INNOVATIVE INTEGRATED WINDOW DESIGN WITH ELECTRIC … · M. Arch., B. Arch. Submitted in fulfilment of the requirements for the degree of Doctor of Philosophy ... drawing window blinds

INNOVATIVE INTEGRATED WINDOW DESIGN WITH ELECTRIC LIGHTING

DESIGN SYSTEM TO REDUCE LIGHTING INTERVENTION IN OFFICE BUILDINGS

Mehdi AmirkhaniM. Arch., B. Arch.

Submitted in fulfilment of the requirements for the degree of Doctor of Philosophy

School of Design

Creative Industries Faculty

Queensland University of Technology

2018

Innovative integrated window design with electric lighting design system to reduce lighting intervention in office buildings i

Keywords

Window design; discomfort glare; luminance contrast; window wall; daylighting systems; office space; LED (light emitting diode); energy consumption; immersive virtual reality

ii Innovative integrated window design with electric lighting design system to reduce lighting intervention in office buildings

Abstract

A high luminance contrast between windows and surrounding surfaces can

increase the risk of visual discomfort, which can diminish office workers’ satisfaction

and productivity. Accordingly, it can lead to negative occupant interventions, such as

drawing window blinds or increasing electric light levels, actions that are intended to

enhance indoor visual comfort but counterproductively increase energy consumption.

This study hypothesizes that increasing the luminance of the areas surrounding the

window using a supplementary lighting system, such as wall-washing with light

emitting diode (LED) linear luminaires, could reduce visual discomfort arising from

windowed walls. It aims to demonstrate the effectiveness of using a proposed LED

wall-washing system on diminishing occupants’ propensity to intervene in lighting

conditions in typical office rooms with different orientations and window sizes. This

investigation reports on the results of one pilot study and two experiments in separate

typical office spaces in Brisbane, Australia, as well as a test in immersive virtual reality

(IVR) office spaces.

The pilot study was carried out in a typical office room facing southwest with

around 15% window-to-exterior-wall ratio (WWR). The outcomes of this study

suggest that an LED wall-washing system with low power level could significantly

reduce the luminance contrast (LC) on the window wall and reduce participants’

intention to intervene in lighting conditions. The results of this study were also used

as the basis to assess annual energy consumption of the test office using the DAYSIM

engine within ECOTECT. This research reports that increased electricity consumption

of an LED wall-washing system with low power level is offset where there is roughly

a one-quarter reduction in users’ intentions to intervene in lighting conditions.

Experiment 1 was conducted in the same office room as the pilot study.

However, unlike in the pilot study, the blind was fully open during all test conditions

in this experiment. Accordingly, the WWR in this study was approximately 27%. This

research suggests that the proposed LED wall-washing system with low power level

could reduce the LC on the window wall from values in the order of 117:1 to 33:1,

leading to enhanced subjective scale appraisal of the window’s appearance. The results

indicate that this LED lighting strategy could decrease the mean users’ intention to

Innovative integrated window design with electric lighting design system to reduce lighting intervention in office buildings iii

turn on ceiling lights by around 27%, as well as to diminish the probability of moving

the blind down by up to 90%.

Experiment 2 was conducted in a different typical office facing northwest with

about 45% WWR. The outcomes of this study suggest that creating an LC of between

11:1 and 12:1 on the window wall using a supplementary LED wall-washing system

leads to improved subjective assessments of the window’s appearance. The results

suggest that such an enhancement could significantly reduce visual discomfort from

windows, as well as diminishing the likelihood of the users intending to turn on the

ceiling lights or to move the blinds down.

The outcomes of the Experiment 2 indicate that while the proposed LED wall-

washing system could reduce the LC on the window wall from about 16:1 to 9:1, it

was not as dramatic as Experiment 1 (around 117:1 to 33:1) or the pilot study (around

215:1 to 26:1). The results of the first three studies of this PhD research suggest that

the proposed LED wall-washing system was less efficient in reducing negative

interventions in lighting conditions in the room with a 45% WWR than in the rooms

with lower window-to-external-wall ratios. The results also indicate that the mean

horizontal illuminance on top of the desk in the second test office room with a 45%

WWR and no LED wall-washing system was more than four times higher than the

office room with around 15% and 27% window-to-external-wall ratios. Overall, The

outcomes of the first three studies of the current PhD research suggest that the WWR

and the light level on the desk might influence the impact of the proposed LED wall-

washing system on reducing the LC between the window and surroundings, and

therefore, on participants’ propensity to intervene in lighting conditions. Therefore,

the final experiment set out to investigate the effectiveness of the proposed electric

wall-washing system on participants’ intentions to change the LC between the window

and surrounding surfaces in rooms with different window sizes without having a

dramatic change in horizontal illuminance on the desk. This experiment was carried

out in the controlled IVR office rooms with different window sizes in which we could

change the WWR and the luminaire power level of the proposed electric linear

luminaires quickly and with low cost. The findings of this research show that a

supplementary electric wall-washing system with a low power level could

significantly reduce the likelihood of the users’ negative interventions in lighting

conditions in rooms with different window-to-external-wall ratios.

iv Innovative integrated window design with electric lighting design system to reduce lighting intervention in office buildings

This PhD research proposes an innovative integrated LED wall-washing system

with windows that could reduce energy bills in buildings and enhance window

appearance through reducing visual discomfort arising from windowed walls. The

outcomes of the experiments, when taken together, demonstrate that the proposed LED

wall-washing system with low power level does efficiently mitigate problematic

interventions in lighting conditions that lead to increased energy consumption in

buildings. The benefit of using such a supplementary LED wall-washing system

introduced in this PhD research is that it can be fitted into existing and future buildings

with minimal construction modifications and at a low cost. Overall, this PhD research

indicates a significant and original contribution to knowledge in the field of window

design in architecture and visual discomfort research. It enhances our understanding

of an integrated lighting design solution for a better understanding of window

appearance that could increase energy savings in office buildings.

Innovative integrated window design with electric lighting design system to reduce lighting intervention in office buildings v

Table of Contents

Keywords .................................................................................................................................. i

Abstract .................................................................................................................................... ii

Table of Contents......................................................................................................................v

List of Figures ....................................................................................................................... viii

List of Tables ............................................................................................................................x

List of Symbols ...................................................................................................................... xii

List of Abbreviations ............................................................................................................ xiii

Statement of Original Authorship......................................................................................... xiv

Publications.............................................................................................................................xv

Acknowledgements............................................................................................................... xvi

Introduction .................................................................................................... 1

1.1 Background.........................................................................................................................1

1.2 Research problem................................................................................................................3

1.3 Innovative integrated lighting design solution....................................................................3

1.4 Purpose................................................................................................................................4

1.5 Research question ...............................................................................................................5

1.6 Outline.................................................................................................................................5

Literature Review......................................................................................... 10

Chapter 2.1 Human factors in lighting....................................................................................10 2.1.1 Visual system....................................................................................................................10 2.1.2 Perceptual system..............................................................................................................12 2.1.3 Visual adaptation ..............................................................................................................16 2.1.4 Visual comfort ..................................................................................................................16 2.1.5 Luminance contrast ...........................................................................................................19 2.1.6 Rating tools .......................................................................................................................19 2.1.7 Summary...........................................................................................................................20

Chapter 2.2 Lighting design....................................................................................................22 2.2.1 Daylighting systems..........................................................................................................23 2.2.2 Impact of window characteristics on visual comfort and energy savings .........................25 2.2.3 Electric lights ....................................................................................................................27 2.2.4 Lighting control systems...................................................................................................30 2.2.5 Summary...........................................................................................................................31 2.2.6 Synthesis of elements from Chapter 2.1 and 2.2...............................................................32

Research Design............................................................................................ 34

3.1 Lighting evaluation methods.............................................................................................34 3.1.1 Subjective assessment of perceived discomfort glare .......................................................35 3.1.2 Questionnaire ....................................................................................................................35 3.1.3 Rating scales .....................................................................................................................36 3.1.4 Magnitude estimation strategies........................................................................................37 3.1.5 Physical lighting measurements........................................................................................37

3.2 Data collection methods....................................................................................................38

vi Innovative integrated window design with electric lighting design system to reduce lighting intervention in office buildings

3.4 Framework ....................................................................................................................... 40

3.5 Participants ....................................................................................................................... 46

3.6 Ethics and Limitations...................................................................................................... 48

3.7 Summary .......................................................................................................................... 48

Published and Submitted Papers ................................................................ 50

Chapter 4.1 LED Lighting Design Strategies to Enhance Window Appearance and Increase Energy Savings in Day-lit Office Spaces ............................................................................... 51

4.1.1 Statement of Contribution of Co-Authors for Thesis by Published Paper ........................53 4.1.2 Abstract.............................................................................................................................54 4.1.3 Introduction.......................................................................................................................54 4.1.4 Small pilot study ...............................................................................................................55 4.1.5 Simulation method............................................................................................................58 4.1.6 Results and discussion ......................................................................................................60

Chapter 4.2 Improving the impact of Luminance Contrast on Window Appearance in a Conventional Office Room: Using Supplementary Lighting Strategies ................................ 62

4.2.1 Statement of Contribution of Co-Authors for Thesis by Published Paper ........................64 4.2.2 Abstract.............................................................................................................................65 4.2.3 Introduction.......................................................................................................................65 4.2.4 Method..............................................................................................................................67 4.2.5 Results and discussion ......................................................................................................70 4.2.6 Conclusion and future work..............................................................................................75

Chapter 4.3 An Energy Efficient Lighting Design Strategy to Enhance Visual Comfort in Offices with Windows............................................................................................................ 77

4.3.1 Statement of Contribution of Co-Authors for Thesis by Published Paper ........................80 4.3.2 Abstract.............................................................................................................................81 4.3.3 Introduction.......................................................................................................................81 4.3.4 Luminance Contrast ..........................................................................................................83 4.3.5 Novel Strategies to Reduce Window Wall Luminance Contrast ......................................84 4.3.6 Method..............................................................................................................................86 4.3.7 Results...............................................................................................................................92 4.3.8 Discussion.........................................................................................................................99 4.3.9 Conclusion ......................................................................................................................101

Chapter 4.4 Innovative Window Design Strategy to Reduce Negative Lighting Interventions in Office buildings................................................................................................................ 102

4.4.1 Statement of Contribution of Co-Authors for Thesis by Published Paper ......................104 4.4.2 Abstract...........................................................................................................................105 4.4.3 Introduction.....................................................................................................................1054.4.4 Advantages of using immersive virtual reality environments in human behaviour

studies...........................................................................................................................107 4.4.5 Limitations of using immersive virtual reality spaces in lighting research.....................108 4.4.6 Method............................................................................................................................110 4.4.7 Results.............................................................................................................................116 4.4.8 Discussion.......................................................................................................................1264.4.9 Conclusion ......................................................................................................................129

General Discussion and Conclusion.......................................................... 130

5.1 General discussion.......................................................................................................... 130

5.2 General conclusion......................................................................................................... 135

Appendices.. ............................................................................................................ 138

Appendix A Lighting evaluation metrics and simulation tools............................................ 138 A.1 Evaluating daylight performance inside buildings............................................................139 A.2 Evaluating discomfort glare from windows ......................................................................140 A.3 Simulation programs to predict discomfort glare..............................................................142

Innovative integrated window design with electric lighting design system to reduce lighting intervention in office buildings vii

Appendix B Questionnaire of the pilot study........................................................................143

Appendix C Questionnaire of the first experiment ...............................................................144

Appendix D Questionnaire of the second experiment ..........................................................145

Appendix E Questionnaire of the third experiment ..............................................................147

References… ........................................................................................................... 151

viii Innovative integrated window design with electric lighting design system to reduce lighting intervention in office buildings

List of Figures

Figure 1 Research map ..............................................................................................9

Figure 2 Schematic diagram of the electromagnetic spectrum (NASA, 2013)....................................................................................................11

Figure 3 Combinations of CCT and illuminance that give a pleasing impression (clear space) ......................................................................15

Figure 4 Penetration of daylight inside buildings ...................................................27

Figure 5 Framework for the study...........................................................................41

Figure 6 Images of the test office rooms during the pilot study and the first two experiments when the proposed LED wall-washing system is on .....................................................................................................42

Figure 7 Examples of the IVR office rooms with different window-to-external-wall ratios when the proposed LED wall-washing system is on .........................................................................................46

Figure 8 Research map of Part 4 .............................................................................50

Figure 9 Test office room at QUT in Brisbane, Australia.......................................56

Figure 10 Survey results..........................................................................................58

Figure 11 Annual electric use of the model in different cases ................................60

Figure 12 Plan and sections of the test office room in Brisbane, Australia ............68

Figure 13 Captured HDR image from the test office room.....................................71

Figure 14 Boxplot of feeling discomfort glare during each stage ...........................72

Figure 15 Mean indoor visual comfort during each stage.......................................73

Figure 16 The relationship between feeling discomfort glare from window and mean indoor visual comfort ..........................................................73

Figure 17 Boxplot of window wall luminance ratio and feeling discomfort glare from window ..............................................................................74

Figure 18 Survey results..........................................................................................75

Figure 19 Plan and sections of the test office room, with details of LED lighting system placement and construction........................................87

Figure 20 The questionnaire measuring participant characteristics, responses to lighting conditions and preferred luminance contrasts...............................................................................................89

Figure 21 Experimental flow in the test office room ..............................................92

Figure 22 Captured high dynamic range (HDR) image from the test office room and window wall luminance contrast (LC) calculation equations..............................................................................................94

Figure 23 Percentage of participants reporting discomfort glare as disturbing during each lighting condition............................................96

Innovative integrated window design with electric lighting design system to reduce lighting intervention in office buildings ix

Figure 24 Percentage of participants indicating that they would turn on overhead lights or move the blinds down during each lighting condition.............................................................................................. 99

Figure 25 Virtual reality office room with four different window-to-exterior-wall ratios ............................................................................ 111

Figure 26 The virtual reality office room with a 30% WWR under different lighting conditions............................................................................. 111

Figure 27 Experimental flow................................................................................. 115

Figure 28 Landolt ring test in the virtual reality office room................................ 116

Figure 29 The error bar of RC scores on the window wall during all lighting conditions .......................................................................................... 117

Figure 30 The error bar of indoor lighting satisfaction during all lighting conditions .......................................................................................... 120

Figure 31 The error bar of participants’ indoor lighting level satisfaction rankings based on luminance contrast scores on the window wall .................................................................................................... 122

Figure 32 Percentage of participants who intended to change the luminance contrast on the window wall based on the luminaire power of the electric wall-washing system and the window-to-exterior-wall ratios .......................................................................................... 123

x Innovative integrated window design with electric lighting design system to reduce lighting intervention in office buildings

List of Tables

Table 1 The light quantities .....................................................................................12

Table 2 Daylight penetration evaluation in the case study patterns (Kevin Van Den & Meek, 2015) .....................................................................27

Table 3 Advantages and disadvantages of different types of questions ..................36

Table 4 Average horizontal illuminance measurements at the desk during each stage of the survey.......................................................................57

Table 5 Average luminance ratio between window and surrounding areas during each stage .................................................................................58

Table 6 Mean horizontal illuminance at the work plane level during each stage.....................................................................................................71

Table 7 Mean luminance ratio between window and surrounding areas during each stage .................................................................................72

Table 8 Demographic data of participants...............................................................75

Table 9 Demographic data of participants...............................................................92

Table 10 Average median luminance contrast of the left and right-hand side on the window wall, as well as the luminance contrast of the whole window wall areas during each lighting condition ...................95

Table 11 Average median luminance contrast of the visual display unit (VDU) and the walls surrounding the window frame, as well as the luminance contrast of the VDU and the window surface..............96

Table 12 Average luminance contrast for each level of perceived glare during experimental lighting conditions..............................................97

Table 13 Average median luminance contrast and participants’ responses for perceived discomfort glare from the window during stage 5 ........98

Table 14 Demographic data of participants...........................................................117

Table 15 Mean RC scores during each lighting condition ....................................119

Table 16 Mean indoor lighting satisfaction scores during each lighting condition ............................................................................................121

Table 17 Median luminaire power percentage of the electric wall-washing system in rooms with different window-to-exterior-wall ratios while setting preferred luminance contrast on the window wall .......124

Table 18 Median luminaire power percentage of the electric wall-washing system in rooms with different window-to-exterior-wall ratios while setting minimum acceptable luminance contrast on the window wall ......................................................................................125

Table 19 LC reduction on the window wall in real office rooms with different window sizes using the proposed LED wall-washing system with low power level .............................................................131

Innovative integrated window design with electric lighting design system to reduce lighting intervention in office buildings xi

Table 20 Some of the existing metrics and criteria that are used to analyse daylight quality.................................................................................. 138

xii Innovative integrated window design with electric lighting design system to reduce lighting intervention in office buildings

List of Symbols

Background luminance determined by taking the average luminance of areas not identified as sources of glare (cd/m2)

Luminance of the sources of glare (cd/m2)

Solid angle of the source of glare (sr)

The solid angle of the glare source modified for its position in the field of view (sr)

Total vertical eye illuminance (lx)

Weight factor based on position in a viewing hemisphere, the position index

Innovative integrated window design with electric lighting design system to reduce lighting intervention in office buildings xiii

List of Abbreviations

BREEAM Building Research Establishment Environmental Assessment Method

CCT Correlated colour temperature

CFL Compact fluorescent lamp

CGI CIE glare index

CIBD Chartered Institution of Building Services Engineers

DA Daylight autonomy

DAcon Continuous daylight autonomy

DAmax Maximum daylight autonomy

DF Daylight factor

DGI Daylight glare index

DGP Daylight glare probability

DGPs Simplified daylight glare probability

EEH Energy efficient halogen

FOV Field of view

GBCA Green Building Council of Australia

HMD Head-mounted display

HDR High dynamic range

IEQ Indoor environment quality

LC Luminance contrast

LED Light emitting diode

LEED Leadership in Energy and Environmental Design

QUT Queensland University of Technology

RC Rated contrast

SD Semantic differential

UDI Useful daylight illuminance

UGR Useful glare rating

VCP Visual comfort probability

IVR Immersive virtual reality

WWR Window-to-exterior-wall ratio

xiv Innovative integrated window design with electric lighting design system to reduce lighting intervention in office buildings

Statement of Original Authorship

The work contained in this thesis has not been previously submitted to meet

requirements for an award at this or any other higher education institution. To the best

of my knowledge and belief, the thesis contains no material previously published or

written by another person except where due reference is made.

Signature:

Date: _14/06/2018_______________

QUT Verified Signature

Innovative integrated window design with electric lighting design system to reduce lighting intervention in office buildings xv

Publications

Journal Publications – Refereed:

Amirkhani, M., Garcia-Hansen, V., Isoardi, G., & Allan, A. (2017). An Energy Efficient Lighting Design Strategy to Enhance Visual Comfort in Offices with Windows. Energies, 10(8), 1126.

Journal Publications – Under review:

Amirkhani, M., Garcia-Hansen, V., Isoardi, G., & Allan, A. (2018). Innovative Window Design Strategies to Reduce Negative Lighting Interventions in Office buildings. Energy and Buildings.

Conference Proceedings – Refereed:

Amirkhani, M., Garcia-Hansen, V., & Isoardi, G. (2016). Reducing luminance contrast on the window wall and users' interventions in an office room. Paper presented at CIE Lighting Quality & Energy Efficiency Conference, Melbourne, Australia, 385-394.

Amirkhani, M., Garcia-Hansen, V. and Isoardi, G. (2015) LED lighting design strategies to enhance window appearance and increase energy savings in daylit office spaces, Asia-Pacific Lighting Systems Workshop, Sydney, Australia, 1-7.

Amirkhani, M., Garcia-Hansen, V., & Isoardi, G. (2015). Improving the impact of luminance contrast on the window appearance in a conventional office room: using supplementary lighting strategies.Paper presented at Living and Learning: Research for a Better Built Environment, 49th International Conference of the Architectural Science Association, Melbourne, Australia, 1129-1138.

Conference Presentation – Refereed (unpublished):

Amirkhani, M., Garcia-Hansen, V., & Isoardi, G. (2015). Integrating LED lighting design strategies with side daylighting systems to improve interior lighting design of office buildings. Paper presented at the IASDR 2015 Doctoral Colloquium Conference, Brisbane, Australia.

Industry Presentation (unpublished):

Amirkhani, M., Garcia-Hansen, V., & Isoardi, G. (2016). Decreasing luminance contrast on window walls as well as users' interventions: Using LED lighting system, IESANZ Queensland Chapter -Technical Meeting.

xvi Innovative integrated window design with electric lighting design system to reduce lighting intervention in office buildings

Acknowledgements

I would like to express my very great appreciation to my supervisory team for

the impact that you have had in shaping my development as a researcher. Dr Veronica

Garcia-Hansen, thank you for the high levels of support and patience you have shown

during my PhD Journey. Dr Gillian Isoardi, thank you for the influence you had on

shaping my research and your support during all stages of my PhD candidature. Dr

Alicia Allan, thank you for your help in developing my analysis skills.

I would like to express my deep gratitude to the panel members of my

Confirmation and Final Seminars: Associate Professor Ian Cowling, Associate

Professor Simon Smith, Professor Vesna Popovic, and Professor John Bell. Your

enthusiastic encouragements, as well as valuable and constructive suggestions, helped

me to improve the quality of my research work. I would also like to extend my thanks

to the staff of the Lighting Lab of Queensland University of Technology (QUT) for

their help in providing the equipment for my research.

I wish to thank those who participated in my experiments; this research would

not have been possible without your contributions. Special thanks should be given to

my friends and fellow research postgraduate students, who made me feel part of a

cohort and helped shape my life forever; thank you for all the precious memories and

friendships. To my parents, Razieh Tasdighi Sani and Mohammad Taghi Amirkhani,

I am so grateful for your strong support and encouragement. To my siblings, Mitra and

Majid, and my brother-in-law, Mohammad Shaygan Mehr, thank you for all your

generous support during the course of my overwhelming PhD journey. Finally, I wish

to thank those who shaped my life, motivated me, and believed in me.

Part 1: Introduction 1

Introduction

1.1 Background

The availability of sustainable indoor environmental quality (IEQ) features is a

significant factor influencing occupants’ satisfaction (Freihoefer, 2012; Zalejska-

Jonsson, Fastigheter och, Bygg- och, Kth, & Skolan för arkitektur och, 2014; Afacan

& Demirkan, 2016). In recent years, there has been significant improvement in

building-design strategies, building technologies, and the development of operational

systems to enhance IEQ while reducing the energy consumption of buildings

(Heydarian, Pantazis, Carneiro, Gerber, & Becerik-Gerber, 2016).

Indoor lighting quality is one of the significant factors that affect IEQ in office

buildings (Brager & Baker, 2009). Research has long suggested that lighting

influences the comfort, productivity, and well-being of office workers (Harris, 1980;

Baron, Rea, & Daniels, 1992; Eilers, Reed, & Works, 1996). In addition to its support

of visual tasks, indoor lighting can have significant non-visual biological impacts on

occupants, such as regulating their circadian rhythms, as well as affecting their

biological clock, alertness, and mood (Mills, Tomkins, & Schlangen, 2007).

While appropriate lighting in office buildings can better meet the visual and

psychological requirements of occupants, problematic workplace lighting can lead to

headache, visual stress, and eyestrain (Boubekri, 1995; Bean, 2012). Thus, there has

been a growing emphasis on a need for building professionals, designers, and

architects to develop a more in-depth understanding of lighting design, the aesthetic

aspects of light, and the quality of ambient illumination (IESNA & Rea, 2000). Since

the end of the 1990s, good indoor lighting has been viewed as that which balances the

needs of individuals, environmental concerns, economic issues, and architectural

design (Bellia, Bisegna, & Spada, 2011). It should also provide the required lighting

level for task performance and safety (IESNA & Rea, 2000).

Indoor lighting quality is influenced by several factors, such as the quantity of

light, luminance distribution, illuminance uniformity, colour characteristics of the

light, glare and flicker rate (Chung & Burnett, 2000). As well as characteristics of the

lighting, characteristics of the space also affect the type of lighting that is appropriate.

For example, the amount of time that users spend in an area has a significant effect on

2 Part 1: Introduction

indoor lighting attributes that should be provided (Bean, 2012). Office workers may

also react in noticeably different ways under the same indoor environment due to

factors beyond environmental parameters, such as personal and psychological aspects

(Kähkönen et al., 2008). However, characteristics of the luminous environment are the

most well-studied factors that influence occupants’ perceptions of indoor lighting

quality. Overall, research suggests that luminance distribution is a major factor

affecting occupant perception of indoor lighting quality (Hawkes, Loe, & Rowlands,

1979; Loe, Mansfield, & Rowlands, 1994).

Lighting in office buildings also accounts for approximately one-third of total

electric consumption of these buildings (Ryckaert, Lootens, Geldof, & Hanselaer,

2010). Daylighting is one the most efficient and sustainable strategies in lighting

design to enhance the indoor environment and to reduce energy consumption in

buildings (Li, 2010). It has been demonstrated that daylighting is more desirable than

electric lighting in office spaces (Heerwagen & Heerwagen, 1986). Previous research

has also established that daylight harvesting inside buildings can lead to significant

electric saving ranging from 30% to 77% (Li, Lam, & Wong, 2006; Doulos,

Tsangrassoulis, & Topalis, 2008; Ihm, Nemri, & Krarti, 2009).

While daylighting in buildings has several advantages, discomfort glarefrom

daylight, is a common problem in office spaces (Osterhaus, 2005; Rodriquez & Pattini,

2014). Discomfort glare is a sensation of annoyance or pain caused by the high-level

or non-uniform brightness in the visual field (Tashiro et al., 2015). However,

discomfort glare is a complex phenomenon and difficult to quantify, and according to

Jakubiec and Reinhart (2012), it is a neglected factor in the architecture design process

due to lack of certainty about the meaning of present metrics, the advantages of such

analyses, and how they should be applied.

Some rating tools, such as the Building Research Establishment Environmental

Assessment Method (BREEAM), the Leadership in Energy and Environmental Design

(LEED), the well Building standard, and Green Star, have been developed to assess

IEQ, with some consideration of indoor lighting quality (Iyer-Raniga, Moore, &

Wasiluk, 2014; "The Well Building Standard," 2018). However, these rating systems

may not capture visual comfort efficiently. As an example, the Green Star

environmental rating system is a recognised rating tool, which has been developed and

implemented by the Green Building Council of Australia (GBCA) since 2003 (GBCA,


2015). Previous research in Brisbane, Australia showed that roughly 50% of full-time

employees who work in buildings that are at least five star rated by GBCA experience

discomfort glare from daylight sources at their computer units (Hirning, Isoardi,

Coyne, Garcia Hansen, & Cowling, 2013; Hirning, Isoardi, & Cowling, 2014).

Another study, with 2540 participants in 36 sustainable (green) buildings across ten

countries, also indicated that glare from daylight is a major issue, despite high ratings

of indoor environmental quality (Baird & Thompson, 2012).

1.2 Research problem

Office buildings usually rely on side daylighting strategies through windows for

daylight harvesting (Huang, Niu, & Chung, 2014), and building occupants find

windows desirable. The outcomes of an extensive field study illustrated that office

workers’ satisfaction with indoor lighting is most strongly influenced by access in their

working environment to windows that can provide both daylight and an outside view

(Leder, Newsham, Veitch, Mancini, & Charles, 2016). Despite the advantages of

vertical windows in office buildings, they usually create high and variable brightness,

particularly when they are limited to a small portion of the window wall. The resulting

luminance contrast (LC) between the bright surface of the window and the

surroundings (e.g., walls and ceiling) can lead to discomfort glare.

A study of 123 buildings with installed photosensor-control systems illustrated

that there is a relatively constant relationship between the amount of illuminance from

windows and turning on the lights by occupants, in particular when dimming control

systems work correctly (Heschong et al., 2006). This study showed that as the window

illuminance increases, the likelihood of turning on the lights also increases, up to 60%,

to diminish LC between the window and surrounding areas. The impact of human

interventions in lighting conditions can reduce energy savings; the most extensive field

study on the effectiveness of side-lighting controls for daylighting showed that less

than 25% of the predicted (modelled) energy savings arising from daylight harvesting

systems were being realised in practice (Heschong, Howlett, McHugh, & Pande,

2005).

1.3 Innovative integrated lighting design solution

Creative side and top daylighting systems have been developed to enhance

indoor lighting quality in office buildings. The main aim of these systems is to send


daylight deeper into the building, while simultaneously reducing glare from sun rays

and excessive solar gains (Rea, 2000). This is mainly achieved by using optical

devices, materials, and elements, including louvres, blinds, light shelves, screens, and

light filters, especially in side-lit office buildings. According to Mayhoub (2014), the

major challenges of existing daylighting strategies are maintaining daylighting quality,

cost-efficiency, applicability, and ease of installation and operation to penetrate the

market. Accordingly, the market penetration of some of the existing innovative

daylighting systems is very limited due to high cost, and risk of discomfort glare

(Tsangrassoulis, 2008).

Because windows are a desirable feature but increase the risk of discomfort

glare, there is a need to establish strategies for improving window appearance (and

therefore visual comfort) in existing office buildings. Because luminance contrast is a

significant cause of discomfort glare, reducing this contrast may enhance visual

comfort. One potential strategy to improve window appearance (and therefore enhance

visual comfort) is to increase the luminance of the areas immediately surrounding the

window. This thesis explores a technique for increasing luminance surrounding

windows by mounting light emitting diode (LED) linear luminaries around the window

frame, the use of which (over time) would result in lower energy consumption than

occupant use of the indoor lighting system to its full capacity. The benefit of using

such a supplementary LED wall-washing system is that it can be fitted into existing

buildings with minimal construction modifications and low cost.

1.4 Purpose

This research hypothesizes that the use of a supplementary LED wall-washing

system will improve visual comfort in office buildings, and therefore will diminish

occupants’ interventions in lighting conditions, which are undesirable due to their

potential for increased energy use. The purpose of this quantitative research is to

investigate the hypothesis in typical office rooms with different orientations and

window-to-external-wall ratios. The objectives of this research are:

to examine the impact of the supplementary LED wall-washing systemon perceived window appearance, as well as on occupants’ intentionsto intervene in lighting conditions in typical office rooms with different orientations and window sizes.

to investigate ideal luminance contrasts on the window wall using theproposed LED wall-washing system in typical office rooms.


1.5 Research question

In existing and future buildings, there will be an increasing focus on energy

savings and IEQ. Thus, based on the research problem previously identified and the

knowledge gap in the literature review, a primary research question is proposed:

Under what conditions does an LED wall-washing system best integrate with windows to reduce negative lighting interventions?

Part 3, Section 3.4 describes the four sub-questions that should be addressed to

answer the primary question.

1.6 Outline

The thesis comprises five parts: an introduction (Part 1), literature review (Part

2), research design (Part 3), published and submitted papers (Part 4), and general

discussion and conclusion (Part 5). Figure 1 illustrates the research map of this study

and a summary of the overall structure is provided below.

Both individual and design factors influence perceived environmental conditions

and visual comfort of building users (Newsham, Veitch, & Aries, 2010) Hua, Oswald,

and Yang (2011). Accordingly, the following literature review is presented in two

chapters (Chapter 2.1 and Chapter 2.2). Chapter 2.1 describes the most critical factors

that affect human visual perception, including the human visual system and human

visual adaptation. It also outlines visual aspects that negatively impact human visual

comfort, such as high LC. Chapter 2.2 addresses the main factors of lighting design in

office buildings, including daylighting systems, electric lighting systems, and light

control strategies. The last section (Section 2.2.6) of this chapter looks at a synthesis

of elements from the two chapters in Part 2.

Lighting researchers have used several approaches to investigate the correlation

between subjective responses and physical stimuli. However, it is challenging to find

predictable, practical relationships between physical stimuli and personal reaction in

the field of lighting (Houser & Tiller, 2003). Part 3 therefore begins by providing

necessary background information relevant to evaluating lighting, followed by the

available strategies to analyse collected data. It provides the experimental framework

to study subjective lighting preferences for the current research. It then discusses the

participants in this study, followed by the ethical considerations of the research and its

problems and limitations. The last section provides a summary.


As outlined in Section 1.3, this research proposes an innovative LED wall-

washing system to reduce the LC between the window and surrounding walls in office

buildings. It hypothesizes that the use of this system will improve visual comfort in

buildings, and therefore, will diminish negative lighting interventions. To adequately

understand how the proposed LED wall-washing system impacts participants’

intentions to intervene in lighting conditions in the rooms with different orientations

and window sizes, it is vital to study such scenarios in real (physical) office spaces.

Therefore, we tested the hypothesis in a room facing southwest with a 15% Window-

to-exterior-wall ratio (WWR) (pilot study). Two more experiments were conducted in

rooms facing southwest and northwest with around 27% and 45% window-to-external-

wall ratios respectively (Experiments 1 and 2). However, while it is possible to

perform such experiments in existing buildings, several factors might influence the

results (e.g., WWR, the reflectance of inner surfaces, different interior space designs,

cloudy/sunny weather on different days, different outside views, different internal

brightness, etc.). These factors, which in some cases are not possible to control, could

cause experimental noise or affect the outcomes. Accordingly, we used immersive

virtual reality (IVR) technology during the last experiment (Experiment 3), allowing

the experimenter to control for most (if not all) potentially confounding features and

to isolate the variables of interest (i.e., lighting scenarios). This technology allowed

the experimenter to change the WWR and the luminaire power level of the proposed

LED wall-washing system quickly and with low cost. It also enabled the experimenter

to provide spaces where users could be fully immersed and feel a parallel sense of

presence in physical environments (Zhao, 2003; Brooks, Brahnam, & Jain, 2014).

However, a limitation associated with IVR environments is an accurate representation

of lighting. Therefore, the last experiment investigates the impact of the proposed LED

wall-washing system on participants’ lighting interventions.

Part 4 of this thesis presents four chapters, the titles of which correspond to three

published papers (pilot study, Experiments 1 and 2) and a submitted paper to be

published (Experiment 3). Each chapter begins with a connecting summary to illustrate

that the papers form a coherent, linked study. A statement of authorship is provided,

explaining the contributions of each author to the paper, as well as details of the

publication. Following that, the research paper is presented verbatim. The first paper

(pilot study), “LED Lighting Design Strategies to Enhance Window Appearance and


Increase Energy Savings in Day-lit Office Spaces” (Chapter 4.1), assesses the

efficiency of the proposed LED wall-washing system to enhance window appearance

in a typical office room facing southwest with a 15% WWR in Brisbane, Australia. A

questionnaire was developed based on previous research (Hirning et al., 2013;

Monette, Sullivan, & DeJong, 2013). Physical data was collected using a luminance

meter, an illuminance meter, and a digital camera with a fisheye lens. Furthermore,

this paper investigates the impact of using the proposed lighting strategy on the annual

energy consumption of the room in which the test was conducted. The second

publication (Experiment 1), “Improving the Impact of Luminance Contrast on Window

Appearance in a Conventional Office Room: Using Supplementary Lighting

Strategies” (Chapter 4.2), takes visual comfort evaluations in the same office room as

the previous study but with a 27% WWR. While the questionnaire of the second paper

was developed based on the outcomes of the previous publication (Chapter 4.1), the

method was not changed. The second paper investigates the influence of the proposed

LED wall-washing system on perceived discomfort glare from the window, as well as

on participants’ propensity to intervene in lighting conditions.

The third paper (experiment 2), “An Energy Efficient Lighting Design Strategy

to Enhance Visual Comfort in Offices with Windows” (Chapter 4.3), assesses visual

comfort in a typical office room facing northwest with a 45% WWR in Brisbane,

Australia. This room was chosen because of its different window size and orientation

to compare with the room that was used in previous experiments (Chapter 4.1

and Chapter 4.2). A modified questionnaire was used for collecting subjective

responses. Physical lighting measurements were used to evaluate the associations

between personal responses and physical stimulus. This study investigates the effect

of the proposed LED wall-washing system on the window appearance, as well as

participants’ propensity to change the lighting conditions. The paper also discusses the

average acceptable LC between the window and surrounding walls using the

supplementary LED linear luminaires.

The final research paper (experiment 3) presents data collected in an IVR office

room designed to be located in Brisbane, Australia. A questionnaire was developed

based on those used in previous experiments (Chapters). The paper, “Innovative

Window Design Strategy to Reduce Negative Lighting Interventions in Office

Buildings”, examines subjective responses concerning the intentions of participants to


change the LC on the window wall under different lighting conditions. It discusses

how participants might respond to different luminance patterns on the window wall

brought about through randomly changing the WWR and the luminaire power of the

supplementary LED linear luminaires. These publications, when taken together,

evaluates the benefits and limitations of the an integrated lighting design solution to

improve window appearance that leads to increased energy savings in office buildings.

Part 5 summaries the significant points of this study as well as their implications

for future research. A brief literature review of existing lighting evaluation metrics and

simulation tools has been added in Appendices (Appendix A). In addition, a sample of

questionnaires used in each experiment is placed in the appendices.


Figure 1 Research map

10 Part 2: Literature Review

Literature Review

Chapter 2.1 Human factors in lighting

The main aim of lighting design in office buildings is to enable occupants to

perform their work efficiently, quickly, safely, and comfortably (IESNA & Rea, 2000).

In order to predict users’ behaviour under different lighting conditions, it is essential

to comprehend the physical, physiological, and perceptual features of the visual system

(IESNA & Rea, 2000). Consequently, the first section of this chapter describes

different factors that work together to help individuals to perceive a scene, including

light, the human optical system, and the human perceptual system.

Even though the human visual system can process data over various ranges of

luminance, it cannot process all of them at once and needs to adjust itself to different

lighting conditions (Boyce, 2003); according to Jameson and Hurvich (1964), this

visual adaptation is one of the most significant factors of visual perception. Thus, the

second part of this chapter defines visual adaptation and its mechanisms. It then

describes the implications of visual adaptation, such as the adaptation level.

It is evident that the presence of visual and psychological comfort conditions in

office buildings increases workers’ motivation, and that this will lead to enhanced

productivity and higher performance (Manav, 2007). Hence, the third section of this

chapter defines visual comfort and the factors that can negatively affect it, such as

problematic light distribution and glare. Finally, the last part of this chapter presents a

summary of this and outlines the knowledge gap.

2.1.1 Visual system

The visual system is an image processing system that consists of the eye (optical

system) and brain (perceptual system) that work together to interpret the visual

environment (IESNA & Rea, 2000). The visual system can only operate when light

exists (Boyce, 2003). The luminous environment allows the creation of the retinal

image that is the stimulus for the process of vision; this image then forms the basis of

the visual perception process to recognise objects and faces for further interpretation

by the individual (Cuttle, 2008).

Part 2: Literature Review 11

2.1.1.1 Light

Light is a fundamental need for individuals and influences their physical,

physiological and psychological behaviour (Bellia et al., 2011). Light is also part of

the electromagnetic spectrum, which is arranged based on the wavelength or frequency

of radiant energy (IESNA & Rea, 2000). Figure 2 shows a schematic comparison

diagram of wavelength and frequency for the electromagnetic spectrum. This figure

also illustrates different types of radiant energy that make up the electromagnetic

spectrum, such as radio waves, microwaves, infrared, visible light, ultraviolet, X-rays

and gamma rays. Wavelength and frequency are measured in meters and cycles per

second respectively, while wavelength is used to quantify two types of radiant energy,

including infrared and visible light (NASA, 2013).

Figure 2 Schematic diagram of the electromagnetic spectrum (NASA, 2013)

2.1.1.2 Quantifying light and minimum lighting levels

Light is often defined in relation to the stimulus it provides to the human visual

system. Aschehoug et al. (2000) defined light as “radiant energy evaluated according

to its capacity to produce a visual sensation.” The human visual system is sensitive to

a narrow band of electromagnetic energy, ranging from about 380 nanometres (nm) to

770 nm (IESNA & Rea, 2000). As a result, there are a range of different units that can

be used to quantify light, including lumen (lm), candela (cd), lx, and candle per square

meter (cd/m2), which are used to describe different types of light measurements

(Tregenza & Wilson, 2011). Table 1 illustrates four terms that are used to describe

light and their units.


Table 1 The light quantities

Measure Definition UnitsLuminous

FluxThe quantity of total light output from a source (lamp or

window) that depends upon wavelength as well aselectromagnetic power to produce a visual sensation (Tregenza

& Wilson, 2011; Bean, 2012).

Lumen (lm)

Luminous intensity

The luminous flux emitted in a very narrow cone holding the given directions divided by the solid angle of the cone (Boyce,

2003).

Candela (cd)

Illuminance The luminous flux falling on a surface per unit area: lumens per square meter (lm/m2) (Aschehoug et al., 2000).

Lx

Luminance The luminance intensity of an element, which can be a small portion of a surface of a light source or a surface transmitting or

reflect light, divided by the space apparently producing the intensity (Bean, 2012).

cd/m2

Lighting levels inside buildings are often described in lx (Bean, 2012).

Satisfactory indoor lighting quantity is influenced by several factors, such as visual

tasks, safety, activities in the space, furnishings, spatial forms, space dimensions, the

age of occupants and the expectations of clients (Steffy, 2008). However, the amount

of illumination needed for a particular task may vary from person to person based on

their visual capacity (Bean, 2012). There are several standards for indoor minimum

lighting level of spaces (Mills & Borg, 1999). For example, the Australian standard

recommends a minimum of 320 lx lighting level on the working plane in office

buildings for general tasks involving reading, typing, and writing, whereas the

European standard recommends 200 to 600 lx and the American standard recommends

500 lx (Mills & Borg, 1999; Garcia-Hansen, 2006; AS/NZS, 2008; Fies T.S. &

Mathers M., 2009). It should also be noted that even though increasing the lighting

level in working environments enhances visual performance, research suggests that,

after reaching a certain level of performance, further increases in the amount of light

bring moderately little enhancement (Bean, 2012).

2.1.2 Perceptual system

The luminous environment and the perceived luminous environment are not the

same thing (Cuttle, 2008). The initial interpretation of the brain is influenced by

several factors, such as previously absorbed data by individuals to analyse the meaning

of a visual scene, their current physical or emotional state, and the physical behaviour


of the eye itself (Bean, 2012; Tregenza & Loe, 2014). The following section outlines

a number of issues relating to the perception of the visual environment.

2.1.2.1 Perceived brightness

Brightness is defined as the perceived intensity of a visual stimulus, regardless

of its source (Corney, Haynes, Rees, & Lotto, 2009). In other words, luminance (the

light entering the eye) raises the sensation of brightness, which is a subjective

experience and depends on more than physical luminance (Gordon, 2003). Although

there is a monotonic relationship between brightness and luminance, the brightness of

different stimuli with the same luminance could alter with their relative saturation as

well as with shifts in the spectral distribution of the stimulus (Pridmore, 2007; Corney

et al., 2009). For instance, grey stimuli appear less bright than strongly coloured

stimuli, and a region looks brighter if it is surrounded by dark areas (Gordon, 2003;

Corney et al., 2009). Moreover, according to Tregenza and Loe (2014), the brightness

of a light source is also influenced by its size; a smaller source appears brighter than a

larger one with the same light output.

It is well understood that sources that are too bright to compare with their

surroundings can impair the human vision and cause glare (BSI, 2011; Bean, 2012).

Perceived brightness of a source is influenced by the real pattern of brightness in the

field of view (FOV) and the state of eye adaptation, besides the luminance of that

source (Tregenza & Loe, 2014). The primary sources of perceived brightness inside

buildings that can cause discomfort glare are daylight through apertures and electric

lights (Bean, 2012). Likewise, the uniformity of luminance distribution inside a room

can affect the perception of brightness (Aschehoug et al., 2000). For instance, a non-

uniformly lit room appears brighter than a room with uniform luminance distribution,

due to the luminance contrast inside a non-uniformly lit room (Tiller & Veitch, 1995).

2.1.2.2 Perceived colour

Ambient illumination colour is another aspect of ambient lighting that should be

considered besides its overall brightness (Cuttle, 2008). The illumination colour of

transmitted daylight or a lamp can be described by two attributes: (1) the colour

appearance of the light and (2) its colour rendering capabilities (BSI, 2011). The

apparent colour of the light emitted is called colour appearance, which is quantified by

correlated colour temperature (CCT) that is expressed in Kelvin (K) (Bean, 2012).

Colour rendering is the usual expression for the impact of a light source on the colour


appearance of objects in subconscious or conscious comparison with their colour

appearance under different light sources (IESNA & Rea, 2000).

2.1.2.2.1 Correlated colour temperature (CCT)

CCT has been described as the one-dimensional definition of the colour of light

sources that are near white (Borbély, Sámson, & Schanda, 2001). In other words, CCT

illustrates the range of illumination colour appearance from yellowish-white to bluish-

white (Cuttle, 2008). There is also a monotonic relationship between the CCT and the

quantity of blue light in the spectrum of light sources (Mills et al., 2007). Furthermore,

the colour appearance of a reference can be defined based on the CCT values: (1) warm

for low CCT values (below 3300K), (2) intermediate (between 3300-5300K), and (3)

Cool for high CCT values (above 5300K) (BSI, 2011; Bean, 2012). It should also be

noted that daylight under clear sky conditions has a CCT around 5500K (Lei et al.,

2007). Finally, although measuring CCT is not as straightforward as measuring

illuminance, an instrument called a chroma meter can be used to measure CCT (Cuttle,

2008).

It is known that the effects of ambient illumination interact with different

independent variables (IESNA & Rea, 2000). Some studies have indicated the

influence of colour appearance on individuals and investigated preferred CCT inside

office buildings, and it has been found that colour temperature can affect human visual

perception and mood even if the illuminance level is constant (Knez, 2001). For

instance, a study found that human mental activity increases when we perceive higher

colour temperature (7500K versus 3000K) (DEGUCHI & SATO, 1992). Another

study has also suggested that drowsiness can be reduced by increasing the colour

temperature of lighting inside buildings when comparing 3000k with 5000k (Noguchi

& Sakaguchi, 1999). Likewise, another study suggested that a combination of cool and

warm colour temperatures can be recommended for the indoor lighting design of office

buildings (Manav, 2007).

The luminance quantity and CCT of references affect human visual perception

and responses that are fundamentally linked to their visual comfort and satisfaction

. Most national and

international lighting design guides recommend a lighting level with high CCT (Ju,

Chen, & Lin, 2012). Figure 3 illustrates the relationship between CCT and illuminance

of an ambient illumination based on Kruithof (1941) study. According to the Kruithof


curve, humans prefer lower CCT when the illuminance is lower, and prefer higher

CCT when the illuminance is higher (Ju et al., 2012). Moreover, combinations of CCT

and illuminance that lie in the upper shaded areas in Figure 3 are perceived as

excessively unnatural and colourful; those in lower shaded spaces are viewed as dim

and cold (Boyce & Cuttle, 1990). Regarding psychological effects, according to

Noguchi and Sakaguchi (1999), high CCT combined with high illuminance creates an

energetic mood, whereas low illuminance and low CCT creates a relaxed atmosphere.

For example, Manav (2007) states that a 4000K colour temperature in an office cell is

preferred to 2700K for the impression of comfort.

Figure 3 Combinations of CCT and illuminance that give a pleasing impression (clear space)

Regarding the association between CCT and perceived brightness, there is a

dependence of brightness on CCT, and a light source with higher CCT reaches the

same perceived brightness as sources with lower CCT, but with less photopic

illuminance (Harrington, 1954). Moreover, some studies advocated that illuminated

rooms with higher CCT lamps will appear brighter than illuminated rooms with lower

CCT lamps, while other light variations are held constant (Berman, 1992; Steffy,

2008). Manav (2007) also suggests that, as perceived brightness is a subjective

appraisal, it is hard to recommend a CCT that improves indoor lighting brightness.

Added to this, one study suggested that colour adaptation influences perceived

brightness (Harrington, 1954). However, other studies have suggested that there is no

association between CCT and perceived brightness (Hu, Houser, & Tiller, 2006; Park,

Chang, Kim, Jeong, & Choi, 2010; Fotios, 2017). Overall, Ju et al. (2012) state that

there is not any consensus on the relationship between CCT and perceived brightness

to enhance indoor visual comfort.


2.1.3 Visual adaptation

Aschehoug et al. (2000) defined adaptation as the process by which the state of

the human visual system is modified by previous and present exposure to stimuli that

may have various luminances, spectral distribution, and angular substance. The optical

system of an individual continuously adjusts itself to the level of light it is receiving

through changing its sensitivity (Boyce, 2003). Thus, human vision can function over

an enormous range of brightness from intense sunshine to faint starlight (Cuttle, 2008).

Nonetheless, regular changes in lighting over a short period may produce visual

annoyance (Lee, Yoon, Baik, & Kim, 2013). Overall, the adaptation of human eyes

from dark to light is moderately rapid to compared with their adaptation from light to

dark (Bean, 2012).

The quantity of the luminance of the visual field as it influences the viewer’s

state of adaptation is called the adaptation level, and a luminance value range of less

than a ratio of 1:100 in the FOV is acceptable within the definition (Cuttle, 2008).

Luminances considerably higher or lower than adaptation level appear brilliant or dark,

respectively (Schreuder, 2008). However, while humans all have the same eye

structure, eye adaptation is subjective and can vary between people (IESNA & Rea,

2000).

2.1.4 Visual comfort

Although we can adapt to lighting over time, we are less effective at adapting to

bright and dark light that is simultaneously present in our field of view. Visual comfort

is described as “a subjective condition of visual well-being induced by the visual

environment” (European standard, 2011). As defining visual comfort is much harder

than discomfort glare, most research has tended to focus on ensuring the absence of

visual discomfort (Waide, Tanishima, Harrington, & Iea, 2006), or discomfort glare as

it is also known (Bean, 2012).

As suggested by the definition, there is a psychological aspect of visual comfort.

However, there are also some physical aspects of the visual environment that can be

used to evaluate it in an objective way (Frontczak & Wargocki, 2011). For instance, it

is suggested that visual comfort can be achieved through providing a sufficient

quantity of light for the expected visual task, adequate directionality to model three-

dimensional surfaces and objects, uniform distribution of luminance and illuminance,


sufficient spectral content to render colours correctly when needed, and absence of

glare (Aschehoug et al., 2000). Overall, according to Linhart and Scartezzini (2011),

to provide visual comfort in office buildings, the horizontal illuminances, particularly

on work-planes, should be adequately high and well distributed, while at the same time

avoiding discomfort glare from light sources.

2.1.4.1 Glare

Aschehoug et al. (2000) have defined glare as (p. 8-4):

a visual condition which results in discomfort, annoyance, interference

with visual efficiency, or eye fatigue because of the brightness of a

portion of the field of view (lamps, luminaires, or other surfaces or

windows that are markedly brighter than the rest of the field).

Glare inside buildings usually occurs by lighting systems, and there is a linear

association between the average luminance of glare patches and the average luminance

of the visual field (Gordon, 2003; Kim & Kim, 2012). Thus, glare can be avoided by

limiting the absolute lighting level of any surface, daylight element or windows, or

electric lights (Steffy, 2008). Glare has also generally been classified into two

categories to assess outdoor lighting and indoor lighting. These are ‘disability glare’

and ‘discomfort glare’, which can occur simultaneously or separately (Vos, 2003;

Jiang, Sun, Chen, Yang, & Chang, 2014). Disability glare is the impact of high

luminances of space in the visual field that impairs the visual system to some extent

(Boyce, 2003). Discomfort glare is a sensation of annoyance or pain caused by high or

non-uniform brightness in the visual field (Aschehoug et al., 2000). The difference

between these two types of glare can perhaps be better described through using the

(translated) German terms of ‘physiological glare’ instead of ‘disability glare’ and

‘psychological glare’ instead of ‘discomfort glare’ (Osterhaus, 2005).

2.1.4.1.1 Disability glare

According to Rea (2000), disability glare is most common during daytime from

the sun and during night-time on roads from oncoming headlights, and is hardly

significant in interior spaces. However, the probability of disability glare by sunlight

or diffuse skylight inside buildings will be highest if glazing constitutes about 40-55%

of the wall area (Muneer, 2000). Disability glare from vertical windows can also occur

through looking at the reflecting wall areas of neighbouring buildings that are


illuminated by the sun (Corrodi, Corrodi, & Spechtenhauser, 2008). According to

Jakubiec and Reinhart (2012), in some conditions, including increasing brightness,

prominence, and size of a source of glare, discomfort glare can turn into disability

glare. Moreover, as the luminance of origin at which disability glare occurs is usually

higher than discomfort glare and causes an immediate reduction in the ability of

viewers to see or to perform a task, occupants usually notice this and may react through

using a shading devices or shifting their position (Osterhaus, 2005). It should also be

noted that even though users always perceive discomfort glare when they experience

disability glare, they may perceive discomfort glare without experiencing disability

glare (Jakubiec & Reinhart, 2012).

2.1.4.1.2 Discomfort glare

Discomfort glare depends on human adaptation luminance and the

characteristics of surrounding light sources (Mainster & Turner, 2012). Discomfort

glare from light sources does not necessarily affect the ability of an individual to

complete a task but may cause specific physiological and psychological symptoms

such as headaches or stress (Tashiro et al., 2015). Likewise, as discomfort glare relies

on subjective evaluation, it is difficult to quantify or forecast perceived discomfort

glare inside buildings (Velds, 2002; Shin, Yun, & Kim, 2012a). Thus, although

disability glare is well-known and can be measured and predicted, the psychological

mechanisms of discomfort glare are not well understood (Aschehoug et al., 2000).

2.1.4.2 Light distribution

The uniformity of illuminance can be described as the ratio of least illuminance

to the average illuminance on a surface (Alrubaih et al., 2013). It is recommended that

the uniformity of the illuminance of a working plane should not be less than 0.8,

whereas higher illuminance uniformity illustrates better visual comfort for workers

(Hannaford, 2002). However, daylight from windows produces a non-uniform

distribution luminance inside buildings (Kim, Kim, & Ahn, 2008). Moreover, Alrubaih

et al. (2013) hold the view that issues of illuminance distribution are especially

significant in daylighting applications because the suggested design intensity for

electric lights has been gradually lowered to minimise energy consumption.

Luminance distribution is another aspect of light distribution. Luminance

distribution is defined as uniformity in brightness over spaces with constant luminance

Fluctuations (Arend, Buehler, & Lockhead, 1971). Öztürk (2003) states that luminance


distribution inside an area depends on the characteristics of the illuminated surfaces

and the uniform illuminance distribution. He also argues that setting up lighting

installations without paying attention to the type of reflection of surfaces and

uncontrolled illuminance distribution can cause indoor high LC. Although it is well

understood that LC is necessary to improve visual performance, great LC should be

avoided to prevent visual discomfort (Alrubaih et al., 2013).

2.1.5 Luminance contrast

According to Bean (2012), vertical surfaces in an environment play a significant

role in the perception of discomfort glare, and a range of specific luminance contrasts

have been suggested for different applications. The Chartered Institution of Building

Services Engineers (CIBSE) and the Illuminating Engineering Society of North

America (IESNA) recommend that the LC between light sources and adjacent areas,

and anywhere within the normal FOV, should be less than 20:1 and 40:1 respectively

(Boubekri & Boyer, 1992; CIBSE, 1994). The Swedish Agency for Economic and

Regional Growth (NUTEK) in Sweden has stricter recommendations: that the

luminance contrasts between any points within FOV should not exceed 1:20 (Boubekri

& Boyer, 1992). Other recommendations suggest that the LC between the work

surface, immediate surrounds, and distant areas should be less than the ratios 1:3:10

(Arup & Royal Institute of British Architects, 2007). Despite these recommendations,

there is a limited investigation of what LC provides in term of optimal visual comfort,

and whether LC changes at different levels of overall illuminance and office settings.

2.1.6 Rating tools

Rating tools exist to assess indoor environment quality and the energy

performance of buildings, such as the Building Research Establishment Environmental


(LEED), and Green Star (Iyer-Raniga et al., 2014). As an example, the Green Star

environmental rating system is a recognised rating tool, which has been launched and

developed by the green building council of Australia (GBCA) since 2003 (GBCA,

2015). However, these rating systems may not capture visual comfort effectively. A

previous study in Brisbane, Australia found that roughly 50% of full-time employees

who work in buildings that are at least five-star rated by GBCA experience discomfort

glare from daylight sources at their computer unit (Hirning et al., 2013; Hirning et al.,


2014). Another study with 2540 participants in 36 sustainable (green) buildings across

11 countries also indicated that glare from daylight is a major issue (Baird &

Thompson, 2012).

2.1.7 Summary

Visual perception can be influenced by the physical parameters of ambient

illumination. Many scholars hold the view that some personal factors, including

previously absorbed data to analyse the meaning of a visual scene, current emotional

and physical states, age, and perceived visual adaptation level can influence human

visual perception. It is also evident that the human visual system continuously adapts

itself to different perceived light levels and brightness. Moreover, findings from

studies suggest that human eye adaptation from light to dark is somewhat slower than

its adaptation from dark to light. Accordingly, Pridmore (1999) gave 10 minutes to

subjects to adapt to perceived indoor ambient light before commencing each

experiment in an entirely dark room. However, there is a paucity of evidence to

determine the adaptability of the human eye from experiences of discomfort glare from

windows.

It is a widely held view among researchers that perceiving discomfort glare

inside buildings can be influenced by vertical and horizontal luminance distribution in

the FOV of occupants. Moreover, according to Nazzal (2005), even minor effects like

discomfort glare can accumulate and cause psychological and functional disorders.

However, studies agree there is a lack of satisfactory knowledge to effectively predict

discomfort glare in practical circumstances (Galasiu & Veitch, 2006; Clear, 2013). He

also states that much uncertainty still exists about the psychological and physiological

basis of the discomfort experienced. Finally, little is known about what LC between

the window and surroundings provides optimal visual comfort, and whether this LC

changes at different levels of overall illuminance and office settings.

This chapter has focused on human factors in lighting and particular human

factors in perceived discomfort glare from daylight. However, as discussed in the

introduction (Part 1), lighting design also considerably affects indoor visual comfort.

Thus, the following chapter will discuss lighting design factors to meet indoor visual

comfort better, as well as saving energy. Moreover, based on the research problems

and research questions of this study, which were explained in Part 1, the next chapter


will focus on lighting design strategies that can be used to improve window appearance

in office buildings. Finally, a synthesis of elements from Chapter 2.1 and Chapter 2.2

will be discussed at the end of Chapter 2.2, which will also outline the knowledge gap.


Chapter 2.2 Lighting design

The lighting design of workplaces ordinarily aims to enhance the quality and

production of a comfortable lighting environment (Fontoynont, 2002). Over the last

decade, identifying and optimising indoor lighting conditions based on both visual and

non-visual functions has become increasingly significant (Borisuit, Linhart,

Scartezzini, & Münch, 2015).

Two sources of light, such as daylight and electric light can be used for interior

lighting design of office buildings (IESNA & Rea, 2000). Access to bright light in the

working environment enhances a worker's mood, productivity, energy, and alertness

(Avery, Kizer, Bolte, & Hellekson, 2001). The significance of utilizing daylight inside

buildings has been highlighted by several studies (Farley & Veitch, 2001; Hourani &

Hammad, 2012; Boubekri, Cheung, Reid, Wang, & Zee, 2014; Ozorhon & Uraz,

2014). However, even though daylight provides illumination across the entire colour

spectrum, the time of day and cloud cover can affect the light from daylight moment

to moment (Ander, 2003), and therefore cannot always be relied upon to consistently

provide appropriate lighting by itself.

Electric lights not only supplement daylight levels when daylight availability is

reduced, but can also aid the visual and psychological comfort of the buildings’ users

by responding to the changeable quality of daylight (Kazanasmaz, Günaydin, & Binol,

2009; Bean, 2012). Consequently, it is necessary to integrate daylighting systems with

electric lights from the beginning of the design process to produce an indoor

environment that effectively restricts the intensity of daylight that penetrates into the

area and provides a controllable and uniform light distribution (IESNA & Rea, 2000).

Currently, lighting research and good lighting design in buildings have three

main directions: 1, design and selection of appropriate daylighting systems, 2,

strategies and development of energy efficient lighting equipment, and 3, using

suitable lighting control systems (Doulos, Tsangrassoulis, & Topalis, 2005). The first

section of this chapter gives a brief overview of the influence of window characteristics

on indoor visual comfort, as well as on the energy consumption of buildings. The

second section presents a study of existing classifications for daylighting systems and

describes the two most common of these strategies that are used in office buildings,

including blinds and light shelves. Furthermore, as explained earlier, it is clear that


side lighting systems should be integrated with energy efficient electric lighting

systems to enhance indoor visual comfort and to decrease the energy consumption of

office buildings. Thus, the third part of this chapter discusses advantages and

disadvantages of using the three most significant existing energy efficient electric

lights, such as energy efficient halogens, compact fluorescent lamps, and LEDs. It will

then go on to describe the most common existing lighting control systems, which are

used in office buildings. The final section outlines the summary of this chapter

following by the synthesis of elements from Chapter 2.1 and Chapter 2.2.

2.2.1 Daylighting systems

Daylighting enhances lighting quality in buildings, providing sufficient light

levels and colour rendering, while also providing a healthy and comfortable

environment for building occupants (Tsangrassoulis, 2008). Reinhart and Weissman

(2012) have defined daylighting as “controlled use of daylight in and around

buildings.” Daylighting systems have been designed to achieve one or more of the

following aims: increasing daylight to under-lit areas, enhancing daylighting for visual

tasks, improving visual comfort, and providing satisfactory shading (Aschehoug et al.,

2000). There are two main daylight strategies, top-lighting (e.g., skylights) and side-

lighting (windows) according to the opening position in the building (Ander, 2003;

Alrubaih et al., 2013; Gago, Muneer, Knez, & Köster, 2015).

Workplaces usually rely on vertical windows for daylight harvesting (Huang et

al., 2014). Different strategies are used to improve the efficiency of vertical windows

in office environments, including light shelves, prismatic panel systems, holographic

optical elements, anidolic ceilings, and louvres and blinds. However, venetian blinds

and light shelves, which will be described in the following sections, are two of the

most widely used daylighting systems to control discomfort glare (Alrubaih et al.,

2013).

2.2.1.1 Light shelves

Light shelves are designed to shade and reflect light towards the surface of the

ceiling and to protect direct glare from the sky (Aschehoug et al., 2000) while

maintaining the outside view from inside the building (Kischkoweit-Lopin, 2002;

Nair, Ramamurthy, & Ganesan, 2014). These systems are suitable for tropical and sub-

tropical regions due to diminishing heat gain and glare as well as significantly


enhancing illuminance levels in the rear of a room (Freewan, Shao, & Riffat, 2008).

Furthermore, the orientation and position of both the light shelves and the ceiling have

significant roles in the performance of these systems (Gago et al., 2015). The best

ceiling shape to improve illuminance uniformity by using the light shelf strategy inside

the building is the curved shape (Freewan et al., 2008). Finally, it is believed that

spaces with installed light shelves consume less electricity than those with

conventional fenestrations, and in comparison to other daylighting strategies, they are

more reliable and effective (Sanati & Utzinger, 2013).

2.2.1.2 Louvres and blinds

Louvres and blinds are designed to limit the sunlight in the vicinity of

fenestrations as well as introducing it to the back of indoor spaces (Gago et al., 2015),

and can be used in different climate regions and all orientations (Aschehoug et al.,

2000). These systems are made up of vertical, horizontal, or sloping slats (Aschehoug

et al., 2000), of which the shape, size, colour, configuration, and rotation angle

influence both glare and external environment visibility from indoor spaces

(Tzempelikos, 2008). In addition, Hammad and Abu-Hijleh (2010) hold the view that

these systems can decrease electric consumption in office buildings by roughly 24%.

Finally, it is shown that reflective louvres can raise working plane illuminance by

approximately 70% under clear sky condition (Leung, Rajagopalan, & Fuller, 2013).

Blinds are one of the traditional shading devices that prevent or adjust discomfort

glare from apertures under varying external sky conditions (Piccolo & Simone, 2009).

They also provide the flexibility of preventing direct sunlight from entering the room

while giving access to the outside view and diffuse daylight (Karlsen, Heiselberg, &

Bryn, 2015). However, several studies have shown that blinds are frequently pulled

down by occupants to avoid discomfort glare in office areas (Galasiu & Veitch, 2006;

Yan et al., 2015). These surveys also showed that the sun position and sky conditions,

as well as the location of the workspace, impact the blind position and the slat angle

chosen by users. For example, it has been reported that blinds are closed under a clear

sky and not operated under an overcast sky (Galasiu & Veitch, 2006). Also, several

studies have shown that there is a significant relationship between orientation and

window blind position (Rea, 1984; Foster & Oreszczyn, 2001; Galasiu & Veitch,

2006). For instance, Foster and Oreszczyn (2001) used a video camera to record the

venetian blind position in three buildings located in the UK and examined the impact


of facade orientation and sunlight on blind usage. They found that, on average,

approximately 40% of all facades are occluded by blinds, with the south facade

representing the most significant amount of occlusion, followed by the east, west, and

north respectively. Finally, Escuyer and Fontoynont (2001) asserted that although it is

likely that occupants will close blinds, it is much less likely that they will raise the

blinds again when there is no glare or overheating problem, especially when they have

a poor outdoor view. Accordingly, these interventions in lighting conditions can lead

to an unnecessary increase in the energy consumption of buildings.

2.2.2 Impact of window characteristics on visual comfort and energy savings

Several studies have stressed the importance of windows and daylight and have

confirmed their abundant positive effects on building occupants (Farley & Veitch,

2001; Hourani & Hammad, 2012; Boubekri et al., 2014; Ozorhon & Uraz, 2014).

Aspects of window design, such as WWR, position, configuration, and overall size are

all important considerations in the design of acceptably daylit buildings (Kevin Van

Den & Meek, 2015). For instance, Gratia and De Herde (2003) demonstrated that the

penetration of daylight inside buildings could be enhanced by designing window in an

upper portion of the wall. Bodart and De Herde (2002) argue that increasing WWR

can reduce electric lighting consumption of office buildings. Conversely, it is reported

that increasing window size does not necessarily lead to diminishing electric usage for

lighting in buildings (Poirazis, Blomsterberg, & Wall, 2008). However, windows can

cause discomfort glare due to the frequent existence of high LC between them

especially when they are limited to a small portion of the wall (Alrubaih et al., 2013).

Currently, a high proportion of glazing is the main characteristic of many green

office buildings due to the desired high levels of daylight penetration (Byrd, 2012).

However, according to Byrd (2012), the highly glazed facade can cause unsatisfactory

IEQ for occupants, as well as increased energy consumption of buildings due to the

use of blinds and electric lights by residents. Large window areas may also directly

influence the energy consumption of the building through allowing large heat gains or

losses (Alrubaih et al., 2013). In the case of fully glazed facades, high window

luminance or direct sunlight from a window could lead to discomfort glare (Shin, Yun,

& Kim, 2012b).


There are some existing recommendations for an appropriate design for

daylighting. According to Saini (1980), to achieve a well day-lit room for sunny

regions, the ratio of glazing area to overall floor area should be about 1/16. It is also

suggested that this ratio should not be less than 1/25 for overcast sky conditions

(Pollock, Roderick, McEwan, & Wheatley, 2009). Furthermore, several guidelines

suggest the use of more advanced sunlight prediction techniques, which forecast the

sunlight level at a given point in a room (Alshaibani, 2015). Overall, it is suggested

that a building with typical façade, which has about 30% WWR, is more probable to

consume less energy than a building with fully glazed façade (Meek & Wymelenberg,

2015).

Dubois et al. (2011) demonstrated some factors that affect designing day-lit

spaces in office buildings, including window characteristics, latitude, and orientation,

the reflectance of inner surfaces, shading devices, partition height in open-plan office

buildings, and ceiling height. Moreover, according to Meek and Wymelenberg (2015),

the construction orientation and footprint geometry have a significant impact on

daylight harvesting. The study also noted that determining an effective depth from

perimeter windows, which provide occupants access to the external view and daylight,

is the primary crucial factor of preferred day-lit spaces. It is also believed that residents

perceive less discomfort glare from a window with an interesting view than from a

window with a less interesting view at the same daylight level (Tuaycharoen &

Tregenza, 2007). Finally, side-lighting fenestration systems provide daylight with a

strong directionality, which decreases as the distance from the window increases

(Alrubaih et al., 2013) (see Figure 4). Thus, the contrast between the adjacent area of

windows and that at the opposite end of the room can lead to some problems for the

workers (Ochoa & Capeluto, 2006).

Table 2 shows daylight penetration assessment in the case study patterns based

on the Banner Bank Building in Boise, United States of America with a 40% WWR

with a window head height at 2.90m, a sill height at 0.91m, and a ceiling height at

3.05m (Meek & Wymelenberg, 2015). Interior reflections in this case study are

approximately 80%, 20%, and 50% for ceiling, floors and walls, respectively.


Figure 4 Penetration of daylight inside buildings

Table 2 Daylight penetration evaluation in the case study patterns (Kevin Van Den & Meek, 2015)

Depth of the indoor space

Percentage of floor area with above 300 lx when there is no furniture on it

3.05 m 100%6.10 m 95%7.92 m 85%9.14 m 80%

12.19 m 50%15.24 m 35%

2.2.3 Electric lights

One of the most effective methods to diminish greenhouse gas emissions and to

save energy is using energy efficient technologies (De Almeida, Santos, Paolo, &

Quicheron, 2014). The efficacy of artificial light is defined as the ratio of the light

output to the input power (lm/W), and lamps with higher efficiency need less energy

to provide a given amount of light (Waide et al., 2006). The difference between the

power consumption of the most and least efficient lighting technologies is about 75 to

85%. Therefore, there can be a dramatic reduction in energy use by switching to

efficient lighting technologies (Corazza, Giorgi, & Massaro, 2012). Energy efficient

halogen (EEH), fluorescent lamp (FL), and LED are the most energy efficient types of

lighting that are currently available on the market (Fenton & Moseley, 2014).

2.2.3.1 Energy efficient halogen

EEH lamps consume at least 20 to 25% less energy for the same light output than

ordinary incandescent lamps (European commission, 2015). EEH, which has an

infrared coating, can enhance the energy efficiency by up to 45% more than the best

common incandescent light bulbs (European commission, 2015). Furthermore,


according to the Department of Industry and Science of the Australian Government

(2015), EEH is less energy efficient than CFL and LED, and the least energy efficient

halogens are being phased out over time.

2.2.3.2 Fluorescent lamps

FL has been designed to fit into light fixtures that were previously used for

incandescent light globes, and it consumes one-fifth to one-third of the electric power

while lasting eight to fifteen times longer (Guan, Berrill, & Brown, 2015). FL also

produces less heat, which can reduce cooling energy consumption inside buildings

during summer (Department of Industry and Science of the Australian Government,

2015). Although there are some issues with its light quality and cost, FL is currently

the most popular energy efficient lamp (Yuen, Sproul, & Dain, 2010). However, it is

estimated that LED lighting will take the lead market position in 2020 by taking about

64% of the overall lighting market while the proportion of other energy efficient

products such as compact fluorescent lamps will be roughly 27% (McKinsey &

Company, 2012).

2.2.3.3 Light emitting diodes

Light emitting diodes have become increasingly important in the last 10-15 years

(De Almeida et al., 2014), and their performance is improving so quickly that the U.S.

Department of Energy has revised its projections more than three times during the last

five years (Sandahl, Cort, & Gordon, 2013). Light emitting diodes can offer flexibility,

high energy efficiency, and long lifetimes, while providing high performance,

including a wider operation temperature (-20°C to 80°C), no low temperature start-up

problems, a wider range of controllable colour temperature (4500 K to 12,000 K), and

rapid response time (Chang, Das, Varde, & Pecht, 2012). Nevertheless, De Almeida

et al. (2014) mentions some disadvantages of light emitting diodes usage, including:

comparatively high initial cost

lack of standardization

risk of discomfort glare due to the small lamp size

a need for thermal management to avoid degradation in lifetime

blue pollution from cool-white light emitting diodes

temperature dependence (ambient temperature greatly affects the LED performance).


However, it is expected that light emitting diodes will deeply improve the

methods in which we use light, and they have already replaced traditional lights in

many lighting systems, such as signs, displays, and traffic lights (Crawford &

Crawford, 2009). The consumption of light emitting diodes instead of other types of

lamps is becoming more popular for the indoor electric lighting of buildings due to

their:

higher efficiency

less radiant heat

the lower waste heat that leads to less cooling energy consumption

minimum maintenance cost

minimum problematic environmental effects

enhanced design possibilities (De Almeida et al., 2014).

The best quality light emitting diodes that are currently available on the market

are four to five times more efficient than normal incandescent globes (Department of

Industry and Science of the Australian Government, 2015). In addition, it is predicted

that the efficacy of light emitting diodes will approach 250 lm/W by 2020 (the efficacy

of normal cool-white LEDs is currently over 100 lm/W) which could lead to a

reduction in global lighting consumption by more than 50% and in total electric use

by around 20% (De Almeida et al., 2014). Finally, the price of LEDs has reduced

rapidly over recent years, and the market is shifting from traditional lighting

technologies to LED (McKinsey & Company, 2012).

Light emitting diodes can be easily controlled with various illuminance and CCT

levels by using dimming technology and a combination of LED devices (Kim, Jang,

Choi, & Sung, 2015). Lei et al. (2007) hold the view that among different kinds of

light emitting diodes, white LED has been identified as the green lighting source of

the 21st century.

White light emitting diodes can be divided into three categories, including warm-

white (2700 K to 3000 K), neutral-white (3500 K to 4000 K) and cool-

K), whilst the daylight under clear sky condition usually has a correlated colour

temperature (CCT) around 5500 K (Lei et al., 2007). It is believed that cool-white light

emitting diodes with high CCT tend to offer higher efficiency at low cost (Office of

Energy Efficiency and Renewable Energy, 2015). However, a study suggested that


neutral-white LED (4000 K) is more preferable than cool-white LED (6400 K) at 500

lx in office spaces based on overall comfort and overall acceptance (Islam et al., 2015).

2.2.4 Lighting control systems

Control systems can provide the right amount of light in the right place at the

appropriate time while reducing the energy consumption of buildings (Arup & Arup,

2007). Furthermore, several studies have confirmed considerable savings from diverse

types of lighting control systems (Galasiu, Newsham, Suvagau, & Sander, 2007).

Lighting control systems have the following components (Tregenza & Loe, 2014):

lamp dimming or switching

light sensors

controllers of other devices, including blinds

occupancy sensors

timers

computer control.

There are several types of lighting control technologies. However, according to

IESNA and Rea (2000), the following two types of lighting controls are commonly

used:

continuous dimming controls: in this case, according to the sunlight, electric lights are brightened or dimmed to save energy and to maintain visual comfort of occupants by using photo-sensors, which are located at the station point

switching controls: in this case, lighting applications can be switched on and off manually, remotely via relays, by occupancy sensors, or switchable circuit breakers through control systems, or by light level switchable ballasts.

Osterhaus (2005) argues that a significant parameter in selecting a suitable

control strategy is the number of occupants in a room. He holds the view that the best

control system to use in individual offices is possibly manual switching control. The

vast majority of recently constructed commercial offices utilize some type of lighting

automation (Haq et al., 2014).

Tregenza and Wilson (2011) have categorised automatic lighting controls into

two forms:

open loop systems: these are usually used for small offices and are controlled by one or more photocells which are positioned either inside or outside the building to sense sunlight


closed loop systems: these are more suitable in larger areas with multiple orientations and are controlled by a photocell that is the room and is exposed to both daylight and electric light.

The expensive process of installing the systems, a lack of technical knowledge,

and technical management problems are the major development barriers to lighting

control systems (Aghemo, Blaso, & Pellegrino, 2014).

2.2.5 Summary

The main aim of any lighting design is to produce good lighting for residents

(Arup & Arup, 2007), which can be achieved through meeting their visual and

psychological needs while ensuring that energy is not wasted (Bean, 2012). This

chapter has reviewed the three critical aspects of indoor lighting, including daylighting

systems, electric lighting strategies, and control systems. It is also evident that office

buildings rely on vertical windows, especially in modern cities for daylight harvesting

(Huang et al., 2014). Accordingly, the primary aim of daylighting strategies in office

buildings is to improve the efficiency of vertical windows that can lead to indoor visual

comfort. As discussed earlier, the most frequently used daylighting systems in office

spaces are: (1) light shelves, and (2) louvres and blinds. According to Osterhaus

(2005), it is critical to ensure precise identification and evaluation of appropriate

shading and glare control devices for office spaces due to the complexity of the non-

uniform luminance distribution of shading devices.

It is a widely held view among researchers that good interior lighting can only

be reached by appropriate integration of energy-efficient electric lighting systems with

daylighting strategies (Vine, Lee, Clear, DiBartolomeo, & Selkowitz, 1998;

Kazanasmaz et al., 2009). The data reported in this chapter appear to support the

assumption that cool-white LED is one of the most highly efficient and

environmentally friendly electric lighting technologies. Moreover, as explained

in Chapter 2.1, there is a clear relationship between preferred perceived illuminance

and CCT, and the effect of indoor visual comfort. Therefore, this research will use

cool-white LEDs, which have similar CCT to sunlight , to integrate with

side-daylighting systems in office buildings to reduce the luminance contrast on the

window wall. Finally, as discussed in the previous section, utilising a suitable lighting

control system can positively influence the energy consumption of buildings.


2.2.6 Synthesis of elements from Chapter 2.1 and 2.2

Chapter 2.1 described the most important human parameters that can influence

perceived discomfort glare from daylight. Furthermore, according to Boyce (2003),

prolonged exposure to poor visual conditions can lead to eye muscle fatigue, irritated

eyes, stress, and headaches. However, as explained in Chapter 2.1, there is little

evidence to determine the influence of age, eyesight-related aspects, and human eye

adaptability on perceived discomfort glare from sunlight inside buildings.

Furthermore, according to Osterhaus (2005), there is little evidence to explain the

underlying principles and mechanisms responsible for the perceived discomfort glare

from windows. Thus, more investigation is needed to identify the numerous

parameters contributing to the perception of discomfort glare from windows and its

influence on the performance of office workers (Osterhaus, 2005).

Chapter 2.2 outlined some daylighting systems that can be used to enhance

indoor visual comfort. Nonetheless, there is a lack of guidelines about how daylighting

strategies can be used in a building design process to improve indoor visual quality, as

well as saving energy (Reinhart, LoVerso, Scartezzini, & Cuttle, 2010). Thus,

architects and building designers usually rely on daylighting rules of thumb, or their

own work experience for interior lighting designs (Reinhart et al., 2010). Although the

focus of this PhD research is on typical buildings and not green buildings, one study

of the latter reports that the parameters of a well day-lit space have not progressed yet,

the indicators of national and international rating systems to rate green buildings are

crude, and the market is not guided by practical information (Mardaljevic, Heschong,

& Lee, 2009). Thus, as discussed in Part 1, studies have illustrated that discomfort

glare from daylight is a common problem in office buildings.

The studies presented thus far provide evidence that the LC between the window

and surrounding immediate surfaces as the result of the high and variable brightness

of vertical windows can cause discomfort glare. Therefore, several innovative projects

and studies in lighting design practice have provided some guides for dealing with

glare from windows in office environments (Osterhaus, 2005). However, while it is

well known that, the LC on the window wall can cause discomfort glare and

consequently can result in attempts by occupants to intervene in lighting conditions to

enhance visual comfort, little research has been conducted into possible ways to reduce

this contrast. Thus, this study will focus on this knowledge gap and investigate how to


enhance window appearance as well as increasing energy savings in office buildings.

The next part will describe the strategies that can be used to evaluate lighting inside

buildings, followed by the framework of this research.

34 Part 3: Research Design

Research Design

This part starts by describing existing lighting evaluation procedures. The next

section will present the framework for this study. The fifth section outlines the

participants in the study, followed by the ethical considerations of the research and its

problems and limitations.

3.1 Lighting evaluation methods

The availability of daylight, which comprises direct and diffuse light from both

sunlight and skylight (Mardaljevic et al., 2009), is predominantly influenced by the

luminance levels and patterns of the sky (Mardaljevic, 2001). Thus, lighting evaluation

for daylight can be significantly more complicated than for electric lights (IESNA &

Rea, 2000), and the most efficient and accurate method of evaluating daylight

illuminance is long-term data measurement (Li, 2010). According to European

Standard EN 12464-1 (BSI, 2011), the most critical parameters to determine the

quality of luminous environment concerning daylight and electric lights are:

luminance distribution

illuminance

colour rendering and colour appearance of the light

glare

flicker

variability of light

the directionality of light, which allows objects in the interior space to be distinguished.

The current investigation focusses on exploring an innovative LED wall-

washing system to minimize glare from daylight that leads to reduced negative

interventions in lighting conditions. As was pointed out in the literature review,

assessing discomfort glare is only possible through examining the association between

the subjective responses and physical parameters. As such, the following section

describes the methods that can be used to evaluate individual responses for indoor

lighting.

Part 3: Research Design 35

3.1.1 Subjective assessment of perceived discomfort glare

Human impressions of discomfort glare are complex and can be affected by other

visual sensations, testing conditions, many psychological variables and individual

variations (Nazzal, 2005). For example, it has been suggested that occupants have

more tolerance to glare from the sky seen through windows than to glare from electric

lighting sources of comparable size (Chauvel, Collins, Dogniaux, & Longmore, 1982).

Human glare perception may also vary from season to season since people could have

a lower acceptance of the presence of daylight in the summer than in the winter

(Wienold & Christoffersen, 2006). Moreover, a study suggested that glare is more

often reported by older people inside buildings (Kent, Altomonte, Tregenza, & Wilson,

2014). Consequently, evaluating discomfort glare is only possible through using

subjective assessment, together with the physical factors (Shin et al., 2012a). However,

it is frequently challenging to find predictable, practical relationships between physical

stimulus and subjective reaction in the field of lighting (Houser & Tiller, 2003).

Psychophysics, which deals with the associations between sensory responses and

physical stimuli, is one of the most challenging branches of psychology (Houser &

Tiller, 2003). Even though there are several studies about the relationships between

perceived brightness and ambient illumination (Heinemann, 1961; Vidovszky-Németh

& Schanda, 2012), very little is known about how to establish practical predictive

associations between the results of these investigations to lighting design (Houser &

Tiller, 2003). Overall, different techniques are used to relate subjective responses to

physical parameters in lighting research, including questionnaires, semantic

differential scales, magnitude estimation strategies, and paired comparisons (Tifler &

Rea, 1992; Houser & Tiller, 2003).

3.1.2 Questionnaire

Kolb (2008) describes a questionnaire as a quantitative study method that

consists of some questions with predetermined response choices that participants

should choose. The questionnaire is one of the most practical methods of evaluating

light exposure in large-scale investigations (Bajaj, Rosner, Lockley, & Schernhammer,

2011). According to Bryman (2004), one of the most critical parts of any quantitative

study is to specify the research questions precisely. Oppenheim (2000) classifies

questions in a questionnaire as either open or closed (pre-coded). He also argues that

both types of question have advantages and disadvantages, and most surveys are a


mixture of the two. Table 3 illustrates the benefits and drawbacks of using open or pre-

coded questions (Oppenheim, 2000).

Table 3 Advantages and disadvantages of different types of questions

Question type

Advantages Disadvantages

Open Spontaneity and freedom of responseChance to probe

Beneficial for testing hypothesis about awareness and ideas

Time-consumingCostly of subjects’ time

Expensive and slow process of coding and sometimes unreliable

Demand more effort from subjects

Closed Need little timeEasy to process

Make comparison between groups easierBeneficial to test particular hypotheses

Less subject training

Loss of spontaneous answersBias in response categories

Occasionally too crudeMay irritate subjects

He argues that the answers to the questions should have the capability to be

converted into numbers in order to calculate statistics such as averages and percentages

during the analysis stage. Furthermore, according to Greasley and NetLibrary (2008),

the questionnaire should have three key features:

It should be brief, clear, and unambiguous.

The questions should be easy to complete (e.g., using checkboxes).

Open questions should be avoided by using alternative strategies (e.g., scaling to measure different dimensions of satisfaction).

Questionnaires can also be more efficient when the researchers already know the

precise research problem (Kolb, 2008).

3.1.3 Rating scales

According to Houser and Tiller (2003), paired comparison and semantic

differential (SD) scaling are two of the more widely used techniques in lighting

research. SD consists of a set of bipolar adjectives. The ends of each scale are defined

through polar opposite adjectives, which are often separated through a seven-point

scale (Monette et al., 2013). However, the number of points to the scale can be seven,

five, or even three (Barbara Sommer, 2006).

Stuart-Hamilton (2007) has defined paired comparison as a “psychological

measure, in which the participant compares every item in a set of stimuli with every

other item in the set (e.g., for brightness, aesthetic value, etc.)” (p. 192). The paired


comparison method is also called the two-alternative-forced-choice method (Bi, 2007).

This approach leads to sharper discriminations than are achievable through other

strategies due to asking participants to compare two items concerning the same

criterion (Böckenholt & Tsai, 2001). Furthermore, the paired comparison approach

can be used for two-sided, one-sided, preference, and intensity tests (Bi, 2007).

3.1.4 Magnitude estimation strategies

According to Stuart-Hamilton (2007), magnitude estimation is “a form of direct

scaling” (p. 161). This method can be used to achieve highly reliable judgments for

the brightness of light sources (Madden, 2008). In this strategy, a random value

(usually 100) is given to a stimulus, and then the subject is asked to rate other stimuli

relative to this value (Madden, 2006; Stuart-Hamilton, 2007). For instance, if a subject

perceives a light to be half as bright as the primary stimulus, he or she rates it as 50

(Stuart-Hamilton, 2007).

3.1.5 Physical lighting measurements

The following is a brief description of tools that can be used for physical light

measurements.

3.1.5.1 Light evaluation portable equipment

Light evaluation portable equipment comprises of light and light source colour

measuring instruments (Nlena & Associates, 2015). The first group is used to assess

luminance and illuminance, and in both cases, light is appraised according to the

photo-adapted visual response which ignores colour (Cuttle, 2008). The second group

is designed to evaluate the colour of light by using three or four filters with the spectral

sensitivity matched to the CIE Tristimulus colour matching functions (Nlena &

Associates, 2015). This group, known as colorimeters, can evaluate at least one of the

photometric measurements, including illuminance, luminance, luminance flux, or

luminance intensity (Nlena & Associates, 2015).

3.1.5.2 High dynamic range imaging techniques

With current high dynamic range (HDR) imaging techniques, the luminance

distribution of any space can be captured and appraised on a pixel-by-pixel basis

(Hirning et al., 2013). HDR imaging techniques can provide qualitative and

quantitative data for visual analysis (Inanici, 2006). These methods are quick and

inexpensive (Ng & Chung, 2011). They have a reasonable accuracy (error margin of


10%) to assess luminance distribution (Inanici, 2006). HDR methods also represent

realistic intensity levels with a larger distinction between the brighter and darker areas

of the registered image (Sarey Khanie, Wienold, & Andersen, 2014). Thus, several

studies have used subjective responses with captured HDR images to assess discomfort

glare from daylight (Inanici, 2006; Borisuit, Scartezzini, & Thanachareonkit, 2010;

Hirning et al., 2013; Sarey Khanie et al., 2014).

HDR images can be created through utilising a digital camera and a fisheye lens

(Suk & Schiler, 2013), which allows a wide FOV that is similar to that of the human

eye (Hirning, 2014). However, the human FOV extends about 135° in the vertical

plane and 200° in the horizontal plane, and the lens can capture 190° FOV in different

directions (Hirning et al., 2013). Thus, HDR images that contain FOV data in the

vertical plane exceed the FOV of the human eye (Hirning et al., 2013). Moreover, it is

necessary to combine multiple exposure pictures of the same scene with the right

software to create a single HDR image with relative luminance (Hirning et al., 2013).

Pictures created by non-HDR cameras have limited contrast and are known as low

dynamic range (LDR) images (Yun, Yoon, & Kim, 2014). These pictures may also be

processed by using Photosphere to create HDR images (Rodriguez & Pattini, 2012).

Photosphere is a photo-editing tool that allows HDR images to be easily made and

supports several formats of output images, including TIFF, JPEG, openEXR, and

Radiance RGBE (HDR) (Yun et al., 2014). HDR images can also be created from

simulation environment software, such as Radiance and Diva-for-Rhino (Yun et al.,

2014). Nonetheless, calibration, which is taken by portable measurement instruments,

is required while converting HDR image luminance values to the real luminance within

the scene to reach more accurate results (Ng & Chung, 2011).

3.2 Data collection methods

The quality of any quantitative study relies to a large extent on the reliability,

sensitivity, and validity of collected data (Roe & Webb, 1998). Reliability refers to the

consistency and purity of a measure that leads to the probability of obtaining the same

results again if the measure were to be duplicated (Oppenheim, 2000). On the other

hand, sensitivity refers to the accuracy of the instruments’ results when measuring

what they are supposed to measure (Roe & Webb, 1998). Finally, validity refers to

whether the variable measures what it is supposed to measure (Oppenheim, 2000).


According to Field (2014), there are two primary methods to collect data:

1. manipulating the independent data using different entities in each test

condition during experiments, such as an independent design that is also

called between-groups design, or between-subjects design

2. manipulating the independent data using the same entities in each test

condition during experiments, including within-subject design or repeated-

measure design.

He further states that there are always two types of variation in both the

independent design and the repeated-measures design, including (1) systematic

variation that occurs when the experimenter does something in one condition but not

in the other condition and (2) unsystematic variation that happens due to random

factors that exist between the test conditions (e.g., the time of day).

It is essential to try to minimise the unsystematic variation as much as possible

in both repeated-measures and independent designs (Field, 2014). The most significant

sources of systematic variation in a repeated-measures design are:

practice effects, in which participants may react differently in the

second condition due to familiarity with the test situation and the

measures being used

boredom results, in which participants may react differently in the

second condition due to being bored or tired from having finished the

first condition (Field, 2014).

Nonetheless, although it is impossible to eliminate these impacts, we can ensure

that they create no systematic variation among our conditions through determining in

which order the conditions can be completed (Field, 2014).

One of the most important attributes of any scientific study is randomisation

(Polit & Hungler, 1991). Randomisation eliminates most sources of systematic

variation, which helps us to ensure that any systematic variation among test conditions

is because of the manipulation of the independent variable (Field, 2014). However,

randomisation is one strategy to minimise the risk of systematic bias in experimental

conditions, as is using a high number of participants, because the risk can increase

when the number of participants decreases (Roe & Webb, 1998).


Most investigations involving discomfort glare have been done in experimental

test rooms using one of two primary data collection methods (Rodriquez & Pattini,

2014). One of these fundamental techniques is a method of adjustment through asking

participants to adjust the brightness of the background so that the source of glare is at

the borderline between discomfort and comfort (Luckiesh & Guth, 1949). Another

widely used technique is SD scaling in which the participant is shown one stimulus

and rates it on a particular scale (Houser & Tiller, 2003). It has been reported that SD

scaling is more reliable than adjustment and results in less variation across participants

(Jacobs, Bullimore, Bailey, & Berman, 1992).

3.4 Framework

As stated in the introduction of this document (Part 1), the primary research

question addressed in this thesis is:

Under what conditions does an LED wall-washing system best integrate with windows to reduce negative lighting interventions?

The following sub-questions should be addressed to answer this question.

Q1: Does the proposed LED wall-washing system reduce LC between the window and surrounding walls?

Q2: Does the proposed LED wall-washing system improve subjective appraisal of the window’s appearance?

Q3: What is the average acceptable LC on the windowed wall?

Q4: How do WWR and LED wall-washing with different power levelsaffect occupants’ intentions to intervene in lighting conditions?

A series of experiments was designed to answer the above questions. These took

place in rooms with different window-to-external-wall ratios and orientations. To

calculate the WWR, we multiplied the window size by 100, and then, divided by the

overall window wall size (window + surrounding walls). An LED wall-washing

system was used to increase the luminance of the areas immediately surrounding the

window to reduce the LC between window and its surrounding walls in these rooms

(and therefore improve window appearance). Increasing the luminaire power of the

proposed LED wall-washing system could also impact the illuminance level on top of

the desk. The series includes a pilot study in a real office space with a 15% WWR to

test the hypothesis. It also involves two tests to explore the impact of a proposed LED

wall-washing system on participants’ scale appraisal of the window’s appearance, as

well as on their negative lighting intervention in real office rooms with a 27% WWR


and a 45% WWR. Finally, the last experiment explores how participants might react

in the IVR office spaces in which we could change the WWR (15%, 30%, 46%, 62%

window-to-external-wall ratios) and the lighting level of the proposed LED wall-

washing system quickly and with low cost. Overall, the tests in the current study were

designed to investigate the effectiveness of the proposed LED wall-washing system on

reducing adverse lighting conditions in rooms with different window sizes and

orientations. Figure 5 outlines the framework for this research, and the association and

rationale of the stages are explained below.

Figure 5 Framework for the study

As stated in the introduction (Part 1), this PhD research hypothesizes that the use

of the proposed LED wall-washing system will improve visual comfort in office

buildings, and therefore, will diminish adverse interventions in lighting conditions by

occupants. The primary aim of conducting a pilot study in this research was to test the

hypothesis of this investigation quickly, as well as improving the questionnaire and

the method before commencing Experiment 1. The pilot study was carried out in a real

office room facing southwest at Gardens Point campus of Queensland University of

Technology (QUT) in Brisbane, Australia. In order to explore the proposed LED wall-

washing system in rooms with different window-to-external-wall ratios, the blind in

the room in which the pilot study was conducted was kept partly down during all

lighting scenarios (see Figure 6). Therefore, while the actual WWR in this room is


27%, the WWR during the pilot study can be considered to be 15%. A questionnaire

was developed based on previous research to be used in the pilot study (see Appendix

B). Each participant was exposed to four default lighting conditions during each test

based on the luminaire power level of the LED linear luminaires. The participants

responded verbally to the questions in the questionnaire during each lighting scenario.

Physical lighting measures were also collected during each lighting condition using a

Nikon Coolpix 4800 digital camera with a fisheye lens, a Konica Minolta LS100

luminance meter, and a Topcon IM-3 illuminance meter.

Figure 6 Images of the test office rooms during the pilot study and the first two experiments when the proposed LED wall-washing system is on

Experiment 1 investigated the influence of the proposed LED wall-washing

system on the LC between the window and surrounding walls, as well as on subjective

ratings for visual comfort. This study also explored the impact of this system on

occupants’ intentions to intervene in lighting conditions. Taken together, Experiment

1 was designed to address sub-questions 1 and 2. It was conducted in the same test

office room in which the pilot study was carried out. However, to test the hypothesis

in a place with a different WWR to compare with the room in which the pilot study

was conducted, we kept the blind fully open during all lighting scenarios in Experiment

1. This allowed us to carry out Experiment 1 in an office room with a 27% WWR and

more daylight penetration to compare with the pilot study. Therefore, we could explore

the impact of WWR on the effectiveness of the proposed LED wall-washing system

to enhance window appearance as well as reducing negative lighting interventions.

Finally, while the questionnaire was improved based on the outcomes of the pilot study

before commencing Experiment 1, the method was similar.

Experiment 2 was designed to answer sub-questions 1 and 2, like Experiment 1.

It also aimed to investigate acceptable LC on the windowed wall (sub-question 3)


through giving participants the opportunity to set the LC between the window and

surrounding walls to their acceptable level using the proposed LED wall-washing

system. While the pilot study and Experiment 1 were carried out in a real office room

facing southwest with a 15% WWR and a 27% WWR respectively, Experiment 2 was

conducted in a typical real office room with different orientation and window size, in

a room facing northwest with a 45% WWR, that is located at Gardens Point campus

of QUT in Brisbane, Australia. Experimenting in this room allowed us to test the

hypothesis of this PhD research in a space in which we could have direct sunlight,

much more daylight penetration, and the sky view in the FOV of participants. To

simulate day-to-day working in the real office spaces, a diagonal arrangement of the

workstation was chosen, and participants were asked to work with a provided laptop

while responding to the questions in the questionnaire verbally. We enhanced and

developed the questionnaire that was used in Experiment 1 before commencing

Experiment 2. The method was similar to Experiment 1; however, the lighting

scenarios were changed randomly during Experiment 2, and the participants were also

given the opportunity to adjust the luminaire power of the proposed LED wall-washing

system themselves.

To adequately understand how the proposed LED wall-washing system impacts

participants’ intention to intervene in lighting conditions in rooms with different

window sizes, it is vital to study such scenarios in real (physical) office spaces. In

previous experiments, we tested the hypothesis of the current PhD research in two

rooms with a 15% WWR, a 27% WWR, and a 45% WWR facing southwest and

northwest. As was pointed in the literature review (Part 2), while it is possible to

perform such experiments in existing buildings, several factors might influence the

results (e.g., WWR, the reflectance of inner surfaces, different interior space designs,

cloudy/sunny weather on different days, different outside views, different internal

brightness, etc.). These factors, which in some cases are not possible to control, could

cause experimental noise or affect the outcomes. Accordingly, during the last

experiment, we used IVR technology that allows the experimenter to control for most

(if not all) potentially confounding features and isolate the variables of interest (i.e.,

lighting scenarios). This technology allowed us to change the WWR and the luminaire

power level of the proposed LED wall-washing system quickly and with low cost. The

technology also enables the experimenter to provide spaces where users can be fully


immersed and feel a parallel sense of presence in physical environments (Zhao, 2003;

Brooks et al., 2014).

While the number of research on immersive IVR environment has increased over

the last two decades, there are very limited studies in the context of architecture (Paes,

Arantes, & Irizarry, 2017). There is also just a few research about lighting using IVR

technology (Heydarian, Carneiro, Gerber, & Becerik-Gerber, 2015; Heydarian et al.,

2016; Heydarian, Pantazis, Wang, Gerber, & Becerik-Gerber, 2017). These studies

have mainly focused on exploring human behaviour under different lighting

conditions. According to Heydarian et al. (2015) human behaviour in an IVR space is

not significantly different from that in a real environment. Heydarian et al. (2015)

investigated the impact of personal control on manual and semi-automatic lighting

options to enhance lighting in an office space using either electric lights or daylight in

an IVR single office room. This study illustrates that participants were more likely to

use daylight to enhance indoor visual comfort specifically when they had access to the

remote control to change the position of the shading device. Heydarian et al. (2016)

studied the impact of default lighting settings on participants’ rate of lighting

adjustment in an office room. The results of this study indicate that participants were

significantly more probable to keep the default lighting setting if they had daylight

available. Another study using IVR technology as an experimental tool demonstrates

that people preferred to have maximum simulated daylighting compared to electric

lighting (Heydarian et al., 2017).

In previous research, rigorous lighting evaluations using Rhinoceros and

Grasshopper have been conducted to make sure that the created IVR environments

represent the actual lighting conditions (Heydarian et al., 2015; Heydarian et al., 2016;

Heydarian et al., 2017). However, according to Natephra, Motamedi, Fukuda, and

Yabuki (2017), there are still some issues in representing realistic lighting scenarios

using IVR technology due to the current technical limitations. They argue that while

IVR technology can provide semi-realistic lighting environment, it is impossible to

create the precise perception of illuminance and glare using HMD with the current

technology. For example, to be able to provide realistic lighting condition in the IVR

environment, the screen resolution of the IVR headset should be around 6000 pixels

horizontally and 8400 pixels vertically, which is not possible with the current

technology (Fuchs, 2017). Natephra et al. (2017) also highlight that it is impossible to


analyse the appearance of lighting design, the distribution of lighting, and

quantification of the amount of lighting in real-time using IVR technology. Studies

have also demonstrated significant differences between distance judgments in the IVR

and the real conditions (Gooch & Willemsen, 2002); virtual dimensions appear

considerably smaller than absolute dimensions (Thompson et al., 2004). Nonetheless,

Paes et al. (2017) hold the view that people can estimate the distance relatively accurate

in a single IVR environment when their reference is in that IVR space only. Even

though perception in the IVR spaces is not like the real environment, as mentioned

earlier there is no significant difference between human behaviour in the IVR

environment and the real spaces. Considering that the main aim of our previous

experiments were to reduce occupants’ interventions in lighting conditions, IVR

technology is an appropriate method to investigate human lighting behaviour in a room

under different lighting scenarios.

The last experiment aimed to examine the impact of the WWR on the efficiency

of the proposed electric wall-washing system to reduce adverse lighting interventions.

This study was conducted in a controlled IVR office room in which we could change

the WWR and the luminaire power level of this system quickly and with low cost. The

final experiment was designed to address sub-questions 2 and 3. The questionnaire in

the last experiment was developed based on the questionnaires used in previous

experiments in the current PhD research. As with previous tests, participants were

exposed to four different lighting conditions based on the luminaire power level of the

proposed LED wall-washing system in each IVR office room with a specific WWR

(15%, 30%, 46%, and 62% window-to-external-wall ratios (see Figure 7)). Overall,

participants were exposed to sixteen default lighting conditions randomly. They were

also given the opportunity to set the LC on the window wall to their preferred and

acceptable level using the proposed LED wall-washing system in the IVR office rooms

with different window sizes.


Figure 7 Examples of the IVR office rooms with different window-to-external-wall ratios when the proposed LED wall-washing system is on

3.5 Participants

According to Aschehoug et al. (2000), about 30 participants, who can be grouped

by age, education, and sex, are needed to compare indices between them and to achieve

statistically significant results. However, Roe and Webb (1998) hold the view that the

risk of systematic bias in experimental conditions can increase when the number of

participants decreases. Thus, the main experiments in this research were conducted

with at least 35 participants to compare indices between them. The following is a brief

description of the target group for this research experiment and the related ethics

parameters.

All the experiments for this research sought participants among professional

staff, academic staff, undergraduate students, postgraduate students, and other

employees, both males and females, 18 years old and over, working at QUT in

Brisbane, Australia. Participants needed to have normal or corrected to normal vision

(corrective lenses or contact glasses). All comments and responses have been treated

confidentially unless required by law. The names of participants were not required in

any of the replies. Any data collected as part of this research have been stored securely

as per QUT’s policy regarding management of research data.


Participation in the pilot study and Experiment 1 took approximately 30 minutes

of participants’ time. Participants also had to spend about 45 minutes taking part in

experiments 2 and 3. It was expected that there might be some minor experience of

glare or uncomfortable conditions for the participants while conducting the pilot study,

and the first and second experiments (regarding both intensity and duration of

exposure). However, the risk of discomfort glare was minimised by reducing the length

of exposure to each lighting condition (exposure times less than two minutes for

comfortable conditions and less than one minute for glare or uncomfortable scenarios).

It was also not expected that these experiments would have any undesirable

consequence (discomfort or inconvenience) outside our studies.

It was expected that participants might experience inconvenience or discomfort

glare during the last experiment in the IVR office room. There was also the potential

for the minor experience of eyestrain, as well as motion sickness. However, the

likelihood of these consequences was low and the severity was small.

We tried to minimise risks by:

using lighting scenarios that do not require any movement on behalf of the user,

incorporating breaks into the experiment between viewing different lighting scenarios, and

informing the participants that they could opt to stop participating altogether in the unlikely event that the discomfort persisted.

Moreover, the risk of discomfort glare was minimised through:

allowing participants to self-pace their responses to each lighting scenario,

including a 5-second break between each condition, where participants could shut their eyes to rest,

ensuring that the device used to show the office environment to participants had inbuilt limits to ensure that participants were not exposed to uncomfortable conditions (i.e., it was limited regarding both intensity and duration of exposure),

designing the experiment so that the virtual reality viewing time wasestimated to be under 20 minutes,

piloting the lighting scenarios with five separate individuals to ensure that there was no discomfort glare.

All participants were informed of the details of the test before conducting each

experiment. The light measurement equipment (in the pilot study, and Experiments 1


and 2) or virtual reality equipment (in Experiment 3) were shown to participants, along

with an explanation of their functionality. Participants were also informed and

reminded that they could decide not to participate at any point before or during the

experiment and that in this case they would immediately be removed from our sample.

However, as the questionnaire was anonymous once it had been stored and saved,

participants could not withdraw after that stage.

3.6 Ethics and Limitations

As mentioned in the previous section, the tests in the current PhD research were

conducted with human participants. Accordingly, it was vital to apply for ethics

approval before commencing the tests. These tests have low-risk classifications, and

the following guidelines and procedures are referred to:

D/6.5 University Human Research Ethics Committee

D/2.6 QUT Code of Conduct for Research

National Statement on Ethical Conduct in Human Research 2007-updated 2014

Australian Code for the Responsible Conduct of Research

The research design of the current PhD investigation includes a pilot study and

two experiments in real office rooms and a study in IVR office spaces. While the

method was quite similar during the tests in the real office environments, we used a

different approach to carry out the last experiment in the IVR office rooms. Therefore,

it was necessary to finalize two ethics clearance applications before commencing the

tests as follows:

1. Completing the first ethics clearance application (number 1500000250)

before starting data collection in the real office rooms, including the pilot

study, Experiment 1, and Experiment 2.

2. Finalizing the second ethics clearance application (number 1700000181)

before commencing data collection for the final experiment in the IVR

office spaces.

3.7 Summary

This part of the thesis began by describing theories of lighting evaluation. It went

on to suggest what kinds of methods can be used to collect data for evaluating lighting

inside buildings. It then described some strategies to analyse collected quantitative


data. The analysis reported in this part appear to support the idea of using closed

questions to make the process of experiments easier. Furthermore, as the tests were

designed to be conducted in office rooms with installed supplementary LED wall-

washing systems, paired comparison methods and magnitude estimation strategies are

not applicable to assess subjective responses. Thus, the current research used the SD

scaling approach to conduct individual assessments. Quantitative data were collected

using a Nikon Coolpix 8400 digital camera, Konica Minolta LS100 luminance meter,

and Konica Minolta T-10 illuminance meter during the pilot study, Experiment 1, and

Experiment 2. The repeated-measure design method, in which one set of participants

is tested more than once, and their scores compared, was used to collect data during

all experiments. The pilot study, Experiment 1, and Experiment 2 were conducted at a

particular time of day to minimise unsystematic variations. The test conditions in

Experiments 2 and 3 were conducted in a random order to reduce any sources of

systematic variation. Finally, quantitative data analysis has been carried out using

SPSS after conducting the experiments.

This part of the thesis has focused on existing methods to evaluate indoor

lighting, as well as the framework of the current PhD research. The following part will

describe each experiment in four chapters.

50 Part 4: Published and Submitted Papers

Published and Submitted Papers

As outlined in Part 3, Section 3.4, this PhD research involves a pilot study and

two experiments that were conducted in real typical office rooms with different

orientations and window sizes. It also involves an experiment that was carried out in

IVR office rooms with different window-to-external-wall ratios. The outcomes of the

tests in real office spaces have been published in two conference proceedings and the

journal Energies. A paper discussing the results of the final experiment in the IVR

office rooms has also been submitted to the journal Building and Environment and is

under review.

This part presents four chapters, the titles of which correspond to the three

published papers and one submitted paper (see Figure 8). Each chapter begins with a

connecting summary to demonstrate that the papers form a coherent, linked study.

Each chapter then provides a statement of authorship that explains the contribution of

each author to the paper, as well as the details of publication. Following that, the

research paper is presented verbatim.

Figure 8 Research map of Part 4

Part 4: Published and Submitted Papers 51

Chapter 4.1 LED Lighting Design Strategies to Enhance Window Appearance and Increase Energy Savings in Day-lit Office

Spaces

This conference paper examines the hypothesis of this PhD thesis through a pilot

study with fifteen participants. It investigates the influence of the proposed LED wall-

washing system on reducing the LC between window and surroundings as well as

diminishing occupants’ lighting interventions. This paper also demonstrates the results

of modelling the energy consumption of the proposed LED wall-washing system in

the same office room over the course of a year, comparing different lighting designs

and user behaviour scenarios to indicate the effectiveness of this system. It suggests

that potential energy savings are offered by the proposed LED wall-washing system

by diminishing negative user interventions.

The tests were conducted in a typical office room facing southwest with a 15%

WWR. While the actual WWR in this office space is about 27%, we kept the blind

partly down during the tests of the pilot study to achieve approximately 15% WWR

(see Figure 6). The luminaire power of the proposed LED wall-washing system was

increased in four default lighting scenarios: no LED wall-washing system, and the

LED wall-washing system with a low, medium, and high power level. A questionnaire

that was developed based on the summary in Part 3 was used to collect participants’

responses about perceived discomfort glare from the windowed wall, as well as their

intention to change the lighting conditions (see Appendix B). Lighting measurements

were collected using a luminance metre, illuminance metre, and a digital camera with

a fisheye lens during each default lighting condition. The outcomes of these tests were

also used to assess under what circumstances the increased annual electricity

consumed by the proposed LED wall-washing system can be offset. Overall, there are

three significant reasons why this paper should be included in the body of work that

makes up this thesis by publication:

The findings of this paper suggest that using proposed LED wall-washing system could efficiently reduce LC between the window and surrounding walls from approximately 215:1 to 26:1 in an office room facing southwest with a 15% WWR.

This study suggests that such a reduction of LC on the window wall could reduce participants’ propensity to intervene in lighting conditions, as well as to enhance indoor lighting satisfaction.


The results of this research paper report that increased electricity consumption of the proposed LED wall-washing system with low power level (approximately 18 W) is offset where there is roughly a one-quarter reduction in users’ negative lighting interventions.

However, the statistical analysis in this study was limited to a descriptive level.

The main value of this research (Chapter 4.1) was to settle the starting point of this

PhD research, and to serve as data exploring, researcher training and method testing

stage. Furthermore, the experience from tests conducted during this study as well as

the outcomes of this paper helped us to improve the questionnaire before commencing

Experiment 1 (Chapter 4.2).


4.1.1 Statement of Contribution of Co-Authors for Thesis by Published Paper

The authors listed below have certified that:

They meet the criteria for authorship in that they have participated in the conception, execution, or interpretation, or at least that part of the publication in their field of expertise;

They take public responsibility for their part of the publication, except for the responsible author who accepts overall responsibility for the publication;

There are no other authors of the publication according to these criteria;

Potential conflicts of interest have been disclosed to (a) granting bodies, (b) the editor or publisher of journals or other publications, and (c) the head of the responsible academic unit, and

They agree to the use of the publication in the student’s thesis and its publication on the QUT’s ePrints site consistent with any limitations set by publisher requirements.

Publication title: LED Lighting Design Strategies to Enhance Window

Appearance and Increase Energy Savings in Day-lit Office Spaces. Paper presented at

CIE Lighting Quality & Energy Efficiency Conference in Melbourne, Australia on

May 27th, 2015.

Contributor Statement of contributionMehdi Amirkhani Conducted literature review. Designed and implemented the

experiments. Performed data analysis and drew initial conclusions from data analysis. Wrote the first draft of the paper. Performed subsequent editing and corresponding author for the paper.

Signature:

Date: 14/06/2018

Dr Veronica Garcia-Hansen Contributed to experimental design and data analysis. Reviewed research paper and helped with the editorial process.

Dr Gillian Isoardi Contributed to experimental design and data analysis. Reviewed research paper and helped with the editorial process.

Principle Supervisor ConfirmationI have sighted email or other correspondence from all Co-authors confirming their certifying authorship. Dr Veronica Garcia-Hansen Signature

_

Date

____14/06/2018_____




4.1.2 Abstract

Vertical windows are the most common and simplest method to introduce

daylight to interior spaces of office buildings, while also providing a view and

connection to the outside. However, high contrast ratios between windows and

surrounding surfaces can cause visual discomfort for occupants and can negatively

influence their health and productivity. Consequently, building occupants may try to

adapt their working environment through closing blinds and turning on lights in order

to improve indoor visual comfort. Such interventions defeat the purpose of daylight

harvesting systems and can increase the forecast electric lighting consumption in

buildings that include such systems. A simple strategy to prevent these problematic

consequences is to reduce the luminance contrasts presented by the window wall by

increasing the luminance of areas surrounding the window through the sparing use of

energy-efficient supplementary lighting, such light emitting diodes (LEDs). This paper

presents the result of a pilot study in typical office in Brisbane, Australia that tests the

effectiveness of a supplementary LED lighting system. The study shows an

improvement in the appraisal of the visual environment is achieved using the

supplementary system, along with up to 88% reductions in luminance contrast at the

window wall. Also observed is a 36% reduction in the likelihood of user interventions

that would increase energy usage. These results are used as the basis of an annual

energy simulation of the test office and indicate that supplementary systems could be

used to save energy beyond what is typically realised in side lit office spaces.

4.1.3 Introduction

Energy use for lighting alone in office buildings is roughly 20-40% of the total

energy consumed in buildings (Freewan, 2014). Office buildings often rely on side

daylighting systems for daylight harvesting through windows, especially in high-rise

cities. It is also well understood that office workers generally prefer to have windows

in their working environment that can provide both natural light and outside view

(Veitch, Hine, & Gifford, 1993). Nonetheless, luminance contrast between the bright

surface of the window and surrounding areas may cause visual discomfort and reduced

visibility for occupants. Thus, occupants may choose to close blinds and turn on the

lights to enhance indoor visual comfort. Escuyer and Fontoynont (2001) asserted that

although it is likely that occupants close blinds, it is quite improbable that they raise

the blinds again when there is no glare or overheating problem especially when they


have poor outdoor view. A study among 123 buildings with installed photosensor-

control systems illustrated that there is a relatively constant relationship between the

amount of illuminance from windows and turning on the lights by occupants in

particular when dimming control systems work perfectly (Heschong et al., 2005). This

study showed that as the window illuminance increases, the likelihood of turning on

the lights will also increase to up to 60% to diminish luminance contrast between the

window and surrounding areas. The impact of human intervention on lighting

conditions can reduce energy savings; the largest field study on the effectiveness of

side-lighting controls for daylighting showed that less than 25% of the predicted

(modelled) energy savings arising from daylight harvesting systems were actually

being realised in practice (Heschong et al., 2005).

In order to reduce these problematic issues, we propose the use of an electric

lighting design that uses small amounts of supplementary light to reduce the luminance

contrast between the bright surface of the window and its surrounding areas. The

purpose of this supplementary lighting design is to diminish occupant intervention that

might reduce the expected energy savings from photosensor controls. This paper

reports on preliminary results of a pilot study and a simulation analysis. The

preliminary pilot study evaluates occupant response to a sample supplementary

lighting design in a conventional office room in Brisbane, Australia. This paper also

demonstrates the results of modelling the lighting energy use in the same office room

over the course of a year, comparing different lighting design and user behaviour

scenarios to demonstrate the effectiveness of the supplementary lighting design

strategy. The goal of this study is to demonstrate the potential energy savings offered

by supplementary lighting strategies, and the opportunities they may present to

increase the effectiveness of photosensor-control dimming systems by diminishing

negative user interventions.

4.1.4 Small pilot study

It is predicted that the efficacy of LEDs will approach 250 lm/W by 2020, which

may lead to a reduction in global lighting electricity consumption by more than 50%

and total electric use by around 20% (De Almeida et al., 2014). Cool-white LEDs,

which have a high correlated colour temperature (CCT > 5000 K) and can achieve

efficacies greater than 100 lm/W, are also a good colour match for natural light under

clear sky condition. This made them a desirable choice for the supplementary lighting


system used in this study. The preliminary pilot study was conducted in a small office

room, with a southwest oriented window, under clear sky conditions in autumn/winter

for Brisbane, Australia. The office was furnished with a desk and had two separately-

switched, suspended linear fluorescent luminaires overhead (28 W each). Room

surfaces were mostly light, with white ceiling, walls, window frame and desk. Figure

9 shows the view of the window wall in the office from the seated desk position.

Figure 9 Test office room at QUT in Brisbane, Australia

A cool white LED lighting system was assembled around the bottom and sides

of the window to achieve a wall-washing effect. The system selected for this study was

an off-the-shelf solution chosen for its ease of use, diffuse output and modularity.

When switched on at maximum power the 6 x 9 W modules consumed 54 W

(according to the manufacturer, providing 57 lm/W), which was relatively high for the

output required for the space. However, the system included a dimmer switch, which

allowed lower levels to be achieved and tested.

A total of 15 participants were asked individually to sit in front of the window

and to imagine this test office room as their own working environment. They then

responded to a simple lighting appraisal survey designed to determine whether the use

of the supplementary lighting system would influence their decision to switch on lights

or close blinds in the room. Each participant was seated at the desk with the overhead

lights off and the blinds open. Four lighting conditions were presented in turn,

beginning with no supplementary wall lighting, and then LED wall-washing of the

window surrounds at 3 different power levels (9 W, 18 W and 27 W). The participant

was asked to rate the visual amenity of the space on a scale of 1-5 (one meaning very

dissatisfied and five meaning very satisfied) and then asked if they wanted to switch

on the lights or close the blinds (yes/no answer). Given that the 2 overhead lights were


separately switched, if the participant answered yes to switching on lights, they were

further asked how many they would like to switch on (i.e. one or two).

The four lighting conditions (stages) according to the total lighting power of

LEDs were: off (stage 1), luminaire power of 9 W (stage 2), luminaire power of 18 W

(stage 3), and luminaire power of 27 W (stage 4). Luminance measurements of key

surfaces in the room were collected at the beginning of each stage using a Konica

Minolta LS100 luminance meter and illuminance measured at the desk with a Konica

Minolta T10. Figure 9 shows a picture of the test office room with overhead lights on,

supplementary system off and the 8 areas that were targeted to measure the luminance

during each stage, as well as the illuminance meter on top of the desk. Table 4

illustrates the average illuminance on top of the desk in the test office room throughout

each stage when overhead lights are off. All values are averaged over all participant

measurements, as there was very little variation in the exterior lighting conditions

across all tests. This was evident in the luminances and illuminances measured in the

space varying by less than 10% across all 15 tests. A key measurement of interest to

this study is the change in the luminance ratio from the window to the window wall at

each stage. To obtain the value of this ratio, readings 1, 2 and 3 are averaged (to give

window luminance) and compared to the average of readings 4, 5 and 6 (for the

surrounding wall luminance).

Stage Average illuminance (lux)1 1452 1513 1414 155

Table 4 Average horizontal illuminance measurements at the desk during each stage of the survey

The survey results shown in Figure 10 indicate that the supplementary lighting

system did affect the decisions the participants made as to whether to switch on or off

the lights, or to close blinds. This study demonstrated that the probability of

participants wanting to turn on the lights or close the blinds in the office has been

reduced by 24% at stage 3 and 36% at stage 4 of the survey compared to stage 1. It

also showed that subjects’ satisfaction increased by about 27% and 36% at the

beginning of stage 3 and 4 respectively. It is believed that the changes to the desired

switching and blind positioning preferences reported by the participants in this study

is attributable to the substantial decrease in the window to wall luminance ratio (from


215 to 26, shown in Table 5), given that the illuminance on the desk remained

consistent across all stages of the study (Table 4). Also of interest is that the

illuminance recorded at the desk (~140 lux) is lower than most photosensor control

systems would use as a setpoint for daylight dimming; however, it appears that the

participants found this an acceptable level to begin reducing overhead lighting where

the luminance contrast was also acceptable.

Stage Average window luminance (cd/m2)

Average wall luminance (cd/m2)

Luminance ratio

1 2403 11 2152 2409 25 953 2257 50 454 2281 86 26

Table 5 Average luminance ratio between window and surrounding areas during each stage

Figure 10 Survey results

4.1.5 Simulation method

A model of the test office room in preliminary study was created in ECOTECT

which is 3.64 m (deep) * 3.17 m (wide) * 3.85 m (high). Window dimensions are 1.235

m * 1.1 m while its sill height is 1 m. The interior reflection of walls, ceiling and floor

are 50%, 70% and 10% respectively. One sensor point was also placed at the desk

height (0.72 m of the floor horizontal) with a 1.5 m distance from the middle of the

window, to correspond to the location of the illuminance meter in the test room.

The DAYSIM engine within ECOTECT was used to analyse the annual electric

energy usage of the model. Five minutes’ time steps for the annual daylight simulation

was chosen. It was assumed that this office will be occupied Monday to Friday from

8am and 5pm, for 48 weeks per year. The lunch time, intermediate break, and daylight


saving time were disabled for lighting analysis. The power density (W/m2) was

calculated dividing total overhead lighting power in the preliminary study (56 W) by

the floor area (11.54 m2). Thus, the total power density of the overhead fluorescent

lighting in this model and a proposed 18 W LED system are 4.85 W/m2 and 1.56 W/m2

respectively. It was also assumed that dimming system has an ideally commissioned

photosensor-control with a ballast loss factor of 20 percent which works perfectly in

this model. Finally, shading device system was assumed to be static in the simulated

model.

Five different cases are presented based on the lighting energy modelling of the

test office over a year. The base case (1) has no photosensor control system. The next

case in terms of user behaviour assumes ideal use – the daylight photosensor control

system is always used as designed – so savings are achieved as sun and sky conditions

permit. However this ideal case (2) has been demonstrated to make an unrealistic

reference case in side-lit spaces, with a realistic value closer to 25% of those savings

(Heschong et al., 2005). So as well as the ‘ideal’ case, the ‘actual’ case (3) will consider

this fraction of the ideal energy savings. Finally, two more cases are presented

considering use of a supplementary lighting system, one conservative (case 4) and one

optimistic scenario (case 5). The minimum illuminance level required at the work-

plane was 320 lux for cases 1 to 4 and 200 lux for case 5. This ‘optimistic’ illuminance

level was based on an observed reduction in acceptable work-plane illuminance

reported by participants in the study.

Annual lighting energy use (in kWh/m2) in cases 1 and 2 is calculated by

ECOTECT based on an installed lighting power density of 4.85 W/m2 and work-plane

illumance of 320 lux. The energy uses in case 3 is calculated simply by assuming only

25% of the energy saved in case 2. Case 3 applied the results observed in the case

study to modify the savings found in cases 2 and 3: a 24% reduction in user

interventions is expected at the cost of an increase of 1.56 W/m2 while dimming

occurs. This is calculated by finding the number of hours that dimming occurs in case

2 (approximately 1190 h), increasing the power density used in case 2 by 1.56 W/m2

for that period of time, and then assuming that changes to user interventions increases

the actual energy savings from 25% to 43% of the modelled value. Case 5 is calculated

again in ECOTECT using an increased overall power density (4.85 + 1.56 W/m2) but

a lower workplace illuminance of 200 lux.


Figure 11 Annual electric use of the model in different cases

4.1.6 Results and discussion

The results of this study indicate that a supplementary lighting system can be

used to significantly reduce the luminance contrast on the window wall of a side lit

space. In this example, the luminance contrast between the window and its surrounding

wall was reduced from an average value of approximately 215 to 26 – a reduction on

88%, or nearly 9-fold. A more modest reduction of 80% was shown to reduce the

likelihood of participants wanting to switch on ceiling lights or close blinds by 24%.

Using this result, annual energy simulations were conducted to show that the increased

power consumed by the supplementary system is offset where a suitable decrease in

negative user intervention is achieved. With no photosensor controls, the test office

was expected to use 10.5 kWh/m2 in lighting energy. Using a realistic scenario where

25% of modelled lighting energy savings are achieved, the test office used

approximately 9.5 kWh/m2. If the supplementary system were to reduce negative user

intervention by 24%, then the annual lighting energy use would also be 9.5 kWh/m2.

The energy saving in this example is not evident; however, it is very important to note

that the LED system employed in this study was not selected for its energy-efficiency,

but rather as an easy method to implement solution at the test level. More efficient

luminaire configurations will be applied to future stages of this research. Also the

luminous efficacy of this system was only 57 lm/W. Both the design of the system,

and the system efficacy could be increased in practice to substantially boost the energy

savings available for this design strategy. Further savings using this supplementary

lighting strategy may lie in the optimistic case scenario. Participants indicated

comfortable working conditions could be achieved at workplane illuminances of

approximately 140 lux when the luminance ratio at the window wall was reduced by


supplementary lighting. This suggests that a decrease in workplane illuminance may

be acceptable where the window to wall contrast was maintained at a relatively low

level in this type of environment. Evidently, this scenario requires further and more

thorough investigation; however, the preliminary modelling shown here for case 5

shows that this is where more substantial energy savings may lie. Further study in this

area will focus on more rigorous testing of occupant perception using supplementary

lighting systems, incorporating a variety of test spaces, window types and office

layouts (including open plan spaces).


Chapter 4.2 Improving the impact of Luminance Contrast on Window Appearance in a Conventional Office Room: Using

Supplementary Lighting Strategies

As was pointed out in Part 3, Section 3.4, this PhD research aims to explore the

impact of the proposed LED wall-washing system on participants’ scale appraisal of

the window appearance, as well as on their propensity to change the lighting

conditions. As part of this investigation, the pilot study (Chapter 4.1) was conducted

in a real typical office room facing southwest with around 15% WWR. However, it

was limited to only 15 participants, and the room had insufficient daylight penetration.

To address the aims of this PhD investigating conditions in which the LED wall-

washing would be useful, we needed to examine the hypothesis in a room with greater

daylight penetration. Therefore, this paper builds on the results of the pilot study by

increasing the WWR to have more daylight, as well as increasing the number of

participants to 35 people. Furthermore, the first question at the beginning of each

lighting condition during the pilot study asked participants to rate window appearance

instead of asking to rate perceived discomfort glare from the window wall. Using the

former question led to some ambiguity about the accuracy of participants’

interpretation of the window appearance. Accordingly, we revised the questionnaire

and asked participants to rate their discomfort glare perception from the window wall,

choosing one of the four options: imperceptible, perceptible, disturbing, and

intolerable (see Appendix C). We also asked participants to rate their indoor lighting

level satisfaction on the scale of 1-5, where one was very dissatisfied, and five was

very satisfied during each lighting scenario.

While Experiment 1 was carried out in the same test office room as the pilot

study, we changed the WWR in the room through keeping the blind fully open during

all lighting conditions (see Figure 6). This method allowed us to conduct Experiment

1 in an office space with about 27% WWR, and therefore, with more daylight

penetration than the room in the pilot study. Iwata and Tokura (1997) illustrated that

individuals have more sensitivity to glare when it is positioned below the line of vision

than when it is located above the line of vision. Additionally, the line of participants’

vision in Experiment 1 was about 1.15-1.25 m from the floor while they were sitting

in front of the window, and the participants’ distance from the window wall was about

1.5 m during the test conditions (see Figure 12). Acc, the upper level of the window


had minimum impact on their ratings for the window appearance. Therefore, we

performed the luminance spot measurements on the lower part of the window, where

we installed the proposed LED wall-washing system. Overall, Experiment 1 aims to

address the sub-questions below to answer the primary question of this PhD

investigation.

Q1: Does the proposed LED wall-washing system reduce the LC betweenthe window and surrounding walls?

Q2: Does the proposed LED wall-washing system improve subjective scale appraisal of the window appearance?

A similar method to the pilot study was used to collect data. The results of

Experiment 1 indicate that the proposed LED wall-washing system with low power

level (around 18 W) could significantly reduce the LC on the window wall from

approximately 117:1 to 33:1. The results suggest that such an LC reduction between

the window and surroundings enhanced participants’ scale appraisal of the window

appearance. It also reduced participants’ intentions to switch on the ceiling lights by

around 27% and to move the blind down by more than 90%. However, the order of the

lighting conditions presented to participants in Experiment 1 was not randomised,

which could have introduced order and fatigue effects that explain the results.

Furthermore, using spot luminance measurements instead of luminance mapping in

the pilot study (Chapter 4.1) and Experiment 1 (Chapter 4.2) results in a poorer

photometric characterization of the window surface and its surrounding walls.

Therefore, Experiment 2 (Chapter 4.3) was carried out with a random ordering of

conditions and the luminance mapping technique was used to address these issues.

Using inferential statistical analysis provides the opportunity to describe and make

inferences about whole the population when it is not possible to examine each member

of that entire population (Ali & Bhaskar, 2016). Accordingly, while we did not use

this technique in the pilot study (Chapter 4.1) and Experiment 1 (Chapter 4.2), we used

that in Experiment 2 (Chapter 4.3).









Publication title: Improving the impact of Luminance Contrast on Window

Appearance in a Conventional Office Room: Using Supplementary Lighting

Strategies. Paper presented at Living and Learning: Research for a Better Built

Environment, 49th International Conference of the Architectural Science Association,

in Melbourne, Australia on December 2nd, 2015.



Signature:

Date: 14/06/2018




_

Date

____14/06/2018_____




4.2.2 Abstract

High contrast ratios between windows and surrounding surfaces could cause

reduced visibility or discomfort for occupants. Consequently, building users may

choose to intervene in lighting conditions through closing blinds and turning on the

lamps in order to enhance indoor visual comfort. Such interventions increase projected

electric lighting use in buildings. One simple method to prevent these problematic

issues is increasing the luminance of the areas surrounding to the bright surface of

windows through the use of energy-efficient supplementary lighting, such Light

Emitting Diodes (LEDs). This paper reports on the results of a pilot study in

conventional office in Brisbane, Australia. The outcomes of this study indicated that a

supplementary LED system of approximately 18 W could reduce the luminance

contrast on the window wall from values in the order of 117:1 to 33:1. In addition, the

results of this experiment suggested that this supplementary strategy could increase

the subjective scale appraisal of window appearance by approximately 33%, as well

as reducing the likelihood of users’ intention to turn on the ceiling lights by about 27%.

It could also diminish the likelihood of occupants’ intention to move the blind down

by more than 90%.

4.2.3 Introduction

Office workers generally spend most of their working time inside the buildings

in which they work (Schweizer et al., 2007). It is well understood that improving

Indoor environmental quality (IEQ) of office buildings can enhance work performance

and reduce absenteeism of office workers, besides reducing energy consumption of

buildings (Fisk, Black, & Brunner, 2011).

Indoor Lighting quality as part of IEQ is one of the most significant attributes of

a working environment (Ne'eman, Sweitzer, & Vine, 1984). Optimal or at least

acceptable indoor lighting quality, which relies on daylight and / or electric lights, can

be achieved through providing high level of visual performance and avoiding visual

discomfort for occupants (Boyce, 2003).

Office buildings generally rely on vertical windows for daylight harvesting,

particularly in high-rise cities (Huang et al., 2014), and they are considerably favoured

in working environments for access to daylight and an outside view (Veitch et al.,

1993). Vertical windows also characterise energy consumption and visual comfort


patterns in buildings (Ochoa, Aries, van Loenen, & Hensen, 2012). For instance,

research suggests that a building with a typical façade, which has about 30% window

to external wall, is likely to consume less energy than a building with fully glazed

façade (Meek & Wymelenberg, 2015).

The ubiquity of high contrast ratios between windows and the surrounding

surfaces of the window especially when they are limited in a portion of wall can lead

to reduced visibility and discomfort glare (Alrubaih et al., 2013). Prolonged exposure

to poor visual conditions may cause headache, visual stress, and eyestrain; besides

negatively affecting satisfaction and productivity of office workers (Boubekri, 1995).

Consequently, building users may intervene by closing blinds and turning on

additional lamps to improve indoor visual comfort (Aschehoug et al., 2000). For

instance, a study among 123 buildings with installed photosensor-control systems

illustrated that there is a comparatively monotonous relationship between the amount

of illuminance from windows and turning on the lights by occupants, in particular

when dimming control systems work perfectly (Heschong et al., 2005). This study

showed that as the window illuminance increases, the probability of switching on the

lights will also increase to up to 60% to reduce luminance contrast between the window

and surrounding areas. Evidently, occupants’ interventions in lighting conditions

increase electricity consumption of buildings.

The aim of this study is to improve user acceptance and visual comfort of typical

day-lit offices, and to reduce negative occupant interventions in these spaces. It is

presumed that one simple and efficient strategy to achieve this is to reduce the

luminance contrast on the window wall by increasing the luminance of the areas

surrounding the window using supplementary lighting, such as LED.

Preliminary small pilot study investigated potential energy saving offered by

using supplementary LED system in a an office room (Amirkhani, Garcia-Hansen, &

Isoardi, 2015b). It evaluated subjective responses, as well as using the DAYSIM

engine within ECOTECT to assess annual energy consumption of the test office room.

The results of this study indicated that increased electricity usage of an approximately

18 W LED lighting strategy, which was not chosen because of its energy efficiency, is

offset where there is roughly one-fourth reduction in users’ intention to intervene in

lighting conditions.


The purpose of this study is assessing subjects’ acceptance for luminance ratios

on the window wall under different lighting conditions using a simple rating scale

(self-reported data). Physical lighting measurements are combined with occupant

surveys to provide a better understanding of discomfort caused by high contrast ratios

between windows and the surrounding window wall when they are in the field of view

of occupants. In addition, different solutions that could reduce any apparent discomfort

have also been tested. The results from this survey present valuable information for

the design of more comfortable and glare-free office environments.

4.2.4 Method

4.2.4.1 Experiment settings

The test is conducted in an individual test office room on the first floor of a 2

storey building located in Brisbane central business distinct (CBD), Australia during

June 2015. The test room is 3.17 m deep by 3.64 m wide and 3.85 m high. Figure 12

illustrates the plan and sections of this room. This room is facing South-West and its

window has ceiling height at 3.6 m and a sill height at 1 m while the width of that is

1.23 m. The walls and ceiling are white and the flooring is grey. Daylight penetration

is controlled by a fabric roller blind, and the room is furnished with a desk and chair,

which are located in front of the window. This room has 2 x 28 W fluorescent

luminaires, suspended 1.3 m from the ceiling. These luminaires can be switched on or

off separately.

Cool-light LED strips, which have matched CCT to sunlight (5600K-7000K),

were chosen to diminish luminance contrast in the field of view of subjects through

distributing light on surfaces around the window. They were pre-assembled in a

channel diffuser to reduce bright spots generally associated with strip LEDs and to

distribute light evenly. Each of pre-assembled LED light strip has 30mm width, 12mm

height, and 513mm length. Each LED strip has luminaire power of 9W and needs a

constant-voltage driver to convert main voltage to 12V. They were also equipped with

a suitable compatible dimmer switch to be able to adjust light level from 0% to 100%.

LED strip cases were mounted on the window sides with sill height of 2.1m and the

bottom of window surface (see Figure 12). It should be noted that the proposed LED

system in this study was chosen as merely an easy method to conduct the test and not

for its energy efficiency.


Figure 12 Plan and sections of the test office room in Brisbane, Australia

The test room relies on reflected sunlight (from adjacent buildings) and diffuse

skylight for indoor daylight harvesting due to its orientation. Thus, the blind was fully

opened during the experiment to have maximum indoor natural light.

4.2.4.2 Questionnaire

The survey was divided into three sections; date and time of conducting the

experiment, basic demographic data from the subject, and some scales to rate

participants’ preferences for window appearance under different lighting conditions.

The number of questions used in this survey was carefully considered to minimise

fatiguing or boring the respondent, while still capturing the significant information

required.

The second part of the survey collected demographic and personal information

relevant to the participant’s glare susceptibility. This included the participant’s age,

whether they wear corrective lenses, and whether the participant considers himself or

herself as a glare-sensitive person.

The third section of the survey related to the participant’s opinion and preference

on the lighting in the test room. It was divided into four stages: no supplementary

lighting, and LED wall-washing of the window surrounds at 3 different power levels

(9 W, 18 W and 27 W). The questions in each stage were designed to find whether the

use of the supplementary lighting system influenced feeling discomfort glare from

windows and subjects’ decision to turn on top lights.

It is frequently challenging to find predictable practical relationships between

physical stimulus and subjective reaction in the field of lighting (Houser & Tiller,

2003). However, some studies have grouped perceived discomfort glare from daylight

into bins of imperceptible, perceptible, disturbing, and intolerable (Suk & Schiler,


2013). The first question at each stage asks participants to rate the level of perceived

discomfort glare from the window when it is in their field of view among these four

groups.

Currently, there are different techniques that can be used to relate subjective

responses to physical parameters in lighting research, including questionnaire, rating

scales, magnitude estimation strategies, and paired comparison (Tifler & Rea, 1992;

Houser & Tiller, 2003). However, according to Houser & Tiller (2003) paired

comparison and semantic differential (SD) scaling are two of the most widely

techniques used in lighting research. SD consists of a set of bipolar adjectives. The

ends of each scale are defined through polar opposite adjectives which are separated

through a seven-point scale (Monette et al., 2013). The number of points to the scale

can be varied between seven, five, or even three (Barbara Sommer, 2006). Therefore,

the second question at each stage uses SD scaling to rate indoor lighting on a scale of

1-5 (one meaning very dissatisfied and five meaning very satisfied).

The last two questions at each stage ask subjects whether they want to move the

blind down or turn on the ceiling lights (yes/no answer). If they respond yes to turn on

lights, further question asks how many they would like to switch on (one or both of

the ceiling lights).

4.2.4.3 Procedure

Thirty-five people participated in this investigation and they were surveyed

individually in the test office room. They were office workers with normal or corrected

to normal vision and representative in age and sex of the general office worker

population. Before starting the experiment, each subject was clearly informed of the

purpose of the research, and shown the light measurement equipment. Each participant

was asked to sit facing the window, around 2.2m from the window surface and the

experimenter stood somewhat behind the subject. They were also asked to fill the first

and second section of the survey themselves; while the researcher led the remainder of

the survey, adjusting light levels and asking questions for a verbal response from the

participant.

To start the first stage of each experiment, all the ceiling lights and the LED

supplementary system were switched off. The experiment followed the same process

during each stage, whereas the luminaire power of the LED supplementary system was


increased by 9W at the start of stages 2 to 4. Participants were given one minute to

adapt to light level changes before the survey started at each stage. Quantitative data

was collected using a Nikon Coolpix 8400 digital camera (calibrated for luminance

measurement (Coyne, Isoardi, Luther, & Hirning, 2008)), Konica Minolta LS100

luminance meter, and Konica Minolta T-10 illuminance meter prior to asking the

questions of each stage.

The digital camera was used to take High Dynamic Range (HDR) images to

observe the luminance distribution at the window and surrounding surfaces. In order

to capture a field of view that is relatively similar to human eye, an FC-E9 fisheye lens

(focal length = 5.6mm, 190° field of view) was used. The camera was located as

practicable as possible to the head of subjects through using a tripod. Multiple pictures

of the same scene were captured during each experiment to achieve a single HDR

image with relative luminance through using Photosphere. In addition, the luminance

meter (LS100) was used to measure the luminance value of a single white spot inside

the room for HDR calibration in Photosphere. Photosphere remembers the response

curve of camera and attached lens. Therefore, it was not essential to measure

luminance values of more than one spot. The illuminance meter was used to record the

illuminance measurement on the working plane (the desk in the test room), which was

0.72 m above the floor and 1.5 m from the window. After collecting quantitative

lighting information at the beginning of each stage while the participant was adapting

to the change in lighting, the experimenter completed the questionnaire by directly

asking the survey questions of the participants.

4.2.5 Results and discussion

Table 6 illustrates mean illuminance measurements at the desk level during each

stage. There was a little variation in exterior lighting conditions across all experiments.

For example, the mean standard variation of horizontal illuminance at the desk level

across all test conditions was 18. Accordingly, about 95% of values were less than 36

lux away from the mean illuminance measurements during each test condition.

Stage

Ceiling lights are off One Ceiling light is on Two Ceiling lights are onmean

illuminance (lux)

Std. deviation

mean illuminance

(lux)

Std. deviation

mean illuminance

(lux)

Std. deviation

1 159 13 250 21 384 252 160 15 251 21 385 23


3 169 17 261 18 397 194 180 18 275 17 409 14

Table 6 Mean horizontal illuminance at the work plane level during each stage

Calibrated HDR images of each stage of all experiments were resized for

calculation. Figure 13 shows an example of a HDR image captured by the digital

camera when overhead lights and supplementary system were off. This image shows

the 12 areas that were targeted for luminance spot measurements using calibrated HDR

images, as well as the illuminance meter located on top of the desk. To obtain the value

of the window to wall luminance ratio, readings 1 to 6 are averaged (to give window

luminance) and compared to the average of readings 7 to 12 (for the surrounding wall

luminance). These ratios are presented in Table 7 below.

Stage 1 Stage 2

Stage 1 Stage 1

Figure 13 Captured HDR image from the test office room and a sample of false colour image during each lighting condition


Table 7 illustrates that as the luminaire power of proposed LED system

increases, the LC between the bright surface of the window and surrounding walls

decreases by about 72% and 81% during stage 3 and 4 to compare with stage 1,

respectively.

Stage Mean window luminance (cd/m2)

Mean wall luminance (cd/m2) luminance ratio

1 2331 20 1172 2406 38 633 2192 66 334 2289 103 22

Table 7 Mean luminance ratio between window and surrounding areas during each stage

Figure 14 plots participants’ response for feeling discomfort glare from the

window at the beginning of each stage during 35 experiments. It illustrates that the

spread of variables during stage 1 generally fall within disturbing and perceptible,

whereas the middle half responses for feeling discomfort glare falls within perceptible

and imperceptible during stage 3 and stage 4. In addition, this figure indicates that

although the median report for feeling discomfort glare during the first three stages

remains the same and is perceptible, it is imperceptible during stage 4. Furthermore,

there is only one person who felt intolerable glare from window throughout all stages.

Accordingly, it has been omitted while analysing data. Overall, this figure suggests

that feeling discomfort glare from windows can be reduced by about 33% through

using proposed LED lighting system.

Figure 14 Boxplot of feeling discomfort glare during each stage

Figure 15 and Figure 16 show the mean subjects’ scores of indoor visual comfort

at the beginning of each stage and also in association with reported discomfort glare


from window. Figure 15 shows that participants’ satisfaction of indoor lighting level

increased by around 17% and 24% throughout stage 3 and 4 to compare with stage 1.

Figure 16 illustrates that the mean indoor visual comfort improved by 24% when

participants did not feel discomfort glare from the window. Finally, these line graphs

indicate that the mean score (about 3.7) for indoor visual comfort during stage 4 is

similar to when reported discomfort glare from windows is imperceptible.

Figure 15 Mean indoor visual comfort during each stage

Figure 16 The relationship between feeling discomfort glare from window and mean indoor visual comfort

Figure 17 plots luminance ratio on the window wall when subjects’ response for

feeling discomfort glare from window is disturbing, perceptible and imperceptible. It

indicates that subjects did not report discomfort glare from window when the median

LC between the window and surrounding surfaces is about 32, which is close to the


mean and median window wall luminance ratio during stage 3 (around 34 and 31

respectively).

Figure 17 Boxplot of window wall luminance ratio and feeling discomfort glare from window

The results of the survey shown in Figure 18 indicate that decreasing the

luminance ratio between the window and immediate walls affect participants wanting

to whether switch on or off top lights or to close blinds. This study suggested that the

mean possibility of subjects’ intention to turn on one or both ceiling lights decreased

by about 27% when their responses for feeling discomfort glare from window were

imperceptible. Approximately 53% of subjects wanted to turn on both overhead

luminaires when they perceived discomfort glare from window. However, only 23%

of subjects wanted to turn on both ceiling lights when they did not perceive discomfort

glare from window. Figure 18 also indicates that the probability of moving the blind

down decreased by about 70% and 96% when subjects’ responses for feeling

discomfort glare from window were perceptible and imperceptible to compare with

when they were disturbing.


Figure 18 Survey results

Table 8 shows some demographic data of participants, including the number of

participants who wore corrective lenses, participant age, and how many considered

themselves to be glare sensitive. The results suggested that there is no relationship

between responses of subjects who wore perception glasses and who did not wear for

feeling discomfort glare at the beginning of each stage. The results also did not indicate

any significant relationship between age and reported discomfort glare in the test room.

In addition, there is not any significant difference between the responses of subjects

who considered themselves to be glare sensitive person and those who did not.

Question Options Number of subjects

Percentage Mode

Perception glasses Reading 4 11.5 NeverDriving 3 8.5

All the time 11 31.5Never 17 48.5

Age Less than 30 19 54.5 Less than 30Between 30 and 50 13 37Between 50 and 65 3 8.5

More than 65 0 0Glare sensitive Yes 22 63 Yes

No 13 37

Table 8 Demographic data of participants

4.2.6 Conclusion and future work

This study investigated users’ acceptance for the luminance ratio on the window

wall using a supplementary lighting strategy. A simple LED system was proposed for

the supplementary lighting strategy. The main aim of this study was to test the impact

of proposed LED system on subjects’ intention to intervene in lighting conditions

through moving the blind down or turning on the ceiling lights. The results from this


study indicated that the proposed LED system could significantly diminish the LC

between the window as a daylight source and surrounding surfaces by about 3.6 fold

(from 117 to 33) during stage 3 and around 5.5 fold (from 117 to 22) during stage 4.

The study also suggested that the mean indoor visual satisfaction increases by about

24% when the luminance ratio of window to wall reduces from values in order of 117:1

during stage 1 to 33:1 during stage 3. In addition, the results of this research indicated

that the median report of discomfort glare from the window is imperceptible, while

using proposed LED lighting system with approximately 18 W luminaire power (stage

3). Consequently, the mean users’ intention to switch on ceiling lights diminished by

about 27% and to move the blind down by more than 90% through using a

supplementary LED strategy with about 18 W luminaire power. Furthermore, this

investigation indicated that there is a monotonous relationship between feeling

discomfort glare from windows and indoor visual comfort.

The tests in this study were not conducted randomly. This research also focused

on a small conventional office room without any specific daylighting system. Further

study is needed to investigate on more rigorous testing of occupants’ perception using

supplementary strategies in various test office environments with different office

layout and window types. In addition, more investigation is needed to improve the

energy efficacy of proposed supplementary system to considerably increase the energy

savings available for this design system.


Chapter 4.3 An Energy Efficient Lighting Design Strategy to Enhance Visual Comfort in Offices with Windows

The pilot study and Experiment 1 were carried out in a real typical office room

facing southwest with around 15% and 27% window-to-external-wall ratios

respectively to test the hypothesis of this PhD research. Nonetheless, the penetration

of direct sunlight inside this room is limited due to its orientation, which could

negatively influence the indoor lighting level on the top of the desk. The results of the

pilot study and Experiment 1 (Chapter 4.1 and Chapter 4.2) illustrate that the average

illuminance on the working plane was about 145 lx and 159 lx respectively when the

proposed LED wall-washing system was off. These illuminance levels are about half

of the minimum recommended level by the Australian standard for general tasks in

office buildings. Therefore, they could affect participants’ intention to intervene in

lighting conditions not only to enhance the window appearance but also to improve

the indoor lighting level.

Experiment 2 was carried out in a typical office room with a different orientation.

It was conducted in a room facing northwest with a 45% WWR that allowed us to have

direct sunlight penetration, and therefore higher illuminance on the working plane in

comparison to the previous studies (Chapter 4.1 and Chapter 4.2). This experiment

was designed to explore whether the proposed LED wall-washing system would be as

efficient as it was during the pilot study and Experiment 1 at reducing LC on the

window wall and diminishing negative lighting interventions in a different typical

office room. This research also investigates the average acceptable LC on the

windowed wall. The purpose of Experiment 2 was to address the following sub-

questions to answer the main question of this PhD research.

Q1: Does the proposed LED wall-washing system reduce the LC betweenthe window and surrounding walls?

Q2: Does the proposed LED wall-washing system improve subjective scale appraisal of the window appearance?

Q3: What is the average acceptable LC on the windowed wall?

The method in Experiment 2 was somewhat similar to the previous studies and

participants were exposed to four default lighting conditions based on the luminaire

power of the proposed LED wall-washing system. The tests were conducted randomly

during each test session to minimise most sources of systematic variation as was


discussed in Part 3, Section 3.2. Moreover, participants were given the opportunity to

set the LC on the window wall to their acceptable level using the LED wall-washing

system. A questionnaire that was developed based on the literature review and the

outcomes of the previous studies (Chapter 4.2) was used to collect subjective responses

during each lighting condition. Physical lighting measures (luminance and

illuminance) were collected during each stage of the experiment using a Nikon Coolpix

8400 digital camera with a fisheye lens, as well as Konica Minolta LS100 luminance

and Topcon IM-3 illuminance meters.

The results indicate that the introduction of the proposed LED wall-washing

system efficiently reduces LC between the window and surrounding walls from

approximately 16:1 to below 11:1. However, there was no significant correlation

between participants’ scale appraisal of the window appearance and the luminaire

power of the LED lighting strategy during all lighting conditions. The outcomes of

Experiment 2 suggest that while it was more likely that participants would turn on the

ceiling lights or move the blind down when the proposed LED wall-washing system

was off, it was less probable that they intervened in lighting conditions when the LED

wall-washing system was on with a low power level. Overall, the results of this study

suggest that an LC of around 11:1 to 12:1 on the window wall using the proposed LED

wall-washing system could be an ideal ratio to achieve visual comfort, as well as

reducing occupants’ intentions to change the lighting conditions.

The average illuminance level on the working plane during Experiment 2 was

about 687 lx when the proposed LED wall-washing system was off due to its

orientation and WWR. This illuminance level was more than four times greater than

the mean illuminance level on the desk top in the room during the pilot study and

Experiment 1 when the LED wall-washing system was off (145 lux and 159 lux,

respectively). Subsequently, the walls surrounding the window in the room during

Experiment 2 were brighter than the room in which previous studies were carried out.

Overall, the LC reduction on the window wall, as well as the decrease of negative

lighting interventions using proposed LED wall-washing system during Experiment 2

were not as dramatic as they were in the pilot study and Experiment 1. Furthermore,

while Experiments 1 and 2 were designed to address sub-question 3 (as mentioned

earlier), the results of these experiments cannot be compared due to the influence of

other room characteristics (e.g., indoor lighting level, outside view, and ceiling height)


on perceived discomfort glare from the window wall. To minimise the impact of these

characteristics on participants’ responses and to better answer sub-question 3, it is

better to conduct an experiment in a typical office room in which we can change the

WWR and the luminaire power of the LED wall-washing system randomly. IVR

technology allows us to change the lighting conditions while maintaining the other

room characteristics quickly and with minimal cost; this would be impossible or

expensive in the real environment. Subsequently, the next study will explore the effect

of the proposed LED wall-washing system on reducing negative lighting interventions

in the controlled IVR typical office room with different window-to-external-wall

ratios.









Publication title: An Energy Efficient Lighting Design Strategy to Enhance

Visual Comfort in Offices with Windows: Published in Energies Journal, Volume 10,

and Issue 8 with the article number of 1126 in August 2017.



Signature:

Date: 14/06/2018



Dr Alicia Allan Contributed to experimental design and data analysis. Reviewed research paper and helped with the editorial process.


_ _____

Date

____14/06/2018_____




4.3.2 Abstract

A high luminance contrast between windows and surrounding surfaces can

increase the risk of discomfort glare, which can diminish office workers’ satisfaction

and productivity. Accordingly, it can lead to occupant interventions, such as drawing

window blinds or increasing electric light levels, which are intended to enhance indoor

visual comfort but counterproductively act to increase energy consumption. Increasing

the luminance of the areas surrounding the windows using a supplementary lighting

system, such as wall-washing with light emitting diode (LED) linear luminaires, could

reduce discomfort glare arising from windowed walls. This paper reports on the results

of a study in a typical office room in Brisbane, Australia. The outcomes of this study

indicate that creating a luminance contrast of between 11:1 and 12:1 on the window

wall in an office room with a 45% window-to-exterior-wall ratio using a

supplementary LED system leads to improved subjective assessments of window

appearance. The results suggest that such an enhancement could significantly reduce

discomfort glare from windows, as well as diminishing the likelihood of the users

intending to turn on the ceiling lights or to move the blinds.

4.3.3 Introduction

The significance of utilizing daylight inside buildings has been highlighted by

several studies (Farley & Veitch, 2001; Hourani & Hammad, 2012; Boubekri et al.,

2014; Ozorhon & Uraz, 2014). Daylight availability in buildings provides a vehicle to

adapt and control the environment in sustainable ways in order to enhance energy

efficiency (Boubekri & Boyer, 1992). While lighting is estimated to represent roughly

40% of total electric usage in office buildings, it has been reported that daylight

harvesting can lead to 20%–77% savings in the lighting consumption of buildings

(Bodart & De Herde, 2002; Krarti, 2016). Daylight availability inside buildings can

also have multiple health advantages for building users, such as decreasing fatigue,

relieving seasonal affective disorder, and diminishing depressive symptoms [8].

Access to daylight can provide information about time and weather, as well as reducing

feelings of isolation, stress, and claustrophobia (Boyce, Hunter, & Howlett, 2003;

Aries, Veitch, & Newsham, 2010). Research suggests that daylighting in working

environments is associated with higher satisfaction and productivity, lower

absenteeism, and a positive attitude (Edwards & Torcellini, 2002; Boccia, Chella, &

Zazzini, 2014).


Office buildings usually rely on side daylighting strategies through vertical

windows for daylight harvesting. The outcomes of a large field study illustrated that

office workers’ satisfaction with indoor lighting were most strongly influenced by

access to windows in their working environment that can provide both natural light

and an outside view (Leder et al., 2016). Vertical windows influence both the energy

consumption of buildings and the visual comfort of the occupants (Ochoa et al., 2012).

A building with a typical facade, which has about a 30%–40% window-to-external-

wall ratio, is likely to consume less energy than a building with a fully glazed façade

(Meek & Wymelenberg, 2015). Meek and Wymelenberg (2015) argue that the

distribution of daylight illumination across a floor area remains similar when the

window wall ratio is between 30% and 40% or above 40%.

Therefore, while vertical windows are often found in existing office buildings,

they usually create high and variable brightness, particularly when they are limited to

a small portion of the window wall (Meek & Wymelenberg, 2015). The resulting

luminance contrast (LC) between the window surface and the immediate surrounds

can cause discomfort glare, especially when windows are in the occupants’ field of

view (FOV). Discomfort glare may not necessarily impair visual performance (unlike

disability glare), but may cause certain physiological and psychological symptoms,

such as headache or stress, over time (Tashiro et al., 2015). It may also lead to

interventions by the occupants in lighting conditions; for example, through closing

blinds and/or turning on the lights to improve indoor visual comfort. Given that

occupants typically leave the blinds closed even when the source of discomfort glare

is long gone, occupant interventions in lighting conditions can reduce the designed

energy savings achieved through the use of daylighting.

As such, there is a significant motivation to explore daylighting strategies that

limit the occupants’ experience of glare. The largest field study on the effectiveness of

side-lighting controls for daylighting (examining 123 buildings with installed

photosensor-control systems) illustrated that there is a relatively constant relationship

between the amount of illuminance from windows and the turning on of lights by

occupants, even when dimming control systems work perfectly (Heschong et al.,

2006). In addition, this study showed that, as the illuminance from windows increases,

the likelihood of turning on the lights to diminish LC increases up to 60%. Heschong

et al. (2006) also illustrated that less than 25% of the predicted (modelled) energy


savings arising from daylight harvesting systems were being realized in practice. This

tendency to turn on indoor lighting as the illuminance from windows increases

suggests an attempt by occupants to reduce the contrast between windows and the

surrounding wall.

4.3.4 Luminance Contrast

While it is essential to provide a good task visibility with no distracting glare in

office spaces (Boubekri & Boyer, 1992), building professionals and lighting designers

are frequently challenged to control glare inside buildings (Suk, Schiler, & Kensek,

2013). Glare inside buildings usually occurs as a result of lighting systems, and there

is a linear association between the average luminance of glare patches and the average

luminance of the visual field (Gordon, 2003; Kim & Kim, 2012). Discomfort glare can

be prevented through the limiting of the absolute lighting level of any surface, daylight

element, windows, or electric lights (Steffy, 2008). It is well understood that some

amount of LC is necessary to improve visual performance (Alrubaih et al., 2013);

however, high LC should be avoided to prevent discomfort glare.

According to Bean (2012), vertical surfaces in an environment play a significant

role in the perception of discomfort glare, and a range of specific luminance contrasts

have been suggested for different applications. The Chartered Institution of Building

Services Engineers (CIBSE) and the Illuminating Engineering Society of North

America (IESNA) recommend that the LC between light sources and adjacent areas,

and anywhere within the normal FOV, should be less than 20:1 and 40:1, respectively

(Boubekri & Boyer, 1992; CIBSE, 1994). The Swedish Agency for Economic and

Regional Growth (NUTEK) in Sweden has stricter recommendations: that the

luminance contrasts between any points within FOV should not exceed 1:20 (Boubekri

& Boyer, 1992). Other recommendations suggest that the LC between the task surface,

immediate surrounds, and distant areas should be less than the ratios 1:3:10 (Arup &

Arup, 2007). Despite these recommendations, there is a limited investigation of what

LC provides in term of optimal visual comfort, and whether this LC changes at

different levels of overall illuminance and office settings.

Rating tools exist to assess indoor environment quality and the energy

performance of buildings, such as the Building Research Establishment Environmental



(LEED), and Green Star (Iyer-Raniga et al., 2014). As an example, the Green Star

environmental rating system is a recognised rating tool, which has been launched and

developed by the green building council of Australia (GBCA) since 2003 (GBCA,

2015). However, these rating systems may not capture visual comfort effectively. A

previous study in Brisbane, Australia found that roughly 50% of full-time employees

who work in buildings that are at least five-star rated by GBCA experience discomfort

glare from daylight sources at their computer unit (Hirning et al., 2013; Hirning et al.,

2014). Another study with 2540 participants in 36 sustainable (green) buildings across

10 countries also indicated that glare from daylight is a major issue (Baird &

Thompson, 2012).

4.3.5 Novel Strategies to Reduce Window Wall Luminance Contrast

Innovative side and top daylighting systems have been developed to enhance

indoor lighting quality in office buildings. The main aim of these systems is to send

daylight deeper into the building, while simultaneously reducing glare from sunrays

and excessive solar gains (Rea, 2000; Garcia-Hansen, 2006). This is mainly achieved

by using optical devices, materials, and elements, including louvres, blinds,

lightshelves, screens, and light filters, especially in side-lit office buildings. According

to Mayhoub (2014), the major challenges of the existing daylighting strategies are the

maintenance of daylighting quality, cost-efficiency, applicability, and ease of

installation and operation to penetrate the market. Accordingly, the market penetration

of some of the existing innovative daylighting systems is very limited due to their high

cost and risk of discomfort glare (Tsangrassoulis, 2008).

For existing office buildings, one potential strategy to reduce the LC between

window and wall (and therefore improve window appearance) is to increase the

luminance of the areas immediately surrounding the window. This could be achieved

by mounting a light emitting diode (LED) linear luminaire around the window frame,

the use of which (over time) would result in lower energy consumption than occupant

use of the indoor lighting system to its full capacity. The benefit of using such a

supplementary LED lighting system is that it can be fitted into existing buildings with

minimal construction modifications and at a low cost.

This study builds on previous work which investigated visual comfort in two

single typical office rooms with dissimilar orientations (southwest and northwest) and


window types (punch window and strip window) in Brisbane, Australia. These rooms

were chosen because of their window wall ratios, which were approximately 40%,

typical of many office environments and likely to consume less energy than a fully

glazed facade. A preliminary study with 35 participants conducted in an office room

facing southwest with a punch window suggested that a supplementary LED system of

approximately 18 W could reduce the LC on the window wall from values in the order of

117:1 to 33:1 under sunny sky conditions (Amirkhani, Garcia-Hansen, & Isoardi,

2015a). It also indicated that this supplementary strategy could diminish the mean

users’ intention to turn on the ceiling lights by approximately 27%, as well as reducing

the probability of moving the blind down to up to 90%. Furthermore, another study in

the same test office room reported that increased electricity consumption of an

approximately 18 W LED lighting system is offset where there is roughly a one-fourth

reduction in users’ intention to intervene in lighting conditions (Amirkhani et al.,

2015b).

The hypothesis of the current study is that the use of linear LED luminaires,

mounted around the window frame so as to wash the window wall with light, would

increase visual comfort in typical single office rooms, as well as reducing negative

occupant interventions in lighting conditions. The aim of this study is to demonstrate

the impact of using a supplementary LED lighting system on the occupants’ scale

appraisal of window appearance, as well as their propensity to intervene in lighting

conditions. This study examines how occupants might respond to different luminance

patterns brought about by changes to lighting design. Specifically, it investigates

whether a mounted LED wall-washing strategy to enhance window appearance by

reducing LC between window and the surrounding wall can reduce occupant visual

discomfort and intention to intervene in lighting conditions. Furthermore, although a

uniform luminance distribution across a room is generally not desirable, the most

acceptable LC for the visual comfort of the occupants is still unknown (Boubekri &

Boyer, 1992). As a result, this study investigates the effects of a range of light intensity

levels from a proposed LED wall-washing system on the perceived window

appearance and indoor visual quality. This investigation will enhance our

understanding of an integrated lighting design solution for a better acceptance of the

window appearance, which could reduce the intention of the occupants to intervene in

lighting conditions. The outcomes of this study present a valuable insight into how a


supplementary LED lighting system could increase energy savings in typical office

buildings through the improvement of window appearance.

4.3.6 Method

This study used a repeated-measures design to assess participant ratings of

discomfort glare and lighting acceptability under four different lighting conditions: no

supplementary lighting (condition 1), and LED wall-washing of the window surrounds

at three different power levels (low power level with about 18 W, medium power level

with around 24 W, and high power level with approximately 30 W: conditions 2, 3 and

4, respectively). Following a random presentation of the lighting conditions,

participants were asked to adjust the luminaire power of the lighting LED system to

their preferred comfort level. Forty participants with normal or corrected to normal

vision participated in this study. Table 9 describes the demographic characteristics of

the participants.

4.3.6.1 Experimental Setting

The experiment was conducted in a typical office room in Brisbane, Australia from

December 2015 to March 2016. The selected test office room was a single office room

facing northwest and located on the seventh floor of a seven-story building at Gardens

Point Campus of Queensland University of Technology (QUT). The central business

distinct (CBD) of Brisbane and the sky can be seen from inside this room. The room

was 4.22 m deep by 2.93 m wide and 2.6 m high with white walls, white ceiling tiles,

and a floor finished with grey carpet. Daylight penetration was controlled by external

shading projecting horizontally from the top of the window wall, as well as a manual

fabric roller blind inside the room. The window of this room had a head height at 2.4 m,

a sill height at 1.05 m, and a window width of 2.38 m, resulting in a window wall ratio

inside this office room of approximately 45%. The room was furnished with an L-shaped

desk and chair, which were located in front of the window, with the chair located at

about a 45° angle to the window surface. The room had two recessed mounted

fluorescent luminaires with a channel diffuser, which could only be turned on or off

together. Figure 19 shows the furniture plan and sections of this office room.

4.3.6.2 Supplementary LED Lighting Intervention

Cool-light LED linear luminaires with a correlated colour temperature (CCT) of

6000 K to 6500 K were chosen to manipulate the lighting contrast in the FOV of


participants by providing a wall-washing light on surfaces around the window. They

were pre-assembled in a channel diffuser to reduce the bright spots generally

associated with LED strip lighting and to distribute light evenly. The proposed LED

strategy was chosen both because of its energy efficiency and for its convenience as an

”off-the-shelf“, pre-assembled luminaire system. Each luminaire contained an LED light

strip with 30 mm width, 12 mm height, and 513 mm length. Each LED strip had

luminaire power of 9 W and needed a constant-voltage driver to convert mains voltage

(240 V) to 12 V. They were also equipped with a suitable compatible dimmer switch

to enable adjustment of the light level from 0 to 100%. Six metal surfaces with 525

mm length and 8 mm width, which were bent at a 45° angle, were used to shield the

participants from direct view of the LED and facilitate the wall-washing function (See

Figure 19). The LED luminaires were mounted on the left window side with head

height at 2.2 m and on the bottom of the window surface to illuminate the walls

surrounding the window (see Figure 19).

Figure 19 Plan and sections of the test office room, with details of LED lighting system placement and construction



Upon presentation of different lighting scenes using the LED system at different

power levels, the participants responded to lighting appraisal questions designed to

assess their ratings of discomfort glare and indoor visual comfort, as well as their

intention to turn on the ceiling lights or move the blind down. The questions in this

study were closed question types to allow the comparability of participant responses

on a standard scale. The number of questions used in this survey was kept to a

minimum to avoid fatiguing or boring the respondent, while still capturing the

information required. It is frequently challenging to find predictable, practical

relationships between a physical stimulus and a subjective reaction in the field of

lighting; however, many studies have asked participants to rate their perceived

discomfort glare from daylight using descriptors of imperceptible, perceptible,

disturbing, and intolerable (Wienold & Christoffersen, 2006; Yamin Garreton,

Rodriguez, & Pattini, 2014; Suk, Schiler, & Kensek, 2017). In this study, these

descriptors were used to elicit participant ratings of perceived discomfort glare, as well

as semantic differential (SD) rating scales using a set of bipolar adjectives, which were

used to gather ratings regarding the suitability of the lighting environment.

Figure 20 shows the questions that were used in this study. The questionnaire

was divided into three sections. The first part of the questionnaire collected

demographic and personal information relevant to the participant’s glare

susceptibility, and addressed (1) the participants’ gender and age group, (2) the use of

prescription glasses or contact lenses, and (3) whether they considered themselves as

a glare-sensitive person (rated using SD scaling).


Figure 20 The questionnaire measuring participant characteristics, responses to lighting conditions and preferred luminance contrasts

The second section of the survey was administered during the experiment, and

researchers verbally asked participants to rate their discomfort and satisfaction of each

of the four different luminaire power conditions as they were presented. In each

section, the first question asked participants to rate the level of perceived discomfort

glare from the window when it was in their field of view on the following scale:

imperceptible, perceptible, disturbing, and intolerable. To reduce uncertainty over the

meaning of each discomfort glare descriptor used in this research, the borderline

between imperceptible and perceptible was defined as the turning point where

discomfort glare would first be noticed. Furthermore, the borderline between

disturbing and intolerable glare was defined as the changeover point where

participants would no longer be able to tolerate the lighting conditions. The second

question asked participants to rate the indoor visual comfort on a scale of 1–5, where

one was very dissatisfied, two was somewhat dissatisfied, three was indifferent, four

was somewhat satisfied, and five was very satisfied. The borderline between somewhat

dissatisfied and very dissatisfied was defined as the changeover point where

participants would no longer tolerate indoor lighting conditions for working with a

laptop. The boundary between somewhat satisfied and very satisfied was defined as


the turning point where indoor visual quality could be slightly improved for working

with a laptop. Finally, the last two questions of each section asked participants whether

they wanted to move the blind down or turn on the ceiling lights, using a yes/ maybe/

no answer format.

Finally, participants were asked to indicate their LC preferences while ceiling

lights are off. In this section, participants were asked to adjust the luminaire power of

the LED strip cases to a setting where they felt more comfortable with the window

appearance, and then the closest luminaire power to the luminaire of LED system was

recorded. The second question asked whether the participants still perceived

discomfort glare from the window (yes/ no answer) after changing the power settings.

Finally, the last two questions asked participants whether they intended to move the

blind down or turn on the ceiling lights to improve the indoor lighting quality

(answering yes, maybe, or no).

4.3.6.4 Lighting Measures

Physical lighting measures (luminance and illuminance) were collected during

each stage of the experiment using a Nikon Coolpix 8400 digital camera (Nikon

Corporation, Tokyo, Japan) with a fisheye lens, as well as Konica Minolta LS100

(Konica Minolta, Tokyo, Japan) luminance and Topcon IM-3 (Topcon Technohouse

Corporation, Tokyo, Japan) illuminance meters. The digital camera was used to take

high dynamic range (HDR) images to observe the luminance distribution at the

window and the surrounding surfaces. To capture a field of view which is relatively

similar to the human eye, an FC-E9 fisheye lens (focal length = 5.6 mm, 190° field of

view) was used. The camera was located as close as practically possible to the heads

of the participants, using a tripod (see Figure 19). Multiple pictures of the same scene

were captured during each experiment to achieve a single HDR image with relative

luminance (using Photosphere software (Anyhere Software, Berkeley, CA, USA)).

The luminance meter (LS100) was used to measure the luminance value of a single

white spot inside the room for HDR calibration in Photosphere, using the response

curve of the camera and attached lens. The illuminance meter was used to record the

illuminance measurement on the working plane (the desk in the test room), which was

0.72 m above the floor and 0.6 m from the window.


4.3.6.5 Procedure

To reduce variation in factors associated with the time of day, such as external

brightness and temperature, experimental sessions were conducted between 11 am and

2 pm under clear sky conditions. However, the participants were not able to see the

sun at any point during the test conditions. Participants completed the study

individually in the test office room, and before commencing the experiment,

participants were asked to sit in the office room for at least five minutes to adapt to the

indoor ambient light. Participants sat at an approximately 45° angle to, and

approximately 1.5 m away from, the window wall surface. The researcher stood behind

at about a 45° angle to the window surface (see Figure 19). A diagonal arrangement of

the workstation was selected to reduce the participants’ intention to move their head

or to deviate their sight between the visual display unit (VDU) and the window during

the experiment, and to ensure that the VDU did not obstruct the window view. During

the first five minutes, each participant was clearly informed of the purpose of the

research and was shown the light measurement tools.

Figure 21 indicates the process of each experimental session. The participants

were asked to complete the first section of the questionnaire themselves. Then,

participants were asked to use the provided laptop while each lighting condition was

presented in a random order. They were asked to surf the internet or to read an article

online. After taking quantitative measurements of luminance and illuminance, during

which participants were adapting to the change in lighting, the researcher completed

the second section of the questionnaire by directly asking the survey questions to the

participants. This enabled participants to remain focused on the VDU during the

experiment. The four different lighting conditions were tested randomly to eliminate

order effects. The same questions were asked over the course of all four conditions,

and the participants were asked to work with the provided laptop while responding to

the questions.

After rating the randomly-presented four lighting conditions, participants were

asked to adjust the lighting level of the LED luminaires to a power setting at which

they perceived the least amount of discomfort glare from the window appearance while

working on the provided laptop. The aim of this stage was to give freedom to the

participants to adjust the LC on the window wall themselves using the proposed LED


system. After collecting physical lighting measures, the researcher filled in the third

part of the questionnaire by directly asking the survey questions to the participants.

Figure 21 Experimental flow in the test office room

4.3.7 Results

Physical measurements and responses to the participant survey questions were

entered into IBM SPSS version 23 for further analysis. Descriptive and inferential analyses

are reported in Sections 5.1–5.6.

4.3.7.1 Participants

Table 9 shows the collected demographic data of the participants, including their

gender, age, use of corrective lenses, and their glare sensitivity. There were slightly

more male than female participants, most were below thirty years old, more than half

reported wearing glasses for at least some tasks, and 44% of participants reported being

moderately or very much sensitive to glare.

Question Options Number of Participants

Percentage Mean, Median, or Mode

Gender Male 22 55% Mode: MaleFemale 18 45%

Age Less than 30 22 55% Mode: Less than 30Between 30 and 50 14 35%Between 50 and 65 2 5%

More than 65 2 5%Prescription

glassesReading 5 12.5% Mode: NeverDriving 2 5%

All the time 16 40%Never 17 42.5%

Glare sensitive

Not at all 4 10% Mean: 3.18Median: 4A little 8 20%

Indifferent 10 25%Moderately 13 32.5%Very much 5 12.5%



Ratings of glare sensitivity did not differ across gender. The average reported

glare sensitivity for males was 3.18 (SD = 0.11, reported on a 1 to 5 scale where 1 was

not at all and 5 was very much), and for females it was 3.17 (SD = 0.129). Perceived

discomfort glare ratings during the first four test conditions were similar for males (M

= 2.00, SD = 0.08) and females (M = 1.85, SD = 0.09), where a rating of 1 meant

imperceptible and 4 meant intolerable). Therefore, both males and females, on

average, considered themselves as indifferent to discomfort glare at the beginning of

each experiment, and reported mild discomfort glare from the window wall while

working on a laptop during the experiment.

4.3.7.2 Window Wall Luminance Contrast Calculation Method

Calibrated HDR images of each test condition during all experiments were used

for the luminance calculation on the window wall. Figure 22 illustrates an example of

an HDR image captured by the digital camera in the absence of all indoor lighting.

This image shows the separate areas for which luminance was calculated from the

calibrated HDR pictures using Photosphere software. The vast majority of the

participants reported that the left-hand side of the window was more in their FOV

compared to the right-hand side of the window wall while working on the laptop.

Additionally, the right-hand side of the window was framed with a perpendicular side

wall, rather than a surrounding wall area (see Figure 22). This meant that the side wall

reflected the entering daylight, and did not suffer from the same contrast problems

typical of a window wall. As such, the LED lighting system was not installed on the

right-hand side of the window, and the LC for this right side was not calculated.

Because the area surrounding the window was not symmetrical, LC was calculated for

the left and right side of the window, as well as for the window as a whole. Finally, as

the distribution of luminance values on the window wall areas was not normal at the

beginning of each stage, the median LC was considered to be the best representation

of the LC during each test condition, and all subsequent results are based on the median

LC for each area.

Figure 22 shows the different areas used to calculate the luminance contrasts of

interest in this study, as well as the calculation equations. To obtain the median

window wall LC on the left-hand side of the window wall, the median window

luminance in area one was compared to the average median wall luminance in areas

three and four. Similarly, the median window luminance in area two was compared to


the median wall luminance in area five to determine the median LC on the right-hand

side of the window wall. Moreover, the median luminance of areas one and two

(window surface) were averaged and compared to the average median luminance of

areas three to five (surrounding walls) to obtain the median LC of the whole window

wall area during each test condition.

Figure 22 Captured high dynamic range (HDR) image from the test office room and window wall luminance contrast (LC) calculation equations

4.3.7.3 Impact of the LED Lighting System on the Window Wall Luminance Contrast

Table 10 summarizes the LC of both the left and right-hand side on the window

wall, calculated as per the previous section, as well as the LC of the whole window

wall areas during each lighting condition. It shows that, on average, as the intensity of

the LED strip lighting increased, the LC of the whole window wall area decreased

from approximately 16:1 when the LED strip lighting was not activated to

approximately 9:1 at the highest (high power level with around 30 W) setting.

Analysis via one-way analysis of variance (ANOVA) showed that the LED

lighting strategy significantly reduced the LC of the window wall (left side: F (3,153)

= 19.26, p < 0.001, right side: F (3,153) = 12.73, p < 0.001, and whole window F

(3,153) = 18.69, p < 0.001). Post-hoc comparisons with bonferroni adjustment showed

that the LC when no LED lighting was used was significantly greater than the LC of

all LED power conditions, but that differences between power levels were not

significant (see Table 10). This shows that the LED lighting strategy did effectively

reduce the average contrast between the window and surrounding wall. However, the

range of LC measured on the window wall (both left and right sides, as well as the

whole window) was wide during each lighting condition; for instance, the range of LC

of the whole window wall areas during Stage 1, when the supplementary LED system


was off, was between 8:1 and 34:1. Horizontal illuminance was also calculated, and as

expected, despite the LED lighting strategy resulting in changes in the LC on the window

wall, horizontal luminance on the desk did not differ significantly across lighting

conditions (M range: 670–687 lux, F (3,156) = 0.10, p = 0.962).

Lighting conditionsWindow Wall Luminance Contrast

Left Side Right Side WholeStage 1 (no LED lighting) 15.8 14.79 15.92

Stage 2 (low power) 10.78 10.23 10.57Stage 3 (medium power) 10.26 9.13 9.77

Stage 4 (high power) 8.96 8.16 8.63

Table 10 Average median luminance contrast of the left and right-hand side on the window wall, as

well as the luminance contrast of the whole window wall areas during each lighting condition

Analysis via one-way ANOVA indicated that the average median luminance of

the VDU did not change considerably during Stages 1 to 4 (M range: 116–134 cd/m2,

F (3,150) = 1.36, p = 0.258). As mentioned in the literature review, the LC between

the task surface and distant areas should be less than 1:10 to avoid visual discomfort.

To obtain the median LC of the task surface (VDU) and the window surface, the

average median window luminance of areas one and two were normalized to the

median task luminance (VDU). Similarly, the average median wall luminance of areas

three to five was normalized to the median VDU luminance to obtain the median LC

of the VDU and the walls surrounding the window frame. Table 11 summarizes the

LC between the VDU and the window surface, as well as the LC of the VDU and the

walls surrounding the window during Stages 1 to 4. It shows that the LC between the

VDU and the window surface during the first four lighting conditions (M range:

1:12.1–1:15.2) was slightly more than the recommended value. However, there was no

significant difference between the LC of the VDU and the window surface during

Stages 1 to 4 (F (3,149) = 0.31, p = 0.817). Overall, Table 11 indicates that the LC of

the VDU and the walls surrounding the window frame (M range: 1:1.04–1:1.54) was

below 1:10 across lighting conditions that minimize the probability of perceived visual

discomfort from the task surface during the tests.

Lighting conditions VDU :Walls Surrounding the Window

Frame:

Window Surface

Stage 1 (no LED lighting)

1 : 1.04 : 15.2

Stage 2 (low power) 1 : 1.43 : 14.0Stage 3 (medium power) 1 : 1.54 : 13.7


Stage 4 (high power) 1 : 1.53 : 12.1

Table 11 Average median luminance contrast of the visual display unit (VDU) and the walls

surrounding the window frame, as well as the luminance contrast of the VDU and the window surface

4.3.7.4 Reported Glare and Satisfaction with the Use of the LED Lighting System

Examination of the participants’ discomfort glare ratings revealed only one

rating of intolerable discomfort glare (the highest discomfort rating) across all

participants and conditions, limiting the meaningful comparison of this rating level.

Therefore, the proportion of participants reporting disturbing discomfort glare from

the window was compared to those reporting imperceptible or perceptible glare. Figure

23 shows the percentage of participants who reported discomfort glare from the

window as disturbing during Stages 1 to 4. Fewer participants reported discomfort

glare as disturbing at the beginning of Stage 2 compared with the other lighting

conditions; however, the proportion of participants rating the glare as disturbing or

intolerable in the absence of LED lighting was not significantly different to that in

Stage 2 (McNemar’s chi-square test, p = 0.219). There was also no significant

difference between these proportions between the low power setting and high power

setting (p = 0.065).

Figure 23 Percentage of participants reporting discomfort glare as disturbing during each lighting condition

Participants also rated their satisfaction with the lighting quality when working

with the provided laptop in each lighting condition. Participants were most satisfied

with the medium power setting (M = 3.65, SD = 0.92), and least satisfied when the

LED system was not on (M = 3.43, SD = 1.01); however, a repeated measures ANOVA


demonstrated no significant differences between ratings of satisfaction with the

lighting quality across lighting conditions (F (3,117) = 0.66, p = 0.582).

As expected, across all presented lighting conditions, participants who reported

increased discomfort glare were less satisfied with the indoor lighting overall. For

example, approximately 8% of participants were very satisfied with the indoor lighting

conditions when their responses for perceiving discomfort glare from daylight were

perceptible, while approximately 43% of participants were very satisfied when they

did not report perceived discomfort glare from the window. There was a significant

negative correlation (Rho p < 0.001) between perceiving discomfort glare

and indoor lighting satisfaction.

4.3.7.5 Acceptable Luminance Contrast on the Window Wall

The low power (around 18 W) setting was rated most comfortable by participants

and, on average, resulted in a LC of approximately 11:1. Table 12 shows the median

LC of the left and right-hand side on the window wall, as well as the median LC of the

whole window wall areas for each level of rated discomfort glare (across all four

manipulated lighting conditions). The correlation between the LC on the window wall

and perceived discomfort from the window was not significant (Spearman’s Rho

correlations with left side rs = 0.02 p = 0.840, right side rs p = 0.154, whole

rs p = 0.444). However, an examination of the average LC associated with

each level of rated discomfort glare revealed that, when participants reported

imperceptible discomfort glare from the window while working on the provided

laptop, the average LC was approximately 12:1.

Discomfort Glare Rating

Number of Responses

Window Wall Luminance ContrastLeft Side Right Side Whole

Intolerable 1 10.82 8.75 9.96Disturbing 36 11.72 9.97 10.94

Perceptible1 73 10.94 9.69 10.43Imperceptible2 47 12.11 12.51 12.32

1Missing n = 1, 2Missing n = 2

Table 12 Average luminance contrast for each level of perceived glare during experimental lighting

conditions

When participants were asked to set their preferred lighting level, the mean LC

on the window wall was 11.06 (SD = 5.67) on the left, 10.51 on the right (SD = 5.86),

and 10.78 (SD = 5.86) overall. Additionally, the mean LC of the VDU and the window


surface was 12.1 (SD = 8.5), and the mean LC of the VDU and the walls surrounding

the window surface was 1.3 (SD = 0.77) when participants were told to set their

preferred lighting level. Table 13 shows the average LC on the window wall and the

participants’ responses for perceived discomfort glare from the window during Stage

5. It indicates that the LC on the left and right-hand side of the window wall was

approximately 12:1 and 11:1, respectively, when participants did not perceive

discomfort. The average median LC of the whole window wall was 11.17 (SD = 6) at

Stage 5 when participants did not perceive discomfort glare from the window. However,

there was no significant difference in the LC on left, right, or the entire window wall

between those who did or did not report glare (Mann-Whitney U = 153, 180, 170, p =

0.630, 0.336, 0.843, respectively).

Perceived Glare During Self-Selected Lighting Level

Number of Responses

Median Window Wall Luminance ContrastLeft Side Right Side Whole

Yes 17 10.39 10.42 10.28No1 22 11.58 10.58 11.17

1Missing n = 1.

Table 13 Average median luminance contrast and participants’ responses for perceived discomfort

glare from the window during stage 5

4.3.7.6 Participants’ Intention to Intervene in Lighting Conditions

During presentation of the four lighting conditions, participants were also asked

whether they would either move the blinds down or turn on the ceiling lights if given

the opportunity. Figure 23 shows that participants were most likely to report they

would either move the blinds or activate the ceiling lights when there was no

supplementary LED lighting, and were least likely to want to intervene at the lowest

LED lighting level; however these differences were not significant (McNemars tests,

p range = 0.143–1.00). Additionally, there was no significant difference between the

LC on the window wall when participants said they would intervene compared to

whether participants said they would maybe intervene or not.

Across all four lighting conditions, when participants rated the glare as

disturbing or intolerable, significantly more intended to move the blinds or turn on the

celling lights than when glare was rated as imperceptible or perceptible (89.19% and 2 (1, N = 160) = 19.08, p < 0.001).


Figure 24 Percentage of participants indicating that they would turn on overhead lights or move the blinds down during each lighting condition

4.3.8 Discussion

The main aim of this study was to examine the effectiveness of a retrofittable

LED wall-washing system to reduce LC and discomfort glare in offices with windows.

We proposed that improving the window appearance could reduce the users’ ratings

of glare, enhance satisfaction with the lighting and reduce the intention to intervene in

lighting conditions, with a potential for savings to the electricity consumption in

buildings. The introduction of the LED wall-washing system did effectively reduce the

corresponding LC on the window wall; however, this reduction in LC was not reflected

in the participants’ perceptions of glare, satisfaction with the lighting in the room, or

intentions to move the blinds or turn on the ceiling lights, and there were no significant

differences between lighting conditions with these measures.

It is possible that a LC of approximately 11:1 to 12:1 between the window, as

the source of daylight, and the surrounding walls may be an ideal ratio to achieve visual

comfort via the proposed LED wall-washing system, as this is the ratio that occurred

when participants were able to adjust the lighting system themselves. This study also

suggests that if the building occupants do not experience glare, their intention to switch

on the ceiling lights or move the blind reduces significantly.

The fact that the LC was not significantly related to perceptions of glare or

satisfaction may explain the lack of the effects of subjective measures. It is possible

that this occurred due to the fact that the LC on the window wall before any lighting

intervention was not particularly high (with an approximately 16:1 ratio). Our previous

research, using a supplementary LED system, found that more dramatic reductions in

LC (from approximately 117:1 to 22:1) could be achieved using a similarly low-


powered system (Amirkhani et al., 2015a). Therefore, it is possible that this system

might be useful in high LC offices, but may have limited effects upon occupant ratings

in lower LC offices.

Future research should therefore assess whether the wall-washing approach

described in this study could improve both the observed luminance contrasts and

subjective ratings in different room conditions. In particular, we recommend the

investigation of this strategy in rooms where the LC with lights off is higher than that

in the room tested in this study. Additionally, this system could also be tested in rooms

with different room layouts, orientations and window types (punch window type and

strip window type), to determine whether it effectively reduces luminance contrasts in

those settings, and whether that reduction is accompanied by changes in occupant

perceptions. More investigation is also needed to examine the impact of participant

lighting preferences on the acceptable LC between the window and surrounding walls

using the supplementary LED lighting system.

Despite the many advantages of using shading devices to limit discomfort glare,

the retrofitting of shading devices to older buildings is not always feasible.

Additionally, shading devices can hinder occupant view and connection to the

outdoors (which occupants typically value highly) and are not as easily adaptable to

changing lighting conditions. It is therefore worthwhile to explore supplementary

electric light strategies, such as that presented here, to reduce the negative impact of

bright windows. A distinct advantage of the tested LED system is not only its low cost

and retrofittable nature, but also its customizability to suit user preferences and

changing outdoor lighting conditions. Although this study provides a suggested

starting point for the optimal window-wall LC, one benefit of such systems is that they

are easily adaptable, and the preferred power setting can be set by the user to ensure

personal visual comfort. Such a system would integrate easily with Internet of Things

technologies, and its easy customizability means that power levels could vary

automatically across the day and in-line with user preferences. Therefore, it could be

considered as an additional option for lighting design that minimizes visual discomfort

while maximizing energy savings. As LED technology advances, increases in the

amount of light produced per unit power input (i.e., luminous efficacy) will also enable

a reduction in the power required to generate the beneficial effects seen from these

supplementary lighting systems. The system demonstrated in this paper showed an


effective design solution using an 18 W LED product; however, the same positive

effect can be achieved with a lower power solution by using purpose-designed, higher-

efficacy LED products.

4.3.9 Conclusion

An integrated lighting design solution to improve window appearance which

potentially leads to increased energy savings of buildings is introduced in this paper.

A supplementary LED lighting system was mounted on the bottom and the left window

side in a single office room in Brisbane, Australia (see Figure 19). This LED lighting

system was designed to illuminate the walls surrounding the window to decrease the

LC on the window wall. Although the system was effective in significantly reducing

the LC on the window wall, it did not reduce the participants’ ratings of glare or

satisfaction with the lighting. The LED wall-washing strategy reduced the LC to below

11:1 without significant changes in horizontal illuminance.

Future research should investigate whether this system is able to match

reductions in LC with changes in subjective ratings, particularly in higher LC

environments, to determine whether there are optimal luminance contrasts to achieve

occupant satisfaction. Because daylight is dynamic and changes in intensity, spectrum,

and direction as the time and weather change, future research could also incorporate

linking supplementary lighting systems to photosensor-based controls that modify

window appearance throughout the day and year. Finally, the proposed system would

be significantly more energy efficient than the use of overhead lighting; however, the

energy savings achievable are contingent on the design of the supplementary system

and the overhead lighting strategies they seek to replace. The optimization of the

product design of supplementary LED lighting systems is required to maximize the

energy savings available from this design strategy.


Chapter 4.4 Innovative Window Design Strategy to Reduce Negative Lighting Interventions in Office buildings

As discussed in Part 1, this PhD research hypothesizes that the use of a

supplementary LED wall-washing system will improve visual comfort in office

buildings, and therefore reduce occupants’ adverse interventions in lighting

conditions. To adequately understand how the proposed LED wall-washing system

impact participants’ intentions to intervene in lighting conditions in the rooms with

different window sizes, it is vital to study such scenarios in real (physical) office

spaces. In previous experiments, we tested the hypothesis of the current PhD research

in two real typical office spaces. The pilot study and Experiment 1 were conducted in

a room facing southwest with around 15% and 27% window-to-external-wall ratios,

respectively (see Figure 6). Experiment 2 was carried out in a room facing northwest

with approximately 45% WWR (see Figure 6). The outcomes of these investigations

suggest that the proposed LED wall-washing system with a low power level could

enhance participants’ scale appraisal of the window appearance, as well as reduce their

propensity to change the lighting conditions.

While it is possible to perform such experiments in existing actual buildings,

several factors might cause experimental noise or affect the outcomes (e.g., different

interior space designs, cloudy/sunny weather on different days, different outside

views, different internal brightness, etc.). As a rationale for the last experiment, we

wanted to be able to manipulate the WWR and the luminaire power level of the

proposed LED wall-washing system while holding other characteristics constant. As

this is difficult to achieve in real life, we used IVR technology during the last

experiment that allows the experimenter to control for most (if not all) potentially

confounding features and isolate the variables of interest (i.e., lighting scenarios). This

technology allowed us to change the lighting conditions quickly and with low cost. It

also enabled us to provide spaces where users can be fully immersed and feel a parallel

sense of presence in physical environments. Experiment 3 was carried out in the IVR

office room facing southwest with different window-to-external-wall ratios.

Experiment 3 set out to address the following sub-question to answer the primary

question of this PhD thesis.


Q4: How do the WWR and the LED wall-washing with different power levels affect occupants’ intentions to intervene in lighting conditions?

The questions in the questionnaire (Appendix E) were adapted from the

questionnaires used in the previous studies of this PhD research. Participants were

exposed to sixteen default lighting conditions based on the luminaire power of the

proposed LED wall-washing system (four conditions) and the WWR in the VR office

room (four scenarios). They were also asked to set the lighting contrast on the window

wall to their preferred level, and to their minimum acceptable level, using the LED

linear luminaires around the window frame in the rooms with different window-to-

external-wall ratios.

This study indicates that the WWR and the luminaire power percentage of the

proposed LED wall-washing system significantly influence participants’ scale

appraisal of the window appearance. The results illustrate that the proposed LED wall-

washing system with a low power level did effectively diminish subjective rated

contrast (RC) scores on the window wall in a room with a 46% WWR. Additionally,

this research suggests that a supplementary LED wall-washing system with a low

power level could improve window appearance in rooms with greater than 30%

window-to-external-wall ratios. The outcomes of this study indicate that participants’

intentions to change the lighting contrast on the window wall significantly reduced

when they reported lower contrast between the window and surroundings.









Publication title: Innovative Window Design Strategies to Reduce Negative

Lighting Interventions in Office buildings. Submitted to Energy and Buildings and is

under review.



Signature:

Date: 14/06/2018



Dr Alicia Allan Contributed to experimental design and data analysis. Reviewed research paper and helped with the editorial process.

Principle Supervisor ConfirmationI have sighted email or other correspondence from all Co-authors confirming their certifying authorship. Dr Veronica Garcia-Hansen Signature Date

____14/06/2018_____




4.4.2 Abstract

Novel lighting strategies have the potential to create luminous environments that

are more satisfactory to building occupants and reduce occupants’ interventions in

lighting conditions. Testing innovative design systems in the immersive virtual reality

(IVR) environment could be a useful approach to investigate these systems quickly

and easily. This paper explores how increasing the luminance of areas surrounding the

window using an electric wall-washing system could improve subjective rated contrast

(RC) scores on the windowed wall, as well as reducing negative lighting interventions

in an IVR office room with different window-to-exterior-wall ratios. The results

indicate that participants report greater lighting contrast between the window and its

surroundings in the room with a 15% window-to-exterior-wall ratio (WWR) compared

with other lighting conditions. The findings of this research also show that the

proposed electric wall-washing system with a low power level could significantly

reduce the likelihood of users’ propensity to intervene in lighting conditions in rooms

with different window sizes.

4.4.3 Introduction

In Australia, it is projected that offices will account for 23% of the total energy

consumption among buildings by 2020 (Department of Climate Change and Energy

Efficiency, 2012). Lighting systems are the second highest energy consumption source

in office buildings (following HVAC system), accounting for 26% of total electricity

consumption over the 1999 – 2012 period (Department of Climate Change and Energy

Efficiency, 2012). Two sources of light, such as daylight and electric light can be used

for lighting design (IESNA & Rea, 2000). Daylighting is an important resource to

enhance the energy efficiency of the buildings through minimizing electric lighting

consumption (Pellegrino, Cammarano, Lo Verso, & Corrado, 2017). Previous research

has established that controlling the integration of daylighting and electric lighting can

lead to significant electric saving ranging from 30% to 77% (Li et al., 2006; Doulos et

al., 2008; Ihm et al., 2009; Pellegrino et al., 2017). Daylight availability inside

buildings can also have several health benefits for occupants, such as relieving

seasonal affective disorder, decreasing fatigue, and diminishing depressive symptoms

[8]. Access to daylight can provide information about time and weather, as well as

decreasing feelings of stress, isolation, and claustrophobia (Boyce et al., 2003; Aries

et al., 2010). Research also suggests that daylighting in working places is associated


with lower absenteeism, higher productivity and satisfaction, and a positive attitude

(Edwards & Torcellini, 2002; Boccia et al., 2014).

Office buildings usually rely on side daylighting strategies through windows, in

particular in high-rise buildings for daylight harvesting. It is now well established by

a variety of studies that office workers desire windows in their working spaces that can

provide both daylight and an outside view (Collins, 1975; Roche, Dewey, & Littlefair,

2000; Galasiu & Veitch, 2006; Garcia-Hansen, Isoardi, & Miller, 2010; Borisuit et al.,

2015). However, vertical windows usually create high and variable luminance contrast

(LC) on the windowed walls, especially when the window-to-exterior-wall ratio

(WWR) is small. High LC on the window wall can cause visual discomfort, mainly

when windows are in the field of view (FOV) of occupants. It may also lead to

occupants’ propensity to intervene in lighting conditions through switching on the

lights and/or moving the blinds to enhance visual comfort. Furthermore, occupants

typically leave the blinds in place even when the source of discomfort glare is long

gone (for days or in some cases weeks) in particular when they have poor outside view

(Sanati & Utzinger, 2013).

Lowry (2016) demonstrated that occupants’ behaviour could significantly affect

lighting energy consumption in buildings. The most significant field study on the

effectiveness of side-lighting controls for daylighting (examining 123 buildings with

installed photosensor-control systems) illustrated that as the LC on the window wall

increases, the tendency to turn on indoor lighting to enhance the LC between the

window and surroundings increases up to 60% (Heschong et al., 2006). The outcomes

of this study showed that occupants’ intervention in lighting conditions reduced

predicted (modelled) energy saving arising from daylight harvesting by more than

75%. Furthermore, research into the impact of human behaviour inside buildings

suggests that motivating occupants to reduce their negative interventions could result

in up to 40% energy savings (Dietz, Gardner, Gilligan, Stern, & Vandenbergh, 2009;

Hong & Lin, 2012; Menassa & Azar, 2012).

An electric wall-washing system has been proposed to improve window

appearance that leads to reduce adverse lighting interventions. This system was

designed to increase the luminance of the areas immediately surrounding the window

to reduce the LC on the windowed walls. The benefit of the proposed electric lighting

strategy is that it can be fitted into existing buildings with little cost and minimal


construction modifications. The outcomes of previous studies in two different real

office rooms with different window-to-exterior-wall ratios (approximately 15%, 27%,

and 45%) indicate that using the proposed lighting system could significantly reduce

negative lighting interventions (Amirkhani et al., 2015b, 2015a; Amirkhani, Garcia-

Hansen, & Isoardi, 2016; Amirkhani, Garcia-Hansen, Isoardi, & Allan, 2017).

However, the results of these studies suggest that the room orientation, outside view,

and the size of the window could influence the outcomes. Rodriquez and Pattini (2014)

also suggest that the size of windows and the position of observers have a significant

influence on perceiving visual discomfort from windows. Interesting outside views

can also increase occupants’ tolerance levels of visual discomfort from the windowed

wall (Tuaycharoen & Tregenza, 2007). Therefore, one objective of the current study

is to explore the influence of the proposed electric wall-washing system on occupant

behaviour and lighting evaluations in rooms with a standard view and with a variety

of window sizes. Immersive virtual reality (IVR) technology was used in this study to

allow us to change the lighting conditions (luminaire power of the proposed wall-

washing system and the window size) quickly and with low cost while providing a

sense of immersion in the physical environment.

4.4.4 Advantages of using immersive virtual reality environments in human behaviour studies

Although it is necessary to assess lighting scenarios in real-life physical settings,

the increasing availability of IVR technology provides a possible avenue for exploring

perceptions of indoor environments quickly and easily while manipulating major

architectural features. Manipulating aspects of a space such as window size or view

type can be time-consuming, costly, and difficult to implement in a repeated-measures

experimental protocol. Additionally, IVR environments allow easy control of factors

such as time of day and cloud cover that can affect the light from daylight moment to

moment and can be difficult to control in experimental settings (Ander, 2003).

The IVR technology can provide virtual spaces where users can be fully

immersed and feel a sense of presence similar to that in physical environments (Zhao,

2003; Brooks et al., 2014). It can also provide an environment in which the

experimenter is less conspicuous to the participant (as participants cannot see the

experimenter), facilitating behaviour that is more natural (Heydarian et al., 2015). This

technology can be used to better understand human decision-making and behaviour in


IVR environments that represent realistic real-world settings (Bosch-Sijtsema &

Haapamaki, 2014). According to Heydarian et al. (2015), human behaviour in an IVR

space is not significantly different from that in a real environment. In summary, using

IVR technology offers the unique advantage of being able to manipulate the control

variables quickly while keeping other design features constant. Therefore, the second

aim of this research is to determine whether IVR may be a useful tool for determining

occupant responses to a range of lighting scenarios and technologies.

4.4.5 Limitations of using immersive virtual reality spaces in lighting research

While the amount of research on IVR environments has increased over the last

two decades, there are very limited studies in the context of architecture (Paes et al.,

2017), and in particular lighting (Heydarian et al., 2015; Heydarian et al., 2016;

Heydarian et al., 2017). These studies have mainly focused on exploring human

behaviour under different lighting conditions. Heydarian et al. (2015) investigated the

impact of personal control on manual and semi-automatic lighting options to enhance

lighting in an office space using either electric lights or daylight in an IVR single office

room. This study illustrates that participants were more likely to use daylight to

enhance indoor visual comfort specifically when they had access to the remote control

to change the position of the shading device. Heydarian et al. (2016) studied the impact

of default lighting settings on participants’ rate of lighting adjustment in an office room

and found that participants were significantly more likely to keep the default lighting

setting if they had daylight available. Another study using IVR technology as an

experimental tool demonstrates that people preferred to have maximum simulated

daylighting compared to electric lighting (Heydarian et al., 2017).

One aspect of IVR that has major implications for lighting research is that the

lighting stimuli presented to participants is not truly reflective of the real-life

environment (Loomis, Blascovich, & Beall, 1999; Chamilothori, Wienold, &

Andersen, 2018). The broad range of luminances that are visible to individuals in a

physical environment are much larger than those able to be presented in an IVR

context. In particular, it is difficult to make observations about visual comfort or

discomfort in an IVR setting, in which the luminance range is constrained, and the

technology, by its nature, avoids the presentation of uncomfortable scenes. The lack

of detail in IVR spaces also means that they are perceived differently to real spaces

(Chamilothori et al., 2018). Therefore, although IVR may be useful for establishing


overall preferences and patterns of luminance distribution, it cannot replicate the

perceptual process that leads to visual discomfort, and results should therefore not be

interpreted in those terms.

In previous research, lighting evaluations using Rhinoceros and Grasshopper

have been conducted to make sure that the created IVR environments represent the

actual lighting conditions (Heydarian et al., 2015; Heydarian et al., 2016; Heydarian

et al., 2017), but there are technical limitations in representing the true luminance

range. This has implications for evaluating experiences such as discomfort, which rely

on the presence of a very large range of luminance values that are not possible to

achieve with IVR headsets. As such, Natephra et al. (2017) suggest that IVR

technology can only provide a semi-realistic lighting environment and that it is

impossible to create the precise perception of illuminance and glare using a head-

mounted display (HMD) with the current technology. For example, to be able to

provide realistic lighting condition in the IVR environment, the screen resolution of

the IVR headset should be around 6000 pixels horizontally and 8400 pixels vertically,

which is not possible with the current technology (Fuchs, 2017). Natephra et al. (2017)

also highlight that it is impossible to analyse the appearance of lighting design, the

distribution of lighting, and quantification of the amount of lighting in real-time using

IVR technology. However, it is possible that IVR could be useful for assessing aspects

of satisfaction with the broad pattern of luminance, and the overall level of contrast,

without addressing the specific question of visual discomfort. Other limitations of the

IVR environments include possible experiences of symptoms of motion (simulator)

sickness and disturbance of balance and eye-hand coordination (Loomis et al., 1999),

but improvements in technology and the use of static scenarios decrease the likelihood

of this occurring (Bailenson & Yee, 2006).

Despite the limitations in the ability of IVR technique to mimic the perceptual

experience of real-life lighting conditions, previous research has suggested that it may

still be a useful tool for investigating responses to lighting scenarios. Therefore, the

IVR environments were used as an experimental tool to place participants in typical

virtual office spaces and easily manipulate window size (four scenarios) and the

luminaire power level of a proposed electric wall-washing system (four scenarios).

The first objective of the current research is to investigate the impact of the WWR and

the power level of an electric wall-washing system on occupants’ lighting preferences


and intended behaviour, and the second is to explore the utility of an IVR space in

examining responses to lighting scenarios.

4.4.6 Method

A repeated-measures design was used to assess participant rated contrast (RC)

on the window wall and participant indoor lighting satisfaction under different lighting

conditions. Figure 27 illustrates the procedure of this study. Fifty-three individuals

with normal or corrected to normal vision participated in this research. Table 14

indicates captured demographic characteristics of the participants. A virtual reality

display room was used to manipulate the WWR and the power level of the electric

wall-washing system.

4.4.6.1 Virtual reality setup

Autodesk 3ds Max was used to create the 3D environment of a virtual reality

office room. To make the virtual reality model as realistic as possible, textures and

materials for every object in the office were also added, through setting up and

including texture/diffuse colour channels, normal channels, specular channels and

occlusion channels using the 3ds Max and V-Ray rendering engine. To achieve linear

workflow, which means that the input voltage to a screen and the output brightness

have a linear correlation, we set the standard settings to work in 2.2 gammas (most

screens have a gamma of approximately 2.2).

The virtual reality office room was designed with four different window-to-

exterior-wall ratios: 15%, 30%, 46%, and 62% (see Figure 25). Cool-white electric

linear luminaires with a correlated colour temperature (CCT) of 6500 K were set up

surrounding the window to manipulate the lighting contrast on the window wall

through providing a wall-washing light on surfaces around the window. The power

level of the electric wall-washing of the window surrounds in each virtual reality office

space could be changed under four default scenarios, including no supplementary

electric lighting, low power level, medium power level, and high power level (see

Figure 26). Participants could also adjust the lighting level of the proposed electric

wall-washing system in the IVR rooms with different window-to-exterior-wall ratios.


Figure 25 Virtual reality office room with four different window-to-exterior-wall ratios

The designed virtual reality office space for this research was placed by itself in

the 3D space and was isolated from the rest of the building components and elements.

A neutral image indicating the central business distinct (CBD) of Brisbane, Australia

with a partial sky view was used for the outside view to minimise the effects of the

outside view on participants’ rating appraisal of the window appearance (see Figure

26). The room was designed to appear to be on the third or the fourth floor of a multi-

story building. This room was 4.1 m deep by 3.05 m wide and 2.6 m high with white

walls (reflection r = 0.6), white ceiling tiles (r = 0.8) and a floor finished with grey

carpet (r = 0.2). The room was furnished with a desk (750 × 1800 mm) and chair,

which was located in front of the window. There was also no particular daylighting

system to control daylight penetration inside the room.

Figure 26 The virtual reality office room with a 30% WWR under different lighting conditions

To maintain all the variables constant that may potentially affect the outcome of

the study, the modelled offices were set to be located in Brisbane, Australia on


February 4th, 2017 with the sun at Azimuth 74 and Altitude 60. Additionally, the

Australian standard recommends a minimum of 320 lux lighting level on the working

plane in office buildings for general tasks involving reading, typing, and writing

(AS/NZS, 2008).

The current experiment uses a relative comparison of lighting contrast on the

window wall in the IVR office rooms with different window sizes to understand the

potential impact of the proposed electric wall-washing system on participants’ lighting

preferences and behaviour. Previous research in real office rooms with different

window sizes and orientations illustrates that low horizontal illuminance on top the

desk could influence participants’ intention to change the lighting conditions

regardless of the lighting contrast on the window wall (Amirkhani et al., 2015a;

Amirkhani et al., 2017). Therefore, the horizontal illuminance on top of the desk was

set to the minimum level of 320 lux when the supplementary electric linear luminaires

were off in rooms with different window-to-exterior-wall ratios through enhancing the

daylight distribution in IVR spaces. Several 360 panorama images were created using

3ds Max for each lighting conditions and were imported into Unity 3D game engine.

The interaction options were then added to Unity. These interaction options were

designed to allow participants to manipulate the luminaire power of the supplementary

electric linear luminaires in IVR office rooms with different window-to-exterior-wall

ratios.

Samsung Gear virtual reality HMD with a 96-degree field of vision and

2560×1140 pixel screen resolution was used in this study. This HMD does not need

wires and works with a phone. A study by Kim, Choe, Hwang, and Kwag (2017)

suggests that the active-matrix organic light-emitting diode (AMOLED) display is the

best fit for IVR spaces as it provides pure black with no flicker or motion blur.

Therefore, we used a Samsung S7 phone as a display with a 2560 × 1440 resolution

screen at roughly 557 pixels per inch pixel density and a 16 million colour AMOLED

display. Its screen has a maximum brightness of 553 lumens, CCT of 6800 K and

gamma of 2.2 (Soneira, 2018). According to the factory information, the Samsung S7

phone screen on automatic brightness mode can provide up to 855 cd/m2 and has the

contrast rating up to 1:186 (Soneira, 2018).



The questionnaire was designed to focus on participants’ lighting satisfaction

and behaviour under different lighting conditions. The questions in the questionnaire

were adapted from those used in previous investigations (Amirkhani et al., 2015a,

2016). The questionnaire was structured as follows:

Time and reference (assessor completed);

Demographic and personal information relating to participants’ glare susceptibility and lighting preference (participant completed);

Rating lighting contrast on the window wall and indoor lighting satisfaction, as well as indicating whether the participant intends to change the lighting contrast on the window wall under 16 different default scenarios randomly (assessor completed based on participant responses);

After every four default lighting conditions, the participants were asked to adjust the lighting contrast between the window and surrounding walls to their preferred lighting level in one of the rooms with different window-to-exterior-wall ratios that were selected automatically and randomly;

The participants were also asked to adjust the lighting contrast on the window wall to a minimum acceptable level in a room similar to the one from the previous stage. This was followed by an assessment of the lighting contrast between the window and surrounding walls and indoor lighting level satisfaction (assessor completed based on participant responses).

Demographic and personal information in the questionnaire was designed to

address the following questions:

Participants’ gender and age group;

Whether they use any forms of eye corrections (prescription glasses or contact lenses) and when (reading, driving, or all the time);

Whether they consider themselves to be glare sensitive by rating between one and five using semantic differential (SD) scale (one means not at all and five means very much);Participants’ preference for lighting while working in an office (choosing one of the following options: daylight, electric light, a combination of daylight and electric light, and not having any preference)

4.4.6.2.1 Rated contrast

At the beginning of the experiment, contrast ratings were explained to

participants with the following instruction: “Rate how much contrast there is between

the window and the wall. High contrast means a big difference between the light level


at the window and the wall surrounding it, and low contrast means little difference

between the light level at the window and the wall surrounding it.” The participants

were asked to rate the lighting contrast on the window wall for each condition, using

a sematic differential scale from one to five, where one means very low contrast, two

means somewhat low contrast, three means neither low nor high contrast, four means

somewhat high contrast, and five means very high contrast. To reduce uncertainty over

the meaning of each RC descriptor used in this research, the borderline between very

low contrast and somewhat low contrast was defined as the turning point where the

lighting contrast would be first noticed. Furthermore, the borderline between

somewhat high contrast and very high contrast was defined as the changeover point

where participants would no longer be able to tolerate the lighting contrast on the

window wall.

4.4.6.2.2 Satisfaction

The participants were also asked to rate their overall satisfaction with the indoor

lighting on a scale from one to five, where one means very dissatisfied, and five means

very satisfied. The borderline between somewhat dissatisfied and very dissatisfied was

defined as the changeover point where participants would no longer tolerate indoor

lighting conditions for working. The boundary between somewhat satisfied and very

satisfied was defined as the turning point where indoor lighting could be slightly

improved. Finally, the participants were asked whether they want to change the

contrast on the window wall by answering yes, maybe, or no during the sixteen default

lighting conditions.

4.4.6.3 Procedure

Figure 27 indicates the process of each experiment session. Participants

completed the study individually in a single office room. They completed the first

section of the questionnaire themselves and were informed about the details of this

research without disclosing any information that might potentially affect their

decisions during the test. The virtual reality equipment was shown to the participants,

and its functionality was explained to them. They were also asked to put on the HMD

and take a moment to adjust the headband size if necessary and to ensure it was

comfortable, and that it did not allow any light to come in around the edges.

They were also informed how to change the luminaire power level of the

proposed electric wall-washing system using the touchpad on the IVR headset, which


they were asked to do during the tests. The researcher could also save the data and

change the lighting conditions using a SteelSeries Stratus XL Wireless Gaming

Controller in the IVR office space. Lighting conditions were presented randomly to

eliminate order effects. Participants were advised that there might be a risk of

experiencing eyestrain or motion sickness while using IVR headset. Nonetheless, these

risks were minimised by using lighting scenarios that were static and did not require

any movement on behalf of the user. Between each lighting condition, participants

took a five to ten seconds break, in which they were presented with a black screen, and

were able to shut their eyes to rest. Upon presentation of each lighting condition,

participants were asked to read aloud a scenario number written on the desk, and look

around the room if they wished for approximately 5 seconds to allow them to adapt to

the new lighting conditions before verbally responding to remaining questions.

Figure 27 Experimental flow

Before beginning with the first lighting condition, participants were provided

with a picture consisting of 30 Landolt rings of different orientation; they adjusted the

focus until they could see the gaps (see Figure 28). Participants then counted the

number of rings with a gap in the right-hand side (seven) and repeated the whole

process if their answer was incorrect. After a five second break, participants were

shown the first lighting condition that was selected randomly from sixteen different

default lighting scenarios.


Figure 28 Landolt ring test in the virtual reality office room

4.4.7 Results

Quantitative data and participants’ responses to the questions of the

questionnaire were entered into IBM SPSS version 23 for further analysis. Descriptive

and inferential analysis of this study is reported in sections 4.4.7.1 – 4.4.7.7.

4.4.7.1 Participants

Table 14 shows collected demographic responses to the questionnaire. Most of

the participants were aged below 50 years. Approximately half (49%) of the

participants had some form of eye correction for at least some tasks, and 56.6% of

participants reported being moderately or very sensitive to glare. The majority of

participants in the current study (54.7%) preferred a combination of daylight and

electric lights in their workplaces, and about one-third (34%) preferred to work under

daylight only.

Options Number of participants

Percentage Mean, Median, or Mode

Gender Male 36 67.9% Mode: MaleFemale 17 32.1%

Age Under 30 25 47.2% Mode: Between 30 and 50Between 30 and 50 27 50.9%

Between 50 and 65 1 1.9%Over 65 0 0%

Prescription glasses or contact lenses

Reading 4 7.5% Mode: NeverDriving 5 9.4%

All the time 17 32.1%Never 27 51%

Glare sensitive Not at all 4 7.5% Mean: 3.43Median: 4A little 9 17%

Indifferent 10 18.9%Moderately 20 37.7%Very much 10 18.9%

Lighting preference Daylight 18 34% Mode: Combination of daylight and

electric lightElectric light 3 5.7%


Combination of daylight and electric light

29 54.7%

Not have preference 3 5.7%


4.4.7.2 Impact of the window-to-exterior-wall ratio and the electric linear luminaires on the window appearance during sixteen default lighting conditions

The WWR and the luminaire power of the electric wall-washing system have a

statistically significant effect on participant RC scores, where higher scores indicated

more contrast on the window wall (Wald: 2(3) = 24.088, 59.625, p < 0.001, 0.001,

respectively). However, a Rank-biserial correlation identified no significant

correlation between the RC scores on the window wall and the WWR when the

supplementary electric wall-washing system was off (rrb (210) = -0.078, p = 0.258).

Figure 29 shows that participants reported higher contrast on the window wall

in the IVR office space with a lower WWR compared with the other groups of window-

to-exterior-wall ratios during the sixteen default lighting scenarios. The probability of

participants’ RC scores on the window wall to be very high versus the probability of

not rating it to be very high in the IVR office room with a 15% WWR was significantly

more than in rooms with a 30% WWR, a 46% WWR, and a 62% WWR during all

lighting conditions (Wald: 2(1) = 8.293, 14.957, 20.895, p = 0.004, p < 0.001, 0.001,

respectively).

Figure 29 The error bar of RC scores on the window wall during all lighting conditions

We conducted comparison analyses to investigate the influence of the

supplementary electric wall-washing system on the window appearance in the IVR


office room with different window-to-exterior-wall ratios. A Friedman test was run to

determine if there were differences in participant scores for the RC on the window wall

in the IVR office space with a 15% WWR between four groups of luminaire power

levels of the electric linear luminaires: "no supplementary electric lighting", "low

power level", "medium power level" and "high power level". This test identified that

the mean RC scores were significantly different in the IVR office space with a 15%

WWR between the four groups of luminaire power levels of the electric linear

luminaires 2(3) = 9.611, p = 0.022. Nonetheless, there were no statistically

significant pairwise comparisons. Likewise, a Rank-biserial correlation indicated that

there was a significant weak, positive correlation between the RC scores on the

window wall in the IVR office space with a 15% WWR and the luminaire power of

the electric linear luminaires (rrb (210) = 0.193, p = 0.005).

Analysis via Friedman test indicated that the mean participant scores for the

contrast on the window wall in the IVR office room with a 30% WWR between the

four groups of luminaire power level of the supplementary electric wall-washing

system were significantly different ( 2(3) = 22.537, p < 0.001). A post hoc analysis

revealed statistically significant differences in mean RC scores between the groups of

“no supplementary electric lighting” and “medium power level” or “high power level”

(p = 0.008, 0.002, respectively), as well as between the groups of “low power level”

(M = 2.74) and “high power level” (p = 0.023). There was also a significant positive

association between participants’ RC scores and the luminaire power level of the

electric wall-washing system in this IVR space (rrb (210) = 0.319, p < 0.001).

A Friedman test indicated that the mean RC scores on the window wall were

significantly different between the four groups of luminaire power level of the electric

linear luminaires in the IVR office space with a 46% WWR 2(3) = 15.183, p = 0.002.

A post hoc test with a Bonferroni correction for multiple comparisons indicated

statistically significant differences in mean RC scores between the groups of “low

power level” and all higher window-to-exterior-wall rations (p = 0.045, 0.009,

respectively), but no significant differences between the other group combinations. A

Rank-biserial correlation indicated that there was a significant positive correlation

between the RC scores on the window wall and the luminaire power of the electric

linear luminaires in the IVR office space with a 46% WWR (rrb (210) = 0.219, p =

0.001).


The mean participants’ RC scores on the window wall in the IVR office room

with a 62% WWR between the four groups of luminaire power level of the

supplementary electric wall-washing system were significantly different (Friedman 2(3) = 18.882, p < 0.001). There were significant differences in mean RC scores

between the groups of “no supplementary electric lighting” and “high power level”, as

well as between the groups of “low power level” and “high power level” (p = 0.003,

0.006, respectively). There was also a significant positive correlation between

participant RC scores and the luminaire power level of the electric wall-washing

system in this IVR office space (rrb (210) = 0.257, p < 0.001).

Overall, Figure 29 and Table 15 indicate that the participants’ RC scores

increases by increasing the luminaire power of the electric wall-washing system during

the sixteen default lighting conditions except in the IVR room with a 46% WWR.

Furthermore, a Rank-biserial correlation test identified a significant negative

correlation between the RC scores on the window wall while the electric linear

luminaires are on with low or medium luminaire power level and the window-to-

exterior-wall ratios in the IVR office space (rrb (210) = -0.207, -0.192, p = 0.002,

0.005, respectively).

Luminaire power level of the electric wall-washing

system

Mean RC scores on the window wall based on the WWR

15% WWR 30% WWR 46% WWR 62% WWR

No supplementary electric lighting

2.98 2.55 2.72 2.49

Low power level 3.15 2.74 2.45 2.51Medium power level 3.55 3.36 3.11 2.89

High power level 3.79 3.49 3.3 3.34

Table 15 Mean RC scores during each lighting condition

4.4.7.3 Impact of the window-to-exterior-wall and the supplementary electricwall-washing system on indoor lighting level satisfaction

The WWR and the luminaire power of the electric wall-washing system have a

statistically significant effect on participant scores for indoor lighting level satisfaction

(Wald: 2(3) = 229.461, 85.866, p < 0.001, 0.001, respectively). Figure 30 and Table

16 indicate that, in general, increasing the lighting level of the supplementary wall-

washing system improves indoor lighting satisfaction. They also show that indoor

lighting satisfaction enhances through increasing the window size in the IVR office

room.


Figure 30 The error bar of indoor lighting satisfaction during all lighting conditions

The assumption of proportional odds was met, as assessed by a full likelihood

ratio test comparing the fit of the proportional odds model to a model with varying 2(18) = 20.449, p = .308. The model was a good fit to the

2(54) = 53.55, p = 0.492. The odds of participants ranking the indoor

lighting level as very satisfied in rooms with a 15% WWR, a 30% WWR, and a 46%

WWR were 0.052, 0.26, and 0.45 times, respectively, that of participants ranking

indoor lighting level in the room with a 62% WWR (Wald: 2(1) = 216.829, 53.772,

19.47, p < 0.001, 0.001, 0.001, respectively). The odds of participants ranking indoor

lighting level as very satisfied in rooms with a 15% WWR, and 46% WWR were 0.199,

and 1.733 times, respectively that of participants ranking indoor lighting level in the

room with a 30% WWR (Wald: 2(1) = 76.187, 9.617, p < 0.001, p = 0.002,

respectively). The odds of participants ranking indoor the lighting level as very

satisfied in a room with a 15% WWR was 0.115 (95% CI, 0.079 to 0.167) times that

of participants ranking the indoor lighting level in the room with a 46% WWR (Wald: 2(1) = 128.646, p < 0.001).

A comparison was performed between the four groups of window-to-exterior-

wall ratios and participants’ indoor lighting level satisfaction scores while the

supplementary electric wall-washing system was off. A Friedman test indicated that

the mean ranks of indoor lighting satisfaction scores were statistically significantly 2(3) = 88.157, p < 0.001. Somers' d test revealed a strong

significant positive correlation between participant indoor lighting satisfaction and


WWR when the supplementary electric wall-washing system was off (d = 0.507, p <

0.001).

Luminaire power level of the electric wall-washing

system

Mean indoor lighting satisfaction scores

15% WWR 30% WWR 46% WWR 62% WWR

No supplementary electric lighting

1.75 2.6 3.09 3.68

Low power level 2.21 3.09 3.49 3.94Medium power level 2.42 3.47 3.6 4.06

High power level 2.75 3.53 3.75 3.92

Table 16 Mean indoor lighting satisfaction scores during each lighting condition

4.4.7.4 Correlation between the window wall rated contrast scores and indoor lighting level satisfaction

Figure 31 indicates the mean scores of satisfaction with indoor lighting levels

for performing office work based on participants’ RC scores on the window wall

during all sixteen default lighting conditions. As expected, participants who rated the

lighting contrast between the window frame and surrounding walls higher were less

satisfied with indoor lighting quality. For instance, approximately 15% of participants

were very satisfied with indoor lighting level when their RC rankings on the window

wall were “very high contrast,” while approximately 42% of participants were very

satisfied when their RC rankings on the window wall were “very low contrast.” On the

other hand, roughly 51% of participants were very dissatisfied with the indoor lighting

level when they reported very high RC on the window wall, while almost 28% of

participants were very dissatisfied when they reported very low contrast on the window

wall. Overall, Kendall's tau-b correlation test identified a statistically significant,

weak, negative association between the RC scores on the window wall (one meaning

very low contrast and five meaning very high contrast) and ratings of satisfaction with

indoor lighting level during sixteen default li b = -0.158, p < 0.001.


Figure 31 The error bar of participants’ indoor lighting level satisfaction rankings based on luminance contrast scores on the window wall

4.4.7.5 Participants’ intention to change luminance contrast on the window wall

During the presentation of the sixteen default lighting conditions, participants

were asked whether they would change the lighting contrast on the window wall if

given the opportunity. Figure 32 shows that participants were most likely to report they

would modify the lighting contrast on the window wall when there was no

supplementary electric wall-washing system, and were least likely to want to intervene

in lighting conditions at the lowest electric wall-washing system.

Across all sixteen default lighting conditions, a Rank-biserial correlation

indicated that participants’ intentions to change the lighting contrast on the window

wall reduced significantly when they ranked lower contrast between the window and

surrounding walls, rrb (845) = -0.235, p < 0.001. Additionally, participants’ intentions

to change the lighting contrast between the window and surrounding walls diminished

significantly when they ranked indoor lighting satisfaction higher, rrb (845) =

0.508, p < 0.001.


Figure 32 Percentage of participants who intended to change the luminance contrast on the window wall based on the luminaire power of the electric wall-washing system and the window-to-exterior-

wall ratios

4.4.7.6 Adjusting luminaire power level of the electric wall-washing system to achieve preferred luminance contrast on the window wall

Participants were asked to set their preferred lighting contrast on the window

wall in the IVR office rooms with different window-to-exterior-wall ratios by

changing the luminaire power percentage of the electric wall-washing system.

Shapiro-Wilk's test showed that adjusted luminaire power percentages of the electric

linear luminaires were not normally distributed in rooms with a 15% WWR, a 30%

WWR, a 46% WWR, and a 62% WWR, (p < 0.001, p = 0.011, 0.035, 0.007,

respectively). Consequently, the median luminaire power percentage was considered

to be the best representative of the luminaire power percentage of the electric wall-

washing system when participants set the lighting contrast on the window wall to their

preferred level (Table 17). A Friedman test showed that there were no significant

differences in median adjusted luminaire power percentages of the electric wall-

washing system in the IVR office room between groups that differed in their window-

to-exterior-wall ratios, 2(3) = 1.529, p = 0.676. There was also no correlation between

the luminaire power percentage of the electric linear luminaires and the WWR in the

IVR office space, rs (210) = 0.082, p = 0.241. Table 17 indicates that participants used

approximately one-third of the power level of the electric linear luminaires while

setting the lighting contrast between the window frame and surrounding walls to their

preferred level under different lighting conditions. This luminaire power level is

similar to the group of “low power level” in the default lighting conditions.


15% WWR 30% WWR 46% WWR 62% WWRMedian preferredluminaire power

percentage of the electriclinear luminaires

29% 33% 38.5% 37.5%

Table 17 Median luminaire power percentage of the electric wall-washing system in rooms with

different window-to-exterior-wall ratios while setting preferred luminance contrast on the window

wall

4.4.7.7 Adjusting luminaire power level of the electric wall-washing system to achieve minimum acceptable luminance contrast on the window wall

Participants were told to set the lighting contrast between the window frame and

surrounding walls to the minimum acceptable level in the IVR office spaces with

different window-to-exterior-wall ratios using the supplementary electric wall-

washing system. Shapiro-Wilk's test showed that adjusted luminaire power level of the

electric linear luminaires was not normally distributed, p < 0.05. Table 18 presents the

median luminaire power percentage of the electric wall-washing system when

participants set their minimum acceptable level. There were no significant differences

in median adjusted luminaire power percentages of the electric wall-washing system

between groups that differed in their window-to-exterior-wall ratios (Kruskal-Wallis:

p = 0.257). Spearman's rank-order correlation test indicated a significant weak,

positive correlation between the luminaire power percentage of the electric linear

luminaires and the WWR in the IVR office space, rs (205) = 0.225, p = 0.001. Table

18 indicates that participants used approximately one-third of the power level of the

electric linear luminaires while setting the lighting contrast between the window frame

and surrounding walls to the minimum acceptable level in the IVR office space with a

15% WWR. However, participants used around one-fifth of the power level of the

electric wall-washing system while setting the lighting contrast on the window wall to

the minimum acceptable level in rooms with more than 15% window-to-exterior-wall

ratios.

15% WWR 30% WWR 46% WWR 62% WWRMedian minimum

acceptable luminaire power percentage of the electric

linear luminaires

29% 18% 20.5% 21.5%


Table 18 Median luminaire power percentage of the electric wall-washing system in rooms with

different window-to-exterior-wall ratios while setting minimum acceptable luminance contrast on the

window wall

After setting the lighting contrast on the window wall to the minimum acceptable

level in rooms with different window-to-exterior-wall ratios using electric linear

luminaires, participants ranked the lighting contrast between the window frame and

surrounding walls on a scale of one to five (one meaning very low contrast and five

meaning very high contrast). A Friedman test was run to determine if there were

differences in RC scores on the window wall between the four groups of window-to-

exterior-wall ratios: "15% WWR", "30% WWR", "46% WWR" and "62% WWR".

Pairwise comparisons were performed with a Bonferroni correction for multiple

comparisons. The RC scores on the window wall were statistically significantly

different between the four groups of window-to-exterior- 2(3) =

15.081, p = 0.002. Post hoc analysis revealed a statistically significant difference in

RC scores between the IVR office room with a 15% WWR (M = 2.95) and 62% WWR

(M = 2.32) (p = 0.018), but not between the other group combinations. There was also

a significant negative correlation between the RC scores on the window wall and the

window-to-exterior-wall ratios (Rank-biserial correlation: rrb (209) = -0.202, p =

0.003). Taken together, the mean subjective scores of the lighting contrast on the

window wall were somewhat low contrast or neither low nor high contrast after setting

the lighting contrast between the window and surrounding walls to the minimum

acceptable level in the IVR office room with different window-to-exterior-wall ratios.

Participants’ ranking for indoor lighting level satisfaction were statistically

significantly different between the four groups of window-to-exterior-wall ratios 2(3) = 57.66, p < 0.001). Post hoc analysis identified statistically

significant differences in indoor lighting level satisfaction scores between the IVR

office rooms with a 15% WWR (M = 2.66) and 46% WWR (M = 3.77) or 62% WWR

(M = 4.02) (p < 0.001, 0.001). There was also a significant difference in indoor lighting

level satisfaction scores between the rooms with a 30% WWR (M = 3.35) and 62%

WWR (p = 0.001). A Rank-biserial correlation identified a statistically significant

positive correlation between indoor lighting level satisfaction scores and the four

groups of window-to-exterior-wall ratios (rrb (209) = 0.475, p < 0.001). In summary,

increasing the WWR when participants set the lighting contrast on the window wall to


the minimum acceptable level significantly increased indoor lighting satisfaction for

performing work in the office.

4.4.8 Discussion

The main aim of this study was to explore the impact of the WWR and the

supplementary electric wall-washing system on the window wall RC scores, indoor

lighting satisfaction, and intention to change the lighting conditions in the IVR office

rooms with four different window-to-exterior-wall ratios. The second objective of this

research was to explore the usefulness of the IVR techniques in determining subjective

responses to a range of lighting conditions. The introduction of the electric lighting

system with a low power level did significantly reduce participants’ intentions to

change the lighting contrast on the window wall. The results indicated that the WWR

did not significantly affect the RC scores on the window wall, which is not consistent

with what we expected. Furthermore, the fact that there was a trend towards a higher

rating for the 15% WWR could be due to the fact that the window is concentrated on

a tiny proportion of the wall in that room which leads to reduced daylight penetration

resulting in lower horizontal illuminance and higher perception of overall contrast in

the presented scene.

The introduction of the electric linear luminaires with a low power level did

effectively reduce participants’ RC scores on the window wall in the IVR office room

with a 46% WWR. Additionally, participants reported somewhat higher RC on the

window wall in the room with a 30% WWR when using the electric lighting system

with a low power level compared with when it was off. However, they adjusted the

luminaire power percentage of this system to roughly 33% and 18% while setting their

preferred and minimum acceptable contrast on the window wall, respectively. These

relationships may partly be explained by the fact that there was no significant

difference between users’ RC scores on the window wall in the room with 30% WWR

when the lighting system was off compared with when it was on with low power level

during default lighting conditions. Taken together, it is possible that using the electric

linear luminaires with a low power level in a room with a 30% WWR could reduce the

RC scores on the window wall.

Another significant finding was that the supplementary electric lighting system

was not effective in reducing the RC scores in the room with a 15% WWR.


Furthermore, the users’ RC scores between the window and surrounding walls

increased when the luminaire power percentage of the electric linear luminaires was

increased (to the medium or the high power percentage) during the sixteen default

lighting scenarios. These outcomes contradict the aim of the supplementary wall-

washing lighting system, which was to reduce RC scores between the window and the

surrounding walls. There are a number of possibilities for this contradictory finding.

The first possibility is that participants might misinterpret the different evaluations of

the RC scores between the window and surroundings. Although rephrasing the

question about rating the contrast on the window wall might reduce this effect, it may

not eliminate its impact. It may also be difficult to convey exactly which part of the

scene the participants are evaluating. The second possibility is that the frame around

the window appears quite dark in the IVR office rooms, therefore, this difference

between the window and the frame might lead to a localised contrast between the frame

and the immediate surrounds. It is probable that using a lighter coloured frame could

have changed the results and reduced the contrast between the lighting immediately

surrounding the frame and the frame itself. However, the findings of this study do not

entirely contradict the findings of previous research investigating the impact of the

proposed wall-washing system on the window appearance, which found that a

supplementary wall-washing strategy with a low power level was the most comfortable

(Amirkhani et al., 2015b, 2015a, 2016; Amirkhani et al., 2017). Overall, this study

highlights the fact that IVR settings might be useful for examining some outcomes and

not others, and that researchers may need to carefully evaluate different aspects of the

IVR built environment they produce and consider unintended perceptual

consequences.

A limitation associated with the IVR environments is an accurate representation

of lighting with the current technology. For example, while the maximum output of

the virtual reality screen in this study is 855 cd/m² (Soneira, 2018), the minimum

average window luminance in our previous study in a real office room with no direct

sunlight under clear sky condition was 2192 cd/m² (Amirkhani et al., 2015a).

Furthermore, to increase the realism of lighting in the IVR space, tone-mapping

techniques, such as Adaptive Reinhard, could be used in future research (Murdoch,

Stokkermans, & Lambooij, 2015). However, according to Chamilothori et al. (2018),

the current tone-mapping operators are static, whereas the content and contrast of


scene in the IVR spaces changes with the users’ head movement. Subsequently, further

research is needed to assess various tone-mapping techniques or other configurations

on Samsung Gear virtual reality HMD and other head-mounted displays. Overall, all

the RC evaluations are relative to the indoor virtual environment setting and the

equipment used in this research, and they cannot be necessarily generalized in the real

conditions.

Our focus in this study was on the impact of the electric linear luminaires and

the window size on participants’ intention to intervene in lighting conditions using the

IVR technology. According to the literature review, the limitations of IVR spaces

might not have a significant impact on participants’ lighting behaviour. However,

several factors such as visual sensations, testing conditions, many psychological

variables and individual variations can influence human lighting interventions

throughout a day within the real spaces (Nazzal, 2005). Therefore, the outcomes of

this study can only represent the initial influence of the proposed wall-washing system

and the WWR on participants’ intention to change the lighting conditions. Further

study should be carried out to use our findings in this research and conduct a field

experiments to better understand the impact of such influences caused by the proposed

wall-washing system and the WWR on participants’ behaviour over a longer period.

The results of this experiment showed that higher luminaire power of the

supplementary wall-washing system led to higher RC scores on the window wall;

however, it enhanced indoor lighting satisfaction. It may be because the RC

evaluations did not drive evaluations of satisfaction. Nonetheless, although the specific

contrast ratings as induced by the supplementary wall-washing system did not result

in expected results; other outcomes were in line with expectations. For instance,

participants’ RC scores between the window and the surrounding walls reduced

through increasing the WWR in the IVR office spaces. Participants were also

increasingly satisfied with a greater power level of the electric wall-washing system,

even though this was not reflected in their RC scores on the window wall. The results

of this study build on previous research in a number of ways. They suggest that there

is no interaction between the power level of the proposed electric wall-washing system

and the WWR. They also suggest that using the IVR spaces might be a useful tool for

evaluating overall room design and evaluations of satisfaction and behaviour, but that


evaluations of more specific perceptual experiences (such as contrast) may need to be

examined with care.

4.4.9 Conclusion

An integrated lighting design solution to enhance window appearance that

potentially leads to reducing the energy consumption of buildings is introduced in this

paper. Supplementary electric linear luminaires were mounted surrounding the

window frame in a single IVR office space with different window-to-exterior-wall

ratios (see Figure 25). The supplementary electric lighting system was designed to

illuminate the walls surrounding the window to reduce the lighting contrast between

the window frame and surrounding walls. The system discussed in this paper

demonstrated an effective design solution to reduce subjective ratings of the RC on the

window wall. This study indicated that the electric lighting system with a low power

percentage could reduce the RC scores on the window wall in the rooms with a 30%

WWR, a 46% WWR, and a 62% WWR; however, this system was not effective in the

room with a 15% WWR (which was less desirable during all lighting conditions). The

results of this study also showed that an electric lighting system with a low power level

could reduce participants’ inclinations to change the lighting contrast on the window

wall in rooms with different window-to-exterior-wall ratios by about 9%. Finally, the

outcomes of this study indicate that identifying perceived lighting contrast could be

best investigated through asking participants “whether they want to change the contrast

between the window and the surrounding walls.” This approach could significantly

minimise misinterpretations about visual discomfort and its evaluations during lighting

research.

Future research should be undertaken to explore whether there are optimal

luminance contrasts between the window and surrounding walls using the electric

lighting system to enhance window appearance. Daylight is dynamic and changes in

direction, spectrum, and intensity as the time and weather change; therefore, further

research might explore linking supplementary electric lighting systems to photosensor-

based controls, which modify the lighting contrast on the window wall throughout the

day and year.

130 Part 5: General Discussion and Conclusion

General Discussion and Conclusion

5.1 General discussion

The current PhD thesis provides an innovative integrated LED lighting design

solution for better acceptance of window appearance in office buildings through

reducing the LC on the window wall. The main aim of this PhD study was to examine

the effectiveness of a retrofittable LED wall-washing system to diminish LC between

the window, as the source of daylight, and surroundings. We proposed that the

introduction of the LED wall-washing system could improve the window appearance

in a way that could reduce users’ intentions to intervene in lighting conditions and that

therefore could lead to savings in the electricity consumption of buildings.

A series of experiments (the pilot study, Experiment 1, and Experiment 2) took

place in actual (real) typical office rooms with different window-to-external-wall ratios

and orientations. The introduction of the LED wall-washing system with low power

level in these rooms did efficiently reduce the LC on the window wall. Table 19

illustrates the LC reduction during each experiment using the proposed LED wall-

washing system with low power level. The outcomes of the pilot study (Chapter 4.1)

also suggest that increased electricity consumption of an approximately 18 W (low

power level) LED wall-washing system is offset where there is roughly a one-quarter

reduction in users’ intentions to intervene in lighting conditions.

Room image when the LED wall-washing is on

Experiment Number

WWR LC reduction on the window wall Orientation From to

Pilot study 15% 215:1 45:1Southwest

Experiment 1 30% 117:1 33:1Southwest

Chapter 6: General Discussion and Conclusion 131

Experiment 2 45% 16:1 9:1Northwest

Table 19 LC reduction on the window wall in real office rooms with different window sizes using the

proposed LED wall-washing system with low power level

These reductions in LC in the rooms with a 15% WWR and a 30% WWR did

efficiently reduce the participants’ reports of discomfort glare from the window wall

and improved indoor lighting satisfaction. They also significantly diminished the

participants’ propensity to change the lighting conditions. However, the LC reduction

in the room with a 45% WWR was not reflected in the participants’ perceptions of

glare, satisfaction with the lighting in the room, or intentions to move the blinds or

turn on the ceiling lights, and there were no significant differences between lighting

conditions with these measures. The fact that the LC in this room (Chapter 4.3) was

not significantly related to perceptions of glare or satisfaction may explain the lack of

the effects of subjective measures. It is possible that this occurred because the LC on

the window wall in the room with a 45% WWR before any lighting intervention was

not particularly high (with an approximately 16:1 ratio). Our experiments in rooms

with a 15% and 30% % window-to-external-wall ratios, using a supplementary LED

system, found that more dramatic reductions in LC could be achieved using a similarly

low-powered system. Therefore, it is possible that this system might be useful in high

LC offices, but may have limited effects on occupant ratings in lower LC offices.

Furthermore, the outcomes of experiment 2 (Chapter 4.3) suggest that it is

possible that an LC of approximately 11:1 to 12:1 between the window and the

surrounding walls may be an ideal ratio to achieve visual comfort via the proposed

LED wall-washing system, as this is the ratio that occurred when participants were

able to adjust the lighting system themselves. The results of the experiments (the pilot

study, Experiment 1, and Experiment 2) in real office rooms also illustrate that if the

building occupants do not experience glare, their intentions to switch on the ceiling

lights or move the blind reduce significantly.


As stated earlier, the primary aim of this study was to test the impact of using

the proposed LED wall-washing system on participants’ scores for visual comfort and

their propensity to intervene in lighting conditions. While it is possible to perform such

experiments in existing buildings, several factors might influence the results (e.g.,

WWR, the reflectance of inner surfaces, different interior space designs, cloudy/sunny

weather on different days, different outside views, different internal brightness, etc.).

These factors, which in some cases are not possible to control, could cause

experimental noise or affect the outcomes. Accordingly, the last experiment of the

current PhD research was conducted in IVR environments with four different window

sizes (approximately 15%, 30%, 46% and 62% window-to-external-wall ratios). Using

IVR technology allowed us to manipulate the control variables quickly while keeping

other design features constant. However, it is hard to make the perceptual experience

in the IVR settings the same as that in an actual physical environment. The primary

objective of Experiment 3 was to explore the impact of the WWR and the

supplementary electric wall-washing system on the window wall rated contrast (RC)

scores, indoor lighting satisfaction, and intention to change the lighting conditions in

the IVR office rooms with four different window-to-exterior-wall ratios.

Supplementary electric linear luminaires were mounted surrounding the window frame

in the single IVR office spaces with different window-to-exterior-wall ratios (see

Figure 7). The outcomes of this study indicate that the introduction of the electric

lighting system with a low power level did significantly reduce participants’ intentions

to change the lighting contrast on the window wall. Furthermore, the results of this

study indicate that the WWR did not significantly affect the RC scores on the window

wall. The fact that there was a trend towards a higher rating for the 15% WWR could

also be due to the fact that the window is concentrated in a tiny proportion of the wall

in that room which leads to reduced daylight penetration resulting in lower horizontal

illuminance and higher perception of overall contrast in the presented scene.

One source of weakness in the pilot study (Chapter 4.1) and Experiment 1

(Chapter 4.2), which could have affected the measurements of lighting, was that they

were not conducted randomly. Large randomised controlled trials could provide

evidence that is more definitive. Future research should also assess whether the wall-

washing approach described in this study could improve both the observed luminance

contrasts and subjective ratings in different real room conditions. Additionally, the


proposed LED wall-washing system could also be tested in actual physical rooms with

different room layouts, orientations and window types (e.g., punch window type and

strip window type), to determine whether it effectively reduces luminance contrasts in

those settings and whether changes in occupant perceptions accompany that reduction.

More investigation is also needed to examine the impact of participant lighting

preferences on the acceptable LC between the window and surrounding walls using

the proposed LED wall-washing system.

An implication of Experiments 1 and 2 might be not applying the vignetting

correction of the HDR images. The vignetting effect is the light falloff that could be

seen toward the edges of a picture, in particular when a fisheye lens is used (Reinhard

et al., 2010). The vignetting effect, which relies on the aperture of the lens (large

apertures yielding more vignetting than small ones), could be as high as a 70%

luminance loss at the periphery of the fisheye images (Cauwerts, PhD, & Deneyer,

2012; Pierson, Jacobs, Wienold, & Bodart, 2017). According to Pierson et al. (2017),

the vignetting correction of an HDR image can be achieved by applying a digital filter

on the fisheye image. It can be done by applying a .cal file to the HDR image using

the pcomb command of Radience. As the result of that, each pixel of the HDR fisheye

picture is divided based on its radial position, by the right vignetting function (Pierson

et al., 2017). In future investigations using HDR images, it is better to apply vignetting

correction to avoid the luminance loss at the periphery of the fisheye HDR images.

Despite the many advantages of using shading devices to limit discomfort glare,

the retrofitting of shading devices to older buildings is not always feasible.

Additionally, shading devices can hinder occupant view and connection to the

outdoors (which occupants typically value highly) and are not as easily adaptable to

changing lighting conditions. It is, therefore, worthwhile to explore supplementary

electric light strategies, such as that presented here, to reduce the negative impact of

bright windows. A distinct advantage of the tested LED system is not only its low cost

and retrofittable nature but also its customizability to suit user preferences and

changing outdoor lighting conditions. Although this study provides a suggested

starting point for the optimal window-wall LC, one benefit of such systems is that they

are easily adaptable, and the preferred power setting can be set by the user to ensure

personal visual comfort. Such a system would integrate easily with the Internet of

Things technologies, and its easy customizability means that power levels could vary


automatically across the day and in-line with user preferences. Therefore, it could be

considered as an additional option for lighting design that minimizes discomfort glare

while maximizing energy savings. As LED technology advances, increases in the

amount of light produced per unit power input (i.e., luminous efficacy) will also enable

a reduction in the power required to generate the beneficial effects seen from these

supplementary lighting systems. The system demonstrated in this paper showed an

effective design solution using an 18 W LED product; however, the same positive

effect can be achieved with a lower power solution by using purpose-designed, higher-

efficiency LED products.

Furthermore, a limitation associated with the IVR environments is an accurate

representation of lighting with the current technology. For example, while the

maximum output of the virtual reality screen in this study is 855 cd/m² (Soneira, 2018),

the minimum average window luminance in our previous study in a real office room

with no direct sunlight under clear sky condition was 2192 cd/m² (Amirkhani et al.,

2015a). Furthermore, to increase the realism of lighting in the IVR space, tone-

mapping techniques, such as Adaptive Reinhard, could be used in future research

(Murdoch et al., 2015). However, according to Chamilothori et al. (2018), the current

tone-mapping operators are static, whereas the content and contrast of scene in the IVR

spaces changes with the users’ head movement. Subsequently, further research is

needed to assess various tone-mapping techniques or other configurations on Samsung

Gear virtual reality HMD and other head-mounted displays. Overall, all the RC

evaluations are relative to the indoor virtual environment setting and the equipment

used in the last experiment (Chapter 4.4), and they cannot be necessarily generalized

in the real conditions.

Our focus in the last experiment (Chapter 4.4) was on the impact of the electric

linear luminaires and the window size on participants’ intention to intervene in lighting

conditions using the IVR technology. According to the literature review, the

limitations of IVR spaces might not have a significant impact on participants’ lighting

behaviour. However, several factors such as visual sensations, testing conditions,

many psychological variables and individual variations can influence human lighting

interventions throughout a day within the real spaces (Nazzal, 2005). Therefore, the

outcomes of this study can only represent the initial influence of the proposed wall-

washing system and the WWR on participants’ intention to change the lighting


conditions. Further study should be carried out to use our findings in this research and

conduct a field experiments to better understand the impact of such influences caused

by the proposed wall-washing system and the WWR on participants’ behaviour over a

longer period.

The results of the last experiment (Chapter 4.4) showed that higher luminaire

power of the supplementary wall-washing system led to higher RC scores on the

window wall; however, it enhanced indoor lighting satisfaction. It may be because the

RC evaluations did not drive evaluations of satisfaction. Nonetheless, although the

specific contrast ratings as induced by the supplementary wall-washing system did not

result in expected results; other outcomes were in line with expectations. For instance,

participants’ RC scores between the window and the surrounding walls reduced

through increasing the WWR in the IVR office spaces. Participants were also

increasingly satisfied with a greater power level of the electric wall-washing system,

even though this was not reflected in their RC scores on the window wall. The results

of this study build on previous research in a number of ways. They suggest that there

is no interaction between the power level of the proposed electric wall-washing system

and the WWR. They also suggest that using the IVR spaces might be a useful tool for

evaluating overall room design and evaluations of satisfaction and behaviour, but that

evaluations of more specific perceptual experiences (such as contrast) may need to be

examined with care.

5.2 General conclusion

An integrated lighting design solution to improve window appearance, which

potentially leads to increased energy savings of buildings, is introduced in this PhD

research. A supplementary LED wall-washing system was mounted surrounding the

window in actual physical office rooms in Brisbane, Australia with different window

sizes and orientations (see Figure 6). This LED wall-washing system was designed to

illuminate the walls surrounding the window to decrease the LC on the window wall.

The results of these studies indicate that the system was effective in significantly

reducing the LC on the window wall in actual office rooms with a 15% and 30%

window-to-exterior-wall ratios (see Table 19). However, the LC reduction in the room

with a 45% WWR was not dramatic. The LED wall-washing strategy reduced the LC

on the window wall in this room to below 11:1 without significant changes in

horizontal illuminance. The LC reduction in rooms with a 15% WWR (Chapter 4.1)


and a 30% WWR (Chapter 4.2) could significantly enhance the participants’ scores for

the window appearance or their indoor lighting satisfaction. Nonetheless, reducing the

LC on the window wall in the room with a 45% WWR (Chapter 4.3) did not reduce

the participants’ ratings of glare or dissatisfaction with indoor lighting. The second

experiment (Chapter 4.4) also explored the acceptable LC on the windowed wall using

the proposed LED wall-washing system. The results suggest that a LC of

approximately 11:1 to 12:1 between the window, as the source of daylight, and the

surrounding walls may be an ideal ratio to achieve visual comfort via the proposed

LED wall-washing system.

The last experiment (Chapter 4.4) was conducted in IVR office rooms with

different window-to-external-wall ratios (see Figure 7). This study indicated that the

proposed electric wall-washing system with a low power percentage could reduce the

participants’ RC scores on the window wall in the rooms with a 30% WWR, a 46%

WWR, and a 62% WWR; however, this system was not effective in the room with a

15% WWR (which was less desirable during all lighting conditions). The results of

this study also showed that an electric lighting system with a low power level could

significantly reduce participants’ propensity to intervene in lighting conditions.

Finally, the outcomes of this study indicate that identifying perceived lighting contrast

on the window wall could be best investigated through asking participants “whether

they want to change the contrast between the window and surroundings.”

The outcomes of the experiments during the current PhD research, when taken

together, demonstrate that the proposed LED wall-washing system with low power

level does efficiently mitigate problematic interventions in lighting conditions that

lead to increased energy consumption in buildings. The benefit of using such a

supplementary LED wall-washing system is that it can be fitted into existing and future

buildings with minimal construction modifications and at a low cost. Overall, this PhD

research indicates a significant and original contribution to knowledge in the field of

window design in architecture and discomfort glare research. It enhances our

understanding of an integrated lighting design solution for better acceptance of the

window appearance that could increase energy savings in office buildings.

Future research should investigate whether this system can match reductions in

LC with changes in subjective ratings, particularly in higher LC environments, to

determine whether there are optimal luminance contrasts to achieve occupant


satisfaction. As daylight is dynamic and changes in intensity, spectrum, and direction

as the time and weather change, future research could also incorporate linking

supplementary lighting systems to photosensor-based controls that modify window

appearance throughout the day and year. Finally, the proposed system would be

significantly more energy efficient than the use of overhead lighting; however, the

energy savings achievable are contingent on the design of the supplementary system

and the overhead lighting strategies they seek to replace. The optimization of the

product design of supplementary LED wall-washing systems is required to maximize

the energy savings available from this design strategy.

138 Appendices

Appendices

Appendix A Lighting evaluation metrics and simulation tools

Currently, there are several metrics and criteria to evaluate daylight parameters,

such as daylight factor (DF), useful daylight illuminance (UDI), daylight autonomy

(DA), continuous daylight autonomy (DAcon), and maximum daylight autonomy

(DAmax) for analysing illuminance (Reinhart, Mardaljevic, & Rogers, 2006). There are

also some metrics to analyse glare parameters, including daylight glare probability

(DGP), simplified daylight glare probability (DGPs), daylight glare index (DGI),

useful glare rating (UGR), visual comfort probability (VCP), CIE glare index (CGI)

for assessing glare (Suk et al., 2013). Table 20 illustrates some of the current metrics

and criteria that are used to evaluate daylight quality.

Table 20 Some of the existing metrics and criteria that are used to analyse daylight quality

Daylight parameter

Metric Criteria Source

Illuminance DF DF 5% or more: The room looks cheerfully lit DF 2-5%: The occupants are likely to use electric

light as a supplement.

Bean (2012)

UDI UDI<100 lx is considered insufficient; 100-500 lx is seen as effective;

500-2000 lx is frequently perceived desirable or tolerable;

UDI>2000 lx is often probable to produce thermal and/ or discomfort glare.

Nabil and Mardaljevic

(2006)

DA Necessary minimum illuminance can be taken from reference documents like the IESNA Lighting

Handbook.

IESNA and Rea (2000)

Distribution Luminance in the field

of view

According to the latest findings, 4000 cd/m2 can be a reasonable threshold for acceptable glare.

Suk et al. (2013)

Glare imperceptible perceptible disturbing intolerable Suk et al. (2013) DGP <0.35 0.35-0.40 0.40-0.45 >0.45

DGI <18 18-24 24-31 >31 VCP 80-100 60-80 40-60 <40 UGR <13 13-22 22-28 >28 CGI <13 13-22 22-28 >28

Directivity Ratio of vector to

scalar illuminance

(Ev/Es)

Acceptable value is in the range of 1.2-1.8 on a scale of 0-4.

Cuttle (2008)

Altitude of illuminance

vector

Cuttle (2008)

Appendices 139

A.1 Evaluating daylight performance inside buildings

The need to have broadly accepted daylighting performance metrics and criteria

which can be referenced has been widely recognised (Reinhart et al., 2006;

Mardaljevic et al., 2009). Thus, the conceptual design of daylighting metrics has been

substantially extended since 2000 (Saxena, Heschong, Wymelenberg, Wayland, &

Analytics, 2010). The following sections describe the most commonly used metrics to

evaluate daylight performance, including DF, DA, and UDI.

A.1.1 Daylight factor

The DF was first proposed in the early 1900s and formalised into building

standards more than 50 years ago (Mardaljevic et al., 2009). It illustrates the ratio of

illuminance at a particular point in the room as a percentage of the illuminance from

the entire unobstructed sky (Bean, 2012).

= indoor illuminanceoutdoor illuminance × 100%

DF is based on the overcast sky condition, where an alteration of orientation

would not influence the internal illuminance (Alshaibani, 2015). Also, DF is still one

of the most commonly used metrics to quantify overall diffuse light (e.g., all skylight

and diffuse-reflected sunlight) within a space (Reinhart et al., 2010; Alshaibani, 2015).

However, Mardaljevic et al. (2009) hold the view that this method does not include the

contribution from sunlight, whereas many practitioners try to comprehend the pattern

of daylight in a room through studying a dynamic solar shading analysis, or the sun-

path diagram. Mardaljevic et al. (2009) further state that the DF only provides some

insight into how sunlight will be distributed in a room and should be integrated with

local weather variation patterns and reflectance of materials, to achieve a better

inference of indoor visual quality. Finally, Cantin and Dubois (2011) suggest that the

UDI should replace the DF for the study of illuminance, and consider that using DF

alone could lead to selecting excessive glazing areas in buildings.

A.1.2 Daylight autonomy

DA is defined as the annual percentage of occupied times that a room can expect

to achieve a minimum target illuminance level on the working plane (EFA, 2014). This

metric includes the influence of a variety of variables (e.g., climate, room geometry,

window size and proportion) and can also be used to forecast electric energy savings

140 Appendices

of buildings if an on/off lighting control system is to be used (Tzempelikos &

Athienitis, 2007). Moreover, according to Saxena et al. (2010), DA can be useful to

define a point to be day-lit if the DA exceeds 50% of the annual occupied times.

Nonetheless, the limitation of DA is that it excludes illuminance levels somewhat

below the threshold and it does not consider the issue of glare because of extreme

daylight (Shen & Tzempelikos, 2012). Therefore, DA should be replaced by other

metrics such as UDI (Shen & Tzempelikos, 2012).

A.1.3 Useful daylight illuminance

UDI mostly resembles DA, though UDI defines lower and upper illuminance

thresholds for sunlight to be useful (Reinhart & Weissman, 2012). UDI is a climate-

based metric that was introduced by Nabil and Mardaljevic (2005). The term climate-

based has been used because the (hourly) sun, and sky conditions are founded on

values from climate datasets in a year (Nabil & Mardaljevic, 2006). UDI is defined as

the annual existence of illuminances across the work plane under day-lit conditions

that are within a range believed useful by building users (Mardaljevic et al., 2009;

CUNDALL, 2014). The useful range of illuminance level in UDI falls within 100-

2000 lx (Nabil & Mardaljevic, 2005). Moreover, Nabil and Mardaljevic (2005) claim

that the UDI offers a framework to interpret daylighting based both on realistic

measures of absolute illuminance and on realistic models for building users’

behaviour. However, as sunlight can be used up to as high as 5000 lx depending on

the activities taking place, there is some debate about the best range to use for UDI

(EFA, 2014).

A.2 Evaluating discomfort glare from windows

Wienold and Christoffersen (2006) hold the view that there is no reliable tool or

descriptor for evaluating discomfort glare from windows. Moreover, Nazzal (2005)

notes that most of the evaluation methods of discomfort glare from daylight only

consider the horizontal illuminances, but this is not enough for occupants’ comfort. He

also claims that all current glare indices are based on uniform sources of light and

should consequently not be applied when discomfort glare is caused by sunlight from

a non-uniform source of light. Moreover, Chauvel et al. (1982) argue that discomfort

glare from windows is independent of the window distance from the observer and its

size. Nonetheless, Rodriquez and Pattini (2014) suggest that the size of windows and

Appendices 141

the position of observers have a significant influence on feeling discomfort glare from

windows. Furthermore, some studies recommend that interesting outside views can

increase occupants’ tolerance levels of discomfort glare (Tuaycharoen & Tregenza,

2007). Thus, glare indices that can be used for electric lighting conditions are not

appropriate for daylight situations (Osterhaus, 2005). Likewise, as the risk of

experiencing glare reduces when indoor electric lights are on, it has been suggested to

evaluate discomfort glare from windows while electric lights are off in order to define

the worst-case conditions (Nazzal, 2005). The following sections describe the most

commonly used indexes to predict discomfort glare from windows, such as DGI and

DGP.

A.2.1 Daylight glare index

The most cited model or index to evaluate discomfort glare from daylight

sources is DGI that was developed by Hopkinson (Chauvel et al., 1982). The formula

of DGI is (Jakubiec & Reinhart, 2012):

= 10 × log 0.48 + ,. ,. + (0.07 ,. , )DGI can be calculated for a person facing the side of the wall or the window at

different distances from the window wall (Hopkinson, 1972). Nonetheless, the DGI

can only be used when there is a homogeneous luminance distribution in large areas

such as a uniform sky luminance view through a window (Aschehoug et al., 2000).

Moreover, research suggests that although DGI has been the standard for several years,

its application may lead to unreliable outcomes through overestimating discomfort

from daylighting systems (Nazzal, 2005; Osterhaus, 2005). However, even though the

limitations of DGI have been accepted, it is still the most widely used indicator for

daylighting systems (Wilks & Osterhaus, 2003).

A.2.2 Daylight glare probability

Another method to evaluate discomfort from daylighting systems is DGP, which

is a modification of DGI and has been developed by Wienold and Christoffersen

(2006). The formula of DGP is (Jakubiec & Reinhart, 2012):

= 5.87 × 10 + 9.18 × 10 × log 2 1 + , ,.

142 Appendices

Wienold and Christoffersen (2006) argue that DGP illustrates a robust

correlation with subjects’ response regarding glare perception. However, it should be

noted that DGP is only valid for values between 0.2 and 0.8 (Hirning et al., 2014). A

study comparing different indices of discomfort glare, such as DGI, UGR, CGI, VCP

and DGP through simulating a real and theoretical building in radiance suggested that

DGP outperformed the other glare indices, particularly when there was direct daylight

within the scene (Jakubiec & Reinhart, 2012). Overall, it should be noted that any

photometric measures can only be meaningful by how well they predict human

perception of glare (Nazzal, 2005).

A.3 Simulation programs to predict discomfort glare

A.3.1 Diva-for-Rhino

Diva-for-Rhino is an energy and daylighting modelling plug-in for Rhinoceros,

which was developed by the Graduate School of Design at Harvard University and is

currently distributed and optimised by Solemma LLC (2015). This plug-in can be used

to appraise the performance of buildings by obtaining photorealistic renderings, annual

and individual time step glare measurement, radiation maps, climate-based daylighting

metrics, single thermal zone energy and load calculation, and LEED and CHPS

daylighting compliance (Yun et al., 2014).

A.3.2 Evalglare

One method to calculate DGI and DGP inside buildings is through using

Evalglare and HDR images. Evalglare is a powerful glare analysis software that can

evaluate glare issues caused by electric lights and daylight through using various glare

metrics, such as DGI, DGP, unified glare rating (UGR), visual comfort probability

(VCP), and CIE glare index (CGI) (Suk & Schiler, 2013). Glare sources can be

automatically detected through using Evalglare based on a threshold value, which can

be (1) a fixed luminance value, (2) a luminance value that is x-times higher than the

mean luminance value of the entire image, or (3) a luminance value that is x-times

higher than the calculated mean luminance of a given zone (task area) (Wienold &

Christoffersen, 2006).

Appendices 143

Appendix B Questionnaire of the pilot study

144 Appendices

Appendix C Questionnaire of the first experiment

Appendices 145

Appendix D Questionnaire of the second experiment

146 Appendices

Appendices 147

Appendix E Questionnaire of the third experiment

148 Appendices

Appendices 149

150 Appendices

References 151

References

Afacan, Y., & Demirkan, H. (2016). The influence of sustainable design features on indoor environmental quality satisfaction in Turkish dwellings. Architectural Science Review, 59(3), 229-238. doi:10.1080/00038628.2015.1056768

Aghemo, C., Blaso, L., & Pellegrino, A. (2014). Building automation and control systems: A case study to evaluate the energy and environmental performances of a lighting control system in offices. Automation in Construction, 43, 10-22. doi:10.1016/j.autcon.2014.02.015

Ali, Z., & Bhaskar, S. B. (2016). Basic statistical tools in research and data analysis. Indian Journal of Anaesthesia, 60(9), 662-669. doi:10.4103/0019-5049.190623

Alrubaih, M. S., Zain, M. F. M., Alghoul, M. A., Ibrahim, N. L. N., Shameri, M. A., & Elayeb, O. (2013). Research and development on aspects of daylighting fundamentals. Renewable and Sustainable Energy Reviews, 21(0), 494-505. doi:http://dx.doi.org/10.1016/j.rser.2012.12.057

Alshaibani, K. (2015). Average daylight factor for the ISO/CIE Standard General Sky. Lighting Research and Technology. doi:10.1177/1477153515572939

Amirkhani, M., Garcia-Hansen, V., & Isoardi, G. (2015a). Improving the impact of luminance contrast on the window appearance in a conventional office room: using supplementary lighting strategies.Paper presented at the Living and Learning: Research for a Better Built Environment, 49th International Conference of the Architectural Science Association, Melbourne, Australia. http://anzasca.net/category/conference-papers/2015-conference-papers/

Amirkhani, M., Garcia-Hansen, V., & Isoardi, G. (2015b). LED lighting design strategies to enhance window appearance and increase energy savings in daylit office spaces. Paper presented at the Asia-Pacific Lighting Systems Workshop 2015, Sydney, Australia. sydney.edu.au/architecture/documents/research/wendy-conference.pdf

Amirkhani, M., Garcia-Hansen, V., & Isoardi, G. (2016). Reducing luminance contrast on the window wall and users' interventions in an office room. Paper presented at the CIE Lighting Quality & Energy Efficiency Conference, Melbourne, Australia.

Amirkhani, M., Garcia-Hansen, V., Isoardi, G., & Allan, A. (2017). An Energy Efficient Lighting Design Strategy to Enhance Visual Comfort in Offices with Windows. ENERGIES, 10(8), 1126. doi:10.3390/en10081126

Ander, G. D. (2003). Daylighting performance and design: John Wiley & Sons.

Arend, L., Buehler, J., & Lockhead, G. (1971). Difference information in brightness perception. Perception & Psychophysics, 9(3), 367-370.

Aries, M. B. C., Veitch, J. A., & Newsham, G. R. (2010). Windows, view, and office characteristics predict physical and psychological discomfort. Journal of Environmental Psychology, 30(4), 533-541. doi:http://dx.doi.org/10.1016/j.jenvp.2009.12.004

Arup, L., & Royal Institute of British Architects. (2007). Lighting: technical review. London: RIBA Publishing.

Arup, O. N., & Arup, L. (2007). Lighting: technical review. London: RIBA Publishing.

AS/NZS. (2008). AS/NZS 1680.2.2:2008 : Interior and workplace lighting - Part 2.2: Specific applications - Office and screen-based tasks. In.

Aschehoug, Ø., Christoffersen, J., Jakobiak, R., Johnsen, K., Lee, E., Ruck, N., . . . Community Systems, E. P. A. (2000). Daylight in Buildings: A Source Book on Daylighting Systems and Components. A Report of IEA SHC Task 21, ECBCS Annex 29: International Energy Agency, IEA, Solar Heating and Cooling Programme, SHC, Energy Conservation in buildings and Community Systems Programme, ECBCS.

Avery, D. H., Kizer, D., Bolte, M. A., & Hellekson, C. (2001). Bright light therapy of subsyndromal seasonal affective disorder in the workplace: morning vs. afternoon exposure. Acta Psychiatrica Scandinavica, 103(4), 267-274. doi:10.1034/j.1600-0447.2001.00078.x

152 References

Bailenson, J. N., & Yee, N. (2006). A longitudinal study of task performance, head movements, subjective report, simulator sickness, and transformed social interaction in collaborative virtual environments. PRESENCE: Teleoperators and Virtual Environments, 15(6), 699-716.

Baird, G., & Thompson, J. (2012). Lighting conditions in sustainable buildings: results of a survey of users' perceptions. Architectural Science Review, 55(2), 102-109. doi:10.1080/00038628.2012.667941

Bajaj, A., Rosner, B., Lockley, S. W., & Schernhammer, E. S. (2011). Validation of a light questionnaire with real-life photopic illuminance measurements: the Harvard Light Exposure Assessment questionnaire. Cancer Epidemiology Biomarkers & Prevention, 20(7), 1341-1349.

Barbara Sommer. (2006). Semantic differential Retrieved from http://psychology.ucdavis.edu/faculty_sites/sommerb/sommerdemo/scaling/semdiff.htm

Baron, R. A., Rea, M. S., & Daniels, S. G. (1992). Effects of indoor lighting (illuminance and spectral distribution) on the performance of cognitive tasks and interpersonal behaviors: The potential mediating role of positive affect. Motivation and Emotion, 16(1), 1-33.

Bean, R. (2012). Lighting: Interior and Exterior: Taylor and Francis.

Bellia, L., Bisegna, F., & Spada, G. (2011). Lighting in indoor environments: Visual and non-visual effects of light sources with different spectral power distributions. Building and Environment, 46(10), 1984-1992. doi:10.1016/j.buildenv.2011.04.007

Berman, S. (1992). Energy efficiency consequences of scotopic sensitivity. Journal of the Illuminating Engineering Society, 21(1), 3-14.

Bi, J. (2007). Similarity testing using paired comparison method. Food Quality and Preference, 18(3), 500-507. doi:10.1016/j.foodqual.2006.06.002

Boccia, A., Chella, F., & Zazzini, P. (2014). Natural light from a wall in buildings: Experimental analysis of the ventilated illuminating wall performances. Solar Energy, 108, 178-188. doi:10.1016/j.solener.2014.07.007

Böckenholt, U., & Tsai, R.-C. (2001). Individual differences in paired comparison data. British Journal of Mathematical and Statistical Psychology, 54(2), 265-277. doi:10.1348/000711001159555

Bodart, M., & De Herde, A. (2002). Global energy savings in offices buildings by the use of daylighting. Energy & Buildings, 34(5), 421-429. doi:10.1016/S0378-7788(01)00117-7

Borbély, Á., Sámson, Á., & Schanda, J. (2001). The concept of correlated colour temperature revisited. Color Research and Application, 26(6), 450-457. doi:10.1002/col.1065

Borisuit, A., Linhart, F., Scartezzini, J. L., & Münch, M. (2015). Effects of realistic office daylighting and electric lighting conditions on visual comfort, alertness and mood. Lighting Research and Technology, 47(2), 192-209. doi:10.1177/1477153514531518

Borisuit, A., Scartezzini, J.-L., & Thanachareonkit, A. (2010). Visual discomfort and glare rating assessment of integrated daylighting and electric lighting systems using HDR imaging techniques. Architectural Science Review, 53(4), 359-373. doi:10.3763/asre.2009.0094

Bosch-Sijtsema, P. M., & Haapamaki, J. (2014). Perceived enablers of 3D virtual environments for virtual team learning and innovation. COMPUTERS IN HUMAN BEHAVIOR, 37, 395-401. doi:10.1016/j.chb.2014.04.035

Boubekri, M. (1995). Appraisal of the lighting conditions in an office building: results of a survey. Indoor and Built Environment, 4(3-4), 162-169.

Boubekri, M., & Boyer, L. L. (1992). Effect of window size and sunlight presence on glare. Lighting Research & Technology, 24(2), 69-74.

Boubekri, M., Cheung, I. N., Reid, K. J., Wang, C.-H., & Zee, P. C. (2014). Impact of Windows and Daylight Exposure on Overall Health and Sleep Quality of Office Workers: A Case-Control Pilot Study. Journal of Clinical Sleep Medicine : JCSM : Official Publication of the American Academy of Sleep Medicine, 10(6), 603-611. doi:10.5664/jcsm.3780

Boyce, P. (2003). Human factors in lighting. New York: Taylor & Francis.

References 153

Boyce, P., & Cuttle, C. (1990). Effect of correlated colour temperature on the perception of interiors and colour discrimination performance. Lighting Research and Technology, 22(1), 19-36.

Boyce, P., Hunter, C., & Howlett, O. (2003). The benefits of daylight through windows. Retrieved from New York:

Brager, G., & Baker, L. (2009). Occupant satisfaction in mixed-mode buildings. Building Research & Information, 37(4), 369-380. doi:10.1080/09613210902899785

Brooks, A. L., Brahnam, S., & Jain, L. C. (2014). Technologies of Inclusive Well-Being : Serious Games, Alternative Realities, and Play Therapy (2014 ed. Vol. 536). Berlin/Heidelberg: Springer.

Bryman, A. (2004). Social research methods (Vol. 2nd). Oxford; New York: Oxford University Press.

BSI. (2011). BS EN 12464-1:2011: Light and lighting. Lighting of work places. Indoor work places. In: British Standards Institute.

Byrd, H. (2012). Post-occupancy evaluation of green buildings: the measured impact of over-glazing. Architectural Science Review, 55(3), 206-212. doi:10.1080/00038628.2012.688017

Cantin, F., & Dubois, M. C. (2011). Daylighting metrics based on illuminance, distribution, glare and directivity. Lighting Research & Technology, 43(3), 291-307. doi:10.1177/1477153510393319

Cauwerts, C., PhD, M. B., & Deneyer, A. (2012). Comparison of the vignetting effects of two identical fisheye lenses. LEUKOS, 8(3), 181-203.

Chamilothori, K., Wienold, J., & Andersen, M. (2018). Adequacy of Immersive Virtual Reality for the Perception of Daylit Spaces: Comparison of Real and Virtual Environments. LEUKOS, 1-24. doi:10.1080/15502724.2017.1404918

Chang, M.-H., Das, D., Varde, P. V., & Pecht, M. (2012). Light emitting diodes reliability review. Microelectronics Reliability, 52(5), 762-782. doi:http://dx.doi.org/10.1016/j.microrel.2011.07.063

Chauvel, P., Collins, J. B., Dogniaux, R., & Longmore, J. (1982). GLARE FROM WINDOWS: CURRENT VIEWS OF THE PROBLEM. Lighting Research & Technology, 14(1), 31-46.

Chung, T. M., & Burnett, J. (2000). Lighting Quality Surveys in Office Premises. Indoor+Built Environment, 9(6), 335-341. doi:10.1159/000057531

CIBSE, C. (1994). Code for interior lighting. London: The Chartered Institute of Building Services Engineers.

Clear, R. D. (2013). Discomfort glare: What do we actually know? Lighting Research & Technology, 45(2), 141-158. doi:10.1177/1477153512444527

Collins, B. L. (1975). Windows and people: a literature survey. Psychological reaction to environments with and without windows (NBS-BSS-70 United States GPO SD Cat. No. C13.29/2:70 HL English). Retrieved from

Corazza, A., Giorgi, S., & Massaro, V. (2012). Materials to Improve Performance of Discharge Lamps. Industry Applications, IEEE Transactions on, 48(5), 1668-1672. doi:10.1109/TIA.2012.2209189

Corney, D., Haynes, J.-D., Rees, G., & Lotto, R. B. (2009). The brightness of colour. PLoS ONE, 4(3), e5091. doi:10.1371/journal.pone.0005091

Corrodi, M., Corrodi, M., & Spechtenhauser, K. (2008). Illuminating: natural light in residential architecture (Vol. 3.). Basel; London: Birkhäuser.

Coyne, S., Isoardi, G., Luther, M., & Hirning, M. (2008). The use of high dynamic range luminance mapping in the assessment, understanding and defining of visual issues in post occupancy building assessments. Paper presented at the International conference on energy efficiency in commercial buildings (IEECB’08). Frankfurt, Germany, 10e11 April.

Crawford, M. H., & Crawford, M. H. (2009). LEDs for Solid-State Lighting: Performance Challenges and Recent Advances. IEEE Journal of Selected Topics in Quantum Electronics, 14; 15(4), 1028-1040. doi:10.1109/JSTQE.2009.2013476

154 References

CUNDALL. (2014). Baseline Design - Daylight Strategy. Retrieved from https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/388138/Baseline_designs_daylight_strategy_Dec_2014.pdf

Cuttle, C. (2008). Lighting by design. Amsterdam; London: Architectural.

De Almeida, A., Santos, B., Paolo, B., & Quicheron, M. (2014). Solid state lighting review – Potential and challenges in Europe. Renewable and Sustainable Energy Reviews, 34(0), 30-48. doi:http://dx.doi.org/10.1016/j.rser.2014.02.029

DEGUCHI, T., & SATO, M. (1992). The effect of color temperature of lighting sources on mental activity level. The Annals of physiological anthropology, 11(1), 37-43.

Department of Climate Change and Energy Efficiency. (2012). Baseline Energy Consumption and Greenhouse Gas Emissions in Commercial Buildings in Australia – Part 1. Retrieved from https://www.energy.gov.au/sites/g/files/net3411/f/baseline-energy-consumption-part_1-report-2012.pdf

Department of Industry and Science of the Australian Government. (2015). Energy-efficient lights. Retrieved from http://yourenergysavings.gov.au/energy/lighting/energy-efficient-lights

Dietz, T., Gardner, G. T., Gilligan, J., Stern, P. C., & Vandenbergh, M. P. (2009). Household Actions Can Provide a Behavioral Wedge to Rapidly Reduce US Carbon Emissions. Proceedings of the National Academy of Sciences of the United States of America, 106(44), 18452-18456. doi:10.1073/pnas.0908738106

Doulos, L., Tsangrassoulis, A., & Topalis, F. (2005). A critical review of simulation techniques for daylight responsive systems. Paper presented at the Proceedings of the European Conference on Dynamic Analysis, Simulation and Testing applied to the Energy and Environmental performance of buildings (DYNASTEE), Athens, Greece.

Doulos, L., Tsangrassoulis, A., & Topalis, F. (2008). Quantifying energy savings in daylight responsive systems: The role of dimming electronic ballasts. Energy & Buildings, 40(1), 36-50. doi:10.1016/j.enbuild.2007.01.019

Dubois, M.-C., Blomsterberg, Å., Lunds tekniska högskola, L. T. H., Institutionen för arkitektur och byggd, m., Energi och, B., Department of, a., . . . Faculty of Engineering, L. T. H. a. L. U. (2011). Energy saving potential and strategies for electric lighting in future North European, low energy office buildings: A literature review. Energy & Buildings, 43(10), 2572-2582. doi:10.1016/j.enbuild.2011.07.001

2007). Impact of lighting arrangements and illuminances on different impressions of a room. Building and Environment, 42(10), 3476-3482. doi:10.1016/j.buildenv.2006.10.048

Edwards, L., & Torcellini, P. A. (2002). A literature review of the effects of natural light on building occupants (NREL/TP-550-30769). Retrieved from Golden, Colorado:

EFA. (2014). EFA daylight design guide - Departmental advice. Retrieved from https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/388373/EFA_Daylight_design_guide.pdf

Eilers, M., Reed, J., & Works, T. (1996). Behavioral aspects of lighting and occupancy sensors in private offices: a case study of a university office building. ACEEE 1996 Summer Study on Energy Efficiency in Buildings.

Escuyer, S., & Fontoynont, M. (2001). Lighting controls: a field study of office workers' reactions. Lighting Research & Technology, 33(2), 77-94. doi:10.1177/136578280103300202

European commission. (2015). Energy-saving light bulbs. Retrieved from http://ec.europa.eu/energy/lumen/overview/avariedchoice/index_en.htm

European standard. (2011). Light and lighting: basic terms and criteria for specifying lighting requirements: Brussels: European committee for standardization.

Farley, K. M. J., & Veitch, J. A. (2001). A Room With A View: A Review of the Effects of Windows on Work and Well-Being (IRC-RR-136). Retrieved from Ottawa:

References 155

Fenton, L., & Moseley, H. (2014). UV emissions from low energy artificial light sources. Photodermatology, photoimmunology & photomedicine, 30(2-3), 153-159. doi:10.1111/phpp.12094

Field, A. P. (2014). Discovering statistics using IBM SPSS statistics: and sex and drugs and rock 'n' roll (Vol. 4thition.). London: SAGE.

Fies T.S., & Mathers M. (2009). The Basics of Efficient Lighting: A Reference Manual for Training in Efficient Lighting Principles. Retrieved from http://www.energyrating.gov.au/wp-content/uploads/2011/02/2009-ref-manual-lighting.pdf

Fisk, W. J., Black, D., & Brunner, G. (2011). Benefits and costs of improved IEQ in U.S. offices. Indoor Air, 21(5), 357-367. doi:10.1111/j.1600-0668.2011.00719.x

Fontoynont, M. (2002). Perceived performance of daylighting systems: lighting efficacy and agreeableness. Solar Energy, 73(2), 83-94. doi:10.1016/S0038-092X(02)00035-X

Foster, M., & Oreszczyn, T. (2001). Occupant control of passive systems: the use of Venetian blinds. Building and Environment, 36(2), 149-155. doi:10.1016/S0360-1323(99)00074-8

Fotios, S. (2017). A Revised Kruithof Graph Based on Empirical Data. LEUKOS, 13(1), 3-17. doi:10.1080/15502724.2016.1159137

Freewan, A. A., Shao, L., & Riffat, S. (2008). Optimizing performance of the lightshelf by modifying ceiling geometry in highly luminous climates. Solar Energy, 82(4), 343-353. doi:http://dx.doi.org/10.1016/j.solener.2007.08.003

Freewan, A. A. Y. (2014). Impact of external shading devices on thermal and daylighting performance of offices in hot climate regions. Solar Energy, 102, 14-30. doi:10.1016/j.solener.2014.01.009

Freihoefer, K. (2012). The relationship between sustainable indoor environmental quality (IEQ) and employees' satisfaction with their office environments. (Dissertation/Thesis), ProQuest Dissertations Publishing, Retrieved from https://conservancy.umn.edu/bitstream/handle/11299/128718/Freihoefer_umn_0130E_12653.pdf?sequence=1&isAllowed=y

Frontczak, M., & Wargocki, P. (2011). Literature survey on how different factors influence human comfort in indoor environments. Building and Environment, 46(4), 922-937. doi:http://dx.doi.org/10.1016/j.buildenv.2010.10.021

Fuchs, P. (2017). Virtual Reality Headsets - A Theoretical and Pragmatic Approach (1 ed.). Milton: CRC Press.

Gago, E. J., Muneer, T., Knez, M., & Köster, H. (2015). Natural light controls and guides in buildings.Energy saving for electrical lighting, reduction of cooling load. Renewable and Sustainable Energy Reviews, 41(0), 1-13. doi:http://dx.doi.org/10.1016/j.rser.2014.08.002

Galasiu, A. D., Newsham, G. R., Suvagau, C., & Sander, D. M. (2007). Energy saving lighting control systems for open-plan offices: a field study. LEUKOS, 4(1), 7-29.

Galasiu, A. D., & Veitch, J. A. (2006). Occupant preferences and satisfaction with the luminous environment and control systems in daylit offices: a literature review. Energy & Buildings, 38(7), 728-742. doi:10.1016/j.enbuild.2006.03.001

Garcia-Hansen, V., Isoardi, G., & Miller, E. (2010). Perceptions of daylight quality delivered by light transport systems. Paper presented at the CIE Conference Lighting Quality and Energy Efficiency, Vienna, Austria.

Garcia-Hansen, V. R. (2006). Innovative daylighting systems for deep-plan commercial buildings.(Dissertation/Thesis), Queensland University of Technology Retrieved fromhttps://eprints.qut.edu.au/16709/1/Veronica_Hansen_Thesis.pdf

GBCA. (2015). What is Green Star? Retrieved from https://www.gbca.org.au/green-star/green-star-overview/

Gooch, A., & Willemsen, P. (2002, 2002). Evaluating space perception in NPR immersive environments.

Gordon, G. (2003). Interior lighting for designers (4th ed.). New York: Wiley.

156 References

Gratia, E., & De Herde, A. (2003). Design of low energy office buildings. Energy & Buildings, 35(5), 473-491. doi:10.1016/S0378-7788(02)00160-3

Greasley, P., & NetLibrary, I. (2008). Quantitative data analysis using SPSS: an introduction for health & social science. New York; Maidenhead: Open University Press Maidenhead ; New York.

Guan, L., Berrill, T., & Brown, R. J. (2015). Measurement of actual efficacy of compact fluorescent lamps (CFLs). Energy and Buildings, 86(0), 601-607. doi:http://dx.doi.org/10.1016/j.enbuild.2014.10.068

Hammad, F., & Abu-Hijleh, B. (2010). The energy savings potential of using dynamic external louvers in an office building. Energy and Buildings, 42(10), 1888-1895. doi:http://dx.doi.org/10.1016/j.enbuild.2010.05.024

Hannaford, B. (2002). The Society of Light and Lighting 2002. Code for Lighting: Lighting Research and Technology

Haq, M. A. u., Hassan, M. Y., Abdullah, H., Rahman, H. A., Abdullah, M. P., Hussin, F., & Said, D. M. (2014). A review on lighting control technologies in commercial buildings, their performance and affecting factors. Renewable and Sustainable Energy Reviews, 33(0), 268-279. doi:http://dx.doi.org/10.1016/j.rser.2014.01.090

Harrington, R. (1954). Effect of color temperature on apparent brightness. JOSA, 44(2), 113-116.

Harris, L. (1980). Office lighting, comfort and productivity–How the workers feel. Lighting Design Application, 10, 35-40.

Hawkes, R., Loe, D., & Rowlands, E. (1979). A note towards the understanding of lighting quality. Journal of the Illuminating Engineering Society, 8(2), 111-120.

Heerwagen, J. H., & Heerwagen, D. (1986). Lighting and psychological comfort. Lighting Design and Application, 16(4), 47-51.

Heinemann, E. G. (1961). The relation of apparent brightness to the threshold for differences in luminance. Journal of experimental psychology, 61(5), 389-399. doi:10.1037/h0047624

Heschong, L., Howlett, O., McHugh, J., & Pande, A. (2005). Sidelighting photocontrols field study.Retrieved from

Heschong, L., McHugh, J., Pande, A., Howlett, O., Reschke, E., & Sipp, B. (2006). Sidelighting Photocontrols Field Study (06-152). Retrieved from Heschong Mahone Group: Fair Oaks, CA.: http://neea.org/docs/reports/sidelightingphotocontrolsfieldstudyb430a8212041.pdf

Heydarian, A., Carneiro, J. P., Gerber, D., & Becerik-Gerber, B. (2015). Immersive virtual environments, understanding the impact of design features and occupant choice upon lighting for building performance. Building and Environment, 89, 217-228. doi:10.1016/j.buildenv.2015.02.038

Heydarian, A., Pantazis, E., Carneiro, J. P., Gerber, D., & Becerik-Gerber, B. (2016). Lights, building, action: Impact of default lighting settings on occupant behaviour. Journal of Environmental Psychology, 48, 212-223. doi:10.1016/j.jenvp.2016.11.001

Heydarian, A., Pantazis, E., Wang, A., Gerber, D., & Becerik-Gerber, B. (2017). Towards user centered building design: Identifying end-user lighting preferences via immersive virtual environments. Automation in Construction, 81, 56-66. doi:10.1016/j.autcon.2017.05.003

Hirning, M. (2014). The application of luminance mapping to discomfort glare : a modified glare index for green buildings. (Dissertation/Thesis), Queensland University of Technology Retrieved from https://eprints.qut.edu.au/72926/4/Michael_Hirning_Thesis.pdf

Hirning, M., Isoardi, G., & Cowling, I. (2014). Discomfort glare in open plan green buildings. Energy and Buildings, 70, 427-440. doi:http://dx.doi.org/10.1016/j.enbuild.2013.11.053

Hirning, M., Isoardi, G., Coyne, S., Garcia Hansen, V., & Cowling, I. (2013). Post occupancy evaluations relating to discomfort glare: A study of green buildings in Brisbane. Building and Environment, 59, 349-357. doi:10.1016/j.buildenv.2012.08.032

Hong, T., & Lin, H.-W. (2012, 2013). Occupant Behavior: Impact on Energy Use of Private Offices,United States.

References 157

Hopkinson, R. G. (1972). Glare from daylighting in buildings. Applied Ergonomics, 3(4), 206-215. doi:10.1016/0003-6870(72)90102-0

Hourani, M. M., & Hammad, R. N. (2012). Impact of daylight quality on architectural space dynamics: Case study: City Mall - Amman, Jordan. Renewable and Sustainable Energy Reviews, 16(6), 3579-3585. doi:10.1016/j.rser.2012.02.074

Houser, K. W., & Tiller, D. K. (2003). Measuring the subjective response to interior lighting: paired comparisons and semantic differential scaling. Lighting Research & Technology, 35(3), 183-195. doi:10.1191/1365782803li073oa

Hu, X., Houser, K. W., & Tiller, D. K. (2006). Higher color temperature lamps may not appear brighter. LEUKOS, 3(1), 69-81.

Hua, Y., Oswald, A., & Yang, X. (2011). Effectiveness of daylighting design and occupant visual satisfaction in a LEED Gold laboratory building. Building and Environment, 46(1), 54-64. doi:10.1016/j.buildenv.2010.06.016

Huang, Y., Niu, J. L., & Chung, T. M. (2014). Comprehensive analysis on thermal and daylighting performance of glazing and shading designs on office building envelope in cooling-dominant climates. Applied Energy, 134, 215-228. doi:10.1016/j.apenergy.2014.07.100

IESNA, & Rea, M. S. (2000). The IESNA lighting handbook: reference & application. New York, NY: Illuminating Engineering Society of North America.

Ihm, P., Nemri, A., & Krarti, M. (2009). Estimation of lighting energy savings from daylighting. Building and Environment, 44(3), 509-514. doi:10.1016/j.buildenv.2008.04.016

Inanici, M. N. (2006). Evaluation of high dynamic range photography as a luminance data acquisition system. Lighting Research & Technology, 38(2), 123-134. doi:10.1191/1365782806li164oa

Islam, M. S., Dangol, R., Hyvärinen, M., Bhusal, P., Puolakka, M., & Halonen, L. (2015). User acceptance studies for LED office lighting: Lamp spectrum, spatial brightness and illuminance. Lighting Research & Technology, 47(1), 54-79. doi:10.1177/1477153513514425

Iwata, T., & Tokura, M. (1997). Position Index for a glare source located below the line of vision.Lighting Research and Technology, 29(3), 172-178.

Iyer-Raniga, U., Moore, T., & Wasiluk, K. (2014). Residential building sustainability rating tools in Australia. Environment Design Guide(80), 1-14.

Jacobs, R., Bullimore, M., Bailey, I., & Berman, S. (1992). Comparing three subjective methods for assessing discomfort glare. Optom Vision Sci, 69, 34.

Jakubiec, J. A., & Reinhart, C. F. (2012). The ‘adaptive zone’ – A concept for assessing discomfort glare throughout daylit spaces. Lighting Research & Technology, 44(2), 149-170. doi:10.1177/1477153511420097

Jameson, D., & Hurvich, L. M. (1964). Theory of brightness and color contrast in human vision. Vision Research, 4(1), 135-154. doi:10.1016/0042-6989(64)90037-9

Jiang, C. J., Sun, C. C., Chen, Y. C., Yang, T. H., & Chang, N. (2014). The correlation of veiling luminance and unified glare rating with a transfer function. Lighting Research & Technology, 46(5), 587-592. doi:10.1177/1477153513493799

Ju, J., Chen, D., & Lin, Y. (2012). Effects of correlated color temperature on spatial brightness perception. Color Research & Application, 37(6), 450-454. doi:10.1002/col.20711

Kähkönen, E., Welling, I., Kostiainen, T., Lahtinen, M., Salmi, K., & Lampinen, J. (2008). Modeling of Subjective Responses to Indoor Air Quality and Thermal Conditions in Office Buildings. HVAC&R Research, 14(6), 905-923. doi:10.1080/10789669.2008.10391046

Karlsen, L., Heiselberg, P., & Bryn, I. (2015). Occupant satisfaction with two blind control strategies: Slats closed and slats in cut-off position. Solar Energy, 115, 166-179. doi:10.1016/j.solener.2015.02.031

Kazanasmaz, T., Günaydin, M., & Binol, S. (2009). Artificial neural networks to predict daylight illuminance in office buildings. Building and Environment, 44(8), 1751-1757. doi:http://dx.doi.org/10.1016/j.buildenv.2008.11.012

158 References

Kent, M. G., Altomonte, S., Tregenza, P., & Wilson, R. (2014). Discomfort glare and time of day. Lighting Research and Technology, 1477153514547291.

Kevin Van Den, W., & Meek, C. (2015). Daylighting and Integrated Lighting Design: Routledge Ltd.

Kim, I. T., Jang, I. H., Choi, A. S., & Sung, M. (2015). Brightness perception of white LED lights with different correlated colour temperatures. Indoor and Built Environment, 24(4), 500-513. doi:10.1177/1420326X14528732

Kim, J. W., Choe, W. J., Hwang, K. H., & Kwag, J. O. (2017). The Optimum Display for Virtual Reality. SID Symposium Digest of Technical Papers, 48(1), 1146-1149. doi:10.1002/sdtp.11845

Kim, W., & Kim, J. T. (2012). The Variation of the Glare Source Luminance According to the Average Luminance of Visual Field. Indoor and Built Environment, 21(1), 98-108. doi:10.1177/1420326X11419926

Kim, W., Kim, J. T., & Ahn, H. T. (2008). A first approach to discomfort glare in the presence of non-uniform luminance. Building and Environment, 43(11), 1953-1960. doi:10.1016/j.buildenv.2007.11.016

Kischkoweit-Lopin, M. (2002). An overview of daylighting systems. Solar Energy, 73(2), 77-82. doi:10.1016/S0038-092X(02)00036-1

Knez, I. (2001). Effects of colour of light on nonvisual psychological processes. Journal of Environmental Psychology, 21(2), 201-208.

Kolb, B. (2008). Questionnaire Design. In. London: SAGE Publications Ltd.

Krarti, M. (2016). Energy Audit of Building Systems: An Engineering Approach, Second Edition (2nd ed.). Baton Rouge: CRC Press.

Kruithof, A. A. (1941). Tubular luminescence lamps for general illumination. Philips Tech. Review, 6,65-96.

Leder, S., Newsham, G. R., Veitch, J. A., Mancini, S., & Charles, K. E. (2016). Effects of office environment on employee satisfaction: a new analysis. Building Research & Information, 44(1), 34-50. doi:10.1080/09613218.2014.1003176

Lee, J.-H., Yoon, Y., Baik, Y.-K., & Kim, S. (2013). Analyses on Human Responses to Illuminance Variations for Resident-Friendly Lighting Environment in a Small Office. Indoor and Built Environment, 22(3), 535-550. doi:10.1177/1420326X12446208

Lei, Z., Xia, G., Ting, L., Xiaoling, G., Qiao Ming, L., & Guangdi, S. (2007). Color rendering and luminous efficacy of trichromatic and tetrachromatic LED-based white LEDs. Microelectronics Journal, 38(1), 1-6. doi:http://dx.doi.org/10.1016/j.mejo.2006.09.004

Leung, T. C. Y., Rajagopalan, P., & Fuller, R. (2013). Performance of a daylight guiding system in an office building. Solar Energy, 94(0), 253-265. doi:http://dx.doi.org/10.1016/j.solener.2013.05.004

Li, D. H. W. (2010). A review of daylight illuminance determinations and energy implications. Applied Energy, 87(7), 2109-2118. doi:http://dx.doi.org/10.1016/j.apenergy.2010.03.004

Li, D. H. W., Lam, T. N. T., & Wong, S. L. (2006). Lighting and energy performance for an office using high frequency dimming controls. Energy Conversion and Management, 47(9), 1133-1145. doi:10.1016/j.enconman.2005.06.016

Linhart, F., & Scartezzini, J.-L. (2011). Evening office lighting – visual comfort vs. energy efficiency vs. performance? Building and Environment, 46(5), 981-989. doi:http://dx.doi.org/10.1016/j.buildenv.2010.10.002

Loe, L., Mansfield, K., & Rowlands, E. (1994). Appearance of lit environment and its relevance in lighting design: experimental study. Lighting Research and Technology, 26(3), 119-133.

Loomis, J. M., Blascovich, J. J., & Beall, A. C. (1999). Immersive virtual environment technology as a basic research tool in psychology. Behavior research methods, instruments, & computers, 31(4), 557-564.

Lowry, G. (2016). Energy saving claims for lighting controls in commercial buildings. Energy and Buildings, 133, 489-497. doi:10.1016/j.enbuild.2016.10.003

References 159

Luckiesh, M., & Guth, S. K. (1949). Brightness in visual field at borderline between comfort and discomfort (BCD). Illuminating Engineering, 44(11), 650-670.

Madden, N. (2006). Magnitude Estimation. Retrieved from http://ucalgary.ca/pip369/mod1/differencethresholds/magnitude

Madden, N. (2008). Magnitude Estimation. Retrieved from http://groups.inf.ed.ac.uk/webexp/release/demos/magnitude-estimation-info.html

Mainster, M. A., & Turner, P. L. (2012). Glare's causes, consequences, and clinical challenges after a century of ophthalmic study. American Journal of Ophthalmology, 153(4), 587-593. doi:10.1016/j.ajo.2012.01.008

Manav, B. (2007). An experimental study on the appraisal of the visual environment at offices in relation to colour temperature and illuminance. Building and Environment, 42(2), 979-983. doi:10.1016/j.buildenv.2005.10.022

Mardaljevic, J. (2001). The BRE-IDMP dataset: a new benchmark for the validation of illuminance prediction techniques. Lighting Research & Technology, 33(2), 117-134. doi:10.1177/136578280103300209

Mardaljevic, J., Heschong, L., & Lee, E. (2009). Daylight metrics and energy savings. Lighting Research and Technology, 41(3), 261-283. doi:http://dx.doi.org/10.1177/1477153509339703

Mayhoub, M. S. (2014). Innovative daylighting systems’ challenges: A critical study. Energy and Buildings, 80(0), 394-405. doi:http://dx.doi.org/10.1016/j.enbuild.2014.04.019

McKinsey & Company. (2012). Lighting the way: perspectives on the global lighting market (Second ed.).

Meek, C., & Wymelenberg, K. V. D. (2015). Daylighting and Integrated Lighting Design. Hoboken: Routledge Ltd.

Menassa, C. C., & Azar, E. (2012). Agent-Based Modeling of Occupants and Their Impact on Energy Use in Commercial Buildings. Journal of Computing in Civil Engineering, 26(4), 506-518. doi:10.1061/(ASCE)CP.1943-5487.0000158

Mills, E., & Borg, N. (1999). Trends in recommended illuminance levels: an international comparison. Journal of the Illuminating Engineering Society, 28(1), 155-163.

Mills, P. R., Tomkins, S. C., & Schlangen, L. J. (2007). The effect of high correlated colour temperature office lighting on employee wellbeing and work performance. Journal of Circadian Rhythms, 5(1), 2.

Monette, D., Sullivan, T., & DeJong, C. (2013). Applied social research: A tool for the human services:Cengage Learning.

Muneer, T. (2000). Windows in buildings: thermal, acoustic, visual, and solar performance : CD-ROM with MS-Excel workbooks includes. Oxford; Boston: Architectural Press.

Murdoch, M. J., Stokkermans, M. G. M., & Lambooij, M. (2015). Towards perceptual accuracy in 3D visualizations of illuminated indoor environments. Journal of Solid State Lighting, 2(1), 12. doi:10.1186/s40539-015-0029-6

Nabil, A., & Mardaljevic, J. (2005). Useful daylight illuminance: a new paradigm for assessing daylight in buildings. Lighting Research and Technology, 37(1), 41-57.

Nabil, A., & Mardaljevic, J. (2006). Useful daylight illuminances: A replacement for daylight factors. Energy and Buildings, 38(7), 905-913. doi:http://dx.doi.org/10.1016/j.enbuild.2006.03.013

Nair, M. G., Ramamurthy, K., & Ganesan, A. R. (2014). Classification of indoor daylight enhancement systems. Lighting Research & Technology, 46(3), 245-267. doi:10.1177/1477153513483299

NASA. (2013). The Electromagnetic Spectrum. Retrieved from http://imagine.gsfc.nasa.gov/science/toolbox/emspectrum1.html

Natephra, W., Motamedi, A., Fukuda, T., & Yabuki, N. (2017). Integrating building information modeling and virtual reality development engines for building indoor lighting design. Visualization in Engineering, 5(1), 1-21. doi:10.1186/s40327-017-0058-x

160 References

Nazzal, A. A. (2005). A new evaluation method for daylight discomfort glare. International Journal of Industrial Ergonomics, 35(4), 295-306. doi:http://dx.doi.org/10.1016/j.ergon.2004.08.010

Ne'eman, E., Sweitzer, G., & Vine, E. (1984). Office worker response to lighting and daylighting issues in workspace environments: A pilot survey. Energy & Buildings, 6(2), 159-171. doi:10.1016/0378-7788(84)90071-9

Newsham, G. R. G., Veitch, J. A. J., & Aries, M. B. C. M. (2010). Windows, view, and office characteristics predict physical and psychological discomfort. VOLUME=30;ISSUE=4;STARTPAGE=533, 30(4), 533-541. doi:10.1016/j.jenvp.2009.12.004

Ng, R. T. H., & Chung, T.-m. (2011). On the calibration of high dynamic range photography for luminance measurements in indoor daylit scenes. Architectural Science Review, 54(1), 39-49. doi:10.3763/asre.2010.0007

Nlena, & Associates. (2015). Lighting Measurements. Retrieved from http://www.ies.org/pdf/education/IES-Color-3-Webcast-Handout.pdf

Noguchi, H., & Sakaguchi, T. (1999). Effect of illuminance and color temperature on lowering of physiological activity. Applied human science, 18(4), 117-123.

Ochoa, C. E., Aries, M. B. C., van Loenen, E. J., & Hensen, J. L. M. (2012). Considerations on design optimization criteria for windows providing low energy consumption and high visual comfort. Applied Energy, 95(0), 238-245. doi:http://dx.doi.org/10.1016/j.apenergy.2012.02.042

Ochoa, C. E., & Capeluto, I. G. (2006). Evaluating visual comfort and performance of three natural lighting systems for deep office buildings in highly luminous climates. Building and Environment, 41(8), 1128-1135. doi:http://dx.doi.org/10.1016/j.buildenv.2005.05.001

Office of Energy Efficiency and Renewable Energy. (2015). LED Basics. Retrieved from http://energy.gov/eere/ssl/led-basics

Oppenheim, A. N. (2000). Questionnaire design, interviewing and attitude measurement: Bloomsbury Publishing.

Osterhaus, W. K. E. (2005). Discomfort glare assessment and prevention for daylight applications in office environments. Solar Energy, 79(2), 140-158. doi:10.1016/j.solener.2004.11.011

Ozorhon, I. F., & Uraz, T. U. (2014). Natural light as a determinant of the identity of architectural space. Journal of Architecture and Urbanism, 38(2), 107-119. doi:10.3846/20297955.2014.916513

Öztürk, L. D. (2003). The Effect of Luminance Distribution on Interior Perception. Architectural Science Review, 46(3), 233-238. doi:10.1080/00038628.2003.9696989

Paes, D., Arantes, E., & Irizarry, J. (2017). Immersive environment for improving the understanding of architectural 3D models: Comparing user spatial perception between immersive and traditional virtual reality systems. Automation in Construction, 84, 292-303. doi:10.1016/j.autcon.2017.09.016

Park, B.-C., Chang, J.-H., Kim, Y.-S., Jeong, J.-W., & Choi, A.-S. (2010). A Study on the Subjective Response for Corrected Colour Temperature Conditions in a Specific Space. Indoor and Built Environment, 19(6), 623-637. doi:10.1177/1420326X10383472

Pellegrino, A., Cammarano, S., Lo Verso, V. R. M., & Corrado, V. (2017). Impact of daylighting on total energy use in offices of varying architectural features in Italy: Results from a parametric study. Building and Environment, 113, 151-162. doi:10.1016/j.buildenv.2016.09.012

Piccolo, A., & Simone, F. (2009). Effect of switchable glazing on discomfort glare from windows. Building and Environment, 44(6), 1171-1180. doi:10.1016/j.buildenv.2008.08.013

Pierson, C., Jacobs, A., Wienold, J., & Bodart, M. (2017). Luminance maps from High Dynamic Range imaging: photometric, radiometric and geometric calibrations. Paper presented at the Lux Europa 2017.

Poirazis, H., Blomsterberg, Å., & Wall, M. (2008). Energy simulations for glazed office buildings in Sweden. Energy and Buildings, 40(7), 1161-1170. doi:http://dx.doi.org/10.1016/j.enbuild.2007.10.011

References 161

Polit, D. F., & Hungler, B. P. (1991). Nursing Research. Principles and Methods. New York, Hagestown: J.B. Lippincott Company, Philadelphia.

Pollock, M., Roderick, Y., McEwan, D., & Wheatley, C. (2009). Building simulation as an assisting tool in designing an energy efficient building: A case study. BUILDING SIMULATION, Glasgow.

Pridmore, R. W. (1999). Bezold–Brucke hue-shift as functions of luminance level, luminance ratio, interstimulus interval and adapting white for aperture and object colors. Vision Research, 39(23), 3873-3891. doi:10.1016/S0042-6989(99)00085-1

Pridmore, R. W. (2007). Effects of luminance, wavelength and purity on the color attributes: Brief review with new data and perspectives. Color Research and Application, 32(3), 208-222. doi:10.1002/col.20312

Rea, M. S. (1984). Window blind occlusion: a pilot study. Building and Environment, 19(2), 133-137.

Rea, M. S. (2000). The IESNA lighting handbook: reference & application (9th ed.). New York, NY: Illuminating Engineering Society of North America.

Reinhard, E., Ward, G., Pattanaik, S., Debevec, P., Heidrich, W., & Myszkowski, K. (2010). HDR Imaging-Acquisition, Display, and Image-Based Lighting. In: Morgan Kaufmann.

Reinhart, C. F., LoVerso, V. R. M., Scartezzini, J. L., & Cuttle, C. (2010). A rules of thumb-based design sequence for diffuse daylight. Lighting Research and Technology, 42(1), 7-31. doi:http://dx.doi.org/10.1177/1477153509104765

Reinhart, C. F., Mardaljevic, J., & Rogers, Z. (2006). Dynamic Daylight Performance Metrics for Sustainable Building Design. LEUKOS, 3(1), 7-31. doi:10.1582/LEUKOS.2006.03.01.001

Reinhart, C. F., & Weissman, D. A. (2012). The daylit area – Correlating architectural student assessments with current and emerging daylight availability metrics. Building and Environment, 50(0), 155-164. doi:http://dx.doi.org/10.1016/j.buildenv.2011.10.024

Roche, L., Dewey, E., & Littlefair, P. (2000). Occupant reactions to daylight in offices. Lighting Research & Technology, 32(3), 119-126. doi:10.1177/096032710003200303

Rodriguez, R. G., & Pattini, A. (2012). Effects of a large area glare source in cognitive efficiency and effectiveness in visual display terminal work. LEUKOS, 8(4), 283-299.

Rodriquez, R. G., & Pattini, A. (2014). Tolerance of discomfort glare from a large area source for work on a visual display. Lighting Research & Technology, 46(2), 157-170. doi:10.1177/1477153512470386

Roe, B., & Webb, C. (1998). Methods of Data Collection for Quantitative Research. In (pp. 135-169). London, England: Whurr Publishers Ltd.

Ryckaert, W. R., Lootens, C., Geldof, J., & Hanselaer, P. (2010). Criteria for energy efficient lighting in buildings. Energy & Buildings, 42(3), 341-347. doi:10.1016/j.enbuild.2009.09.012

Saini, B. S. (1980). Building in Hot Dry Climates: John Wiley and Sons.

Sanati, L., & Utzinger, M. (2013). The effect of window shading design on occupant use of blinds and electric lighting. Building and Environment, 64(0), 67-76. doi:http://dx.doi.org/10.1016/j.buildenv.2013.02.013

Sandahl, L. J., Cort, K. A., & Gordon, K. L. (2013). Solid-State Lighting: Early Lessons Learned on the Way to Market. Retrieved from

Sarey Khanie, M., Wienold, J., & Andersen, M. (2014). A sensitivity analysis on glare detection parameters. Paper presented at the 13th International Radiance Workshop 2014.

Saxena, M., Heschong, L., Wymelenberg, K. V. D., Wayland, S., & Analytics, I. P. (2010). 61 Flavors of daylight. In: Washington (DC): American Council for an Energy-Efficient Economy Asilomar, CA.

Schreuder, D. (2008). Outdoor lighting: physics, vision and perception: Springer.

Schweizer, C., Edwards, R. D., Bayer-Oglesby, L., Gauderman, W. J., Ilacqua, V., Juhani Jantunen, M., . . . Künzli, N. (2007). Indoor time-microenvironment-activity patterns in seven regions of Europe.

162 References

Journal of Exposure Science and Environmental Epidemiology, 17(2), 170-181. doi:10.1038/sj.jes.7500490

Shen, H., & Tzempelikos, A. P. (2012). An Experimental and Simulation Study of Lighting Performance in Offices with Automated Roller Shades. ASHRAE Transactions, 118, 66-73.

Shin, J. Y., Yun, G. Y., & Kim, J. T. (2012a). View types and luminance effects on discomfort glare assessment from windows. Energy and Buildings, 46(0), 139-145. doi:http://dx.doi.org/10.1016/j.enbuild.2011.10.036

Shin, J. Y., Yun, G. Y., & Kim, J. T. (2012b). View types and luminance effects on discomfort glare assessment from windows. Energy & Buildings, 46, 139-145. doi:10.1016/j.enbuild.2011.10.036

Solemma LLC. (2015). DIVA for Rhino. Retrieved from http://diva4rhino.com/

Soneira, R. M. (2018). Galaxy S7 OLED Display Technology Shoot-Out. Retrieved from http://www.displaymate.com/Galaxy_S7_ShootOut_1.htm

Steffy, G. R. (2008). Architectural lighting design (Vol. 3rd). Hoboken, N.J: John Wiley & Sons.

Stuart-Hamilton, I. (2007). Paired comparison. In.

Suk, J., & Schiler, M. (2013). Investigation of Evalglare software, daylight glare probability and high dynamic range imaging for daylight glare analysis. Lighting Research and Technology, 45(4), 450-463. doi:http://dx.doi.org/10.1177/1477153512458671

Suk, J. Y., Schiler, M., & Kensek, K. (2013). Development of new daylight glare analysis methodology using absolute glare factor and relative glare factor. Energy and Buildings, 64(0), 113-122. doi:http://dx.doi.org/10.1016/j.enbuild.2013.04.020

Suk, J. Y., Schiler, M., & Kensek, K. (2017). Investigation of existing discomfort glare indices using human subject study data. Building and Environment, 113, 121-130. doi:10.1016/j.buildenv.2016.09.018

Tashiro, T., Kawanobe, S., Kimura-Minoda, T., Kohko, S., Ishikawa, T., & Ayama, M. (2015). Discomfort glare for white LED light sources with different spatial arrangements. Lighting Research & Technology, 47(3), 316-337. doi:10.1177/1477153514532122

Thompson, W. B., Willemsen, P., Gooch, A. A., Creem-Regehr, S. H., Loomis, J. M., & Beall, A. C. (2004). Does the Quality of the Computer Graphics Matter when Judging Distances in Visually Immersive Environments? Presence: Teleoperators & Virtual Environments, 13(5), 560-571. doi:10.1162/1054746042545292

Tifler, D. K., & Rea, M. S. (1992). Semantic differential scaling: Prospects in lighting research. Lighting Research and Technology, 24(1), 43-51.

Tiller, D., & Veitch, A. J. (1995). Perceived room brightness: Pilot study on the effect of luminance distribution. Lighting Research & Technology, 27(2), 93-101.

Tregenza, P., & Loe, D. (2014). The design of lighting (Vol. 2nd). Milton Park, Abingdon, Oxon: Routledge.

Tregenza, P., & Wilson, M. (2011). Daylighting: architecture and lighting design. Abingdon, Oxon: Routledge.

Tsangrassoulis, A. (2008). A Review of Innovative Daylighting Systems. Advances in Building Energy Research, 2(1), 33-33. doi:10.3763/aber.2008.0202

Tuaycharoen, N., & Tregenza, P. R. (2007). View and discomfort glare from windows. Lighting Research & Technology, 39(2), 185-200. doi:10.1177/1365782807077193

Tzempelikos, A. (2008). The impact of venetian blind geometry and tilt angle on view, direct light transmission and interior illuminance. Solar Energy, 82(12), 1172-1191. doi:http://dx.doi.org/10.1016/j.solener.2008.05.014

Tzempelikos, A., & Athienitis, A. K. (2007). The impact of shading design and control on building cooling and lighting demand. Solar Energy, 81(3), 369-382.

Veitch, J. A., Hine, D. W., & Gifford, R. (1993). End Users ‘Knowledge, Beliefs, and Preferences for Lighting. Journal of Interior Design, 19(2), 15-26.

References 163

Velds, M. (2002). User acceptance studies to evaluate discomfort glare in daylit rooms. Solar Energy, 73(2), 95-103. doi:10.1016/S0038-092X(02)00037-3

Vidovszky-Németh, A., & Schanda, J. (2012). White light brightness–luminance relationship. Lighting Research & Technology, 44(1), 55-68. doi:10.1177/1477153512437013

Vine, E., Lee, E., Clear, R., DiBartolomeo, D., & Selkowitz, S. (1998). Office worker response to an automated Venetian blind and electric lighting system: a pilot study. Energy and Buildings, 28(2), 205-218. doi:http://dx.doi.org/10.1016/S0378-7788(98)00023-1

Vos, J. J. (2003). Reflections on glare. Lighting Research and Technology, 35(2), 163-175.

Waide, b. P., Tanishima, S., Harrington, P., & Iea. (2006). Light's labour's lost: policies for energy-efficient lighting. FR: International Energy Agency.

The Well Building Standard. (2018). In (Vol. 1). New York.

Wienold, J., & Christoffersen, J. (2006). Evaluation methods and development of a new glare prediction model for daylight environments with the use of CCD cameras. Energy and Buildings, 38(7), 743-757.

Wilks, A., & Osterhaus, W. (2003). Towards an assessment method for visual comfort in daylit offices. IEA SHC Task, 31.

Yamin Garreton, J. A., Rodriguez, R. G., & Pattini, A. E. (2014). Glare indicators: an analysis of ocular behaviour in an office equipped with venetian blinds. Indoor and Built Environment.doi:10.1177/1420326X14538082

Yan, D., O’Brien, W., Hong, T., Feng, X., Burak Gunay, H., Tahmasebi, F., & Mahdavi, A. (2015). Occupant behavior modeling for building performance simulation: Current state and future challenges. Energy & Buildings, 107, 264-278. doi:10.1016/j.enbuild.2015.08.032

Yuen, G. S. C., Sproul, A. B., & Dain, S. J. (2010). Performance of energy efficient compact fluorescent lamps. Clinical & Experimental Optometry, 93(2), 66-76. doi:10.1111/j.1444-0938.2010.00462.x

Yun, G., Yoon, K. C., & Kim, K. S. (2014). The influence of shading control strategies on the visual comfort and energy demand of office buildings. Energy and Buildings, 84(0), 70-85. doi:http://dx.doi.org/10.1016/j.enbuild.2014.07.040

Zalejska-Jonsson, A., Fastigheter och, b., Bygg- och, f., Kth, & Skolan för arkitektur och, s. (2014). Parameters contributing to occupants' satisfaction: Green and conventional residential buildings. Facilities, 32(7-8), 411-437. doi:10.1108/F-03-2013-0021

Zhao, S. (2003). Toward a Taxonomy of Copresence. Presence: Teleoperators & Virtual Environments, 12(5), 445-455. doi:10.1162/105474603322761261

Documents

INNOVATIVE INTEGRATED WINDOW DESIGN WITH ELECTRIC … · M. Arch., B. Arch. Submitted in fulfilment of the requirements for the degree of Doctor of Philosophy ... drawing window blinds