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
/
Enhanced post occupancy evaluation (POE) for office building: improvement of current methodology to identify impact of ambient environment
(USC Thesis Other)
Enhanced post occupancy evaluation (POE) for office building: improvement of current methodology to identify impact of ambient environment
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
Enhanced Post Occupancy Evaluation(POE) for Office Building:
Improvement of current methodology to identify impact of ambient environment
by
KYEONGSUK LEE
A Thesis
SCHOOL OF ARCHITECTURE
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF BUILDING SCIENCE
MAY 2018
ii
COMMITTEE
CHAIR:
Joon-ho Choi
Assistant Professor
USC School of Architecture
joonhoch@usc.edu
COMMITTEE MEMBER #2:
Marc Schiler
Professor
USC School of Architecture
marcs@usc.edu
(213)740-4591
COMMITTEE MEMBER #3:
Selwyn Ting
Associate Professor of Practice
USC School of Architecture
sting@usc.edu
(310)902-2998
COMMITTEE MEMBER #4:
Heidi Creighton, AIA, LEED AP BD+C, O+M, WELL AP
Associate
Buro Happold Engineering
Heidi.Creighton@burohappold.com
(310)945-4870
iii
ABSTRACT
Post-occupancy evaluation (POE), an architectural design tool for making better indoor
environmental quality to enhance occupants’ satisfaction and productivity, has been conducted to
find out the impact of the variable ambient environment. Since indoor environmental quality (IEQ)
is fully related to occupants’ satisfaction and productivity, many research projects about IEQ have
been launched. POE is one of the methods to determine the quality of the indoor environment
based on occupants’ feedback; a user satisfaction survey is conducted and connected to various
environmental strategies and building performance. However, a POE study always has the critical
limitation that it mainly relies on a subjective survey. To overcome this limitation, many
researchers have developed methods that integrated survey feedback with quantitative IEQ data
gathered by measurement sensors. Despite the effort of developed POE methodology, it is still
necessary to consider more factors such as season, time of day, and day of the week to improve
the quality of the POE studies. Three different offices were selected based on climate condition
and building attributes. Data collection using IEQ measurements and multiple-occupant
satisfaction surveys were conducted for a year. Occupant satisfaction surveys and IEQ
measurements were performed at the same time to aggregate the data of occupants’ feedback and
their indoor condition. To analyze the dataset, various methodologies including statistical analysis
and advanced data mining were adopted. The results show that occupant has an adjustability to
their ambient environments and have a constant satisfaction level with IEQ condition by different
months. However, this result was not constant depending on the offices. The result of Office C
indicated that occupant had a significant change of thermal satisfaction between different
measurement times. Moreover, the study also found that human factors and time functions affected
the variety of occupant’s environmental satisfaction. Based on the result, it is possible to conclude
that one-time data collection is hard to fully reliable and multiple data collection is required, as a
POE method.
iv
HYPOTHESIS
1. A common POE method that uses one-time measurement data is not adequate to
diagnose IEQ condition and occupants’ environmental satisfaction fully.
2. Variable factors including season and a time of day affect user satisfaction.
3. Multiple data collections are a more accurate method to consider impacts of various
features on human perception.
v
TABLE OF CONTENTS
COMMITTEE ............................................................................................................................... ii
ABSTRACT .................................................................................................................................. iii
HYPOTHESIS ............................................................................................................................. iv
TABLE OF CONTENTS ............................................................................................................. v
LIST OF FIGURES .................................................................................................................... vii
LIST OF TABLES ..................................................................................................................... viii
CHAPTER 1: INTRODUCTION ................................................................................................ 1
1.1 Indoor Environment Quality (IEQ) .................................................................................. 1
1.1.1 Indoor Environment Quality ..................................................................................... 1
1.1.2 IEQ Standards ........................................................................................................... 1
1.2 Human Comfort in Indoor Environment .......................................................................... 2
1.3 Post Occupancy Evaluation (POE) as an IEQ Research Tool ......................................... 4
1.3.1 Post-Occupancy Evaluation (POE) ........................................................................... 5
1.3.2 POE in Office Environments .................................................................................... 5
1.3.3 Limitation of POE ..................................................................................................... 5
1.3.4 Developing POE method .......................................................................................... 6
1.4 Scope of the thesis ............................................................................................................ 7
1.5 Structure of the Thesis ...................................................................................................... 7
CHAPTER 2: PREVIOUS WORK: BACKGROUND AND LITERATURE REVIEW ...... 9
2.1 Importance of Indoor Environmental Quality (IEQ) ........................................................ 9
2.2 Concept of Post Occupancy Evaluation (POE) .............................................................. 10
2.3 Previous POE research by using survey data ................................................................. 11
2.4 POE integrated with IEQ data ........................................................................................ 14
2.5 Occupant’s environmental perception ............................................................................ 15
CHAPTER 3: METHODOLOGY ............................................................................................ 17
3.1 Building selection ........................................................................................................... 18
vi
3.2 IEQ measurement ........................................................................................................... 19
3.3 Satisfaction survey ......................................................................................................... 20
3.4 Statistical analysis .......................................................................................................... 23
3.4.1 Correlation analysis ........................................................................................................ 24
3.4.2 Two-sample T-test ................................................................................................ 24
3.4.3 Analysis of Variance (ANOVA) ............................................................................. 24
3.4.4 Decision Tree (J48) ................................................................................................. 24
CHAPTER 4: RESULT .............................................................................................................. 26
4.1 Overview data analysis ................................................................................................... 26
4.2 Case 1: Office A ............................................................................................................. 28
4.2.1 The change of occupant’s responses ....................................................................... 34
4.2.2 Impact of human factors ......................................................................................... 35
4.2.3 Impact of time functions ......................................................................................... 40
4.3 Case 2: Office B ............................................................................................................. 41
4.3.1 The change of occupant’s responses ....................................................................... 47
4.3.2 Impact of human factors ......................................................................................... 48
4.3.3 Impact of time functions ......................................................................................... 52
4.4 Case 3: Office C ............................................................................................................. 53
4.4.1 The change of occupant’s responses ....................................................................... 59
4.4.2 Impact of human factors ......................................................................................... 60
4.4.3 Impact of time functions ......................................................................................... 63
4.5 Discussion ..................................................................................................................... 64
4.6 Developing design suggestion by using data mining tool .............................................. 66
CHAPTER 5: CONCLUSION .................................................................................................. 70
BIBLIOGRAPHY ....................................................................................................................... 73
vii
LIST OF FIGURES
FIGURE 1. RESEARCH METHODOLOGY DIAGRAM .......................................................................... 18
FIGURE 2. IMAGE OF EACH BUILDING ............................................................................................. 19
FIGURE 3. INDOOR ENVIRONMENT QUALITY MEASUREMENT CART AND SENSING INTERFACE ....... 20
FIGURE 4. THE IMAGE OF HOBO SENSOR ...................................................................................... 20
FIGURE 5. DISTRIBUTION OF TEMPERATURE AT FLOOR LEVEL AND TEMPERATURE AT 1.2M ......... 29
FIGURE 6. DISTRIBUTION OF RADIANT TEMPERATURE ASYMMETRY WALL AND RADIANT
TEMPERATURE ASYMMETRY CEILING .................................................................................... 29
FIGURE 7. DISTRIBUTION OF RELATIVE HUMIDITY AND CO
2
CONCENTRATION ............................. 30
FIGURE 8. DISTRIBUTION OF WORK SURFACE ILLUMINANCE AND UGR ........................................ 31
FIGURE 9. DISTRIBUTION OF NOISE LEVEL ..................................................................................... 31
FIGURE 10. ROSE CHART OF THE SURVEY QUESTION BY MONTH .................................................... 33
FIGURE 11. DISTRIBUTION OF TEMPERATURE AT FLOOR LEVEL AND TEMPERATURE AT 1.2M ....... 41
FIGURE 12. DISTRIBUTION OF RADIANT TEMPERATURE ASYMMETRY WALL AND RADIANT
TEMPERATURE ASYMMETRY CEILING .................................................................................... 42
FIGURE 13. DISTRIBUTION OF RELATIVE HUMIDITY AND CO
2
CONCENTRATION ........................... 42
FIGURE 14. DISTRIBUTION OF WORK SURFACE ILLUMINANCE AND UGR ...................................... 43
FIGURE 15. DISTRIBUTION OF NOISE LEVEL ................................................................................... 44
FIGURE 16. ROSE CHART OF THE SURVEY QUESTION BY MONTH .................................................... 46
FIGURE 17. DISTRIBUTION OF TEMPERATURE AT FLOOR LEVEL AND TEMPERATURE AT 1.2M ....... 53
FIGURE 18. DISTRIBUTION OF RADIANT TEMPERATURE ASYMMETRY WALL AND RADIANT
TEMPERATURE ASYMMETRY CEILING .................................................................................... 54
FIGURE 19. DISTRIBUTION OF RELATIVE HUMIDITY AND CO
2
CONCENTRATION ........................... 55
FIGURE 20. DISTRIBUTION OF WORK SURFACE ILLUMINANCE AND UGR ...................................... 55
FIGURE 21. DISTRIBUTION OF NOISE LEVEL ................................................................................... 56
FIGURE 22. ROSE CHART OF THE SURVEY QUESTION BY MONTH .................................................... 58
FIGURE 23. DECISION TREE (J48) FROM SELECTED SURVEY QUESTIONS AND IEQ FACTORS .......... 67
FIGURE 24. DECISION TREE (J48) FROM SELECTED SURVEY QUESTIONS AND IEQ FACTORS .......... 67
FIGURE 25. DECISION TREE (J48) FROM SELECTED SURVEY QUESTIONS AND IEQ FACTORS .......... 68
FIGURE 26. DECISION TREE (J48) FROM SELECTED SURVEY QUESTIONS AND IEQ FACTORS ......... 69
viii
LIST OF TABLES
TABLE 1. SUMMARY OF IEQ STANDARDS ........................................................................................ 2
TABLE 2. OCCUPANT SATISFACTION SURVEY ................................................................................ 23
TABLE 3. SUMMARY OF SCALE POINTS OF OCCUPANT SATISFACTION SURVEY ............................... 23
TABLE 4.. SUMMARY OF MEASUREMENT TIME AND WEATHER CONDITION ..................................... 26
TABLE 5. DEMOGRAPHIC INFORMATION OF DATASET BY HUMAN FACTORS ................................... 27
TABLE 6. DEMOGRAPHIC INFORMATION OF DATASET BY TIME FACTORS ....................................... 28
TABLE 7. SUMMARY OF MEASURED IEQ DATA BY MONTH ............................................................. 32
TABLE 8. SUMMARY OF P-VALUE FROM ANOVA TEST BY DIFFERENT DATA CATEGORIES ............ 35
TABLE 9. CONFIDENCE INTERVAL OF ACOUSTIC QUALITY SATISFACTION (A); MEASURED ACOUSTIC
DECIBEL (B); ACOUSTIC QUALITY SATISFACTION (C); MEASURED ACOUSTIC DECIBEL (D) BY
MONTH .................................................................................................................................... 36
TABLE 10. CONFIDENCE INTERVAL OF ACOUSTIC QUALITY SATISFACTION (A) AND MEASURED
ACOUSTIC DECIBEL (B) BY MONTH ......................................................................................... 37
TABLE 11. CONFIDENCE INTERVAL OF LIGHTING CONDITION SATISFACTION (A); DIRECT GLARE
SATISFACTION (B); LIGHTING QUALITY SATISFACTION (C); MEASURED UGR (D); MEASURED
WORK SURFACE ILLUMINANCE (E) BY MONTH ........................................................................ 38
TABLE 12. SUMMARY OF CORRELATION ANALYSIS ........................................................................ 39
TABLE 13.SUMMARY OF P-VALUE FROM STATISTICAL ANALYSIS BY DIFFERENT DATA CATEGORIES
............................................................................................................................................... 40
TABLE 14. SUMMARY OF MEASURED IEQ DATA BY MONTH ........................................................... 45
TABLE 15. SUMMARY OF P-VALUE FROM ANOVA TEST BY DIFFERENT DATA CATEGORIES .......... 47
TABLE 16. CONFIDENCE INTERVAL OF THERMAL QUALITY SATISFACTION (A); MEASURED
TEMPERATURE (B); THERMAL QUALITY SATISFACTION (C); MEASURED TEMPERATURE (D) BY
MONTH .................................................................................................................................... 48
TABLE 17. CONFIDENCE INTERVAL OF GLARE CONDITION SATISFACTION (A); MEASURED UGR (B);
GLARE CONDITION SATISFACTION (C); MEASURED UGR (D) BY MONTH ................................ 50
TABLE 18. CONFIDENCE INTERVAL OF ACOUSTIC QUALITY SATISFACTION (A); MEASURED NOISE
LEVEL (B); ACOUSTIC QUALITY SATISFACTION (C); MEASURED NOISE LEVEL (D) BY MONTH . 51
TABLE 19.SUMMARY OF P-VALUE FROM STATISTICAL ANALYSIS BY DIFFERENT DATA CATEGORIES
............................................................................................................................................... 52
TABLE 20. SUMMARY OF MEASURED IEQ DATA BY MONTH .......................................................... 57
TABLE 21. SUMMARY OF P-VALUE FROM ANOVA TEST BY DIFFERENT DATA CATEGORIES .......... 59
ix
TABLE 22. CONFIDENCE INTERVAL OF THERMAL QUALITY SATISFACTION (A); MEASURED
TEMPERATURE (B); THERMAL QUALITY SATISFACTION (C); MEASURED TEMPERATURE (D) BY
MONTH .................................................................................................................................... 60
TABLE 23. CONFIDENCE INTERVAL OF GLARE CONDITION SATISFACTION (A); MEASURED UGR (B);
GLARE CONDITION SATISFACTION (C); MEASURED UGR (D) BY MONTH ................................ 61
TABLE 24.CONFIDENCE INTERVAL OF ACOUSTIC QUALITY SATISFACTION (A); MEASURED NOISE
LEVEL (B); ACOUSTIC QUALITY SATISFACTION (C); MEASURED NOISE LEVEL (D) BY MONTH . 62
TABLE 25. SUMMARY OF P-VALUE FROM STATISTICAL ANALYSIS BY DIFFERENT DATA CATEGORIES
............................................................................................................................................... 63
1
CHAPTER 1: INTRODUCTION
This research focuses on advancing the current post occupancy evaluation (POE) methods to
enable revealing the impact of seasonal changes on occupants’ satisfaction and environmental
perception. Also, the research validates the effect of repetitive on-site measurement and user
satisfaction surveys to get more sophisticated evaluation results and find out what kind of external
forces affect users’ satisfaction. This chapter provides an introduction to understanding the basic
elements of indoor environmental quality (IEQ) and POE study. Moreover, this chapter includes
current limitations of POE study and describes how developing methodology can overcome these
limitations.
1.1 Indoor Environment Quality (IEQ)
1.1.1 Indoor Environment Quality
Humans spend 80 to 90% of the day in enclosed buildings [1]. Therefore, it was necessary to
provide a comfortable indoor environment for occupants. Indoor Environmental Quality (IEQ) is
simply explained as the condition of the inside of the building. It can affect occupants’ comfort
and has various categories including thermal, lighting, acoustics, and air quality. Since building
occupants are easily influenced by IEQ factors, myriads of researchers have conducted IEQ studies
to identify the impact of each element. Multiple studies reveal that improvement of IEQ conditions
has a remarkable impact on users’ health and satisfaction. Moreover, it also enhances the
productivity of workers in office buildings [2]–[4]. Also, achieving a high quality indoor
environment not only creates a comfortable indoor space, but also has an indirect effect by saving
energy through cutting-edge technologies. Based on the reasons mentioned above, the
consideration of IEQ factors in building design and control is continuously necessary.
1.1.2 IEQ Standards
To make people feel satisfied with their indoor space, it is necessary to provide a proper level of
IEQ factors. Since the building industry has known about the importance of IEQ to improve users’
2
satisfaction and productivity, various industrial organizations related to IEQ factors have
suggested standards which provide minimum requirements for acceptable IEQ levels. For instance,
the American Society of Heating, Refrigerating and Air-conditioning Engineers (ASHRAE) offers
ASHRAE standard 55 and 62.1 which are the basic guideline for thermal comfort and air quality
[5], [6]. Moreover, the Illuminating Engineering Society of North America (IESNA) and the
International Commission on Illumination (CIE) offer standards for comfortable lighting levels [7],
[8]. Table 1 provides a summary of IEQ standards.
Variable Guideline
Temperature floor ('C) between 19 and 29 'C (ASHRAE 55)
Temperature 1.2m ('C) between 23.3 and 27.8 'C (ASHRAE 55)
Vertical Air Temperature Difference ('C) less than 3 'C (ASHRAE 55)
Radiant Temperature Asymmetry_Ceiling ('C) less than 5 'C (ASHRAE 55)
Radiant Temperature Asymmetry Wall ('C) less than 10 'C (ASHRAE 55)
Relative Humidity (%) 65 % or less (ASHRAE 62)
CO2 level (ppm) less than 1000 ppm (ASHRAE)
Work surface illuminance (lux) between 200 and 500 lux (ANSI/IES RP-1-12)
UGR between 13 and 19 (CIE)
Acoustic decibel (dBA) less than 40 Dba (ASHRAE)
Table 1. Summary of IEQ Standards
1.2 Human Comfort in Indoor Environment
As a primary goal of indoor environmental design, lots of sustainable building designs have
considered IEQ conditions and the occupant’s comfort. Since a comfortable indoor environment
directly affects user’s health issues and productivity, its importance has frequently been pointed
3
out. To create a high quality indoor space, it is necessary to focus on four major IEQ elements:
thermal, lighting, indoor air, and acoustic quality.
Thermal comfort is the occupant’s satisfaction regarding the ambient thermal conditions and is
fundamental to consider when designing a building. Since it is highly subjective, it is difficult to
measure accurate thermal satisfaction. According to the ANSI/ASHRAE Standard 55 which is
commonly used as a thermal comfort guideline, there are six factors to take into consideration
when aiming for thermal comfort. Its determining criteria include the (1)temperature of the air
surrounding the occupant, (2)relative humidity, (3)radiant temperatures, (4)occupants’ metabolic
rates (met), (5)clothing insulation (clo), and (6)air velocity across body surfaces [9].
Maintaining lighting comfort means ensuring that people have comfortable lighting level for their
works, that the light has the appropriate quality and balance, and that people can access great views.
Comfortable lighting conditions help to enhance occupants’ well-being and productivity. There
are two main design strategies related to lighting quality: daylighting and artificial lighting. Users
tend to feel better and have higher satisfaction levels with daylighting. Therefore, it is significantly
necessary to make a balance between artificial lighting and daylighting for providing a high quality
of visual comfort without glare. Moreover, occupants’ lighting perception are subjective in the
same way as thermal comfort. Therefore, an individual control system is required to achieve visual
comfort at each workstation [10].
Indoor Air Quality (IAQ) refers to the air quality within and around buildings and structures. Since
IAQ primarily relates to the occupant’s health, it is essential to understand and control common
pollutants including Particulate Matter (less than 10 micrometers) (PM10) and TVOC to help
reduce your risk of indoor health concerns. To achieve healthy IAQ condition, the indoor area
needs to bring in 100% outside air by using natural ventilation or HVAC system which has high-
efficiency Minimum Efficiency Reporting Value (MERV) filters. In addition, it is recommended
to use non-toxic material and products because furniture or materials release pollutants. As a
4
primary industrial guideline, ASHRAE Standard 62.1 provides information that how to create
good IAQ [11].
Controlling acoustic level is also an essential method of improving users’ satisfaction with the
indoor environment. Occupants are normally more productive when they are not distracted by
noises from the outside or surrounding environment. Acoustic comfort is especially important for
schools and office buildings. Although occupants’ acoustic perception is also subjective, a
comfortable environment can be created by controlling objective measures like decibel level,
reverberation time, and the sound reflection and damping properties of materials. To reduce
background noise, acoustic tiles on ceilings and walls can be used to dampen sound [10].
1.3 Post Occupancy Evaluation (POE) as an IEQ Research Tool
Post occupancy evaluation (POE) has been used for several years to evaluate the performance of
buildings after they have been built and occupied for some time. The fundamental concept of this
study is based on the requirement of occupants including satisfaction, health, efficiency, safety,
and psychological comfort. To conduct POE studies, there are many methods such as
questionnaires, interview, and physical measurement.
As a primary method of POE, user satisfaction surveys were mainly used to understand how
occupants are satisfied with their indoor environment. In one U.S. institution’s study, an online
user satisfaction survey was established, which include 68-questions[12]. The purpose of the
survey is to evaluate occupant satisfaction with thermal comfort, air quality, individual
workstation’s spatial characteristics, lighting, acoustic quality, and maintenance. Moreover,
additional questions about functionality, community, and satisfaction beyond the environmental
factors were also added to the survey [13], [14].
5
1.3.1 Post-Occupancy Evaluation (POE)
Many buildings do not perform as designed and it can cause various problems regarding running
costs, occupant’s environmental satisfaction, building performance, health and safety. In order to
avoid this problem, various construction companies have tried to learn from and correct past
mistakes in design and commissioning of buildings to improve workplace productivity and
enhance the performance of new buildings. To identify the mistakes and develop the better design,
Post Occupancy Evaluation (POE) is commonly used. The value of POE is increasingly recognized,
and it is becoming mandatory on many public projects. It is valuable in all construction sectors,
especially healthcare, education, offices, commercial building, and housing, where poor building
performance will impact running costs, occupant well-being, and business efficiency[15].
1.3.2 POE in Office Environments
For building occupants meanwhile, a healthy workforce is a crucial elements of a productive,
successful business in the long-term. For the companies, salaries and benefits typically account for
about 90% of a business’ operating costs. It indicates that the productivity of workers, or anything
that influences their ability to be productive, should be a primary concern for any organization. In
addition, although it is a small difference or development regarding users’ health and productivity,
the outcome can be significant and become a major financial implication for employers. According
to World Green Building Council (WGBC), a range of office design factors, from indoor air quality,
thermal comfort and daylighting, to acoustics, interior layout, views, bio-philia, location, and
amenities need to be considered to enhance the workplace [16]. Because of the importance of these
factors, POE has frequently been used to provide optimal designs for office building [17].
1.3.3 Limitation of POE
Most previous POE studies have entirely relied on survey responses. Since the survey data is
subjective, it may exaggerate study result if the study uses surveys alone. This limitation is
6
frequently pointed to by many researchers as a weakness of POE research. Thus, ambient IEQ
conditions can easily affect the satisfaction and perception of occupants. To prevent this limitation,
it is necessary to understand the impact of real IEQ conditions of the workstation to integrate this
with survey data.
Since current sensing technology has been significantly developed, there has been another
scientific trend to overcome the limitation of previous POE studies by using IEQ measuring
sensors. One study from the Center for Building Performance and Diagnostics (CBPD) at Carnegie
Mellon University showed that on-site measurement could enhance the objectivity of POE studies.
Thus, it helped to understand the relationship between IEQ condition of workstations and
occupants’ satisfaction.
Despite objective data from on-site IEQ measurements, it is still hard to say that those data can be
sufficiently reliable because of the variety of IEQ conditions. Most previous research which
conducted on-site measurement only did a one-time measurement. However, ambient IEQ
conditions which can easily influence the satisfaction and perception of occupants may vary
depending on the season, day and time of day. Moreover, even occupants’ perception can be
different based on the physiological & psychological condition of occupants.
1.3.4 Developing POE method
To compensate for the uncertainty of a one-time survey and measurement of data, the objective of
this paper is to understand a variety of actual IEQ condition and suggest effective POE methods
using data collection at multiple times. This study carried out a repetitive on-site measurement to
gather environment performance data. The database used in this study contained measures of
indoor thermal, lighting, acoustic, and air condition, and was collected from three different office
buildings in Southern California. In addition, on-site user satisfaction surveys were performed in
7
each building to integrate with the IEQ database. Based on the database, the study team used
statistical analysis and comparative analysis to determine the most reliable POE methodology that
can supplement existing limitation of POE.
1.4 Scope of the thesis
The research focuses on the impact of factors including season, time of day, and day of the week
on occupants’ environmental satisfaction. Comprehensive POE was repetitively conducted at three
offices in Southern California to collect data through user satisfaction surveys and other IEQ
measurements. To identify the impact of the factors mentioned above, data collections are carried
out every two or three months during different times and days. Based on aggregated data, statistical
analysis was performed to diagnose differences of human IEQ perception and satisfaction. To
evaluate human comfort, five different criteria were observed: thermal, visual, acoustic, air quality
and spatial comfort. In addition, human factors like gender and three different range of age (young,
mid-age, elderly) are also considered.
1.5 Structure of the Thesis
Chapter 1 introduces basic knowledge about indoor environmental quality (IEQ) research and post
occupancy evaluation (POE) in a modern office environment. Thus, the chapter also includes
current limitations of POE and suggestions of how to overcome them. Chapter 2 illustrates the
background and literature review of previous POE research to show the importance of POE and
how its methodology has been developed. In addition, the impact of various factors such as human
factors and the environmental element are also explained. Chapter 3 shows the improved POE
methodologies: building selection, on-site measurement, occupant satisfaction surveys and
statistical analysis. Chapter 4 describes the overall IEQ condition of each office, and user
satisfaction based on aggregated data. Moreover, the chapter also illustrates the change of
occupant’s response and the impact of human factors and time functions on occupant’s
8
environmental perception. Finally, Chapter 5 illustrates conclusion of the thesis with identified
weakness/limitations.
9
CHAPTER 2: PREVIOUS WORK: BACKGROUND AND LITERATURE REVIEW
Many researchers have already studied POE and revealed a significant impact of IEQ on occupants
and residents. This chapter shows previous researches related to POE and shows how POE
methods have been developed to overcome existing limitation. Thus, some studies explained how
human factors such as gender and age and/or specific IEQ factors could influence on occupants’
perception.
2.1 Importance of Indoor Environmental Quality (IEQ)
IEQ study has been conducted for several decades to identify a comfortable indoor environment
range. IEQ includes various factors such as thermal comfort, air quality, lighting, and acoustic
quality. These elements can affect occupants’ satisfaction. Moreover, IEQ also has an impact on
the productivity of office occupants. Therefore, it is very critical to maintain the high quality of
the indoor environment to enhance employees’ productivity in office building [18].
Tiller investigated the combined impact of background acoustic levels and temperature on human
comfort and performance. Using a sample of 16 females and 14 males, researchers observed as the
subjects were exposed to a matrix of combinations between two noise qualities, three noise levels,
and five thermal conditions. It is noted that since no environment will please everyone at once it
was important to identify the conditions which were satisfactory to the largest group of people.
The research revealed that the subjects’ thermal comfort can be affected by acoustical disruptions
while the subjects’ impression of acoustics was not affected by the thermal conditions. Moreover,
the test subject perceived temperatures as colder as noise levels increased. The female group rated
the lower temperatures colder than males and the higher temperatures more comfortable than males.
The thermal composite ratings for males and females meet at 72 degrees. In addition, rumbly
noises were disliked more than white noise sounds and the subjects could or would adapt to their
conditions meaning that a sensitivity to initial condition impressions was also observed [19].
10
Brager’s research examined how a wider range of temperatures inside an office space can improve
productivity and decrease health problems associated with keeping an indoor temperature ranges
within a one to the two-degree range. The researchers used productivity as a means to evaluate the
effects of losses in energy efficiency because energy costs are very low; only 3% of a standard
business budget. To achieve the goal of a wider global temperature range, this experiment provided
personal comfort systems (PCS) to each occupant thus creating indoor environments that were
individually controlled and thus improved productivity and well being. The PCS includes a low-
cost desktop fan and under-desk radiant foot warmer which enabled the workers to have more
control over the air speeds and temperatures. Moreover, the research illustrated that there is a
correlation between workplace sicknesses and air speeds. It is suggested that if building managers
moved to PCS conditioning environment approaches, workers could make adjustments according
to their preferences and it would become unnecessary to maintain a specific temperature using a
centralized system [20].
As shown in the above reviews, IEQ conditions are significantly relevant to occupant’s satisfaction
and productivity. Since IEQ factors and user’s environmental satisfaction are remarkably
complicated, it is necessary to conduct many types of research to acquire significant findings
including the impact of IEQ factors, the correlation between each factor, and how human factors
(gender and age group) affect occupant’s environmental comfort. Ultimately, it would be possible
to provide an optimal indoor condition for the human
2.2 Concept of Post Occupancy Evaluation (POE)
Post Occupancy Evaluation (POE) as a diagnostic design tool helps facility managers or architects
evaluate critical aspects of building performance systematically. This system has been used to
identify problems of existing buildings, to test new building prototypes, and to develop future
design guidelines. To perform a POE study, user feedback needs to be collected to diagnose the
environmental condition and performance of the building. The first POE studies started in the mid-
1960s. However, they were only conducted when serious problems happened in buildings. In
11
addition, the early POE model was applied to housing, and has been expanding to office buildings
and other commercial real estate. Moreover, not only building type but also research focus has
been developed to show a relationship between human behavior and building design. Building
performance consists of several elements including technical, functional, and behavioral elements
and each element contributes to different environmental strategies. Technical elements deal with
safety, health, security, and performance of building systems. Functional elements which include
workflow, productivity, and operational efficiency assess the fit between the building and the
occupants’ activity. Behavioral elements consider perception and psychological needs of building
users. Based on collected information which takes care of various strategies, it is possible to
understand current building performance and improve IEQ conditions [21]–[24].
2.3 Previous POE research by using survey data
As a main methodology of POE, user satisfaction surveys that ask occupants’ satisfaction of IEQ
and building performance have been used. One finding describes that there are not linear
relationships between individual IEQ factors and overall user satisfaction using survey data from
the Center for the Built Environment (CBE). This survey was conducted since 2000 and
accumulated data from more than 600 buildings with diverse usages. The study analyzed 43,021
respondent samples from 351 different office buildings extracted from the CBE database and
suggested categorization of IEQ factors [25]. Another study investigated occupant satisfaction in
a Building Research Establishment Environmental Assessment Method (BREEAM)-Certified
office building compared with the Non-BREEAM-Certified building based on 203 survey
response collected from selected buildings in the United Kingdom. The study showed noticeably
lower IEQ satisfaction when occupants had spent more than 24 months at a BREEAM office. It
suggested further research to continue the benefits of sustainable building over time[26].
Frontczak performed a questionnaire survey in Danish homes to identify the elements which can
affect occupant’s environmental comfort. The questionnaire included different types of questions
regarding user’s behavior, their knowledge of building control system, and the way to achieve
12
satisfaction. The research team sent a total of 2499 questionnaires to occupants who lived in
common Danish housing. With a response rate of 26%, data analysis was conducted. The result
revealed that occupants believed four main IEQ factors (thermal, acoustic, lighting condition, and
air quality) were the most important parameters determining their comfort. In addition, the
respondents indicated that they preferred manual control of indoor environment except for thermal
condition. For thermal comfort, users noted that they could accept both automatic and manual
control. The respondents also indicated that they did not try to search a solution even if any
problem relevant to indoor environment condition occurred in their home. Based on these results,
the study suggested that it is necessary to enhance people’s awareness related to the consequences
of bad IEQ conditions on human health and enhancing the knowledge of a good indoor climate
[27].
Clara conducted a web-based occupant satisfaction survey to evaluate indoor environment quality
of office buildings in Northern Italy. The surveys had a 7-point scale and were distributed to the
occupants of an energy efficient Italian office building. A total of 78 people participated in the
survey. With a collected dataset, the research team made a comparison study against a large
database of 66 surveyed buildings which were LEED certificated. The results showed that the
analyzed building had higher scores than the median of the 66 LEED buildings in the categories
of: acoustic quality, thermal comfort, general satisfaction, air quality, and office furnishings, and
lower scores than the median in cleanliness and maintenance, office layout, and lighting. The
survey also illustrated that high simulated energy efficiency and high measured IEQ could be
achieved simultaneously [28].
Similar to the previous review, Zagrues also conducted a web-based indoor environmental quality
survey. He demonstrated that building occupants could be used as a rich source of information
about IEQ and its impact on comfort and productivity. The research team developed an online
survey and reporting tools to quickly and inexpensively acquire, analyze, and present the data. To
create an extensive database as a standard, the survey, which assesses user’s environmental
13
satisfaction with IEQ components: thermal, acoustic, visual, comfort, and building cleanliness and
maintenance, was performed in more than 70 buildings. Based on the large collected dataset, the
study showed how the survey could be used depending on the application. Three different cases
such as pre/post analysis of users in new building, identifying an impact of environmental factors
on occupant’s satisfaction and productivity, and an example of the survey used to establish how
new buildings are meeting a client’s design objectives were presented as a representative
application of survey [29].
Langton investigated the nature and extent of user satisfaction with the built environment in a
different organizational setting in Australia. The study pointed out that most research relevant to
IEQ tended to focus on the impact of the built environment, instead of the actual needs of occupants
working in different organizational settings. To conduct a comparison study, a survey was
conducted in 41 buildings including six government buildings, 14 educational buildings, and 21
commercial buildings. The results illustrated that there were significant differences in aspects of
air, temperature, space suitability, flexibility, usability, and controllability based on different
building types. Users in the educational buildings showed the highest satisfaction with most
variables in the workspace design and management category. Employees of Government buildings
noted a lower level of satisfaction with their physical work environment and workspace design
and management. Moreover, the government and educational group displayed more similarity with
each other, while the commercial employees showed significant difference [30].
Although these previous studies show powerful finding from a large number of data, some
limitations were found. Since the survey data is subjective, it can exaggerate study results if the
study uses surveys alone [31]. Thus, because a user’s ambient IEQ condition could easily affect
the satisfaction and perception of individual occupants, an occupant’s environmental comfort
cannot be readily understood without identifying relevant ambient environmental quality
conditions. Therefore, it is necessary to understand the impact of IEQ conditions on user
satisfaction by integrating environmental and survey data.
14
2.4 POE integrated with IEQ data
As mentioned above, POE’s excessive reliance on subjective surveys is frequently pointed out as
the weakness of the POE by researchers. To overcome the limitation of POE, IEQ measurement
which brings objectivity can be conducted to integrate with survey data. IEQ conditions can be
defined with spot and continuous measurement, while the users’ satisfaction of the environmental
condition with their workstations is simultaneously surveyed [32].
Choi’s recent study showed improved POE method by integrating IEQ measurement of buildings
and user surveys. Rather than limiting data collection to distributing surveys that can lead to
qualitative general solutions, using data to cross-examine data results improves the quality of the
data results. By integrating human and IEQ conditions data, the results were shown to be of
hierarchal nature and thus the operational response could be improved to better meet the needs of
the occupants in several ways. Data were collected from several office and educational buildings
in southern California including nine buildings at USC where the climate is warm and dry. Three
types of data were collected. The first consisted of observing the spatial factors of the building’s
attributes. The second data set included IEQ factors such as illuminance levels, temperature, noise
level, and glare. The third data set included surveys given to occupants to rate their perceived
satisfaction of the space. Characteristics of 411 occupants were organized based on three age
brackets and the sex of the individual. Lighting was found to be impacted by age, and another
interesting observation is that men preferred lower air speeds than the woman. In addition, the
senior group preferred lower illuminance level and lower air velocity. Based on aggregated data,
a decision tree was established indicating the order of importance of IEQ conditions to suggest
how office environment should be improved [33].
Liang conducted occupant satisfaction surveys integrated with monitored IEQ variables to
compare indoor environment of the green and commercial building. The study illustrated that
Taiwan’s green building certification system established IEQ criteria to evaluate the performance
of buildings regarding lighting, acoustics, ventilation, and decoration. To perform the data
15
acquisition, building monitoring was conducted in Central Taiwan during a period of the cooling
season. According to the monitored variables, the background noise, illumination, and CO
2
in
both building types were close to the international and Taiwan’s regulatory guideline, while
volatile organic compounds were not. In addition, occupants of the green buildings showed higher
degrees of overall IEQ satisfaction compared to the conventional buildings. The result also
revealed that there is the statistically significant difference between the mean score of satisfaction
in the green buildings and conventional buildings [34].
Zhe’s research investigated the indoor environment conditions of Chinese large-hub airport
terminal buildings. The study described that airport terminal buildings have unique IEQ
performance because of their operational and architectural features. The subjective surveys and
objective measurements were conducted in eight major Chinese large-hub airport terminal
buildings. The surveys showed that people were “just satisfied” and “satisfied” with the IEQ of
airport buildings. Among the IEQ variables, thermal condition and air quality were the most
important factors for travelers’ overall satisfaction. In addition, people's air quality satisfaction
level is highly correlated with the CO
2
concentration of airport terminals [35].
2.5 Occupant’s environmental perception
Although current scientific POE methods have revealed numerous significant findings, many
existing POE studies principally depend on subjective survey data and on-site IEQ measurements,
which are limited to one-time data acquisition. Since humans are sensitively affected by time-
varying indoor and outdoor conditions because of their circadian rhythms and daily and seasonal
changes, it is difficult to accept that these one-time data collections are fully trustworthy.
Considering the fact that user’s environmental comfort is constantly changing, it is possible to say
that current POE methods only evaluated the moment in time of the measurements and surveys,
instead of taking consideration the user’s satisfaction which varies over time.
16
Ali investigated the time-dependent variation in individual thermal satisfaction by using an
adaptive stochastic modeling technique which is based on dynamic Bayesian networks. The author
illustrated that the majority of the research effort in thermal comfort adopted time-invariant
learning algorithms, although thermal comfort has been shown to change from person to person
and vary over time because of climatic variations or acclimation. To identify the variation of
occupant’s thermal perception, the statistical analysis was conducted based on the data from 33
human subjects. The result showed that personal comfort change over time with an average of
0.061ºC per day. In addition, the study suggested that not only tracking personal comfort over
time, but also considering the comfort variation of each individual person [36].
Jazizadeh developed a method which collects continuous occupant environmental perception
based on a smart phone application to improve both building energy efficiency and occupant
comfort. The study showed that user’s perceptions vary from actual IEQ condition due to various
factors such as building schedules and occupancy, occupant activity and preferences, weather and
climate, and the placement of sensors. Although current industry guidelines improve the quality
of the indoor environment and occupant comfort surveys are sometimes carried out, they are
commonly limited to one-time that do not fully represent user’s feedback throughout building
operation. Unlike conventional occupant satisfaction surveys, the new participatory sensing
application included a few question rather than a comprehensive list of questions to encourage fast
and frequent input. In addition, a pilot study was conducted in eight rooms, on different floors of
a university building to validate the application [37].
To compensate for the uncertainty of a one-time survey and measurement of data, the objective of
this research is to suggest an effective POE method that uses multiple data collections, with
consideration of time-varying IEQ conditions. For this study, a series of on-site measurements are
made to collect environmental quality data. Based on the database, the authors conducted statistical
and comparison analyses to determine more trustworthy POE methodologies, that could counteract
current limitations.
17
CHAPTER 3: METHODOLOGY
The goal was to validate current POE and IEQ study methods to enhance a technical idea for data
collection and user satisfaction survey. To do this, three objectives were established:
1. Identify a time of day and/or seasonal effect on user satisfaction depending on climate
and IEQ conditions
2. Investigate the correlation between IEQ and user satisfaction as a function of
climate/season/time of day
3. Validate these research findings via using multiple site studies with repetitive on-site
measurement and surveys
To perform advanced POE research, datasets were collected by IEQ measurement and satisfaction
surveys. Before gathering datasets, three offices which are located in different cities in Southern
California were selected as sample offices. Building selection was conducted based on
environmental conditions and building attributes. First data was assembled through an on-site
measure of IEQ factors (e.g., air temperature, radiant temperature, illuminance, unified glare rating
(UGR), noise level, air quality, etc.) at the workstation. At the same time, a continuous IEQ
measurement device called HOBO was installed in five different zones of each office. In addition,
occupant satisfaction surveys were carried out at the same time as IEQ measurement to understand
the employees’ subjective environmental satisfaction with their ambient environment. Based on
the collected dataset, statistical analyses were conducted to evaluate and demonstrate the
effectiveness of the enhanced POE method. Figure 1 illustrates the process of research
methodology.
18
Figure 1. Research Methodology Diagram
3.1 Building selection
The data collection of the study occurred in an office environment of modern buildings in
California. Three buildings were finally chosen as sample building based on environmental
conditions including climate zone and building attributes. Two of the offices are located in
downtown Los Angeles surrounding by high rise buildings. This area belongs to climate zone 9
which is established by the State of California. The third office is in the City of Irvine which is
climate zone 8. Compared with building #1 (Figure 2 (A)) and #2 (Figure 2 (B)), building #3
(Figure 2 (C)) has less urban density. Although the sampled buildings all in the Southern California
area which has a mostly mild or moderate climate, it might show different impact integrated with
building features. Figure 2 shows the image of each building.
19
Building #1 (A) [38] Building #2 (B) [39] Building 3 (C) [40]
Figure 2. Image of each building
3.2 IEQ measurement
To figure out the environmental condition of each workstation, actual IEQ measurements such as
thermal, lighting, air quality, and acoustics were collected by two types of sensing equipment. First
was an IEQ cart (Figure 3 (left)), called “e-BOT”, that was developed by the Human-Building
Integration Lab (HBI LAB) [41] to collect IEQ condition of each workstation. This cart was
designed to gather temperature data at four different levels from the floor. The cart also measured
relative humidity (RH), carbon-dioxide (CO2), particulate matters (PM), noise level and total
volatile organic compounds (TVOC) at 1.1 m height, defined as the “breathing zone” by the
American Society of Heating, Refrigerating, and Air-Conditioning Engineers (ASHRAE). To
transfer actual IEQ measurements to a database, the research team also designed a data logger
program using Lab VIEW (Figure 3 (right)) [42]. The interface is coded to collect IEQ data for 5
min and save the average data of last 2 minutes with the IEQ cart. In addition to the IEQ cart, some
hand-held sensors and a high dynamic range (HDR) camera were simultaneously used to measure
lighting levels, radiant temperature, UGR, and air velocity at the workstation.
Moreover, 5 HOBO sensors (Figure 4) were installed in five different zones of each office for a
week to collect long-term IEQ data which can reveal the variation of IEQ conditions. The sensors
can collect three IEQ factors such as temperature, relative humidity, and CO2. IEQ data was
automatically saved in memory with 10 minutes sensing intervals for recording. The measurement
was conducted every two months. To achieve a variety of ambient IEQ conditions, the
measurements were performed in different times and conditions such as work week within the
month, day of the week and time of day.
20
Figure 3. Indoor Environment Quality measurement cart (left) and sensing interface (right)
Figure 4. The image of HOBO sensor [43]
3.3 Satisfaction survey
To determine occupants’ satisfaction with their indoor condition, occupant satisfaction surveys
were conducted to collect the users’ feedback regarding their ambient environment. The paper-
based survey was designed and distributed at the same time as the on-site measurement. The survey
was made based on Cost-Effective Open-Plan Environments (COPE) environmental satisfaction
questionnaire developed by the National Research Council Canada to support the COPE project in
21
2003 [44]. Based on the COPE survey, the study customized and add some questions to fit the
research direction and goal. The final form of satisfaction survey consists of a 30-question
questionnaire which asks about the satisfaction level of various IEQ condition, the physical layout
of the workspace and human factors. Table 2 shows a summary of survey questions and their
categories.
Category Question
Q01
Occupant
Information /
Human
Factor
How many years have you been working in this building?
Q02 In a week, numbers of hours do you spend at your workstation?
Q03 What is your age?
Q04 What is your gender?
Q05 What is your job category?
Q06
Workspace
Condition
How satisfied are you with your job/institution?
Q07
How satisfied are you with the size of your personal workstation to
accommodate your work, materials, and visitors?
Q08 How satisfied are you with the level of privacy in your workspace?
Q09
How satisfied are you with your ability to alter physical conditions in
your work area? (e.g. operable windows, blinds)
Q10
How satisfied are you with your access to a view of outside from where
you sit?
Q11
How satisfied are you with the distance between you and other people
you work with?
Q12
How satisfied are you with the degree of enclosure of your work area by
walls, screens, or furniture?
22
Q13
Thermal
Quality
How satisfied are you with the current temperature at your workspace?
Q14 How satisfied are you with the thermostats? (e.g. operability)
Q15 Air Quality
How satisfied are you with the current air quality in your workspace? (i.e.
stuffy/stale air, cleanliness, odors)
Q16
Acoustic
Quality
How satisfied are you with the amount of noise from other people's
conversations while you are at your workstation?
Q17
How satisfied are you with the amount of background noise you hear at
your workstation? (i.e. not speech, noise from mechanical systems)
Q18
Visual
Quality
How satisfied are you with the light for doing computer work?
Q19
How satisfied are you with the amount of reflected light or glare on the
computer screen?
Q20
How satisfied are you with the amount of direct glare from light fixtures?
(ex: high luminance’s that are visible from a viewer’s position)
Q21 How satisfied are you with the amount of direct glare from daylight?
Q22
How satisfied are you with the quality of the lighting in your work area?
(combined artificial and daylighting)
Q23
Spatial
Factor
How satisfied are you with the general building and office layout?
(workstation layout)
Q24
How satisfied are you with the colors and textures of the flooring,
furniture, and surface finishes?
Q25 Productivity
Effect of environmental conditions in your workstation on personal
productivity?
Q26
Overall
Satisfaction
How satisfied are you with the indoor environment of your workspace
Q27
IEQ
Condition
Would rate the current thermal conditions as
Q28 Would rate the current air condition as
23
Q29 Would rate the current acoustic condition as
Q30 Would rate the current lighting condition as
Table 2. Occupant Satisfaction Survey
A 7-point scale was used to get detailed satisfaction levels for each IEQ component: -3: very
dissatisfied, -2: dissatisfied, -1: slightly dissatisfied, 0: neutral, +1: slightly satisfied, +2: satisfied,
+3: very satisfied. However, the responses were transformed into the lower level of point scale to
carry out a statistical analysis that better represents the tendency and correlation of result. Table 3
illustrates the points of each scale.
7-point Scale -3 -2 -1 0 +1 +2 +3
3-point Scale Negative Neutral Positive
Table 3. Summary of scale points of occupant satisfaction survey
3.4 Statistical analysis
With above-mentioned industry standards and the collected dataset, cross-sectional analyses were
conducted. The aggregated data was strategically categorized by selected months and human
factors like age group and gender to conduct comparative analysis. To define the variation of user’s
environmental perception in five different measurement time, the one-way analysis of variance
(ANOVA) and Two-sample T-test were mainly adopted.
24
3.4.1 Correlation analysis
Correlation Analysis estimates a sample correlation coefficient, more specifically, the Pearson
Product Moment correlation coefficient. The sample correlation coefficient denoted “r”, ranges
between !1 and þ1 and quantifies the direction and strength of the linear association between the
two variables. The correlation between two variables can be either positive or negative[45].
3.4.2 Two-sample T-test
The two-sample T-test is a hypothesis test that calculates a confidence interval and the difference
between two random sample means based on the t-distribution. Two hypotheses including null
hypothesis and alternative hypothesis are set when the T-test is performed. To conduct a two-
sample T-test, the two samples must be drawn independently from each other. It is commonly used
with small samples to examine the difference between the samples when the variance of two
normal distributions is not known. Confidence in the results are enhanced when the sample size is
increased. The test is commonly adopted to test and determine whether a new development or
procedure is more advanced than current process or development [46].
3.4.3 Analysis of Variance (ANOVA)
The one-way Analysis of Variance (ANOVA) is used to determine whether the mean of a
dependent variable is the same in two or more unrelated, independent groups of an independent
variable. However, it is typically only used when you have three or more independent, unrelated
groups, since an independent t-test is more commonly used when you have just two groups [47].
3.4.4 Decision Tree (J48)
Data mining is the computational process that diagnoses large data sets and finds patterns. The
overall goal of data mining is to figure out valuable information from collected data and transform
25
it into a visualized format which the user can easily understand. As a data mining model, Decision
Tree (J48) decides the dependent variable of a new sample based on various categories of the data.
It provides tree shape decision guideline which represents driving factors for making a specific
decision. Weka that was developed by a machine learning group at the University of Waikato in
New Zealand is used to run decision tree algorithms. The software automatically analyses the data
by different attributes and decides what information is most important[48], [49].
26
CHAPTER 4: RESULT
4.1 Overview data analysis
The on-site measurements and user’s satisfaction surveys were simultaneously carried out in the
three sample offices in Southern California during five different months (one time per each month).
Table 4 summaries the data collection time and weather condition for each office. Based on the
aggregated data which consists of measured IEQ data and survey results, statistical analysis was
conducted to investigate and define the relationship between IEQ elements, each building’s
occupants’ satisfaction, and ambient factors.
January April June September November
Office A
Date 2018.1.16 (Tue) 2017.4.11 (Tue) 2017.6.22 (Thur) 2017.9.14 (Thur) 2017.11.17 (Fri)
Time 14:00 - 16:30 14:00 - 18:00 10:00 - 15:00 13:00 - 15:00 10:00 - 13:00
Weather condition Sunny Sunny Partly cloudy Sunny Sunny
Office B
Date 2018.1.23 (Tue) 2017.4.18 (Tue) 2017.6.8 (Thur) 2017.9.28 (Thur) 2017.11.27 (Mon)
Time 14:00 - 16:00 14:00 - 17:00 10:00 - 14:00 11:00 - 14:00 13:00 - 16:00
Weather condition Sunny Partly cloudy Partly sunny Sunny Partly cloudy
Office C
Date 2018.1.30 (Tue) N 2017.6.14 (Thur) 2017.9.21 (Thur) 2017.11.10 (Fri)
Time 13:30 - 14:00 N 14:00 - 17:30 14:00 - 16:30 09:30 - 12:00
Weather condition Sunny N Sunny Sunny Partly cloudy
Table 4.. Summary of measurement time and weather condition
To help the understanding of the dataset, the study categorized the collected dataset for each
selected office: Office A, which is the largest office and located in Los Angeles, Office B, located
in Irvine, California and Office C, located in Los Angeles. Furthermore, to show a distribution of
the collected data, the whole dataset was grouped by the offices and the human factors of the data.
As summarized in Table 5, a total of 315 subjects of three offices participated in the study. For the
office A, there were 76 male users and 59 female users, ranging in the age from 18 to 50+ years
old. A total of 67 occupants was in the age group 18 to 29 years old, 63 in the Mid-age group (30
– 49 years old), and 5 in the Senior group (50+ years old). Overall, 25 or more occupants joined
27
the research in each measurement time. For the office B, a total of 89 occupants including 63 male
users and 35 female users answered the satisfaction survey. In total, 34 users were categorized in
the Junior group (18 – 29 years old), 39 in the Mid-age group (30 – 49 years old), and 25 in the
Senior group (50+ years old). Comparing to other offices, the office B had the larger number of
the senior group. For the office C, a total of 82 subjects which consists of 20 female occupants and
62 male occupants reported their satisfaction level with IEQ condition. A total of 36 occupants
was in the Junior group (18 - 29 years old), 38 in the Mid-age group (30 – 49 years old), and 8 in
the Senior group (50+ years old). In addition, due to schedule issue, the data acquisition of the
office C was not conducted in April.
January April June September November
Company Age group Age Female Male Female Male Female Male Female Male Female Male Total
Office A
Junior 18-29 7 9 9 5 7 4 6 9 5 6 67
Mid-age 30-39 2 2 2 8 4 6 3 8 2 6 43
40-49 2 2 2 3 3 2 1 2 1 2 20
Senior 50+ 1 0 1 0 1 1 0 1 0 0 5
Subtotal 12 13 14 16 15 13 10 20 8 14 135
Office B
Junior 18-29 2 4 3 6 3 3 3 4 2 4 34
Mid-age 30-39 0 4 0 2 0 4 0 8 0 6 24
40-49 2 0 3 4 2 1 1 0 2 0 15
Senior 50+ 3 4 1 3 3 1 2 2 3 3 25
Subtotal 7 12 7 15 8 9 6 14 7 13 98
Office C
Junior 18-29 3 8 0 0 1 7 1 7 2 7 36
Mid-age 30-39 2 5 0 0 2 6 3 7 2 3 30
40-49 1 1 0 0 0 1 0 3 0 2 8
Senior 50+ 0 0 0 0 1 3 1 0 1 2 8
Subtotal 6 14 0 0 4 17 5 17 5 14 82
Total 25 39 21 31 27 39 21 51 20 41 315
Table 5. Demographic Information of dataset by human factors
28
Moreover, to identify the impact of time functions, the aggregated dataset was also categorized by
the offices and time factors. The time factors include a day (Monday, Tuesday, Thursday, and
Friday), a time (Morning and Afternoon), and a week (Early of the week and Late of the week).
Early of the week consists of Monday and Tuesday, and Thursday and Friday belong to Late of
the week. Table 6 illustrates the number of data depending on the time factors by three sampled
offices.
Day Time Week
Company Monday Tuesday Thursday Friday Morning Afternoon Early of week Late of week
Office A 0 55 58 22 50 85 55 80
Office B 20 41 37 0 17 81 61 37
Office C 0 20 43 19 19 63 20 62
Table 6. Demographic Information of dataset by time factors
4.2 Case 1: Office A
Various IEQ elements were measured to diagnose the environmental statue of each workstation
and to perform a comparative study with the occupant’s satisfaction survey responses by using
statistical analysis. In this section, the distribution of the measured IEQ factors is illustrated. The
red box indicates the comfortable range or recommended guideline, of the current industry
organizations.
29
Figure 5. Distribution of Temperature at floor level (left) and Temperature at 1.2m (right)
Distribution of temperature at two different levels is illustrated in Figure 5. In general, the
temperature at floor level was within the standard among five different months (Figure 5 (left)).
The data of January, April, and June shows the similar distribution. In addition, the temperature in
September was slightly lower than other measurement times, while the temperature in November
was marginally higher than other months. However, unlike the temperature at floor level, the
thermal condition at 1.2m was not entirely within the guideline, which is established by ASHRAE
(Figure 5 (right)). The June and September’s temperature were lower than the minimum
recommended temperature especially June’ temperature.
Figure 6. Distribution of Radiant Temperature Asymmetry wall (left) and Radiant Temperature
Asymmetry Ceiling (right)
30
Distribution of radiant temperature asymmetry wall and ceiling are shown in Fig. Overall, the
office A had a good thermal condition regarding radiant temperature. The majority of data are
within the comfort zone, even if there are some exceptions. According to the ASHRAE standards,
the radiant temperature asymmetry cool wall and warm ceiling should be lower than 5
o
C and 10
o
C. In November, few data of radiant temperature asymmetry wall were higher than the
recommended zone (Figure 6 (left)). In addition, the data of radiant temperature asymmetry ceiling
indicated that some workstations reported the high-temperature gap which is higher than 5
o
C in
January and September (Figure 6 (right)).
Figure 7. Distribution of Relative Humidity (left) and CO
2
concentration (right)
Distribution of Relative Humidity and CO
2
concentration are illustrated in Fig. The whole data of
RH is placed in the comfort zone, which is the maximum 65% for the office building (Figure 7
(left)). However, those data measured in January and April showed relatively lower value
comparing to other three months (June, September, and November). Similar to RH, CO
2
levels
were also fully within suggested guideline (Figure 7 (right)). The CO
2
data for January were
relatively higher than other four months, with a range of 840ppm to 900ppm.
31
Figure 8. Distribution of Work Surface Illuminance (left) and UGR (right)
Distribution of lighting quality data is shown in Figure 8 The office had a poor lighting condition
based on the corresponding standards. There were some measured illuminance data which were
lower than the minimum suggested requirement of the IESNA in every measured time (Figure 8
(left)). In addition, the UGR data illustrated similar problem same as illuminance (Figure 8 (right)).
The International Commission on Illumination (CIE) established the guideline of UGR which
suggests the value between 13 to 19. However, many workstations’ UGR value in five different
months were located on the outside of the guideline especially in September.
Figure 9. Distribution of Noise level
32
Distribution of acoustic decibel is illustrated in Figure 9. Although the measured noise levels in
January, September, and November were significantly lower than the data of April and June, the
whole data were out of the comfort range which is the maximum 40 dBA. The majority of April
and June’s acoustic data were in a range between 60 to 68 dBA, which is notably higher than the
current standard. Those distributions indicated that occupants of the office A were exposed in a
noisy indoor environment.
January April June September November
Variables M S W M S W M S W M S W M S W
Temperature
floor ('C)
22.06 0.50 100% 23.44 0.44 100% 22.65 0.50 100% 21.97 0.53 100% 23.96 0.56 100%
Temperature
1.2m ('C)
23.79 0.40 96% 23.83 0.57 94% 23.32 0.31 54% 22.73 0.33 0% 24.30 0.63 96%
Vertical Air
Temperature
Difference
0.75 0.33 100% 0.40 0.26 100% 0.68 0.40 100% 0.78 0.44 100% 0.34 0.27 100%
Radiant
Temperature
Asymmetry
Wall ('C)
2.24 2.03 100% 2.07 1.85 100% 2.11 1.38 100% 1.92 1.71 100% 4.25 5.11 86%
Radiant
Temperature
Asymmetry
Ceiling ('C)
2.16 2.47 92% 1.95 1.34 94% 2.14 1.21 100% 3.46 2.42 87% 1.30 0.75 100%
Relative
Humidity (%)
37.13 2.12 100% 35.42 0.85 100% 54.98 0.88 100% 54.83 1.49 100% 49.25 1.44 100%
CO2 level
(ppm)
867.9 19.84 100% 714.9 20.48 100% 684.5 27.05 100% 709.1 44.72 100% 719.5 60.94 100%
Work surface
illuminace (lux)
N N N 241.9 148.1 57% 233.8 123.1 54% 251.5 131.0 63% 203.9 116.8 46%
Unified Glare
Rating (UGR)
16.36 2.77 78% 17.6 2.57 63% 15.08 2.83 62% 19.3 3.03 37% 16.71 3.1 63%
Acoustic
decibel (dBA)
46.2 3.26 0% 65.0 1.10 0% 64.4 1.05 0% 49.5 3.84 0% 46.3 3.09 0%
Table 7. Summary of measured IEQ data by month (M = Mean, S = StDev, W = Within
guideline)
Table 7 illustrates the summary of the measured IEQ data by five different months. Most
temperature data were within ASHRAE’s thermal comfort zone [6]. However, a temperature,
measured at the height of 1.2m, indicates that 46% of those measured in June and 100% of data in
September were out of the recommended range with a mean value of 23.32
o
C and 22.73
o
C. Those
values were marginally lower than the minimum recommended temperature. CO
2
concentrations
and relative humidity levels were fully within the relevant industry guidelines, and the mean values
of five months were 739ppm and 46.32%, respectively. The average illuminance for the four
33
months was 232.77 lux, and 57%, 54%, 63% and 46% of the measured data were within the
comfort range of the IESNA [7], respectively. 78%, 63%, 62%, 37% and 63% of the UGR data of
the chosen five measurement times were within the CIE’s standard [8]. In general, around 50% of
workstations were placed in the area which had a poor lighting condition based on IESNA’s
standard which suggests illuminance level between 200lux to 500lux. Furthermore, even if
acoustic levels in January, September and November were noticeably dropped compared with
April and June’s acoustic level, all the measured noise levels were higher than 40 dBA which is
the allowed maximum level of ASHRAE’s open plan office acoustic guideline [50]. The average
values of the acoustic level in five different months were 46.2 dBA, 65.0 dBA, 64.4 dBA, 49.5
dBA, and 46.3 dBA. This result indicated that users of the office A were surrounded by relatively
excessive background noise, as compared to levels suggested by current guidelines.
Figure 10. Rose chart of the survey question by month
-1.00
-0.50
0.00
0.50
1.00
1.50
2.00
2.50
Job/Institution
Size of workstation
Level of privacy
Alterability of physical condition
Accessibility of outside view
Distance between others
Degree of enclosure of work area
Temparature
Operability of thermostats
Air quality
Background noise (conversation) Background noise (mech system)
Lighting condition for computer
work
Glare in the computer screen
Direct glare from lighting fixture
Direct glare from daylight
Lighting quality
Workstation layout
Color and textures of finishes
Productivity
IEQ
01_January 04_April 06_June 09_September 11_November
34
Figure 10 shows an average score for the occupants’ responses to environmental satisfaction
survey questions. The survey used a 7-point scale which ranged from -3 (very uncomfortable) to
+3 (very comfortable), with “0” for neutral. On the average, the occupants reported neutral or
positive satisfaction with their IEQ conditions and spatial criteria. Although the average response
was positive, satisfaction with some IEQ factors, including operability of thermostats, background
noise, and level of privacy, was relatively lower than that for other components. Overall,
occupant’s responses in April and November were slightly lower than other selected months
regarding IEQ elements. The patterns of the each of five months’ lines were notably similar. In
most parameters, the overall average satisfaction in January and June were slightly higher than
that in other months, even though there was no remarkable difference in the measured IEQ data
reported for these months. However, the average satisfaction on overall IEQ condition is slightly
lower in June than other months, while the satisfaction with individual IEQ components and spatial
factors was relatively higher. In addition, this study also found that a few specific IEQ parameters
including thermal, acoustic and lighting quality showed significantly different results within five
months. In addition, statistical analysis revealed that human factors, such as gender and age,
seemed to affect the user’s environmental satisfaction. A detailed discussion is available in Section
4.2.2.
4.2.1 The change of occupant’s responses
Table 8 illustrates how much occupant’s satisfaction levels and actual IEQ conditions were
changed by using statistical analysis. The numeric values indicate a p-value which shows the
difference of the data by group categories. In addition, the numeric values with an asterisk describe
the p-values which report there is a remarkable difference between variable groups (i.e., different
months in this analysis). Those p-values were mainly generated by ANOVA test. In general,
occupant’s satisfaction levels on specific IEQ factors such as thermal, air quality, and acoustic
condition showed constancy, even though there was a change of the answers in male group and
the age group (18 to 29 years old) regarding acoustic condition. However, unlike environmental
35
satisfaction, the actual IEQ conditions were significantly varied. This result reveals that occupant
might have an ability to adjust their ambient environment, which is unstable, and have a constant
environmental satisfaction level concerning thermal, air quality, and acoustic condition. In
addition, it might be possible to conclude that one-time data collection for current POE study can
be dependable, although it would be necessary to perform multiple data acquisition which can
consider the variety of occupant’s environmental satisfaction, affected by age and gender, and
provide a more sophisticated result. A detailed discussion about the effect of human factors on
occupant’s satisfaction is available in Section 4.2.2.
Occupant satisfaction survey
Thermal satisfaction IAQ satisfaction Acoustic satisfaction
Data Categories Q13 Q14 Q15 Q16 Q17
ALL 0.8659 0.8384 0.2197 0.4654 0.0365*
FEMALE 0.7609 0.476 0.7134 0.7052 0.7024
MALE 0.6704 0.637 0.5134 0.229 0.0325*
AGE (18-29) 0.5654 0.8197 0.7881 0.8672 0.0628*
AGE (30-39) 0.5747 0.4654 0.5414 0.4983 0.2036
AGE (40-49) 0.9681 0.9654 0.236 0.8746 0.6803
Actual IEQ
Thermal condition Air quality Acoustic condition
Data Categories Temperature CO2 Humidity Acoustic
ALL <0.000* <0.000* <0.000* <0.000*
FEMALE <0.000* <0.000* <0.000* <0.000*
MALE <0.000* <0.000* <0.000* <0.000*
AGE (18-29) <0.000* <0.000* <0.000* <0.000*
AGE (30-39) <0.000* <0.000* <0.000* <0.000*
AGE (40-49) 0.0009* <0.000* <0.000* <0.000*
Table 8. Summary of P-value from ANOVA test by different data categories (* indicates a P-
value below 0.1)
4.2.2 Impact of human factors
As mentioned in the previous section, although the overall patterns of the survey illustrated
significant similarities between datasets of the selected five months, there were some differences
in survey responses based on human factors.
36
Male Group
(A) Interval plot of acoustic quality satisfaction (B) Interval plot of actual acoustic decibel
Female Group
(C) Interval plot of acoustic quality satisfaction (D) Interval plot of actual acoustic decibel
Table 9. Confidence interval of acoustic quality satisfaction (A); measured acoustic decibel (B);
acoustic quality satisfaction (C); measured acoustic decibel (D) by month
Table 9 (B) and Table 9 (D) illustrates actual noise level of each gender group. Both graphs
describe a significantly similar pattern, and there was a remarkable increase of acoustic decibel in
June and a significant decrease of the noise level in September. Even if the data of January,
September, and November were notably lower than other two months, the data were still above 40
dBA, which is the maximum value of the recommended range. The analyses found that there was
an effect of gender on occupant’s acoustic perception. Table 9 (A) shows that the male group’s
acoustic satisfaction especially background noise from the mechanical system was inconstant with
a p-value of 0.0325 which is statistically significant. The male group was notably more satisfied
with their acoustic condition in January and June, even though the actual background noise was
higher than the recommended guideline. In addition, despite of the similar pattern of acoustic
satisfaction in January and June, those measured noise levels in both months were significantly
different. However, as shown in Table 9 (C), female group’s responses were persistent (with a p-
37
value of 0.7024) with the almost same acoustic condition as the male group. This result reveals
that the male users might be less sensitive to low sound pressure, which is typically created from
mechanical system, duct, and fan, than female users. Since this study only measured general
acoustic levels of the workstation in dBA, it is hard to classify the type of sound sources. However,
it might be possible to conclude that gender group affects occupant’s acoustic perception which
can cause the change of user’s satisfaction.
(A) Interval plot of acoustic quality satisfaction (B) Interval plot of actual acoustic decibel
Table 10. Confidence interval of acoustic quality satisfaction (A) and measured acoustic decibel
(B) by month (18 to 29 years old only)
Similar to the male group, the junior group also showed the variation of acoustic satisfaction, while
another age groups reported the constant satisfaction levels with background noise especially the
noise from the mechanical system. As shown in Table 10 (A), the age group’s (18 to 29 years old)
answers were significantly varied in January and June with a p-value of 0.0628, even though all
age group’s acoustic conditions were almost same. Table 10 (B) illustrated the actual noise level
of the junior group’s workstations. The whole data were above 40 dBA, which is higher than the
maximum of the recommended range, and April and June’s data were notably higher than other
selected months. This result indicates that the junior group might be less sensitive to noise from
mechanical system, duct, and fan, than other age groups. In addition, although the study cannot
identify the type of sound source, it may be able to conclude that age group has an impact on user’s
acoustic perception.
38
(A) Interval plot of lighting condition satisfaction (B) Interval plot of direct glare satisfaction
(C) Interval plot of lighting quality satisfaction (D) Interval plot of UGR
(E) Interval plot of work surface illuminance
Table 11. Confidence interval of lighting condition satisfaction (A); direct glare satisfaction (B);
lighting quality satisfaction (C); measured UGR (D); measured work surface illuminance (E) by
month (Age group 40 to 49 only)
39
Q18 Q21 Q22
Lighting factors Spearman rho P-value Spearman rho P-value Spearman rho P-value
Work surface illuminance -0.3378596 0.2021 -0.524039 0.0660 -0.677089 0.0110
UGR 0.508212 0.0313 0.248449 0.3202 0.118313 0.6401
Table 12. Summary of correlation analysis
In general, subjects had a persistent satisfaction on their lighting condition in five different months.
However, as shown in Table 11, the mid-age group (40-49 years old) reported a notably different
satisfaction with a p-value of 0.0029 (Table 11 (A)). In addition, the analyses illustrate that this
pattern was found in most user’s feedback relevant to lighting quality. Table 11 (B) indicates the
mid-age group’s responses about the satisfaction with direct glare from daylighting was inconstant
(with a p-value of 0.00767) in different data collection times, even though all user’s responses
about direct glare were relatively constant. Moreover, the 40 to 49 age group noted remarkable
change of the answers for Q22 (Table 11 (C)) which asked the satisfaction with overall lighting
quality. However, even if the mid-age group indicated inconstant satisfaction rate with their
lighting condition, actual lighting conditions of their workstations were relatively constant in five
different measurement times. As shown in Table 11 (D), the UGR values for five selected months
were relatively persistent, even though there was a slight increase in September. Furthermore, the
illuminance level of the mid-age group’s work surface was also significantly similar among
different months (with a p-value of 0.6708), although the distribution of each month was
marginally changed (Table 11 (E)). The result indicates that the age group (especially the 40 to 49
age group) has an impact on the variation of occupants’ lighting satisfaction. Moreover, other
possible components may influence the mid-age group’s lighting perception. Those components
might include other environmental or physical factors such as climate, circadian rhythm, and time
functions. In addition, the correlation result indicates that direct glare satisfaction and lighting
quality satisfaction had a significant negative correlation with work surface illuminance, while the
UGR had a notably positive correlation with lighting condition satisfaction (Table 12). This result
reveals that the 40 to 49 age group tended to have higher satisfaction levels with direct glare and
40
lighting quality when the work surface illuminance levels were lower. However, with common
sense, the mid-age group supposes to prefer higher illuminance level. Therefore, it might be
possible to conclude that the mid-age group desires low illuminance level for computer-based work.
In addition, there are other ambient factors such as monitor’s brightness and contrast which might
affect the 40 to 49 age group’s lighting preference.
4.2.3 Impact of time functions
Occupant satisfaction survey
Thermal satisfaction IAQ satisfaction Acoustic satisfaction Lighting satisfaction
Data
categories
Q13 Q14 Q15 Q16 Q17 Q18 Q19 Q20 Q21 Q22
DAY 0.9304 0.5921 0.1306 0.4667 0.4666 0.6614 0.5098 0.8602 0.1985 0.4969
TIME 0.9445 0.6325 0.2178 0.9547 0.4925 0.7888 0.6992 0.2938 0.8375 0.9397
WEEK 0.946 0.7502 0.3253 0.9259 0.5691 0.6996 0.4078 0.6108 0.8328 0.7575
Actual IEQ
Thermal condition Air quality Acoustic condition Lighting condition
Data
categories
Temperature CO 2 Humidity Acoustic Work Surface illuminance UGR
DAY <0.000* <0.000* <0.000* <0.000* 0.4617 0.7645
TIME 0.0133* <0.000* <0.000* 0.134 0.2979 0.0003*
WEEK <0.000* <0.000* <0.000* 0.1091 0.8208 0.8076
Table 13.Summary of P-value from statistical analysis by different data categories (* indicates a
P-value below 0.1)
Table 13 describes how much user’s satisfaction levels and actual IEQ conditions were varied by
using statistical analysis. The numeric values mean a p-value which can indicate the change of the
data by group categories (i.e., time factors in this analysis). Furthermore, the numeric values with
an asterisk illustrate the p-values which are lower than the value of 0.1. Those p-values were
mainly made by Two-sample t-test and ANOVA test depending on the number of variables.
Overall, occupants reported constant satisfaction levels regarding all IEQ components, even
though some corresponding IEQ factors such as air quality, thermal, and acoustic conditions were
remarkably changed. This result reveals that occupant might be possible to have relatively
consistent environmental satisfaction, even though occupant’s IEQ perception can be affected by
41
time functions. Although human’s physical and psychological conditions which influence
subject’s perception can be changed depending on different times, it might be possible that
occupants in modern offices can easily adjust to their indoor environment or are less sensitive
about the change of ambient environment because they already have a rigid idea about it. Therefore,
it might be possible to conclude that one-time data acquisition can be reliable because of
occupant’s persistent feedback at different times.
4.3 Case 2: Office B
Figure 11. Distribution of Temperature at floor level (left) and Temperature at 1.2m (right)
Distribution of temperature at two different levels is illustrated in Figure 11. Overall, the
temperature data measured at floor level were fully placed in the recommended guideline among
five different measurement times (Figure 11 (left)). The January’s data indicates that the
temperature in January was marginally lower than other four months. However, the temperature
measured at 1.2m level was partially out of the ASHRAE’ comfort zone in each data collection
time (Figure 11 (right)). Although most of November’s temperature was lower than the minimum
recommended temperature which is the value of 23.3
o
C, those data were slightly lower than the
comfort zone with a mean value of 23.02
o
C. In addition, even if there were some exceptions, a
large portion of September’s temperature was located in the suggested temperature range.
42
Figure 12. Distribution of Radiant Temperature Asymmetry wall (left) and Radiant Temperature
Asymmetry Ceiling (right)
Distribution of radiant temperature asymmetry wall and ceiling are described in Figure 12. Both
of charts illustrate that all the data were within the guideline. For radiant temperature asymmetry
wall, the data showed similar a pattern of distribution in different months, even though a few
April’s temperature difference were relatively higher than other data (Figure 12 (left)). Moreover,
the data of radiant temperature asymmetry ceiling indicated that the data of June, September, and
November showed almost same distribution, while January and April’s data were similar (Figure
12 (right)). Furthermore, some workstations in January and April had significantly higher
temperature gap than other workstations.
Figure 13. Distribution of Relative Humidity (left) and CO
2
concentration (right)
43
Distribution of Relative Humidity and CO
2
concentration are illustrated in Figure 13. In general,
the office had a good air condition among five different months. All humidity and CO
2
data except
to few workstations in April were within the relevant comfort zone. The RH data of January were
notably lower than other months’ data (Figure 13 (left)). For CO
2
concentration, June and
November’s CO
2
levels were slightly lower than CO
2
data in January, April, and September
(Figure 13 (right)). In addition, as mentioned above, few workstations in April had higher CO
2
level than the suggested guideline, which is the maximum value of 1000ppm.
Figure 14. Distribution of Work Surface Illuminance (left) and UGR (right)
Distribution of lighting quality data is shown in Figure 14. Comparing to thermal condition and
air quality, lighting quality of the office was significantly worse. The majority of illuminance data
in five different months were placed in outside of the comfort zone (Figure 14 (left)). The data
showed that those workstations which had poor work surface illuminance levels reported notably
lower illuminance levels than the minimum requirement. In addition, the UGR data also described
similar problem same as illuminance (Figure 14 (right)). The large portion of workstations’ UGR
values in five measurement times was out of the guideline, and those data were mostly lower than
the value of 13 which is the minimum recommended range by CIE.
44
Figure 15. Distribution of Noise level
Distribution of acoustic decibel is illustrated in Figure 15. Even if the distribution of each months’
data was varied, all measured acoustic data were out of the comfort zone which is the maximum
value of 40 dBA. The April and June’s noise levels were relatively higher than other three months.
In addition, although the whole acoustic data of November was placed in the outside of the
recommended range, the majority of the data were closed to 40 dBA. Those distributions indicated
that occupants of the office B were also exposed in noisy condition same as office A.
45
January April June September November
Variables M S W M S W M S W M S W M S W
Temperature
floor ('C)
22.60 0.35 100% 23.02 0.42 100% 23.14 0.19 100% 23.35 0.27 100% 22.84 0.31 100%
Temperature
1.2m ('C)
23.32 0.32 47% 23.31 0.38 68% 23.31 0.20 59% 23.52 0.26 85% 23.02 0.28 10%
Vertical Air
Temperature
Difference
0.72 0.33 100% 0.30 0.20 100% 0.21 0.10 100% 0.28 0.25 100% 0.20 0.13 100%
Radiant
Temperature
Asymmetry
Wall ('C)
0.64 0.93 100% 1.57 1.47 100% 1.04 0.66 100% 0.81 0.63 100% 0.54 0.45 100%
Radiant
Temperature
Asymmetry
Ceiling ('C)
1.18 1.14 100% 0.74 0.73 100% 0.25 0.27 100% 0.47 0.42 100% 0.26 0.23 100%
Relative
Humidity (%)
27.10 0.45 100% 50.74 1.89 100% 57.58 0.53 100% 53.06 1.80 100% 45.18 0.68 100%
CO2 level
(ppm)
782.4 56.09 100% 726.8 66.55 95% 660.2 29.04 100% 745.8 17.09 100% 622.0 45.94 100%
Work surface
illuminace (lux)
119.0 60.58 5% 132.9 86.39 14% 136.5 78.05 18% 119.9 86.90 20% 129.1 107.4 25%
Unified Glare
Rating (UGR)
12.55 1.48 33% 13.52 1.94 60% 13.24 1.55 53% 13.51 2.87 38% 14.81 2.87 75%
Acoustic
decibel (dBA)
54.3 2.25 0% 62.6 3.78 0% 62.4 2.44 0% 48.4 4.03 0% 44.5 2.53 0%
Table 14. Summary of measured IEQ data by month (M = Mean, S = StDev, W = Within
guideline)
Table 14 illustrates the statistical summary of measured IEQ elements by five different months:
January, April, June, September, and November. Overall, the data regarding thermal condition
were mostly within the comfort zone, which is established by ASHRAE. However, the
temperatures measured at 1.2m were significantly placed in outside of the recommended zone. The
average values of the temperatures measured at 1.2m were approximately around 23.3
o
C.
Therefore, despite the large percentage of the data was out of the comfort zone, the actual
temperatures were only marginally lower than the minimum value of the comfortable range. The
relative humidity levels and CO
2
concentrations fall within the comfort zones. Unlike thermal
condition and air quality, the work surface illuminance, UGR, and Acoustic decibel were
significantly out of the recommended guidelines especially acoustic levels were fully out of the
guideline. The average illuminance levels of five measured times were 119.0 lux, 132.9 lux, 136.5
lux, 119.9 lux, and 129.1 lux, which are relatively lower than the minimum suggested. In addition,
the UGR data show 67%, 40%, 47%, 62% and 25% of the workstations were located in the outside
of the CIE’s recommended comfort range. However, there was a significant increase of the UGR
in November with an average value of 14.81, which is slightly higher than the minimum
46
requirement of CIE’s standard. Similar to Office A, there were no workstations which had a
comfortable noise level with the average value of 54.3 dBA, 62.6 dBA, 62.4 dBA, 48.4 dBA, and
44.5 dBA, which is remarkably higher than 40 dBA, the maximum level suggested by the
ASHRAE.
Figure 16. Rose chart of the survey question by month
As shown in Figure 16, the users were mostly satisfied or neutral with their indoor environment.
Overall, most criteria reach a positive value of satisfaction, even though some questions regarding
background noise, operability of thermostats, alterability of physical condition and accessibility of
view indicated relatively lower satisfaction levels than other factors. Even if the satisfaction levels
in different measurement times were marginally different, the patterns of linear lines are
-1.00
-0.50
0.00
0.50
1.00
1.50
2.00
2.50
Job/Institution
Size of workstation
Level of privacy
Alterability of physical condition
Accessibility of outside view
Distance between others
Degree of enclosure of work area
Temparature
Operability of thermostats
Air quality
Background noise (conversation) Background noise (mech system)
Lighting condition for computer
work
Glare in the computer screen
Direct glare from lighting fixture
Direct glare from daylight
Lighting quality
Workstation layout
Color and textures of finishes
Productivity
IEQ
01_January 04_April 06_June 09_September 11_November
47
considerably similar. Occupants had relatively lower satisfaction on lighting condition in
November. In addition, users were significantly less satisfied with their overall IEQ condition in
June. Similar to the results for Office A, this study also found that human factors have an impact
on the environmental perceptions of a building’s occupant.
4.3.1 The change of occupant’s responses
Occupant satisfaction survey
Thermal satisfaction IAQ satisfaction Acoustic satisfaction
Data Categories Q13 Q14 Q15 Q16 Q17
ALL 0.3832 0.8986 0.5909 0.5148 0.9286
FEMALE 0.039* 0.7327 0.7392 0.902 0.8581
MALE 0.2208 0.8571 0.4564 0.6167 0.8147
AGE (18-29) 0.1721 0.7427 0.5612 0.2251 0.8148
AGE (30-39) 0.6491 0.6059 0.9936 0.879 0.2136
AGE (40-49) 0.4438 0.8237 0.8871 0.8418 0.8326
AGE (50+) 0.9674 0.3453 0.5561 0.0778* 0.7773
Actual IEQ
Thermal condition Air quality Acoustic condition
Data Categories Temperature CO2 Humidity Acoustic
ALL <0.000* <0.000* <0.000* <0.000*
FEMALE 0.1153 0.0002* <0.000* <0.000*
MALE <0.000* <0.000* <0.000* <0.000*
AGE (18-29) 0.4696 <0.000* <0.000* <0.000*
AGE (30-39) 0.000* 0.0002* <0.000* <0.000*
AGE (40-49) 0.3381 0.018* <0.000* <0.000*
AGE (50+) 0.0128* <0.000* <0.000* <0.000*
Table 15. Summary of P-value from ANOVA test by different data categories (* indicates a P-
value below 0.1)
Table 15 illustrates the variation of occupant’s response and actual IEQ condition between five
different measured times by using ANOVA test. The numeric values with the asterisk mean the p-
values which had the value lower than 0.1. Although the female group’s thermal satisfaction and
the 50+ age group’s acoustic satisfaction showed the variety of satisfaction in five months, overall
user’s satisfaction with their indoor conditions concerning thermal, air quality, and acoustic
condition was relatively constant. However, similar to Office A, the actual IEQ elements including
48
temperature, CO
2
concentrations, humidity, and acoustic decibel were significantly varied among
five different data collection times. Even if the actual IEQ conditions were notably changed, the
actual value of the differences was not huge. This result also reveals that occupant’s environmental
satisfaction might be insensitive to the change of IEQ conditions if the variation is not extreme.
Therefore, it might be possible to believe that current POE method which uses one-time data
acquisition is dependable, even though some ambient components can affect occupant’s
environmental satisfaction.
4.3.2 Impact of human factors
Male Group
(A) Interval plot of thermal quality satisfaction (B) Interval plot of actual temperature
Female Group
(C) Interval plot of thermal quality satisfaction (D) Interval plot of actual temperature
Table 16. Confidence interval of thermal quality satisfaction (A); measured temperature (B);
thermal quality satisfaction (C); measured temperature (D) by month
49
Table 16 (B) illustrates the measured temperature of male group’s workstation showed notable
changes by different five different months (with a p-value of <0.0001). However, as shown in
Table 16 (D), the actual temperature of female group’s workstation was relatively less varied
compared to the male group (with a p-value of 0.1153), even though there was a significant
temperature drop in November. Despite the variety of the actual thermal condition, the answers of
user’s thermal comfort indicated the opposite pattern. Table 16 (A) represents that the male group’s
satisfaction with their thermal environment was respectably persistent with a p-value of 0.2208,
while the female group’s thermal comfort level (Table 16 (C)) was remarkably varied between
five different months (with a p-value of 0.039). This result reveals that female group might be
relatively more sensitive to the change of thermal condition than the male group, even though the
variation in temperature is not huge. In addition, external factors including a seasonal effect, or the
effect of the time functions might have a large effect on a female’s thermal perception.
50
All Users
(A) Interval plot of glare condition satisfaction (B) Interval plot of actual UGR
Age Group 30-39
(C) Interval plot of glare condition satisfaction (D) Interval plot of actual UGR
Table 17. Confidence interval of glare condition satisfaction (A); measured UGR (B); glare
condition satisfaction (C); measured UGR (D) by month
Table 17 (B) and Table 17 (D) illustrate the UGR data for five different months. The graphs show
that the UGR data of all user group was relatively more widespread and inconstant (with a p-value
of 0.0496) than the mid-age group’s data, even though the 30 to 39 age group’s UGR data also
showed a variation by different measured times (with a p-value of 0.1502). However, the survey
result reports that the mid-age group’s responses (Table 17 (C)) were notably inconstant between
five months (with a p-value of 0.0214), although their real glare condition had fewer changes than
the overall glare condition. In addition, although the mid-age group’s feedback showed the
significant change of lighting satisfaction, whole user group’s satisfaction levels (Table 17 (A))
were relatively persistent (with a p-value of 0.6567). This result reveals that the mid-age group (30
to 39 years old) might have notably more sensitive glare perception, while the other age groups
have some adjustable range of lighting perception. Moreover, the correlation analysis indicates
that the mid-age group’s glare satisfaction had a negative correlation with their actual UGR data
51
(with a Spearman rho of -0.370188 and p-value of 0.0899). It indicates that the age group (30 to
39 years old) might tend to prefer and have higher glare satisfaction with low UGR value, while
other age groups do not show any correlation.
All Users
(A) Interval plot of acoustic quality satisfaction (B) Interval plot of actual acoustic decibel
Age Group 30-39
(C) Interval plot of acoustic quality satisfaction (D) Interval plot of actual acoustic decibel
Table 18. Confidence interval of acoustic quality satisfaction (A); measured noise level (B);
acoustic quality satisfaction (C); measured noise level (D) by month
Table 18 (B) and Table 18 (D) illustrates measured acoustic decibel of all users and mid-age
group’s workstations. Overall, the graphs show a remarkably similar pattern, even though the 30
to 39 age group’s noise level was more widespread. Both data illustrate that there was a significant
drop in noise level (approximately 15dBA) in September and all data were higher than the
maximum value of the standard. Moreover, the acoustic conditions of both groups were
significantly inconstant in five selected months. The analyses revealed that there was an impact of
age group on user’s acoustic perception. Table 18 (A) shows that the all user group’s acoustic
52
satisfaction concerning the background noise from the conversation was relatively constant with a
p-value of 0.5148, even though their actual noise levels were varied. However, as described in
Table 18 (C), the mid-age group’s responses were significantly inconstant (with a p-value of
0.0778) with the almost same acoustic condition as the other groups. This result indicates that the
age group 30 to 39 might be more sensitive to the noise from the conversation. Since the office
had an open office plan, the occupants of the office B were possibly exposed to the other colleagues’
conversation. In addition, the correlation result shows that the mid-age group’s and their noise
level had a marginally positive correlation (with a Spearman rho of 0.344553 and a p-value of
0.0917). Therefore, it might be possible to conclude that the mid-age group tends to prefer slightly
noisy condition because of some potential factors such as psychological and/or physical condition,
and acoustic privacy.
4.3.3 Impact of time functions
Occupant satisfaction survey
Thermal satisfaction IAQ satisfaction Acoustic satisfaction Lighting satisfaction
Data
categories
Q13 Q14 Q15 Q16 Q17 Q18 Q19 Q20 Q21 Q22
DAY 0.848 0.9779 0.6083 0.3857 0.9843 0.7558 0.5207 0.7416 0.7307 0.3676
TIME 0.2498 0.3422 0.592 0.2571 0.5827 0.9492 0.5587 0.9452 0.6472 0.9113
WEEK 0.9667 0.8313 0.4327 0.3208 0.9033 0.4726 0.4118 0.4379 0.7041 0.8023
Actual IEQ
Thermal condition Air quality Acoustic condition Lighting condition
Data
categories
Temperature CO 2 Humidity Acoustic Work Surface illuminance UGR
DAY <0.000* <0.000* <0.000* <0.000* 0.9934 0.0184*
TIME 0.8579 <0.000* <0.000* <0.000* 0.6062 0.4306
WEEK 0.0013* 0.8091 <0.000* 0.6668 0.9906 0.5799
Table 19.Summary of P-value from statistical analysis by different data categories (* indicates a
P-value below 0.1)
Table 19 describes how much user’s satisfaction levels and actual IEQ conditions were changed
based on different time categories by using statistical analysis. The numeric values indicate a p-
value which can show the difference of the dataset by group variable. Two-sample t-test and
53
ANOVA test were adopted to generate the p-value. In general, subjects reported a relatively
consistent satisfaction levels concerning all IEQ elements, even though some corresponding IEQ
components including thermal, acoustic condition and air quality were notably changed. Similar
to the result of office A. the result of office B also indicates that users might be possible to have
stable environmental satisfaction at different times, even though their IEQ conditions are not
constant. Moreover, even if user’s circadian rhythm might have an impact on their environmental
perception depending on different times, occupants might be possible to adjust to the change of
ambient environment unless it is not extremely varied. Based on the result above, it might be
possible to conclude that time functions do not have a huge effect on occupant’s environmental
satisfaction and cause the change of satisfaction with the indoor condition.
4.4 Case 3: Office C
Figure 17. Distribution of Temperature at floor level (left) and Temperature at 1.2m (right)
Distribution of temperature at two different levels is illustrated in Figure 17. The graphs show that
temperature data of selected months were within the ASHRAE’s comfort zone, while the office A
and the office B showed relatively poor thermal condition. Overall, the temperature data measured
at floor level showed the relatively similar distribution in four different measurement times, even
though November’s temperature were slightly lower than other months (Figure 17 (left)). In
addition, although all workstations’ temperature measured at 1.2m level was fully placed in the
recommended zone, the temperature in September and November were marginally lower than
54
January and June’s temperature data (Figure 17 (right)). Moreover, one of the workstation in June
had significantly higher temperature compared to the other workstations.
Figure 18. Distribution of Radiant Temperature Asymmetry wall (left) and Radiant Temperature
Asymmetry Ceiling (right)
Distribution of radiant temperature asymmetry wall and ceiling are described in Figure 18. For
radiant temperature asymmetry wall, the data of different months showed a notably similar pattern
of distribution, even though June’s temperature difference were slightly more widespread than
other data and had a workstation’s data, which is out of the guideline (Figure 18 (left)). Moreover,
the data of radiant temperature asymmetry ceiling indicated that most data were within the comfort
zone, although there were some exceptions in January (Figure 18 (right)).
55
Figure 19. Distribution of Relative Humidity (left) and CO
2
concentration (right)
Distribution of Relative Humidity and CO
2
concentration are illustrated in Figure 19. The office
had a good air quality among four selected months and all humidity and CO
2
data were within the
recommended comfort zone. The January’s RH data were significantly lower than other months’
data (Figure 19 (left)). For CO
2
concentration, September and November’s CO
2
levels were
notably lower than CO
2
data in January and June (Figure 19 (right)).
Figure 20. Distribution of Work Surface Illuminance (left) and UGR (right)
Distribution of lighting quality data is shown in Figure 20. Comparing to the previous two offices,
the office C had a better lighting condition, especially regarding work surface illuminance level.
Although there was some workstation which had lower or higher illuminance levels than the
56
recommended guideline, the majority of illuminance data in four different months were within the
comfort zone (Figure 20 (left)). Overall, workstations had a notably constant illuminance among
the four months. However, although the office C had a good lighting condition, the large
percentage of UGR data were out of the CIE’s guideline (Figure 20 (right)). In general, those data
which were placed in outside of comfort zone had the values lower than the minimum value of
suggested guideline.
Figure 21. Distribution of Noise level
Distribution of acoustic decibel is illustrated in Figure 21. Although there was a significant
difference of noise level by different four months, all measured acoustic decibel was fully above
the maximum value of ASHRAE’s standard. The June’s acoustic condition was remarkably noisier
than other three months with a range between 64 dBA and 72 dBA. In addition, although January,
September and November’s acoustic data were placed on the outside of the recommended range,
some workstation’s noise levels were closed to 40 dBA.
57
January June September November
Variables M S W M S W M S W M S W
Temperature
floor ('C)
25.41 0.51 100% 25.19 0.68 100% 24.88 0.67 100% 24.24 0.78 100%
Temperature
1.2m ('C)
25.53 0.55 100% 25.46 0.70 100% 24.98 0.62 100% 24.50 0.56 100%
Vertical Air
Temperature
Difference
0.27 0.15 100% 0.32 0.25 100% 0.25 0.26 100% 0.37 0.32 100%
Radiant
Temperature
Asymmetry
Wall ('C)
1.60 1.16 100% 3.16 2.70 95% 0.66 0.57 100% 1.05 0.84 100%
Radiant
Temperature
Asymmetry
Ceiling ('C)
0.91 0.81 100% 0.65 0.41 100% 1.08 0.69 100% 1.05 0.66 100%
Relative
Humidity (%)
20.33 0.54 100% 42.61 1.69 100% 46.97 2.90 100% 45.43 2.68 100%
CO2 level
(ppm)
753.63 26.70 100% 777.97 17.17 100% 571.38 35.65 100% 592.09 25.48 100%
Work surface
illuminace (lux)
235.55 37.38 80% 291.95 105.48 95% 219.30 26.58 70% 227.85 59.99 55%
Unified Glare
Rating (UGR)
12.69 1.29 42% 13.44 1.58 64% 12.62 1.50 44% 12.5 1.76 33%
Acoustic
decibel (dBA)
47.6 2.63 0% 65.4 2.01 0% 51.5 3.27 0% 49.5 3.57 0%
Table 20. Summary of measured IEQ data by month (M = Mean, S = StDev, W = Within
guideline)
Table 20 shows the statistical summary of measured IEQ components by four selected months:
January, June, September, and November. Overall, the data relevant to thermal quality were mostly
within the ASHRAE’s thermal comfort zone. The mean values of temperatures measured at 1.2m
were 25.53
o
C, 25.46
o
C, 24.98
o
C, and 24.50
o
C, which are approximately 1
o
C to 2
o
C higher than
the minimum value of the guideline. The relative humidity levels and CO
2
concentrations were
fully within the comfort zones. However, unlike thermal quality and air quality, lighting condition
and acoustic quality were relatively out of the corresponding guidelines especially acoustic levels
were fully out of the guideline. 20%, 5%, 30%, and 45% of the illuminance levels in four measured
times were placed in outside of the suggested range. In addition, the UGR data also showed 58%,
36%, 56%, and 67% of the workstations were located in the outside of the CIE’s recommended
comfort range. However, although the work surface illuminace data were not fullu within the
standard, the lighting conditions of office C was relatively better than other offices. For the
acoustic quality, there were no workstations which had a comfortable acoustic decibel with the
average values of 47.6 dBA, 65.4 dBA, 51.5 dBA, and 49.5 dBA, which are notably higher than
the maximum level of the standard.
58
Figure 22. Rose chart of the survey question by month
As shown in Figure 22, although the occupants showed dissatisfaction with some questions such
as level of privacy, background noise, and temperature in September and November, overall
satisfaction with most criteria reach a positive or neutral value of satisfaction. In the previous
sections, the office A and the office B illustrated the notably similar patterns of linear lines by
different months. However, the occupants of the office C reported remarkably inconstant responses
regarding thermal satisfaction and lighting satisfaction. Occupants had significantly higher
satisfaction on the thermal condition in January. Moreover, users were remarkably less satisfied
with their lighting condition in November. In addition, occupants mostly indicated higher
satisfaction levels with all questions in January. Similar to other two offices, the study also found
that there was an effect of human factors on the building occupant’s environmental perception
which affects the change of satisfaction.
-1.00
-0.50
0.00
0.50
1.00
1.50
2.00
2.50
Job/Institution
Size of workstation
Level of privacy
Alterability of physical condition
Accessibility of outside view
Distance between others
Degree of enclosure of work area
Temparature
Operability of thermostats
Air quality
Background noise (conversation) Background noise (mech system)
Lighting condition for computer
work
Glare in the computer screen
Direct glare from lighting fixture
Direct glare from daylight
Lighting quality
Workstation layout
Color and textures of finishes
Productivity
IEQ
01_January 06_June 09_September 11_November
59
4.4.1 The change of occupant’s responses
Occupant satisfaction survey
Thermal satisfaction IAQ satisfaction Acoustic satisfaction
Data Categories Q13 Q14 Q15 Q16 Q17
ALL 0.0049* 0.0659* 0.6976 0.1252 0.5708
FEMALE 0.4059 0.3384 0.5878 0.8142 0.455
MALE 0.0214* 0.207 0.6869 0.1836 0.8887
AGE (18-29) 0.0327* 0.2249 0.3521 0.4477 0.0085*
AGE (30-39) 0.1514 0.2126 0.9401 0.3325 0.5183
Actual IEQ
Thermal condition Air quality Acoustic condition
Data Categories Temperature CO2 Humidity Acoustic
ALL <0.000* <0.000* <0.000* <0.000*
FEMALE 0.1197 <0.000* <0.000* <0.000*
MALE <0.000* <0.000* <0.000* <0.000*
AGE (18-29) 0.0005* <0.000* <0.000* <0.000*
AGE (30-39) 0.0095* <0.000* <0.000* <0.000*
Table 21. Summary of P-value from ANOVA test by different data categories (* indicates a P-
value below 0.1)
Table 21 summaries the change of occupant’s response and actual IEQ condition between four
selected months by using statistical analysis. According to previous two offices, occupants might
be possible to adjust their indoor condition and have a persistent satisfaction level, even though
the actual IEQ conditions are varied. However, the result of the office C indicates that users’
satisfaction with IEQ condition can be changeable. Occupants of the office C reported inconstant
satisfaction with their thermal condition by different measurement times. In addition, the 18 to 29
age group and the male group also showed the change of satisfaction with ambient temperature.
Furthermore, the junior age group’s acoustic satisfaction showed the variety of satisfaction in four
months. Except for those specific age or gender group, the occupant showed the constant
satisfaction regarding air quality, and acoustic condition, even though the real IEQ element had
the variety. This result reveals that occupant’s environmental satisfaction might be not persistent
all the time and have a possibility of the change based on the case group, although users’
60
satisfaction concerning air quality and acoustic condition showed the consistency. Therefore, it
might be possible to conclude that one-time data acquisition is reliable depending on the group.
However, since it is hard to define the specific group which might report the variety of the
satisfaction level, multiple data collection may be required to provide a high quality of evaluation.
4.4.2 Impact of human factors
Male Group
(A) Interval plot of thermal quality satisfaction (B) Interval plot of actual temperature
Female Group
(C) Interval plot of thermal quality satisfaction (D) Interval plot of actual temperature
Table 22. Confidence interval of thermal quality satisfaction (A); measured temperature (B);
thermal quality satisfaction (C); measured temperature (D) by month
Table 22 (B) and Table 22 (D) showed the measured temperature of both gender groups. Although
both data had a relatively similar distribution pattern, the graphs illustrate that the male group’s
workstations had a larger variation of temperature by four different months (with a p-value of
<0.0001) compared to the female group’s workstations (with a p-value of 0.1197). Similar to IEQ
61
condition, the occupants’ satisfaction levels with the thermal condition were also varied. However,
as illustrated in Table 22 (A) and Table 22 (C), only the male group’s satisfaction levels were
significantly changed between selected months (with a p-value of 0.0214), while the female
group’s responses were relatively consistent (with a p-value of 0.4059). This result reveals that
female group might be less sensitive to the variety of thermal condition unless the actual
temperature is within the ASHRAE’s comfort zone. Moreover, it might be possible that male group
can easily be affected by external factors which possibly influence the male group’s environmental
perception and make them sensitive to the change of temperature, even though the temperature is
within the recommended range.
Male Group
(A) Interval plot of glare condition satisfaction (B) Interval plot of UGR
Female Group
(C) Interval plot of glare condition satisfaction (D) Interval plot of UGR
Table 23. Confidence interval of glare condition satisfaction (A); measured UGR (B); glare
condition satisfaction (C); measured UGR (D) by month
62
As shown in Table 23 (B) and Table 23 (D), the measured UGR values were persistent among the
four different measurement times (with a p-value of 0.4172 and 0.2753), However, unlike the
distribution of UGR data, the occupant’s glare satisfaction reported the change of satisfaction level.
Table 23 (A) represents the male group’s glare satisfaction were significantly varied between four
different months (with a p-value of 0.082). However, the female group answered the constant
satisfaction regarding glare condition of their workstation with a p-value of 0.4334 (Table 23 (C)).
This result reveals that the male group’s lighting perception might be more sensitive than the
female group’s perception. In addition, although the male group’s satisfaction and actual UGR
values illustrated the relatively similar pattern, there was no correlation between those two data.
Therefore, it might be possible that the male occupants are easily affected by external factors which
cause the variety of lighting perception.
All Users
(A) Interval plot of acoustic quality satisfaction (B) Interval plot of actual acoustic decibel
Age Group 18-29
(C) Interval plot of acoustic quality satisfaction (D) Interval plot of actual acoustic decibel
Table 24.Confidence interval of acoustic quality satisfaction (A); measured noise level (B);
acoustic quality satisfaction (C); measured noise level (D) by month
63
Table 24 (B) and Table 24 (D) shows the actual noise level of all user group and the 18 to 29 age
group. Both group had a significantly similar noise condition in four different months, and there
was a significant increase of acoustic decibel in June. All acoustic decibel data were above 40 dBA,
which is the maximum value of the recommended guideline. Table 24 (A) illustrates that the all
user group’s acoustic satisfaction especially background noise from the mechanical system was
consistent with a p-value of 0.5708. However, as shown in Table 24 (C) the junior group’s
responses were significantly varied between selected month (with a p-value of 0.0085) with the
almost same acoustic condition as all user group. This result reveals that the age group (18 to 29
years old) might be more sensitive to low sound pressure than other age groups. Since this study
only measured general acoustic levels of the workstation in dBA, it is hard to identify that the
increase of acoustic decibel was because of the louder noise from mechanical systems. However,
it might be possible to conclude that age group influence user’s acoustic perception which can
affect the change of user’s acoustic satisfaction.
4.4.3 Impact of time functions
Occupant satisfaction survey
Thermal satisfaction IAQ satisfaction Acoustic satisfaction Lighting satisfaction
Data
categories
Q13 Q14 Q15 Q16 Q17 Q18 Q19 Q20 Q21 Q22
DAY 0.0044* 0.0639* 0.5071 0.6515 0.4053 0.2088 0.2696 0.039* 0.3175 0.674
TIME 0.2917 0.5682 0.4619 0.6929 0.942 0.079* 0.1261 0.018* 0.165 0.4943
WEEK 0.0006* 0.0144* 0.2424 0.3562 0.1691 0.3756 0.4732 0.4981 0.9382 0.4058
Actual IEQ
Thermal condition Air quality Acoustic condition Lighting condition
Data
categories
Temperature CO 2 Humidity Acoustic Work Surface illuminance UGR
DAY <0.000* <0.000* <0.000* <0.000* 0.3253 0.1096
TIME <0.000* <0.000* <0.000* <0.000* 0.2153 0.3243
WEEK 0.0011* <0.000* <0.000* <0.000* 0.4237 0.7887
Table 25. Summary of P-value from statistical analysis by different data categories (* indicates a
P-value below 0.1)
64
Table 25 shows the variation of user’s satisfaction levels and actual IEQ conditions depending on
different time categories by using statistical analysis. The numeric values indicate a p-value and
Two-sample t-test and ANOVA test were used to generate the p-value. In the previous sections,
other two offices illustrated that subjects reported a relatively constant satisfaction levels
concerning all IEQ factors, even though some related IEQ elements including thermal, acoustic
condition and air quality were significantly varied. However, the result of the office C reveals that
time functions had an impact on occupant’s environmental satisfaction. Although users of the
office A and office B illustrated the adaptability and had constant satisfaction levels by different
times, the users of the office C reported the different satisfaction levels with the thermal and
lighting condition by different time factors. This result reveals that occupants might have the
different environmental perception based on the time factors because of their circadian rhythm or
psychological factors. Since the result of three offices are inconstant, it might be hard to conclude
that occupant always have persistent satisfaction levels and time function does not have an effect
on the variety of user’s satisfaction.
4.5 Discussion
In this section, the study investigates and validates the effectiveness of the current POE method,
which uses one-time data collection, by adopting multiple data collection method. The office A
and office B showed the relatively similar result, which illustrated that occupant can adjust their
indoor environment and have a constant satisfaction level. However, the occupants of the office C
described the variety of IEQ satisfaction. Moreover, although the users of that two offices
generally reported consistent satisfaction levels with their IEQ condition, there was an impact of
human factors such as gender and age on occupant’s environmental perception concerning specific
IEQ elements including thermal, lighting, and acoustic. The results indicate that the age and gender
affect the variety of user’s satisfaction. Similar to this result, previous researches already revealed
that human factors influence occupant’s environmental satisfaction.
65
Choi’s study [51] investigated the effect of human factors on satisfaction with the thermal
environment in the office buildings. The study aggregated 40 subjects’ data from 38 floors in 20
office buildings in the U.S. Based on objective IEQ data and subjective survey data, the data
analysis was conducted. The result revealed that the female group was more dissatisfied with their
thermal condition than the male group, especially in the summer season. Moreover, the age group
(over 40 years old) showed the higher satisfaction level than under the age group (under 40 years
old) in the cooling season with slight significance. In addition, Kim [25] also investigated that
there are gender differences in office occupant perception of IEQ. Based on 38.257 occupant
sample, the statistical analyses were conducted. The result reported that gender differences were
existed, especially in sick building syndrome and thermal discomfort. Moreover, the female group
consistently had a lower satisfaction regarding most IEQ components such as thermal comfort, air
quality, lighting, acoustic, office layout & furnishings, and cleanliness & maintenance.
Based on the results and literature reviews, it might be possible to conclude that the current POE
method which uses one-time data collection is not fully reliable to evaluate the indoor condition
of the office buildings. Moreover, even if the occupants of the office A and office B showed the
consistency of satisfaction levels regarding thermal comfort, air quality and acoustic condition, it
is hard to demonstrate that the one-time data acquisition is dependable because the result can be
different based on the buildings (the office C in this study). Moreover, since all offices illustrated
that human factors affected the change of occupant’s satisfaction with specific IEQ factors, current
method might need to be developed to consider the effect of human factors. Therefore, the multiple
data collection might be necessary to get more sophisticated results and provide a better design
solution.
In addition to the validation of POE method, the study also investigated the impact of time
functions on user’s environmental satisfaction. Overall, occupants reported persistent feedbacks,
even though their IEQ conditions were notably varied. However, the occupants of office C had a
variety of satisfaction with thermal and lighting condition based on the time factors such as the
66
day of the week and the time of day. Since the results of three offices were different, it might be
hard to define that time factors do not affect the occupants’ satisfaction. In addition, since the scope
of this study is to identify the change of user’s environmental satisfaction depending on the time
difference, it is difficult to diagnose the specific reason that why the occupants of the office C
showed the variation of satisfaction by different times. However, it might be possible to conclude
that time functions affect the occupant’s environmental perception and cause the change of
satisfaction.
4.6 Developing design suggestion by using data mining tool
As a design guideline, a decision tree is frequently generated by adopting data mining tools. Data
mining is the computational process of finding patterns in a large size of the dataset by using
various methods including artificial intelligence, machine learning, statistics, and database systems
to identify what information is the most relevant. In Section 4.5, the study describes that multiple
data collection might provide the better design solution. Since multiple data acquisition can
enhance the number of data, it is helpful to make the decision tree. However, there is another
reason why multiple data collection is necessary for making a design guideline. Figure 23
illustrates the decision tree which was generated based on the dataset collected for three different
months. It shows that air quality is the most significant factor in overall IEQ satisfaction. Then,
lighting condition for computer work is the second important factor to enhance the overall
satisfaction. However, as described in Figure 24, lighting quality is the driving factor from the
decision tree which used five months’ data. In addition, depending on the answer to lighting quality
satisfaction, thermal comfort or acoustic satisfaction remain as the next significant elements. Those
two datasets were collected from the same office and the only difference was the size of the dataset.
However, the decision tree of two datasets was significantly different. Since individual IEQ
satisfaction and overall IEQ satisfaction have a nonlinear relationship, it is complex to find out the
key factors on user’s environmental satisfaction [52]. Moreover, IEQ components also affect each
IEQ factors differently and cause a variety of occupant’s satisfaction, the large size of data not
only increase the number of samples but also helps to define the complex relationship between all
IEQ elements and classify the driving factor.
67
Figure 23. Decision Tree (J48) from selected survey questions and IEQ factors (Dataset includes
three months’ data)
Figure 24. Decision Tree (J48) from selected survey questions and IEQ factors (Dataset includes
five months’ data)
68
In addition, the decision guideline also can be different depending on the season that data collection
is conducted. Figure 25 shows the decision tree which was generated based on summer data. The
tree indicates that air quality satisfaction is the most crucial factor for overall occupant’s
satisfaction. Moreover, depending on the answer to IAQ satisfaction, acoustic satisfaction is the
next key factor. However, as illustrated in Figure 26, the decision tree which used the winter data
describes that thermal satisfaction and lighting satisfaction are the most important factors which
have a huge impact on occupant’s overall IEQ satisfaction. Although two datasets were aggregated
at the same office, the design guidelines were totally different based on the data acquisition time.
Therefore, it might be possible to conclude that multiple data acquisition enhances the accuracy
and quality of design guideline. In addition, multiple data collection can consider the variation of
design solution, which is caused by seasonal impact.
Figure 25. Decision Tree (J48) from selected survey questions and IEQ factors (Dataset
collected in summer season)
69
Figure 26. Decision Tree (J48) from selected survey questions and IEQ factors (Dataset
collected in winter season)
70
CHAPTER 5: CONCLUSION
This research validated the effectiveness of current POE method which mainly uses one-time data
acquisition and identified the time-varying user’s IEQ perception by adopting multiple data
collection. Overall, occupants of the office A and office B showed a similar pattern concerning the
variety of environmental satisfaction, while users of the office C reported the opposite result. The
occupants of two offices (A and B) described a constant satisfaction level with specific IEQ
categories such as thermal, air quality and acoustic in five different measured times. However,
users of the office C showed the change of satisfaction on thermal and lighting quality. In addition,
the study also found that human factors affected the change of user’s environmental perception.
The case of office A indicated that the male group was less sensitive with background noise from
mechanical system and it influenced the variety of the male group’s acoustic satisfaction, even
though their actual acoustic decibel was almost similar to female group’s noise level. In addition,
Junior group also showed the same result that they were less sensitive to low-pressure sound and
it affected the change of the junior group’s acoustic satisfaction. However, the other age groups
did not report the variety of satisfaction with background noise. Moreover, the 40 to 49 age group
showed the significant change of their answers regarding lighting conditions such as overall
lighting quality, lighting condition for computer work, and direct glare. However, the other users
except for the mid-age group (40 to 49 years old) reported relatively consistent satisfaction with
lighting condition. The case of Office B revealed similar findings like those for Office A. The
female group illustrated more sensitive thermal perception and showed the different thermal
comfort level between four selected months. However, unlike the female group, the male group
reported a persistent thermal satisfaction, although their thermal condition was notably inconstant.
In addition, the 30 to 39 age group noted the change of the satisfaction with glare condition in
different months, while the measured UGR was pretty constant. Besides, this finding did not occur
in the other age group, even though their glare condition was less stable than the mid-age group
(30 to 39 years old). Furthermore, the 30 to 39 age group also indicated more sensitive to the
background noise from the conversation. Although all occupants had an almost same acoustic
condition, only the mid-age group showed the variety of acoustic satisfaction. Lastly, the case of
71
office C revealed that the male group’s thermal satisfaction levels were varied in different months,
while the female group illustrated the constant responses regarding thermal condition. In addition,
the male group also indicated that their lighting perception is more sensitive than the female
group’s perception. The male group reported the variety of glare satisfaction, while the female
occupants had persistent satisfaction levels between four months. Moreover, the 18 to 29 age group
showed the significant change of acoustic satisfaction, even though this result did not occur in the
other age group. It indicated that junior group is more sensitive to background noise from the
mechanical system than other age groups.
These three case studies from sampled offices described that overall occupant’s environmental
satisfaction was persistent. However, since the results were inconstant among the three offices, it
is hard to generalize that occupants always have constant satisfaction levels. Moreover, although
occupants tended to show that they could adjust their indoor environment to maintain the
satisfaction level unless the variation of IEQ condition is notably huge, human factors significantly
affected the occupants’ IEQ perception and caused the change of the satisfaction. Therefore, this
study revealed that one-time IEQ measurement and occupant satisfaction surveys are hard to fully
reliable because the effectiveness of the one-time data collection is inconstant based on cases. In
addition, since occupants also showed the variety of their comfort level depending on human
factors, it might be possible to conclude that the suggested multiple data collection is remarkably
required to conduct more accurate analysis and provide an optimal design solution. In addition,
the study also illustrated that there is an impact of time functions on occupant’s environmental
perception. Therefore, repetitive data acquisition is necessary to consider the effect of time
functions.
Despite there are research findings, it still has some limitations which need to be developed and
considered for future study. Although this research selected five different months (January, April,
June, September, and November) that covered various seasons, the climatic conditions of five
months were not significantly different due to its location. Three sampled office buildings are
72
located in Southern California where the weather is mostly sunny and moderate. In this area, it
might be difficult that occupants of three offices are significantly affected by the seasonal or
climatic effect. Therefore, the result might be possibly changed in the area where significant
differences in climate condition occur. Furthermore, additional sampled offices should be
considered to extend the variety of age, gender, ethnic origin group. The study already illustrated
that human factors remarkably affected user’s environmental satisfaction on specific IEQ elements.
Therefore, if the additional office has a greater variety of human factors, it might help to diagnose
detailed impact of those components.
These extra research elements and conditions will be able to help establish building design or
control guideline, with an extended consideration of human factors and time functions, based on
more specific intellectual evidence.
73
BIBLIOGRAPHY
[1] U.S Department of Labor, “Bureau of Labor Statistics,” Occup. Outlook Handbook, 2012-
13 Ed. Med. Assist., 2012.
[2] W. J. Fisk, “How IEQ affects health, productivity,” ASHRAE J., vol. 44, no. 5, 2002.
[3] Y. Geng, W. Ji, B. Lin, and Y. Zhu, “The impact of thermal environment on occupant IEQ
perception and productivity,” Build. Environ., vol. 121, pp. 158–167, 2017.
[4] V. Loftness et al., “Sustainability and health are integral goals for the built environment,”
Heal. Build., vol. 1, pp. 1–17, 2006.
[5] ASHRAE, ASHRAE Standard 62.1-2013 Ventilation for Acceptable Indoor Air Quality, vol.
2013. Atlanta, GA, 2013.
[6] ASHRAE, ASHRAE Standard 55-2013 Thermal Environmental Conditions for Human
Occupancy, vol. 2004. Atlanta, GA, 2013.
[7] Illuminating Engineering Society of North America, The IESNA lighting handbook :
reference & application. New York, NY, 2000.
[8] CIE 1995, CIE 117-1995 Discomfort Glare in Interior Lighting. 1995.
[9] “Human Thermal Comfort | Sustainability Workshop.” [Online]. Available:
https://sustainabilityworkshop.autodesk.com/buildings/human-thermal-comfort. [Accessed:
01-Nov-2017].
[10] “Occupant Comfort | Sustainability Workshop.” [Online]. Available:
https://sustainabilityworkshop.autodesk.com/buildings/occupant-comfort. [Accessed: 01-
Nov-2017].
[11] O. US EPA, “Introduction to Indoor Air Quality.”
[12] “Center for the Built Environment: Occupant IEQ Survey.” [Online]. Available:
https://www.cbe.berkeley.edu/research/briefs-survey.htm. [Accessed: 06-Oct-2017].
[13] “Center for the Built Environment: Occupant Indoor Environmental Quality (IEQ) Survey.”
[Online]. Available: https://www.cbe.berkeley.edu/research/survey.htm. [Accessed: 28-
Jan-2018].
[14] “Center for the Built Environment: Occupant IEQ Survey.” [Online]. Available:
https://www.cbe.berkeley.edu/research/briefs-survey.htm. [Accessed: 28-Jan-2018].
74
[15] “BRE Group: Post-Occupancy Evaluation (POE).” [Online]. Available:
https://www.bre.co.uk/page.jsp?id=1793. [Accessed: 01-Nov-2017].
[16] WGBC, “Health, Wellbeing & Productivity in Offices,” 2014.
[17] A. F. Marmot, “The Good Office:: POST-OCCUPANCY EVALUATION OF OFFICE
BUILDINGS,” Facilities, vol. 9, no. 12, pp. 10–13, 1991.
[18] S. Abbaszadeh, L. Zagreus, D. Lehrer, and C. Huizenga, “Occupant Satisfaction with Indoor
Environmental Quality in Green Buildings,” Heal. Build., vol. 3, pp. 365–370, 2006.
[19] D. K. Tiller, L. M. Wang, A. Musser, and M. J. Radik, “Combined effects of noise and
temperature on human comfort and performance,” 2010.
[20] G. Brager, H. Zhang, and E. Arens, “Evolving opportunities for providing thermal comfort,”
Build. Res. Inf., vol. 43, no. 3, pp. 274–287, 2015.
[21] W. F. E. Preiser, “Post-occupancy evaluation: how to make buildings work better,”
Facilities, vol. 13, no. 11, pp. 19–28, 1995.
[22] W. F. E. Preiser, J. Vischer, C. Zimring, T. Rosenheck, and A. Kaplan, “Learning from our
buildings. A State-of-the-Practice Summary of Post-Occupancy Evaluation.,” 2001.
[23] C. Watson, “Review of Building Quality Using Post Occupancy Evaluation,” J. Program.
Educ. Build., vol. 53, no. 9, pp. 1689–1699, 2013.
[24] V. Loftness et al., “The value of post-occupancy evaluation for building occupants and
facility managers,” Intell. Build. Int., vol. 1, no. 4, pp. 249–268, 2009.
[25] J. Kim, R. de Dear, C. C??ndido, H. Zhang, and E. Arens, “Gender differences in office
occupant perception of indoor environmental quality (IEQ),” Build. Environ., vol. 70, pp.
245–256, 2013.
[26] S. Altomonte, S. Saadouni, and S. Schiavon, “Occupant Satisfaction in LEED and
BREEAM-Certified Office Buildings,” in PLEA 2016 Los Angeles - 36th International
Conference on Passive and Low Energy Architecture: Cities, Buildings, People: Towards
Regenerative Environments., 2016.
[27] M. Frontczak, R. V. Andersen, and P. Wargocki, “Questionnaire survey on factors
influencing comfort with indoor environmental quality in Danish housing,” Build. Environ.,
vol. 50, pp. 56–64, 2012.
75
[28] C. Peretti, S. Schiavon, J. Goins, E. Arens, and M. De Carli, “Evaluation of Indoor
Environment Quality with a Web-based Occupant Satisfaction Survey : a Case Study in
Northern Italy,” Cent. Built Environ., 2010.
[29] L. Zagreus, C. Huizenga, E. Arens, and D. Lehrer, “Listening to the occupants: a Web-
based indoor environmental quality survey,” Indoor Air, vol. 14, no. s8, pp. 65–74, 2004.
[30] C. Langston, Y. Song, and B. Purdey, “Perceived conditions of workers in different
organizational settings,” Facilities, vol. 26, no. 1/2, pp. 54–67, 2008.
[31] K. Hadjri and C. Crozier, “Post-occupancy evaluation: purpose, benefits and barriers,”
Facilities, vol. 27, no. 1/2, pp. 21–33, 2009.
[32] J. H. Choi, V. Loftness, and A. Aziz, “Post-occupancy evaluation of 20 office buildings as
basis for future IEQ standards and guidelines,” Energy Build., vol. 46, pp. 167–175, 2012.
[33] J. H. Choi and J. Moon, “Impacts of human and spatial factors on user satisfaction in office
environments,” Build. Environ., vol. 114, pp. 23–35, 2017.
[34] H. H. Liang, C. P. Chen, R. L. Hwang, W. M. Shih, S. C. Lo, and H. Y. Liao, “Satisfaction
of occupants toward indoor environment quality of certified green office buildings in
Taiwan,” Build. Environ., vol. 72, pp. 232–242, 2014.
[35] Z. Wang, H. Zhao, B. Lin, Y. Zhu, Q. Ouyang, and J. Yu, “Investigation of indoor
environment quality of Chinese large-hub airport terminal buildings through longitudinal
field measurement and subjective survey,” Build. Environ., vol. 94, pp. 593–605, 2015.
[36] A. Ghahramani, C. Tang, Z. Yang, and B. Becerik-Gerber, “A Study of Time-Dependent
Variations in Personal Thermal Comfort via a Dynamic Bayesian Network,” Sustain.
Human–Building Ecosyst., pp. 99–107, 2015.
[37] F. Jazizadeh, G. Kavulya, L. Klein, and B. BecerikGerber, “Continuous Sensing of
Occupant Perception of Indoor Ambient Factors,” Comput. Civ. Eng., no. 413, pp. 760–767,
2011.
[38] “Our LA office wins two USGBCLA Awards - BuroHappold Engineering.” [Online].
Available: https://www.burohappold.com/news/burohappolds-la-offices-wins-two-awards/.
[Accessed: 01-Mar-2018].
[39] “Los Angeles - Glumac.” [Online]. Available: https://www.glumac.com/locations/los-
angeles/. [Accessed: 01-Mar-2018].
76
[40] “Glumac Irvine: LEED Platinum ... - Glumac Office Photo | Glassdoor.” [Online]. Available:
https://www.glassdoor.com/Photos/Glumac-Office-Photos-IMG309033.htm. [Accessed:
01-Mar-2018].
[41] “uschbilab.” [Online]. Available: http://joonhoc2.wixsite.com/uschbilab. [Accessed: 19-
Mar-2018].
[42] “LabVIEW - National Instruments.” [Online]. Available: http://www.ni.com/en-
us/shop/labview.html. [Accessed: 06-Oct-2017].
[43] “HOBO MX1102 Carbon Dioxide | Onset HOBO Data Logger.” [Online]. Available:
http://www.onsetcomp.com/products/data-loggers/mx1102. [Accessed: 19-Mar-2018].
[44] G. R. Newsham, “Making the open-plan office a better place to work,” Constr. Technol.
Updat., no. 60, pp. 1–6, 2003.
[45] “Introduction to Correlation and Regression Analysis.” [Online]. Available:
http://sphweb.bumc.bu.edu/otlt/MPH-
Modules/BS/BS704_Multivariable/BS704_Multivariable5.html. [Accessed: 16-Oct-2017].
[46] “Why should I use a 2-sample t test?” [Online]. Available: http://support.minitab.com/en-
us/minitab/17/topic-library/basic-statistics-and-graphs/hypothesis-tests/tests-of-
means/why-use-2-sample-t/. [Accessed: 06-Oct-2017].
[47] “One-way ANOVA in Minitab | Procedure, output and interpretation of the output using a
relevant example.” [Online]. Available: https://statistics.laerd.com/minitab-tutorials/one-
way-anova-using-minitab.php. [Accessed: 16-Oct-2017].
[48] “Weka 3 - Data Mining with Open Source Machine Learning Software in Java.” [Online].
Available: https://www.cs.waikato.ac.nz/ml/weka/index.html. [Accessed: 27-Oct-2017].
[49] “Decision Trees in Machine Learning – Towards Data Science – Medium.” [Online].
Available: https://medium.com/towards-data-science/decision-trees-in-machine-learning-
641b9c4e8052. [Accessed: 27-Oct-2017].
[50] ASHRAE, ASHRAE Handbook - HVAC Applications. 2011.
[51] J. H. Choi, A. Aziz, and V. Loftness, “Investigation on the impacts of different genders and
ages on satisfaction with thermal environments in office buildings,” Build. Environ., vol.
45, no. 6, pp. 1529–1535, 2010.
[52] J. Kim and R. de Dear, “Nonlinear relationships between individual IEQ factors and overall
workspace satisfaction,” Build. Environ., vol. 49, no. 1, pp. 33–40, 2012.
Abstract (if available)
Abstract
Post-occupancy evaluation (POE), an architectural design tool for making better indoor environmental quality to enhance occupants’ satisfaction and productivity, has been conducted to find out the impact of the variable ambient environment. Since indoor environmental quality (IEQ) is fully related to occupants’ satisfaction and productivity, many research projects about IEQ have been launched. POE is one of the methods to determine the quality of the indoor environment based on occupants’ feedback
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Impacts of building performance on occupants' work productivity: a post occupancy evaluation study
PDF
Considering occupants: comprehensive POE research on office environment of Southern California
PDF
Impacts of indoor environmental quality on occupants environmental comfort: a post occupancy evaluation study
PDF
Human-building integration based on biometric signal analysis: investigation of the relationships between human comfort and IEQ in a multi-occupancy condition
PDF
IEQ, sleep quality, and IoT: meta-analysis on improving IEQ and sleep quality using IoT
PDF
Human–building integration: machine learning–based and occupant eye pupil size–driven lighting control as an applicable visual comfort tool in the office environment
PDF
Nursing comfort study: are hospitals performing for nurses? IEQ analysis of working conditions inside hospitals
PDF
Developing environmental controls using a data-driven approach for enhancing environmental comfort and energy performance
PDF
Impact of occupants in building performance: extracting information from building data
PDF
Exploration for the prediction of thermal comfort & sensation with application of building HVAC automation
PDF
Daylighting study of a LEED platinum laboratory building: a post-occupancy evaluation comparing performance in use to design intent
PDF
Impact of bamboo materials on indoor air quality
PDF
Developing a data-driven model of overall thermal sensation based on the use of human physiological information in a built environment
PDF
Indoor environmental quality and comfort: IEQ adaptation and human physiological responses in commercial buildings
PDF
Post occupancy energy evaluation of Ronald Tutor Hall using eQUEST; computer based simulation of existing building and comparison of data
PDF
Human-environmental interaction: potential use of pupil size for office lighting controls
PDF
Natural ventilation in tall buildings: development of design guidelines based on climate and building height
PDF
Multi-occupancy environmental control for smart connected communities
PDF
Human-building integration: Investigation of human eye pupil sizes as a measure of visual sensation in the workstation environment
PDF
Environmental adaptive design: building performance analysis considering change
Asset Metadata
Creator
Lee, Kyeongsuk
(author)
Core Title
Enhanced post occupancy evaluation (POE) for office building: improvement of current methodology to identify impact of ambient environment
School
School of Architecture
Degree
Master of Building Science
Degree Program
Building Science
Publication Date
04/30/2019
Defense Date
04/30/2018
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
environmental comfort,human factor,IEQ,OAI-PMH Harvest,POE,time factor,well-being
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Choi, Joon-ho (
committee chair
), Creighton, Heidi (
committee member
), Schiler, Marc (
committee member
), Ting, Selwyn (
committee member
)
Creator Email
elrmehrm@gmail.com,kyeongsl@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c40-498120
Unique identifier
UC11266975
Identifier
etd-LeeKyeongs-6288.pdf (filename),usctheses-c40-498120 (legacy record id)
Legacy Identifier
etd-LeeKyeongs-6288.pdf
Dmrecord
498120
Document Type
Thesis
Format
application/pdf (imt)
Rights
Lee, Kyeongsuk
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
environmental comfort
human factor
IEQ
POE
time factor
well-being