Vous êtes sur la page 1sur 16

sustainability

Article
Feasibility Study on Parametric Optimization of
Daylighting in Building Shading Design
Kyung Sun Lee 1, *, Ki Jun Han 2 and Jae Wook Lee 3, *
1 School of Architecture, Hongik University, 94 Wausan-ro, Mapo-gu, Seoul 02481, Korea
2 Digit, 12, Dongmak-ro 2-gil, Mapo-gu, Seoul 04071, Korea; dbxkrvk2@naver.com
3 School of Architecture, University of Illinois at Urbana-Champaign, Champaign, IL 61820, USA
* Correspondence: ksunlee01@gmail.com (K.S.L.); jlee764@illinois.edu (J.W.L.); Tel.: +82-10-3731-2170 (K.S.L.);
+1-217-402-4335 (J.W.L.)

Academic Editors: Francisco Lozano and Marc A. Rosen


Received: 1 August 2016; Accepted: 18 November 2016; Published: 24 November 2016

Abstract: Shading design to optimize daylighting is in many cases achieved through a designer’s
sense based on prior knowledge and experience. However, computer-assisted parametric techniques
can be utilized for daylighting design in an easy and much more accurate way. If such tools are
utilized in the early stages of a project, this can be more effective for sustainable design. This study
compares the conventional approach, which depends on a designer’s sense of judgment to create
optimal indoor lighting conditions by adjusting louver shapes and window patterns, with the
approach of making use of genetic algorithms. Ultimately, this study discusses the advantages and
disadvantages of those two approaches. As a starting point, 30 designers were instructed to design
a facade by manually adjusting several input parameters of shading. The parameters govern six
kinds of louver and window types, with the ratio of analysis grid surface area achieving a daylight
factor of 2%–5%. Secondly, input parameters were automatically created by using genetic algorithm
optimization methods to find optimal fitness data. As a conclusion, conventional approaches result
in a strong disposition toward designing certain shading types represented by linear relationships.
Computer-assisted daylight simulation can help influence this, being effective when dealing with a
large amount of data and non-linear relationships.

Keywords: parametric optimization; daylight; genetic algorithm; facade design; computer simulation

1. Introduction
The importance of sustainability in green building design is increasing. About 70% of all energy
usage is consumed in urban areas, and lighting alone takes up around 15% of total building energy
consumption [1–3]. Therefore, architects and engineers are actively looking for solutions to reduce
lighting energy. Beside energy issues, lighting plays a significant role in buildings and urban areas
more generally. Lighting, which can be divided into natural and artificial sources, can affect occupants’
comfort and satisfaction levels [4].
Inside a building, access to daylight makes the indoor environment healthier and more
comfortable for occupants; natural lighting is also linked with greater productivity [5]. When designed
with proper glare control and minimized solar heat gain, daylighting provides high-quality light
while significantly reducing energy consumption for lighting and cooling. An effective facade shading
design should contribute to the creation of such an environment that will reduce building energy
expenditure and optimize daylight distribution [6–8].
There are several factors involved in the control of lighting energy, but the impact of the building
skin on lighting is considerable. Formerly, architects typically designed the building skin based on
previous experience and intuition. In order to achieve low-energy design, environmental engineers

Sustainability 2016, 8, 1220; doi:10.3390/su8121220 www.mdpi.com/journal/sustainability


Sustainability 2016, 8, 1220 2 of 16

sometimes support the design by mathematical analysis. However, their involvement is often requested
later in the design process. In the early design stages, passive design strategies can be optimized by
considering an active design, which potentially improves the energy efficiency of buildings much
more effectively [9–12]. Therefore, there is a need to integrate scientific methods with design to help a
designer achieve low-energy design easily and effectively.
With the development of a design-integrated energy simulation, architects and designers can
now more easily construct and design a building. Simulation programs which are combined with
architectural daylighting design tools, such as DIVA-for-Grasshopper/Rhino were tested and used to
design a building envelope [13]. With daylighting design, research-related building energy simulations
such as Esp-r [14] or EnergyPlus [15] for energy-efficient buildings were also implemented in various
areas. Furthermore, some simulation programs provide parametric methods or algorithms for
optimization, which can help to find optimal solutions in a faster and easier way. Several pieces of
research-related optimization techniques for transparent building facades [16] and genetic algorithms
for design variables [17] show that those techniques can be an effective way to harmonize design and
energy fields. Recently, utilizing digital technology based on computer hardware and software has
been implemented in design projects, and this is known as parametric design [18,19]. Such technology
can be actively and positively used in the early design stage for design optimization. For example,
parametric modeling is related to geometric information which includes several parameters [20,21],
and it can automatically modify parameters such as elevation proportion, louver size, and angle
without reorganizing the whole model. Furthermore, other related research on energy performance of
a building envelope and optimization of an engine or heat exchanger’s performance successfully adopts
an optimization genetic algorithm [22–24], demonstrating how the application of a computer-aided
algorithm has very important potential in architectural design.
In this study, other viewpoints and methods are introduced to clarify the distinction and
importance compared to previous research. Above all, differences in daylighting between two design
approaches will be tested: (A) design alternatives performed by an architectural designer’s experience
and intuition; (B) design optimization by computer simulation will be tested. By comparing the
architectural designer and the computer simulation, it can help to identify: (1) how much more
daylighting performance a designer can achieve with a simulation program; (2) which louver and
window types (defined as horizontal, vertical, random horizontal, random vertical louver types,
and Delaunay and Voronoi screen types) can affect daylighting performance.
Secondly, this study explains how parametric optimization methods can be applied in daylighting
design simulation. As a result, it can help to identify the advantages and disadvantages of daylighting
optimization between designers and computer simulations. Also, this paper suggests how a designer’s
manual and simulation methods can harmonize daylighting performance and facade design.
Architects design buildings through an iterative process using light simulation feedback.
Light analysis can be simulated for either a simple or complex building facade. In other words,
design processes are inseparable from simulations, to the extent that design can be controlled by the
results of simulations. Thus, the main purpose of this research is to analyze and compare the uses of
simulation results in buildings, specifically regarding the aspect of daylighting design. As a result,
the processes of finding optimal solutions for daylighting design can be compared using a computer
algorithm. Also, the proposed solutions can be tested on both a conventional envelope design as well as
a computer-aided one, such as with Delaunay and Voronoi screens.

2. Methodology

2.1. Research Procedures and Methods


The main goal of this research is to find input parameters which can maximize fitness, i.e., the ratio
of analysis grid surface area achieving a daylight factor (DF) of 2%–5% with variation of shape and
combination of shading and window size in given conditions. This DF metric is a rather simplistic
Sustainability 2016, 8, 1220 3 of 16

measure of daylighting performance, thus it is relatively easy to understand and be applied by an


architectural designer in terms of simulation and analysis. The research was performed in the following
three steps:

• To find optimal fitness with manually input parameters adjusted by 30 persons,


• To find optimal fitness data with Galapagos, a component inside Grasshopper that can
optimize shape,
• To find efficiency and improvement of optimization in green building facade design through a
comparative analysis of manually adjusting input parameters and genetic algorithms.

2.1.1. Base Simulation Model and Input Data


For simulation, a typical office space facing south is used as a base model, which measures 5 m
(16.4 ft) depth by 7.2 m (23.6 ft) width, with a floor-to-floor height of 3 m (9.8 ft), as shown in Figure 1.
South orientation was selected, which is most favorable, since there are no significant differences in DF
by changes in orientation. Input materials for floor, wall, and ceiling were implemented as default
values which are supported by DIVA in Table 1, and the selected glazing type is 8 mm double-pane
clear glass.

Figure 1. Base model.

Table 1. Combination of input parameters for louver/window type.

Input Materials Material Name in DIVA Material Properties


This is a purely diffuse reflector with a
1 Wall GenericInteriorWall_50
standard wall reflectivity of 60%
Material for typical ceilings as
2 Ceiling GenericCeiling_70
suggested by IES-LM-83 [25]
Visual transmittance: 80%
3 Window Glazing_DoublePane_Clear_80
Visual transmissivity: 87%
This is a purely diffuse reflector with a
4 Floor GenericFloor_20
standard floor reflectivity of 20%

2.1.2. Simulation Program and Basic Concept


As tools for parametric design and indoor lighting analysis, Grasshopper and DIVA are used [26].
Grasshopper is a graphical algorithm editor, a plug-in for Rhinoceros, a NURBS (Non-uniform rational
B-spline curve) modeling software. It does not require knowledge of programming or scripting, and it
allows designers to create simple to complex shapes. Once a definition for the Grasshopper is created,
Sustainability 2016, 8, 1220 4 of 16

designers can easily adjust input parameter values and generate a variety of outputs for deformation
and modification. Grasshopper is used for this experiment in order to easily adjust input parameters
to find optimal fitness (output) manually. Using the Galapagos component, which is linked to those
programs, an optimization method based on genetic algorithms in daylight fields can be applied using
a personal computer.
DIVA-for-Grasshopper is a daylighting and energy modeling add-on for Grasshopper and an
environmental simulation tool which can calculate data such as DF or solar irradiation parametrically.
If it is done manually, people should adjust the value of the input parameters to find optimal fitness.
When using Galapagos, associated input parameters execute an algorithm to automatically calculate
the optimized output.

2.1.3. Daylight Factor (DF)


There are different types of metrics for daylighting. For instance, DF, daylight autonomy,
and useful daylight illuminance are well-known metrics [7]. However, each metric comes with
disadvantages as well as unique advantages. Daylight autonomy counts all the values over 300
lux, which is regarded as excessive light level. Also, useful daylight illuminance does not count any
value over 2000 lux or below 100 lux. We have acknowledged those contradictory phenomena and
have chosen DF for the research. Although it is not perfect, it is widely known and easy to understand.
Even though the metrics are measured for either static or dynamic conditions, they are commonly
considered static at a single point in time. In fact, the stability of DF is itself one of the benefits,
regardless of the time of day or year, even assuming an overcast sky [27,28]. Also, DF is easy to
calculate for buildings in the real world or models in the lab with illumination meters. Most European
and Asian countries, such as the UK (BREEAM), France (HQE), Japan (CASBEE), and South Korea
(G-SEED), use DF for green building assessment as part of their guidelines [29–32].
DF is a metric used to quantify the amount of diffuse daylight in a space. Diffuse daylight is
light that has been scattered in the atmosphere before reaching the Earth’s surface. It is a ratio that
represents the amount of illumination available indoors relative to the illumination present outdoors
at the same time under an overcast sky. The design day for DF calculations is based on the standard
International Commission on Illumination (CIE) overcast sky for 21 September at 12:00 p.m. [27,28].
It is usually measured at the height of the work plane (i.e., a desktop) under a standardized CIE
overcast sky [33].

DF = (Ein/Eext) × 100 (1)

Ein: Interior illuminance at a fixed point on the work plane. Eext: Exterior illuminance under an
overcast sky.
In the UK (CIBSE Lighting Guide 10), DF is categorized into three levels [34]:

(1) DF under 2 is not adequately lit and artificial lighting will be required,
(2) DF between 2 and 5 is adequately lit but artificial lighting may be in use for part of the time,
(3) DF over 5 is well-lit and artificial lighting is generally not required except at dawn and dusk,
although glare and solar gain may cause problems.

According to the British Standards Institution (BS 8206 part 2 CIBSE), a space with a mean DF
between 2% and 5% is considered well-lit and requires little or no additional lighting during the
day [35].
Also, the Indoor Environmental Quality in the LEED rating system recommends a 2% DF in
at least 75% of all spaces occupied for critical visual tasks (LEED NC-v2.2 EQc8.1: Daylight and
Views) [36].
Sustainability 2016, 8, 1220 5 of 16

2.1.4. Louver/Window Type and Input Parameters


In this experiment, louver and window types are formed from a combination of eight kinds of
input parameters: (1). Angle: horizontal/vertical angle of the louver, from −60 to 60 degrees; (2). Num:
number of louvers; (3). Depth: louver depth; (4). U Num.: number of window panel divisions in
the horizontal direction; (5). V Num.: number of window panel divisions in the vertical direction;
(6). WWR: window to wall ratio; (7). Pattern: Delaunay, Voronoi division pattern number (number
of seed points of patterns); (8). Random: Delaunay, a random variable to be added to the Voronoi
type. Each louver and window type has 125 variations with a combination of the three types of input
parameters out of eight input parameters (Table 2). The parameters govern six kinds of louver and
window types (horizontal, vertical, random horizontal, random vertical louver types, and Delaunay
and Voronoi screen types) (Tables 3 and 4).

Table 2. Input parameters.

Input Range Range


Unit Step Size Variables
Parameters (min.) (max.)
Horizontal/vertical angle
1 Angle (degree) −60 60 30 5
adjustment of the louver
2 Num. (integer) The number of louvers 5 13 2 5
3 Depth (10 cm) Louver depth 2 6 1 5
U Num. Number of window panel
4 5 13 2 5
(integer) divisions in the U direction
V Num. Number of window panel
5 5 13 2 5
(integer) divisions in the V direction
Ratio of remove from the entire
6 Remove (-) 0.2 (20%) 0.8 (80%) 0.15 (15%) 5
window panel
7 Pattern (step) Number of seed point patterns 10 50 10 5
8 Random (-) Random variable for seed points 1 5 1 5

Table 3. Combination of input parameters for louver/window types.

1 2 3 4 5 6
Random Random Delaunay Voronoi
Input Horizontal Vertical
Horizontal Vertical Pattern Pattern
Parameters Louver Louver
Louver Louver Screen Screen
√ √
1 Angle √ √ - - - -
2 Num. √ √ - - - -
3 Depth -
√ -
√ - -
4 U Num. - - √ √ - -
5 V Num. - - √ √ -
√ -

6 Remove - - √ √
7 Pattern - - - - √ √
8 Random - - - -
No. total variations 125 125 125 125 125 125

: Applied parameters.
55 VVNum.
Num. -- -- √√ √√ -- --
66 Remove
Remove -- -- √√ √√ √√ √√
77 Pattern
Pattern -- -- -- -- √√ √√
88 Random
Random -- -- -- -- √√ √√
No.
No.total
totalvariations
variations 125
125 125
125 125
125 125
125 125
125 125
125
Sustainability 2016, 8, 1220 6 of 16
√:√:Applied
Appliedparameters.
parameters.

Table
Table4. Louver/window types.
Table 4.
4. Louver/window
Louver/windowtypes.
types.
(a)
(a)Horizontal
Horizontallouver
louver (b)
(b)Vertical
Verticallouver
louver
(a) Horizontal
Parameters:
Parameters: louver
Angle,
Angle, Num.,
Num.,Depth
Depth (b) Vertical
Parameters:
Parameters: louver
Angle,
Angle,Num.,
Num.,Depth
Depth
Parameters: Angle, Num., Depth Parameters: Angle, Num., Depth

Sustainability
Sustainability2016,
2016,8,
8,1220
1220 66ofof16
16
Sustainability 2016,
Sustainability 8,8,
2016, 1220
1220 6 6ofof
1616
(c) Random horizontal louver (d) Random vertical louver
(c)
(c)Random
Randomhorizontal
Parameters: U Num., louver
horizontal
Vlouver
Num., WWR (d)
(d)Random
Randomvertical
Parameters: Num.,louver
Uvertical louver
V Num., WWR
Parameters:
Parameters: UNum.,
U Num.,VVNum.,
Num.,WWR
WWR Parameters:
Parameters: U Num.,VVNum.,
U Num., Num.,WWR
WWR

(e)
(e) Delaunay
Delaunay pattern
pattern screen
screen (Figure
(Figure A1)
A1) ** (f)
(f) Voronoi
Voronoi pattern
pattern screen
screen (Figure
(Figure A2)
A2) **
(e) DelaunayWWR,
Parameters: patternPattern,
screen (Figure
Random A1) * (f)Parameters:
Voronoi pattern
WWR, screen (Figure
Pattern, A2) *
Random
(e)(e)Delaunay pattern
DelaunayWWR,
Parameters: patternscreen
screen(Figure
Pattern, (FigureA1)
Random A1)* * (f)(f)Voronoi
Voronoipattern
Parameters: WWR,screen
pattern screen(Figure
Pattern, A2)
(Figure
Random A2)* *
Parameters: WWR, Pattern, Random Parameters: WWR, Pattern, Random
Parameters:
Parameters:WWR,
WWR,Pattern,
Pattern,Random
Random Parameters:
Parameters:WWR,
WWR,Pattern,
Pattern,Random
Random

Green: Solid wall; Blue: Window glass. * The detailed procedures described in Figures are found in Appendix A.
Green:
Green:Solid
Solidwall;
wall;Blue:
Blue:Window
Windowglass.
glass.**The
Thedetailed
detailedprocedures
proceduresdescribed
describedininFigures
Figuresare
arefound
foundin
inAppendix
AppendixA.
A.
2.2. Green: Solid
Green: Solid
Methods wall;
and wall; Blue: Window
Blue: Window
Procedures glass.
glass. * The detailed procedures described in Figures are found in AppendixA.A.
* The
of the Experiment detailed procedures described in Figures are found in Appendix
2.2.
2.2. Methods
Methods and and Procedures
Procedures of of the
the Experiment
Experiment
2.2.
2.2.Methods
2.2.1. Methods
Manual and
and Procedures
Procedures
Approach ofof ofthetheExperiment
Adjusting Experiment
Input Parameters
2.2.1.
2.2.1. Manual
Manual Approach
Approach of of Adjusting
Adjusting Input Input Parameters
Parameters
2.2.1. Thirty
2.2.1.Manual
Manual students
Approach
Approach with of at least
Adjusting
of Adjusting four years
Input
Input of architectural design education participated in this
Parameters
Parameters
Thirty
Thirty students
experiment. students
They have
with
with at least
least four
four years
fairatunderstanding years ofGrasshopper
ofof
architectural
architecturaland design
design
DIVA.
education
education participated
They areparticipated
in this
also familiarinwith this
experiment.
Thirty
Thirty
experiment. They
students
students
They have
have fair
with
with
fair understanding
atat least
least four
understanding four of
years
years
of Grasshopper
of architectural
of and
architectural
Grasshopper and DIVA.
design
design
DIVA.
the concept of DF and indoor lighting environments through their sustainable architecture classes.
They
educationare
education
They are also familiar
participated
participated
also familiar with
inin
with the
this
this
the
concept
experiment. of
experiment.
concept DFTheyand
They
of DF and indoor
have
have fair
indoor lighting
fairunderstanding environments
understanding
lightingare ofof
environments through
Grasshopper
Grasshopper their
and
and sustainable
DIVA.
DIVA. They
They architecture
are
arealso
alsofamiliarclasses.
familiar
through their sustainable architecture classes. The with
with The
the
the
The guidelines for the experiment given as:
guidelines
concept
conceptofofDF
guidelines for
for the
DFand experiment
indoor
indoorlighting
andexperiment
the are
lighting given
are given as:
environments
environments
as: through
throughtheirtheirsustainable
sustainablearchitecture
architectureclasses.
classes.The The

guidelines
Adjust
guidelines
 Adjust for
for the
three the experiment
parameters,
experiment are
and
aregiven
given as:
proceed as: with DIVA simulations. Check visual information of DF
Adjust three
three parameters,
parameters, and and proceed
proceed withwith DIVA
DIVA simulations.
simulations. CheckCheck visual
visual information
information of of DF
DF
range
range being
being 2%–5% of the analysis grid surface area in the color index (Figures 2 and 3).
  Adjust
Adjust
range three2%–5%
three
being parameters, of
of the
parameters,
2%–5% theandanalysis
and proceed grid
proceed
analysis grid surface
with
with DIVA
surface area
DIVAarea in
in the
the color
simulations. index
colorCheck
simulations. (Figures
indexvisual
Check 22 and
and 3).
visualinformation
(Figures information3). ofofDF DF
 range
Calculate
Calculate the output (fitness: the ratio of analysis grid surface area having a DF value of 2% to
range being
beingthe output
2%–5%
2%–5% of(fitness:
ofthethe the
analysis
analysisratio
grid
gridof analysis
surface
surface area grid
areain surface
the
in thecolor area
colorindex having
index a
(Figures DF
(Figures 2 value
and
2 and 3).of
3). 2% to
5%)
  Calculateas
as operation
Calculate
5%) thetheoutput
operation result.
output (fitness:
result.(fitness:the theratio
ratioofofanalysis
analysisgridgridsurface
surfacearea
areahaving
havinga aDF DFvalue
valueofof2% 2%toto
 5%) Adjust the
5%)asasoperation
Adjust input parameters
inputresult.
operation
the result.
parameters appropriately
appropriately and and choose
choose the the form
form of of louvers
louvers which
which create
create aa high
high
fitness.
  Adjust
Adjust
fitness. thethe input
input parameters
parameters appropriately
appropriately and
and choose
choose the
theform
form of louvers
of louvers which
which create
create a ahigh
high
 fitness.
Check if three kinds of parameters are set properly using visual
fitness.if three kinds of parameters are set properly using visual information of DF in the color
Check information of DF in the color
Sustainability 2016, 8, 1220 7 of 16

• Calculate the output (fitness: the ratio of analysis grid surface area having a DF value of 2% to 5%)
as operation result.
• Adjust the input parameters appropriately and choose the form of louvers which create a
high fitness.
• Check if three kinds of parameters are set properly using visual information of DF in the color
index (Figures 2 and 3). This assists participants in determining which parameter combination
needs to be used for the next run and if the space needs more or less light.
• Repeat at least 10 times for each window/louver type and adjust the input parameters in order to
get to 2016,
Sustainability the highest
8, 1220 possible optimal value. 7 of 16
Sustainability 2016, 8, 1220 7 of 16

Figure 2.
2. Analysis grid
grid surface of
of DF range
range between 2%
2% to 5%.
5%.
Figure 2. Analysis
Figure Analysis grid surface
surface of DF
DF range between
between 2% to
to 5%.

Figure 3. Analysis grid surface of DF.


Figure 3.
Figure 3. Analysis
Analysis grid
grid surface
surface of
of DF.
DF.

2.2.2. Genetic Algorithm Optimization Methods Using Galapagos


2.2.2. Genetic
2.2.2. Genetic Algorithm
Algorithm Optimization
Optimization Methods
Methods Using
Using Galapagos
Galapagos
Galapagos is used as the other experiment method to find optimum fitness values and three
Galapagos is
Galapagos is used
used asas the
the other
other experiment
experiment method
method to to find
find optimum
optimum fitnessfitness values
values and
and three
three
parameter values for each louver/window type using genetic algorithm optimization methods. The
parameter
parameter values for each louver/window type using genetic algorithm optimization
for each louver/window type using genetic algorithm optimization methods. methods. The
main reason why the genetic algorithm was selected is that it is the only algorithm to find optimal
mainmain
The reason why why
reason the genetic algorithm
the genetic was selected
algorithm is that isit that
was selected is theit only
is thealgorithm to find optimal
only algorithm to find
solutions with Rhino and Grasshopper software. In addition, this software is very widely used for
solutions with Rhino and Grasshopper software. In addition, this software
optimal solutions with Rhino and Grasshopper software. In addition, this software is very widely is very widely used for
architectural drawings [37]. In order to have equal conditions to those of manual methods, Galapagos
architectural
used drawingsdrawings
for architectural [37]. In order
[37].to
Inhave
order equal conditions
to have to those ofto
equal conditions manual
those methods,
of manualGalapagos
methods,
is set with three parameter values with four variables: Max. Stagnant: 50, Population: 20, Maintain:
is set with is
Galapagos three parameter
set with values with
three parameter four variables:
values Max. Stagnant:
with four variables: Max. 50, Population:
Stagnant: 20, Maintain:
50, Population: 20,
20%, Inbreeding: 50%. The Population, set as 20, means that evolution for each generation DIVA
20%, Inbreeding: 50%. The Population, set as 20, means that evolution for
Maintain: 20%, Inbreeding: 50%. The Population, set as 20, means that evolution for each generation each generation DIVA
simulates 20 times. In this case, Max. Stagnant, set as 50, calculates up to 50 generations. The total
simulates
DIVA 20 times.
simulates In this case,
20 times. In thisMax.
case,Stagnant, set as 50,set
Max. Stagnant, calculates up to 50 generations.
as 50, calculates The total
up to 50 generations.
number of simulations should be 1000, and the skip and filtering-method algorithm was
number
The of simulations
total number shouldshould
of simulations be 1000, and and
be 1000, the the
skip
skip andandfiltering-method
filtering-method algorithm was
algorithm was
programmed to avoid duplications. Stagnant by 20 of population. However, each louver/window
programmed to avoid duplications. Stagnant by 20 of population. However,
programmed to avoid duplications. Stagnant by 20 of population. However, each louver/window each louver/window
type is composed with 125 cases in this research, and the maximum value of fitness could be found
type is composed with 125 cases in this research, and the maximum value of fitness could be found
within ten generations.
within ten generations.
3. Result
3. Result
Table 5 shows the results of fitness (the ratio of analysis grid surface area having a DF value of
Table 5 shows the results of fitness (the ratio of analysis grid surface area having a DF value of
Sustainability 2016, 8, 1220 8 of 16

type is composed with 125 cases in this research, and the maximum value of fitness could be found
within ten generations.

3. Result
Table 5 shows the results of fitness (the ratio of analysis grid surface area having a DF value of
2% to 5%) by manual judgment against optimal fitness data from Galapagos. A total of 30 people
ran the simulation, ten times for each louver or window type, to reach the highest fitness values.
The average person’s data for each trial were compared, and the result (Table 5) includes the maximum
and minimum of average, range, and standard deviation value from the average of the ten simulations.
In order to compare the ratios to approach the maximum value for each type of louver or window
under the same conditions, the original data need to be adjusted. The maximum fitness value for
each type results from using Galapagos. Therefore, each type’s fitness is adjusted to 100 based on
the Galapagos result for the comparison. Similarly, the other type’s fitness was also readjusted by
applying the same ratio. The maximum of the average values for manual input ranges from 65% to
94%, and the minimum values of the average ranges from 32% to 87% compared to the optimization
value by Galapagos. Among the six types, the vertical louver type reached the highest value (94%)
by manual input, and the Voronoi screen type reached the lowest value (65%). The horizontal louver
type has the highest average range (51%), and the random vertical and random horizontal louver
type have the lowest average (3%). The average of the SD (standard deviation) of each trial ranges
from 6 to 27. The horizontal louver type has the highest average of SD (27%) and the random vertical
louver type has the lowest average of SD (6%). The maximum fitness value from the Galapagos
optimization method ranges from 44% to 86%. The horizontal louver type has the highest value (86%),
and the Voronoi screen type has the lowest value (44%). The minimum original fitness value from the
Galapagos optimization method ranges from 0% to 36%.

Table 5. Statistical data from manual input vs. Galapagos program.

Method Manual Approach (Adjusted Data) Galapagos (Original Data)


Fitness Value (%) Max. of Min. of Max.
Aver. SD Min.
Louver/WindowType Avg. Avg. Parameter Value
Horizontal 86
83 32 27 0
(Type 1) 0/5/5
Vertical 46
94 76 9 26
(Type 2) 0/11/6, 0/13/0
Random horizontal 75
70 67 16 20
(Type 3) 5/11/0.35
Random vertical 47
90 87 6 36
(Type 4) 11/5/0.2
Delaunay 47
79 61 19 6
(Type 5) 0.65/30/2
Voronoi 44
(Type 6) 65 55 18 0
0.2/10/4, 0.2/10/5

3.1. Manual Approach of Adjusting Input Parameters


The result of the experiment in which participants find optimal values by adjusting input
parameters manually is illustrated in Figures 4–6. Each graph shows the optimal value from
125 combinations of variables and the average value of ten tries by 30 participants. In the case
of horizontal and vertical louver types, the more tries that are undertaken, the more optimal values
are obtained (Figure 4). As an example, in the horizontal louver, the value of the first try was 36.
This rose to 83 at the tenth try, which is an improvement of 2.3 times the original value. After ten
repeated experiments, 83 was reached, which is 83% of the maximum value 86 from 125 combinations.
In the vertical louver, the value of the first try was 76, which rose to 94 at the tenth try, which is
an improvement of 1.2 times. After ten repeated experiments, 94 was reached, which is 94% of the
Sustainability 2016, 8, 1220 9 of 16

maximum value 46 from 125 combinations. Participants learned the linear relationship between
optimized output values and input parameters after repeated efforts. This test confirms that two
types of louvers and windows are composed of linear input parameters, and the average value of 30
participants approaches the maximum value gradually. In other words, as the number of tries increases,
participants become able to predict the results, and the most optimized facade design becomes possible.
The minimum to maximum value of fitness spectrum of the horizontal louver type is larger than the
vertical louver’s spectrum. It is interpreted that the combination of three input parameters influences
the fitness values more critically in the horizontal louver type. The average values show that vertical
is nearer the maximum value than horizontal. The basic fitness values are high regardless of input
parameters. In addition, the variances of the horizontal louver are larger than those of the vertical
louver, which
Sustainability 2016,is8,the
1220unique characteristic of those louver types. As the louver is facing south in the
9 of 16
base model of the experiment, daylight control of the horizontal louver is excellent compared to the
vertical louver.
is excellent Therefore,
compared thevertical
to the weight louver.
of the horizontal
Therefore,louver is larger
the weight of than the verticallouver
the horizontal louverisfor the
larger
input parameters.
than the vertical louver for the input parameters.

Figure 4.
Figure 4. Horizontal
Horizontal louver
louver vs.
vs. Vertical louver.
Vertical louver.

However, the results of the random horizontal and random vertical with non-linear components
However, the results of the random horizontal and random vertical with non-linear components
remain steady regardless of the number of calculations (Figure 5). In the case of the random
remain steady regardless of the number of calculations (Figure 5). In the case of the random horizontal
horizontal and random vertical, as the relationship between fitness and input parameters is non-
and random vertical, as the relationship between fitness and input parameters is non-linear, unlike the
linear, unlike the horizontal and vertical, the graph runs parallel to the x-axis. The random horizontal
horizontal and vertical, the graph runs parallel to the x-axis. The random horizontal louver shows that
louver shows that the values of the first and tenth try are 70 and 69, respectively. The variance ranges
the values of the first and tenth try are 70 and 69, respectively. The variance ranges from 67 minimum
from 67 minimum to 70 maximum overall. It only reaches 70% of 75, the maximum of 125
to 70 maximum overall. It only reaches 70% of 75, the maximum of 125 combinations. The random
combinations. The random vertical louver shows that the difference between the first and tenth try is
vertical louver shows that the difference between the first and tenth try is negligible, 87 for the first,
negligible, 87 for the first, and 89 for the tenth, respectively. Overall, the minimum value is 87 and
and 89 for the tenth, respectively. Overall, the minimum value is 87 and the maximum is 90. However,
the maximum is 90. However, it reaches 90% of 47, the maximum of 125 combinations. The reason is
it reaches 90% of 47, the maximum of 125 combinations. The reason is that the relationship between
that the relationship between optimized output values and input parameters is unclear. Therefore,
optimized output values and input parameters is unclear. Therefore, this unpredictability makes it
this unpredictability makes it difficult to improve fitness value toward the optimization. Due to
difficult to improve fitness value toward the optimization. Due to irrelevancy, it is not useful when
irrelevancy, it is not useful when input parameters are adjusted manually.
input parameters are adjusted manually.
Lastly, in the cases of the Delaunay and Voronoi screen types, which include non-linear
Lastly, in the cases of the Delaunay and Voronoi screen types, which include non-linear
components, the optimized values are not achievable since the relationship between optimized
components, the optimized values are not achievable since the relationship between optimized output
output values and input parameters are neither concrete nor predictable (Figure 6). The Delaunay
values and input parameters are neither concrete nor predictable (Figure 6). The Delaunay screen type
screen type shows that the first try results in 61, and 79 occurs at the tenth try, which is an
shows that the first try results in 61, and 79 occurs at the tenth try, which is an improvement of 1.3 times.
improvement of 1.3 times. After the tenth try, it reaches 79% of 47, the maximum value of 125
After the tenth try, it reaches 79% of 47, the maximum value of 125 combinations. The Voronoi screen
combinations. The Voronoi screen type indicates that the first try results in 65, while the tenth try
type indicates that the first try results in 65, while the tenth try gives a lower value of 64. When the
gives a lower value of 64. When the unit sizes are compared between the Delaunay and Voronoi
screen types, the Delaunay pattern has lower average values than the Voronoi pattern (Figure 7). The
Delaunay pattern is more advantageous over the Voronoi pattern for the fact that subtle facade
adjustment is possible and light is diffused evenly over the facade. Therefore, fitness values of the
Delaunay screen type are larger than those of the Voronoi screen type.
Sustainability 2016, 8, 1220 10 of 16

unit sizes are compared between the Delaunay and Voronoi screen types, the Delaunay pattern has
lower average values than the Voronoi pattern (Figure 7). The Delaunay pattern is more advantageous
over the Voronoi pattern for the fact that subtle facade adjustment is possible and light is diffused
evenly over the facade. Therefore, fitness values of the Delaunay screen type are larger than those of
the Voronoi screen type.
Sustainability 2016, 8, 1220 10 of 16
Sustainability 2016, 8, 1220 10 of 16
Sustainability 2016, 8, 1220 10 of 16

Figure 5. Random horizontal louver vs. Random vertical louver.


Figure 5. Random horizontal louver vs. Random vertical louver.
Figure 5. Random horizontal louver vs. Random vertical louver.
Figure 5. Random horizontal louver vs. Random vertical louver.

Figure 6. Delaunay screen type vs. Voronoi screen type.


Figure 6. Delaunay screen type vs. Voronoi screen type.
Figure 6. Delaunay
Figure 6. Delaunayscreen typevs.
screen type vs.Voronoi
Voronoi screen
screen type.type.

Figure 7. Unit size variation by panel division of Delaunay window screen vs. Voronoi screen type.
Figure 7. Unit size variation by panel division of Delaunay window screen vs. Voronoi screen type.
Figure
seen7. size
Unit size variation by panel division
results ofofDelaunay windowsixscreen vs.types
Voronoi screen type.
FigureAs7. Unit in Table 6, theby
variation statistical
panel division of the previous
Delaunay window louver
screen were
vs. Voronoiplotted.
screenThetype.
horizontal and vertical louver types have relatively high R-squared value, 0.585
As seen in Table 6, the statistical results of the previous six louver types were plotted. and 0.301, which
The
As seen
indicates
horizontalthat
andin Table
those two
vertical6,louver
the statistical
types are results
on have
types the of theline.
regression
relatively previous
high The six louver
coefficient
R-squared types
values
value, were
0.585of plotted.
those
and The
two which
0.301, types
Ashorizontal
seen
seem inthat
to be
indicates
Table
and 6,two
vertical
significant
those bythe
the statistical
louver types
p-value
types have
are ontest,
results
theshowing
of
relatively
regressiontheythe
highhave
line.
previous
R-squared
Thea much sixvalues
value,
more
coefficient
louver types
0.585relationship.
linear ofand 0.301,
those
were
twowhich
types
plotted.
The horizontal
seem to beand
indicates that vertical
those
significanttwo louver
types
by the are types
on
p-value the have relatively
regression
test, showing they have high
line. Thea muchR-squared
more linear value,
coefficient values of those 0.585 and 0.301,
two
relationship. types
seem to be significant by the p-value test, showing they have a much more linear relationship.
Sustainability 2016, 8, 1220 11 of 16

which indicates that those two types are on the regression line. The coefficient values of those two
types seem to be significant by the p-value test, showing they have a much more linear relationship.

Table 6. Statistical results of the six louver/window types.


Sustainability 2016, 8, 1220 11 of 16

TypeTable 6. Statistical results of the Constant


Coefficients (S.E.)
six louver/window types. p-Value
Horizontal
Type 5.71 **
Coefficients 23.04
Constant (0.28)
(S.E.) 0.001
p-Value
R-squared: 0.585
Horizontal
Vertical 5.71 ** 23.04 (0.28) 0.001
R-squared: 0.585 1.96 ** 73.71 (0.17) 0.001
R-squared: 0.301
Vertical
Random horizontal 1.96 ** 73.71 (0.17) 0.001
R-squared: 0.301 −0.14 69.60 (0.32) 0.65
R-squared: 0.001
Random horizontal
Random vertical −0.14 69.60 (0.32) 0.65
R-squared: 0.001 0.19 87.57 (0.12) 0.12
R-squared: 0.008
Random vertical
Delaunay 0.008 0.19 87.57 (0.12) 0.12
R-squared: 1.53 ** 64.93 (0.39) 0.001
R-squared: 0.049
Delaunay
1.53 ** 64.93 (0.39) 0.001
Voronoi 0.049
R-squared:
−0.47 63.10 (0.36) 0.19
R-squared: 0.006
Voronoi
−0.47
**: p < 0.01.
63.10 (0.36) 0.19
R-squared: 0.006
**: p < 0.01.
3.2. Genetic Algorithm Optimization Methods Using Galapagos
3.2. Genetic Algorithm Optimization Methods Using Galapagos
The results of using Galapagos for the six types of facades are as follows: The row on the left
The results of using Galapagos for the six types of facades are as follows: The row on the left
shows the input parameters’ combination of 125 cases (genome combination). The row on the right
shows the input parameters’ combination of 125 cases (genome combination). The row on the right
indicates the input parameters’ combination of the upper 10% fitness values (Figure 8).
indicates the input parameters’ combination of the upper 10% fitness values (Figure 8).

(a) (b)

(c) (d)

(e) (f)

(g) (h)

Figure 8. Cont.
Sustainability 2016, 8, 1220 12 of 16
Sustainability 2016, 8, 1220 12 of 16

(i) (j)

(k) (l)
Figure 8. Galapagos results. (a) Horizontal louver (Type 1); (b) Horizontal louver (Type 1)—Top 10%;
Figure 8. Galapagos results. (a) Horizontal louver (Type 1); (b) Horizontal louver (Type 1)—Top 10%;
(c) Vertical louver (Type 2); (d) Vertical louver (Type 2)—Top 10%; (e) Random horizontal (Type 3);
(c) Vertical louverhorizontal
(f) Random (Type 2);(Type
(d) Vertical louver
3)—Top 10%; (Type 2)—Top
(g) Random 10%; (Type
vertical louver (e) Random horizontal
4); (h) Random (Type 3);
vertical
(f) Random
louverhorizontal
(Type 4)—Top(Type 3)—Top
10%; 10%;pattern
(i) Delaunay (g) Random vertical
(Type 5); louver
(j) Delaunay (Type(Type
pattern 4); (h) Random
5)—Top 10%;vertical
(k) Voronoi
louver (Type 4)—Toppattern screen;
10%; (i) (l) Voronoi pattern
Delaunay patternscreen—Top
(Type 5); 10%.
(j) Delaunay pattern (Type 5)—Top 10%;
(k) Voronoi pattern screen; (l) Voronoi pattern screen—Top 10%.
The area with black dots on the left is a genome map (combination map). Each point refers to a
gene (a parameter), and the distance between genes refers to a relationship. A genome (parameter
The area with black
combination) is madedots
up ofonnumerous
the left is a genome
genes. mapis(combination
If a genome decomposed into map). Each
N genes, thepoint refers to a
distance
gene (a count
parameter), andisthe
of two genes distance between
N-dimensional, which meansgenesthatrefers tocases
all the a relationship.
are a count ofA“Ngenome
raised to(parameter
the
power of
combination) isN” [38].up of numerous genes. If a genome is decomposed into N genes, the distance
made
It is impossible to render N-dimensional values with a two-dimensional plane. Therefore, this
count of two genes is N-dimensional, which means that all the cases are a count of “N raised to the
genome map only shows approximate relationships between genes. The definitive information on
power of N” [38].
this map is how similar genes are, and how different they are. The relationship of genes is shown as
It is
theimpossible to point
distance. If each render N-dimensional
is grouped valuesit means
in a single point, with that
a two-dimensional plane. Therefore,
the number of combinations of
this genome map only
parameters shows
that make upapproximate relationships
the louver is less. However, if between genes.
each point The definitive
is distributed information on
across multiple
this map points,
is how it means
similarthat thereare,
genes are fewer
and howparameters
different of the combination
they of parametersof
are. The relationship to genes
achieveisthe
shown as
optimal illumination of the parameters, i.e., the optimized illumination environment.
the distance. If each point is grouped in a single point, it means that the number of combinations of
parameters that make
According to up the8,louver
Figure is less.
the horizontal axisHowever, if eachi.e.,point
is each parameter, count,isangle,
distributed
depth. Theacross
verticalmultiple
axis points,
is the
it means that number
there areoffewer
cases (5parameters
× 5 × 5) for eachof
parameter. The upper 10%
the combination ofresults for each type
parameters to confirm
achieve thethe optimal
aggregation parts. It is apparent that the upper 10% fitness parameter combination gives only one
illumination of the parameters, i.e., the optimized illumination environment.
kind in the case of the horizontal louver (Type 1) (Figure 8a). On the contrary, the different vertical
louver types have diverse upper 10% fitness parameters. The random vertical louver (Figure 8h)
According to Figure 8, the horizontal axis is each parameter, i.e., count, angle, depth.
provides six kinds of parameter combinations. In summary, the Galapagos results show that the
The vertical axis islouver
random vertical the number of cases
has more design (5 × 5than
alternatives × 5)thefor each parameter.
horizontal The
louver when it is upper
set to find 10%
resultsthe
forupper
each10%
type confirm
fitness values. the aggregation parts. It is apparent that the upper 10% fitness
parameter combination gives only one kind in the case of the horizontal louver (Type 1)
4. Conclusions
(Figure 8a). On the contrary, the different vertical louver types have diverse upper 10%
fitnessAparameters.
feasibility study
The on parametric
random optimization
vertical of daylighting
louver (Figure was performed,
8h) provides six kindsand the main
of parameter
findings can be summarized as follows:
combinations. In summary, the Galapagos results show that the random vertical louver
has more designdesign
Parametric alternatives than
technology tothe horizontal
create louver
optimal indoor whenconditions
lighting it is set tobyfind the upper
adjusting 10%
shading
fitnessshapes can help to improve daylighting quality in early design stages.
values.
 Conventional methods which depend on designers’ experience and knowledge can sufficiently
be applied with computer simulation techniques in several design types, notably horizontal and
4. Conclusions
vertical louver types, which represent linear relationships in daylight simulation.
A feasibility study on parametric optimization of daylighting was performed, and the main
findings can be summarized as follows:

• Parametric design technology to create optimal indoor lighting conditions by adjusting shading
shapes can help to improve daylighting quality in early design stages.
• Conventional methods which depend on designers’ experience and knowledge can sufficiently
be applied with computer simulation techniques in several design types, notably horizontal and
vertical louver types, which represent linear relationships in daylight simulation.
Sustainability 2016, 8, 1220 13 of 16

• Computer-assisted daylight simulation can help and assist the conventional approach, which can
be maximized when dealing with large amounts of data and non-liner algorithms such as in the
case of random,
Sustainability Delaunay, and Voronoi.
2016, 8, 1220 13 of 16


When Computer-assisted
the experimentdaylight simulation
is executed can help
manually, it and assist thetoconventional
is possible approach, which
find the relationship between
parameters and fitness. However, when the cases of parameter combination are large such
can be maximized when dealing with large amounts of data and non-liner algorithms and as in
complex,
the case of random, Delaunay, and Voronoi.
it takes considerable time to obtain the results, or sometimes it is difficult to reach optimal values.
While theWhenrelationship betweenisparameters
the experiment and fitness
executed manually, it isispossible
unobtainable
to find using the geneticbetween
the relationship algorithms
parametersitand
of Galapagos, fitness. However,
is effective to find thewhen the cases of parameter
combination of parameters combination
to reach are alarge and fitness
certain complex,value.
it takes considerable
In particular, it is useful whentime tothe
obtain
numberthe results, or sometimes
of parameters it isand
is large difficult to reach optimal
the combination values.
or algorithm is
While the relationship between parameters and fitness is unobtainable
complex. Moreover, the optimization method using Galapagos provides various design alternatives using the genetic algorithms
of Galapagos, it is effective to find the combination of parameters to reach a certain fitness value. In
in a scientific manner, which is executable on a personal computer. However, the downside is that
particular, it is useful when the number of parameters is large and the combination or algorithm is
the optimization method using Galapagos still needs the same amount of time even if the number of
complex. Moreover, the optimization method using Galapagos provides various design alternatives
combinations is small.
in a scientific manner,In that case,
which it is necessary
is executable on ato set population
personal computer. and generation
However, values in Galapagos.
the downside is that
The calculation
the optimization method using Galapagos still needs the same amount of time even and
continues until a certain number of calculations, i.e., population if thegeneration,
number of are
reached. In other words,
combinations is small. even if thecase,
In that combinations of parameters
it is necessary are found
to set population andtogeneration
produce the upper
values in 10%
or above of fitness,
Galapagos. The the calculation
calculation continues.
continues until a Therefore,
certain number as seen in this experiment,
of calculations, when the
i.e., population and total
numbergeneration,
of casesare is reached.
only 125,Initother
mightwords, even if the to
be inefficient combinations
use Galapagos. of parameters are found
Nevertheless, thetooptimization
produce
the upper 10% or above of fitness, the calculation continues. Therefore, as
method is useful for the fact that it is accessible without special skills or experience in daylighting seen in this experiment,
when
design. the total number
A designer of cases is only
with insufficient 125, it might
knowledge be inefficient
of computer to use Galapagos.
programming can applyNevertheless, the in
facade design
optimization method is useful for the fact that it is accessible without special skills or experience in
a logical procedure. Instant views of results enable designers to develop their initial designs further,
daylighting design. A designer with insufficient knowledge of computer programming can apply
making Grasshopper a useful resource.
facade design in a logical procedure. Instant views of results enable designers to develop their initial
Currently,
designs further,buildings
makingare designedausing
Grasshopper useful computers
resource. and various software, as for daylighting.
NURBS orCurrently,
randomized patterns
buildings which might
are designed using have been hard
computers to implement
and various software,previously, can now be
as for daylighting.
applied.
NURBSTherefore, this research
or randomized patternshas revealed
which might how thosehard
have been shading design methods
to implement previously,are canevaluated
now be by
comparing
applied. computer-aided simulations
Therefore, this research with conventional
has revealed how those shading ones. design methods are evaluated by
comparing computer-aided simulations with conventional ones.
Acknowledgments: This research was supported by the Basic Science Research Program through the
Acknowledgments: This research was supported by the Basic Science Research Program through the National
National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning
Research Foundation ofIl Hyun
(NRF-2014R1A1A1002268). Korea Whang,
(NRF) funded by the
a student MinistryUniversity,
of Hongik of Science,assisted
ICT & in
Future Planning of
the collection (NRF-
materials.
2014R1A1A1002268). Il Hyun Whang, a student of Hongik University, assisted in the collection of materials.
Author Contributions: Kyung Sun Lee and Jae Wook Lee conceived and designed the experiments; Ki Jun Han
Author
performed theContributions:
experiments;Kyung
KyungSun
SunLee
Leeand JaeJae
and Wook LeeLee
Wook conceived andthe
analyzed designed theJun
data; Ki experiments; Ki Jun Han
Han contributed analysis
tools; performed
Kyung Sunthe Leeexperiments; Kyung
and Jae Wook LeeSun Leethe
wrote andpaper.
Jae Wook Lee analyzed the data; Ki Jun Han contributed
analysis tools; Kyung Sun Lee and Jae Wook Lee wrote the paper.
Conflicts of Interest: The authors declare no conflict of interest.
Conflicts of Interest: The authors declare no conflict of interest.
Appendix A
Appendix A

(a) (b)

(c) (d)

Figure A1. Cont.


Sustainability 2016, 8, 1220 14 of 16
Sustainability 2016,
Sustainability 8, 8,
2016, 1220
1220 1416
14 of of 16

(e)
(e)
Figure A1. The procedure for making the Delaunay pattern. (a) Selecting surface for analysis;
Figure A1. The procedure for making the Delaunay pattern. (a) Selecting surface for analysis;
Figure A1. Theseed
(b) Selecting procedure for making
points (Pattern the Delaunay
parameter, pattern. (a) (c)
Random parameter); Selecting
Creatingsurface for pattern
Delaunay analysis;
(b) Selecting seed points (Pattern parameter, Random parameter); (c) Creating Delaunay pattern
(b)division;
Selecting
(d)seed points
Removing (Patterncells
Delaunay parameter, Random parameter);
by WWR parameter; (c) material:
(e) Application Creatingglass
Delaunay pattern
(80% visible
division; (d)(d)
division; Removing
RemovingDelaunay
transmittance)/wall Delaunay
(50%
cells
cells by
reflectance).by WWR parameter;(e)
WWR parameter; (e)Application
Applicationmaterial:
material: glass
glass (80%
(80% visible
visible
transmittance)/wall (50% reflectance).
transmittance)/wall (50% reflectance).

(a) (b)
(a) (b)

(c) (d)

(c) (d)

(e)
Figure A2. The procedure for making the Voronoi pattern. (a) Selecting surface for analysis; (b)
Selecting seed points (Pattern parameter, Random
(e)parameter); (c) Creating Voronoi pattern division;
(d) Removing Voronoi cells by WWR parameter; (e) Application material: glass (80% visible
Figure A2. The procedure
transmittance)/wall for making the Voronoi pattern. (a) Selecting surface for analysis; (b)
(50% reflectance).
Figure A2. The procedure for making the Voronoi pattern. (a) Selecting surface for analysis;
Selecting seed points (Pattern parameter, Random parameter); (c) Creating Voronoi pattern division;
(b) Selecting seed points (Pattern parameter, Random parameter); (c) Creating Voronoi pattern
(d) Removing Voronoi cells by WWR parameter; (e) Application material: glass (80% visible
division; (d) Removing Voronoi cells by WWR parameter; (e) Application material: glass (80% visible
transmittance)/wall (50% reflectance).
transmittance)/wall (50% reflectance).
Sustainability 2016, 8, 1220 15 of 16

References
1. Lo, K. A critical review of China’s rapidly developing renewable energy and energy efficiency policies.
Renew. Sustain. Energy Rev. 2014, 29, 508–516. [CrossRef]
2. Deal, B.; Schunk, D. Spatial dynamic modeling and urban land use transformation: A simulation approach
to assessing the costs of urban sprawl. Ecol. Econ. 2004, 51, 79–95. [CrossRef]
3. Pérez-Lombard, L.; Ortiz, J.; Pout, C. A review on buildings energy consumption information. Energy Build.
2008, 40, 394–398. [CrossRef]
4. Boubekri, M.; Hull, R.B.; Boyer, L.L. Impact of window size and sunlight penetration on office workers’
mood and satisfaction a novel way of assessing sunlight. Environ. Behav. 1991, 23, 474–493. [CrossRef]
5. Boubekri, M. Daylighting, Architecture and Health; Routledge: Abingdon, UK, 2008.
6. Boyce, P.; Hunter, C.; Howlett, O. The Benefits of Daylight through Windows; Rensselaer Polytechnic Institute:
Troy, NY, USA, 2003.
7. Reinhart, C.F.; Mardaljevic, J.; Rogers, Z. Dynamic daylight performance metrics for sustainable building
design. Leukos 2006, 3, 7–31.
8. Munaaim, M.A.C.; Al-Obaidi, K.M.; Ismail, M.R.; Rahman, A.M.A. Empirical evaluation of the effect of heat
gain from fiber optic daylighting system on tropical building interiors. Sustainability 2014, 6, 9231–9243.
[CrossRef]
9. Lee, K.S.; Lee, J.; Lee, J.S. Low-energy design methods and its implementation in architectural practice:
Strategies for energy-efficient housing of various densities in temperate climates. J. Green Build. 2013, 8,
164–183. [CrossRef]
10. Augenbroe, G. Integrated building performance evaluation in the early design stages. Build. Environ. 1992,
27, 149–161. [CrossRef]
11. Cheong, C.H.; Kim, T.; Leigh, S.B. Thermal and daylighting performance of energy-efficient windows in
highly glazed residential buildings: Case study in Korea. Sustainability 2014, 6, 7311–7333. [CrossRef]
12. Piderit Moreno, M.B.; Labarca, C.Y. Methodology for Assessing Daylighting Design Strategies in Classroom
with a Climate-Based Method. Sustainability 2015, 7, 880–897. [CrossRef]
13. Jakubiec, J.A.; Reinhart, C.F. DIVA 2.0: Integrating daylight and thermal simulations using Rhinoceros 3D,
Daysim and EnergyPlus. In Proceedings of Building Simulation 2011: 12th Conference of International
Building Performance Simulation Association, Sydney, Australia, 14–16 November 2011; Volume 20,
pp. 2202–2209.
14. Bourgeois, D.; Reinhart, C.; Macdonald, I. Adding advanced behavioural models in whole building energy
simulation: A study on the total energy impact of manual and automated lighting control. Energy Build.
2006, 38, 814–823. [CrossRef]
15. Crawley, D.B.; Lawrie, L.K.; Winkelmann, F.C.; Buhl, W.F.; Huang, Y.J.; Pedersen, C.O.; Glazer, J. EnergyPlus:
Creating a new-generation building energy simulation program. Energy Build. 2001, 33, 319–331. [CrossRef]
16. Lee, J.W.; Jung, H.J.; Park, J.Y.; Lee, J.B.; Yoon, Y. Optimization of building window system in Asian regions
by analyzing solar heat gain and daylighting elements. Renew. Energy 2013, 50, 522–531. [CrossRef]
17. Caldas, L.G.; Norford, L.K. A design optimization tool based on a genetic algorithm. Autom. Constr. 2002, 11,
173–184. [CrossRef]
18. Iqbal, I.; Al-Homoud, M.S. Parametric analysis of alternative energy conservation measures in an office
building in hot and humid climate. Build. Environ. 2007, 42, 2166–2177. [CrossRef]
19. Attia, S.; Gratia, E.; De Herde, A.; Hensen, J.L. Simulation-based decision support tool for early stages of
zero-energy building design. Energy Build. 2012, 49, 2–15. [CrossRef]
20. Ward, G.; Shakespeare, R. Rendering with Radiance: The Art and Science of Lighting Visualization; Morgan
Kaufmann Publishers: Burlington, MA, USA, 1998.
21. Blackwell, O.M.; Blackwell, H.R. Individual responses to lighting parameters for a population of 235
observers of varying ages. J. Illum. Eng. Soc. 1980, 9, 205–232. [CrossRef]
22. Ascione, F.; Bianco, N.; de Masi, R.F.; Mauro, G.M.; Vanoli, G.P. Design of the building envelope: A novel
multi-objective approach for the optimization of energy performance and thermal comfort. Sustainability
2015, 7, 10809–10836. [CrossRef]
23. Sadeghzadeh, H.; Aliehyaei, M.; Rosen, M.A. Optimization of a Finned Shell and Tube Heat Exchanger
Using a Multi-Objective Optimization Genetic Algorithm. Sustainability 2015, 7, 11679–11695. [CrossRef]
Sustainability 2016, 8, 1220 16 of 16

24. Turkson, R.F.; Yan, F.; Ali, M.K.A.; Liu, B.; Hu, J. Modeling and multi-objective optimization of engine
performance and hydrocarbon emissions via the use of a computer aided engineering code and the NSGA-II
genetic algorithm. Sustainability 2016, 8, 72. [CrossRef]
25. IES Daylight Metrics Committee. IES Spatial Daylight Autonomy (sDA) and Annual Sunlight Exposure (ASE),
Daylight Metrics Committee. Approved Method IES LM-83-12; Illuminating Engineering Society of North
America: New York, NY, USA, 2012.
26. Website of Solemma. Available online: http://www.solemma.net/DIVA-for-Rhino/DIVA-for-Rhino.html
(accessed on 14 October 2016).
27. Ismail, E.D.; Ibrahim, N.; Hajar, N.H. Daylight Factor on Natural Lighting Analysis Simulation of an
Adaptive Reuse Building: Penaga Hotel, Penang. Adv. Sci. Lett. 2016, 22, 1120–1124. [CrossRef]
28. Galatioto, A.; Beccali, M. Aspects and issues of daylighting assessment: A review study. Renew. Sustain.
Energy Rev. 2016, 66, 852–860. [CrossRef]
29. Building Research Establishment. BREEAM—The BRE Environmental Assessment Method. Available online:
http://www.breeam.org (accessed on 12 October 2016).
30. Kudryashova, A. Certification Schemes for Sustainable Buildings: Assessment of BREEAM, LEED and
LBC from a Strategic Sustainable Development Perspective. Ph.D. Thesis, Blekinge Institute of Technology,
Blekinge, Sweden, 2015.
31. Lim, Y.W. Building Information Modeling for Indoor Environmental Performance Analysis. Am. J.
Environ. Sci. 2015, 11, 55–61. [CrossRef]
32. CASBEE for Building (New Construction). 2014 Edition. Available online: http://www.ibec.or.jp/CASBEE/
english/ (accessed on 12 October 2016).
33. Day Lighting Rules of Thumb. Available online: http://www.gsd.harvard.edu/research/gsdsquare/
Publications/DiffuseDaylightingDesignSequenceTutorial.pdf (accessed on 12 December 2015).
34. CIBSE Lighting Guide. 10: Daylighting and Window Design; The Chartered Institution of Building Services
Engineers: London, UK, 1999.
35. British Standards Institution. Available online: http://www.cibse.org/pdfs/GPG245.pdf (accessed on
5 January 2016).
36. LEED for New Construction & Major Renovations. Available online: http://www.usgbc.org/Docs/Archive/
General/Docs1095.pdf (accessed on 8 February 2016).
37. Yi, Y.K.; Kim, H. Agent-based geometry optimization with Genetic Algorithm (GA) for tall apartment’s solar
right. Sol. Energy 2015, 113, 236–250. [CrossRef]
38. Rutten, D. Galapagos: On the logic and limitations of generic solvers. Arch. Des. 2013, 83, 132–135. [CrossRef]

© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).