RESEARCH

MULTIOBJECTIVE OPTIMIZATION OF THE

INTERGENERATIONAL RESIDENTIAL SPACE WITH THE
GOAL OF DAYLIGHTING AND THERMAL COMFORT

Jiehui Wang1 and Juanli Guo1,2,*

ABSTRACT
(1) Background: An ageing population and two-child policy have led to the trans-
formation of China’s family structure, and multigenerational residences account for
an increasing proportion of mainstream family residences. Different generations
of residents have great differences in behaviour patterns and health requirements,
but existing residential buildings are not especially designed for health needs. (2)
Methods: First, based on relevant codes and a questionnaire survey, the spatial needs
and behaviour patterns of different generations of residents are obtained, and the
benchmark model is established based on the Grasshopper (GH) parametric plat-
form. Then, based on the GH platform and the building simulation plug-in, which
are Ladybug Tools and a multiobjective optimization tool named Wallacei, daylight
autonomy (DA), useful daylight illuminance (UDI), and the proportion of thermal
discomfort hours (PDH) are taken as daylighting and thermal comfort indicators,
and the bedroom location, orientation, bay size, depth to bay ratio, window form,
window-to-wall ratio, and horizontal shading width are optimized. Finally, the Pareto
front is analysed by the Technique for Order of Preference by Similarity to Ideal
Solution (TOPSIS) comprehensive evaluation method, and the optimal solution is
compared with existing multigenerational residences in Tianjin. (3) Results: First,
the middle room, oriented south or south by east, large bay, small window-to-wall
ratio, and bay window can effectively be improved for indoor thermal comfort, while
rooms that are oriented south or south by east, small depth bay ratio, large window-
to-wall ratio and balcony can effectively be improved for daylighting quality. Second,
compared with existing buildings in Tianjin, the DA, UDI, and PDH of the adult
bedroom with the ideal solution are increased by 33.7%, 3.5%, and 10.8%, respec-
tively; the DA, UDI, and PDH of the child bedroom with the ideal solution are
increased by 15.5%, 4.2%, and 4.9%, respectively; and the DA, UDI, and PDH of
the elderly individual bedroom with the ideal solution are increased by 42.7%, 4.9%,
and 1.7%, respectively. (4) Conclusions: The optimization scheme is substantially
improved for the health of the indoor daylight and thermal environment of existing
housing, and it provides a scientific and quantitative decision-making basis for the
healthy design of multigenerational residences.

1. School of Architecture, Tianjin University, Tianjin 300072, China; wangjiehui@[Link]

2. Key Laboratory of Architectural Physics and Environmental Technology, Tianjin 300072, China; guojuanli@[Link]
* Correspondence: guojuanli@[Link]; Tel.: +86-13820262201

Journal of Green Building 225

 KEYWORDS
multigenerational residence; daylight autonomy; useful daylight illuminance; propor-
tion of thermal discomfort hours; space optimization

1. INTRODUCTION
As the population aged more rapidly and the birth policy was implemented, the proportion of
multigenerational residences increased in China. The traditional housing form does not adapt
to the family structure under intergenerational relationships, and the problems of elderly people
living alone and social isolation are becoming increasingly serious. China’s traditional culture
has always attached importance to family affection and filial piety, and the phenomenon of
different generations of people living together is common [1]. At the same time, as housing
is the main place for people to rest and live, research on housing is not limited to the division
of spatial function. The space environment has an important impact on people’s behaviours
and mental health, and people, as the main experiencers of space, also have a negative effect
on the indoor environment. Changes in residential use patterns, such as population ageing
and working at home, have put forward new requirements for residence designs, spatial pat-
terns, colour textures, etc. These problems directly affect people’s physical and mental health
in housing. However, in the design of existing residential buildings in China, a designer only
pays attention to whether the window-to-floor ratio meets the requirements of the specification
for the design of residential indoor light environments and does not consider the impact on
human health caused by the uneven indoor light intensity and glare caused by its design. At the
same time, most previous studies are about the impact of a single indoor physical environment
on human health. There are few studies that comprehensively consider the impact of different
influencing factors on human health, and most of them focus on adults and elderly individuals.
There is less research on the health of different generations of people considering the indoor
physical environment. Therefore, the study of the impact of the indoor physical environment
on the health of different generations of people has become a problem that cannot be ignored
in current urban housing exploration.
Research on existing residential units in China shows that when architects design build-
ings, they basically set children’s bedrooms and adult bedrooms in the south and elderly people’s
bedrooms in the north, which is determined by current social factors. However, the location
of such rooms does not take into account the use time of different generations of people for
residential bedrooms. At the same time, long-term living in the northern bedroom will affect
the physical and mental health of elderly individuals.
The impact of the residential indoor physical environment on human health is mainly
reflected in the indoor thermal environment, daylighting environment, acoustic environment
and space environment. To determine the impact of different factors on human health, the
analytic hierarchy process (AHP) is used to determine the relative importance of different
factors. Taking human health as the target layer and the indoor light environment as the crite-
rion layer, the importance of every two indicators in the four criteria layers can be compared,
an online questionnaire is prepared, and the questionnaire is distributed to industry experts

226 Volume 18, Number 2

for evaluation. The Saaty 1–9 scale method is used for importance comparison, and 30 valid
questionnaires are retrieved. The analytic hierarchy process is used to calculate the evaluation
results of each expert, and finally, the average of the evaluation results of 30 experts is taken as
the final weight. The influence weights of four different factors are as follows: indoor thermal
environment 0.293, indoor light environment 0.265, indoor acoustic environment 0.209, and
indoor space environment 0.233. The expert survey results show that the thermal environment
and light environment are more important for human health. At the same time, changing the
residential room orientation, window-to-wall ratio, window form, etc., has a more direct impact
on the indoor thermal environment and light environment. Therefore, it is necessary to deter-
mine how to change the residential room orientation, window-to-wall ratio, etc., to make the
indoor light and thermal environment reach a comfortable range.
Although the relevant research based on the characteristics of people of different ages in
the field of healthy buildings is increasing yearly, there are few articles that consider multigen-
erational housing and focus on how to allocate rooms based on the characteristics of people of
different ages. Due to physiological and behavioural differences, the needs of elderly individu-
als, adults and children for light and heat vary greatly. Therefore, through the investigation of
current multigenerational housing in Tianjin and the combination with various parametric
design methods, such as parametric modelling, performance simulation, and multiobjective
optimization [2], this paper takes the bedroom, which has the closest relationship with people
in housing, as the research object to provide guidance for the design of multigenerational resi-
dences guided by the demand for daylighting and thermal comfort. The specific research con-
tents include (1) proposing the optimization method of bedrooms in multigenerational housing
for different generations; (2) analysing the influence of different design parameters on light
and heat through parametric modelling, building performance simulation and optimization
algorithms, and the best matching relationship between different rooms and different people
is proposed; and (3) optimizing the actual case with the optimal solution and comparing the
effects before and after optimization.

2. REVIEW
Existing research on healthy housing mainly focuses on the influencing factors of the indoor
physical environment and mostly focuses on adult and elderly individuals, less on children, and
less on considering different intergenerational groups at the same time. Mao summarized 30
factors that affect health from the four stages of design, construction, operation and demolition
and finally built a framework of factors that influence health for the whole life cycle of build-
ings [3]. Yao Jingwen paid attention to factors such as sound, light, thermal environment and
air quality that affect indoor life and explored technical measures for applying health concepts
to the indoor physical environment in residential areas [4]. Sujanova et al. analysed and sum-
marized 12 studies on health influencing factors and summarized the importance level of four
factors on health: thermal environment, light environment, acoustic environment and indoor
air quality. In terms of the impact of the indoor thermal environment on human health [5],
Monika et al. found that compared with other indoor environmental factors (sound comfort,
visual comfort, indoor air quality, etc.), building users pay more attention to indoor thermal
environment comfort [6]; Wouter et al. found that compared with constant indoor temperature,
indoor environment with temperature fluctuations will have a positive impact on human health
[7]; Givoni found through experimental research that a good indoor thermal environment

Journal of Green Building 227

can make people feel happy physically and mentally and improve work efficiency at the same
time [8][9]; Fisk found through research that the appropriate indoor thermal environment
can increase productivity by 18% [10]; Chow et al. found that women are more sensitive to
temperature than men and more adaptable to slightly warmer environments, but men and
women have similar neutral temperature ranges when using the climate chamber for research
[11]; Katafygiotou et al.’s actual tests on middle school classrooms and laboratories confirmed
this view and attributed the differences to the differences in metabolism and skin surface char-
acteristics between men and women [12]; Ahrentzen explored the impact of residential indoor
factors on the active and healthy life of elderly individuals, including indoor obstacles, space
organization, air quality, lighting quality, sound quality, indoor greening, etc.[13]; taking elderly
individuals in elderly care facilities in Shanghai as the research object, Jiao collected data through
indoor environmental parameter measurement and questionnaire survey, established a multiple
regression model and analysed the influencing factors of indoor thermal satisfaction [14]. Yukari
Iino et al. found that room orientation, layout and shading facilities will have a great impact
on the indoor thermal environment after studying 12 elderly residences in Japan [15]. Based
on architectural medical theory that emphasizes the natural healing function of the environ-
ment, Zheng Hua proposed an optimization scheme of the elderly residential environment for
11 chronic diseases [16]. Hyunjun Yun and Yue Peng studied the indoor thermal environment
of kindergartens and found that the thermal comfort temperature range of children was lower
than that of adults [17]. Konis K found that the use of internal shading will reduce the indoor
lighting coefficient through the investigation of the indoor natural lighting coefficient and the
comfort of indoor staff, while most staff believe that good indoor natural lighting can improve
their office efficiency [18]. Through research, P T Stone found that only when elderly people
receive more than 1000 lx of natural light for approximately one hour every day can they main-
tain a relatively stable periodic daily melatonin level that ensures that elderly people can obtain
high-quality sleep at night. Letting elderly people live in a bedroom with a good orientation to
meet the lighting needs of the bedroom is more conducive to maintaining their own normal
biological rhythm and their physical and mental health [19].
From the existing research, there are many studies on the single demand of indoor light
environment or thermal environment for elderly people and other special groups, while there are
few studies on intergenerational cohabitation housing considering the indoor light and thermal
environment and targeting the needs of different generations of residents. Due to the differences
in physiology and behaviour, the demand for light and heat among elderly individuals, adults
and children of intergenerational residents varies greatly. How to simultaneously solve the light
and heat demand of different intergenerational people in residential design as much as possible
to ensure that the indoor light and heat environment is more conducive to the health needs of
different intergenerational residents is an urgent problem to be solved.

3. METHODS
This research takes the bedroom space of different generations of people in intergenerational
housing in Tianjin as the research object. Based on the ladybug building performance simulation
plug-in of the Grasshopper platform, combined with the multiobjective optimization method
and Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) comprehen-
sive evaluation method, this research optimizes the daylighting and thermal comfort of the

228 Volume 18, Number 2

 bedroom and explores the impact of different design parameters on the indoor environment of
the bedroom, and the optimal solution of spatial parameters for bedrooms of different genera-
tions in Tianjin is obtained.
First, a questionnaire survey was conducted among the residents living in intergenera-
tional housing in Tianjin to obtain information about the residential population structure and
the behavioural characterisitcs of people of different ages. Second, combined with the relevant
standards and the needs analysis of different generations of people, the simplified bedroom
space benchmark model is obtained. Then, based on the Ladybug Tools plug-in and Wallacei
multiobjective optimization plug-in, taking the bedroom location, orientation, bay, depth-bay
ratio, window form, window-to-wall ratio, and horizontal shading width as design variables and
taking daylight autonomy (DA), useful daylight illuminance (UDI) and proportion of thermal
discomfort hours (PDH) as objectives, optimization is carried out to explore the influence of
different design variables on daylighting and thermal comfort. Finally, the Pareto solution set
is analysed based on the TOPSIS comprehensive evaluation method, and the optimal solution
for bedrooms of different generations is obtained. The results are compared with the existing
intergenerational housing to prove its effectiveness.

3.1 Multiobjective optimization

Generally, the optimization problem is to obtain the optimal solution of the objective function
through a certain optimization algorithm. When there are two or more optimization objectives,
it becomes a multiobjective optimization problem. Its core is to make at least one optimization
objective better without deteriorating other objectives. In general, multiobjective optimization
problems can be defined as follows [20]:

min F (x) = ( f1(x), f 2 (x),! f n (x))

To obtain the optimal results of bedroom space design parameters of different generations
under the coupling of light and heat, this paper takes the three indoor physical environment
indicators DA, UDI, and PDH as the optimization objectives. Since three optimization objec-
tives are selected for this optimization and there is no obvious positive correlation between the
three optimization objectives, it is likely to weaken the other two objectives when optimizing
one of them. Therefore, it is suitable to use a multiobjective optimization algorithm in this case
to ensure the rationality of the optimization results. It can be expressed as follows:

FIGURE 1. Framework for optimizing bedroom space in multigenerational residences

Figure 1. Framework for optimizing bedroom space in multigenerational residences

3.1 Multiobjective optimization

Journal of Green Building 229

Generally, the optimization problem is to obtain the optimal solution of the objective function through a certain
optimization algorithm. When there are two or more optimization objectives, it becomes a multiobjective optimization
problem. Its core is to make at least one optimization objective better without deteriorating other objectives. In general,
 ⎧− f DA (x)
⎪
min ⎨− f UDI (x)
⎪
⎩ f PDH (x)

When the optimization reaches the Pareto optimal state, the Pareto optimal solution set
is generated. There is no unique global optimal solution for the multiobjective optimization
problem, but a set of compromise optimal solutions is generated through the trade-off between
various objectives, which is called the Pareto solution set [21]. The surface distribution of the
optimal solution set in space is called the Pareto frontier.
To achieve the multiobjective optimization problem, this paper uses the Wallacei multiob-
jective tool [22], which is based on the Grasshopper platform, takes the NSGA-II algorithm as
the main evolutionary algorithm, and integrates the functions of data recording, comparison,
analysis and visualization to help users better understand the optimization process and make
wise decisions according to the process data. NSGA-II performs well when dealing with more
than 3 objectives [23], and it has the advantages of a fast-running speed and good convergence
of the solution set. In addition, it has been proven to be one of the most effective algorithms
for solving multiobjective optimization and multicriteria decision-making to improve build-
ing performance.

3.2 TOPSIS comprehensive evaluation method

Since the Pareto solution obtained from multiobjective optimization is the result of the compro-
mise of three objectives and each of the three objectives in the solution has its own evaluation
criteria, it is impossible to directly obtain which solution in the solution set is the relatively
optimal solution, so it is necessary to use the comprehensive evaluation method to sort the
Pareto solution set. The full name of TOPSIS is a technology for order preference by similar-
ity to an ideal solution. The TOPSIS method was first proposed by C.L. Hwang and K. Yoon
in 1981. It is a method of ranking according to the proximity between a limited number of
evaluation objects and idealized objectives. It is a method of evaluating the relative advantages
and disadvantages of existing objects. As a sort method approaching the ideal solution, this
method only requires that each utility function be monotonically increasing (or decreasing).
It is a commonly used and effective method in multiobjective decision analysis, also known as
the good and bad solution distance method [24]. In this paper, with the help of SPSS statisti-
cal analysis software, the Pareto frontier solution set is sorted by the TOPSIS method, and the
ideal solution for bedrooms for different generations of people is obtained.

3.3 Ladybug Tools

Building performance simulation is achieved by Ladybug Tools, a plug-in based on the
Grasshopper platform. Among all available simulation software, Ladybug Tools is the most com-
prehensive tool. It integrates mainstream performance simulation engines worldwide, such as
Radiance, EnergyPlus, Therm, and Openfoam, and breaks the data transmission barrier between
parametric modelling and performance simulation. Therefore, model making, simulation, and
visualization can be carried out in one interface. Ladybug Tools include four modules: ladybug,
honeybee, butterfly, and dragonfly. Each plug-in has its own function. Honeybee is mainly used
for energy consumption and lighting analysis, which can calculate the energy consumption,
comfort, and lighting of buildings.

230 Volume 18, Number 2

 luating the relative advantages and disadvantages of existing objects. As a sort method approaching
his method only requires that each utility function be monotonically increasing (or decreasing). It is
nd effective method in multiobjective decision analysis, also known as the good and bad solution
]. In this paper, with the help of SPSS statistical analysis software, the Pareto frontier solution set is
IS method, and the ideal
FIGURE 2. solution for
Pareto front, bedrooms
optimal solution,for
anddifferent generations
ideal solution [25] of people is obtained.

Figure 2. Pareto front, optimal solution, and ideal solution[25]

4. MODEL
4.1 The establishment of a bedroom model for multigenerational residences
Based on the relevant specifications of residential design [26][27], this paper establishes three
rmance simulation is achieved by Ladybug Tools, a plug-in based on the Grasshopper platform.
benchmark models of the south facing bedroom in the multigenerational house, corresponding
e simulation software,
to three ageLadybug Toolselderly
groups, namely, is theindividuals,
most comprehensive
adults and children. tool. It integrates
Among mainstream
them, the adult
ation engines worldwide,
bedroom and such as Radiance,
elderly EnergyPlus,
individual bedroom adopt theTherm, andofOpenfoam,
size range and breaks
the master bedroom in the the data
specification,
r between parametric modellingand theand
childperformance
bedroom adoptssimulation.
the size rangeTherefore,
of the common secondary
model bedroom.
making, simulation,
Compared with adults and children, the physical conditions of the different individuals within
n be carried out in
theone interface.
elderly Ladybug
group often Tools
vary greatly, andinclude
bedroomfour designmodules:
for elderlyladybug, honeybee,
people needs more con-butterfly,
h plug-in has its own function.
sideration. Based on theHoneybee
considerationis of
mainly
the whole used for design
life cycle energy consumption
theory of housing, theand
age lighting
calculate the energy
changeconsumption,
of residents should comfort,
be takenand
intolighting
account inofthe buildings.
design of intergenerational housing,
and the spatial characteristics of the ageing design should be considered in the bedrooms for
elderly individuals. Therefore, when establishing the model, the elderly individual bedroom
design should mainly consider elderly people with two lifestyles, including independent living
andfor
of a bedroom model assisted living. The bedroom
multigenerational for elderly people should consider the spatial characteristics
residences
needed by the ageing process. Therefore, the bay and depth of the room for independent-living
elderly people who use single beds are taken as the minimum value of the bedroom space size,
levant specifications of residential design[26][27], this paper establishes three benchmark models of
and the bay and depth of the space for assisted-living elderly people who use double beds are
edroom in the taken
multigenerational
as the maximum value. house, corresponding
As shown in Table 1. to three age groups, namely, elderly
and children. AmongTothem, improvethethe
adult bedroom
pertinence of theand elderly
research individual
results, bedroomaadopt
this paper conducted the size range
questionnaire
survey on the
oom in the specification, andcurrent
the intergenerational
child bedroom cohabitation
adopts the project
sizeresidents
rangeinofChina. The question-
the common secondary
naire was distributed through the online questionnaire platform Questionnaire Star. A total of
d with adults and children, the physical conditions of the different individuals within the elderly
748 online questionnaires were recovered, and 11 invalid questionnaires were removed. A total
eatly, and bedroom
of 737design for elderlywere
valid questionnaires people needs
obtained, more 367
including consideration.
male subjects andBased on thesubjects,
370 female consideration
cle design theorywith
of housing, the age
a sex ratio close change
to 1:1; of residents
710 adults aged 19~60,should
includingbe311taken intoaged
children account
0~18 andin 27
the design
elderly people aged 61 and above.
housing, and the spatial characteristics of the ageing design should be considered in the bedrooms
After screening and analysing the indoor conditions of people of different ages and differ-
als. Therefore, when establishing the model, the elderly individual bedroom design should mainly
ent time periods in the questionnaire, it is found that adults usually go out from 8:00 to 18:00;
ople with two lifestyles, including
children usually independent
go out from living
8:00 to 17:00; and assisted
and elderly living.
people usually The
spend bedroom
most for elderly
of their time
ider the spatial in
characteristics
the house, as shown needed by the
in Figures 3–[Link] process. Therefore, the bay and depth of the
ent-living elderly people who use single beds are taken as the minimum value of the bedroom space
nd depth of the space for assisted-living elderly people who use double beds are taken as the
Journal of Green Building 231
s shown in Table 1.
Table 1. Size range of bedrooms for adults, elderly people and children.
TABLE 1. Size range of bedrooms for adults, elderly individuals and children.

Elderly individual
Adult bedroom Children’s bedroom bedroom
Minimum size

Minimum size

Maximum size

Maximum size

To improve the pertinence of the research results, this paper conducted a questionnaire survey on the current
intergenerational cohabitation project residents in China. The questionnaire was distributed through the online
questionnaire platform Questionnaire Star. A total of 748 online questionnaires were recovered, and 11 invalid
questionnaires were removed. A total of 737 valid questionnaires were obtained, including 367 male subjects and 370
female subjects, with a sex ratio close to 1:1; 710 adults aged 19~60, including 311 children aged 0~18 and 27 elderly
people aged 61 and above.
After screening and analysing the indoor conditions of people of different ages and different time periods in the
questionnaire, it is found that adults usually go out from 8:00 to 18:00; children usually go out from 8:00 to 17:00; and
FIGURE
elderly 3. Behaviour
people usually spendofmost
adults in different
of their time
time in the periods
house, as shown in Figs. 3-5.

Figure 3. Behaviour of the adults in different time periods

232 Volume 18, Number 2

FIGURE 4. Behaviour of children in different time periods

Figure 4. Behaviour of the children in different time periods

FIGURE 5. Behaviour ofFigure 4. Behaviour
elderly of the
individuals in children
differentin time
different time periods
periods

Figure 5. Behaviour of elderly individuals in different time periods

Figure 5. Behaviour of elderly individuals in different time periods
Table 2. Time for adults, elderly individuals and children in the room.
TABLE 2. Time for adults, elderly
Table 2. Time forindividuals
adults, elderlyand children
individuals and in the room.
children
Occupancy period
in the room.

Adult 18:00-8:00
Occupancy period
ElderlyAdult
individuals Occupancy period
24:00-24:00
18:00-8:00
Children
Elderly individuals 17:00-8:00
24:00-24:00
Adult 18:00–8:00
In this study, the benchmark model Children is established based on the south-facing bedrooms of multigenerational
17:00-8:00
housing. The
In this
Elderly daylighting
study,
individuals optimization
the benchmark sky model
model adopts the
is established all-weather
based on the sky
24:00–24:00 condition bedrooms
south-facing proposed by of Richard Perez[28].
multigenerational
The PerezThe
housing. skydaylighting
model establishes hourlysky
optimization skymodel
models for typical
adopts years in different
the all-weather regions
sky condition based on
proposed bythe climate
Richard data in
Perez[28].
TheChildren
previous
Perezyears to ensure
sky model that the simulation
establishes hourly sky is as close
models the 17:00–8:00
fortotypicalreal situation
years as possible[29][30].
in different regions basedLadybug Tools can
on the climate datause
in
the local years
previous data files in thethat
to ensure Energy Plus database
the simulation to build
is as close to thethis
realsky model as
situation during the year-round
possible[29][30]. dynamic
Ladybug Toolslighting
can use
simulation.
the local dataIn this
filespaper,
in thetheEnergy
EPW file in Tianjin
Plus database is used to ensure
to build the model
this sky accuracy of thethe
during simulation. The dynamic
year-round limit value in the
lighting
Design Standard
simulation. In thisfor Energy
paper, theEfficiency
EPW file inof Tianjin
Residential Buildings
is used to ensurein Severe Cold and
the accuracy Cold
of the Regions (JGJ
simulation. 26-2018)
The limit shall
value be
in the
DesignIn
adopted forthis
Standard study,
the setting ofthe
for Energy benchmark
theEfficiency model
external enclosure isBuildings
structure
of Residential established
of the in based
building,
Severe and on
Cold andthe
the south-facing
setting
Cold of bedroom
Regions bedrooms
(JGJsurface shall of
materials
26-2018) be
refers
adoptedto the
for Standard
the settingfor
ofLighting Design
the external of Buildings
enclosure structure(GBof50033-2013).
the building,Asandshown in Table
the setting of3.
bedroom surface materials
multigenerational housing. The daylighting optimization sky model adopts the all-weather sky
refers to the Standard for Lighting Design of Buildings
Table (GB 50033-2013).
3. Environment As shown in Table 3.
parameter setting.
condition proposed by Richard Perez [28]. The Perez sky model establishes hourly sky models
Table 3. Environment parameter setting.
for typical years in different regions based on the climate data in previous years to ensure that the
7
simulation is as close to the real situation as possible [29][30]. Ladybug Tools can use the local
data files in the Energy Plus database to build 7this sky model during the year-round dynamic

Journal of Green Building 233

TABLE 3. Environment parameter setting.

Thermal Transmittance SHGC Reflectance

Exterior wall 0.45 w/m2 · k
Exterior window 2.0 w/m2 · k 0.50
Indoor floor 0.3
Interior wall 0.6
Indoor ceiling 0.75

lighting simulation. In this paper, the EPW file in Tianjin is used to ensure the accuracy of
the simulation. The limit value in the Design Standard for Energy Efficiency of Residential
Buildings in Severe Cold and Cold Regions (JGJ 26-2018) shall be adopted for the setting of
the external enclosure structure of the building, and the setting of bedroom surface materials
refers to the Standard for Lighting Design of Buildings (GB 50033-2013). As shown in Table 3.

4.2 Design variables

From the perspective of indoor light and thermal environment, there are usually three types
of design factors that will have a direct impact on it, including indoor space design, building
facade design and thermal performance of the building envelope. As this paper mainly discusses
the impact of architectural design on the physical environment of multigenerational bedroom
space, this paper only considers the first two types of design factors. The parametric model uses
7 variables to change the room shape, room location, orientation, bay, depth-bay ratio, window-
to-wall ratio, horizontal shading depth, and window type to ensure that one model can cover the
possibility of all geometric changes in the shape. Among them, the window types are discrete
variables, including ordinary windows, bay windows and balconies, and the other variables are
continuous variables. The value range of the design variables is determined according to research
data, design experience and relevant specifications. As shown in Table 4.

4.3 Objectives
Based on the WELL V2 healthy building standard, Evaluation Standard for Healthy Buildings
(t/asc 02-2021) and Evaluation Standard for Healthy Housing (t/cecs 462-2017), three indi-
cators of the daylight and thermal environment are selected as the optimization objectives to
obtain the most comprehensive optimization results.

1) Daylight autonomy
Daylight not only plays the role of illumination but can also transmit information to human
nerves and regulate human emotions [31]. Khademagha P found that in addition to illuminance
and lighting spectrum, multiple factors such as light source direction, exposure time, and expo-
sure duration can affect the nonvisual effect of light and affect physiological and psychological
indicators such as people’s sleep, work efficiency, alertness, blood pressure and heart rate [32].
Therefore, a suitable light environment plays a positive role in ensuring the normal operation
of body functions. Daylight autonomy (DA) is selected as the first daylighting index. The
definition of DA is that for each calculation point on the working face, the proportion of the
time in the whole year when the design illumination standard value can be reached by relying

234 Volume 18, Number 2

TABLE 4. Variables and range of room design.

variables subvariables type range step length

Room location Discrete east, middle, west ——
Room orientation Continuous 30 degrees south 5°
by East
30 degrees south
by West
Room Bay Continuous 3000–3900 mm 300 mm
(Adult)
2400–3300 mm
(Children)
2700–4200 mm
(Elderly)
Depth-bay ratio Continuous 1.2:1–1.5:1 0.1
Window form Common window —— ——
Bay window depth Continuous 400–600 mm 100 mm
Balcony depth Continuous 900–1500 mm 100 mm
Window-to-wall Continuous 30%–50% 5%
ratio
Horizontal Continuous 100–500 mm 100 mm
shading width

solely on natural lighting, expressed as a percentage. DA is calculated in hours, which can be

expressed as follows:

DA = (TE 8760) ×100%

where TE is the number of hours when the illuminance at the calculation point is greater than
the standard illuminance (300 lux).

2) Useful daylight illuminance

The useful daylight illuminance (UDI) is selected as the supplementary daylighting index.
UDI was put forward by Mardaljevic and Nabil in 2005 and refers to the percentage of time
when the illuminance value of each calculation point on the working face is within the range
of useful daylight illuminance in the whole year. The effective daylighting illumination range is
100 lx~2000 lx. If the illumination value is less than 100 lx, the lighting is insufficient, and the
room is too dark. If the illumination exceeds 2000 lx, daylighting is excessive, which tends to
cause glare. The UDI can accurately describe the changes in natural light in space and time and
take into account the visual discomfort caused by high illumination values. Compared with the
natural lighting coefficient, it is more adaptable and effective [33]. Combining DA and UDI

Journal of Green Building 235

can effectively improve the performance of natural lighting in the room and reduce the risk of
glare [34]. UDI is also calculated in hours, which can be expressed as follows:

UDI = (TEU 8760) ×100%

where TEU is the number of hours when the illuminance at the calculation point is within the
given useful illuminance range.

3) Proportion of thermal discomfort hours (PDH)

The indoor thermal environment has a significant impact on human health. Temperature,
humidity, ventilation conditions and other factors are untimely, and workers will have itching,
tears, headache, throat inflammation and other symptoms [35]. When the indoor thermal
environment temperature is high, the symptoms of sick building syndrome, respiratory dis-
eases, negative emotions, heart rate and fatigue will increase correspondingly [36]. The thermal
environment adopts the proportion of annual thermal uncomfortable hours as the optimiza-
tion objective [37]. With reference to the Design Standard for Energy Efficiency of Residential
Buildings in Severe Cold and Cold Regions (JGJ 26-2018), the indoor thermal comfort thresh-
old is that the room temperature is lower than 18°C in winter and higher than 26°C in summer.
The formula for annual thermal discomfort hours is PDH = (H18 + H26) ÷ 8760 where H18 is
the number of hours when the annual indoor operating temperature is lower than 18°C, and
H26 is the number of hours when the annual indoor operating temperature is higher than 26°C.

5. RESULTS

5.1 Multiobjective optimization based on Wallacei

The Wallacei multiobjective optimization plug-in of the Grasshopper platform is used as the
optimization platform. Based on the variables and objectives mentioned above, the simulation
optimization is started. The generation will be set to 40, and the number of each population
is set to 25. The average single optimization simulation duration is 40 seconds, and the three
experiments are 36 hours in total.
Table 6 shows the parallel coordinate plot (PCP) of the multiobjective optimization results.
Each column represents the simulation result of one objective; from left to right, it represents
the PDH, DA, and UDI, respectively, and the angle of the polyline represents the value of the
objective. This PCP includes all the solutions in the multiobjective optimization and gives an
overview of the convergence of the whole calculation process. In the figure, the red line repre-
sents the solution generated in the early stage of the operation, and the blue line represents the
solution generated in the late stage of the operation. The intensity of the polylines reflects the

TABLE 5. Parameter setting of multiobjective optimization.

Optimization objectives DA UDI PDH

Optimization direction max Max min
Optimization purpose Increase daylighting Improve useful Improved thermal
daylighting illumination comfort

236 Volume 18, Number 2

 optimization process, that is, the Pareto optimal solution, while the red space surface represents the Pareto front surface
of the optimization process. The three axes of the coordinate system represent the three objectives of this optimization.
Adult
The closer the cube is to the origin of the three-dimensional coordinates, the better the indoor physical environment of
optimization process, that is, the Pareto optimal solution, while the red space surface represents the Pareto front surface
the solution corresponds to this point.
of the optimization process. The three axes of the coordinate system represent the three objectives of this optimization.
Table
The closer the cube is to the 6. Multiobjective
origin optimization results
of the three-dimensional of the different
coordinates, the generations.
better the indoor physical environment of
the solution corresponds to thiscoordinate
Parallel point. plot Pareto solution set
Table 6. Multiobjective optimization results of the different generations.
TABLE 6. Multiobjective optimization results of the different generations.
Parallel coordinate plot Pareto solution set

Adult Children Elderly individual

ParallelAdult
coordinate
ization process, that is, the Pareto optimal solution, while the red space surface represents the Pareto front surface
Elderly
optimization process. The three axes ofplot Adult
the coordinate system represent the three objectives of thisChildren
optimization.
individual
oser the cube is to the origin of the three-dimensional coordinates, the better the indoor physical environment of
lution corresponds to this point.
Table 6. Multiobjective optimization results of the different generations.

Parallel coordinate plot Pareto solution set

Pareto
solution set
5.2 Correlation analysis of daylight and thermal performance

Taking the multiobjective optimization results of the elderly individual bedroom for the whole
the three optimization objectives are compared based on the parallel coordinate plot, and the corre
Elderly performance of the three objectives is analysed. As shown in Table 7, with the increase in DA
Children individual decreasing trend, and the performance of the two shows a negative correlation. Similarly, with the
dult
UDI values, the PDH also showed an opposite trend. The results show that excessive daylighting
Children thermal comfort and increase energy consumption. In summary, in the design of multigenerational b
we should not simply pursue an increase in the sunlight inlet but should consider the combinati
window-to-wall ratio, increasing the heat buffer zone (such as bay windows and balconies) and pro
flexible shading methods to provide a healthy and comfortable environment for the bedroom.
Table 7. Correlation analysis of DA, UDI and PDH.

5.2 Correlation analysis of daylight and thermal performance Parallel coordinate plot

Taking the multiobjective optimization results of the elderly

10 individual bedroom for the whole year as an example,
the three optimization objectives are compared based on the parallel coordinate plot, and the correlation between the
performance of the three objectives is analysed. As shown in Table 7, with the increase in DA, the UDI shows a
decreasing trend, and the performance of the two shows a negative correlation. Similarly, with the increase in DA and
convergence effect. It can be seen from the figure that the convergence effect of this multiobjec-
UDI values, the PDH also showed an opposite trend. The results show that excessive daylighting can reduce indoor
DA and UDI

ldren
tive optimization is good. thermal comfort and increase energy consumption. In summary, in the design of multigenerational bedrooms in Tianjin,
we should not simply pursue an increase in the sunlight inlet but should consider the combination of reducing the
Then, in the Pareto solution set graph, each small green cube represents the feasible solu-
window-to-wall ratio, increasing the heat buffer zone (such as bay windows and balconies) and providing a variety of
flexible shading methods to provide a healthy and comfortable environment for the bedroom.
tion calculated in each simulation. Each green cube Table connected
7. Correlation analysisby
of DA,the red
UDI and [Link] represents the

nondominated solution calculated by the optimization process,Parallel thatcoordinate

is, theplotPareto optimal solu-
tion, while the red space surface represents
10
the Pareto front surface of the optimization process.
The three axes of the coordinate system represent the three objectives of this optimization.
The closer the cube is to the originDAofandthe
10

UDI
three-dimensional coordinates, DA and PDH
the better the indoor
physical environment of the solution corresponds to this point.

5.2 Correlation analysis of daylight and thermal performance

Taking the multiobjective optimization results of the elderly individual bedroom for the whole
11

year as an example, the three optimization objectives are compared based on the parallel coor-
dinate plot, and the correlation between the performance of the three objectives is analysed.
As shown10 in Table 7, with the increase in DA, the UDI shows a decreasing trend, and the
DA and PDH

performance of the two shows a negative correlation. Similarly, with the increase in DA and
UDI values, the PDH also showed an opposite trend. The results show that excessive daylight-
ing can reduce indoor thermal comfort and increase energy consumption. In summary, in the
design of multigenerational bedrooms in Tianjin, we should not 11
simply pursue an increase in
the sunlight inlet but should consider the combination of reducing the window-to-wall ratio,
increasing the heat buffer zone (such as bay windows and balconies) and providing a variety of
flexible shading methods to provide a healthy and comfortable environment for the bedroom.

5.3 Optimal solution selection

After finishing the optimization, a total of 33 frontier solutions were obtained for the adult’s
bedroom, 28 frontier solutions for the children’s bedroom and 39 frontier solutions for the

Journal of Green Building 237

 performance of the three objectives is analysed. As shown in Table 7, with the increase in DA, the UDI shows a
decreasing trend, and the performance of the two shows a negative correlation. Similarly, with the increase in DA and
UDI values, the PDH also showed an opposite trend. The results show that excessive daylighting can reduce indoor
thermal comfort and increase energy consumption. In summary, in the design of multigenerational bedrooms in Tianjin,
we should not simply pursue an increase in the sunlight inlet but should consider the combination of reducing the
window-to-wall ratio, increasing the heat buffer zone (such as bay windows and balconies) and providing a variety of
flexible shading methods to provide a healthy and comfortable environment for the bedroom.
5.2 Correlation analysis of daylight and thermal performance Table 7. Correlation analysis of DA, UDI and PDH.

Taking the multiobjective optimization results of the elderly individual bedroom for the whole year as an example, Parallel coordinate plot
the three optimization objectives are compared based on the parallel coordinate plot, and the correlation between the
performance of the three objectives is analysed. As shown in Table 7, with the increase in DA, the UDI shows a
decreasing trend, and the performance of the two shows a negative correlation. Similarly, with the increase in DA and
UDI values, the PDH also showed an opposite trend. The results show that excessive daylighting can reduce indoor
TABLE 7. Correlation analysis of DA, UDI and PDH.
thermal comfort and increase energy consumption. In summary, in the design of multigenerational bedrooms in Tianjin,
we should not simply pursue an increase in the sunlight inlet but should consider the combination of reducing the
DA and UDI
window-to-wall ratio, increasing the heat buffer zone (such as bay windows and balconies) and providing a variety of
flexible shading methods to provide a healthy and comfortable environment for the bedroom.

DA and UDI
Table 7. Correlation analysis of DA, UDI and PDH.
DA and PDH UDI and PDH
Parallel coordinate plot

Parallel
coordinate
plot
DA and UDI DA and PDH UDI and PDH

11 5.3 Optimal solution selection

After finishing the optimization, a total of 33 frontier solutions were obtained for the adult’s bedroom, 28 frontier
solutions for the children’s bedroom and 39 frontier solutions for the elderly individual bedroom. However, any
solution that falls on the Pareto frontier does not mean that it is better than the others, so mathematical evaluation
DA and PDH
methods are needed to select the appropriate solution among these optimal solutions[38][39]. Based on SPSS statistical

elderly individual’s bedroom. However, any solution that falls on the Pareto frontier does not analysis software, this paper normalizes the data of the three objectives, including DA, UDI and PDH, and sets the
weights of the three objectives to be equal to obtain the optimal solution for the bedroom in intergenerational housing
in cold regions under the same weight. The results are shown in Table 8. For easy observation, only the top three optimal
mean that it is better than the others, so mathematical evaluation methods are needed to select solutions from TOPSIS as well as the optimal daylighting solution and the optimum thermal comfort solution are shown
in the table. The ideal solution represents the optimization result obtained when the room comprehensively considers

the appropriate solution among these optimal solutions [38][39]. Based on SPSS statistical
11
the three indicators of DA, UDI and PDH, while the daylighting optimal solution represents the optimization result
obtained when the room comprehensively considers DA and UDI, which can also be considered as the daylighting

analysis software, this paper normalizes the data of the three objectives, including DA, UDI and potential of the room, and the thermal comfort optimal solution represents the optimization result obtained when the
room only considers PDH, which can also be considered as the thermal comfort potential of the room.

PDH, and sets the weights of the three objectives to be equal to obtain the optimal solution for
Table 8. Bedroom optimization parameters of different generations.

Win Hori

the bedroom in intergenerational housing in cold regions under the same weight. The results are TOPSIS
Rankin
Locat
ion
Orienta
tion
Bay
Depth
bay
Windo
w type
dow-
to-
zonta
l
DA
（%））
UDI
（%
PDH
（%）
）
gs ratio wall shadi ）
shown in Table 8. For easy observation, only the top three optimal solutions from TOPSIS as Bay
ratio ng

well as the optimal daylighting solution and the optimum thermal comfort solution are shown
East South
1 3m 1.2 window 0.35 0.5 m 89.8 85.9 69.3
room
-0.5 m

in the table. The ideal solution represents the optimization result obtained when the room
Bay
Ideal Middle
2 South 3m 1.2 window 0.35 0.4 m 90.0 85.6 70.0
Solutions room
-0.5 m

comprehensively considers the three indicators of DA, UDI and PDH, while the daylighting A
d 3
Middle
room
South 3m 1.2
Bay
window 0.45 0.3 m 93.8 73.2 71.9
ul
optimal solution represents the optimization result obtained when the room comprehensively
-0.5 m
t Optimal
East Balcony
Daylightin 20 South 3m 1.2 0.45 0.5 m 92.5 81.9 75.6

considers DA and UDI, which can also be considered as the daylighting potential of the room,
room 1.2 m
g Solution
Optimal
South by Bay

and the thermal comfort optimal solution represents the optimization result obtained when thermal
Comfort
Solution
7
Middle
room
East
5°
3.3 m 1.2 window
-0.5 m
0.3 0.5 m 88.7 82.2 68.8

the room only considers PDH, which can also be considered as the thermal comfort potential Ideal
Solutions
1
Middle
room
South by
East
2.7 m 1.2
Bay
window
0.4 0.5 m 92.3 79.7 71.4

of the room. 12

5.4 Discussion of the solution

According to the analysis in Table 8, the following conclusions can be drawn: (1) Room Location:
As the room location has no impact on daylighting, it is not considered in this part. From the
perspective of optimal heat comfort, the middle room is the best one for the three groups of
people. The reason is that the middle room has a small contact surface area with the outside, and
there are rooms on the left and right as thermal buffers. Therefore, the temperature fluctuation
in this room is the smallest, and the thermal environment is suitable. (2) Room Orientation:
From the perspective of optimal daylighting, the optimal orientation for the bedrooms of dif-
ferent generations is south, which means that the south direction is the most favourable one for
indoor daylighting environments. From the perspective of optimal comfort, the orientation of
all rooms is south or south to east, which indicates that west sunlight cannot provide a better
indoor environment in Tianjin and should be avoided as much as possible. (3) Bay and Depth:
All solutions favour smaller bays and depths, suggesting that smaller rooms provide a better
indoor environment. From the daylighting perspective, a lower depth-bay ratio will result in
better daylighting quality. Therefore, when designing a residence, functions with a high demand
for light should be placed in this area as much as possible, while functions with a relatively low
demand for light should be placed deeper in the room. For elderly individual living rooms,
large depths of rooms are sometimes unavoidable and should be designed with artificial light
sources for supplementary lighting. From the thermal comfort point of view, the optimal room

238 Volume 18, Number 2

 TABLE 8. Bedroom optimization parameters of different generations.

TOPSIS Depth Window-to- Horizontal UDI PDH

Rankings Location Orientation Bay bay ratio Window type wall ratio shading DA (%) (%) (%)
Ideal Solutions 1 East room South 3m 1.2 Bay window 0.35 0.5 m 89.8 85.9 69.3
–0.5 m
2 Middle South 3m 1.2 Bay window 0.35 0.4 m 90.0 85.6 70.0
room –0.5 m
3 Middle South 3m 1.2 Bay window 0.45 0.3 m 93.8 73.2 71.9

Adult
room –0.5 m
Optimal Daylighting 20 East room South 3m 1.2 Balcony 0.45 0.5 m 92.5 81.9 75.6
Solution 1.2 m
Optimal thermal 7 Middle South by East 3.3 m 1.2 Bay window 0.3 0.5 m 88.7 82.2 68.8
Comfort Solution room 5° –0.5 m
Ideal Solutions 1 Middle South by East 2.7 m 1.2 Bay window 0.4 0.5 m 92.3 79.7 71.4
room 5° –0.5 m
2 West South 2.4 m 1.2 Bay window 0.5 0.3 m 93.3 79.1 73.7
room –0.5 m
3 East room South 2.4 m 1.2 Ordinary 0.5 0.3 m 94.6 66.3 73.4
Window

Children
Optimal Daylighting 16 East room South 2.4 m 1.2 Balcony 0.5 0.1 m 93.7 78.0 77.9

Journal of Green Building

Solution 1.4 m
Optimal thermal 18 Middle South 3m 1.3 Bay window 0.3 0.5 m 86.7 86.5 70.9
Comfort Solution room –0.5 m
Ideal Solutions 1 Middle South 2.7 m 1.2 Bay window 0.5 0.4 m 92.9 80.7 74.9
room –0.5 m
2 Middle South by East 2.7 m 1.2 Bay window 0.45 0.4 m 93.2 76.5 74.5
room 5° –0.5 m
3 Middle South 2.7 m 1.2 Bay window 0.4 0.3 m 93.0 79.9 75.7
room –0.5 m
Optimal Daylighting 23 East room South 3m 1.2 Balcony 0.5 0.4 m 93.2 79.1 78.6

Elderly individual
Solution 1.1 m
Optimal thermal 14 Middle South by East 3.3 m 1.2 Bay window 0.3 0.5 m 90.7 73.4 72.2
Solution room 15° –0.5 m

239
bays are larger than the other solutions, indicating that a larger bay can improve indoor thermal
comfort. (4) Window Type: According to the optimal daylighting solutions and optimal thermal
solutions, all schemes tend to use the bay window, which means that it can effectively improve
the performance of the indoor environment. (5) Window-to-wall ratio: From the perspective
of thermal comfort and daylighting, the results show that under the current code, a lower
window-to-wall ratio can bring a better indoor comfort environment. From the perspective of
daylighting, a larger window-to-wall ratio will undoubtedly bring better natural daylighting.
The window-to-wall ratio of the ideal solution is between the two. The results show that when
the combined effects of the two factors are considered, the results of the window-to-wall ratio
are within the middle of the upper and lower limits.

5.5 Case study

This paper takes an actual multigenerational residential project in Tianjin as an example and
optimizes it according to the conclusions above, as shown in Table 9. Compared with the

TABLE 9. Scheme comparison before and after optimization.

Original scheme Optimized scheme

Floor plan

Parameter Adult Child Elderly Adult Child Elderly

bedroom bedroom individual bedroom bedroom individual
bedroom bedroom
Opening 3750 mm 3300 mm 3300 mm 3750 mm 3300 mm 3300 mm
Depth- 1.45:1 1.45:1 1.28:1 1.2:1 1.2:1 1.2:1
bay ratio
Window- 0.4 0.46 0.46 0.4 0.4 0.4
to-wall
ratio
Window Balcony Common Bay Bay Bay Bay
type depth window window window window window
–0.5 m –0.5 m –0.5 m –0.5 m

240 Volume 18, Number 2

original scheme, the optimized scheme is modified as follows: (1) According to the above, the
indoor thermal environment of the middle room is optimal, so the elderly individual room is
adjusted to the middle; (2) Considering the actual functional needs, the building plan should
not be changed too much, so the room opening after optimization should be consistent with
the original room; (3) According to the above, reducing the depth-bay ratio can promote the
indoor physical environment, so the optimized depth-bay ratio is set to 1.2:1; and (4) The
window-to-wall ratio has a great impact on the indoor physical environment. According to the
multiobjective optimization results, when the window-to-wall ratio is 0.4, it is most beneficial
to improve the room’s daylighting and thermal environment. (5) When the specific needs of
residents are not considered, the use of a bay window is the most beneficial measure of the
room’s daylight and thermal environment.
A comparison of the results before and after optimization is shown in Table 10. The com-
parison shows that daylight autonomy (DA) of the optimized solution has increased signifi-
cantly, ranging from 31.4% to 38.8%, and useful daylight illuminance (UDI) has increased to
a certain extent, ranging from 2.5% to 10.4%. At the same time, the yearly thermal discomfort
illuminance (UDI) has increased to a certain extent, ranging from 2.5% to 10.4%. At the same time, the yearly thermal
discomfort hours (PDH) ratio is also improved by 2.1% to 8.4%. The optimized solution, therefore, provides an effective
improvement in both daylighting and thermal comfort, which is important for creating a healthy indoor physical
environment.
TABLE 10. Comparative analysis between original and optimized room.
Table 10. Comparative analysis between original and optimized room.

DA
Distribution PDH
Floor Plan DA（
（%）
） Distribution UDI（
（%）
）
Maps （%）
）
Maps

Original
67.1 83.0 77.7
solution

Adult
Bedroom
optimized
89.7 85.1 73.3
solution

promotion 33.7% 2.5% 5.6%

Original
79.9 76.5 75.1
solution

Children’s
Bedroom
optimized
90.4 84.9 73.5
solution

promotion 31.4% 9.8% 2.1%

Original
65.1 76.9 76.2
solution
Elderly
individual
Bedroom optimized
90.4 84.9 69.8
solution

promotion 38.8% 10.4% 8.4%

6. Conclusion

Relying on the Grasshopper parametric platform and Ladybug Tools performance simulation plug-in,
multiobjective optimization of the daylightingJournal
and thermal environment
of Green Buildingof multigenerational residences in Tianjin 241
was
carried out by taking the room location, orientation, bay, depth-bay ratio, window-to-wall ratio, window type and
horizontal shading width as design variables. The research conclusions are as follows:
(1) By analysing the results of the multiobjective optimization of the spatial design variables of bedrooms in
hours (PDH) ratio is also improved by 2.1% to 8.4%. The optimized solution, therefore, pro-
vides an effective improvement in both daylighting and thermal comfort, which is important
for creating a healthy indoor physical environment.

6. CONCLUSION
Relying on the Grasshopper parametric platform and Ladybug Tools performance simulation
plug-in, multiobjective optimization of the daylighting and thermal environment of multigen-
erational residences in Tianjin was carried out by taking the room location, orientation, bay,
depth-bay ratio, window-to-wall ratio, window type and horizontal shading width as design
variables. The research conclusions are as follows:
(1) By analysing the results of the multiobjective optimization of the spatial design vari-
ables of bedrooms in intergenerational housing in Tianjin, the results show that the comfort of
the middle room is the best, and elderly people can be arranged in the middle room first. The
building orientation should be from south to 5° south by east. Sunlight from the west cannot
provide a good indoor environment for the bedroom, so it should be avoided. When designing
the bay and depth, a small value should be taken when meeting the requirements. A bedroom
with a small bay and depth can provide a better indoor environment. Compared with ordi-
nary windows and balconies, bay windows can provide a better environment for the bedroom.
In addition to the window type, a lower window-to-wall ratio can also bring a better indoor
environment. The results can provide a reference for future multigenerational bedroom design.
(2) The optimized solution was used to optimize intergenerational residences in Tianjin.
The results show that the adult bedroom has a 33.7% increase in DA, a 2.5% increase in UDI
and a 5.6% increase in PDH; the children’s bedroom has a 31.4% increase in DA, a 9.8%
increase in UDI and a 2.1% increase in PDH; and the elderly individual bedroom has a 38.8%
increase in DA, a 10.4% increase in UDI and an 8.4% increase in PDH. The optimized solu-
tion can substantially improve the health of the indoor light and heat environment in existing
intergenerational residential buildings. At the same time, it should be noted that the space
parameters obtained from the equilibrium optimal solution and the optimal solution for light
and heat comfort can only represent the optimal values under ideal conditions. In practical
design, appropriate spatial parameter values of the optimal solution set should be used for
optimization according to the actual requirements of the project.
(3) This paper takes the bedrooms in multigenerational residential buildings as the research
object, investigates the behaviour laws of residents in existing multigenerational housing in cold
regions of China, discusses the space needs of different generations of people at different ages,
and proposes the space prototype and size data. On this basis, multiobjective optimization is
carried out for the bedroom space prototype for different generations of people with the goal
of optimizing the daylight and thermal environment indicators and the room size parameters
as variables, and the optimization strategy and the optimal solution are proposed. Finally, the
optimization of the actual project is used to determine the effectiveness of the optimization
strategy. However, when optimizing the daylight and thermal environment, this paper focuses
on the impact of household behaviour characterstics on the results and does not consider the
different thermal comfort zones for people of different ages. Therefore, in the future, the thermal
and light environment needs of different people should be taken into account to optimize the
physical environment for multigenerational residential space.

242 Volume 18, Number 2

REFERENCES
1. Zhou YM, Gong MY. Discussion on aging design of multi generation residential houses. China Housing
Facilities 2014, 04, 46–55.
2. Lv S, Xu WG, Yan X. A Methodological Study of Architectural Digital Design Based on Rapid Feedback.
Architectural Journal 2017, 05, 18–23.
3. Mao P, Qi J, Tan YT, et al. An examination of factors affecting healthy building: An empirical study in east
China. Journal of Cleaner Production 2017, 162, 1266–1274.
4. Yao JW, Lin CX, Shao X. Study on the design strategy of indoor physical environment in cold area. Low
Temperature Architecture Technology 2017, 39(04), 138–141.
5. Sujanova P, Rychtarikova M, Mayor T S, et al. A Healthy, Energy-Efficient and Comfortable Indoor
Environment, a Review. Energies 2019, 12(8), 1414.
6. Monika Frontczak, Pawel Wargocki. Literature survey on how different factors influence human comfort in
indoor environments. Building and Environment, 2010, 46(4):922–937.
7. Wouter D. Van Marken Lichtenbelt, Boris R. Kingma. Building and occupant energetics: a physiological
hypothesis. Architectural Science Review, 2013, 56(1):48–53.
8. Givoni B, Rim Y. Effect of the Thermal Environment and psychological factors upon subject’s responses and
performance of mental work. Ergonomics, 1962, 5: 99–114.
9. Givoni B. A new method for evaluating industrial heat exposure and maximum permissible work load. Int.
J. Biometeor, 1964, 8(2): 115–124.
10. William J. Fisk. How IEQ affects health, productivity. ASHRAE Journal May, 2002.
11. T.T. Chow, K.F. Fong, B. Givoni, Zhang Lin, A.L.S. Chan. Thermal sensation of Hong Kong people with
increased air speed, temperature and humidity in air-conditioned environment. Building and Environment,
2010, 45(10):2177–2183.
12. Martha C. Katafygiotou, Despina K. Serghides. Thermal comfort of a typical secondary school building in
Cyprus. Sustainable Cities and Society, 2014, 13:303–312.
13. Ahrentzen S, Tural E. The role of building design and interiors in ageing actively at home. Building Research
and Information 2015, 43(5), 582–601.
14. Jiao Y, Yu H, Wang Z. et al. Influence of individual factors on thermal satisfaction of the elderly in free
running environments. Building and Environment 2017, 116, 218–227.
15. Iino Y, Igarashi Y, Yamagishi A. Study on the improvement of environmental humidity in houses for the
elderly: Part 1—Actual conditions of daily behavior and thermal environment. Environmental Ergonomics,
2005, 3:231–237.
16. Zheng H. Research on the optimization of residential environment based on the theory of geriatric medicine.
Art Panorama 2020, 06, 120–126.
17. Yun Hyunjun, Nam Insick. A field study of thermal comfort for kindergarten children in Korea An assess-
ment of existing models and preferences of children. Building and Environment, 2014, 75(5): 182–189.
18. Kyle Konis. Evaluating daylighting effectiveness and occupant visual comfort in a side-lit open-plan office
building in San Francisco, California. Building and Environment, 2013, 59:662–677.
19. P T Stone. The effects of environment illumination on melatonin bodily rhythms and mood states: a review.
Lighting Research and Technology, Vol 31 No. 3. 1999.
20. Hou D, Liu G, Zhang Q. et al. Integrated Building Envelope Design Process Combining Parametric
Modelling and Multi-Objective Optimization. Trans. Tianjin Univ 2017, 23, 138–146.
21. Gou S, Nik V M, Scartezzini J-L, et al. Passive design optimization of newly-built residential buildings
in Shanghai for improving indoor thermal comfort while reducing building energy demand. Energy and
Buildings 2018, 169, 484–506.
22. Makki M, Showkatbakhsh M, Song Y. Wallacei: An evolutionary and Analytic Engine for Grasshopper 3D.
Wallacei, 2018.
23. Showkatbakhsh M, Kaviani S, Weinstock M. Evolutionary Design Processes with Embedded Homeostatic
Principles—Adaptation of Architectural Form and Skin to Excessive Solar Radiation. Computer-Aided Design
and Applications 2021, 18(5):914–953.
24. Qian W, Dezhi Z, Chong W. Study of the selection of old industrial building renovation based on aph topsis.
Industrial Construction 2014, 44(9), 26–29.

Journal of Green Building 243

25. Bre F, Fachinotti V. A computational multi-objective optimization method to improve energy efficiency and
thermal comfort in dwellings. Energy and Buildings 2017, 154.
26. Ministry of Housing and Urban-Rural Development of the People’s Republic of China. Code for design of
residential building for the aged GB 50340. Beijing: China Architecture & Building Press, 2016.
27. Ministry of Housing and Urban-Rural Development of the People’s Republic of China. Design code for
residential buildings GB 50096. Beijing: China Architecture & Building Press, 2011.
28. Perez R, Seals R, Michalsky J. To All-weather model for sky luminance distribution-Preliminary configuration
and validation. Solar Energy 1993, 50(3), 235–245.
29. Liu L, Liang J, Sun C. A study on daylighting performance prediction model for office buildings in mid-
latitude cold region of china. New Architecture 2015, 05, 46–51.
30. Guan Y. Research progress of daylighting and investigation of classroom lighting in chongqing. Light &
Lighting 2020, 44(04), 16–19.
31. Wirz-Justice A, Skene DJ, Munch M. The relevance of daylight for humans. Biochemical Pharmacology 2021,
191, 114304.
32. Khademagha P, Aries MBC, Rosemann ALP, et al. Implementing non-image-forming effects of light in the
built environment: A review on what we need. Building and Environment 2016, 108, 263–272.
33. F Cantin, MC Dubois. Daylighting metrics based on illuminance, distribution, glare and directivity. Lighting
Research & Technology 2011, 43(3), 291–307.
34. Wu W, Liu KP. An introduction of new daylighting evaluation criteria: a replacement for daylight factor.
China Illuminating Engineering Journal 2012, 23(02), 1–7+24.
35. Bluyssen PM, C Roda, C Mandin, et al. Self-reported health and comfort in “modern’ office buildings: first
results from the European OFFICAIR study. Indoor Air 2016, 26(2), 298–317.
36. Lan L, P Wargocki, DP Wyon, et al. Effects of Thermal Discomfort in an Office on Perceived Air Quality,
SBS Symptoms, Physiological Responses, and Human Performance. Indoor Air 2011, 21(5), 376–390.
37. Gao Y, Chi JY, Luo SL, et al. Multi-objective optimization technology framework of passive energy-saving
design for Tianjin rural houses through Octopus+Honeybee. Journal of Xi’an University of Technology 2021,
37(04), 488–497.
38. Gao Y, Hu K, Yue XP, et al. Shape parameters design of northern rural houses for mulit-objective optimization
of energy performance and thermal comfort. Journal of Huaqiao University (Natural Science) 2021, 42(05),
619–627.
39. Ren FK, Wei ZQ, Zhai XQ. Multi-objective optimization and evaluation of hybrid CCHP systems for dif-
ferent building types. Energy 2021, 215(A),119096.

244 Volume 18, Number 2

Copyright of Journal of Green Building is the property of College Publishing and its content
may not be copied or emailed to multiple sites or posted to a listserv without the copyright
holder's express written permission. However, users may print, download, or email articles for
individual use.