Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Exploring the influence of scheduling on daylighting performance evaluation in assisted living: testing two new metrics based on sDA and ASE
(USC Thesis Other)
Exploring the influence of scheduling on daylighting performance evaluation in assisted living: testing two new metrics based on sDA and ASE
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
i
EXPLORING THE INFLUENCE OF SCHEDULING ON
DAYLIGHTING PERFORMANCE EVALUATION IN ASSISTED
LIVING:
Testing Two New Metrics Based on sDA and ASE
By
Ruoxiao Zeng
Presented to the
FACULTY OF THE
SCHOOL OF ARCHITECTURE
UNIVERSITY OF SOUTHERN CALIFORNIA
In partial fulfillment of the
Requirements of degree
MASTER OF BUILDING SCIENCE
August 2019
ii
ACKNOWLEDGEMENTS
First, I wish to present my special thanks to my chair, Professor Kyle Konis, who gave me a lot of advice on
daylight metrics and Grasshopper, and introduced professionals to help on me thesis.
I would also show my gratitude to my committee members, Professor Karen Kensek, Professor Lauren
Dandridge, and Professor Victor Regnier. Karen gave me much help on thesis idea and writing. Lauren
helped on professional lighting issues and Victor provided suggestions on assisted living related issues.
Last but not least, I would like to thank all the MBS students and my family. They gave me a lot of support
on my study and my life.
iii
COMMITTEE MEMBERS
Kyle Konis, Ph.D., AIA
Assistant Professor
University of Southern California
kkonis@usc.edu
Karen Kensek, LEED BD+C
Professor of Practice
University of Southern California
kensek@usc.edu
Lauren Dandridge Gaines
Adjunct Assistant Professor
University of Southern California
ldandrid@usc.edu
Victor Albert Regnier, FAIA
Professor
University of Southern California
regnier@usc.edu
iv
ABSTRACT
Daylighting is important for occupants in buildings. To evaluate the lighting performance, metrics are needed
to measure daylighting. Knowing when and where people are in the building can be an important
consideration when analyzing daylighting. IES LM-83-12 gives an approved method on the evaluation of
the lighting performance of interior space. Two metrics, one for daylighting, Spatial Daylight Autonomy
(sDA) and one for glare, Annual Sunlight Exposure (ASE), are used to describe the lighting performance of
“regularly occupied space” of the common working space
SDA and ASE are widely-used in daylighting performance evaluation. Thus, their scope may be able to be
extended to other building types. However, according to IES LM-83-12, the analysis period for both sDA
and ASE is fixed from 8am to 6pm (10 hours). People in specific building types such as assisted living may
not follow that schedule, which may lead to an inaccurate lighting performance prediction. Therefore, new
metrics may be needed to reveal the lighting performance more accurately according to real operating
schedule. New metrics based on sDA and ASE using a flexible analysis period can determine the lighting
performance more fully. The two new interior lighting metrics proposed are called flexible analysis period
spatial Daylight Autonomy (fsDA) and flexible analysis period Annual Sunlight Exposure (fASE).
In addition, IES LM-83-12 does not define “regularly occupied space” in detail. There is also no hierarchy
within “regularly occupied space.” Space where more people spend more time in is more important, which
means the more important spaces have more priority to achieve better lighting condition. Therefore, spaces
with different occupied levels may need different daylighting requirement. As a result, a standard on fsDA
and fASE or sDA and ASE of spaces with different occupied levels may be needed.
First, a Rhino model of a case study building was built. The space of the model was divided into 6 function
spaces. The schedules of assisted living were obtained by visiting assisted living communities, as well as by
talking with experts and staff of senior livings. Then the sDA
300,50%
and the ASE
1000,250h
of the case building
were calculated according to the method suggested by IES LM-83-12. After that, fsDA
300,50%
and
fASE
1000,250h
were calculated based on the operating hours rather than 8am to 6 pm (the fixed 10 hours). Next,
the new calculation results were compared with that calculated based on the 10-hour period and a small
difference was found between the two results. Thus, fsDA and fASE, who have a flexible analysis period,
are not necessarily needed to replace sDA and ASE even though the operation hour changes in assisted living.
Occupied Hour (O) and Hour Percentage (P) were developed to define the occupied hierarchy of function
spaces. It was tested if the values of the two metrics could be predicted by operation schedules by finding
their values under different operation schedules. If they are predictable, they can be used to define the
occupied hierarchy. Otherwise, they cannot be used to decide the occupied hierarchy. It was found that both
of the two metrics were influenced by operation schedules while Hour Percentage (P) can be influenced by
occupant spatial distribution as well. Therefore, Occupied Hour (O) can be used as one factor to define
occupied hierarchy while the feasibility if Hour Percentage (P) can decide occupied hierarchy need to be
discussed after a sensitivity test. It was tested if Hour Percentage (P) was sensitive to occupant spatial
distribution by finding the value of the metric under the same operation schedule while with different
occupant spatial distributions. It was found Hour Percentage (P) was proportionally sensitive to occupant
spatial distribution so whether the metric can be used to define occupied hierarchy of function spaces cannot
be decided at this stage. The variation range of occupant spatial distribution should be explored in the future.
Based on the findings, “regularly occupied” space can be defined based on Occupied Hour (O) while whether
there is a predicted occupied hierarchy within “regularly occupied” space based on Hour Percentage (P) is
not decided yet. A reference on sDA and ASE for “regularly occupied” function should be suggested. If there
is a more detailed occupied hierarchy within “regularly occupied” space, a more detailed standard on sDA
and ASE for the function spaces with different occupied levels may be suggested in the future. The
management group can set up an operation schedule to make more people staying in the spaces with better
daylighting condition to meet the sDA and ASE requirement.
KEYWORDS
sDA, ASE, fsDA, fASE, occupied hierarchy, assisted living
v
HYPOTHESIS
• The sDA and ASE can vary significantly if they are calculated based on the real operating schedule, rather
than using the standard 8am to 6 pm period.
• Occupied Hour (O) and Hour Percentage (P) can predictable based on the operation schedule of the building
so O and P can be be used to define occupied hierarchy of function spaces.
RESEARCH OBJECTIVES
• Prove if fsDA and fASE, the metrics using real occupied hours as the analysis period are
necessary to describe the lighting performance of the building.
• Demonstrate that the occupied hierarchy of fsDA and fASE may be defined based on the real
Occupied Hours (O) and Hour Percentage (P).
vi
TABLE OF CONTENTS
ACKNOWLEDGEMENTS ........................................................................................................................... ii
COMMITTEE MEMBERS .......................................................................................................................... iii
ABSTRACT ................................................................................................................................................. iv
HYPOTHESIS ................................................................................................................................................ v
RESEARCH OBJECTIVES ........................................................................................................................... v
1. INTRODUCTION ...................................................................................................................................... 1
1.1 Daylighting ........................................................................................................................................... 1
1.1.1 Importance of Daylighting ............................................................................................................ 1
1.1.2 Measurement ................................................................................................................................. 2
1.1.3 Daylight Autonomy (DA) ............................................................................................................. 3
1.1.4 Simulation: TMY and epw weather file ........................................................................................ 3
1.2 Daylight performance evaluation ......................................................................................................... 4
1.2.1 IES LM-83-12 ............................................................................................................................... 4
1.2.2 Spatial Daylight Autonomy (sDA) ............................................................................................... 4
1.2.3 Annual Sunlight Exposure (ASE) ................................................................................................. 5
1.2.4 LEED BD+C daylight evaluation ................................................................................................. 6
1.3 Specific Problem of sDA and ASE ...................................................................................................... 7
1.3.1 Fixed analysis period .................................................................................................................... 7
1.3.2 “Regularly occupied space” defined unclear; no occupied hierarchy ........................................... 8
1.3.3 Solution ......................................................................................................................................... 8
1.4 Assisted living ...................................................................................................................................... 8
1.4.1 Definition ...................................................................................................................................... 9
1.4.2 Assisted living overview ............................................................................................................... 9
1.4.3 Terms ............................................................................................................................................ 9
1.4.4 The case building ........................................................................................................................ 12
1.5 Software Tools ................................................................................................................................... 18
1.6 Summary ............................................................................................................................................ 21
2. BACKGROUND AND LITERATURE VIEW ........................................................................................ 23
2.1 Assisted living .................................................................................................................................... 23
2.1.1 The importance of daylight to assisted living ............................................................................. 23
2.1.2 Residents in assisted living ......................................................................................................... 23
2.1.3 Assisted living schedules ............................................................................................................ 25
2.2 Daylight evaluation metrics ................................................................................................................ 27
2.3 Spatial Daylight Autonomy (sDA) and Annual Sunlight Exposure (ASE) ........................................ 29
2.3.1 Spatial Daylight Autonomy (sDA) ............................................................................................. 30
2.3.2 Annual Sunlight Exposure (ASE) ............................................................................................... 31
2.4 LEED BD+C daylight evaluation ....................................................................................................... 33
2.4.1 Option 1 ...................................................................................................................................... 33
2.4.2 Option 2 ...................................................................................................................................... 33
2.4.3 Option 3 ...................................................................................................................................... 34
vii
2.5 Method to capture the behavior patterns of older adults .................................................................... 34
2.6 Summary ............................................................................................................................................ 35
3. METHODOLOGY ................................................................................................................................... 37
3.1 Methodology diagram ........................................................................................................................ 37
3.2 Preparation ......................................................................................................................................... 38
3.3 Explore the influence of analysis period on sDA and ASE ................................................................ 41
3.3.1 Calculate sDA
by Honeybee ....................................................................................................... 43
3.3.2 Calculate ASE by Ladybug ......................................................................................................... 54
3.3.3 Calculate fsDA and fASE ........................................................................................................... 62
3.3.4 Compare fsDA with sDA, and fASE with ASE .......................................................................... 63
3.3.5 Suggest whether fsDA and fASE are necessary .......................................................................... 66
3.4 Explore occupied hierarchy of the function spaces ............................................................................ 66
3.4.1 Simulate occupant spatial distribution ........................................................................................ 67
3.4.2 Sensitivity test for Hour Percentage (P) to occupant spatial distribution .................................... 73
3.4.3 Analyze the result of the occupant spatial distribution of the three schedules ............................ 74
3.4.4 Analyze the result of the sensitivity test ..................................................................................... 74
3.4.5 Suggest a hierarchy for function space ....................................................................................... 76
3.5 Summary ............................................................................................................................................ 77
4. RESULTS ................................................................................................................................................. 78
4.1 Influence of analysis period on sDA and ASE Simulation Outcomes................................................ 79
4.2 Occupant spatial distribution of three schedules ................................................................................ 81
4.2.1 Occupant spatial distribution under schedule 1 ........................................................................... 82
4.2.2 Occupant spatial distribution of schedule 2 ................................................................................ 83
4.2.3 Occupant spatial distribution of schedule 3 ................................................................................ 85
4.3 Sensitivity test .................................................................................................................................... 86
4.3.1 Test people move from apartment to activity space .................................................................... 86
4.3.2 Test people move from apartment to office ................................................................................ 90
4.4 Summary ............................................................................................................................................ 92
5. DISCUSSION ........................................................................................................................................... 94
5.1 Compare fsDA with sDA, and fASE with ASE ................................................................................. 95
5.1.1 Compare sDA and fSDA ............................................................................................................ 96
5.1.2 Compare ASE and fASE ........................................................................................................... 102
5.1.3 Conclusion ................................................................................................................................ 108
5. 2 Compare occupant spatial distribution under the three schedules ................................................... 108
5.3 Sensitivity test .................................................................................................................................. 110
5.3.1 Sensitivity test result ................................................................................................................. 110
5.3.2 Analyze the result of the sensitivity test: Hour Percentage Change .......................................... 111
5.3.3 Suggest a sample occupied hierarchy for the space .................................................................. 112
5.4 Call for requirement on sDA and ASE for function spaces with different occupied levels ............. 113
5.5 Summary .......................................................................................................................................... 114
viii
6. CONCLUSION AND FUTURE WORK ............................................................................................... 116
6.1 Conclusions ...................................................................................................................................... 116
6.2 Future work ...................................................................................................................................... 119
6.2.1 Expand sample quantity and variety ......................................................................................... 119
6.2.3 Use new technics to detect occupied status ............................................................................... 119
6.2.4 Simulate sDA in higher quality ................................................................................................. 119
6.2.5 Explore the variation range of Hour Percentage (P) ................................................................. 120
6.2.6 Define occupied hierarchy ........................................................................................................ 120
6.2.7 Find out suggested sDA and ASE for function space with different occupied levels ............... 120
6.2.8 Circadian lighting and occupied hierarchy................................................................................ 120
6.3 Summary .......................................................................................................................................... 121
APPENDIX A ............................................................................................................................................ 124
APPENDIX B ............................................................................................................................................. 138
REFERENCES ........................................................................................................................................... 139
1
1. INTRODUCTION
This chapter is about daylighting, daylight performance evaluation, specific problem of sDA and ASE,
assisted living, and software tools. Daylighting is important to buildings. Two measurements of daylighting,
including illuminance and glare are introduced. A daylight metric Daylight Autonomy (DA) is described.
Typical Meteorological Year and epw weather file are explained. Two daylighting metrics, spatial Daylight
Autonomy (sDA) and Annual Sunlight Exposure (ASE) are introduced and analyzed while some limitations
are found. Assisted living can be a good example to study daylighting and schedules. Terms are introduced
for assisted living research. A case building is presented by using the terms. Software tools including Rhino,
Grasshopper and the plugins for Grasshopper including Ladybug and Honeybee, and Microsoft Excel are
described.
1.1 Daylighting
This section is about the importance of daylighting, the measurement of daylighting, Daylight Autonomy
(DA), and two terms in daylighting performance simulation. Daylighting is important to interior space. Some
spaces have good access to daylight while others do not. Illuminance and glare are two important
measurements of daylight. Illuminance is to measure the luminous flux striking a surface. Glare is the visual
sensation caused by uncontrolled and overly brightness. Daylight Autonomy (DA) measures the annual
sufficiency of ambient daylight levels. Typical Meteorological Year (TMY) and epw weather file are used
for the simulation of daylight metrics.
1.1.1 Importance of Daylighting
Daylighting has many aesthetic and health benefits. For example, daylighting can increase people’s
productivity and comfort, as well as help to regulate human circadian rhythms (Wymelenberg 2014). The
rooms without windows can lead to stress (Boubekri 2008). A research found a minimum of 3 hours of
exposure to daylight reduced stress and burnout while the exact extent of the reduction was not specified
(Boubekri 2008). Another research found that medical errors among nurses were more often in midwinter
than in the fall or summer (Boubekri 2008). A strong relationship between outside darkness and the rate of
medical errors was found while whether there was a causality between them was not revealed (Boubekri
2008). In addition, utilizing daylight can lead to significant energy savings (Wymelenberg 2014). Financial
benefits can also be achieved by buildings with good natural lighting condition. Researchers found the
commercial real estate without windows leases for around 20% less than the spaces with windows
(Wymelenberg 2014). Some interior spaces have good access to daylight while others do not (Figure 1-1, 1-
2, 1-3, 1-4). Thus, studying daylighting of buildings is important.
Figure 1-1: An interior residential space with Figure 1-2: An interior space of a library with good
good access to daylight (Bristolite Team 2014) access to daylight
2
Figure 1-3: A reading space of an assisted living Figure 1-4: A bedroom of an assisted living with
with no daylight access daylight access
1.1.2 Measurement
Two measurements, illuminance and glare, are introduced. Illuminance is to measure the luminous flux
striking a surface. Glare is the visual sensation caused by uncontrolled and overly brightness.
• Illuminance
Illuminance is the luminous flux striking a given surface (Figure 1-5). It is the quotient of lumens and area.
For example, if the luminous flux of 1000 lumens beamed uniformly on a surface which is 100 m
2
, the
illuminance is 10 lux. Within the field of vision, the performance of human eye largely depends on
illuminance. Generally, people’s performance improves with growing illuminance while the mistake
decreases (Knowledge Base DIALux 4 2016)
Figure 1-5: Illuminance (Jrh.main 2018)
• Glare
Glare means a visual sensation caused by uncontrolled and overly brightness. It can be uncomfortable or
even disabling (Figure 1-6). The sensitivity to glare can be quite different among the subjective. The elderly
is usually more sensitive to glare because of the aging eye. Even though some of them had cataract surgery
which resolves the aging eye/color distortion and glare problem reasonably well, some of them are still
suffering from the problems of aging eyes (Regnier, Senior living related issues 2019). Discomfort glare is
the sensation of pain or just simply annoyance caused by excessively bright sources. Disability glare
describes the reduction in visibility induced by intense light sources in the field of view (Rensselaer
Polytechnic Institude 2007).
3
Figure 1-6: The example that daylight can lead to discomfort and disability glare (Astrobob 2009)
1.1.3 Daylight Autonomy (DA)
Daylight Autonomy (DA) is a dynamic daylight performance metrics, which is based on illuminance within
a building measures (Reinhart, Mardaljevic and Rogers 2006). DA measures the percentage of occupied time
that the daylight levels exceed a specified illuminance (Max 2012) The subscripts represent the illuminance
level. For example, DA
300
represents the percentage of the occupied time that the illuminance meets or
exceeds 300 lux. The three inputs for DA are the target illuminance, climate-based daylighting levels over
the entire year, and the real hours of operation of the space (Max 2012)(Figure 1-7).
There are several good points about DA:
• DA considers the real operation period and real weather condition of the site (Max 2012).
• DA provides an intuitive overview of how well daylight penetrating in to the space (Max 2012).
• DA can help to determine which fixtures can benefit from automatic daylight harvesting (Max
2012).
• DA can be visualized so the lighting condition of each point can be identified clearly (Figure 1-7).
Figure 1-7: An example of visualized DA
300
However, because DA only evaluates the daylighting performance of a single point, it is hard to describe the
daylighting performance of a space by a single number. Numerous DA values are needed to illustrate the
daylighting condition of a tested space.
1.1.4 Simulation: TMY and epw weather file
Typical Meteorological Year (TMY) is a series of meteorological data whose data values for every hour in
a year for certain geographical location. The data are selected from a much longer time period (normally 10
years or more). The typical data selected for a specific month can be from different years. For instance,
4
January may be from 2005, February from 2016 and so on. TMY are used for many fields. For example,
EnergyPlus, a software for calculating the energy performance of buildings. The output format can be chosen,
either in Energy Plus Weather File format (EPW) or a generic CSV format (European Energy Efficiency
Platform n.d.). The calculation of the daylight metrics, DA, sDA and ASE require using TMY and epw
weather file. Epw format was created by the US Department of Energy (DoE) as a standard weather data
format. Some other data formats can be converted into epw format. The epw weather file includes all 8760
hours of weather data of a 365-day year. The location information, wind data, temperature, humidity,
enthalpy and solar radiation data are covered by epw weather file (Paulwintour 2016).
1.2 Daylight performance evaluation
This section covers IES LM-83-12, spatial Daylight Autonomy (sDA), Annual Sunlight Exposure (ASE)
and LEED BD+C daylight evaluation. IES LM-83-12 developed two daylight metrics to advance the
predictive performance of existing daylighting metrics. The first metric sDA is based on DA to measure the
spatial performance of annual sufficiency of ambient daylight levels (IES Daylight Metrics Committee 2012).
The other metric ASE is to predict glare (IES Daylight Metrics Committee 2012). IES LM-83-12 suggested
using sDA and ASE to evaluate interior daylighting condition (IES Daylight Metrics Committee 2012).
LEED borrowed this idea (USGBC 2019).
1.2.1 IES LM-83-12
IES stands for Illuminating Engineering Society. An important job for IES is to create Lighting Standards
for the benefits of the public (Illuminating Engineering Society n.d.). The document “Approved Method:
LES Spatial Daylight autonomy (sDA) and Annual Sunlight Exposure (ASE)” is numbered as IES LM-83-
12. The document developed a new series of metrics to describe significant scopes of daylighting
performance in an existing building. The new metrics are to advance the predictive performance of existing
metrics, such as Daylight Factor and define a standard methodology which enables some design alternatives
of daylit buildings, proposed designed, and/or climatic locations to be compared in a constant manner.
During the metric development process, historical accepted daylight performance metrics were evaluated. It
was found that no individual metric could fully describe all of the dimensions included in a successful
daylighting system. The committee proposed that useful metrics must assess the lighting condition of the
whole daylit area of a full year, taking daily and yearly climatic variation into consideration, rather than the
existing method of assessing a single point in time. Finally, two metrics were developed. The two metrics
enable a daylit space to be assessed for a year-long period. The first metric is spatial Daylight Autonomy
(sDA), which illustrates the sufficiency of daylight illuminance. The other metric is Annual Sunlight
Exposure (ASE), which predicts the potential risk of excessive sunlight penetration (IES Daylight Metrics
Committee 2012).
1.2.2 Spatial Daylight Autonomy (sDA)
All definitions in 1.2.2 are from IES Daylight Metrics Committee 2012 (IES Daylight Metrics Committee
2012).
IES LM-83-1 defined Spatial Daylight Autonomy (sDA). The definition, standard threshold, analysis criteria,
analysis area, analysis period, and analysis points are important parts of it.
• Definition
Spatial Daylight Autonomy (sDA) is a metric to indicate the annual sufficiency of ambient daylight levels
in interior space. sDA is defined as the percentage of a studied area that meets a minimum daylight
illuminance threshold for a specified fraction of the operation hours per year. The metrics is intended be
applied to common workplace environments such as open offices, multi-purpose rooms, service arear in
libraries and lobbies, and so are most applicable to areas with similar visual tasks. The ideal analysis area is
a coherent “space” defined by opaque walls and access to daylighting through at least one wall or ceiling
surface. The metrics can also be applied to the “regularly occupied” floor area of a building, or some part of
a building such as one floor plate.
• Suggested sDA Performance Criteria
sDA
300,50%
is the suggested metric for daylight sufficiency analysis. The subscripts represent the illuminance
level and time fraction. SDA
300,50%
means the percentage of the analysis points meet or exceeds 300 lux for
5
at least 50% of the analyzed hour over a year. To realize nominally accepted daylight sufficiency, sDA
300,50%
must be no lower than 55%. To achieve preferred daylight sufficiency, sDA
300,50%
must be no lower than
75%.
• Analysis area
To analyze whole buildings, the area to be analyzed should be “regularly occupied work areas,” which means
spaces with very low occupancy such as storage rooms and copy rooms can be excluded from the analyzed
area.
• Analysis period
The period for analysis is fixed, from 8am to 6pm local clock time. The result is 10 hours per day (3650
hours for a complete annual analysis). Although the operating schedules can vary in different buildings, the
10-hour day was regarded as the standard analysis period because by using the fixed time, so different
building designs can be compared. sDA can be used to describe the “asset value.” Other time periods of
analysis can be used for study only. The daylight conditions are rest on typical meteorological year (TMY)
data. Typical Meteorological Year (TMY) formatted hourly weather data by hourly average, with data
centered on the half-hour. For instance, hour 9 on a day is an average of 8am to 9am. When modelling sDA,
the time period uses the solar position on 8:30 am.
• Analysis points
Within the analysis areas, the points to be analyzed should be continuous, specified no larger than 2 feet on
center. The points should be from 1 to 2 feet offset from the walls, with as height of 2.5 feet above finished
floor. A two-foot grid is the maximum scale for the tiling of the sDA analysis (Figure 1-8).
Figure 1-8: Analysis points of sDA
1.2.3 Annual Sunlight Exposure (ASE)
All definitions in 1.2.3 are from IES Daylight Metrics Committee 2012.
IES LM-83-1 also defined Annual Sunlight Exposure (ASE). The definition, standard threshold, analysis
criteria, analysis area, analysis period and analysis points were introduced in the section.
• Definition
Annual Sunlight Exposure (ASE) is a metric that expresses the possibility for visual discomfort(glare) in
interior space. ASE is defined as the percentage of an analyzed area that surpasses a specified direct sunlight
illuminance level more than a specified number of hours per year. The metrics is intended be applicable to
6
common workplace environments such as open offices, multi-purpose rooms, service arear in libraries and
lobbies, and so are most applicable to areas with similar visual tasks. The ideal analysis area is a coherent
“space” defined by opaque walls and access to daylighting through at least one wall or ceiling surface. The
metrics can also be applied to the “regularly occupied” floor area of a building, or some part of a building
such as one floor plate.
• Suggested ASE Performance Criteria
The subscripts represent the illuminance level and hours. ASE
1000,250h
is the suggested metric for visual
discomfort analysis. ASE
1000,250h
represent the percentage of the analyzed space where are exposed to more
than 1000 lux of direct sunlight for at least 250 hours over a year. Space with more than 10% ASE
1000,250h
is
regarded to have unsatisfactory visual comfort. Space with ASE
1000,250h
less than 7% is nominally acceptable.
When the value is less than 3% ASE
1000,250h
, the space can be clearly acceptable. Typical Meteorological
Year (TMY) formatted hourly weather data by hourly average, with data centered on the half-hour. For
instance, hour 9 on a day is an average of 8am to 9am. To modelling sDA, the time period uses the solar
position on 8:30 am.
• Analysis area
To analyze whole buildings, the area to be analyzed should be “regularly occupied work areas,” which means
spaces with very low occupancy such as storage rooms and copy rooms can be excluded from the analyzed
area.
• Analysis period
The period for analysis is fixed, from 8am to 6pm local clock time. The result is 10 hours per day (3650
hours for a complete annual analysis). The analysis period is the same as that of Spatial Daylight Autonomy
(sDA).
• Analysis points
Within the analysis areas, the points to be analyzed should be continuous, specified no larger than2 feet on
center. The points should be from 1 to 2 feet offset from the walls, with as height of 2.5 feet above finished
floor. A two-foot grid is the maximum scale for the tiling of the ASE analysis (Figure 1-9).
Figure 1-9: Analysis points of ASE
1.2.4 LEED BD+C daylight evaluation
Except for IES LM-83-1, there are some other rating systems such as LEED (Leadership in Energy and
Environmental Design), the most popular green building rating system all over the world (USGBC n.d.) .
7
Framework to make economic, efficient and healthy green buildings was provided by LEED (USGBC n.d.)
For LEED BD+C v4, there are 110 points in total. Several categories are considered by LEED: Indoor
Environment Quality, Location and Transportation, Water Efficiency, Sustainable Sites, Materials and
Resources, Energy and Atmosphere, Innovation and more (USGBC n.d.). The certification level is decided
by the credits a project earns: Certified (40-49 points), Silver (50-59 points), Gold (60-79 points) and
Platinum (80+) (USGBC n.d.). LEED v4 for Building Design and Construction gives 16 points to Indoor
Environmental Quality (USGBC 2016).
Within the Indoor Environmental Quality category, 1-3 points can be given to daylight, aiming to connect
building occupants with the outdoor environment, strengthening circadian rhythms and to reduce the
consumption of electrical lighting by using daylight (USGBC 2019). Three options are provided to evaluate
the daylighting performance of interior space. Only one of them needs to be selected (USGBC 2019).
• Option 1 is about spatial Daylight Autonomy(sDA) and Annual Sunlight Exposure (ASE)
(normally 2-3 points, 1-2 points for healthcare). LEED borrows the suggested sDA and ASE from
IES LM-83-12.
• Option 2 requires calculating illuminance by computer modelling at two time points at the
equinox.
• Similar to option 2, option 3 is also about illuminance simulation at two time points. However,
option 3 requires fixtures, furniture and equipment in place. And option 3 takes first measurement
by computer modelling in a regularly occupied month and then take the second measurement of
another required time (USGBC 2019). However, none of the three options take the real operation
schedule of the building into account.
1.3 Specific Problem of sDA and ASE
This section is about the specific problem of sDA and ASE, including their limited scope, fixed analysis
period, the unclearly defined “regularly occupied space, the lack of occupied hierarchy and the solution. sDA
and ASE does not consider the operating schedule of the building. According to IES LM-83-12, the analyzed
period for sDA and ASE is fixed at 8am to 6pm (10 hours). In addition, the document does not define
“regularly occupied space” in detail. However, people in specific building types such as senior housing may
not follow the schedule and the occupied condition of the space can vary remarkably. sDA and ASE may
not be an accurate enough lighting performance description because of this.
1.3.1 Limited scope
SDA and ASE can describe the annual daylighting performance of interior space by a single number. The
two metrics are very useful in the description of annual daylighting performance. However, the two metrics
are originally intended to be applied to common working environment such as offices, meeting room and
class rooms. As the two metrics are effective in daylighting performance description, the measurement scope
may be able to extend to varies of building types such as assisted living.
1.3.1 Fixed analysis period
If the scope for sDA and ASE are extended to assisted living, the analysis period for the two metrics becomes
a problem. The analyzed period is fixed, from 8am to 6pm local clock time. The result is 10 hours per day
(3650 hours for a complete annual analysis). However, the real operating hours of specific building types
such as assisted livingmay not be 8am to 6 pm. The standard analysis period cannot illustrate the lighting
performance of the space accurately because of this.
IES LM-83-12 gives its reason why choose the fixed analysis period. According to the document, the 10-
hour day was regarded as the standard analysis period because by using the fixed time, different building
designs can be compared although the operating schedules can vary in different buildings. sDA and ASE can
be used to describe the “asset value.” Other time periods of analysis can be used for study only (IES Daylight
Metrics Committee 2012).
Calculating sDA and ASE with the fixed analysis period could describe the “asset value.” However, it can
be unfair to compare the space with different operating schedules by using the same analysis period. It is not
8
true that a space with low sDA and high ASE must have a poor lighting performance. For example, if a room
is only used for 1 pm to 3 pm, the lighting conditions of the space may be good during the period while it
can be poor at 8 am to12pm. Whatever, the lighting performance of the space can be regarded poor according
to IES LM-83-12. However, the occupants experience can be good as the lighting condition is good when
they are in the room. ASE has the similar problem in not taking time into account.
New metrics such as flexible analysis period spatial Daylight Autonomy (fsDA) and flexible analysis period
Annual Sunlight Exposure (fASE) that uses the real occupied hours as the analysis period for the results may
be used to describe the lighting performance because they can reflect the lighting performance of the space
more accurately.
1.3.2 “Regularly occupied space” defined unclear; no occupied hierarchy
IES LM-83-12 does not clearly define what “regularly occupied space” is. All the space except for those
with a really low occupancy such as storage and circulation space are regarded as “regularly occupied space”
(IES Daylight Metrics Committee 2012).
• “Regularly occupied space” not defined based on real occupied hours
Sometime, the normally lowly-occupied space can be highly occupied while the regular highly-occupied
space can be lowly occupied. For example, people may use the circulation space as a relaxing area, and they
may spend a lot of time there. Normally, office is regarded as “regularly occupied space.” Nevertheless, the
occupants may only stay in the office for an hour every day. Therefore, the “regularly occupied space” should
be defined based on the real occupied hours of the room. If at least one person stays in the space for more
than certain hours, the space should be regarded as a “regularly occupied space” and the space need to
achieve the LEED lighting requirement.
• No occupied hierarchy
In addition, there is no hierarchy for the occupied level. There are only “regularly occupied space” and “the
space with very low occupancy” suggested by IES LM-83-12. Assume an activity room and an office have
the same SDA
300,50%
and ASE
1000,250h
. 20 people stay in the activity room for 4 hours a day. Only 1 person
stay in an office for 8 hours a day. Both of the rooms are “regularly occupied space.” In total, 80 hours are
spent in the activity room, and 8 hours are spent in the office. However, Iif the lighting performance of the
activity room, rather than the office, is improved, more hours in total with better lighting condition can
benefit from that.
Thus, the space with more hours that all people spend in may be regarded as more densely occupied space
than the “regularly occupied space.” Higher lighting requirement of the space may be necessary. The space
which takes up more than certain percentage of the total hours all the occupants spend in the building can be
highly occupied space. Since there are hierarchy among the occupied levels, the fsDA
300,50%
and fASE
1000,250h
requirements may be different for the spaces with different occupied levels.
1.3.3 Solution
New metrics based on sDA and ASE, except for using the real operating hours as the analysis period, may
describe the lighting performance much more fully than sDA and ASE for an assisted living. The new metrics
are called flexible analysis period spatial Daylight Autonomy (fsDA) and flexible Annual Sunlight Exposure
(fASE). New standard for fSDA
300,50%
and fASE
1000,250h
may be set up based on different occupied hierarchy.
As operating hours are the main concerns of the new metrics, the architect may suggest an operation schedule
for the management group of a building to achieve good daylighting performance of the interior space.
1.4 Assisted living
This section is about assisted living, including the definition, the overview, terms that will be used and the
case building. Assisted living usually has a regular and flexible schedule. Therefore, assisted living can be a
suitable example to explore daylighting and operation schedule. Terms including “architectural program,”
“function space,” “Occupied Hour (O)” and “Hour Percentage (P)” are introduced for the assisted living
daylighting study.
9
1.4.1 Definition
In 2001, the Assisted Living Federation of America (ALFA) developed a widely accepted definition of
assisted living:
…[A] special combination of housing, supportive services, personalized assistance and healthcare
designed to respond to the individual needs of those who require help with activities of daily living
(ADL) and instrumental activities of daily living (IADL). Supportive services are available, 24 hours
a day, to meet scheduled and unscheduled needs, in a way that promotes maximum dignity and
independence for each resident and involves the resident’s family, neighbors and friends. ... (ALFA,
2000) (Regnier, 2002)
Assisted living serve for both the mentally and physically frail though the two types of people have different
service needs (Regnier, 2002). The mentally frail occupants usually live in a separate unit designed for their
special need (Regnier, 2002). For example, dementia units are often designed around a cluster with 12-20
residents (Regnier, 2002). Aimless wandering can happen often to people with dementia (Regnier, 2002).
The residents with physical problems often have chronic disability of mobility impairment, arthritis or
balance control (Regnier, 2002). Most of them have walking problems. Therefore, most of them rely on four-
prong cane, walking stick or walker and a few are in wheelchairs (Regnier, 2002). Many of them have other
limitations such as diabetes, hypertension, heart disease m hearing loss, incontinence or visual impairment
(Regnier, 2002). When the limitations are not very serious, they can cook, drive. When the limitations get
more disabling, it becomes hard for them to take a shower, dress and eat without any assistance. In assisted
living, the disabilities are easily managing unless they are quite disabling when the occupants need 24-hour
supervised nursing care (Regnier, 2002).
1.4.2 Assisted living overview
Licensed assisted living communities provide three meals every day at specific times, and the management
group set up daily activity schedules. The schedule can be generally projected. The spatial distribution of the
occupants and the occupied status of the rooms can be predicted. Therefore, the real occupied hours of each
space can be found as the analysis period for fsDA and fASE. The Occupied Hour (O) and Hour Percentage
(P) of each function space can be assumed, becoming a reference for the occupancy hierarchy. Occupies
Hour (O) and Hour Percentage (P) are introduces in 1.4.3. The method to calculate the values are introduced
in chapter 3.
In addition, since the elderly usually have a much earlier schedule than the young, the sDA
300,50%
and
ASE
1000,250h
may be quite different from fsDA
300,50%
and fASE
1000,250h
. As fsDA
300,50%
and fASE
1000,250h
describe the lighting performance more accurately, if there is a large gap between the two results, it can
prove that fsDA and fASE can be useful when evaluating lighting performance. Otherwise, since it takes
more work to predict the real operation hour than to use the fixed 10-hour schedule, the two new metrics
may not be necessarily needed.
1.4.3 Terms
“Architectural program” and “function space” are defined. Two new terms, “Occupied Hour (O)” and “Hour
Percentage (P)” were developed to explore the schedules.
• Architectural program
The interior space of assisted living can be analyzed by evaluating each function space defined by
architectural program. Program is usually the first thing an architect needs to figure out. Programming is an
architecture way of asking “what is your wish list?” “Programming” is a term which architects use to describe
how they refine the client’s wish list. Most architects study the program, make the space layout, and define
how the building functions. When it seems the function is defined, the architects come into the design stage
(David Balber Architect n.d.). Studying the architectural program can be an effective method to analyze the
space layout of the case building (Figure 1-10).
• Function space
The term “Function space” is developed to represent the space with a specific function. The “Function space”
is defined according to architectural program. For example, office, activity space and apartment are three
10
function spaces (Figure 1-10).
• Occupied Hour (O)
The term “Occupied Hour (O)” is developed to describe the hours when a function space is occupied by at
least one person. The value can be calculated after occupant spatial distribution simulation, which is
introduced in chapter 3.
• Hour Percentage (P)
“Hour Percentage (P)” describes the total time all the people spend in a function space comparing to other
function spaces. The value of P can be calculated after occupant spatial distribution simulation. The method
is introduced in chapter 3.
11
Figure 1-10: Example of an interior space divided into different function space according to architectural
program (Based on Archdaily 2018)
12
1.4.4 The case building
(1) Description
All the content in 1.4.4 (1) is from Arch Daily 2018 (Arch Daily 2018).
The project is Retirement and Nursing Home Wilder Kaiser, which is located 6351 Scheffau am Wilden
Kaiser, Austria. The site is 78 km from the large city, Innsbruck, where has the nearest weather station. The
project was finished in 2017. The architect is Dü rschinger Architekten, SRAP SeDAk Rissland. The area is
5120 m2. The building consists of two volumes. The larger volume on the east with a green daylight atrium
is the study subject (Figure 1-11). The project includes both assisted living and nursing home, but only the
assisted living part (The block on the right) is considered.
The atrium is the main reason why the project was chosen as the case building. Except for the windows, the
daylight atrium also connects the interior space with the exterior environment. Because much of the surfaces
are connected to the outdoor space, much interior space may be influenced by daylighting, which may help
with daylighting research.
The first floor serves as a public area, with Café -Lounge, Event Hall and Chapel adjoining the main entrance.
The administrative offices, service rooms and kitchen are in the rear area (Figure 1-12,1-13). Wooden
formwork made with untreated larch and broom finished plaster are used as the main construction materials
(Figure 1-14, 1-15). The windows are covered by wooden inlays of profiles larch boards, a reference to local
handcraft traditions.
Figure 1-11: Site Plan (Arch Daily 2018)
The studied volume
13
Figure 1-12: Ground floor plan (Arch Daily 2018)
14
Figure 1-13: Second and third floor plan (Arch Daily 2018)
15
Figure 1-14: Exterior Picture (Arch Daily 2018) Figure 1-15: Interior Picture (Arch Daily 2018)
(2) Function space defined by architectural programming
The case study includes all the space of the main building on the east side except for the function spaces
where are very lowly occupied such as the storage and bathroom. The studied space is divided into different
parts according to architectural programming: office, activity, circulation, apartment, dining, and care (care
bath, nursing station and therapy). The diagrams illustrate the architectural programs by different colors
(Figure 1-16, 1-17).
16
Figure 1-16: Function spaces of Retirement and Nursing Home Wilder Kaiser, divided according to
architectural program, first floor (Based on Archdaily 2018)
17
Figure 1-17: Function spaces of Retirement and Nursing Home Wilder Kaiser, divided according to
architectural program, second floor (Based on Archdaily, 2018)
18
1.5 Software Tools
This section introduces the software tools that will be used, including Rhino, Grasshopper, Honeybee,
Ladybug and Excel. Rhino was used to build up the model of the case building. Grasshopper was to connect
the Rhino model to Ladybug and Honeybee. Ladybug and Honeybee were used to do daylighting simulation.
Excel was for the scheduling study.
(1) Rhino
Rhino is a 3d model building tool. It can model any shape people can imagine accurately (Figure 1-18). More
importantly, the tool is compatible with many other software tools such as Grasshopper.
Figure 1-18: An example of a Rhino model
(2) Grasshopper
Grasshopper is an algorithmic modelling plugin for Rhinoceros that uses a visual programming language
(Alonso 2014). An algorithm is a set of instructions which take as serious of inputs, manipulate them and
give some output (VanderLeest, Nyhoff and Zylstra 2005). Plug-in is a software add-on installed on a
program to improve its capabilities. For instance, if a user wants to watch videos on a website, a plugin is
needed to play it because the browser cannot play it without additional added features (Computer Hope 2018).
Grasshopper provides new methods to expand and control the 3D design and modelling process. It can
automate repetitive processing, create geometry via mathematical functions, make changes to complex
geometries quickly, and generate complex geometries by repeating simple geometries (Reilly 2016).
Parametric design can be achieved by using Grasshopper. The tool makes it possible to reference Rhino
geometry objects. It can bake or produce Grasshopper objects back into Rhino (Figure 1-19, 1-20). Many
plugins for Grasshopper are available for downloading (Alonso 2014) Ladybug and Honeybee, two
environmental plug-ins for Grasshopper were originally created by Mostapha Sadeghipour Roudasri. But
they are open-source and maintained by some people, including Chris Mackey (Paulwintour 2016).
Figure 1-19: An example of Grasshopper script
19
Figure 1-20: An example of Grasshopper script
• Ladybug
Ladybug is a plug-in for Grasshopper. It enables users to import epw weather data into Grasshopper, draw
diagrams such as sun-path, edit the diagrams, do radiation simulation, study shadows, and analyze views
(Sadeghipour n.d.) (Figure 1-21, 1-22).
Figure 1-21: An example of Ladybug script (Mackey 2017)
20
Figure 1-22: An example of Ladybug result (Mackey 2017)
• Honeybee
Honeybee is a plug-in for Grasshopper to connect Grasshopper with simulation engines such as Radiance,
Daysim, Energyplus and OpenStudio for building daylighting, lighting, energy, and comfort simulation
(Sadeghipour n.d.) (Figure 1-23, 1-24).
Figure 1-23: An example of Honeybee script (Roudsari 2015)
21
Figure 1-24: An example of Honeybee result (Roudsari 2015)
(3) Microsoft Excel
Microsoft Excel is a spreadsheet application with the functions of calculation, graphing tools, pivot tables
(Figure 1-25). A macro programming language called Visual Basic (VBA) is also one of the features
(Software-Matters n.d.). The tool can help with data analysis.
Figure 1-25: An example of Excel
1.6 Summary
Daylighting is important to interior space. Two measurements of daylighting, including illuminance and
glare were introduced. Illuminance was to measure the luminous flux striking a surface. Glare was the visual
22
sensation caused by uncontrolled and overly brightness. Daylight autonomy was suggested to describe
daylight. Meteorological Year and epw weather file were useful in daylighting metrics simulation. Two
daylighting metrics, sDA and ASE were suggested by IES LM-83-12. However, the two metrics have limited
scope and lacked the consideration of operation schedule. LEED borrowed the idea of sDA and ASE for its
daylighting rating system. There are also another two options for LEED daylighting evaluation while none
of the three options think about operation schedules. Therefore, fsDA and fASE, two new metrics based on
sDA and ASE while with flexible analysis period were suggested to describe the daylighting performance
of interior space for varies building types such as assisted living.
Assisted living could be a good example to study daylighting and schedules because of their regular and
early schedule. Terms, including architectural program, function space, Occupied Hour (O) and Hour
Percentage (P) were introduced for assisted living theoretical project. The interior space was divided into
different function spaces according to architectural program so the lighting performance of each function
space can be analyzed individually. The Occupied Hour (P) and Hour Percentage (P) for each function space
are calculated after occupant spatial distribution simulation. Software tools including Rhino, Grasshopper
and the plugins for Grasshopper including Ladybug and Honeybee, and Microsoft Excel were described.
23
2. BACKGROUND AND LITERATURE VIEW
This section covers assisted living, daylight evaluation metrics, spatial Daylight Autonomy (sDA) and
Annual Sunlight Exposure (ASE), LEED BD+C daylight evaluation and the method to capture the behavior
patterns of older adults.
2.1 Assisted living
This section includes the importance of daylight to assisted living, the residents in assisted living and the
schedules of assisted living.
2.1.1 The importance of daylight to assisted living
(This paragraph is from IES Lighting for the Elderly and Partially Sighted Committee 2016) (IES Lighting
for the Elderly and Partially Sighted Committee 2016). Daylight is beneficial to senior housing. First,
daylight helps to improve visibility. According to the 2013 version of ASHRAE 90.1, people with visual
impairment need better lighting condition. Second, daylight also provides high color rendering. Third,
daylight provides exterior views for residents. Exterior views connect occupants with nature so the isolation
that older adults usually experience is reduced. The view also connects residents with the changing sun
patterns and weather and season, as well as the coming and going pedestrians. This reinforcement the timing
and orientation is essential for the occupants. Exterior views can also lengthen the eye’s focal distance so
eye muscles can relax. In addition, pain and stress can be reduced by exterior views. Also, the adaption
discomfort via transition areas can be reduced by daylight. Sometimes exterior spaces are much brighter than
the interior space. Therefore, the visual acuity decreases when their eyes adapt from the bright exterior to
the darker interior. Older eyes take much longer to adapt the changes in light levels. Thus, graduated levels
of daylight in transition spaces can minimize the adaption discomfort. In addition, energy can be saved by
utilizing daylight as less electric light is used.
(This paragraph is from Daylighting, Architecture and Health 2008) (Boubekri 2008). The access to outdoors
and walking opportunities outdoors are also important. Vitamin D is essential to the health of human as it
regulates the absorption of nutrients and it helps to keep phosphorus and serum calcium concentrations in
regular range. Phosphorus and calcium are essential for the growth and development of bone structure.
Because Vitamin D is formed in the skin by the action of ultraviolet rays from the sun, vitamin D is called
the “sunshine vitamin”. Ultraviolet light is divided into three wavelength spectra: UV-A, UV-B, and UV-C
according to the wavelength. UV-B (290nm to 315 nm) is responsible for photosynthesis and stimulates the
skin to produce vitamin D. UV-B can lead to skin burning and aging as well. Glass filters out approximately
95% of the UV-B radiation in the atmosphere. Therefore, the people in the buildings receive nine to ten time
less UV-B radiation than if they are outdoors. Thus, the exposure to daylighting without the glass filter is
necessary for occupants.
(This paragraph is from IES Lighting for the Elderly and Partially Sighted Committee 2016) (IES Lighting
for the Elderly and Partially Sighted Committee 2016). Daylight can also help to maintain circadian rhythms.
People follow a 24-hour rhythm daily, which is known as circadian rhythm. Light entering the eye not only
influence visual system, but also affects the circadian (body) clock. Because of disrupted circadian rhythm,
older adults usually have sleep disturbances and seasonal depression or seasonal affective disorders. Being
exposed to daylight can help to mitigate the problems. Short-term exposure to sunlight three times per week
can lead to sufficient synthesis of vitamin D3, which enable the body to efficiently utilize calcium and
enhance muscle function, as well as to prevent cancer.
2.1.2 Residents in assisted living
All the content in 2.1. 2 is from Design for Assisted Living – Guidelines for Housing the Physically and
Mentally Frail, 2002 (Regnier, Design for Assisted Living- Guidelines for Housing the Physically and
Mentally Frail 2002).
Who the residents in assisted living are, their service requirement and their daily activities are introduced in
2. 2. The information about the occupants can be a reference for assuming the schedules for the residents.
• Who they are?
Knowing who the residents are in assisted living can help to explore their daily activities in the building. As
24
woman usually outlive men in the US, in 2000, 75-78% people in assisted living are female. More women
end up in assisted living since many of them are widows so no spouse can take care for them. Woman are
usually caregivers because they often outlive their spouse. The average age of females in assisted living is
84.3 and that of males is 82.5. Around two-thirds of the occupants had an income of less than $25,000 per
year while the average annual fee was about $24,433. Some of the residents rely on financial help from
family members. Married couples are not very common in assisted living. On average, there are 1.8 couples
in each assisted living while the average size for an assisted living is 52 units. Married couples usually require
less care because they can take care of each other. More than half of the occupants in assisted living need
care due to their mental and/or physical frailty.
• Service requirement
Approximately, only 7% to 10% can living independently with no services. The residents need services of
different levels. Around two-thirds occupants have complex drug regiments so they need professional care.
Nursing personnel are essential in assisted living to help identify drug interactions and side effects. Apart
from this, bathing assistance is another service that two-third of the residents need. Around 45% occupants
have cognitive impairment in assisted living and the values reach 54 in nursing homes. About 40% residents
need dressing assistance and about 30% need toileting assistance. About 10% residents have special
problems such as difficulty with eating and chronic diseases such as Parkinson’s disease and diabetes, as
well as wheelchair dependence.
• Daily activities
The table illustrates 14 different activities of the residents on assisted living (Table 2-1). The most common
activities can be carried out in common areas or in the residents; dwelling unit. Talking or conversing, the
social activity is the most popular, with around three-fourths of the population engaging in it. Social activity
can contribute to friendship is a necessary factor that can lead to well-being. If a building supports more
social activities, the building is more friendly and the interaction between residents are enhanced. Watching
TV, listening to music, as well as reading and writing are the core activities. The middle popular activities
include more group activities such as card playing, crafts and religious services, as well as activities which
require much physical movement like trips and shopping, walking and wheeling outdoors and exercise sports.
The less common activities need an activity room, an indoor room for exercise, and pathways surrounding
the site. The activity can be a reference for resident schedule assumption. However, the data are from 20
years ago. The activities may have changed now. Therefore, the real schedules given by the managers in
assisted living may be more useful.
25
Table 2-1: Assisted living: activities available for residents (Regnier, 2002)
2.1.3 Assisted living schedules
There is often no standard schedule for an assisted living community. The schedule depends on the
management group (Regnier, Senior living schedule 2019). Three sample schedules for an assisted living
are given. One is from Victor Regnier, an expert in assisted living. Two of them are schedules of two real
assisted livings: Belmont Village, Hollywood Hills and Sunrise of Beverly Hills.
(1) Sample schedule 1 from Victor Regnier
Victor Regnier, an assisted living expert gives a sample schedule of the buildings (Table 2-2). But he did not
give a schedule for the staff shift. The staff schedule was borrowed from Belmont Village, Hollywood Hills
as an example for the case study (Table 2-3) (Manager of Belmont Village 2019). The staff can have 3 shifts
26
a day to ensure the residents are cared for 24 hours a day.
Table 2-2: Residents schedule 1 (Regnier, Senior living schedule 2019)
Time Activity
8:00-10:00 Breakfast
10:00-12:00 Shower/ Stay in bedroom or activity
room/Take snacks
12:00-14:00 Lunch
13:30-17:00 Sleep/Stay in bedroom or activity
room/Take snacks
17:00-19:00 Dinner
Table 2-3: Staff schedule 1 (Manager of Belmont Village 2019)
Time Number of staffs
8:00-17:00 9
17:00-1:00 5
1:00-8:00 4
(2) Sample schedule 2 from Belmont Village, Hollywood Hills
All the content in 2.1.3 (2) is from the conversation with manager of Belmont Village, 2019 (Manager of
Belmont Village 2019)
Belmont Village, Hollywood Hills is located in Los Angeles, USA. It is a senior living community mainly
focus on the people older than 80. Some of the residents have memory and they can take care of themselves
while others can have dementia. Special care is provided for the residents who really need that. The staff
have 3 shifts to make sure the residents have care 24 hours a day. The staff, except for the manager, do not
stay in their office all day long. They mainly walk around and help the elderly. Although they have several
offices, the space is usually unoccupied. The ratio of the residents to the staff is 3:1 to 5:1. The residents do
not stay in their bedroom all day long as well. They spend most of the time in the activity room with others
and do some training. There is no fixed schedule for every resident as they can choose what to do by
themselves. A general schedule is given by the manager (Table 2-4, 2-5).
Table 2-4: Residents schedule 2 (Manager of Belmont Village 2019)
Time Activity
7:00-9:00 Breakfast
9:30-11:00 Chair exercise/other activities
11:00-13:00 Lunch
13:00-14:00 Nap/Exercise
14:00-16:30 Activity
16:30-19:00 Dinner
From the values given by the manager, one can theoretically create resident and staff schedules. By using
the theoretically schedules, occupant spatial distribution can be simulated if real values are not available.
Thus, Occupied Hour (O) and Hour Percentage (P) were able to be calculated. Then the occupied hierarchy
of different function spaces can be discussed based on O and P. Assume there are 54 residents in the assisted
living, according to the staff-residents ratio, the number of staffs in the assisted living would be the following
(Table 2-5).
Table 2-5: Staff schedule 2 (Manager of Belmont Village 2019).
Time Number of staffs
8:00-17:00 9
17:00-1:00 5
1:00-8:00 4
27
(3) Sample schedule 3 from Sunrise of Beverly Hills
All the content in 2.1.3(3) is from the conversation with manager of Sunrise of Beverly Hills, 2019 (Manager
of Sunrise of Beverly Hills 2019).
Sunrise of Beverly Hills is a senior living located in Los Angeles, USA. The schedule of the residents is
given (Table 2-6). The staff have 8-hour shift to make sure the residents are cared 24 hours a day. There are
8 staff during the day and 5 at night. 2 of the working staff are in the office and the rest spend half of the
time in the office and half of the time walk around. The ratio of staff to residents in assisted living is 1:4.
Assume there are 54 residents in the assisted living, according to the staff-residents ratio, the number of
staffs in the assisted living would be the following (Table 2-7).
Table 2-6: Residents schedule 3 (Manager of Sunrise of Beverly Hills 2019)
Time Activity
6:30-8:00 Breakfast
9:30-11:00 Activity
12:00-13:30 Lunch
13:00-14:00 Nap/Exercise
14:00-16:30 Activity
17:00-18:30 Dinner
Table 2-7: Staff schedule 3 (Manager of Sunrise of Beverly Hills 2019)
Time Number of staffs
9:00-18:00 7
18:00-2:00 4
2:00-9:00 4
2.2 Daylight evaluation metrics
Six daylight metrics, including Daylight Autonomy (DA), continuous Daylight Autonomy (cDA), Useful
Daylight Illuminance (UDI), Daylight Glare Probability (DGP), spatial Daylight Autonomy (sDA) and
Annual Sunlight Exposure (ASE), were analyzed first to find out the most useful metrics in daylight
performance analysis.
(1) Daylight Autonomy (DA)
Daylight Autonomy is an annual measurement of the percentage of occupied time that the daylight levels
exceed a specified illuminance (Max 2012). The three inputs for DA are the target illuminance, climate-
based daylighting levels over the entire year, and the real hours of operation of the space (Max 2012). There
are several good points about DA:
• DA considers the real operation period and real weather condition of the site (Max 2012).
• DA provides an intuitive overview of how well daylight penetrating in to the space (Max 2012).
• DA can help to determine which fixtures can benefit from automatic daylight harvesting (Max
2012).
• DA can be visualized so the lighting condition of each point can be identified clearly (Figure 2-1).
28
Figure 2-1: An example of visualized DA
300
However, because DA only evaluates the daylighting performance of a single point, it is hard to describe
the daylighting performance of a space by a single number. Numerous DA values are needed to illustrate the
daylighting condition of a tested space.
(2) Continuous Daylight Autonomy (cDA)
All the content of 2.2(2) is from Max, 2012 (Max 2012).
Continuous Daylight Autonomy (cDA) is similar to DA, but for the point with an illuminance is less than
the target value, partial credit is given. For instance, if the target illuminance is 300 lux, but the point has
150 lux for 100% of the operating time. The cDA for the point would be 0.5. There are several advantages
to using cDA:
• cDA is aimed to be used as a compliance metric more than a design tool.
• The space with enough daylight while needs to reduce electric lighting loads does well on cDA.
• Problem times and climate scenarios are not very apparent.
However, cDA is not a suitable metric for comparing two potential design, since if a point gets 0.1, it is not
clear whether that is because there were 300 lux for 100% of the time, or 3000 lux for 10% of the time, or
what proportion in between.
(3) Useful Daylight Illuminance (UDI) (Max 2012)
All the content of 2.2(3) is from Max, 2012 (Max 2012).
It is another variation on Daylight Autonomy. The metric is to penalize for direct sunlight which penetrate
the space as it can cause glare. UDI is a set of three numbers for every point in the space, the fraction of time
that a point is lower than a minimum threshold, between the minimum and maximum threshold, and above
a maximum value, which can lead to glare or thermal discomfort. Normally, 10 fc is used as the lower bond
and 250 is the upper bond. There are several advantages to using UDI:
• UDI not only owns most of the advantages of DA, but also includes additional dimension for
glare and thermal discomfort.
• It is helpful for comparing the performance of two designs versions.
However, as three data are generated for each point, it can be hard to evaluate the performance at a glance.
(4) Daylight Glare Probability (DGP)
All the content of 2.2(4) is from Max, 2012 (Max 2012).
It can be difficult to quantify glare as it depends on the observer (age and personal preference) and the
29
position relative to sources and the specific task being performed. Apart from that, glare can happen in
numerous ways such as an extreme luminance ratio between windows and adjacent surfaces, or reflections
from the sun. The reference method to assess glare is to generate an HDR image for the viewpoint for every
daylight hour of over the year. Then then luminance ratios and adjacencies are compared to find out a
possibility that an average observer at the point would find it uncomfortable. An advance of DGP is that
projects possible glare by evaluating luminance ratios, rather than using a proxy metric such as ASE .
There are several disadvantages to using DGP:
• The calculation is time-consuming.
• The results are only meaningful for the tested location and orientation.
• It can be difficult to tell what the DGP number means for the specific qualitative problem.
(5) Spatial Daylight Autonomy (sDA)
Spatial Daylight Autonomy (sDA) is a metric to indicate the annual sufficiency of ambient daylight levels
in interior space. sDA is defined as the percentage of a studied area that meets a minimum daylight
illuminance threshold for a specified fraction of the operation hours per year. The illuminance level and time
fraction are expressed as subscripts, as in sDA
300,50%
. The sDA value is described as a percentage of area
(IES Daylight Metrics Committee 2012). There are several advantages to using SDA:
• Unlike DA, which gives a series of data points for every test point of the space, sDA is a single
number.
• sDA is experimentally verified to predict occupants’ satisfaction (Max 2012).
However, sDA cannot reveal glare problems (Max 2012).
(6) Annual Sunlight Exposure (ASE)
Annual Sunlight Exposure (ASE) is a metric that illustrates the possibility for visual discomfort in interior
space. ASE is defined as the percentage of an analysis area that surpasses a specified direct sunlight
illuminance level more than a specified number of hours per year. The subscripts represent the illuminance
level and hours. As in ASE
1000, 250h
. The ASE value is described as a percentage of area (IES Daylight Metrics
Committee 2012).
There are several advantages to using ASE:
• It is fast to calculate as no radiosity solutions or luminance maps for an observer point is needed
(Max 2012)..
• It can be a helpful design tool because it suggests where the problem areas might be (Max 2012)..
• It covers possible thermal discomfort issues (Max 2012).
There are several disadvantages to using cDA:
• ASE is not a glare metric, strictly speaking (Max 2012)..
• It is a proxy which help to predict glare (Max 2012)..
• It does not calculate glare due to specular reflections or veiling glare from high luminance ratios
(Max 2012).
2.3 Spatial Daylight Autonomy (sDA) and Annual Sunlight Exposure (ASE)
All the content of 2.3 is from IES Daylight Metrics Committee 2012 (IES Daylight Metrics Committee
2012).
This section introduces spatial Daylight Autonomy (sDA) and Annual Sunlight Exposure (ASE). IES LM-
83-12 gives an approved method on the evaluation of the lighting performance of an interior space. This
document suggests two metrics, one for daylighting, Spatial Daylight Autonomy (sDA) and one for glare,
Annual Sunlight Exposure (ASE), to describe the lighting performance of “regularly occupied space” of a
building.
30
2.3.1 Spatial Daylight Autonomy (sDA)
IES LM-83-1 defined Spatial Daylight Autonomy (sDA). The definition, standard threshold for analysis,
analysis period, analysis area, analysis points, interior surface reflectance, modelling parameters and
suggested sDA performance criteria were introduced.
(1) The Definition of sDA
Spatial Daylight Autonomy (sDA) is a metric to indicate the annual sufficiency of ambient daylight levels
in interior space. sDA is defined as the percentage of a studied area that meets a minimum daylight
illuminance threshold for a specified fraction of the operation hours per year. The illuminance level and time
fraction are expressed as subscripts, as in sDA
300,50%
. The sDA value is described as a percentage of area.
SDA
300,50%
means the percentage of the analysis points meet or exceeds 300 lux for at least 50% of the
analyzed hour over a year . SDA is intended be applied to common workplace environments such as open
offices, multi-purpose rooms, service arear in libraries and lobbies, and so are most applicable to areas with
similar visual tasks. The ideal analysis area is a coherent “space” defined by opaque walls and access to
daylighting through at least one wall or ceiling surface. The metric can also be applied to the “regularly
occupied” floor area of a building, or some part of a building such as one floor plate.
(2) Standard Threshold for analysis
sDA
300,50%
is the suggested metric for daylight sufficiency analysis.
• 300lux
300 lux is the standard illuminance threshold for analysis on a horizontal workplan, which should be 2.5 feet
above the floor. 300 lux is chosen as the threshold based on many other recommendations and standards. For
example, IES recommendation for tack level illuminance for many spaces is 300 lux. Nevertheless, 300 lux
is not a fixed goal for daylight illumination as the daylight illumination can vary throughout the day and
seasons. Therefore, any other indicator value (such as 100lux, 200 lux, 1000 lux) may be used to indicate
the daylight performance over time. The 300-lux indicator value is used to represent a slightly better
correlation to occupant preference for daylight sufficiency.
Different space such as circulation pathways and storage may have different illuminance goal. However, the
Committee does not have a research basis to provide a suggestion for the area. In addition, there is no upper
limits to annual daylight autonomy values to describe the dissatisfaction of the occupants. Indeed, it is found
that the higher the annual Daylight Autonomy values in the analyzed space, the less likely occupants would
report visual discomfort, the more satisfaction could be reported.
• 50%
The 50% time indicator valued was determined based on regression analysis of some alternatives (e.g. 40%,
75%, 95%). Although all the indicator value is possible to describe the annual daylight performance of the
space, the 50% value expresses more consistent correlation to occupant preference for daylight sufficiency
in the three space types: offices, classrooms, library/lobbies based on researches.
(3) Simulation method
The period for analysis is fixed, from 8am to 6pm local clock time. The result is 10 hours per day (3650
hours for a complete annual analysis). Although the operating schedules can vary in different buildings, the
10-hour day was regarded as the standard analysis period because by using the fixed time, different building
designs can be compared. sDA can be used to describe the “asset value”. Other time periods of analysis can
be used for study only.
The daylight conditions are rest on typical meteorological year (TMY) data. Typical Meteorological Year
(TMY) formatted hourly weather data by hourly average, with data centered on the half-hour. For instance,
hour 9 on a day is an average of 8am to 9am. To modelling sDA, the time period uses the solar position on
8:30 am.
To analyze whole buildings, the area to be analyzed should be “regularly occupied work areas,” which means
spaces with very low occupancy such as storage rooms and cop rooms can be excluded from the analyzed
area. Within the analysis areas, the points to be analyzed should be continuous, specified no larger than2 feet
31
on center. The points should be from 1 to 2 feet offset from the walls, with as height of 2.5 feet above finished
floor. A two-foot grid is the maximum scale for the tiling of the sDA analysis (Figure 2-1).
Figure 2-1: Analysis points of sDA
(4) Interior Surface Reflectance
The actual surface reflectance values should be used for analysis. However, if the actual reflectance is
unknown, the following default reflectance values can be used.
20% floor
50% walls
70% ceiling
50% furniture
(5) Modelling Parameters
The modelling parameters can be complex and they vary based on the analysis program algorithms and the
model’s complexity. The following parameters are sample values, which are used in this research (Table 2-
8).
Table 2-8: Sample parameters used in a research
Ambient bounces 3
Ambient division 512
Direct threshold 0
(6) Suggested sDA Performance Criteria
• Preferred Daylight Sufficiency using sDA
300,50%
.
sDA
300,50%
must be no lower than 75%.
• Nominally Accepted Daylight Sufficiency using sDA
300,50%
.
sDA
300,50%
must be no lower than 55%.
2.3.2 Annual Sunlight Exposure (ASE)
IES LM-83-1 defined Annual Sunlight Exposure (ASE). The definition, standard threshold for analysis,
analysis period, analysis area, analysis points and suggested sDA performance criteria are introduced.
(1) Definition
Annual Sunlight Exposure (ASE) is a metric that illustrates the possibility for visual discomfort in interior
32
space. ASE is defined as the percentage of an analysis area that surpasses a specified direct sunlight
illuminance level more than a specified number of hours per year. The subscripts represent the illuminance
level and hours. As in ASE
1000, 250h
. The ASE value is described as a percentage of area. ASE
1000,250h
represent
the percentage of the analyzed space where are exposed to more than 1000 lux of direct sunlight for at least
250 hours over a year. ASE is intended be applied to common workplace environments such as open offices,
multi-purpose rooms, service arear in libraries and lobbies, and so are most applicable to areas with similar
visual tasks. The ideal analysis area is a coherent “space” defined by opaque walls and access to daylighting
through at least one wall or ceiling surface. The metric can also be applied to the “regularly occupied” floor
area of a building, or some part of a building such as one floor plate.
(2) Standard Threshold for analysis
ASE
1000,250h
is the suggested metric for daylight sufficiency analysis.
ASE
1000,250h
means more than 1000 lux of direct sunlight for more than 250 hours per year before the shade
are used is the threshold.
• 1000 lux:
1000 lux was selected as the indicator threshold.
• 250 hours:
250 hours is selected as the indicator value. Although values such as 100 hours, 300 hours can be used to
compare spaces, the higher the threshold, the less potential that the analysis will report any area, and the
lower the threshold, the more unstable the result will be.
(3) Simulation method
The analysis period is the same as that of sDA. The period for analysis is fixed, from 8am to 6pm local clock
time. The result is 10 hours per day (3650 hours for a complete annual analysis).
The area to be analyzed is the same as that of sDA. To analyze whole buildings, the area to be analyzed
should be “regularly occupied work areas,” which means spaces with very low occupancy such as storage
rooms and cop rooms can be excluded from the analyzed area.
The analysis points regulation is the same as that of sDA. Within the analysis areas, the points to be analyzed
should be continuous, specified no larger than2 feet on center. The points should be from 1 to 2 feet offset
from the walls, with as height of 2.5 feet above finished floor. A two-foot grid is the maximum scale for the
tiling of the ASE analysis (IES Daylight Metrics Committee 2012) (Figure 2-2).
Figure 2-2: Analysis points of ASE
33
(4) Suggested ASE Performance Criteria
Space with more than 10% ASE
1000,250h
is regarded to have unsatisfactory visual comfort.
Space with ASE
1000,250h
less than 7% is neutral (nominally acceptable).
The space with less than 3% ASE
1000,250h
can be clearly acceptable.
The values are considered across single spaces while it is not for larger area, or entire building.
2.4 LEED BD+C daylight evaluation
This section introduces the three daylight evaluation options of LEED BD+C. In addition to IES LM-83-1,
other rating systems on daylight exist. LEED (Leadership in Energy and Environmental Design), is the most
popular green building rating system in the world. LEED offers framework to make economic, efficient and
healthy green buildings (USGBC n.d.)
All the flowing content of 2.4 is from USGBC, 2019 (USGBC 2019).
LEED v4 for Building Design and Construction gives 3 options to evaluation to find out if any of them
considers operating scheduling. 1-3 points can be given to daylight, aiming to connect building occupants
with the outdoor environment, strengthening circadian rhythms and to reduce the consumption of electrical
lighting by using daylight. Three options are provided by the standard. Only one of them needs to be selected
(USGBC 2019). However, none of the three options takes the operation schedule of the building into
consideration.
2.4.1 Option 1
Option 1 is about spatial Daylight Autonomy
300,50%
(sDA
300,50%
) and Annual Sunlight Exposure
100,250h
(ASE
1000,250h
) (normally 2-3 points, 1-2 points for healthcare). LEED borrows the suggested sDA and ASE
from IES LM-83-12 (Table 2-9).
• SDA requirement:
Table 2-9: Spatial Daylight Autonomy
300,50%
(sDA
300,50%
) requirement under option 1
New Construction, Core and Shell, Retail, Schools,
Warehouses, Data Centers and Distribution Centers,
Hospitality
Healthcare
sDA (for regularly
occupied floor area)
Points sDA (for perimeter floor
area)
Points
55% 2 75% 1
75% 3 90% 2
• ASE requirement:
ASE
1000,250h
should be no more than 10%
2.4.2 Option 2
Option 2 requires calculating illuminance at two time points at the equinox. Simulate the illuminance of
regularly occupied floor area will be between 300 lux to 3000 lux for 9 am and 3 pm, both on a clear-sky
day at the equinox. Typical meteorological year data is used. The weather file should be that of the nearest
weather station. Choose one day within 15 days of September 21 and another day within 15 days of March
21 to represent the clearest sky condition. Blinds, shades, movable furniture and partitions are excluded in
the model. The illuminance requirement is given below (Table 2-10).
Table 2-10: Illuminance requirement
New Construction, Core and Shell, Retail, Schools,
Warehouses, Data Centers and Distribution Centers,
Hospitality
Healthcare
Percentage of regularly
occupied floor area
Points Percentage of perimeter
floor area
Points
75% 1 75% 1
90% 2 90% 2
34
2.4.3 Option 3
Similar to option 2, option 3 is also about illuminance simulation at two time points. Illuminance levels
between 300 lux and 3000 lux must be realized for the floor area illustrated (Table 2-11). However, option
3 requires fixtures, furniture and equipment in place. Take first measurement in a regularly occupied month,
and then take the second measurement of another required time (Table 2-12).
Table 2-11: Illuminance standard under option 3
New Construction, Core and Shell, Retail, Schools,
Warehouses, Data Centers and Distribution Centers,
Hospitality
Healthcare
Percentage of regularly
occupied floor area
Points Percentage of perimeter
floor area
Points
75% 2 75% 1
90% 3 90% 2
Table 2-12: The time to take measurement under option 3
If the first measurement is taken in… The second measurement should be taken in…
January May-September
February June-October
March June-July, November-December
April August-December
May September-January
June October-February
July November-March
August December-April
September December-January, May-June
October February-June
November March-July
December April-August
2.5 Method to capture the behavior patterns of older adults
All the content from 2.6 is from Virone et al. 2008 (Virone, et al. 2008).
This section introduces a method to capture the behavior patterns of older adults. The system to detect people
behavior patterns used in a research gives a method to find out the occupied statues of interior spaces. Sensors
were used to find out the precise occupied status of each space. In future work, the method may help to find
the accurate occupied status of each function space and the result can be more convincing. The researchers
tested whether a software program called Software for Automatic Measurement of Circadian Activity
Deviation (SAMCAD) can figure out behavior patterns and differentiate between normal and abnormal
behaviors of the older adults in assisted living. People usually follow an approximate 24-hour daily rhythm
called circadian rhythm. A normal model of Circadian activity rhythm (CAR) was built. The motion of
people was captured to find out the deviation to the normal CAR.
One method to collect the motion activity data was described. The motion activity data were collected by an
in-home monitoring system (IMS), which consists of wireless passive infrared motion sensors installed in
every room, as well as a stovetop temperature sensor and a passive bed-based vital sign monitor.
Nevertheless, only the motion sensor data was used to build the CAR model. The data manager collected
data from the separate sensor modules. When the residents moved from on space to another, their transitions
trigger the motion sensors in both rooms. The difference in the time results in the period of time the resident
spent in the room. How active the residents were during their presence period in each room was calculated
by the number of motion sensors firing event per hour.
By analyzing the average behavior patterns of all the residents, a general occupied status of each room can
be predictable. The data manager gathered the data from IMS. Then the data logs were sent to a secure central
data analysis server. After preprocessing the raw activity data, it was transformed into a format suitable for
the SAMCAD software which extracted the CAR to decide whether the activity monitoring enables people
to figure out behavior patterns, and differentiate between normal and abnormal behaviors (Figure 2-3).
35
However, the method is only valid for individuals. Testing for multiple people cannot be achieved by the
method. This system may be used for further study to explore the accurate occupied status of each room.
Figure 2-3: Connection between IMS and SAMCAD (Virone et al. 2008)
2.6 Summary
Daylight is important to assisted living. The residents in assisted living usually have mentally and/or
physically frail. The identity, the service requirement, and the daily activities of the residents in assisted
living were introduced. Three assisted living schedules were provided. Six daylight metrics, including DA,
cDA, UDI, DGP, sDA and ASE were evaluated. It was found that, sDA and ASE, were very important. The
method to simulate sDA and ASE were provided by IES-LM-83-12. However, operation schedule was not
considered by the document. LEED provides 3 options for daylight evaluation. One of the options was to
assess sDA and ASE, which borrowed the advice from IES-LM-83-12. Nevertheless, none of the three
options took schedule into consideration. One method to obtain occupants’ behavior patterns was illustrated.
Spatial Daylight Autonomy (sDA) is a metric to indicate the annual sufficiency of ambient daylight levels
in interior space. sDA is defined as the percentage of a studied area that meets a minimum daylight
illuminance threshold for a specified fraction of the operation hours per year. sDA
300,50%
is the suggested
metric for daylight sufficiency analysis. To realize nominally accepted daylight sufficiency, sDA
300,50%
must
be no lower than 55%. To achieve preferred daylight sufficiency, sDA
300,50%
must be no lower than 75%
(Table 2-13). Annual Sunlight Exposure (ASE) is a metric that expresses the possibility for visual discomfort
(glare) in interior space. ASE is defined as the percentage of an analyzed area that surpasses a specified
direct sunlight illuminance level more than a specified number of hours per year. ASE
1000,250h
is the suggested
metric for visual discomfort analysis. Space with more than 10% ASE
1000,250h
is regarded to have
unsatisfactory visual comfort. Space with ASE
1000,250h
less than 7% is nominally acceptable. When the value
is less than 3% ASE
1000,250h
, the space can be clearly acceptable (Table 2-14). The analysis period for sDA
and ASE are fixed, from 8am to 6pm local clock time. The result is 10 hours per day (3650 hours for a
complete annual analysis). However, the scope for sDA and ASE are limited to common working space. The
two metrics are useful in daylighting performance description so they may be able to describe the lighting
performance of different building types such as assisted living. However, when it comes to assisted living,
the building type with regular and early operation schedules, the fixed analysis period becomes a problem.
The fixed schedule may not reflect the daylighting performance fully. Flexible analysis period based on real
36
operation schedules may be useful in daylighting performance description.
Table 2-13: sDA
300,50%
Evaluation Criteria
sDA300,50%
Nominally accepted >=55%
Preferred >=75%
Table 2-14: ASE
1000,250h
Evaluation Criteria
ASE1000,250h
Unsatisfactory visual comfort >10%
Nominally acceptable <7%
Clearly acceptable <3%
37
3. METHODOLOGY
This chapter is about methodology diagram, preparation, exploring the influence of sDA and ASE, and
exploring of occupied hierarchy of the function spaces.
3.1 Methodology diagram
There are three main parts of the methodology: preparation, exploring the influence of analysis period on
sDA and ASE and exploring the approach to define occupied hierarchy of the function spaces (Figure 3-1).
Figure 3-1: Methodology diagram
• Preparation:
The preparation includes building the Rhino model of the case building, dividing the case building into
different function spaces according to architectural program and doing research on operation schedules of
assisted living.
• Part 1:
Step 1:
To explore the influence of analysis period on sDA and ASE, first, the sDA
300,50%
and ASE
1000,250h
of a case
building were calculated according to the method suggested by IES LM-83-12.
38
Step 2;
After that, fsDA
300,50%
and fASE
1000,250h
were calculated based on the real operating hours rather than 8am to
6 pm (the fixed 10 hours).
Discussion:
Then, the new calculation results were compared with that calculated based on the 10-hour. fsDA and fASE,
who have a flexible analysis period, can describe the lighting performance more fully than sDA and ASE.
Thus, if there is a large difference between the simulation result of the new metrics and that of the old metrics,
it reveals that the two new metrics can be useful in lighting performance evaluation of an assisted living. If
the difference is slight, the two new metrics may be unnecessary in lighting performance evaluation of an
assisted living as there is not much difference between the result of the existing and the new metrics.
• Part 2:
It will be good if all the spaces in a building have good lighting performance. But there are usually some
limitations so not all spaces can achieve good lighting performance. Therefore, priority should be given to
more critical spaces. The spaces where people spend more time in can be more important and better lighting
condition is needed for the spaces (Regnier, Senior living schedule 2019). There can be different
requirements on sDA and ASE for spaces with different importance. Therefore, deciding which function
space is more important can be necessary.
Step 1:
To explore the occupied hierarchy of the space, two influential factors, Occupied Hour (O) and Hour
Percentage (P) of the function spaces were calculated by simulating occupant spatial distribution under the
three operation schedules.
Step 2:
With the same schedule, Occupied Hour (O) does not change much with different occupant spatial
distribution while Hour Percentage (P) may change significantly. It is necessary to find out how sensitive
Hour Percentage (P) to occupant spatial distribution is. Therefore, a sensitivity test including two parts was
carried out. Schedule 3 was selected for the test.
Test 1:
The first part was to test how Hour Percentage (P) change with people moving between two highly populated
space.
Test 2:
Then the condition where people moving from a highly populated space to a lowly populated space was
tested.
Discussion:
The Occupied Hour (O) and Hour Percentage (P) of the three occupant spatial distributions were analyzed
to find out if they vary with different schedules. Then it was discussed whether the two factors could decide
occupied hierarchy. Next, the new Hour Percentage (P) of the sensitivity test were compared with the initial
value of Hour Percentage (P) under schedule 3. If Hour Percentage (P) was sensitive to occupant spatial
distribution, the metric may not be able to be used as a factor to decide the occupied hierarchy of the function
spaces. Otherwise, it may be used as a factor to decide the occupied hierarchy of the function spaces. Finally,
a sample occupied hierarchy was suggested and requirements on sDA
300,50%
/fsDA
300,50%
and ASE
1000,250h
/fASE
1000,250h
of function spaces with different occupied levels were called for.
3.2 Preparation
The preparation method includes building the case model, dividing the case building into different function
spaces and doing research about the schedule (Figure 3-2). A Rhino model of the case building was built
(Figure 3-3, 3-4). The interior space was divided into 6 parts according to the functions: office, activity,
circulation, apartment, care and dining (figure 3-5, 3-6, 3-7). The space with very low occupied time such as
bathroom and storage were excluded from the study. Three operation schedules for assisted living were
39
found by visiting assisted living communities, as well as by talking with experts and staff of assisted livings.
Figure 3-2: Preparation methodology
Figure 3-3: Rhino model of the case building
Figure 3-4: Rhino model of the case building
40
Figure 3-5: the 1st floor divided into different function spaces
41
Figure 3-6: the 2nd and 3rd floor divided into different function spaces
Figure 3-7: Legend of function spaces
3.3 Explore the influence of analysis period on sDA and ASE
This section is about calculating sDA by Honeybee, calculating ASE by Ladybug, calculating fsDA and
fASE, comparing fsDA with sDA, and comparing fASE with ASE, as well as suggesting whether fsDA and
fASE are useful. The influence of analysis period on sDA and ASE was explored by comparing the fsDA
300,50%
with sDA
300,50%
, and comparing fASE
1000,250h
with ASE
1000,250h
(Figure 3-8). The analysis period is decided
according to the real operation hours (Table 3-1, 3-2, 3-3).
42
Figure 3-8: The methodology of exploring the influence of analysis period on sDA and ASE
Table 3-1: The analysis period under schedule 1
Space Analysis period
Dining 8:00-10:00
12:00-14:00
17:00-19:00
Activity 8:00-19:00
Apartment 8:00-19:00
Office 8:00-9:00
14:00-15:00
Care 14:00-16:00
Circulation 8:00-19:00
Table 3-2: The analysis period under schedule 2
Space Analysis period
Dining 7:00-9:00
12:00-14:00
17:00-19:00
Activity 7:00-19:00
Apartment 7:00-19:00
Office 7:00-9:00
14:00-15:00
Care 14:00-16:00
Circulation 7:00-19:00
Table 3-3: The analysis period under schedule 3
Space Analysis period
Dining 6:30-8:00
12:00-13:30
17:00-18:30
Activity 6:30-18:30
Apartment 6:30-18:30
Office 6:30-18:30
Care 13:30-15:30
43
Circulation 6:30-18:30
3.3.1 Calculate sDA by Honeybee
To calculate sDA, first, each of the surface of the Rhino model was converted into Honeybee surface. Six
tests were carried out for different function spaces: the whole building, office, activity, circulation, apartment,
and care (care bath, nursing station and therapy). The epw weather file of Innsbruck was imported to
Honeybee. By using Honeybee Run Daylighting Simulation, the results were exported to Honeybee_Read
Annual Result Ⅰ. The sDA
300,50%
of the analyzed area at each floor can be found by using branch. DA can
be visualized by using reColorMesh and it was moved to the right side of the view. The analysis period is 8
am to 6 pm all over the year. The step is repeated for different function spaces. The Grasshopper script to do
the simulation was based on the method given by Mostapha Roudsari (Roudsari 2015).
• Script
The SDA simulation includes 5 parts: grid size, result background, simulation settings, converting the
building into Honeybee surfaces, and sDA simulation for each function space (Figure 3-9).
Figure 3-9: The overview of sDA simulation script
(1) Grid size and (2) Result background
The grid size was 2’ (Figure 3-10). The floor panels and walls were used as the background for the result
mesh (Figure 3-10). Thus, the result mesh could be placed at the top of the floor plan (Figure 3-11).
(5) sDA simulation for each function
space
(3) Simulation
Settings
(1) Grid size
(2) Result
Background
(4) Convert the building
into Honeybee surfaces
44
Figure 3-10: Script for ①Grid size and ② Result background
Figure 3-11: Result background
(3) Simulation settings
A north sign was set. The radiance quality was 0, which means low. The ambient bounces for radiance was
3. A Boolean toggle controlled the sDA simulation running (Figure 3-12).
45
Figure 3-12: (3) Simulation settings
(4) Convert the building into Honeybee Surfaces
To convert building surfaces into Honeybee surfaces, materials were set for each surface. (Figure 3-13). The
surface type was set for each surface. Four types of materials were used: Larch, plaster, double glazing clear
float and grey material for the site (Figure 3-14). The surfaces of the Rhino model were converted into
Honeybee Surfaces by using “Create HBsrfs” (Figure 3-15, 3-16).
Figure 3-13: (4) Convert the building into Honeybee surfaces
Ⅱ. Convert floors into
Honeybee surfaces
Ⅰ. Set
materials
46
Figure3-14: ⅠSet materials
Figure 3-15: ⅡConvert the floors into Honeybee Surfaces.
47
Figure 3-16: the building converted into Honeybee surfaces
(5) SDA simulation for each function space
Daylight simulations were run for each function space (the whole building, activity, office, circulation,
dining, care and apartment) separately (Figure 3-17,3-18, 3-19, 3-20). SDA was obtained from the simulation.
The method to calculate sDA for each function space was the same (Figure 3-21). The activity space was
used as the example to explain the method to calculate sDA for a function space. The method was divided
into 6 steps (Figure 3-22).
Figure 3-17: The Rhino model of the case building
48
Figure 3-18: the 1st floor divided into different function spaces
49
Figure 3-19 the 2nd and 3rd floor divided into different function spaces
Figure 3-20: Legend of function spaces
50
Figure 3-21: (5) sDA simulation for each function space
Figure 3-22: sDA simulation for activity space
① The test surfaces of each floor were selected as boundary representation using the Brep command in
Grasshopper, and each floor was given a path number (Figure 3-23). The analysis surface was 2.5’ from the
selected surfaces. The test points, north, epw weather file of Innsbruck, Austria, test mesh, radiance
parameters were imported into the component “Annual Daylight Simulation” and the component exported
an analysis recipe to another component called “Run Daylight Autonomy” (Figure 3-24). ②The analysis
①
②
③
⑤
④
③
⑥
③
Ⅰ. sDA simulation for activity space
The whole building
Activity
Office
Circulation
Dining
Care
Apartment
51
period was imported to a component called “Read Annual Result Ⅰ” as an occupancy file (Figure 3-25). By
importing annual analysis files, test points, points vectors and the occupancy files, the sDA of each test
surface can be exported by “Read Annual Result Ⅰ”.
Figure 3-23: Test meshes for activity space
Figure 3-24: sDA simulation for activity space step ①
52
Figure 3-25: sDA simulation for activity space step ②
③ The results were visualized by showing Daylight Autonomy (DA). DA can be visualized by “Recolor
Mesh” (Figure 3-26). Different colors represent different Daylight Autonomy (Figure 3-27). ④ Then the
visualized meshes were moved to the right side of the canvas (Figure 3-28, 3-29).
Figure 3-26: Visualize DA
53
Figure 3-27: Visualized DA of activity space
Figure 3-28: Move the DA mesh to the right side of the canvas
Figure 3-29: the visualized sDA meshes moved to the right side of the canvas
54
⑤ Because the simulation only gave out the sDA of each test surface, the sDA of each floor and the average
sDA of the whole building was calculated by some other script. By using branch, the sDA of each floor were
found (Figure 3-30). ⑥ Similarly, the average sDA of the whole building was obtained (Figure 3-31).
Figure 3-30: sDA of each floor
Figure 3-31: Calculate the average sDA
3.3.2 Calculate ASE by Ladybug
To calculate ASE, all the solid walls of the case building were selected as the context. Six tests were carried
for different function spaces: the whole building, office, activity, circulation, apartment, and care (care bath,
nursing station and therapy). The epw weather file of Innsbruck was imported to Ladybug. By using Ladybug
“Sun path” component, the direct illuminance over 1000 lux was calculated. Next, “Sunlight Hour Analysis”
gives out a sunlight hour mesh. The percentage of the points of the mesh over 250 hours during a year was
calculated as ASE. The result could be visualized by a sunlight hour mesh and it was moved to the right side
of the view. The analysis period was 8 am to 6 pm all over the year. The step was repeated for different
floors of function spaces. The method was based on the script for ASE simulation given by Chris Mackey
(Mackey 2017).
• Script
ASE simulation includes 3 parts: control, climate and context input; grid size and result background; and
ASE simulation. The simulations were carried out for the whole building and the 6 function spaces (activity,
office, circulation, dining, care and apartment) separately (Figure 3-32). The method to do the ASE
simulation of the activity space was used as an example to demonstrate the approach to calculate ASE. The
approach for the ASE calculation of other spaces were similar. The method to calculate the ASE for activity
55
space was divided into 7 parts: grid size and result background; control climate and context input; settings
and the average ASE of the 3 floors; 1st floor ASE; 2nd floor ASE; and 3rd floor ASE (Figure 3-33).
Figure 3-32: The overview of ASE simulation script
Figure 3-33: ASE simulation for activity space
① Grid size and Result background
The grid size was 2’ (Figure 3-34). The floor panels and walls were used as the background for the result
mesh (Figure 3-34). Thus, the result mesh could be placed at the top of the floor plan (Figure 3-35).
① Grid size
and result
background
Activity space
ASE
③
Settings
⑤ 1
st
floor
ASE
⑥ 2
nd
floor
ASE
⑦ 3
rd
floor
ASE
② Control,
climate and
context
input
④ The
ASE
of the
whole
building
ASE
Settings
The ASE of
the 1
st
floor
The ASE of
the 2
nd
floor
The ASE of
the 3
rd
floor
(1) Control,
climate and
context input
The
ASE
of the
whole
building
(2) Grid size and
result background
The whole
building
Activit
y
Office
Circulation
Dining
Care
Apartment
(3) ASE
simulation
56
Figure 3-34: Script for ①Grid size and Result background
Figure 3-35: Result background
② Control, climate and context input
To calculate ASE, all the surfaces of the model except for the glazing surface were selected as part of the
context for ASE simulation (Figure 3-36). The epw weather file of Innsbruck, Austria was imported. A
Boolean Toggle controls where or not the ASE calculation is running (Figure 3-37).
57
Figure 3-36: the context for ASE simulation
Figure 3-37: ② Control, climate and context input
58
③ Simulation Settings
The settings for ASE simulation was introduced in 3 parts, A, B, C (Figure 3-38). The epw weather file was
imported to Ladybug. Then the direct illuminance in the horizontal plane was found (Figure 3-39). The
hourly illuminance was calculated by averaging the illuminance at the start and the end of each hour. The
annual hourly data, the building, and a threshold of 1000 lux, were imported to the component of “Sun path”.
The analysis period was set to 8am to 6 pm everyday all over the year (Figure 3-40). A sunlight hour mesh
would be obtained from the simulation, which would be introduced in the following content in this section.
A color scheme was selected here for the mesh (Figure 3-41). Then the sunlight hours hitting the test surface
was obtained.
Figure 3-38: ③ Simulation settings
Figure 3-39: Simulation setting A
Figure 3-40: Simulation setting B
C
A
B
59
Figure 3-41: Simulation setting C
60
④ The ASE of the whole building ⑤ 1st floor ASE ⑥ 2nd floor ASE ⑦ 3rd floor ASE
The method to do ASE simulation for each floor were the same for each floor and function space (Figure 3-
42, 3-43, 3-44, 3-45|). The ASE simulation process for the 1st floor of activity space was used as an example
to explain the approach (Figure 3-43). The settings, the context (all the solid walls) and the test geometry
were imported into “Sunlight Hours Analysis” and a sunlight hours mesh was obtained (Figure 3-46, 3-47).
The portion of the space with direct illuminance over 1000 lux above 250 hours was calculated (Figure 3-
43). The result was visualized and moved to the right side of the canvas (Figure 3-47, 3-48). The model at
top left is the case building and the red background with white interior walls are the result meshes (Figure
3-49).
Figure 3-42: ④ Calculate the ASE of the activity space of the whole building
Figure 3-43: ⑤ Calculate the ASE of the 1st floor activity space
Figure 3-44: ⑥ Calculate the ASE of the 2nd floor activity space
Figure 3-45: ⑦ Calculate the ASE of the 3rd floor activity space
61
Figure 3-46: the test points for activity space
Figure 3-47: Visualized ASE
Figure 3-48: zoom in view of ASE result meshes moved to the right side of the canvas
62
Figure 3-49: Visualized ASE moved to the right side of the canvas
3.3.3 Calculate fsDA and fASE
The calculation of fsDA
300,50%
and fASE
1000,250h
Used the same method of calculating sDA and ASE except
for setting the analysis period according to the real operation time. The analysis period of fsDA can be
changed by editing occupancy files (Figure 3-50). The analysis period of ASE can be changed by editing the
hour of sun path analysis (Figure 3-51).
Figure 3-50: Analysis period for fsDA
sDA result meshes ASE result meshes
Case
building
63
Figure 3-51: Analysis period for fASE
3.3.4 Compare fsDA with sDA, and fASE with ASE
Compare fsDA
300,50%
with sDA
300,50%
, and fASE
1000,250h
with ASE
1000,250h
. Find out if there is a large
difference between the metrics with fixed analysis period and the new metrics with flexible analysis period.
(1) Compare sDA and fsDA.
The table was used to compare sDA
300,50%
and fsDA
300,50%
(Table 3-2).
The absolute error of fsDA is calculated by the following formula.
Δx
fsDA
= 𝑓𝑠𝐷𝐴 𝑖 − 𝑠𝐷𝐴 𝑖
MAE
fsDA
=
1
𝑛 ∑ │(𝑓𝑠𝐷𝐴 𝑖 − 𝑠𝐷𝐴 𝑖 )│
𝑛 𝑖 =1
When
Δx
fsDA
= The absolute errors of fsDA
n = The number of errors
MAE
fsDA
=Mean absolute error of fsDA
Δx
fsDA
in different categories were shaded by different colors (Figure 3-52).
Table 3-2: Sample result of Δx
fsDA
Function space Schedule Average 1st floor 2nd floor 3rd floor
The whole building
Fixed Schedule 0.00% 0.00% 0.00% 0.00%
Schedule 1 -1.14% -0.96% -1.19% -1.20%
Schedule 2 -2.37% -2.04% -2.42% -2.57%
Schedule 3 -2.78% -2.38% -2.67% -3.19%
Activity
Fixed Schedule 0.00% 0.00% 0.00% 0.00%
Schedule 1 -0.65% -0.53% -0.76% -0.77%
Schedule 2 -1.66% -1.06% -2.12% -2.30%
Schedule 3 -1.71% -1.06% -2.11% -2.49%
Office
Fixed Schedule 0.00% 0.00% 0.00% 0.00%
64
Schedule 1 1.10% 0.48% 0.00% 3.75%
Schedule 2 -1.06% -2.89% 0.00% 2.50%
Schedule 3 -4.06% -4.81% 0.00% -6.25%
Circulation
Fixed Schedule 0.00% 0.00% 0.00% 0.00%
Schedule 1 -0.89% -1.11% -0.99% -0.68%
Schedule 2 -2.08% -2.85% -1.78% -1.92%
Schedule 3 -2.11% -2.43% -1.90% -2.15%
Dining
Fixed Schedule 0.00% / 0.00% 0.00%
Schedule 1 -5.00% / -6.25% -3.75%
Schedule 2 0.25% / 0.57% -0.07%
Schedule 3 0.91% / 2.74% -0.92%
Care
Fixed Schedule 0.00% / 0.00% 0.00%
Schedule 1 0.66% / 2.17% -0.85%
Schedule 2 0.46% / 2.17% -1.24%
Schedule 3 9.92% / 9.52% 10.31%
Apartment
Fixed Schedule 0.00% / 0.00% 0.00%
Schedule 1 -1.35% / -1.12% -1.57%
Schedule 2 -3.87% / -3.06% -4.66%
Schedule 3 -4.64% / -3.94% -5.32%
0.0%<=│ Δx
fsDA
│<=5.0%
5.0%<│ Δx
fsDA
│<=10.0%
10%<│ Δx
fsDA
│
Figure 3-52: Legend of Δx
fsDA
(2) Compare ASE and fASE
A table will be used to compare ASE
1000,250h
and fASE
1000,250h
(Table 3-3).
The absolute error of fASE is calculated by the following formula.
Δx
fASE
= 𝑓𝐴𝑆𝐸 𝑖 − 𝐴𝑆𝐸 𝑖
MAE
fASE
=
1
𝑛 ∑ │(𝑓𝐴𝑆𝐸 𝑖 − 𝐴𝑆𝐸 𝑖 )│
𝑛 𝑖 =1
When
Δx
fASE
= the absolute errors of fASE
n = the number of errors
MAE
fASE
=Mean absolute error of fASE
The ASE changes in different categories were shaded by different colors (Figure 3-53).
Table 3-3: Sample result of Δx
fASE
Function space Schedule Average 1st floor 2nd floor 3rd floor
The whole building
Fixed Schedule 0.00% 0.00% 0.00% 0.00%
Schedule 1 0.64% 0.16% 0.74% 0.87%
65
Schedule 2 1.61% 2.03% 1.40% 1.52%
Schedule 3 1.23% 2.58% 0.79% 0.72%
Activity
Fixed Schedule 0.00% 0.00% 0.00% 0.00%
Schedule 1 1.40% 0.00% 4.07% 1.28%
Schedule 2 4.23% 4.96% 5.51% 1.62%
Schedule 3 3.44% 6.15% 1.62% 0.35%
Office
Fixed Schedule 0.00% 0.00% 0.00% 0.00%
Schedule 1 -3.98% -4.65% 0.00% -6.25%
Schedule 2 -3.43% -4.65% 0.00% -3.75%
Schedule 3 0.55% 0.01% 0.00% 2.50%
Circulation
Fixed Schedule 0.00% 0.00% 0.00% 0.00%
Schedule 1 0.45% 0.16% 0.61% 0.49%
Schedule 2 0.45% 0.16% 0.61% 0.49%
Schedule 3 0.15% 0.63% 0.00% 0.00%
Dining
Fixed Schedule 0.00% / 0.00% 0.00%
Schedule 1 -7.50% / -5.00% -10.00%
Schedule 2 -7.50% / -5.00% -10.00%
Schedule 3 -6.87% / -5.00% -8.75%
Care
Fixed Schedule 0.00% / 0.00% 0.00%
Schedule 1 -0.58% / 0.00% -1.16%
Schedule 2 -0.58% / 0.00% -1.16%
Schedule 3 -0.58% / 0.00% -1.16%
Apartment
Fixed Schedule 0.00% / 0.00% 0.00%
Schedule 1 0.41% / 0.48% 0.34%
Schedule 2 1.95% / 2.00% 1.91%
Schedule 3 1.89% / 1.79% 1.99%
0.0%<=│ Δx
fASE
│<=5.0%
5.0%<│ Δx
fASE
│<=10.0%
10%<│ Δx
fASE
│
Figure 3-53: Legend of Δx
fASE
(3) Get conclusion
First, the distribution of Δx
fsDA
and Δx
fASE
were visualized by scatter chart (Figure 3-54). The table counts
the number of Δx
fsDA
and Δx
fASE
in different value ranges (Table 3-4). The standard deviation was calculated
to suggest whether sDA and ASE change much with different analysis period. A sample result is given (Table
3-4).
To decide whether the metrics with flexible analysis period make large difference to those with fixed analysis
period, the table counts the number of Δx
fsDA
and Δx
fASE
in different value ranges (Table 3-4). The standard
deviation of Δx
fsDA
and Δx
fASE
were calculated. The mean absolute error (MAE) of fsDA and fASE were
calculated as well (Table 3-4). Thus, the conclusion whether fsDA and ASE were necessary could be reached.
66
Figure 3-54: Sample scatter chart of Δx
fsDA
and Δx
fASE
Table 3-4: Sample of the number of sDA and ASE changes
Δx ΔxfsDA ΔxfASE
│Δx│<=5% 69 66
5%<│Δx │<=10% 5 9
10%<│ Δx │ 1 0
Standard deviation 2.97% 3.33%
Mean absolute error (MAE) 2.36% 2.22%
3.3.5 Suggest whether fsDA and fASE are necessary
After getting sDA and ASE change, the conclusion whether fsDA and fASE are necessary for lighting
performance evaluation can be reached. If the change is significant, the new metrics can be essential to
describe the lighting performance. Otherwise, the new metrics may not be necessarily needed.
3.4 Explore occupied hierarchy of the function spaces
This section covers simulating occupant spatial distribution, sensitivity test for Hour Percentage (P) to
occupant spatial distribution, analyzing the result of the occupant spatial distribution of the three schedules,
analyzing the result of the sensitivity test and suggesting a hierarchy for function space.
The occupied hierarchy was explored to find out which function space people spend more and less time in.
It would be good if all the function spaces achieve good daylighting performance. However, this may not be
always realistic so the priority should be given to more important spaces, where people spend more time in.
(Regnier, 2018). Thus, the requirements on sDA
300,50%
and ASE
1000,250h
for spaces with different importance
could vary. Therefore, deciding which space was more important could be necessary.
Occupied Hour (O) and Hour Percentage (P) of the function spaces could be two influential factors. The
occupied hierarchy of the function spaces was discussed by comparing Occupied Hour (O) and Hour
Percentage (P) of each function space when the operating schedules were different. The three schedules
mentioned in chapter 2 were used. Although the number of residents and staff were different in the three
cases, they were assumed to be the same for the research. The method to obtain Occupied Hour (O) and Hour
Percentage (P) will be explained in 3.4.1. The occupant spatial distribution under the three schedules were
simulated to find out the Occupied Hour (O) and the Hour Percentage (P) under each schedule (Figure 3-55).
With the same schedule, Occupied Hours (O) may be predictable according to operation and it may not
-15.00%
-10.00%
-5.00%
0.00%
5.00%
10.00%
15.00%
sDA/ASE change
fsDA/fASE absolute error
ΔxfsDA
ΔxfASE
67
change much with different occupant spatial distribution. If so, Occupied Hour (O) could be a factor to define
“regularly occupied” space. However, Hour Percentage (P) may not be predictable based on operation
schedule as it can change if occupant spatial distribution changed. To decide whether the metric can define
occupied hierarchy, it was necessary to find out how sensitive Hour Percentage (P) to occupant spatial
distribution was. Therefore, 2 sensitivity tests were carried out. Schedule 3 was used for the test (Figure 3-
55). If Hour Percentage (P) was sensitive to occupant spatial distribution change, P might not be able to be
used as a factor to decide the occupied hierarchy of the function spaces. Otherwise, P might be used as a
factor to decide the occupied hierarchy of the function spaces. Finally, a sample occupied hierarchy was
suggested and different requirements on sDA/fsDA and ASE/fASE were called for.
Figure 3-55: Method to explore occupied hierarchy of the function spaces
3.4.1 Simulate occupant spatial distribution
Three occupant spatial distributions were simulated according the three schedules mentioned in chapter 2.
Although the number of residents and staff were different in the three cases, they were assumed to be the
same for the research. The following assumptions were made for the simulation:
• The occupants have a regular daily schedule.
• There are 54 residents living in the assisted living.
• In total, there are 18 staff in the analyzed area during the analyzed period.
• The staff make 3 shifts to make sure the residents are taken care 24 hours a day.
(1) Simulate occupant spatial distribution under schedule 1
• Make assumptions
The staff do not stay in the office for long. They walk around and take care of the residents (Table 3-5).
Table 3-5: Assumed staff schedule 1
68
Time Number of staffs
8:00-17:00 9
17:00-1:00 5
1:00-8:00 4
• Use the residents schedule given by Victor Regnier
Victor Regnier, an expert in assisted living, provided a sample schedule for assisted living (Table 3-6)
(Regnier, Senior living schedule 2019)
Table 3-6: Residents schedule 1 (Regnier, Senior living schedule 2019)
Time Activity
8:00-10:00 Breakfast
10:00-12:00 Shower/ Stay in bedroom or activity
room/Take snacks
12:00-14:00 Lunch
13:30-17:00 Sleep/Stay in bedroom or activity
room/Take snacks
17:00-19:00 Dinner
• Make an assumed people distribution table without thinking about circulation time (Table 3-7)
The time period from breakfast until dinner was listed by every half an hour. The number of people in each
space would be filled in the shaded cells. Because people hardly ever spent a continuous 30 minutes in
circulation space, it could be difficult to put in the number of people in circulation space. Therefore, the
condition without considering circulation time was assumed first. Then it was assumed people spend certain
percentage of time in circulation space and the time spent in different function spaces was deducted by the
time in circulation space.
Assume people do not spend time in circulation space.
Total hours: 63*9+59*2=685 (∑number of people * time)
The shaded area should be filled with the number of people
The minimum Occupied Hours (O) was found.
Table 3-7: Occupant spatial distribution under schedule 1
Time Spatial people distribution Total
Dining Activity Apartment Office Care Circulation
8:00-8:30 / 63
8:30-9:00 / 63
9:00-9:30 / 63
9:30-10:00 / 63
10:00-10:30 / 63
10:30-11:00 / 63
11:00-11:30 / 63
11:30-12:00 / 63
12:00-12:30 / 63
12:30-13:00 / 63
13:00-13:30 / 63
13:30-14:00 / 63
14:00-14:30 / 63
69
14:30-15:00 / 63
15:00-15:30 / 63
15:30-16:00 / 63
16:00-16:30 / 63
16:30-17:00 / 63
17:00-17:30 / 59
17:30-18:00 / 59
18:00-18:30 / 59
18:30-19:00 / 59
Occupied Hour
(O)
/
Total hours a b c d e / 685.0
No circulation
Hour Percentage
(P)
a/685 b/685 c/685 d/685 e/685 / 100%
• Make an assumed time distribution for each program with the consideration of circulation time
It was assumed that everyone spent x% of their time in circulation space. Then the total time in circulation
space was 685*x%=x%*685
The time people spent in each space(W) should be subtracted by the time they spent in circulation space (C)
according to the weight. Thus, the Real hour (R) could be reached. Finally, the Hour Percentage (P) of each
function space can be calculated (Table 3-8).
Table 3-8: Hour Percentage of each function space under schedule 1
Assumed hours
without thinking about
circulation time (W)
Assumed circulation
hours that takes up the
time in the space (C)
Real hour (R) Hour Percentage (P)
of each function space
Dining
a (a/685)*(685*x%) a-C(dining) R(dining)/685
Activity
b (b/685)*(685*x%) b-C(activity) R(activity)/685
Apartment
c (c/685)*(685*x%) c-C(apartment) R(apartment)/685
Office
d (d/685)*(685*x%) d-C(office) R(office)/685
Care
e (e/685)*(685*x%) e-C(care) R(care)/685
Circulation
/ f f R(circulation)/685
(2) Simulate occupant spatial distribution under schedule 2
• Make assumptions
-The staff do not stay in the office for long. They walk around and take care of the residents (Table 3-9).
Table 3-9: Assumed staff schedule 2
Time Number of staffs
8:00-17:00 9
17:00-1:00 5
1:00-8:00 4
• Use the schedule of Belmont Village, Hollywood Hills.
-The residents use the schedule from Belmont village, Hollywood Hills (Manager of Belmont Village
2019)(Table 3-10)
70
Table 3-10: Residents schedule 2 (Manager of Belmont Village 2019)
Time Activity
7:00-9:00 Breakfast
9:30-11:00 Chair exercise/other activities
11:00-13:00 Lunch
13:00-14:00 Nap/Exercise
14:00-16:30 Activity
16:30-19:00 Dinner
• Make an assumed people distribution table without thinking about circulation time (Table 3-11)
Assume people do not spend time in circulation space.
Total hours: 58*1+63*9+59*2=743 (∑number of people * time)
The shaded area should be filled with the number of people.
The minimum Occupied Hours (O) was found.
Time Spatial people distribution Total
Dining Activity Apartment Office Care Circulation
7:00-7:30 / 58
7:30-8:00 / 58
8:00-8:30 / 63
8:30-9:00 / 63
9:00-9:30 / 63
9:30-10:00 / 63
10:00-10:30 / 63
10:30-11:00 / 63
11:00-11:30 / 63
11:30-12:00 / 63
12:00-12:30 / 63
12:30-13:00 / 63
13:00-13:30 / 63
13:30-14:00 / 63
14:00-14:30 / 63
14:30-15:00 / 63
15:00-15:30 / 63
15:30-16:00 / 63
16:00-16:30 / 63
16:30-17:00 / 63
17:00-17:30 / 59
17:30-18:00 / 59
18:00-18:30 / 59
18:30-19:00 / 59
Occupied Hour
(O)
/
Total Hours a b c d e / 743.0
No circulation
Hour
Percentage (P)
a/743 b/743 c/743 d/743 e/743 / 100%
71
Table 3-11: Occupant spatial distribution under schedule 2
• Make an assumed time distribution for each program with the consideration of circulation time
It was assumed that everyone spent x% of their time in circulation space. Then the total time in circulation
space was 743*x%=x%*743
The time people spent in each space(W) should be subtracted by the time they spend in circulation space (C)
according to the weight. Thus, the Real hour (R) could be calculated.
Finally, the Hour Percentage (P) of each program was obtained (Table 3-12).
Assumed hours
without thinking about
circulation time (W)
Assumed circulation
hours that takes up the
time in the space (C)
Real hour (R) Hour Percentage (P) of
each function space
Dining
a (a/743)*(743*x%) a-C(dining) R(dining)/743
Activity
b (b/743)*(743*x%) b-C(activity) R(activity)/743
Apartment
c (c/743)*(743*x%) c-C(apartment) R(apartment)/743
Office
d (d/743)*(743*x%) d-C(office) R(office)/743
Care
e (e/743)*(743*x%) e-C(care) R(care)/743
Circulation
/ f f R(circulation)/ 743
Table 3-12: Hour Percentage (P) of each function space under schedule 2
(3) Simulate occupant spatial distribution under schedule 3
• Make assumptions
-The staff stay in the office for long. The office is occupied all day long (Table 3-13).
Table 3-13: Assumed staff schedule 2
Time Number of staffs
9:00-18:00 9
18:00-2:00 5
2:00-9:00 4
• Use the schedule of Sunrise of Beverly Hills.
-The residents use the schedule from Sunrise of Beverly Hills (Table 3-14) (Manager of Sunrise of Beverly
Hills 2019).
Table 3-14: Residents schedule 2 (Manager of Sunrise of Beverly Hills 2019)
Time Activity
6:30-8:00 Breakfast
9:30-11:00 Activity
12:00-13:30 Lunch
13:00-14:00 Nap/Exercise
14:00-16:30 Activity
17:00-18:30 Dinner
• Make an assumed people distribution table without thinking about circulation time (Table 3-15)
Assume people do not spend time in circulation space.
Total hours: 58*2.5+63*9+59*0.5=741.5 (∑number of people * time)
The shaded area should be filled with the number of people.
The minimum Occupied Hours (O) was found.
Table 3-15: Occupant spatial distribution under schedule 3
Time Spatial people distribution Total
72
Dining Activity Apartment Office Care Circulation
6:30-7:00 / 58
7:00-7:30 / 58
7:30-8:00 / 58
8:00-8:30 / 58
8:30-9:00 / 58
9:00-9:30 / 63
9:30-10:00 / 63
10:00-10:30 / 63
10:30-11:00 / 63
11:00-11:30 / 63
11:30-12:00 / 63
12:00-12:30 / 63
12:30-13:00 / 63
13:00-13:30 / 63
13:30-14:00 / 63
14:00-14:30 / 63
14:30-15:00 / 63
15:00-15:30 / 63
15:30-16:00 / 63
16:00-16:30 / 63
16:30-17:00 / 63
17:00-17:30 / 63
17:30-18:00 / 63
18:00-18:30 / 59
Occupied Hour
(O)
/
Total hours a b c d e / 741.5
No circulation
Hour Percentage
(P)
a/741.5 b/741.5 c/741.5 d/741.5 e/741.5 / 100%
• Make an assumed time distribution for each program with the consideration of circulation time
Then it was assumed that everyone spent x% of their time in circulation space. Then the total time in
circulation space was 741.5*x%=x%*741.5
The time people spent in each space(W) should be subtracted by the time they spend in circulation space (C)
according to the weight. Thus, the Real hour (R) could be calculated.
Finally, the Hour Percentage (P) of each program was obtained (Table 3-16).
Table 3-16: Hour Percentage of each function space under schedule 3
Assumed hours
without thinking about
circulation time (W)
Assumed circulation hours
that takes up the time in the
space (C)
Real hour (R) Hour Percentage (P) of
each function space
73
Dining
a (a/741.5)*(741.5*x%) a-C(dining) R(dining)/741.5
Activity
b (b/741.5)*(741.5*x%) b-C(activity) R(activity)/741.5
Apartment
c (c/741.5)*(741.5*x%) c-C(apartment) R(apartment)/741.5
Office
d (d/741.5)*(741.5*x%) d-C(office) R(office)/741.5
Care
e (e/741.5)*(741.5*x%) e-C(care) R(care)/741.5
Circulation
/ f f R(circulation)/ 741.5
3.4.2 Sensitivity test for Hour Percentage (P) to occupant spatial distribution
With the same schedule, Occupied Hours (O) may not change much with different occupant spatial
distribution while Hour Percentage (P) may change significantly. It is necessary to find out how sensitive
Hour Percentage (P) to occupant spatial distribution is. Therefore, sensitivity tests were carried out. Schedule
3 was selected for the test. If the Hour Percentage (P) can vary significantly with slight people distribution
change, the metric of Hour Percentage (P) may not be a very useful factor to decide the importance of a
function space. If the Hour Percentage (P) does not change significantly with moderate people distribution
difference, the metric of Hour Percentage (P) can be a helpful factor to decide the hierarchy of function
spaces.
Schedule 3 was selected for the test. Two sensitivity tests were carried out. The first test was to explore the
change of Hour Percentage (P) between two function space, both of which had high population density. First,
it was tested how much the Hour Percentage (P) could change if 1 person was moved from the apartment
space to the activity space for the whole day. Then the number of people removed became 5 and finally 10.
The second test was to figure out the change of Hour Percentage (P) between another 2 function spaces, one
of which had high population density while the other one had low population density. It was tested how
much the Hour Percentage (P) could change if one person was moved from the apartment space to the office
for the whole day. Then the condition when 2 people were moved from the apartment space to the office
space was tested. Because office was small, it was unlikely 5 or 10 people move, so only 1 and 2 people
were moved.
The test used a similar table to that to simulate occupant spatial distribution, but the percentage of people
moving was calculated. The cells of occupant spatial distribution were shaded in yellow and the percentage
of people moving was shaded in green (Table 3-17).
74
Table 3-17: Table for sensitivity test
3.4.3 Analyze the result of the occupant spatial distribution of the three schedules
The Occupied Hour (O) and Hour percentage (P) found in the result of occupant spatial distribution
simulation under the three schedules were compared to each other. It was discussed whether the results vary
with changing operation schedules. If it is true, can the management group control the value of the two
metrics by adjusting schedule? If they can be controlled, they can be the factors to decide the occupied
hierarchy of function spaces. Otherwise, the two metrics may not be the right factors to decide occupied
hierarchy. Occupied Hours (O) and Hour Percentage (P) of the function spaces are compared by using the
table (Table 3-18).
Table 3-18: Occupied Hour (O) and Hour Percentage (P) under the three schedules
Occupied Hour (O) Hour Percentage (P)
Schedule 1 Schedule 2 Schedule 3 Schedule 1 Schedule 2 Schedule 3
Dining
Activity
Apartment
Office
Care
Circulation / / /
3.4.4 Analyze the result of the sensitivity test
The result of sensitivity test was listed. The Hour Percentage (P) of activity space and apartment were
changed in test 1. The result of apartment and office were changed in test 2. These changed figures are
shaded in grey (Table 3-19).
Dining Activity Apartment Office Care Circulation Number of people
moving/People in
apartment
Number of people
moving/People in
activity space
Number of people
moving/People in
total
6:30-7:00 / 58
7:00-7:30 / 58
7:30-8:00 / 58
8:00-8:30 / 58
8:30-9:00 / 58
9:00-9:30 / 63
9:30-10:00 / 63
10:00-10:30 / 63
10:30-11:00 / 63
11:00-11:30 / 63
11:30-12:00 / 63
12:00-12:30 / 63
12:30-13:00 / 63
13:00-13:30 / 63
13:30-14:00 / 63
14:00-14:30 / 63
14:30-15:00 / 63
15:00-15:30 / 63
15:30-16:00 / 63
16:00-16:30 / 63
16:30-17:00 / 63
17:00-17:30 / 63
17:30-18:00 / 63
18:00-18:30 / 59
Occupied Hour (O) /
Total hours a b c d e / 741.5
No circulation Hour
Percentage (P)
a/741.5 b/741.5 c/741.5 d/741.5 e/741.5 /
Occupant spatial distribution Total Sensitivity test people moving percentage Time
75
Hour Percentage Change =New Hour percentage- Original Hour Percentage
The extent the Hour percentage changes can show how sensitive it is to occupant spatial distribution. After
getting the result, the changes in different value categories are shaded in different colors (Figure 3-56).
Table 3-19: Hour Percentage Change
Original Hour
Percentage (P)
Test 1
part 1
Test 1
part 2
Test 1
part 3
Test 2 part
1
Test 2
part 2
Dining
Activity
Apartment
Office
Care
Circulation
0%<=│Hour Percentage Change│<5%
5%<=│Hour Percentage Change│<10%
10%<=│Hour Percentage Change│<15%
Figure 3-56: The legend of result change
The percentages of people moving were calculated. The number of people moving was divided by different
figures. If people move from apartment to activity space, the number of people moving was divided by the
number of people in apartment, activity space, and total number of people in the assisted living. If people
move from apartment to office, the number of people moving was divided by the number of people in
apartment, office, and total number of people in the assisted living. It was discussed whether the percentage
of people moving influenced Hour Percentage (P)’s sensitivity to occupant spatial distribution. People
moving percentages can change with time. The percentages were shown by scattered charts. A sample scatter
chart is given (Figure 3-57). The average, minimum value, maximum value and median were listed for each
item (Table 3-20).
Figure 3-57: A sample people moving percentage scatter chart
0%
20%
40%
60%
80%
100%
120%
People moving percentage
5/People in apartment 5/People in activity space 5/People in total
76
Table 3-20: A sample people moving data
Item Statistic Value
5/People in apartment
Average
Min
Max
Median
5/People in activity space
Average
Min
Max
Median
5/People in total
Average
Min
Max
Median
3.4.5 Suggest a hierarchy for function space
According to the analysis result of occupant spatial distribution of the three schedules and the analysis result
of the sensitivity test, suggest a sample occupied hierarchy.
• The “regularly occupied space” can be defined according to the minimum Occupied Hours (O) found in
the schedule studies (Table 3-21).
Space hierarchy
Definition
Regularly occupied space A space which is occupied for at least n hours a day.
Table 3-21: Sample definition of “regularly occupied space”
• If Hour Percentage (P) is significantly sensitive to occupant spatial distribution, Occupied Hours (O)
becomes the only factor to decide the occupied hierarchy.
• If Hour Percentage (P) is not significantly sensitive to occupant spatial distribution, within the “regularly
occupied space,” suggest a sample hierarchy for each function space according to Hour Percentage (P) (Table
3-22).
Table 3-22: Sample occupied hierarchy
Occupancy level Hour Percentage (P)
1 …
2 …
3 …
… …
… …
… …
• Call for requirement on sDA/fsDA and ASE/fASE for function spaces with different occupied levels
A space with higher occupied level requires higher fsDA
300,50%
and lower fASE
1000,250h
. A reference for
fsDA
300,50%
and fASE
1000,250h
can be suggested (Table 3-23, 3-24). Whether to use sDA and ASE or fsDA and
fASE were decided after the analysis which suggest whether fsDA and ASE were necessarily needed.
Table 3-23: Sample sDA/fsDA and ASE/fASE requirement for “regularly occupied” space
77
Reference for sDA/fsDA and ASE/fASE for “regularly occupied function
sDA300,50% range
ASE1000,250h range
Table 3-24: Sample sDA/fsDA and ASE/fASE requirement for function spaces with different occupied levels
Occupied level
Hour Percentage (P)
sDA/fsDA requirement ASE/fASE requirement
1 …
… …
2 …
… …
3 …
… …
4 …
… …
… …
… …
… …
… …
…
3.5 Summary
The methodology consisted of three main steps (Figure 3-58).
Figure 3-58: Methodology diagram
78
Preparations, including building the case model by Rhino, Dividing the case building into different function
spaces and doing research about the schedules were done at the beginning. First, to explore the influence of
the analysis period on sDA and ASE, sDA and ASE, and fsDA and fASE under the three schedules were
calculated. Then the two new metrics with flexible analysis period were compared to the two with fixed
analysis period. Next, it was discussed whether fsDA and fASE were useful in daylighting performance
based on the value difference between the new and the existing metrics. Then, based on Regnier’s theory
that the space where people spend more time in are more important and need better lighting condition, there
should be a hierarchy of space importance. Thus, the factors that could predict occupied hierarchy of function
spaces were explored. As Occupied Hour (O) and Hour Percentage (P) were important factors to predict
occupied hierarchy, occupant spatial distributions under the three schedules were simulated to find out the
values of O and P. Occupied Hour (O) may be one of the deciding factors while whether Hour Percentage
(P) should be one of the factors was not certain because the factor might be quite sensitive to occupant spatial
distribution even though the schedule kept the same. Thus, sensitivity tests of Hour Percentage (P) to
occupant spatial distribution were carried out. If P is not sensitive to occupant spatial distribution, P could
be used as one factor to decide the occupied hierarchy of function spaces. Then, a sample ranking method
based on O and P for occupied hierarchy was suggested.
The assumptions for occupant spatial distribution are for three schedules (Table 3-25)
Table 3-25: Assumptions for occupant spatial distribution are for three schedules
Schedules Residents Staff
Schedule 1 Some like staying in apartments Do not stay in the office for long
Schedule 2 Do not stay in apartments for long Do not stay in the office for long
Schedule 3 Some like staying in apartments Office is always occupied by staff
4. RESULTS
Chapter 4 lists the outcomes of the test, which are marked with red rectangles (Figure 4-1): influence of
analysis period on sDA and ASE simulation outcomes, occupant spatial distribution of three schedules and
sensitivity test. This chapter explores if fsDA and fASE, the metrics using real occupied hours as the analysis
period, are necessary to describe the lighting performance of the building. It will also demonstrate if the
occupied hierarchy of fsDA and fASE may be defined based on the real Occupied Hours (O) and Hour
Percentage (P).
79
Figure 4-1: Methodology
4.1 Influence of analysis period on sDA and ASE Simulation Outcomes
To explore the influence of analysis period on sDA, fsDA, ASE and fASE with different analysis period
were simulated by Ladybug and Honeybee (Figure 4-2). The simulation results are given in 4.1. SDA and
fsDA are listed first (Table 4-2). ASE and fASE are provided as well (Table 4-3)
80
Figure 4-2: 4.1 Methodology
Table 4-2: sDA and fsDA simulation result
Function space Schedule Average 1st floor 2nd floor 3rd floor
The whole
building
Fixed Schedule 42.4% 33.2% 44.0% 47.5%
Schedule 1 41.3% 32.3% 42.8% 46.3%
Schedule 2 40.1% 31.2% 41.6% 44.9%
Schedule 3 39.6% 30.8% 41.3% 44.3%
Activity
Fixed Schedule 72.9% 58.0% 86.2% 86.8%
Schedule 1 72.2% 57.4% 85.4% 86.0%
Schedule 2 71.2% 56.9% 84.0% 84.5%
Schedule 3 71.2% 56.9% 84.0% 84.3%
Office
Fixed Schedule 36.7% 35.6% 30.0% 46.3%
Schedule 1 37.8% 36.1% 30.0% 50.0%
Schedule 2 35.6% 32.7% 30.0% 48.8%
Schedule 3 32.6% 30.8% 30.0% 40.0%
Circulation
Fixed Schedule 22.7% 19.6% 22.2% 25.2%
Schedule 1 21.8% 18.5% 21.2% 24.5%
Schedule 2 20.6% 16.8% 20.4% 23.3%
Schedule 3 20.6% 17.2% 20.3% 23.1%
Dining
Fixed Schedule 26.3% / 21.3% 31.3%
Schedule 1 23.1% / 17.5% 28.8%
Schedule 2 23.1% / 17.5% 28.8%
Schedule 3 21.3% / 15.0% 27.5%
Care
Fixed Schedule 26.5% / 21.8% 31.2%
Schedule 1 27.2% / 24.0% 30.3%
Schedule 2 27.0% / 24.0% 29.9%
Schedule 3 36.4% / 31.3% 41.5%
Apartment
Fixed Schedule 56.5% / 54.6% 58.4%
Schedule 1 55.2% / 53.5% 56.8%
Schedule 2 52.6% / 51.6% 53.7%
81
Schedule 3 51.9% / 50.7% 53.1%
Table 4-3: ASE and fASE simulation result
Function space Schedule Average 1st floor 2nd floor 3rd floor
The whole building
Fixed Schedule 11.7% 4.5% 13.6% 15.1%
Schedule 1 12.4% 4.7% 14.3% 15.9%
Schedule 2 13.4% 6.5% 15.0% 16.6%
Schedule 3 13.0% 7.1% 14.4% 15.8%
Activity
Fixed Schedule 35.2% 10.5% 57.3% 57.9%
Schedule 1 36.6% 10.5% 61.3% 59.2%
Schedule 2 39.4% 15.5% 62.8% 59.6%
Schedule 3 38.6% 16.7% 58.9% 58.3%
Office
Fixed Schedule 4.0% 4.7% 0.0% 6.3%
Schedule 1 0.0% 0.0% 0.0% 0.0%
Schedule 2 0.6% 0.0% 0.0% 2.5%
Schedule 3 4.5% 4.7% 0.0% 8.8%
Circulation
Fixed Schedule 5.1% 3.1% 4.9% 6.5%
Schedule 1 5.5% 3.2% 5.5% 7.0%
Schedule 2 5.5% 3.2% 5.5% 7.0%
Schedule 3 5.2% 3.7% 4.9% 6.5%
Dining
Fixed Schedule 17.5% / 12.5% 22.5%
Schedule 1 10.0% / 7.5% 12.5%
Schedule 2 10.0% / 7.5% 12.5%
Schedule 3 10.6% / 7.5% 13.8%
Care
Fixed Schedule 0.6% / 0.0% 1.2%
Schedule 1 0.0% / 0.0% 0.0%
Schedule 2 0.0% / 0.0% 0.0%
Schedule 3 0.0% / 0.0% 0.0%
Apartment
Fixed Schedule 13.5% / 12.8% 14.2%
Schedule 1 13.9% / 13.3% 14.6%
Schedule 2 15.5% / 14.8% 16.2%
Schedule 3 15.4% / 14.6% 16.2%
4.2 Occupant spatial distribution of three schedules
To explore occupied hierarchy of the function spaces, the result of step 1 and step 2 are given in 4.2 (Figure
4-3). Three occupant spatial distributions were simulated according to the three schedules mentioned in
chapter 2. Two sensitivity tests were carried out to test the sensitivity of Hour Percentage to occupant spatial
distribution.
82
Figure 4-3: 4.2 Methodology
4.2.1 Occupant spatial distribution under schedule 1
A simulated Occupied Hour (O) and Hour Percentage (P) under schedule 1 of each function space can be
calculated. The circulate hour is assumed first. Then the occupant spatial distribution without considering
circulation hours is simulated. After that, with thinking about the circulation hour, a more accurate occupant
spatial distribution outcome is obtained.
• Circulation hour
It is assumed that people spend 20% of the total time in circulation space. Therefore, the total circulation
hour for the occupants is 137 hours (Table 4-4).
Table 4-4: Circulation Hour under schedule 1
Percentage Total hour Total circulation hour
0.2 685.0 137
• Occupant spatial distribution without considering circulation hours
The Occupied Hour (O) and the Hour Percentage (P) of each function space without thinking about
circulation hours are obtained (Table 4-5).
Table 4-5: Occupant spatial distribution without thinking about circulation hours under schedule 1
Time Spatial people distribution Total
Dining Activity Apartment Office Care Circulation
8:00-8:30 16 1 44 2 0 / 63
83
8:30-9:00 16 14 30 3 0 / 63
9:00-9:30 16 31 16 0 0 / 63
9:30-10:00 16 35 12 0 0 / 63
10:00-10:30 0 51 12 0 0 / 63
10:30-11:00 0 51 12 0 0 / 63
11:00-11:30 0 51 12 0 0 / 63
11:30-12:00 0 51 12 0 0 / 63
12:00-12:30 16 39 8 0 0 / 63
12:30-13:00 16 35 12 0 0 / 63
13:00-13:30 16 31 16 0 0 / 63
13:30-14:00 16 26 21 0 0 / 63
14:00-14:30 0 31 26 2 4 / 63
14:30-15:00 0 36 21 2 4 / 63
15:00-15:30 0 43 16 0 4 / 63
15:30-16:00 0 48 11 0 4 / 63
16:00-16:30 0 55 8 0 0 / 63
16:30-17:00 0 55 8 0 0 / 63
17:00-17:30 16 37 6 0 0 / 59
17:30-18:00 16 35 8 0 0 / 59
18:00-18:30 16 30 13 0 0 / 59
18:30-19:00 16 24 19 0 0 / 59
Occupied Hour (O) 6 11 11 2 2 / 32
Total hours 96.0 405.0 171.5 4.5 8.0 / 685.0
No circulation Hour
Percentage (P)
14% 59% 25% 1% 1% / 100%
• Hour Percentage (P) with the consideration of circulation hours
Hour Percentage (P) with the consideration of circulation hours is calculated after having the circulation
hours (Table 4-6).
Table 4-6: Hour Percentage (P) with consideration of circulation hours under schedule 1
Assumed hours without
thinking about circulation
time (W)
Assumed circulation hours
that takes up the time in the
space (C)
Real hour
(R)
Hour
percentage (P)
Dining 96.0 19.2 76.8 11%
Activity 405.0 81.0 324.0 47%
Apartment 171.5 34.3 137.2 20%
Office 4.5 0.9 3.6 1%
Care 8.0 1.6 6.4 1%
Circulation 0 137.0 137.0 20%
4.2.2 Occupant spatial distribution of schedule 2
A simulated Occupied Hour (O) and Hour Percentage (P) under schedule 2 of each function space can be
obtained by using the same method of simulating occupant spatial distribution under schedule 1.
84
• Circulation hour
It is assumed that people spend 20% of the total time in circulation space. Therefore, the total circulation
hour for the occupants is 148.6 hours (Table 4-7).
Table 4-7: Circulation Hour under schedule 2
Percentage Total hour Total circulation hour
0.2 743.0 148.6
• Occupant spatial distribution without considering circulation hours
The Occupied Hour (O) and the Hour Percentage (P) without thinking about circulation hours of each
function space can be calculated (Table 4-8).
Table 4-8: Occupant spatial distribution without thinking about circulation hours under schedule 2
Time Spatial people distribution Total
Dining Activity Apartment Office Care Circulation
7:00-7:30 16 2 38 2 0 / 58
7:30-8:00 16 16 24 2 0 / 58
8:00-8:30 16 31 13 3 0 / 63
8:30-9:00 16 40 4 3 0 / 63
9:00-9:30 0 57 6 0 0 / 63
9:30-10:00 0 57 6 0 0 / 63
10:00-10:30 0 57 6 0 0 / 63
10:30-11:00 0 57 6 0 0 / 63
11:00-11:30 0 57 6 0 0 / 63
11:30-12:00 0 57 6 0 0 / 63
12:00-12:30 16 39 8 0 0 / 63
12:30-13:00 16 39 8 0 0 / 63
13:00-13:30 16 25 22 0 0 / 63
13:30-14:00 16 22 25 0 0 / 63
14:00-14:30 0 27 30 2 4 / 63
14:30-15:00 0 27 30 2 4 / 63
15:00-15:30 0 47 12 0 4 / 63
15:30-16:00 0 53 6 0 4 / 63
16:00-16:30 0 55 8 0 0 / 63
16:30-17:00 0 55 8 0 0 / 63
17:00-17:30 16 38 5 0 0 / 59
17:30-18:00 16 32 11 0 0 / 59
18:00-18:30 16 26 17 0 0 / 59
18:30-19:00 16 20 23 0 0 / 59
Occupied Hour (O) 6 12 12 3 2 / 35
Total Hour 96.0 468.0 164.0 7.0 8.0 / 743.0
85
No circulation Hour
Percentage (P)
13% 63% 22% 1% 1% / 100%
• Hour percentage (P) with consideration of circulation hours
Hour Percentage (P) with consideration of circulation hours is given (Table 4-9).
Table 4-9: Hour Percentage (P) with consideration of circulation hours under schedule 2
Assumed hours without
thinking about circulation
time (W)
Assumed circulation hours
that takes up the time in the
space (C)
Real hour
(R)
Hour
percentage (P)
Dining 96.0 19.2 76.8 10%
Activity 468.0 93.6 374.4 50%
Apartment 164.0 32.8 131.2 18%
Office 7.0 1.4 5.6 1%
Care 8.0 1.6 6.4 1%
Circulation 0 148.6 148.6 20%
4.2.3 Occupant spatial distribution of schedule 3
A simulated Occupied Hour (O) and Hour Percentage (P) under schedule 3 of each function space can be
obtained by using the same method to simulate occupant spatial distribution under schedule 1.
• Circulation hour
It is assumed that people spend 20% of the total time in circulation space. Therefore, the total circulation
hour for the occupants is 148.3 hours (Table 4-10)
Table 4-10: Circulation Hour under schedule 3
Percentage Total hour Total circulation hour
0.2 741.5 148.3
• Occupant spatial distribution without considering circulation hours
Occupied Hour (O) and Hour Percentage (P) of each function space without thinking about circulation hours
can be calculated (Table 4-11).
Table 4-11: Occupant spatial distribution without thinking about circulation hours under schedule 3
Time Spatial people distribution Total
Dining Activity Apartment Office Care Circulation
6:30-7:00 20 2 35 1 0 / 58
7:00-7:30 20 19 18 1 0 / 58
7:30-8:00 20 23 14 1 0 / 58
8:00-8:30 0 39 18 1 0 / 58
8:30-9:00 0 39 18 1 0 / 58
9:00-9:30 0 41 19 3 0 / 63
9:30-10:00 0 41 19 3 0 / 63
10:00-10:30 0 41 19 3 0 / 63
10:30-11:00 0 41 19 3 0 / 63
86
11:00-11:30 0 41 19 3 0 / 63
11:30-12:00 0 41 19 3 0 / 63
12:00-12:30 20 28 13 2 0 / 63
12:30-13:00 20 25 16 2 0 / 63
13:00-13:30 20 27 14 2 0 / 63
13:30-14:00 0 33 24 2 4 / 63
14:00-14:30 0 43 14 2 4 / 63
14:30-15:00 0 43 14 2 4 / 63
15:00-15:30 0 43 14 2 4 / 63
15:30-16:00 0 45 16 2 0 / 63
16:00-16:30 0 45 16 2 0 / 63
16:30-17:00 0 45 16 2 0 / 63
17:00-17:30 20 31 10 2 0 / 63
17:30-18:00 20 25 16 2 0 / 63
18:00-18:30 20 15 22 2 0 / 59
Occupied Hour (O) 6 12 12 12 2 / 44
Total hours 90.0 408.0 211.0 24.5 8.0 / 741.5
No circulation Hour
Percentage (P)
12% 55% 28% 3% 1% / 100%
• Hour percentage (P) with consideration of circulation hours
Hour Percentage (P) with the consideration of circulation hours is given (Table 4-12).
Table 4-12: Hour Percentage (P) with consideration of circulation hours under schedule 3
Assumed hours without
thinking about circulation
time (W)
Assumed circulation hours
that takes up the time in the
space (C)
Real hour
(R)
Hour percentage
(P)
Dining 90.0 18.0 72.0 10%
Activity 408.0 81.6 326.4 44%
Apartment 211.0 42.2 168.8 23%
Office 24.5 4.9 19.6 3%
Care 8.0 1.6 6.4 1%
Circulation 0 148.3 148.3 20%
4.3 Sensitivity test
Sensitivity tests were carried out for occupant spatial distribution under schedule 3. For schedule 3, the
circulation hour is assumed (Table 4-13).
Table 4-13: Circulation Hour under schedule 3
Percentage Total hour Total circulation hour
0.2 741.5 148.3
4.3.1 Test people move from apartment to activity space
The first part of the test is to calculate Occupied Hour (O) and Hour Percentage (P) when people move from
apartment to activity space.
87
(1) Move 1 person from apartment to activity space.
First, it was tested when 1 person move from apartment to activity space.
• Occupant spatial distribution without considering circulation hours (Table 4-14)
From the occupant spatial distribution without considering circulation hours, Occupies Hour (O) of each
function space, Hour Percentage (P) without thinking about circulation hours and the percentage of people
moving can be calculated.
Table 4-14: Occupant spatial distribution without thinking about circulation hours of sensitivity test 1-1
• Hour percentage (P) with consideration of circulation hours
Hour Percentage (P) with consideration of circulation hours is given (Table 4-15).
Table 4-15: Hour Percentage (P) with consideration of circulation hours of sensitivity test 1-1
Assumed hours without
thinking about circulation
time (W)
Assumed circulation hours
that takes up the time in the
space (C)
Real hour
(R)
Hour
percentage (P)
Dining 90.0 18.0 72.0 10%
Activity 420.0 84.0 336.0 45%
Apartment 199.0 39.8 159.2 21%
Dining Activity Apartment Office Care Circulation 1/People in
apartment
1/People in
activity space
1/People in
total
6:30-7:00 20 3 34 1 0 / 58 3% 33% 2%
7:00-7:30 20 20 17 1 0 / 58 6% 5% 2%
7:30-8:00 20 24 13 1 0 / 58 8% 4% 2%
8:00-8:30 0 40 17 1 0 / 58 6% 3% 2%
8:30-9:00 0 40 17 1 0 / 58 6% 3% 2%
9:00-9:30 0 42 18 3 0 / 63 6% 2% 2%
9:30-10:00 0 42 18 3 0 / 63 6% 2% 2%
10:00-10:30 0 42 18 3 0 / 63 6% 2% 2%
10:30-11:00 0 42 18 3 0 / 63 6% 2% 2%
11:00-11:30 0 42 18 3 0 / 63 6% 2% 2%
11:30-12:00 0 42 18 3 0 / 63 6% 2% 2%
12:00-12:30 20 29 12 2 0 / 63 8% 3% 2%
12:30-13:00 20 26 15 2 0 / 63 7% 4% 2%
13:00-13:30 20 28 13 2 0 / 63 8% 4% 2%
13:30-14:00 0 34 23 2 4 / 63 4% 3% 2%
14:00-14:30 0 44 13 2 4 / 63 8% 2% 2%
14:30-15:00 0 44 13 2 4 / 63 8% 2% 2%
15:00-15:30 0 44 13 2 4 / 63 8% 2% 2%
15:30-16:00 0 46 15 2 0 / 63 7% 2% 2%
16:00-16:30 0 46 15 2 0 / 63 7% 2% 2%
16:30-17:00 0 46 15 2 0 / 63 7% 2% 2%
17:00-17:30 20 32 9 2 0 / 63 11% 3% 2%
17:30-18:00 20 26 15 2 0 / 63 7% 4% 2%
18:00-18:30 20 16 21 2 0 / 59 5% 6% 2%
Occupied Hour (O) 6 12 12 12 2 / 44
Total hours 90.0 420.0 199.0 24.5 8.0 / 741.5
No circulation Hour
Percentage (P)
12% 57% 27% 3% 1% / 100%
Occupant spatial distribution Sensitivity test people moving
percentage
Time Total
88
Office 24.5 4.9 19.6 3%
Care 8.0 1.6 6.4 1%
Circulation 0 148.3 148.3 20%
(2) Move 5 people from apartment to activity space
Then, it was tested when 5 people move from apartment to activity space.
• Occupant spatial distribution without considering circulation hours (Table 4-16)
From the occupant spatial distribution without considering circulation hours, Occupies Hour (O) of each
function space, Hour Percentage (P) without thinking about circulation hours and the percentage of people
moving can be calculated.
Table 4-16: Occupant spatial distribution without thinking about circulation hours of sensitivity test 1-2
• Hour percentage (P) with consideration of circulation hours
Hour Percentage (P) with consideration of circulation hours is given (Table 4-17).
Dining Activity Apartment Office Care Circulation 5/People in
apartment
5/People in
activity
space
5/People in
total
6:30-7:00 20 7 30 1 0 / 58 17% 71% 9%
7:00-7:30 20 24 13 1 0 / 58 38% 21% 9%
7:30-8:00 20 28 9 1 0 / 58 56% 18% 9%
8:00-8:30 0 44 13 1 0 / 58 38% 11% 9%
8:30-9:00 0 44 13 1 0 / 58 38% 11% 9%
9:00-9:30 0 46 14 3 0 / 63 36% 11% 8%
9:30-10:00 0 46 14 3 0 / 63 36% 11% 8%
10:00-10:30 0 46 14 3 0 / 63 36% 11% 8%
10:30-11:00 0 46 14 3 0 / 63 36% 11% 8%
11:00-11:30 0 46 14 3 0 / 63 36% 11% 8%
11:30-12:00 0 46 14 3 0 / 63 36% 11% 8%
12:00-12:30 20 33 8 2 0 / 63 63% 15% 8%
12:30-13:00 20 30 11 2 0 / 63 45% 17% 8%
13:00-13:30 20 32 9 2 0 / 63 56% 16% 8%
13:30-14:00 0 38 19 2 4 / 63 26% 13% 8%
14:00-14:30 0 48 9 2 4 / 63 56% 10% 8%
14:30-15:00 0 48 9 2 4 / 63 56% 10% 8%
15:00-15:30 0 48 9 2 4 / 63 56% 10% 8%
15:30-16:00 0 50 11 2 0 / 63 45% 10% 8%
16:00-16:30 0 50 11 2 0 / 63 45% 10% 8%
16:30-17:00 0 50 11 2 0 / 63 45% 10% 8%
17:00-17:30 20 36 5 2 0 / 63 100% 14% 8%
17:30-18:00 20 30 11 2 0 / 63 45% 17% 8%
18:00-18:30 20 20 17 2 0 / 59 29% 25% 8%
Occupied Hour (O) 6 12 12 12 2 / 44
Total hours 90.0 468.0 151.0 24.5 8.0 / 741.5
Occupant spatial distribution Sensitivity test people moving
percentage
Time Total
89
Table 4-17: Hour Percentage (P) with consideration of circulation hours of sensitivity test 1-2
Assumed hours without
thinking about circulation
time (W)
Assumed circulation hours
that takes up the time in the
space (C)
Real hour
(R)
Hour
percentage
(P)
Dining 90.0 18.0 72.0 10%
Activity 468.0 93.6 374.4 50%
Apartment 151.0 30.2 120.8 16%
Office 24.5 4.9 19.6 3%
Care 8.0 1.6 6.4 1%
Circulation 0 148.3 148.3 20%
(3) Move 10 people from apartment to activity space
Next, it was tested when 10 people move from apartment to activity space.
• Occupant spatial distribution without considering circulation hours (Table 4-18)
From the occupant spatial distribution without considering circulation hours, Occupies Hour (O) of each
function space, Hour Percentage (P) without thinking about circulation hours and the percentage of people
moving can be calculated.
Table 4-18: Occupant spatial distribution without thinking about circulation hours of sensitivity test 1-3
Dining Activity Apartment Office Care Circulation 10/People in
apartment
10/People in
activity space
10/People in
total
6:30-7:00 20 12 25 1 0 / 58 40% 83% 17%
7:00-7:30 20 29 8 1 0 / 58 125% 34% 17%
7:30-8:00 20 33 4 1 0 / 58 250% 30% 17%
8:00-8:30 0 49 8 1 0 / 58 125% 20% 17%
8:30-9:00 0 49 8 1 0 / 58 125% 20% 17%
9:00-9:30 0 51 9 3 0 / 63 111% 20% 16%
9:30-10:00 0 51 9 3 0 / 63 111% 20% 16%
10:00-10:30 0 51 9 3 0 / 63 111% 20% 16%
10:30-11:00 0 51 9 3 0 / 63 111% 20% 16%
11:00-11:30 0 51 9 3 0 / 63 111% 20% 16%
11:30-12:00 0 51 9 3 0 / 63 111% 20% 16%
12:00-12:30 20 38 3 2 0 / 63 333% 26% 16%
12:30-13:00 20 35 6 2 0 / 63 167% 29% 16%
13:00-13:30 20 37 4 2 0 / 63 250% 27% 16%
13:30-14:00 0 43 14 2 4 / 63 71% 23% 16%
14:00-14:30 0 53 4 2 4 / 63 250% 19% 16%
14:30-15:00 0 53 4 2 4 / 63 250% 19% 16%
15:00-15:30 0 53 4 2 4 / 63 250% 19% 16%
15:30-16:00 0 55 6 2 0 / 63 167% 18% 16%
16:00-16:30 0 55 6 2 0 / 63 167% 18% 16%
16:30-17:00 0 55 6 2 0 / 63 167% 18% 16%
17:00-17:30 20 41 0 2 0 / 63 N/A 24% 16%
17:30-18:00 20 35 6 2 0 / 63 167% 29% 16%
18:00-18:30 20 25 12 2 0 / 59 83% 40% 17%
Occupied Hour (O) 6 12 12 12 2 / 44
Total hours 90.0 528.0 91.0 24.5 8.0 / 741.5
No circulation Hour
Percentage (P)
12% 71% 12% 3% 1% / 100%
Time Occupant spatial distribution Sensitivity test people moving percentage Total
90
• Hour percentage (P) with consideration of circulation hours
Hour Percentage (P) with consideration of circulation hours is given (Table 4-19).
Table 4-19: Hour Percentage (P) with consideration of circulation hours of sensitivity test 1-3
Assumed hours without
thinking about circulation
time (W)
Assumed circulation hours
that takes up the time in the
space (C)
Real hour
(R)
Hour
percentage
(P)
Dining 90.0 18.0 72.0 10%
Activity 528.0 105.6 422.4 57%
Apartment 91.0 18.2 72.8 10%
Office 24.5 4.9 19.6 3%
Care 8.0 1.6 6.4 1%
Circulation 0 148.3 148.3 20%
4.3.2 Test people move from apartment to office
The second part of the sensitivity test was to move people from apartment to office.
(1) Move one person from apartment to office
First, it was tested when 1 person move from apartment to office.
• Occupant spatial distribution without considering circulation hours (Table 4-20)
From the occupant spatial distribution without considering circulation hours, Occupied Hour (O) of each
function space, Hour Percentage (P) without thinking about circulation hours and the percentage of people
moving can be calculated.
91
Table 4-20: Occupant spatial distribution without thinking about circulation hours of sensitivity test 2-1
• Hour percentage (P) with consideration of circulation hours
Hour Percentage (P) with consideration of circulation hours is given (Table 4-21).
Table 4-21: Hour Percentage (P) with consideration of circulation hours of sensitivity test 2-1
Assumed hours without
thinking about circulation
time (W)
Assumed circulation hours
that takes up the time in the
space (C)
Real hour
(R)
Hour
percentage (P)
Dining 90.0 18.0 72.0 10%
Activity 408.0 81.6 326.4 44%
Apartment 199.0 39.8 159.2 21%
Office 36.5 7.3 29.2 4%
Care 8.0 1.6 6.4 1%
Circulation 0 148.3 148.3 20%
(2) Move 2 people from apartment to office
Then, it was tested when 2 people move from apartment to office.
• Occupant spatial distribution without considering circulation hours (Table 4-22)
From the occupant spatial distribution without considering circulation hours, Occupies Hour (O) of each
function space, Hour Percentage (P) without thinking about circulation hours and the percentage of people
moving can be calculated.
Dining Activity Apartment Office Care Circulation 1/People in
apartment
1/People in
office
1/People in
total
6:30-7:00 20 2 34 2 0 / 58 3% 50% 2%
7:00-7:30 20 19 17 2 0 / 58 6% 50% 2%
7:30-8:00 20 23 13 2 0 / 58 8% 50% 2%
8:00-8:30 0 39 17 2 0 / 58 6% 50% 2%
8:30-9:00 0 39 17 2 0 / 58 6% 50% 2%
9:00-9:30 0 41 18 4 0 / 63 6% 25% 2%
9:30-10:00 0 41 18 4 0 / 63 6% 25% 2%
10:00-10:30 0 41 18 4 0 / 63 6% 25% 2%
10:30-11:00 0 41 18 4 0 / 63 6% 25% 2%
11:00-11:30 0 41 18 4 0 / 63 6% 25% 2%
11:30-12:00 0 41 18 4 0 / 63 6% 25% 2%
12:00-12:30 20 28 12 3 0 / 63 8% 33% 2%
12:30-13:00 20 25 15 3 0 / 63 7% 33% 2%
13:00-13:30 20 27 13 3 0 / 63 8% 33% 2%
13:30-14:00 0 33 23 3 4 / 63 4% 33% 2%
14:00-14:30 0 43 13 3 4 / 63 8% 33% 2%
14:30-15:00 0 43 13 3 4 / 63 8% 33% 2%
15:00-15:30 0 43 13 3 4 / 63 8% 33% 2%
15:30-16:00 0 45 15 3 0 / 63 7% 33% 2%
16:00-16:30 0 45 15 3 0 / 63 7% 33% 2%
16:30-17:00 0 45 15 3 0 / 63 7% 33% 2%
17:00-17:30 20 31 9 3 0 / 63 11% 33% 2%
17:30-18:00 20 25 15 3 0 / 63 7% 33% 2%
18:00-18:30 20 15 21 3 0 / 59 5% 33% 2%
Occupied Hour (O) 6 12 12 12 2 / 44
Total hours 90.0 408.0 199.0 36.5 8.0 / 741.5
No circulation Hour
Percentage (P)
12% 55% 27% 5% 1% / 100%
Occupant spatial distribution Sensitivity test people moving
percentage
Time Total
92
Table 4-22: Occupant spatial distribution without thinking about circulation hours of sensitivity test 2-2
• Hour percentage (P) with consideration of circulation hours
Hour Percentage (P) with consideration of circulation hours is given (Table 4-23).
Table 4-23: Hour Percentage (P) with consideration of circulation hours of sensitivity test 2-2
Assumed hours without
thinking about circulation
time (W)
Assumed circulation hours
that takes up the time in the
space (C)
Real hour
(R)
Hour
percentage (P)
Dining 90.0 18.0 72.0 10%
Activity 408.0 81.6 326.4 44%
Apartment 187.0 37.4 149.6 20%
Office 48.5 9.7 38.8 5%
Care 8.0 1.6 6.4 1%
Circulation 0 148.3 148.3 20%
4.4 Summary
In this chapter, to explore the influence of analysis period on sDA and ASE, the simulation outcomes of
average fsDA and average fASE, as well as fsDA and fASE on each floor were given. The corresponding
sDA and ASE were given as well. To explore the occupied hierarchy of the function spaces, occupant spatial
distribution simulation under the three given schedules were shown. Occupied Hours (O) and Hour
Dining Activity Apartment Office Care Circulation 2/People in
apartment
2/People in
office
2/People in
total
6:30-7:00 20 2 33 3 0 / 58 6% 67% 3%
7:00-7:30 20 19 16 3 0 / 58 13% 67% 3%
7:30-8:00 20 23 12 3 0 / 58 17% 67% 3%
8:00-8:30 0 39 16 3 0 / 58 13% 67% 3%
8:30-9:00 0 39 16 3 0 / 58 13% 67% 3%
9:00-9:30 0 41 17 5 0 / 63 12% 40% 3%
9:30-10:00 0 41 17 5 0 / 63 12% 40% 3%
10:00-10:30 0 41 17 5 0 / 63 12% 40% 3%
10:30-11:00 0 41 17 5 0 / 63 12% 40% 3%
11:00-11:30 0 41 17 5 0 / 63 12% 40% 3%
11:30-12:00 0 41 17 5 0 / 63 12% 40% 3%
12:00-12:30 20 28 11 4 0 / 63 18% 50% 3%
12:30-13:00 20 25 14 4 0 / 63 14% 50% 3%
13:00-13:30 20 27 12 4 0 / 63 17% 50% 3%
13:30-14:00 0 33 22 4 4 / 63 9% 50% 3%
14:00-14:30 0 43 12 4 4 / 63 17% 50% 3%
14:30-15:00 0 43 12 4 4 / 63 17% 50% 3%
15:00-15:30 0 43 12 4 4 / 63 17% 50% 3%
15:30-16:00 0 45 14 4 0 / 63 14% 50% 3%
16:00-16:30 0 45 14 4 0 / 63 14% 50% 3%
16:30-17:00 0 45 14 4 0 / 63 14% 50% 3%
17:00-17:30 20 31 8 4 0 / 63 25% 50% 3%
17:30-18:00 20 25 14 4 0 / 63 14% 50% 3%
18:00-18:30 20 15 20 4 0 / 59 10% 50% 3%
Occupied Hour (O) 6 12 12 12 2 / 44
Total hours 90.0 408.0 187.0 48.5 8.0 / 741.5
No circulation Hour
Percentage (P)
12% 55% 25% 7% 1% / 100%
Occupant spatial distribution Sensitivity test people moving
percentage
Time Total
93
Percentage (P) were listed. The result of sensitivity tests under schedule 3 were provided. Hour Percentage
(P) and the percentage of people moving found in the sensitivity tests were given as well. The findings and
implications are summarized and discussed in Chapter 5 (Table 4-24).
Table 4-24: Findings and implications summary
Findings Implications Conclusion
From the visualized result meshes, it is
found that the value of fsDA and fASE
of each analysis point are close to each
other
analysis period may not
influence the value of sDA and
ASE significantly.
fsDA and fASE were not necessary to
describe the daylighting performance of
a function space in the senior living.
ΔxfsDA and ΔxfASE are around 2%,
which is small.
Occupied Hour (O) changes
significantly according to the
operation schedule.
The value of Occupied Hour (O)
of each function space is mostly
depend on Operation schedule
Occupied Hour (O) can be easily
controlled by the management group
and its value can be used to defined
“regularly occupied” space.
With higher portion of people moving,
the Hour Percentage (P) Change
becomes larger.
Hour Percentage (P) is
proportionally sensitive to
people spatial distribution
Research about the variation range of
occupant spatial distribution needs to be
carried out in the future to decide
whether Hour Percentage (P) can be
used to define occupied hierarchy
within “regularly occupied” function
spaces.
A standard on sDA and ASE of function spaces with different occupied level should be suggested by an institute such
as Illuminating Engineering Society. The requirement for sDA and ASE suggested by IES-LM-83-12 may renew
according to that. LEED may renew their lighting evaluation options according to the standard as well.
94
5. DISCUSSION
Chapter 5 discusses the result listed in chapter 4, including comparing fsDA with sDA, comparing fASE
with ASE, comparing occupant spatial distribution under the three schedules, sensitivity test and calling for
requirement on sDA and ASE for function spaces with different occupied levels (Figure 5-1). In this chapter,
the value of fsDA and fASE are compared with the value of sDA and ASE, to explore if the analysis period
influences the value of the two metrics with flexible analysis period.
As spaces where more people spend more time in owns higher priority to achieve good lighting condition,
the occupied hierarchy among function space are explored, Occupied Hour (O) and Hour Percentage (P) can
be the deciding factors if they are predictable. Occupied Hour (O) and Hour Percentage (P) of each of the
three schedules are compared with each other. Thus, whether the schedule can determine O and O can be
found so it can be decided whether O and P are predictable. Sensitivity tests are carried out for Hour
Percentage (P) to explore whether P is sensitive to occupant spatial distribution. Then Hour Percentage (P)
found in the two sensitivity tests, are compared with each other to find out if its value is sensitive to occupant
spatial distribution. Based on the findings above, it suggests whether Occupied Hour (O) and Hour
Percentage (P) can be the factors to decide the occupied level of a function space. Finally, a sample standard
on lighting performance of function spaces with different occupied levels is suggested. The findings and
implications are listed (Table 5-1).
Table 5-1: Findings and implications summary
Findings Implications Conclusion
From the visualized result meshes, it is
found that the value of fsDA and fASE
of each analysis point are close to each
other
analysis period may not
influence the value of sDA and
ASE significantly.
fsDA and fASE were not necessary to
describe the daylighting performance of
a function space in the senior living.
ΔxfsDA and ΔxfASE are around 2%,
which is small.
Occupied Hour (O) changes
significantly according to the
operation schedule.
The value of Occupied Hour (O)
of each function space is mostly
depend on Operation schedule
Occupied Hour (O) can be easily
controlled by the management group
and its value can be used to defined
“regularly occupied” space.
With higher portion of people moving,
the Hour Percentage (P) Change
becomes larger.
Hour Percentage (P) is
proportionally sensitive to
people spatial distribution
Research about the variation range of
occupant spatial distribution needs to be
carried out in the future to decide
whether Hour Percentage (P) can be
used to define occupied hierarchy
within “regularly occupied” function
spaces.
A standard on sDA and ASE of function spaces with different occupied level should be suggested by an institute such
as Illuminating Engineering Society. The requirement for sDA and ASE suggested by IES-LM-83-12 may renew
according to that. LEED may renew their lighting evaluation options according to the standard as well.
95
Figure 5-1: Chapter 5 methodology
5.1 Compare fsDA with sDA, and fASE with ASE
The value of fsDA and fASE are compared with the value of sDA and ASE to explore if there is significant
difference between their values (Figure 5-2). Then the conclusion whether analysis period influences the
value of sDA and ASE significantly can be reached.
96
Figure 5-2: 5.1 methodology
5.1.1 Compare sDA and fSDA
The absolute error of fsDA is calculated by the following formula.
Δx
fsDA
= 𝑓𝑠𝐷𝐴 𝑖 − 𝑠𝐷𝐴 𝑖
MAE
fsDA
=
1
𝑛 ∑ │(𝑓𝑠𝐷𝐴 𝑖 − 𝑠𝐷𝐴 𝑖 )│
𝑛 𝑖 =1
Where
Δx
fsDA
= The absolute errors of fsDA
n = The number of errors
MAE
fsDA
=Mean absolute error of fsDA
The analysis period for each schedule introduced in chapter 3 are listed (Table 5-2). Δx
fsDA
under the three
schedules are shown (Table 5-3). Δx
fsDA
are shaded by different colors according to their values (Figure 5-
4). There are 75 Δx
fsDA
. The absolute values of most of the Δx
fsDA
are less than 5%, which are slight changes
(Table 5-3). Daylight Autonomy (DA) is visualized by different colors. The values of average fsDA and the
fsDA of each floor of the whole building is given by the numbers (Figure 5-4, 5-5). From the visualized
Daylight Autonomy of the interior space, it can be found that there is no significant change in Daylight
Autonomy of each test point under different schedules as the result meshes are quite similar (Figure 5-4, 5-
5). The three biggest Δx
fsDA
are in care space under schedule 3 (Table 5-3). This can also be found in
visualized DA (Figure 5-6, 5-7).
In care space, schedule 3 has a much shorter operation hour than that under the fixed schedule. The operation
hour of care space under schedule 3 is about half an hour earlier than that of schedule 1 and schedule 2 (Table
5-2). The operation hour of care space under schedule 3 may be the time with relatively higher daylight
sufficiency among the day. Different operation hours can cause varied fsDA values. However, in this case,
the values do not change significantly. Therefore, for senior living, the type of building with regular and
relatively earlier operation schedules than normal buildings, analysis period may not influence the value of
sDA significantly.
Table 5-2: Operation schedules
Fixed Schedule Schedule 1 Schedule 2 Schedule 3
Total analysis period 8:00-18:00 8:00-19:00 7:00-19:00 6:30-18:30
Activity 8:00-18:00 8:00-19:00 7:00-19:00 6:30-18:30
Office 8:00-18:00 8:00-9:00
14:00-15:00
7:00-9:00
14:00-15:00
6:30-18:30
97
Circulation 8:00-18:00 8:00-19:00 7:00-19:00 6:30-18:30
Dining 8:00-18:00 8:00-10:00
12:00-14:00
17:00-19:00
7:00-9:00
12:00-14:00
17:00-19:00
6:30-8:00
12:00-13:30
17:00-18:30
Care 8:00-18:00 14:00-16:00 14:00-16:00 13:30-15:30
Apartment 8:00-18:00 8:00-19:00 7:00-19:00 6:30-18:30
Table 5-3: The absolute error of sDA
Function space Schedule Average 1st floor 2nd floor 3rd floor
The whole building
Fixed Schedule 0.0% 0.0% 0.0% 0.0%
Schedule 1 -1.1% -1.0% -1.2% -1.2%
Schedule 2 -2.4% -2.0% -2.4% -2.6%
Schedule 3 -2.8% -2.4% -2.7% -3.2%
Activity
Fixed Schedule 0.0% 0.0% 0.0% 0.0%
Schedule 1 -0.7% -0.5% -0.8% -0.8%
Schedule 2 -1.7% -1.1% -2.1% -2.3%
Schedule 3 -1.7% -1.1% -2.1% -2.5%
Office
Fixed Schedule 0.0% 0.0% 0.0% 0.0%
Schedule 1 1.1% 0.5% 0.0% 3.8%
Schedule 2 -1.1% -2.9% 0.0% 2.5%
Schedule 3 -4.1% -4.8% 0.0% -6.3%
Circulation
Fixed Schedule 0.0% 0.0% 0.0% 0.0%
Schedule 1 -0.9% -1.1% -1.0% -0.7%
Schedule 2 -2.1% -2.9% -1.8% -1.9%
Schedule 3 -2.1% -2.4% -1.9% -2.2%
Dining
Fixed Schedule 0.0% / 0.0% 0.0%
Schedule 1 -5.0% / -6.3% -3.8%
Schedule 2 0.3% / 0.6% -0.1%
Schedule 3 0.9% / 2.7% -0.9%
Care
Fixed Schedule 0.0% / 0.0% 0.0%
Schedule 1 0.7% / 2.2% -0.9%
Schedule 2 0.5% / 2.2% -1.2%
Schedule 3 9.9% / 9.5% 10.3%
Apartment
Fixed Schedule 0.0% / 0.0% 0.0%
Schedule 1 -1.4% / -1.1% -1.6%
Schedule 2 -3.9% / -3.1% -4.7%
Schedule 3 -4.6% / -3.9% -5.3%
0.0%<=│ Δx
fsDA
│<=5.0%
5.0%<│ Δx
fsDA
│<=10.0%
10.0%<│ Δx
fsDA
│
Figure 5-3: Legend of Δx
fsDA
98
Figure 5-4: Visualized DA and fsDA of the whole building under the fixed and the first schedule
99
Figure 5-5: Visualized DA and fsDA of the whole building under the second and the third schedule
100
Figure 5-6: Visualized DA and fsDA of care space under the fixed and the first schedule
101
Figure 5-7: Visualized DA and fsDA of care space under the second and the third schedule
102
5.1.2 Compare ASE and fASE
The absolute error of fASE is calculated by the following formula.
Δx
fASE
= 𝑓𝐴𝑆𝐸 𝑖 − 𝐴𝑆𝐸 𝑖
MAE
fASE
=
1
𝑛 ∑ │(𝑓𝐴𝑆𝐸 𝑖 − 𝐴𝑆𝐸 𝑖 )│
𝑛 𝑖 =1
When
Δx
fASE
= the absolute errors of fASE
n = the number of errors
MAE
fASE
=Mean absolute error of fASE
The analysis period for each schedule introduced in chapter 3 are listed (Table 5-4). Δx
fASE
under the three
schedules are shown (Table 5-5). Δx
fASE
are shaded by different colors according to their values (Figure 5-
8). There are 75 Δx
fASE
. The absolute values of most of the Δx
fASE
are less than 5%, which are slight changes
(Table 5-5). ASE is visualized by different colors. The meshes show sunlight exposure, the number of hours
that the test points have illuminance meet or exceed 1000 lux (Figure 5-9, 5-10). The value of average fASE
and the fASE of each floor of the whole building are given by the numbers (Figure 5-9, 5-10). From the
visualized ASE of the interior space, it can be found that there is no significant change on the value of each
test point under different schedules (Figure 5-11, 5-12).
The fASE of the dining area for all the 3 schedules have the absolute value of Δx
fASE
of at least 5%, and the
values are quite close to each other (Table 5-5). This can also be seen from the visualized ASE (Figure 5-11,
5-12). The 3 schedules have close dining hours while the time is quite different from that of the fixed
schedule (Table 5-4). Different operation hours can cause varied fASE values. However, in this case, the
values do not change significantly. Therefore, for senior living, the type of building with regular and
relatively earlier operation schedules than normal buildings, analysis period may not influence the value of
ASE significantly.
Table 5-4: Operation schedules
Fixed Schedule Schedule 1 Schedule 2 Schedule 3
Total analysis period 8:00-18:00 8:00-19:00 7:00-19:00 6:30-18:30
Activity 8:00-18:00 8:00-19:00 7:00-19:00 6:30-18:30
Office 8:00-18:00 8:00-9:00
14:00-15:00
7:00-9:00
14:00-15:00
6:30-18:30
Circulation 8:00-18:00 8:00-19:00 7:00-19:00 6:30-18:30
Dining 8:00-18:00 8:00-10:00
12:00-14:00
17:00-19:00
7:00-9:00
12:00-14:00
17:00-19:00
6:30-8:00
12:00-13:30
17:00-18:30
Care 8:00-18:00 14:00-16:00 14:00-16:00 13:30-15:30
Apartment 8:00-18:00 8:00-19:00 7:00-19:00 6:30-18:30
103
Table 5-5: Absolute error of fASE
Function space Schedule Average 1st floor 2nd floor 3rd floor
The whole building
Fixed Schedule 0.0% 0.0% 0.0% 0.0%
Schedule 1 0.6% 0.2% 0.7% 0.9%
Schedule 2 1.6% 2.0% 1.4% 1.5%
Schedule 3 1.2% 2.6% 0.8% 0.7%
Activity
Fixed Schedule 0.0% 0.0% 0.0% 0.0%
Schedule 1 1.4% 0.0% 4.1% 1.3%
Schedule 2 4.2% 5.0% 5.5% 1.6%
Schedule 3 3.4% 6.2% 1.6% 0.3%
Office
Fixed Schedule 0.0% 0.0% 0.0% 0.0%
Schedule 1 -4.0% -4.7% 0.0% -6.3%
Schedule 2 -3.4% -4.7% 0.0% -3.8%
Schedule 3 0.6% 0.0% 0.0% 2.5%
Circulation
Fixed Schedule 0.0% 0.0% 0.0% 0.0%
Schedule 1 0.5% 0.2% 0.6% 0.5%
Schedule 2 0.5% 0.2% 0.6% 0.5%
Schedule 3 0.1% 0.6% 0.0% 0.0%
Dining
Fixed Schedule 0.0% / 0.0% 0.0%
Schedule 1 -7.5% / -5.0% -10.0%
Schedule 2 -7.5% / -5.0% -10.0%
Schedule 3 -6.9% / -5.0% -8.8%
Care
Fixed Schedule 0.0% / 0.0% 0.0%
Schedule 1 -0.6% / 0.0% -1.2%
Schedule 2 -0.6% / 0.0% -1.2%
Schedule 3 -0.6% / 0.0% -1.2%
Apartment
Fixed Schedule 0.0% / 0.0% 0.0%
Schedule 1 0.4% / 0.5% 0.3%
Schedule 2 2.0% / 2.0% 1.9%
Schedule 3 1.9% / 1.8% 2.0%
0.0%<=│ Δx
fASE
│<=5.0%
5.0%<│ Δx
fASE
│<=10.0%
10%<│ Δx
fASE
│
Figure 5-8: Legend of Δx
fASE
104
Figure 5-9: Visualized fASE of the whole building under the fixed and the first schedule
105
Figure 5-10: Visualized fASE of the whole building under the second and the third schedule
106
Figure 5-11: Visualized fASE of the dining space under the fixed and the first schedule
107
Figure 5-12: Visualized fASE of the dining space under the second and the third schedule
108
5.1.3 Conclusion
The chart shows the distribution of the absolute error of fsDA(Δx
fsDA
) and fASE (Δx
fASE
) (Figure 5-13). Most
of the Δx
fsDA
are around -5% to 5%. Several of them are around 10%, which is much larger than the other
absolute errors. All the Δx
fASE
range from -10% to 10%. In general, there is no significant change in the
value of fsDA and fASE.
Figure 5-13 the absolute error of fsDA and fASE
The table counts the number of Δx
fsDA
and Δx
fASE
in different value ranges (Table 5-6). There are 75 Δx
fsDA
and 75 Δx
fASE
. 69 of the Δx
fsDA
are no more than 5% an0d 5 of them are no more than 10%. Only 1 of them
is larger than 10%. The changes of fASE are slight as well. 66 of the Δx
fASE
are no more than 5% and the
rest 9 of them are between 5% and 10%. The standard deviation of both Δx
fsDA
and Δx
fASE
are around 3%.
The mean absolute error (MAE) of fsDA and fASE are both around 2%. As a result, for the three schedules
in senior living, using a fixed schedule for sDA and ASE can generally describe the real lighting performance.
fsDA and fASE are not necessary for daylight performance description.
Table 5-6: The number of Δx
fsDA
and Δx
fASE
Δx ΔxfsDA ΔxfASE
│Δx│<=5% 69 66
5%<│Δx │<=10% 5 9
10%<│ Δx │ 1 0
Standard deviation of
absolute error
3.0% 3.3%
Mean absolute error
(MAE)
2.4% 2.2%
5. 2 Compare occupant spatial distribution under the three schedules
This section is to analyze the result of 4.2 (Figure 5-14). The Occupied Hour (O) and Hour percentage (P)
of the three schedules were compared to each other (Table 5-6). The more Occupied Hour (O) and Hour
percentage(P) the function space has, the more important the space is. Thus, the space is given more priority
to have good lighting performance.
-15.00%
-10.00%
-5.00%
0.00%
5.00%
10.00%
15.00%
sDA/ASE change
fsDA/fASE absolute error
ΔxfsDA
ΔxfASE
109
Figure 5-14: Methodology of 5.2
• Occupied Hour (O)
- Occupied Hour (O) can be controlled by operation schedule
Occupied Hour (O) can change significantly according to the operation schedule. For example, the occupied
hour of office under schedule 3 is 12, which is marked in red. It is much larger than that under schedule 1
and schedule 2 because the management group 3 want some of the staff keep staying in office during the
working time while the other 2 management groups try to make staff walk around to take care of the residents
(Table 5-7). Therefore, according to Occupied Hour (O), the office under schedule 3 can be regarded as
“regularly occupied space” while the space under schedule 1 and schedule 2 may not be a “regularly occupied
space”. Whatever, Occupied Hour (O) is predictable based on operation schedules and it is easy to control
Occupied Hour (O) by the building management group by adjusting the operation schedule.
• Hour Percentage (P)
- Operation schedule can influence Hour Percentage (P)
Operation schedule can influence the value of Hour Percentage (P) of each function space. For example,
schedule 2 try to make the residents stay in activity room rather than in their apartment so the Hour
Percentage (P) of the apartment in schedule 2 (18%, marked in red) is smaller than the other two schedules
(Table 5-7).
Table 5-7: Occupied Hour (O) and Hour Percentage (P) under the three schedules
Occupied Hour (O) Hour percentage (P)
Schedule 1 Schedule 2 Schedule 3 Schedule 1 Schedule 2 Schedule 3
Dining 6.0 6.0 6.0 11% 10% 10%
Activity 11.0 12.0 12.0 47% 50% 44%
Apartment 11.0 12.0 12.0 20% 18% 23%
Office 2.0 3.0 12.0 1% 1% 3%
110
Care 2.0 2.0 2.0 1% 1% 1%
Circulation 11 12 12 20% 20% 20%
- Precise value of Hour Percentage (P) is hard to predict
However, the precise value of Hour Percentage (P) can be hard to predict according to operation schedule as
occupant spatial distribution may influence the value of Hour Percentage (P). Occupant spatial distribution
cannot be fully controlled by the management group.
5.3 Sensitivity test
To test whether Hour Percentage (P) change significantly with different occupant spatial distribution,
sensitivity tests were carried on for P to occupant spatial distribution under schedule 3 (Figure 5-15).
Figure 5-15: methodology of 5.3
5.3.1 Sensitivity test result
As sensitivity test 1 moved people from apartment to activity area, the Hour Percentage (P) of activity space
and apartment changed. Sensitivity test 2 moved people from apartment to office. Thus, the Hour Percentage
(P) of apartment and office changed (Table 5-8).
Table 5-8: Hour Percentage Change
Original Hour
Percentage (P)
Test 1
part 1
Test 1
part 2
Test 1
part 3
Test 2
part 1
Test 2
part 2
Dining 10% 10% 10% 10% 10% 10%
Activity 44% 45% 50% 57% 44% 44%
Apartment 23% 21% 16% 10% 21% 20%
Office 3% 3% 3% 3% 4% 5%
Care 1% 1% 1% 1% 1% 1%
111
Circulation 20% 20% 20% 20% 20% 20%
5.3.2 Analyze the result of the sensitivity test: Hour Percentage Change
It is discussed how much Hour Percentage (P) changes when different number of people were moved. How
much percentage did the moved people took up of the total number of people in the senior living? To what
extend Hour Percentage (P) was sensitive to occupant spatial distribution change?
(1) Hour Percentage change and the number of people moving
1 people moved in test 1 part 1 and test 2 part 1. 2 people moved in test 2 part 2. 5 people moved in test 1
part 2. 10 people moved in test 1 part 3.
In this case, when 1 or two people move, the Hour Percentage (P) does not change significantly. When 5
people move from apartment to activity space, the Hour Percentage Change reaches 6%, which is not a slight
change. When the number of people moving becomes 10, the change reaches 13% (Table 5-9, Figure 5-16).
Table 5-9: Shaded Hour Percentage Change
Original Hour
Percentage (P)
Test 1 part
1
Test 1 part
2
Test 1 part
3
Test 2 part
1
Test 2 part
2
Dining 10% 0% 0% 0% 0% 0%
Activity 44%
1% 6% 13% 0% 0%
Apartment 23%
-1% -6% -13% -1% -3%
Office 3% 0% 0% 0%
1% 3%
Care 1% 0% 0% 0% 0% 0%
Circulation 20% 0% 0% 0% 0% 0%
Hour Percentage Change=New Hour percentage- Original Hour Percentage (P)
0%<=│Hour Percentage Change│<5%
5%<=│Hour Percentage Change│<10%
10%<=│Hour Percentage Change│<15%
Figure 5-16: The legend of result change
(2) Hour Percentage change and the percentage of people moving
As mentioned, when 1 or 2 people move, Hour Percentage (P) does not change much while when 5 or 10
people move, Hour Percentage (P) varies significantly. What percentage of the total people do 1, 2, 5 and 10
people take up? The percentage of moved people to the number of people in each function space changes
significantly, but it does not influence Hour Percentage Change. The percentage of moved people to the total
number of people in the senior living can influence the value of Hour Percentage Change. For schedule 3, 1
person takes up about 2% of the total number of people and 2 people accounts for 3% (Table 5-10). 2% and
3% are not significant percentage. However, 5 people comprises about 8% and 10 people occupies
approximately 16% of the total number of people (Table 5-10). These two percentages are not small amount.
Therefore, when a small percentage, such as 2% or 3% people move, Hour Percentage (P) does not change
much while when a larger percentage of people move, Hour Percentage (P) can vary more obviously.
Table 5-10: The percentage of people moving
Test Average Min Max Median
Test 1
1/People in apartment 6% 3% 11% 6%
1/People in activity space 4% 2% 33% 2%
112
1/People in total 2% 2% 2% 2%
5/People in apartment 45% 17% 100% 42%
5/People in activity space 16% 10% 71% 11%
5/People in total 8% 8% 9% 8%
10/People in apartment 159% 40% 333% 125%
10/People in activity space 26% 18% 83% 20%
10/People in total 16% 16% 17% 16%
Test 2
1/People in apartment 6% 3% 11% 6%
1/People in office 35% 25% 50% 33%
1/People in total 2% 2% 2% 2%
2/People in apartment 14% 6% 25% 13%
2/People in office 51% 40% 67% 50%
2/People in total 3% 3% 3% 3%
(3) Hour Percentage (P) is proportionally sensitive to occupant spatial distribution
As a result, Hour Percentage (P) is proportionally sensitive to occupant spatial distribution. When a small
percentage of the total number of people move, Hour Percentage (P) does not change much. However, when
larger percentage of the total number of people move, Hour Percentage Change becomes larger. Therefore,
Hour Percentage can be proportionally sensitive to occupant spatial distribution.
5.3.3 Suggest a sample occupied hierarchy for the space
The functions spaces were given different occupied levels based on Occupied Hour (O) and Hour Percentage
(P). First, “regularly occupied” spaces were defined according to Occupied Hour (O). As Hour Percentage
(P) is proportionally sensitive to occupant spatial distribution, the variation range of occupant spatial
distribution should be explored in the future. If the variation range is small, P can be used to decide the more
detailed occupied hierarchy within “regularly occupied” space. If the variation range is large, there is no
more detailed occupied hierarchy within “regularly occupied” space.
•Define “regularly occupied” space
Based on current result, as Occupied Hour (O) can be easily controlled by management group, “regularly
occupied” space can be defined according to Occupied Hour (O) (Table 5-11). But the specific threshold
needs to be set up after doing more research about lighting and health.
Table 5-11: Sample definition of “regularly occupied space”
Space hierarchy
Definition
Regularly occupied space A space which is occupied for at least n hours a day.
•Define detailed occupied hierarchy within “regularly occupied” space
Within “regularly occupied” space, there may be an occupied hierarchy defined based on Hour Percentage
(P). The higher Hour Percentage the space has, the higher occupied level the space has. However, Hour
113
Percentage (P) can be hard to predict because it can change with different occupant spatial distribution. As
Hour Percentage (P) is proportionally sensitive to occupant spatial distribution, Hour Percentage (P) may be
a factor to decide occupied hierarchy if the spatial distribution change is slight. However, if the spatial
distribution change is significant, Hour Percentage (P) may not be a suitable metric to define the occupied
level of a function space.
Thus, more survey should be done in the future to find out the variation range of occupant spatial distribution.
If the variation range of occupant spatial distribution is not significant, Hour Percentage (P) can be used to
define the occupied level of a function space. Thus, within “regularly occupied space”, the occupied level of
each function space will be defined according to Hour Percentage (P). A sample occupied hierarchy is given
(Table 5-12).
Table 5-12: Sample occupied hierarchy
Occupied level
Hour Percentage (P)
1 …
2 …
3 …
4 …
… …
… …
… …
5.4 Call for requirement on sDA and ASE for function spaces with different occupied levels
As spaces with higher importance have more priority to get better daylighting performance, the requirement
on daylighting performance can be higher on function spaces with higher occupied level. A basic daylighting
requirement is necessary for “regularly occupied” spaces (Figure 5-17). Based on that, more detailed
requirements are given on functions spaces with different occupied levels.
Figure 5-17: 5.4 methodology
114
• Use the metric of sDA and ASE
For this case study, it is found that the value of sDA and ASE do not change much with varied operation
schedule. Therefore, the new metric fsDA and fASE are not necessary for daylight performance description,
at least for this case. SDA and ASE can be still used to evaluate the daylight performance of the senior living.
• Daylighting requirement for “regularly occupied” space
“Regularly occupied” space may meet the minimum requirement of LES-LM-83:
sDA may achieve Nominally Accepted Daylight Sufficiency and ASE may achieve satisfactory visual
comfort (IES Daylight Metrics Committee 2012) (Table 5-13). As the value for Nominally Accepted
Daylight Sufficiency and satisfactory visual comfort were defined based on common working space, the
values for assisted living may not be the same. The specific values should be decided after more research
and literature review.
Table 5-13: Sample reference for sDA and ASE for “regularly occupied function
Reference for sDA and ASE for “regularly occupied function
sDA300,50%>=55%
ASE1000,250h<= 10%
• Daylighting requirement for function spaces with different occupied levels within “regularly
occupied” spaces
Within “regularly occupied” spaces, the functions spaces with different occupied levels need different
daylighting requirements. It is found that sDA and ASE can describe the daylighting condition of senior
living generally correctly. Therefore, sDA and ASE are used to describe the daylighting performance of
senior living. The specific value of Hour Percentage (P), sDA requirement and ASE requirement need to be
defined after more research is done.
Within “regularly occupied” space, the function spaces may need to meet the following requirement based
on occupied hierarchy. A sample requirement on sDA and ASE for function spaces with different occupied
levels are given (Table 5-14). The specific number should be decided after doing more research.
Table 5-14: Sample sDA and fASE requirement for function spaces with different occupied levels
Occupied level Hour Percentage (P) sDA requirement ASE requirement
1 …
… …
2 …
… …
3 …
… …
4 …
… …
… …
… …
… …
… …
…
5.5 Summary
In summary, after comparing the values of fsDA and fASE to sDA and ASE, from the visualized result
meshes, it was found the value of each test point does not change much. It was found that Δx
fsDA
and Δx
fASE
are around 2%, which is small. Thus, fsDA and fASE do not change significantly with varied analysis period.
As sDA and ASE are easier to calculate and the result is close to that of fsDA and fASE, fsDA and fASE
were not necessary to describe the daylighting performance of a function space in the senior living.
As the value of Occupied Hour (O) changes significantly with varied operation schedule, O depends on
operation schedules significantly. Therefore, O can be easily controlled by the management group and its
value can be used to defined “regularly occupied” space. With higher portion of people moving, the Hour
Percentage (P) Change becomes larger. Thus, Hour Percentage (P) is proportionally sensitive to occupant
spatial distribution. Research about the variation range of occupant spatial distribution needs to be carried
115
out in the future to decide whether Hour Percentage (P) can be used to define occupied hierarchy within
“regularly occupied” function spaces. “Regularly Occupied” space should be defined as a space which is
occupied for at least n hours a day. A sample standard on sDA and ASE of “regularly occupied” space and
that for function spaces with different occupied level was suggested. A formal standard should be presented
in the future by an institute such as Illuminating Engineering Society. The requirement for sDA and ASE
suggested by IES-LM-83-12 may renew according to that. LEED may renew their lighting evaluation options
according to the standard as well. The findings and implications are summarized (Table 5-15).
Table 5-15: Findings and implications
Findings Implications Conclusion
From the visualized result meshes, it is
found that the value of fsDA and fASE
of each analysis point are close to each
other
analysis period may not
influence the value of sDA and
ASE significantly.
fsDA and fASE were not necessary to
describe the daylighting performance of
a function space in the senior living.
ΔxfsDA and ΔxfASE are around 2%,
which is small.
Occupied Hour (O) changes
significantly according to the
operation schedule.
The value of Occupied Hour (O)
of each function space is mostly
depend on Operation schedule
Occupied Hour (O) can be easily
controlled by the management group
and its value can be used to defined
“regularly occupied” space.
With higher portion of people moving,
the Hour Percentage (P) Change
becomes larger.
Hour Percentage (P) is
proportionally sensitive to
people spatial distribution
Research about the variation range of
occupant spatial distribution needs to be
carried out in the future to decide
whether Hour Percentage (P) can be
used to define occupied hierarchy
within “regularly occupied” function
spaces.
A standard on sDA and ASE of function spaces with different occupied level should be suggested by an institute such
as Illuminating Engineering Society. The requirement for sDA and ASE suggested by IES-LM-83-12 may renew
according to that. LEED may renew their lighting evaluation options according to the standard as well.
116
6. CONCLUSION AND FUTURE WORK
This chapter presents conclusions based on the work that has been done and suggests the fields that can be
explored in the future.
6.1 Conclusions
Daylighting is important to interior space. IES-LM-83-12 suggested two metrics, spatial Daylight Autonomy
(sDA) and Annual Sunlight Exposure (ASE) to describe annual daylighting performance of interior space.
SDA indicates the annual sufficiency of ambient daylight levels in interior space and ASE expresses the
possibility for visual discomfort (glare) in interior space. However, the measurement scope for the two
metrics are limited to common working space. As they are useful in daylighting performance description,
the scope may be extended to different building types such as assisted living. However, different buildings
types may have different operation schedules. The two metrics have an analysis period fixed to 8:00 am to
6:00 pm. They lacked the consideration of real operation schedule. LEED borrowed the idea of sDA and
ASE for its daylighting rating system. There are also another two options for LEED daylighting evaluation,
but none of the three options considers operation schedules. Therefore, flexible analysis period spatial
Daylight Autonomy (fsDA) and flexible analysis period Annual Sunlight Exposure (fASE), two new metrics
based on sDA and ASE, were proposed to describe the daylighting performance of interior spaces. Assisted
living was used as the case study to explore daylighting and schedules because of its regular and early
schedule.
There were three main parts of the methodology: preparation, exploring the influence of analysis period on
sDA and ASE and exploring the approach to define occupied hierarchy of the function spaces (Figure 6-1).
The preparation included building the Rhino model of the case building, dividing the case building into
different function spaces and doing research about the operation schedules of assisted living (Figure 6-1).
117
Figure 6-1: Methodology diagram
Part 1 explored the influence of analysis period on sDA and ASE value to find out whether fsDA and fASE
were useful in describing daylight performance of an interior space (Figure 6-1). First, sDA and ASE with
the fixed analysis period were calculated for each function space. Then the fsDA and fASE with the analysis
period based on the real operation schedule were calculated. Next, the two results were compared with each
other. As a result, the differences between the values of the metrics with flexible analysis period and that
with the fixed analysis period were not significant (Figure 6-2). It was found that the standard deviation of
ΔxfsDA and ΔxfASE were around 3% and the mean absolute error for fsDA and fASE were around 2%. It
is easier to calculate sDA and ASE than calculating fsDA and fASE. Therefore, fsDA and fASE are not
necessary for describing the daylight performance of an interior space. SDA and ASE are useful in
daylighting performance evaluation.
Because the spaces where more people spend more time in should have more priority to provide good
daylight to occupants, the importance of each function space was discussed in part 2. The second part was
to demonstrate that the occupied hierarchy of interior space might be defined based on the real Occupied
Hours (O) and Hour Percentage (P), which were calculated based on occupant spatial distribution simulation
results (Figure 6-1).
The occupant spatial distribution simulation results were simulated according to the three operation
118
schedules. If the occupied hierarchy of interior space can be predicted based on the real Occupied Hours (O)
and Hour Percentage (P), O and P must be predictable according to the known operation schedules. To prove
that the occupied hierarchy can be predicted by O and P, it was tested if the values of the two metrics were
decided by operation schedules through finding their values under the three different operation schedules. It
was found that Occupied Hour (O) was decided by the operation schedule because under the same operation
schedule, the Occupied Hour (O) of the same function space remain the same. Under different schedules for
the same function space, Occupied Hour (O) changes. Therefore, Occupied Hour (O) can be used to define
“regularly occupied” space. Hour Percentage (P) could be influenced by both operation schedule and
occupant spatial distribution. Therefore, whether Hour Percentage (P) can decide occupied hierarchy within
“regularly occupied” space need to be decided after a sensitivity test. The sensitivity test explored if Hour
Percentage (P) was sensitive to occupant spatial distribution by calculating the value of the metric under the
same operation schedule while with different occupant spatial distributions (Figure 6-1). It was found that
Hour Percentage (P) was proportionally sensitive to occupant spatial distribution as when more portion of
occupants move, the larger Hour Percentage Changes were found. As a result, additional research about the
variation range of occupant spatial distribution needs to be carried out in the future to decide whether Hour
Percentage (P) can be used to define occupied hierarchy within “regularly occupied” function spaces.
In conclusion, sDA and ASE are currently used in evaluating daylight performance, and fsDA and fASE are
not necessarily more useful. Occupied Hours (O) can be used to define “regularly occupied” space while
whether Hour Percentage (P) can be used to define the occupied hierarchy within “regularly occupied” space
is not decided yet. “Regularly Occupied” space should be defined as a space which is occupied for at least n
hours a day. A sample reference on sDA and ASE for “regularly occupied function spaces was suggested
according to IES-LM-83-12 (Table 6-1). The specific values should be decided after doing more research
and literature review for assisted living.
Table 6-1: Sample reference for sDA and ASE for “regularly occupied function
Reference for sDA and ASE for “regularly occupied function
sDA300,50%>=55%
ASE1000,250h<= 10%
If the variation range of occupant spatial distribution is small, Hour Percentage (P) can be used to decide the
occupied hierarchy within “regularly occupied” space. Thus, a reference on sDA and ASE of function spaces
with different occupied levels can be suggested in the future by an institute such as Illuminating Engineering
Society. The requirement for sDA and ASE suggested by IES-LM-83-12 may renew according to that. LEED may
renew their lighting evaluation options according to the standard as well (Table 6-2). The building management
group can set up a schedule that makes more people spend more time in space with good daylight condition
so the standard can be satisfied. The conclusions were summarized (Table 6-3).
Table 6-2: Sample sDA and fASE requirement for function spaces with different occupied levels
Occupied level Hour Percentage (P) sDA requirement ASE requirement
1 …
… …
2 …
… …
3 …
… …
4 …
… …
… …
… …
… …
… …
…
Table 6-3: Findings and implications
Findings Implications Conclusion
From the visualized result meshes, it is
found that the value of fsDA and fASE
of each analysis point are close to each
other
Analysis period may not
influence the value of sDA
and ASE significantly.
fsDA and fASE were not necessary to
describe the daylighting performance of
a function space in the senior living case
study.
119
ΔxfsDA and ΔxfASE are around 2%,
which is small.
Occupied Hour (O) changes
significantly according to the
operation schedule.
The value of Occupied Hour (O)
of each function space mostly
depends on operation schedule
Occupied Hour (O) can be easily
controlled by the management group,
and its value can be used to defined
“regularly occupied” space.
With higher portion of people
moving, the Hour Percentage (P)
Change becomes larger.
Hour Percentage (P) is
proportionally sensitive to
people spatial distribution.
Additional research about the variation
range of occupant spatial distribution
needs to be carried out in the future to
decide whether Hour Percentage (P)
can be used to define occupied
hierarchy within “regularly occupied”
function spaces.
A standard on sDA and ASE of function spaces with different occupied level should be suggested by an institute
such as Illuminating Engineering Society. The requirement for sDA and ASE suggested by IES-LM-83-12 may
renew according to that. LEED may renew their lighting evaluation options according to the standard as well.
6.2 Future work
Future work includes expanding sample quantity and variety, recording the occupants’ number in different
function spaces in site, using new technics to detect occupied status, simulating sDA in higher quality,
exploring the variation range of Hour Percentage (P), defining occupied hierarchy, and discovering
suggested sDA and ASE for function spaces with different occupied levels. The relationship between
circadian lighting and occupied hierarchy can be considered in further study as well.
6.2.1 Expand sample quantity and variety
Only one assisted living and three schedules were tested. The results may not be convincing enough with
limited sample quantity as special conditions may occur with limited sample number. Therefore, more case
studies of different buildings with different operation schedules should be tested. These buildings can be in
different styles in different locations all over the world. The unit size can be one factor to influence the time
people spend in the unit or the common area (Regnier, Senior living related issues 2019). Therefore, take
different unit size into consideration is essential. The operation schedules can be borrowed from more
different assisted livings in the same region of the case building.
6.2.2 Record the occupant’s number in different function spaces on site
As the occupant spatial distribution is estimated according to the operation schedule, the result may not be
precise. Therefore, spend several whole days in different assisted living and record the actual number of
people (residents, staff and visitors) in each common space can help to get a general idea of when and where
people are. This can prove whether the occupant spatial distribution based on operation schedule is correct.
6.2.3 Use new technics to detect occupied status
The occupied status, including the status of occupied or unoccupied as well as the number of people of each
function space over time was estimated based on the result of occupant spatial distribution simulation. The
simulation is not the real condition. Thus, the simulation result may not follow the real occupant spatial
distribution. In-home monitoring system (IMS), which consists of wireless passive infrared motion sensors
installed in every room, can be used to detect the occupied status of each function space (Virone, et al. 2008).
By this means, the real occupant spatial distribution can be recorded so Occupied Hour (O) and Hour
Percentage (P) can be more precise based on this than the simulated occupant spatial distribution.
6.2.4 Simulate sDA in higher quality
The simulation quality for sDA was low. The Radiance quality was 0, which meant low. The ambient
bounces for radiance was 3. The simulation quality was not very high so the result may not be very accurate.
The quality was not very high because higher quality took longer time to calculate. With more time for
120
computer simulations, the analysis quality could be more convincing.
6.2.5 Explore the variation range of Hour Percentage (P)
It was found that Hour Percentage (P) was proportionally sensitive to occupant spatial distribution. Therefore,
whether or not the Hour Percentage (P) can be used to decide the occupied hierarchy within “regularly
occupied” space is not yet decided. The variation range of Hour Percentage (P) should be explored more
thoroughly in the future. If the variation range is small, P can be used to decide the occupied hierarchy.
Otherwise, no detailed occupied hierarchy within “regularly occupied” space can be predicted. The variation
range can be explored by comparing the Hour Percentage (P) under more different schedules in different
case buildings all over the world. The in-home monitoring system (IMS) can be used to detect the real
occupied status of each space.
6.2.6 Define occupied hierarchy
It was suggested that “regularly occupied” space can be defined based on the value of Occupied Hour (O)
while the specific value is not defined. Literature review should be more thorough to determine what other
researchers think what specific hour to define a “regularly occupied” hour in the future. If Hour Percentage
(P) can be used to define the detailed occupied hierarchy within the “regularly occupied” space, the specific
value for the Hour Percentage (P) for function spaces with different occupied level might be decided by
doing a literature review to see if more information is available on this.
6.2.7 Find out suggested sDA and ASE for function space with different occupied levels
The suggested value of sDA and ASE should be defined for “regularly occupied” in assisted living after
doing more literature review or study. If there is a detailed occupied hierarchy within the “regularly occupied”
space, the sDA and ASE for the function spaces with different occupied levels might be suggested by doing
more literature review or more original research can be done on this specific topic.
6.2.8 Circadian lighting and occupied hierarchy
Circadian lighting is a lighting system that tap into human bodies every day. The color, angle and position
of natural sunlight at any given time can influence human body (Figure 6-2) (Herman 2016). As a result,
human body follows an approximately 24-hour rhythm, which is known as “circadian rhythms” (Herman
2016). A circadian lighting system to mimic the natural light progression over time can be designed to
maximize people’s functionality (Konis 2017). Without a proper circadian system, the space can be regarded
as biologically dark, contributing to the disruption of biology clock (Konis 2017). The Equivalent Melanopic
Lux (EML) is a measurement of circadian lighting effect (Delos Living LLC 2014). WELL Building
Standard is a performance -based system for measuring, certifying, and monitoring features of the built
environment that impact human health and wellbeing, through light, water, air, nourishment, comfort, fitness
and mind (Knox 2015). According to WELL Building Standard’s Circadian Lighting Design precondition
(option 1), the minimum threshold for circadian lighting is 250 EML and it has to be available for at least 4
hour per day within at least 75% of workstation on the vertical plane facing forward 1.2 m (4ft) above
finished floor (Delos Living LLC 2014). Standard can be met with daylight, electrical light, or a combination
of both light sources (Konis 2017). By meeting this standard, the interior space helps occupants to reinforce
the natural patterns of circadian cycle (Delos Living LLC 2014).
Circadian lighting can significantly influence people’s daily schedules. Therefore, whether the circadian
lighting condition is good for each function space of an assisted living can be explored. As the circadian
lighting condition can change among different seasons, the study will be carried out for each season and each
hour individually. The idea that spaces where more people spend more time in have more priority to achieve
good lighting condition can be applied to the study as well. The circadian lighting of each function space in
an assisted living over time are calculated. Then calculate the hours of circadian lighting which is over 250
EML for over 75% area of each function space. After that, it can be discussed whether the more important
spaces get more hours with good circadian lighting conditions (at least 250 EML and over 75% area).
121
Figure 6-2: The natural light changes in color, angle and intensity with the day progressing (Herman 2016).
6.3 Summary
Daylight is important for senior living facilities. Spatial Daylight Autonomy (sDA) and Annual Sunlight
Exposure (ASE) were two important metrics suggested by IES-LM-83-12 to describe daylight performance
of interior space. However, they are originally intended to measure common working space. If they are
applied to assisted living, the fixed analysis period may lead to an inaccurate result as assisted living has a n
earlier operation schedules. Simulations with different operation schedules were done to see if the flexible
analysis period Spatial Daylight Autonomy (fsDA) and flexible analysis period Annual Sunlight Exposure
(fASE) could replace sDA and ASE. A Rhino model of an assisted living facility was built, and the interior
spaces were divided into different function spaces. Three operation schedules for senior living were
investigated. By comparing the value of fsDA and fASE with sDA and ASE of each function spaces under
different operation schedules, it was found that there is no significant change between the values of the
metrics with fixed and flexible analysis period. Therefore, fsDA and fASE were not necessary for daylight
description in an assisted living. As the space where more people spend more time in have priority to achieve
better daylight condition, the occupied hierarchy among function spaces was explored by discussing if the
hierarchy was predictable by knowing the operation schedules. Occupied Hour (O) and Hour Percentage (P)
under different operation schedules were calculated and compared with each other. It was found that
Occupied Hour (O) was predictable based on the operation schedules. Thus, “regularly occupied” space can
be predicted based on Occupied Hour (O). The “regularly occupied” space can be defined as a space which
is occupied for at least n hours a day. The specific value for n needs to be decided based on more literature
review and research. The basic sDA and ASE requirement suggested by IES-LM-83-12 was used as an
example for “regularly occupied” space (Table 6-4). More research and literature review for assisted living
needs to be done to define the specific values.
Table 6-4: Sample reference for sDA and ASE for “regularly occupied function
Reference for sDA and ASE for “regularly occupied function
sDA300,50%>=55%
ASE1000,250h<= 10%
However, operation schedule is not the only factor to influence Hour Percentage. Hour Percentage (P) could
be influenced by occupant spatial distribution as well. Then a sensitivity test was carried out to see if P
changes significantly with different people spatial distribution while under the same schedule. It was found
that P was proportionally sensitive to occupant spatial distribution, which means the changes of P becomes
higher when more portion of occupants move. Therefore, more research about the variation range of the
portion people move should be carried out in the future. If the variation range is small, P can be used to
define the occupied hierarchy within “regularly occupied” space, otherwise, it could be hard to predict the
occupied hierarchy within “regularly occupied” space. If P can be used to define the occupied hierarchy,
there can be a reference on sDA and ASE for function spaces with different occupied levels. A sample sDA
and ASE requirement was suggested while the specific values of P, sDA and ASE need to be decided after
doing more literature review or failing in that, additional original research (Table 6-5). The building
122
management group can set up a schedule which makes more people spend more time in space with good
daylight condition so the standard can be satisfied. The conclusions were summarized in Table 6-6.
Table 6-5: Sample sDA and fASE requirement for function spaces with different occupied levels
Occupied level Hour Percentage (P) sDA requirement ASE requirement
1 …
… …
2 …
… …
3 …
… …
4 …
… …
… …
… …
… …
… …
…
Table 6-6: Findings and implications
Findings Implications Conclusion
From the visualized result meshes, it is
found that the value of fsDA and fASE
of each analysis point are close to each
other
The analysis period may not
influence the value of sDA
and ASE significantly.
fsDA and fASE were not necessary to
describe the daylighting performance of
a function space in the senior living.
ΔxfsDA and ΔxfASE are around 2%,
which is small.
Occupied Hour (O) changes
significantly according to the
operation schedule.
The value of Occupied Hour (O)
of each function space is mostly
depend on Operation schedule.
Occupied Hour (O) can be easily
controlled by the management group
and its value can be used to defined
“regularly occupied” space.
With higher portion of people
moving, the Hour Percentage (P)
Change becomes larger.
Hour Percentage (P) is
proportionally sensitive to
people spatial distribution.
Research about the variation range of
occupant spatial distribution needs to be
carried out in the future to decide
whether Hour Percentage (P) can be
used to define occupied hierarchy
within “regularly occupied” function
spaces.
A standard on sDA and ASE of function spaces with different occupied level should be suggested by an institute
such as Illuminating Engineering Society. The requirement for sDA and ASE suggested by IES-LM-83-12 may
renew according to that. LEED may renew their lighting evaluation options according to the standard as well.
Some additional work can be done in the future to improve the research. The sample quantities, including
the quantity and variety of case buildings and operation schedules, can be expanded. Recording occupant
spatial distribution on site can prove whether the occupant spatial distribution simulation result is correct.
New technics, in-home monitoring system (IMS), can be used to detect the occupied status of spaces. The
sDA simulation quality can be higher. The variation range of Hour Percentage (P) need to be explored to
decide whether Hour Percentage (P) can be the factor to define the detailed occupied hierarchy among
function spaces. The specific value of Occupied Hour (O) and Hour Percentage (P) to define occupied
hierarchy should be decided. The suggested sDA and ASE for function spaces with different occupied levels
should be figured out in the future. The relationship between circadian lighting and the occupied hierarchy
among function spaces can be explored in the future. By doing more research about daylighting in senior
living facilities, the health of older adults will be benefit.
123
See Appendix A for Visualized fsDA and fASE.
See Appendix B for Operation schedules.
124
APPENDIX A
125
126
127
128
129
130
131
132
133
134
135
136
137
138
APPENDIX B
139
REFERENCES
Alonso, Nono Martí nez. 2014. 4 Reasons to Use Grasshopper in Architecture. 2 10. Accessed 12 30, 2018.
http://nono.ma/says/4-reasons-to-use-grasshopper-in-architecture.
Arch Daily. 2018. Arch Daily. 3 11. Accessed 8 1, 2018. https://www.archdaily.com/889737/retirement-and-
nursing-home-wilder-kaiser-srap-sedlak-rissland-plus-durschinger-architekten.
Astrobob. 2009. The most Radiant Time of Year. 12 7. Accessed 11 3, 2018.
https://astrobob.areavoices.com/2009/12/07/the-most-radiant-time-of-year/.
Boubekri, Mohamed. 2008. Daylighting, Architecture and Health- Building Design Strategies. Oxford:
Elsevier.
Bristolite Team. 2014. 7 Architectural and Interior Design Benefits of Using Daylighting Systems. 4 8.
Accessed 3 1, 2019. http://www.bristolite.com/blog/7-architectural-and-interior-design-benefits-
of-using-daylighting-systems/.
Computer Hope. 2018. Plugin. 11 13. Accessed 2 1, 2019.
https://www.computerhope.com/jargon/p/plugin.htm.
David Balber Architect. n.d. What is an architectural program? Accessed 9 5, 2018.
http://balberarchitect.com/architectural-program/.
Delos Living LLC. 2014. "WELL Building Standard."
European Energy Efficiency Platform. n.d. What is a Typical Meteorological Year (TMY)? Accessed 3 15,
2019. https://e3p.jrc.ec.europa.eu/articles/typical-meteorological-year-tmy.
Herman, Jake. 2016. Working by Daylight: How Circadian Lighting Increases Productivity. 5 3.
https://www.glumac.com/circadian-lighting/.
IES Daylight Metrics Committee. 2012. IES LM-83-12. New York: the Illuminating Engineering Society of
North America.
IES Lighting for the Elderly and Partially Sighted Committee. 2016. ANSI/IES RP-28-16`. New York: the
Illuminating Engineering Society of North America.
Illuminating Engineering Society. n.d. About IES Standards. Accessed 2 1, 2019.
https://www.ies.org/standards/.
Jrh.main. 2018. Illuminance. 5 25. Accessed 1 20, 2019.
https://commons.wikimedia.org/wiki/File:Illuminance_Diagram.tif.
Knowledge Base DIALux 4. 2016. Illuminance (E)- Definition. 6 29. Accessed 12 2, 2018.
https://dialux4.support-en.dial.de/support/solutions/articles/9000078069-illuminance-e-definition.
Knox, Nora. 2015. What is a WELL? 4 2. Accessed 4 17, 2019. https://www.usgbc.org/articles/what-well.
Konis, Kyle. 2017. "A novel circadian daylight metric for building design and evaluation." Building and
Environment 113: 22-38.
https://www.sciencedirect.com/science/article/pii/S0360132316304498?via%3Dihub.
Mackey, Chris. 2017. Calculate ASE for LEED. 11 27. Accessed 12 2, 2018.
http://hydrashare.github.io/hydra/viewer?owner=chriswmackey&fork=hydra_2&id=Calculate_AS
E_for_LEED&slide=0&scale=1&offset=0,0.
Manager of Belmont Village, interview by Ruoxiao Zeng. 2019. Belmont village schedule (1 29).
Manager of Sunrise of Beverly Hills, interview by Ruoxiao Zeng. 2019. Sunrise of Beverly Hills schedule
(2 11).
Max. 2012. An Overview of Daylighting Metrics, with Examples. 5. Accessed 2 26, 2018.
https://www.maxpierson.me/2012/05/19/an-overview-of-daylighting-metrics-with-examples/.
Paulwintour. 2016. Ladybug & Honeybee. 3 13. Accessed 3 20, 2019.
https://parametricmonkey.com/2016/03/13/ladybug-honeybee/.
Regnier, Victor. 2002. Design for Assisted Living- Guidelines for Housing the Physically and Mentally Frail.
New York: John Wiley & Sons.
Regnier, Victor, interview by Ruoxiao Zeng. 2019. Senior living related issues (4 20).
Regnier, Victor, interview by Ruoxiao Zeng. 2019. Senior living schedule (1 25).
Reilly, Chris. 2016. What is Grasshopper. 9 26. Accessed 2 3, 2019. https://www.lynda.com/Grasshopper-
tutorials/What-Grasshopper/174491/194087-4.html.
Reinhart, Christophn F., John Mardaljevic, and ZAck Rogers. 2006. "Dynamic Daylight Performance
Metrics for." LEUKOS-The Journal of the Illuminating Engineering Society of North America 3 (1):
7-31. https://ies-tandfonline-
com.libproxy1.usc.edu/doi/abs/10.1582/LEUKOS.2006.03.01.001#.XKalAFxKiUk.
Rensselaer Polytechnic Institude. 2007. What is glare. 2. Accessed 1 25, 2018.
140
https://www.lrc.rpi.edu/programs/nlpip/lightinganswers/lightpollution/glare.asp.
Rhinoceros. n.d. Rhino 6 Features. Accessed 11 6, 2018. https://www.rhino3d.com/en/6/features.
Roudsari, Mostapha. 2015. Honeybee Annual Daylight Simulation Example. 11 9. Accessed 10 10, 2018.
http://hydrashare.github.io/hydra/viewer?owner=mostaphaRoudsari&fork=hydra_1&id=Honeybe
e_Annual_Daylight_Simulation_Example&slide=0&scale=3.031433133020804&offset=-
497.34512554037485,-467.6911620989838.
Sadeghipour, Roudsari Mostapha. n.d. Ladybug Tools. Accessed 9 2, 2018.
https://www.food4rhino.com/app/ladybug-tools.
Software-Matters. n.d. A Brief History of Excel. Accessed 1 3, 2019. http://www.software-
matters.co.uk/spreadsheets.html.
USGBC. n.d. Achieve better buildings with LEED. Accessed 4 1, 2019. https://new.usgbc.org/leed.
—. 2016. Checklist: LEED v4 for Building Design and Construction. 4 5. Accessed 4 1, 2019.
https://www.usgbc.org/resources/leed-v4-building-design-and-construction-checklist.
USGBC. 2019. "LEEED v4 for Building Design and Construction."
—. n.d. What is LEED. Accessed 2 20, 2019. https://new.usgbc.org/leed.
VanderLeest, Steven H., Jeffrey Nyhoff, and Nancy Zylstra. 2005. Algorithmic modeling. Accessed 3 25,
2019. https://www.calvin.edu/academic/rit/webBook/chapter5/algorithm.htm.
Virone, Gilles, Siddharth Dalad, Steven W Kell, Beverely Turner, John A Stankovic, and Robin Felder. 2008.
"Behavioral Patterns of Older Adults." IEEE TRANSACTIONS ON INFORMATION
TECHNOLOGY IN BIOMEDICINE 12 (3): 387-398.
Wymelenberg, Kevin Van Den. 2014. The Benefits of Natural Light. 3 19. Accessed 10 7, 2018.
https://www.archlighting.com/technology/the-benefits-of-natural-light_o.
Abstract (if available)
Abstract
Daylighting is important for occupants in buildings. To evaluate the lighting performance, metrics are needed to measure daylighting. Knowing when and where people are in the building can be an important consideration when analyzing daylighting. IES LM-83-12 gives an approved method on the evaluation of the lighting performance of interior space. Two metrics, one for daylighting, Spatial Daylight Autonomy (sDA) and one for glare, Annual Sunlight Exposure (ASE), are used to describe the lighting performance of “regularly occupied space” of the common working space. ❧ SDA and ASE are widely-used in daylighting performance evaluation. Thus, their scope may be able to be extended to other building types. However, according to IES LM-83-12, the analysis period for both sDA and ASE is fixed from 8am to 6pm (10 hours). People in specific building types such as assisted living may not follow that schedule, which may lead to an inaccurate lighting performance prediction. Therefore, new metrics may be needed to reveal the lighting performance more accurately according to real operating schedule. New metrics based on sDA and ASE using a flexible analysis period can determine the lighting performance more fully. The two new interior lighting metrics proposed are called flexible analysis period spatial Daylight Autonomy (fsDA) and flexible analysis period Annual Sunlight Exposure (fASE). ❧ In addition, IES LM-83-12 does not define “regularly occupied space” in detail. There is also no hierarchy within “regularly occupied space.” Space where more people spend more time in is more important, which means the more important spaces have more priority to achieve better lighting condition. Therefore, spaces with different occupied levels may need different daylighting requirement. As a result, a standard on fsDA and fASE or sDA and ASE of spaces with different occupied levels may be needed. ❧ First, a Rhino model of a case study building was built. The space of the model was divided into 6 function spaces. The schedules of assisted living were obtained by visiting assisted living communities, as well as by talking with experts and staff of senior livings. Then the sDA300,50% and the ASE1000,250h of the case building were calculated according to the method suggested by IES LM-83-12. After that, fsDA₃₀₀,₅₀﹪ and fASE₁₀₀,₂₅₀ₕ were calculated based on the operating hours rather than 8am to 6 pm (the fixed 10 hours). Next, the new calculation results were compared with that calculated based on the 10-hour period and a small difference was found between the two results. Thus, fsDA and fASE, who have a flexible analysis period, are not necessarily needed to replace sDA and ASE even though the operation hour changes in assisted living. ❧ Occupied Hour (O) and Hour Percentage (P) were developed to define the occupied hierarchy of function spaces. It was tested if the values of the two metrics could be predicted by operation schedules by finding their values under different operation schedules. If they are predictable, they can be used to define the occupied hierarchy. Otherwise, they cannot be used to decide the occupied hierarchy. It was found that both of the two metrics were influenced by operation schedules while Hour Percentage (P) can be influenced by occupant spatial distribution as well. Therefore, Occupied Hour (O) can be used as one factor to define occupied hierarchy while the feasibility if Hour Percentage (P) can decide occupied hierarchy need to be discussed after a sensitivity test. It was tested if Hour Percentage (P) was sensitive to occupant spatial distribution by finding the value of the metric under the same operation schedule while with different occupant spatial distributions. It was found Hour Percentage (P) was proportionally sensitive to occupant spatial distribution so whether the metric can be used to define occupied hierarchy of function spaces cannot be decided at this stage. The variation range of occupant spatial distribution should be explored in the future. ❧ Based on the findings, “regularly occupied” space can be defined based on Occupied Hour (O) while whether there is a predicted occupied hierarchy within “regularly occupied” space based on Hour Percentage (P) is not decided yet. A reference on sDA and ASE for “regularly occupied” function should be suggested. If there is a more detailed occupied hierarchy within “regularly occupied” space, a more detailed standard on sDA and ASE for the function spaces with different occupied levels may be suggested in the future. The management group can set up an operation schedule to make more people staying in the spaces with better daylighting condition to meet the sDA and ASE requirement.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Daylight and health: exploring the relationship between established daylighting metrics for green building compliance and new metrics for human health
PDF
Evaluation of daylighting circadian effects: Integrating non-visual effects of lighting in the evaluation of daylighting designs
PDF
Development of circadian-effective toplighting strategies: using multiple daylighting performance goals for dementia care communities
PDF
Simulation-based electric lighting control algorithm: integrating daylight simulation with daylight-linked lighting control
PDF
Multi-domain assessment of a kinetic facade: determining the control strategy of a kinetic façade using BIM based on energy performance, daylighting, and occupants’ preferences; Multi-domain asse...
PDF
Adaptive façade controls: A methodology based on occupant visual comfort preferences and cluster analysis
PDF
Performative shading design: parametric based measurement of shading system configuration effectiveness and trends
PDF
The intelligent control strategy of kinetic façades for daylight and energy performance: evaluating the daylight effect of adaptive systems based on parametric workflow
PDF
Double skin façades performance: effects on daylight and visual comfort in office spaces
PDF
Lateral design with mass timber: examination of structural wood in high-rise timber construction
PDF
Real-time simulation-based feedback on carbon impacts for user-engaged temperature management
PDF
Absolute glare and relative glare factors: predicting and quantifying levels of interior glare and exterior glare caused by sunlight and daylight
PDF
Impacts of building performance on occupants' work productivity: a post occupancy evaluation study
PDF
Net-zero cultural urbanism: the implementation of traditional and cultural net-zero urban housing design in Lagos
PDF
Building retrofitting evaluation: Energy savings and cost effectiveness of building retrofits on Graduate Art Studios at the University of California, Los Angeles
Asset Metadata
Creator
Zeng, Ruoxiao
(author)
Core Title
Exploring the influence of scheduling on daylighting performance evaluation in assisted living: testing two new metrics based on sDA and ASE
School
School of Architecture
Degree
Master of Building Science
Degree Program
Building Science
Publication Date
07/03/2019
Defense Date
05/06/2019
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
OAI-PMH Harvest,sDA, ASE, fsDA, fASE, occupied hierarchy, assisted living
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Konis, Kyle (
committee chair
), Gaines, Lauren Dandridge (
committee member
), Kensek, Karen (
committee member
), Regnier, Victor Albert (
committee member
)
Creator Email
ruoxiaoz@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c89-180244
Unique identifier
UC11662237
Identifier
etd-ZengRuoxia-7528.pdf (filename),usctheses-c89-180244 (legacy record id)
Legacy Identifier
etd-ZengRuoxia-7528.pdf
Dmrecord
180244
Document Type
Thesis
Format
application/pdf (imt)
Rights
Zeng, Ruoxiao
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
sDA, ASE, fsDA, fASE, occupied hierarchy, assisted living