
Malaysia, 2017, pp. 105–110. Francisco, CA, USA, 2017, pp. 1720–1729.
[6] L.C. Ng, A.K. Persily, S.J. Emmerich, Improving infiltration modeling in commercial [34] T. Dogan, C. Reinhart, P. Michalatos, Autozoner: an algorithm for automatic
building energy models, Energy Build. 88 (2015) 316–323. thermal zoning of buildings with unknown interior space definitions, J. Build.
[7] Z. Pezeshki, S.A.S. Ivari, Applications of BIM: a brief review and future outline, Perform. Simul. 9 (2015) 176–189.
Arch. Comput. Methods Eng. 25 (2) (2018) 273–312. [35] R. Caro-Martinez, J.J. Sendra, Implementation of urban building energy modeling
[8] C. Bianchi, S.M. Lucich, and A.D. Smith, Influence of weather boundary conditions in historic districts. seville as case-study, Int. J. Sustain. Dev. Plan. 13 (2018)
on building energy modeling, in: Proceedings of the 2015 IEEE Conference on 528–540.
Technologies for Sustainability (SusTech) Ogden, UT, USA, 2015, pp. 35–41. [36] E. Kamel, A.M. Memari, Automated building energy modeling and assessment tool
[9] Y.K. Yi, Dynamic coupling between a kriging-based daylight model and building (ABEMAT), Energy 147 (2018) 15–24.
energy model, Energy Build. 128 (2016) 798–808. [37] A.V. Prada-Hernández, J.S. Rojas-Quintero, J.A. Vallejo-Borda, and J.L. Ponz-
[10] J. Jeon and Y. Ham, Evaluating the impact of reflecting the as-is building condition Tienda, Interoperability of building energy modeling (BEM) with building in-
in the BIM-based energy modeling process on building energy simulations, in: formation modeling (BIM), in: SIBRAGEC ELAGEC 2015 Sao Carlos, Brasil, 2015,
Proceedings of the Construction Research Congress 2016, 2016, pp. 2360–2370. pp. 519–526.
[11] V.S.K.V. Harish, A. Kumar, Modeling and simulation of a simple building energy [38] C. Cui, T. Wu, M. Hu, J.D. Weir, X. Li, Short-term building energy model re-
system, in: Proceedings of the 2016 International Conference on Microelectronics, commendation system: a meta-learning approach, Appl. Energy 172 (2016)
Computing and Communications (MicroCom) Durgapur, India, 2016. 251–263.
[12] C. Duarte, P. Raftery, S. Schiavon, Development of whole-building energy models [39] B. Zhang, Y. Liu, R. Rai, V. Krovi, Invariant probabilistic sensitivity analysis for
for detailed energy insights of a large office building with green certification rating building energy models, J. Build. Perform. Simul. 10 (4) (2016) 392–405.
in Singapore, Energy Technol. 6 (1) (2017) 84–93. [40] Y. Chen, T. Hong, M.A. Piette, Automatic generation and simulation of urban
[13] J. Yuan, V. Nian, B. Su, A meta model based bayesian approach for building energy building energy models based on city datasets for city-scale building retrofit ana-
models calibration, Energy Procedia 143 (2017) 161–166. lysis, Appl. Energy 205 (2017) 323–335.
[14] D. Mauree, S. Coccolo, J. Kaempf, J.L. Scartezzini, Multi-scale modelling to evaluate [41] M. Heidarinejad, N. Mattise, M. Dahlhausen, K. Sharma, K. Benne, D. Macumber,
building energy consumption at the neighbourhood scale, PLoS One 12 (9) (2017) et al., Demonstration of reduced-order urban scale building energy models, Energy
1–21. Build. 156 (2017) 17–28.
[15] D. Robinson, F. Haldi, J. Kämpf, P. Leroux, D. Perez, A. Rasheed, et al., CitySim: [42] A.M. Lewis, R. Valdes-Vasquez, C. Clevenger, T. Shealy, BIM energy modeling: case
comprehensive microsimulation of resource flows for sustainable urban planning, study of a teaching module for sustainable design and construction courses, J. Prof.
in: Proceedings of the Eleventh International IBPSA Conference Glasgow, Scotland, Issues Eng. Educ. Pract. 141 (2) (2015) C5014005.
2009, pp. 1083–1090. [43] L. Bottaccioli, A. Aliberti, F. Ugliotti, E. Patti, A. Osello, E. Macii, et al., Building
[16] Y. Chen, T. Hong, Impacts of building geometry modeling methods on the simu- energy modelling and monitoring by integration of IoT devices and building in-
lation results of urban building energy models, Appl. Energy 215 (2018) 717–735. formation models, in: Proceedings of the 2017 IEEE 41st Annual Computer Software
[17] A. Andriamamonjy, D. Saelens, R. Klein, An automated IFC-based workflow for and Applications Conference (COMPSAC) Turin, Italy, 2017, pp. 914–922.
building energy performance simulation with modelica, Autom. Constr. 91 (2018) [44] J. Allegrini, J. Carmeliet, Simulations of local heat islands in Zürich with coupled
166–181. CFD and building energy models, Urban Clim. 24 (2018) 340–359.
[18] A. Mohammadpour, M. Mottahedi, S.S. Amiri, S. Asadi, D. Riley, A. Shafaghat, [45] H. Tanaka, Building energy management: performance verification and system si-
Parametric analysis of building elements on building energy use, J. Teknol. (Sci. mulation, in: B. Muller, H. Shimizu (Eds.), Towards the Implementation of the New
Eng.) 74 (4) (2015) 79–83. Urban Agenda, Springer, Cham, 2018.
[19] M. Martin, D.J.C. Hii, M. Ignatius, and N.H. Wong, Comparison between a simpli- [46] O. Kramer, Genetic Algorithm Essentials in Studies in Computational Intelligence,
fied and detailed building energy model coupled with an urban canopy model, in: Springer International Publishing, 2017.
Proceedings of the 4th International Conference on Countermeasures to Urban Heat [47] T. Hong, J. Kim, J. Jeong, M. Lee, C. Ji, Automatic calibration model of a building
Island National University of Singapore, Singapore, 2016, pp. 1–16. energy simulation using optimization algorithm, Energy Procedia 105 (2017)
[20] W. Tian, S. Yang, Z. Li, S. Wei, W. Pan, Y. Liu, Identifying informative energy data 3698–3704.
in bayesian calibration of building energy models, Energy Build. 119 (2016) [48] R.A. Lara, E. Naboni, G. Pernigotto, F. Cappelletti, Y. Zhang, F. Barzon, et al.,
363–376. Optimization tools for building energy model calibration, Energy Procedia 111
[21] F. Monari, Sensitivity Analysis and Bayesian Calibration of Building Energy Models, (2017) 1060–1069.
University of Strathclyde, Glasgow, United Kingdom, 2016 (Ph.D., Energy System [49] A. Jafari, V. Valentin, An optimization framework for building energy retrofits
Research Unit). decision-making, Build. Environ. 115 (2017) 118–129.
[22] Z. Shi, W. O’Brien, Building energy model reduction using model-cluster-reduce [50] M. Arida, N. Nassif, R. Talib, T. Abu-Lebdeh, Building energy modeling using ar-
pipeline, J. Build. Perform. Simul. 11 (5) (2017) 553–567. tificial, Neural Netw. Energy Res. J. 7 (2) (2017) 24–34.
[23] W. Li, Y. Zhou, K.S. Cetin, S. Yu, Y. Wang, B. Liang, Developing a landscape of urban [51] S. Nagpal, C. Mueller, A. Aijazi, C.F. Reinhart, A methodology for auto-calibrating
building energy use with improved spatiotemporal representations in a cool-humid urban building energy models using surrogate modeling techniques, J. Build.
climate, Build. Environ. 136 (2018) 107–117. Perform. Simul. (2018) 1–16.
[24] X. Sun, J.J. McArthur, Facilitating early stage energy modeling using BIM, in: [52] S. Salakij, N. Yu, S. Paolucci, P. Antsaklis, Model-based predictive control for
Proceedings of the Sustainable Built Environment 2016 (SBE16), Regenerative and building energy management. I: energy modeling and optimal control, Energy
Resilient Urban Environments Toronto, 2016, pp. 1–9. Build. 133 (2016) 345–358.
[25] T. Gerrish, K. Ruikar, M. Cook, M. Johnson, M. Phillip, Using BIM capabilities to [53] S. Banihashemi, G. Ding, J. Wang, Developing a hybrid model of prediction and
improve existing building energy modelling practices, Eng. Constr. Archit. Manag. classification algorithms for building energy consumption, Energy Procedia 110
24 (2) (2017) 190–208. (2017) 371–376.
[26] Y. Bai, P.A. Zadeh, S. Staub-French, R. Pottinger, Integrating GIS and BIM for [54] L.M. Khodeir, A.A. Nessim, BIM2BEM integrated approach: examining status of the
community-scale energy modeling, in: Proceedings of the International Conference adoption of building information modelling and building energy models in egyptian
on Sustainable Infrastructure 2017, 2017, pp. 185–196. architectural firms, Ain Shams Eng. J.l (2017).
[27] J. Yuan, V. Nian, B. Su, Q. Meng, A simultaneous calibration and parameter ranking [55] A. Farzaneh, J. Carriere, D. Forgues, D. Monfet, Framework for using building in-
method for building energy models, Appl. Energy 206 (2017) 657–666. formation modeling to create a building energy model, J. Archit. Eng. 24 (2) (2018)
[28] M. Hall, E. Frank, G. Holmes, B. Pfahringer, P. Reutemann, I.H. Witten, The weka 1–13 (05018001).
data mining software: an update, ACM SIGKDD Explor. Newsl. 11 (1) (2009) 10–18. [56] R. Maskil-Leitan, I. Reychav, BIM's social role in building energy modeling, Clean
[29] W.O. Collinge, J.C. DeBlois, A.E. Landis, L.A. Schaefer, M.M. Bilec, Hybrid dynamic- Technol. Environ. Policy (2018) 1–32.
empirical building energy modeling approach for an existing campus building, J. [57] Z. Pezeshki, Energy Modeling for Estimation of Indoor Temperature of the Building
Archit. Eng. 22 (1) (2016) 1–12 (04015010). on the BIM Platform, Shahrood University of Technology, Shahrud, Iran, 2018, pp.
[30] K.U. Ahn, C.S. Park, Artificial neural network models for building energy predic- 1–166 (M.Sc Faculty of Electrical and Robotics Engineering), http://shahroodut.ac.
tion, in: Proceedings of the 2017 Winter Simulation Conference (WSC) Las Vegas, ir/en/thesis/thesis.php?thid=TK658.
NV, USA, 2017, pp. 2708–2716. [58] Z. Pezeshki, A. Soleimani, A. Darabi, S.M. Mazinani, Thermal transport in: building
[31] Z. Ming-ming, Y. Guang, Q. Shu-nian, Z. Wei-jie, L. Zheng-wei, Energy saving materials, Construct. Build. Mater. 181 (2018) 238–252.
17

Application of BEM and using BIM database for BEM_ A review

Uploaded by

Copyright:

Available Formats

Application of BEM and using BIM database for BEM_ A review

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Application of BEM and using BIM database for BEM_ A review

Uploaded by

Copyright:

Available Formats

Journal of Building Engineering 23 (2019) 1–17

Contents lists available at ScienceDirect

Journal of Building Engineering

Application of BEM and using BIM database for BEM: A review T

1. Introduction factors such as productivity characteristics, thermal eﬃciency, energy

Fig. 2. Overall structure of Building Energy System (BES) model [11].

Fig. 5. An overview of the data integration process [26].

Fig. 8. Framework of Building Energy Model Recommendation (BEMR) system [38].

Ng et al. [6] 2015 Energy consumption EnergyPlus

based model order reduction to a lower order system model (RE-BEAM)

Prada-Hernndez et al. [37] 2015 1. Detailed designs of building projects in BIM.

You might also like