See discussions, stats, and author profiles for this publication at: [Link]

net/publication/283708072

Retrofitting the Electric Lighting and Daylighting Systems to Reduce Energy

Use in Buildings: A Literature Review

Article in Energy Research Journal · August 2015

DOI: 10.3844/erjsp.2015.25.41

CITATIONS READS

41 1,946

7 authors, including:

Marie-Claude Dubois Fabio Bisegna

Lund University Sapienza University of Rome
58 PUBLICATIONS 1,185 CITATIONS 134 PUBLICATIONS 1,693 CITATIONS

SEE PROFILE SEE PROFILE

Niko Gentile Martine Knoop

Lund University Technische Universität Berlin
34 PUBLICATIONS 262 CITATIONS 57 PUBLICATIONS 343 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Building Façade Integrated Photovoltaic Architectural methods in urban context View project

Spectral Sky Models. Development and Application of Spectral Sky Models in Urban planning View project

All content following this page was uploaded by Barbara Matusiak on 11 November 2015.

The user has requested enhancement of the downloaded file.

 Energy Research Journal

Literature Reviews

Retrofitting the Electric Lighting and Daylighting Systems to

Reduce Energy Use in Buildings: A Literature Review
1
Marie-Claude Dubois, 2Fabio Bisegna, 1Niko Gentile,
3
Martine Knoop, 4Barbara Matusiak, 5Werner Osterhaus and 6Eino Tetri
1
Division of Energy and Building Design, Institute for Architecture and the Built Environment,
Lund University, Lund, Sweden, Box 118, 221 00 Lund, Sweden
2
Department of Astronautical Electrical and Energetic Engineering,
Faculty of Engineering, SAPIENZA University of Rome, Via Eudossiana 18-00184, Rome, Italy
3
Technische Universität Berlin, Einsteinufer 19, 10587 Berlin, Germany
4
Faculty of Architecture, NTNU Norwegian University of Science and Technology,
Alfred Getz vei 3, 7491 Trondheim, Norway
5
Department of Engineering, Lighting Research Design Laboratory, Aarhus University,
Dalgas Avenue 2, DK-8000 Aarhus C, Denmark
6
Aalto University, PO Box 13340, 00076 Aalto, Finland

Article history Abstract: This paper presents a literature review about energy-efficient
Received: 05-07-2015 retrofit of electric lighting and daylighting systems in buildings. The
Revised: 10-07-2015 review, which covers around 160 research articles, addresses the following
Accepted: 28-07-2015 themes: (1) retrofitting electric lighting in buildings, (2) electric lighting
energy use and saving potential and (3) lighting retrofit strategies. The
Corresponding Author: retrofit strategies covered in the review are: replacement of lamp, ballast or
Marie-Claude Dubois
luminaire; use of task-ambient lighting design; improvement in
Division of Energy and
Building Design, Institute for maintenance; reduction of maintained illuminance levels; improvement in
Architecture and the Built spectral quality of light sources; improvement in occupant behavior;
Environment, Lund University, use of control systems; and use of daylighting systems. The review
Lund, Sweden, Box 118, 221 indicates that existing general knowledge about lighting retrofit is
00 Lund, Sweden currently very limited and that there is a significant lack of information
concerning the actual energy performance of lighting systems installed
E-mail: [Link]@[Link] in the existing building stock. The resulting key directions for future
research highlights issues for which a better understanding is required
for the spread and development of lighting retrofit.

Keywords: Lighting Retrofit, Energy-Efficiency, Luminaires, Lamps,

Lighting Controls, Daylighting Systems, Occupant Behavior

Introduction rebound effect, which means that with the reduced

system power in lighting-and fixed energy prices-the
Projections by the International Energy Agency tendency is to use more light because it is cheaper and
(IEA, 2006) show that if governments only rely on by that absolute energy consumption is ultimately
current policies, global electricity use for lighting will increased. Increase in efficiency will thus not
grow to around 4250 TWh by 2030, an alarming necessarily produce an absolute reduction of energy
increase of more than 40%. Due to the world’s use, as observed by William Stanley Jevons in 1865.
growing population and the increasing demand for The Jevons Paradox, which was first expressed in relation
electrically driven services in emerging economies, to use of coal, states that an increase in efficiency in using a
this increase will occur despite constant improvements resource leads to increased use of that resource rather than
in energy efficiency of electric systems. Tsao and to a reduction (Polimeni et al., 2007). However,
Waide (2010) claimed that there is a massive potential Saunders and Tsao (2012) argued that even in this scenario,
for growth in the consumption of light if new lighting the increase in efficiency should be pursued because it will
technologies are developed with higher luminous at least lead to economic benefits.
efficacies and lower cost of light. Many authors (e.g. To sum up: There is a high saving potential with new
Porritt et al., 2013) have warned about this so-called lighting technologies. High quality lighting can be

© 2015 Marie-Claude Dubois, Fabio Bisegna, Niko Gentile, Martine Knoop, Barbara Matusiak, Werner Osterhaus and Eino
Tetri. This open access article is distributed under a Creative Commons Attribution (CC-BY) 3.0 license.
Marie-Claude Dubois et al. / Energy Research Journal 2015, ■ (■): ■■.■■
DOI: 10.3844/erjsp.2015.■■.■■

achieved in energy-efficient and more sustainable buildings by the new-build is only around 1.0-3.0%
ways with appropriate retrofits. However, awareness per annum (see also Eames et al., 2013; Itani et al.,
of possible rebound effects is important; therefore, 2013) and it is 2.2% per year in the commercial
actions to raise awareness, measures targeting building sector. Current renovation and refurbishment
absolute energy savings and understanding of the rates are somewhat higher-between 2.9% and 5% in
potential to improve lighting quality in existing the UK of existing stock for domestic buildings and 2-
buildings are all urgently needed. This article focuses 8% for commercial stock, depending on the sector
on the available energy saving strategies in lighting or (Stafford et al., 2011). In the UK, for instance, some
daylighting (façade and roofs) retrofit and their 70% of total 2010 building stock is expected still to be
potential as described in the scientific literature. in use in 2050 (BBP, 2010). Therefore, rapid
improvement of energy efficiency in existing
Lighting Offers Great Energy Saving Potential buildings is needed for a timely reduction in global
energy use and promotion of environmental
Lighting has been pointed out as one of the areas
sustainability (Zhenjun et al., 2012).
with significant improvement potential in energy- During the last decade, many governments (e.g.
efficiency. One study (Enkvist et al., 2007) pointed USA, UK, Australia and others) and international
out that investment in energy‐efficient lighting is one organisations (e.g. IEA) have put significant efforts
of the most cost‐effective ways for improving energy- towards energy efficiency improvement in existing
efficiency in buildings and reduce CO2 emissions. buildings (Zhenjun et al., 2012). For instance, the
According to Krarti et al. (2005), energy retrofits of International Energy Agency (IEA) has launched a
lighting equipment are very cost-effective with typical series of Annex projects (Annex 46, 50, 55, 56) to
payback periods of less than two years. In addition, promote energy efficiency of existing buildings. In the
higher costs of electricity compared to most other field of lighting, the IEA launched Annex 45-Energy
energy sources (e.g. natural gas) in most countries efficient electric lighting for buildings under the
further justifies ranking lighting retrofit measures high umbrella of the Energy Conservation in Buildings and
on the list of options as suggested by Boyano et al. Community Systems (ECBCS) programme. More
(2013). One report (AEFEETS, 2010) even stated that recently, the IEA launched a new task within the Solar
the cost of saving 1 kWh lighting energy through Heating and Cooling (SHC) Programme entitled ‘IEA
efficiency is less than 20% of the average price of 1 Task 50-Advanced Lighting Solutions for Retrofitting
kWh electricity in the United States. Buildings’. This article was prepared as part of the
In a recent article, Dubois and Blomsterberg dissemination effort of this international Task, which
(2011) presented key energy use figures and saving will be completed at the end of 2015. Further
information can be found at the Task website
potential for electric lighting in office buildings based
[Link]
on a literature review. This review concluded that an
annual energy intensity of 10kWh/m2 is a realistic Objectives of this Literature Review
target for electric lighting in future low energy office
This literature review was achieved with the aim to
buildings. According to these authors, this would
analyze and summarize existing information found in
yield a significant reduction in energy intensity of at the scientific literature, previous European and
least 50% compared to the actual average electricity international research projects, websites of national
use for office lighting. This review also discusses projects, etc. The specific objectives of this literature
different strategies for reducing lighting energy use in review are listed below:
addition to providing a summary of saving potential.
Many of these strategies are further discussed and • Identify existing databases of case studies
updated in the present article. • Identify previous research about lighting and/or
daylighting retrofit
Importance of Retrofitting the Existing Building • Update key information regarding energy saving
Stock strategies and solutions demonstrated in the past
by research, monitoring or demonstration projects
According to Zhenjun et al. (2012), retrofitting
should be considered as one of the main approaches to • Summarize the energy saving potential according to
achieving sustainability in the built environment at measure or strategy
relatively low cost and high uptake rates. These Table 1 summarizes some of the key ideas
authors point out that most energy is consumed by presented in the introduction as outlined by this
existing buildings and the replacement rate of existing literature review.

■■
Marie-Claude Dubois et al. / Energy Research Journal 2015, ■ (■): ■■.■■
DOI: 10.3844/erjsp.2015.■■.■■

Table 1. SWOT analysis of the lighting retrofit situation

Strengths High demonstrated energy savings of new lighting technologies and control systems
Demonstrated short payback times
Minimal disruption of lighting retrofit compared to many other retrofit measures
Weaknesses Lack of knowledge and hands-on experience about many retrofit measures in a retrofit
context (reduced illuminance, task-ambient lighting design, improvement of spectral quality
of light source, improved occupant behavior etc.)
Uncertainty of predicted energy savings and lack of reliability of some control systems
(occupancy and daylight- linked)
Opportunities Retrofit of the existing building stock for timely reduction of energy use and promotion of
environmental sustainability
General need to improve and modernize indoor environmental quality including lighting
qualityHigher cost of electricity ranking lighting high on the list of priorities
Obsolescence of existing lighting installations in developed countries
Threats Potential increase of lighting energy use by 40% in 2030
Rebound effect

Method net energy savings were about 70% of the gross lighting
energy savings for most cases of recessed fluorescent
This review is based on information mainly found in fixtures, due to an increase in heating demand. They
peer-reviewed journal articles, conference articles, argued that the improvement of the lighting system
reports, past relevant IEA and European or national might thus be less cost effective than expected
projects published during the last 20 years (i.e. in the initially when only considering the gross energy
period 1993-2013). The articles were identified by first savings by the lighting system alone.
performing a search in several databases (Science Direct, Hestnes and Kofoed (2002) suggested that
Compendex, Inspec, etc.) with the following keywords: improvements in lighting should be combined with
lighting retrofit, energy-efficient lighting, relamping, building envelope improvements to avoid the related
luminaires, lamps, lighting control, light sensors,
increase in heating loads. However, they argued that
dimming, daylight retrofit, daylighting systems, etc.
since electricity is usually more expensive than thermal
Subsequently, the bibliography of each article was
scrutinized to find other relevant sources. The authors energy, any saving in electricity use is a more valuable
also asked all experts of IEA Task 50 for articles related improvement. In addition to this, when heating is
to retrofitting and lighting. Finally, the main author provided by a heating system (instead of a lighting
registered for automatic email alerts in various fields in system), it is targeted, controllable, more effective and
order to get updates on new relevant articles. This thorough does provide better thermal comfort. Electric lighting
process allowed finding over 350 publications of which systems are not designed to provide thermal comfort and
about half were judged directly or indirectly relevant and they are generally not effective at heating in a building.
further analyzed after reading the abstract. This article thus Finally, it is also worth mentioning that a reduction in
summarizes information found in some 160 research lighting energy use also yields a reduction of internal
articles on lighting and/or daylighting retrofit. heat gains and thus cooling needs are reduced while
summertime thermal comfort is generally improved.
Main Findings
Electric Lighting Energy use and Saving Potential
Retrofitting Electric Lighting in Buildings
Lighting consumes 15-60% of the final energy use
Generally, previous research (Dascalaki and in buildings according to Spyropoulos and Balaras
Santamouris, 2002; Erhorn-Kluttig et al., 2004) (2011), a wide variation, which depends on many
suggested that it is necessary to look at energy saving factors such as building type, function, technology
measures in a holistic way since electric lighting used, climate, etc. Previous research suggests that the
reductions normally entail an increase in heating potential for energy savings depends significantly on
demand, which can make lighting retrofit measures less the initial energy demand for lighting and on the
cost effective considering all other end-uses. For building type. Higher initial lighting load and more
example, Zmeureanu and Peragine (1999) investigated compact building shapes generally present higher
by computer simulations the net energy impact of saving potential since these cases generally have higher
lighting system retrofit taking into account the electric lighting demand compared to other installations
interactions with HVAC systems for an existing, 28- and building types (Dascalaki and Santamouris, 2002).
floor, 100 000 m2 office building from 1983 built in Note that daylight provisions are also less likely to be
Montreal, Canada. Overall, the results indicated that the sufficient in such buildings.

■■
Marie-Claude Dubois et al. / Energy Research Journal 2015, ■ (■): ■■.■■
DOI: 10.3844/erjsp.2015.■■.■■

One study (Chidiac et al., 2011) demonstrated that (approximately 50%), a life time typically two to three
reductions in energy use were not necessarily a linear times higher, a comparable color rendering and a beam
addition of the savings from individual Energy Retrofit angle of around 140°.
Measures (ERMs). For example, the implementation of Labayrade and Avouac (2013) recently evaluated the
light dimming features with more efficient lighting performance of 10 000 samples of a customized LED
fixtures is not as effective in reducing consumption as a solution, which were optimized to replace low voltage
linear addition would indicate. Generally, the overall halogen lamps (4 W equivalent to a 20 W halogen and a
reduction in energy consumption is less than the linear 5,5 W equivalent to a 35 W halogen lamp). A total of
addition of the individual ERMs. 9300 retrofits were evaluated in uncontrolled
environments (restaurants, cafes and shops), in which
Lighting Retrofit Strategies
more than 85% of the users were satisfied with the light
Lamp, Ballast and Luminaire Replacement produced by the LED spots and would consider
replacing their halogen lamps with it.
Studies about energy-efficient lighting retrofit
However, the CALIPER study in the USA (U.S.
generally suggest that most existing lighting installations
Department of Energy, 2010) investigated 14 LED
consist of fluorescent lighting (with conventional
retrofit downlights that were equivalent to typical CFL
ballasts). One American study (Baker, 2013) stated that
downlights (32 W) and incandescent downlights (65
the most commonly retrofitted fixtures (in the USA)
W), which are typically applied for ambient lighting in
are the 4-lamp T12 while parabolic and lenses troffers
normal ceiling heights. This study indicated that the
with T12 or older T8 lamps are the primary lamp
luminous flux was too low for all tested ‘MR16
types to replace. This study also outlined that energy
equivalent’ LED retrofit solutions, having a product
savings are generally decreasing over time, due to
efficacy of 16-35 lm/W. Color rendering ranged from
increases in new construction baselines and decreases
61 to 96 and color temperature was not near the target
in the number of existing very inefficient lighting
CCT or Planckian locus. This study also pointed out
systems (‘low-hanging fruit’).
that the low wattage lamps might not provide enough
According to another American author (Vogel,
load to the existing transformer, dimmers or related
2012), specifiers have four traditional options to consider
controls. In that case, the retrofits may not work or
in lighting retrofit:
cause flicker or stroboscopic effects. However, note
1. Relamp and reballast that this study is already six years old, a period with
2. Delamp and reballast much development in SSL technology.
3. One-for-one fixture replacement A more recent study (Poplawski and Miller, 2013)
4. Complete redesign nevertheless showed that a wide variation in flicker
performance and unfamiliar flicker characteristics can
Options 3 and 4 represent a higher investment since still be found amongst LED lighting solutions (also in
entry into the plenum is required-a key factor affecting Lehman et al., 2011; U.S. Department of Energy, 2010).
the cost of retrofitting-but they also present a higher Their study was based on the evaluation of 22 traditional
saving potential. A new generation of lighting retrofit lighting technology sources (incandescent, halogen,
kits is available in energy efficient LED options. metal halide and fluorescent lamps) and 93 LED
According to Vogel (2012), these kits enable component products (mainly retrofit lamps). Since flicker can affect
parts to be installed in 15 min or less into the housing of well-being and performance, it seems to be a relevant
old fixtures, provide better quality and better looking quality criterion to take into consideration in the choice
fixtures and involve minimal disruption because they are of LED retrofit solutions. The IES (2010) recommends a
installed below the ceiling. minimum driver output frequency of 120 Hz to avoid
perceptible flicker, but the analysis by Poplawski and
Retrofits with LED Lamps
Miller (2013) indicates that this is insufficient to ensure
The hottest topic in lighting today is probably the quality. As no standard procedure for evaluation of
possible replacement of different lamp types by highly flicker is currently available (CIE, 2013; Lehman et al.,
efficient LED lamps. Rapid developments in the area of 2011), Poplawski and Miller (2013) proposed a light
Solid-State Lighting (SSL) technology have created a source evaluation using a flicker frequency dependent
real reorganization of the lighting industry worldwide maximum flicker index. Another author (Osterhaus,
with great emphasis on enormous potential savings. An 2014) stressed the need for appropriate combinations of
analysis of LED retrofit lamps offered on the market (as LED sources and LED drivers. Inappropriate
alternative and equivalent to linear fluorescent solutions) combinations can lead to flicker problems, which
carried out as part of IEA Task 50 Subtask B indicated motivates the need to test sources and drivers as a unit,
that these lamps have a reduced energy consumption not as two separate components.

■■
Marie-Claude Dubois et al. / Energy Research Journal 2015, ■ (■): ■■.■■
DOI: 10.3844/erjsp.2015.■■.■■

In addition, a study performed by NC (2012) fixture while the use of T5 system requires electronic
indicated that LED lamps and equivalent compact ballasts with high efficiency version that can reach a
fluorescent lamps have comparable average life-cycle lamp luminous efficacy superior to 100 lm/W. However,
energy consumption (approximately 3,900 MJ per 20 the reader should consider that replacement to T5 tubes
million lumen-hours). For the purpose of the analysis, might require more controlled luminaire outputs (e.g.
a LED lamp luminous efficacy of 64lm/W was used. more louvers or baffles) to prevent glare due to the
But as the efficacy of LED lamps increases, the life- higher luminance of the light source, thus perhaps
cycle energy consumption will diminish, since the reducing efficacy. According to these authors, T5 lamps
energy consumption in use represents the significant can last at least 18,000 h with 5% reduction of lumen
portion of the total life-cycle energy consumption output in the lifetime. In contrast, a T8 light tube usually
lasts about 20,000 h but it loses about 20% output in its
(approximately 90 percent).
life. Mahlia et al. (2011) found that using T8 electronic
Earlier, Ryckaert et al. (2011) evaluated 12 different
system, HPT8 system and T5 lamps with electronic
brands of LED retrofit lamps as alternative for a linear ballasts could reduce the energy consumption and LCC
fluorescent solution (T8/36W, 3500-4000 K, 3350 lm). by 17%, 31% and 40% respectively at 100% retrofitting.
They assessed the quality of the retrofit lamps through Assuming an increase in electricity tariff of 2% per year,
laboratory measurements at the beginning of the project they also found that if retrofitting was fully done
and after 2000 h. The lamp efficacy of the LED retrofit (100%), the payback period for T8 electronic would only
was between 50,8 lm/W and 89,5 lm/W, compared to 75 be 0.689 years while it would be 1.24 years for HPT8
and 95 lm/W for the linear fluorescent solution, and 1.95 years for T5 electronic alternative.
depending on the ballast chosen. The majority of the Table 2 summarizes the potential energy savings
retrofits had a CRI below 80 and would therefore not be reported with replacement of lighting technology.
suitable for office applications. Lumen depreciation over
2000 h varied from -38,7 % to +7,1% amongst the Task-Ambient Lighting Design
different brands. In addition to the product evaluation, Use of task-ambient lighting design has proven to
the application of three selected retrofits was studied in a provide better quality lighting and a 22-25% reduction in
small office room. The authors concluded that, at that electricity use compared to a standard general energy-
time, replacing T8 fluorescent lamps with ‘equivalent’ efficient lighting installation, as discussed by Dubois and
LED retrofits would indeed bring energy savings up to Blomsterberg (2011). However, no study has been found
70%, but would reduce at the same time the illuminance here about task-ambient lighting design approach in a
levels by about 50%, which is consistent with findings retrofit context. This strategy certainly needs to be
from a recent field study (Osterhaus, 2014). The latter studied in a retrofit context.
was noticed by nearly all of the 44 subjects that
evaluated the lighting conditions in the small office Improvement in Maintenance
room. In addition to this major limitation, they noted According to Hanselaer et al. (2007), a high
that the luminous intensity distribution of the luminaire maintenance factor (cleaning) together with an effective
with all three retrofits changed considerably, which maintenance programme promotes energy efficient design
affected the illuminance distribution and uniformity as and limits the installed lighting power requirements.
well as the impression of the room. Gasparovsky and Raditschova (2013) studied the
In summary, despite the promising savings that LED luminous properties of old type luminaires after decades
retrofits may bring in the future, many serious issues such of their operation. They measured their efficacy in their
as flicker, low illuminance levels, poor beam distribution actual conditions and after cleaning, with inserted and
and color rendering have been reported and should be given reference lamps. The measurements included luminous
serious consideration in real retrofit projects. flux, luminous efficiency (in integrating sphere), luminous
intensity distribution curve (with goniophotometer), spectral
Retrofits with T8 and T5 Lamps
transmittance of diffuser in order to assess the yellowish
In the context of Malaysia, Mahlia et al. (2011) effect of UV radiation (using a spectrophotometer) and
investigated the potential energy savings, Life Cycle electrical characteristics of the lamp-ballast system. They
Cost (LCC) and payback period of the lighting system in concluded that luminaires from the 1980s have significant
the campus buildings of the University of Malaya, by non-recoverable losses of about 10% in case of interior
using theoretical calculations and standard cost and luminaires. Pollutants (e.g. dust) collecting on the
payback equations. They compared retrofitting the surfaces of the luminaire are responsible for another 10-
existing standard fluorescent lighting systems (T12) with 20% reduction in performance, but these can be
T8 magnetic (18/36W), T8 electronic (18/36W), HPT8 recovered through cleaning. The luminous efficacy of
electronic (17/32W) and T5 electronic (14/28W) ballasts. old-type T12 fluorescent lamps is 15% lower than
According to these authors, T8 lamps can replace the old catalogue values but in comparison with recent
T12 fluorescent lamps without any modification of the technology their efficacy was only half.

■■
Marie-Claude Dubois et al. / Energy Research Journal 2015, ■ (■): ■■.■■
DOI: 10.3844/erjsp.2015.■■.■■

Table 2. Potential energy savings by using more energy-efficient lighting technology in retrofit projects
Technology Potential energy savings Issues Source
Linear fluorescenta to LED 50% Flicker, reduced illuminance, IEA Task 50
poor beam distribution and color rendering
T8 to LED 70% Ryckaert et al. (2011)
T12 to T8b 17% Mahlia et al. (2011)
T12 to HPT8 31%
T12 to T5 40%
a unspecified
b electronic

Mucklejohn et al. (2013) reported on fundamental time (Escuyer and Fontoynont, 2001). For example, a
basics of lighting design and the dimensioning of Canadian study (Veitch and Newsham, 2000) conducted
lighting taking the specific factors and aspects of light in an open-plan office laboratory where forty-seven
conversion and delivery into account. They presented matched pairs of participants spent a day completing
case studies for a warehouse area (70m×54m) without various simulated office tasks and questionnaires showed
any windows or skylights. In six configurations, HID- that individually preferred light levels varied widely
MH (High-intensity discharge metal halide) luminaires (mean desktop illuminance 423 lx, s.d. 152 lx, min. 83
were compared with High Efficiency Plasma (HEP) light lx, max 725 lx), but on average required 10-15% less
sources. Under the same maintenance conditions, they power than prevailing energy code recommendations.
claimed that HEP light sources can fulfil the lighting Another study (Schuler, 1995) in a computer hardware
design requirements with a total installed power of and software distribution company, where each of the
22kW while the design with HID-MH light sources offices contained at least two computers, showed
requires 38.6kW. Assumptions on the cleaning interval, through measurements that most employees felt
e.g. extending the cleaning interval from 1 to 3 years led comfortable with a lighting level of around 100 lux (as
to an increase of 9.5% in the installed power in order to opposed to the standard regulations of workplaces
always guarantee the required light levels. Changes in demanding 300 to 500 lux at desk level). Meanwhile, a
the reflectances and their impact were demonstrated by French field study (Escuyer and Fontoynont, 2001)
changing the wall reflectance from 50% to 10%. For the involving worker interviews in three office buildings,
HID-MH configuration for instance this again asked for distinguished between two distinct groups: a small group
a design with more fittings and therefore an increased spending more than 70% of their time working on the
installed power by 4.8% compared to the base case. computer, for which light levels were low (100-300 lux)
and a bigger group spending less than 70% of their time
Reduction of Maintained Illuminance Levels working on the computer for which light levels were
Boyce et al. (2006) claimed that lighting practice that higher (300-600 lx). These results are in line with those
uses 500 lx as the target for maintained illuminance is of an earlier French study by Berrutto et al. (1997).
excessive. According to these authors, by using 400 lx as However, in a more recent Finnish study by
a design criterion, a 20% decrease in energy Viitanen et al. (2013), lighting quality parameters were
consumption could be gained together with a likely studied in an office lighting setting for three different
increase in the percentage of office workers who are luminaire types: (1) square LED panel luminaire
within 100 lx of their preferred illuminance. (Sq_LED); (2) round LED downlight luminaire
No study was found where the issue of reduced (Ro_LED); and (3) rectangular recessed T5 fluorescent
illuminance level was specifically addressed in a lamp luminaire (Re_T5). Re_T5 lighting was
retrofit context. However, Baker (2013) discussed compared to Sq_LED lighting at 300, 600 and 1000
trends in new construction and retrofit lighting lx. Ro_LED lighting was studied at three different
projects as seen in four years of energy efficiency color temperatures: 3000, 4500 and 6000 K. The
incentive programs in Texas, USA and reported that subjects evaluated 600 lx to be equally pleasant to
in most cases, the number of lamps was reduced, 1000 lx and the reading task was evaluated to be
which is partly due to the ‘education around proper equally easy at these two illuminance levels.
light level as many facilities are currently over-lit’. However, 1000 lx caused slightly more glare and 300
Indeed, many studies (e.g. Galasiu et al., 2007; lx was considered to be less pleasant. Visual
Boyce et al., 2006; Moore et al., 2002; 2001; Veitch and performance regarding reading and detail distinction
Newsham, 2000; Newsham et al., 2008) have indicated on the wall was more difficult at 300 lx than at higher
that office workers generally prefer illuminance levels illuminance levels. At 600 lx, the amount of light was
that are lower than recommended by the standards considered to be more optimal than at 300 or 1000 lx.
particularly if they work with a computer most of the When the users adjusted illuminance, the overall

■■
Marie-Claude Dubois et al. / Energy Research Journal 2015, ■ (■): ■■.■■
DOI: 10.3844/erjsp.2015.■■.■■

average preferred illuminance was 648 lx for Re_T5 and how they are supposed to operate and when they
lighting and 517 lx for the LED lighting; but the know how to get short-comings of a system rectified,
authors noted that there were large variations in the they will be less likely to disable systems and become
preferred illuminances between subjects. more aware of the energy-saving mentality. Manuals
Finally, it might be worthwhile pointing out the should also explain the purpose of the light energy
potential impact of vertical illumination, rather than just saving technology or control system.
horizontal illuminance: if vertical surfaces appear well- Unfortunately, only a few studies have been found
lit, lower horizontal illuminance values might be addressing the energy saving potential related to
tolerated more easily. occupant behavior or evaluating the consequences for
human performance, health and well-being of energy-
Improvement in Spectral Quality of Light Sources saving lighting strategies in a retrofit context.
Better match between the lighting system’s spectral One study (Mahdavi et al., 2008) analyzed
qualities and the user’s visual response can provide an occupants’ operation of lighting and shading systems by
optimal, energy-efficient lighting solution. Rea et al. monitoring three office buildings from nine months to a
(2009) have shown, for instance, that they could achieve year. They found that the probability of switching the
energy savings (of the order of 37% according to Rea, light on upon arrival increased significantly when the
2010) in outdoor lighting applications by using Metal horizontal illuminance at the proximity of the
Halide (MH) lamps instead of the more common High- workstation was less than 200 lx. The same authors also
Pressure Sodium (HPS) since MH spectra are better obtained a probability model for switching the lights off
tuned to the spectral sensitivity of the human retina at as a function of the duration of absence from the offices.
mesopic light levels. Note that the illuminance ratio Another study (Coleman et al., 2013) demonstrated
between an MH and an HPS light source has been that an installed wireless system was found to help
measured to be about 0.7 for equivalent brightness individuals evaluate their energy-related behaviors and
perception in the high end of the mesopic luminance identify personal actions that are not apparent from
range (>0.1 cd/m2) (Rea, 1996; Fotios and Cheal, 2007). aggregated building-level feedback. Neither study
The same logic can be applied to indoor lighting provides clear data about the potential energy savings
situations. Rea (2010) indicated, for instance, that at the achievable through improved occupant behavior.
same brightness level, 6500 K T8 fluorescent lamps use Table 3 provides a summary of the potential energy
35% less energy than 3000 K T8 fluorescent lamps. savings achievable with the lighting retrofit strategies
A recent field study (Osterhaus, 2014) carried out at discussed in the last five sections of this literature review.
Horsens Hall in Denmark, where 2700 K fluorescent
lamps were retrofitted with 6000 K LED panels,
Use of Control Systems
indicated that the 6000 K lamps were judged to be The use of electric lighting control systems-in order
brighter than the 2700 K source despite the fact that they to provide light exactly at the right time, to the right
provided slightly lower illuminance values on the work level and in the right place - can significantly contribute
surface. Although this retrofit case involved two to reduce the consumption of electricity for lighting.
different types of light sources, it still suggests that the Recently, Boyano et al. (2013) presented key energy use
spectral light distribution of the light source is very figures and explored the energy saving potential in office
critical in terms of subjective brightness perception. buildings across Europe by simulating (with Energy
Plus) several currently available Energy Conservation
Occupant Behavior Measures (ECMs) for three representative locations
across Europe (cold, mild and warm climate). With
Masoso and Grobler (2010) claimed that ‘behavioural
partial daylight-linked dimming control (on 50% of
change has energy saving potential comparable and in office building facilities), they obtained a potential
most cases higher than that of technological solutions’. energy saving between 9 and 37% of the total energy
The most salient feature of behavioural change is that it consumed and 18 to 37% with total lighting control
is largely no cost, it needs no hi-tech knowledge, it is (100% of office building facilities).
readily applicable to both new and existing buildings, it Earlier, Fostervold et al. (2010) investigated the
is largely appreciated by many (though not practiced) potential for energy savings and possible consequences
and it has a self-perpetuating potential in that once for the workers by implementing new luminaires and a
occupants of a building have developed an energy new lighting control system in a large hospital building.
conservation culture, they spread it to their new They obtained reductions in lighting energy use by 55-
comers as well as take it with them to other places. It 75% (depending on the control system) with neither
might even be worthwhile addressing the need for positive nor negative effect on individual well-being and
good user manuals for buildings and their systems. concentration or negative outcomes of adaptive lighting
When occupants know how the systems are designed systems due to reduced degree of perceived control.

■■
Marie-Claude Dubois et al. / Energy Research Journal 2015, ■ (■): ■■.■■
DOI: 10.3844/erjsp.2015.■■.■■

Table 3. Potential energy savings by using specific lighting retrofit strategies

Potential energy savings Retrofit studies Source
Task-ambient 22-25% No Dubois and Blomsterberg (2011)
Improved maintenance 5-20%a Yes Gasparovsky and Raditschova (2013)
Mucklejohn et al. (2013)
Reduced maintained illuminance 20%b No Boyce et al. (2006)
Improved spectral quality of light source 35%c No Rea (2010)
Improved occupant behavior n/a No Mahdavi et al. (2008) Coleman et al. (2013)
a light loss if no maintenance program is applied
b from 500 to 400 lux
c for a specific lighting technology

Prior to this, Granderson and Agogino (2006) The authors also found that there is a consistent effect of
developed an intelligent lighting dimming system in the switch-off delay in the occupancy strategies.
order to balance user comfort, energy savings and retrofit It might be worthwhile pointing out that Simpson
costs using an influence diagram approach. This system (2003) claimed that office buildings are probably the
utilizes wireless sensing and actuation technology to most important application for lighting control systems,
relieve much of the expense associated with retrofitting. but also an application where individuals will likely have
In contrast to traditional systems that use a single strong opinions about lighting control. While systems
ceiling-mounted photosensor per control zone, the might have been installed with the best of intentions, he
intelligent system uses Smart Dust motes placed directly states that some ‘have been unsuccessful to the extent
on each work surface. Illuminance sensing is performed that users have disconnected the automatic element or
with photodiodes embedded on the motes, while even the entire lighting control system’. He argues that
occupancy sensing is accomplished with mote specifiers of lighting control systems need to be aware of
accelerometers fixed to occupants’ chairs, or with various factors when selecting a lighting control system:
commercial personal occupancy sensors. Validation and
fusion algorithms are used to mitigate interference from • People behave differently when lighting is under
the users. Smart dust motes offer significantly reduced central control
retrofit costs since they are wireless and directly • People of different ages and visual abilities have
interfaced with ballasts. They avoid the need to access different requirements and even those with the same
power lines behind the walls and ceiling of an office. age and ability might have different preferences
The authors (Granderson and Agogino, 2006) tested this • Occupancy times of spaces vary widely, especially
system by simulation and found that replacing the for private office and other work spaces
existing non-dimming system designed under previous • Unpredictability of lighting system behaviour is
Illuminating Engineering Society (IES) guidelines with a generally disliked
commercial dimming system would generate 13% • Very quick and very large changes in illuminance
energy savings and a 15% cost savings. On the other
levels are difficult to handle for the human eye
hand, the intelligent system increased the energy savings
to 26% and the cost savings to 20%. • The extent of daylight contributing to the workplace
In general, the saving potential varies greatly illumination typically varies significantly with the
according to context and building: different studies show distance to windows
different energy savings, which leads to difficulties in • The orientation of the workspace’s daylight
calculating the payback time of a lighting retrofit action. openings can result in highly seasonal or diurnal
Williams et al. (2012) tried to overcome this difficulty problems affecting the users
through a meta-analysis based on the review of 88 • The introduction of blinds and other shading
scientific papers and reports which included the potential devices affects the way in which automated
savings from lighting control systems. The authors lighting control systems work
categorized the different strategies and listed the study • Appropriate placement of light and/or presence
typology as well as the key features of each document. sensors is crucial for achieving user satisfaction
Applying increasing restrictive filters, they concluded and energy savings
that there is a potential saving of 24-38% for different • The type of occupancy of the space is a considerable
lighting control systems in actual installation. The study ‘human’ factor
also pointed out that the simulations generally
Manual Controls
overestimate the savings compared to field studies. The
effect, in this case, is higher when the lighting control Manual control systems, such as door switches,
system has a higher level of automation and/or manual task lamps and manual dimmers, can offer an
technology, such as daylight harvesting technologies. unexpectedly high saving potential. For example, a

■■
Marie-Claude Dubois et al. / Energy Research Journal 2015, ■ (■): ■■.■■
DOI: 10.3844/erjsp.2015.■■.■■

survey conducted in France (IEA, 2006) reported energy A recent article by Motta Cabrera and Zareipour
savings of up to 77% by installing more manual switches in (2013) presented an experimental research aiming to
open plan offices. Besides energy savings, the possibility of quantify and understand lighting energy waste patterns
controlling the light environment has a positive effect on in a post-secondary educational institute located in
the users’ mood according to Moore et al. (2002), which Calgary, Canada. They collected data over a full
is also related to monetary savings. Juslén et al. (2007) academic year in three typical classrooms. Data
have shown an increase of 4,5% in productivity in a association mining was used in order to extract
factory hall where manually dimmable task lights were association rules and explore lighting waste patterns.
provided to the workers. They made an energy assessment to account for the
In an earlier study carried out in the USA amount of energy, money and CO2 emissions spent by each
(Jennings et al., 2000), five different lighting control classroom throughout the year and obtained energy wastes
scenarios were tested in an office building located in of 126.4kWh/seat, 49.2kWh/seat and 62.8kWh/seat
San Francisco. Among these scenarios, two considered respectively for the three classrooms. The average number
the use of manual controls. In the first case, a bi-level of waste instances for all three classrooms was 44.24%,
switching gave the possibility to choose to turn on only a which means that the lights were turned on with no one in
part of the light fixtures. In the second scenario, the the classroom for 10.6 h in an average day. They finally
electric lighting was turning on automatically when demonstrated by simulation that if the waste patterns were
people were entering the offices, but it could be avoided, significant savings, up to 70% of the current
manually dimmed afterwards. The bi-level switching energy use, could be achieved.
offered about 23% energy savings compared with a
Another recent article (Itani et al., 2013) concerned
classic switch, while the second scenario (automatic on
the effect of Energy Conservation Measures (ECMs) for
with dimming) provided about 26% energy savings.
Nevertheless, the authors analysed the behaviour of the an existing eight-storey building located in Beirut,
users and found quite important differences in individual Lebanon. The authors analysed the impact of low
preferences. For example, in the bi-level switching case, investment and minimal disruption ECMs that can
about 63% of the occupants used mostly the full-light maintain thermal comfort and good indoor air quality.
setting, 13% used mainly 2/3 of the light fixtures and the The ECMs were Investigated by using a commercial
remaining used mostly 1/3 of the light fixtures. energy analysis software (IES-VE) and varying the
In general, the saving potential is not predictable indoor temperature cooling set point, lighting control,
because it largely depends on individual behaviours etc. They used a standard system audit methodology and
according to Boyce et al. (2000). In small office rooms, advanced energy modelling techniques to replicate the
the occupants tend to adjust the light level, which leads existing building base case. The lighting energy use in
to both a more pleasant lighting environment and energy the building studied used less percentage of total energy
savings (see also Love, 1998; Gentile et al., 2013). In compared to most ordinary office buildings due to the
open space offices, a strategy could be to provide use of efficient lights (T5 and CFLs with an average
manually adjustable task lighting, while keeping some LPD of 10.23 W/m2) and the large daylighting available
automatic controls for the general electric lighting. from the glazed façade. Substantial energy savings were
Recently, a solution that combined automation with achieved by implementing scheduled lighting controls
individual preferences was proposed by Wen and Agogino and by placing occupancy sensors in meeting rooms and
(2011). They proposed a lighting design method enabling private offices. The schedule of lighting was also adjusted
dynamic, personalized and optimal horizontal illumination by turning off some lights during unoccupied and low
of open-plan offices by using an elaborated control occupancy hours, which yielded reductions in the lighting
mechanism to tune each lamp in the office according to energy corresponding to 11.8% savings in lighting energy
each occupant’s preference and need. The prototype or a 2.6% saving in overall building energy consumption
lighting system was tested in an open-plan office. The due to a simultaneous decrease in lighting and cooling
overall energy savings for the year analyzed was 51% loads. The economic analysis showed that lighting control
compared to the original all on/off lighting configuration. and increasing temperature set point are two ECMs that
Occupancy Controls should be implemented because of short payback period of
1.3±0.2 years, respectively.
One of the most effective approaches to minimize According to Motta Cabrera and Zareipour (2013),
energy use in the non-residential sectors is by using one obstacle of implementing an occupancy-based
occupancy based lighting control systems (IEA, 2006; lighting-control is the uncertainty on the amount of
Garg and Bansal, 2000; Galasiu et al., 2007). As a result energy that could be saved. Previous research papers
of occupants not turning the lights off when they no report differences in expected savings, typically ranging
longer need them, more energy is spent on non-working from 25% to 75% (Garg and Bansal, 2000; Moore et al.,
hours than during scheduled time according to a study by 2003; Richman et al., 1995; Granderson and Agogino,
Masoso and Grobler (2010). 2006). This could be due to the fact that each space has a

■■
Marie-Claude Dubois et al. / Energy Research Journal 2015, ■ (■): ■■.■■
DOI: 10.3844/erjsp.2015.■■.■■

different occupancy profile based on the schedules and Daylight-Linked Control Systems
activities of people in the building (see e.g. Guo et al.,
2010; Rubinstein et al., 2003). Several studies have indicated that daylighting can
In addition to this, the switching strategy seems to provide a cost-effective alternative to electrical lighting
play a significant role. The occupancy control system for commercial and institutional buildings (Ihm et al.,
could automatically turn on/off the electric lighting when 2009). In addition, it is generally acknowledged that
the presence is detected (presence switches), or only daylight is preferred to electric light, fosters higher
switch off a manually turned on system when any productivity and performance (Plympton et al., 2000;
movement is recognized (absence switches). The Säter, 2010). According to many authors, simulation
differences in savings between these two approaches studies as well as field monitoring, daylighting controls
could be high, especially in individual or small offices, can result in significant lighting savings ranging from 30
as shown by Gentile et al. (2012). These authors showed to 77% (Doulos et al., 2008; Li et al., 2006; Lee and
that using a presence (on/off) control system in small Selkowitz, 2006; Onaygil and Guler, 2003; Ihm et al.,
offices would yield higher energy use for lighting than a 2009; Kobav and Bizjak, 2010). However, previous
simple manual switch at the door with an absence surveys have indicated that daylighting control strategies
detector (switch off), confirming results from previous are not commonly integrated in buildings (Li and Lam,
research (Voss et al., 2006). 2003). According to Ihm et al. (2009) and Krarti et al.
In an earlier study (Garg and Bansal, 2000), it was (2005), this may be explained by the lack of
found that by optimizing the time delay, energy savings simplified prediction tools.
from using an occupancy sensor increased from 20% to A recent field study (Chow et al., 2013) considered a
25%. In an earlier study in eight buildings (Richman et al., corridor space adjacent to a large skylight atrium. The
1995), which included conference rooms, mail room, combination of high daylight availability and low
restrooms, one training room and laboratory areas, it was illuminance requirements for corridor spaces, suggested
found that occupancy sensors had the potential to save the use of daylight-linked control system with efficient
between 24% and 79% of energy consumption by using T5 light fixtures. The solution led to an overall energy
a 10-min time delay. When the time delay was decreased saving of 93% compared to the existing lighting
to a 2-min setting, the potential savings range increased installation. The payback period for the proposed
to 76-93%, with negative impact on user comfort. solution was calculated to be 3.42 years.
Guo et al. (2010) presented a review of occupancy- Another study achieved in Hong Kong (Li et al.,
based lighting control systems where they analyzed the 2006) investigated a fully air conditioned side-lighted
typologies of sensors generally used for this kind of open plan office with an initial power density for
lighting control systems, as well as the settings generally lighting of 16.7 W/m2. The original two rows of
fluorescent lamps closer to the window were improved
applied during the installation. Regarding the sensors,
by adding new high frequency ballasts and a photo
while several technologies are available, the market
sensor for daylight harvesting. A single photo sensor
often offers only PIR (Passive Infrared), ultrasonic or
served all the upgraded fixtures. The authors obtained
hybrid PIR/ultrasonic presence sensors, which are average energy savings of 33% compared to the non-
offering a good compromise between feasibility, dimmable fixtures, with better performances during the
accuracy and costs. Each of these systems presents some central part of the day and the summer months.
limitation regarding the position of the sensor, the room In the Canadian climate, Galasiu et al. (2004) tested
area, the geometry, etc. This makes the savings different combinations of lighting and shading control
conditioned by proper installation and post-installation systems. For the electric lighting, the authors used
commissioning. Guo et al. (2010) concluded that a cheap dimmable and on/off daylight systems, which were
and feasible solution could be to have a network of combined with photo controlled blinds as well as with
sensors rather than a single expensive one. With focus on different cases of static positions of the blinds. The best
the settings, this review shows that the savings with 20 case scenario offered possible energy savings of 50-60%
min of delay could be as high as 46%, while it grows up with windows without blinds, which dropped by 5-80%
to 86% when delay is reduced to 5 min. with different static (predefined blind positions, not
As mentioned previously, shorter time delays for the automatically controlled) window blinds settings.
switch-off reduce the energy consumption, but could be An Italian study (Gugliermetti and Bisegna, 2005)
unacceptable for the occupants. The general performed in the Mediterranean area investigated the
recommendation is to keep 10-20 min time delay, never luminous and energy aspects related to the integration of
accepting shorter time delays than 7 min. Note also that control systems with different Electro Chromic (EC) and
the effectiveness of those systems is largely dependent double glazed systems equipped with motorized internal
on the pattern of use of the space. Generally, irregularly shading devices. On/off and linear control strategies
occupied spaces offer higher saving potential. were used to change the transparency of EC systems

■■
Marie-Claude Dubois et al. / Energy Research Journal 2015, ■ (■): ■■.■■
DOI: 10.3844/erjsp.2015.■■.■■

from clear to dark state and to close the indoor curtains, systems. A lack of awareness by the designers about the
while dimming and on/off strategies for managing the comprehensive performance of the real installations has
electric lighting. The study showed the significant also been identified (Ehrlich et al., 2002). Gentile et al.
impact of highly flexible controls of both electric and (2013) also pointed out the need for training installers to
natural light also in climates where overheating and calibrate these systems properly or simply to be trained
visual comfort problems are of great importance owing about the calibration settings of the systems they install.
to the high level of daylight. They found that the Table 4 summarizes the findings regarding potential
difference in lighting electric power demand can be largely energy savings using lighting control systems.
affected by the variation in the number of dimming zones.
And still more important are the different situations Use of Daylighting Systems
obtained by changing the daylighting control: a finer Building facades and roofs, by their glass area ratio,
regulation, such as that obtained by the linear control, shading or daylighting systems, may greatly affect
involves a more uniform and continuous, but lighter, electricity use for lighting, provided of course that electric
presence of electric lights with respect to simpler and less lighting is switched off in the presence of daylight. Many
flexible regulation. This has been proved for both internal studies report results related to daylight utilisation i.e. the
curtain control and ECs, from both energy efficiency and replacement of electric light by daylight.
visual comfort points of view. Sanati and Utzinger (2013) examined the effect of an
In an earlier study in a sub-tropical environment interior light shelf system fixed in the upper part of
(To et al., 2002) considered a side-lit classroom with windows on occupants’ use of blinds in the lower part
two rows of fluorescent tubes parallel to the windows. and on the consumption of electric lighting. The results
The closest row to the windows was replaced with a high suggest that in otherwise identical environmental
frequency electronic daylight linked dimming system. In conditions, occupants working in the ‘light shelf zone’
addition, the general horizontal illuminance was reduced demonstrated a lower window occlusion than those located
from about 1000 lux to around 800 lux. The authors in the area with conventional windows. Light shelves
extrapolated the potential annual savings using a 16- distributed daylight more evenly, consequently, occupants
weeks data collection period, which demonstrated in the ‘light shelf zone’ used less electric lighting.
savings of the order of 40% compared to the full-power Previously, daylighting systems have been developed
scenario. Considering the actual installation costs, the to enhance daylight penetration or utilization. A large
energy savings led to a payback period of 4.9 years for number of daylighting systems were evaluated within
the tested installation, with projection of possible IEA Task 21 ‘Daylight in Buildings’ (IEA, 2000) based
reduction up to 2.2 years for larger spaces. on their ability to block or redirect daylight. Depending
Koyle and Papamichael (2010) installed an on the geographical location and its predominant
innovative dual-loop photo sensor control system in a daylighting conditions different daylighting systems
150000 ft2 retail store. The system received readings seem to be suitable. Energy savings for electric lighting
from both open- and closed-loop sensors. It combined can be achieved with (angular) selective systems, such as
the information through an algorithm and determined the anidolic solar blinds, using direct sunlight without glare,
relative requested electric light output. Over a 12-month which is of specific relevance for mild and sunny
observation period, the authors found that the system climates. Sunlight-redirecting daylighting systems
was able to match the requested light levels 63.7% of the positioned in the upper part of the window plane, such as
time and saved 36.6% energy compared to a retail store laser-cut-panels and prismatic panels, present energy
without daylight harvesting strategy. The payback time saving potential as well, but need consideration with
of this installation was determined to be 2.4 years. respect to position and angle to avoid glare. These
In addition to the uncertainty in predicting the energy systems are typically applied in sunny climates
saving potential, several studies reported difficulties in according to Edmonds and Greenup (2002), even though
real installation of daylight-linked dimming systems (e.g. the laser-cut panel has shown its applicability in
Lee and Selkowitz, 2006; Gentile et al., 2012). The temperate climates as well (IEA, 2000). Other, more
practical difficulties regard mainly the performance of invasive, light redirecting systems, like light shelves and
the photo sensor (Ehrlich et al., 2002), since the whole anidolic systems, redirect both diffuse and sunlight. Both
system is based on its reading of the light environment. can increase daylight penetration, but might reduce the
Frequent light switching under unstable sky conditions sun-and daylight contribution near the façade. They have
may compromise the savings (Li et al., 2010). In limited application in high-latitude countries, because of
addition, difficulties in matching the illuminance design the additionally required shading device for a prevalent
levels because of overestimated number of light fixtures, time of the year. The most efficient daylighting systems
changes in the space purpose (Choi and Sung, 2000) or for moderate climates seem to be automatically
actual occupancy rate of the space (Roisin et al., 2008) controlled blinds and louvres, because of their flexibility
are also factors that may compromise the efficiency of the to respond to different daylighting conditions.

■■
Marie-Claude Dubois et al. / Energy Research Journal 2015, ■ (■): ■■.■■
DOI: 10.3844/erjsp.2015.■■.■■

Table 4. Potential energy savings by using different types of lighting control systems
Potential energy savings Retrofit studies Source
Manual controls 23-77% Yes IEA (2006); Jennings et al. (2000)
Scheduling 12% No Itani et al. (2013)
Occupancy control 20-93%a Yes Motta Cabrera and Zareipour (2013); Garg and Bansal (2000);
Moore et al. (2003); Richman et al. (1995); Guo et al. (2010)
Daylight-linked 10-93% Yes Boyano et al. (2013); Doulos et al. (2008); Li et al. (2006);
dimming Lee and Selkowitz (2006); Onaygil and Guler (2003);
Ihm et al. (2009); Kobav and Bizjak (2010);
Chow et al. (2013); Li et al. (2006); To et al. (2002);
Koyle and Papamichael (2010)
Combined daylight 26% No Granderson and Agonino (2006)
-linked and occupancy
a Highly dependent on space occupancy and time delay

Ehling (2000) investigated the energy savings • The so-called rebound effect may create a tendency
potential and economical aspects of daylighting systems to use more light because it is cheaper and by that
under moderate climate conditions and concluded the absolute consumption could be ultimately increased.
pay back times for daylighting systems are typically • Energy savings measures should be considered in a
extensive therefore also pleading for simple, cost holistic way since electric lighting reductions
effective, daylighting systems, such as blinds and normally entail an increase in heating demand.
louvres. These systems perform well under predominant Improvements in lighting should be planned along
sunny sky conditions as well.
with building envelope improvements to
A recent literature review (Nair et al., 2013)
compensate for the related increase in heating loads
concluded that active systems, with for example sun
tracking mirrors or lenses, can optimally collect daylight Reductions in energy use are not necessarily a linear
and with this offer high energy saving potential. addition of the savings from individual Energy Retrofit
Nonetheless, they are typically complex and relatively Measures (ERMs). The overall reduction in energy
expensive and require regular maintenance. Passive consumption is generally less than the linear addition of
systems have a poorer performance, but are typically the individual ERMs With respect to lighting retrofit
cheap, simple and require less maintenance. Positive strategies, the literature review shows the following
attributes of a good daylighting enhancement system are
said to be passiveness, ease of installation, visual Electric Lighting
acceptance, solar shading against direct radiation and
well-controlled output distribution. • Replacement of lamp, ballast and luminaire appears
as the most often reported lighting retrofit strategy,
with a great saving potential. The most common
Conclusion existing lighting installations consist of fluorescent
As part of the international research ‘IEA-Task 50- lighting (with conventional ballasts) and most
Advanced Lighting Solutions for Retrofitting Buildings’, commonly retrofitted fixtures are the 4-lamp T12
this literature review pursued the aim to analyze and parabolic and lenses troffers with T12 or older
information found in the scientific literature, previous T8 lamps (data from the USA)
international, European and national research projects in • Compared to fluorescent lighting, LED lamps have
order to summarize the state of knowledge on lighting reduced energy consumption (approximately 50%) and
and daylighting retrofit. a longer life time. Although good products are
Key conclusions that should be remembered from available, lighting quality aspects such as
this literature review are stated below: unsatisfactory color rendering, low light load, flicker
General and poor light distribution have been reported and need
to be considered seriously to ensure user satisfaction
• Electric lighting is one of the major sources of • Reducing the maintained illuminance level is
electricity consumption in buildings representing 15- another promising strategy since previous research
60% of the final energy use. It has a high saving indicated lower preferred illuminance levels
potential at a reasonable pay-back period, especially compared to those recommended by the standards
due to the development of new lighting technologies particularly in areas where computers are used.
with higher luminous efficacies and lower cost of light There are indications of a tendency to reduce the
• Reported energy savings through lighting retrofit number of lamps (by ‘delamping’) partly due to the
vary widely depending on initial energy use, education around proper light levels and the fact that
building type, usage, etc many facilities are currently overlit

■■
Marie-Claude Dubois et al. / Energy Research Journal 2015, ■ (■): ■■.■■
DOI: 10.3844/erjsp.2015.■■.■■

• Use of task-ambient lighting design has proven to was limited to the topic of energy efficiency but the
provide better quality lighting and a 22-25% reader should be reminded that retrofitting a lighting
reduction in electricity use compared to a standard installation offers several advantages besides energy
general energy-efficient lighting installation but no savings: improvement in lighting quality, occupant
study has been found about task-ambient lighting satisfaction and productivity, improved corporate image,
design approach in a retrofit context energy security, etc. The review generally shows that
studies of lighting retrofit in real context with monitored
Occupant Behavior data are surprisingly rare and most of the existing studies
• Occupant behavior offers substantial energy saving target either lamp-ballast-luminaire replacement or
potential but this strategy has not been sufficiently implementation of advanced control systems. Monitoring
explored in retrofit context studies where simple and robust retrofit strategies such
as task-ambient lighting design, improved occupant
Lighting Control behavior, improvement in the spectral quality of light
sources, or even a simple reduction of maintained
• The use of electric lighting control systems can
illuminance levels have not been reported extensively in
also significantly reduce the consumption of
the literature despite their promising saving potential.
electric lighting but the saving potential varies
This review suggests that research efforts addressing
greatly according to context and building, which
these specific strategies should be emphasized in the
leads to difficulties in estimating the payback
context of retrofitting buildings.
time of a lighting retrofit
One of the next tasks of IEA Task 50, Subtask D, will
• Simulations generally overestimate the savings hopefully contribute to bridge this gap by providing
compared to field studies; especially when the measured data on a number of real retrofitted buildings.
control system involves advanced automation and/or This will provide more knowledge in this area, which
technology, such as daylight harvesting technologies will pave the way for further developments and
• Manual control systems, such as door switches, acceleration of the lighting and daylighting retrofit
manual task lamps and manual dimmers, can offer activities worldwide.
an unexpectedly high saving potential with increase
in occupant satisfaction and productivity Acknowledgement
• Occupancy based lighting control systems are also very
promising with high expected savings (20-93%) The authors thank Jan de Boer and Anna Hoier from
• Irregularly occupied spaces offer higher saving the Fraunhofer Institute for Building Physics (Stuttgart)
potential for reviewing the last version of this article and
• Optimizing the time delay has a significant impact providing fruitful comments and discussion.
on the energy savings
• Using a presence (on/off) control system could yield Funding Information
higher energy use for lighting than a simple manual The authors thank their respective funding agencies
switch at the door combined with absence detection for supporting their work as part of IEA Task 50:
(switch off), especially in individual or small offices
• Daylight-linked control systems can result in • Swedish Energy Agency (Statens energimyndighet),
significant lighting savings, but several studies Sweden
reported difficulties in real installations and in • Finnish RYM Indoor Environment Program, Finland
estimating the payback period at the design stage ([Link]
program/)
Daylighting Systems • The Research Council of Norway, Norway
• Building facades, by their glass area ratio, shading • Bundesministerium für Wirtschaft und Technologie,
or daylighting systems, may greatly affect electricity Germany
use for lighting provided that electric lights are
switched off in presence of sufficient daylight Author’s Contributions
• Payback times for daylighting systems are typically
Marie-Claude Dubois: The article was written,
extensive while passive daylighting or shading
coordinated and edited by.
systems have a poorer performance, but are typically
Niko Gentile: The section on control systems was
cheap, simple and require less maintenance, leading
written by.
to better payback times
Fabio Bisegna: Contributed with some reading notes.
This review discussed several strategies for reducing Eino Tetri: Contributed to the section on occupant
electricity use in lighting retrofit projects. The review behavior.

■■
Marie-Claude Dubois et al. / Energy Research Journal 2015, ■ (■): ■■.■■
DOI: 10.3844/erjsp.2015.■■.■■

Barbara Matusiak: Mainly contributed to the CIE, 2013. Review of lighting quality measures for
section on daylighting systems. interior lighting with LED lighting systems.
Martine Knoop: The section about light technologies. Proceedings of the CIE Lighting Quality and Energy
Werner Osterhaus: Provided contributions in Efficiency Conference, Mar. 3-5, Central Bureau,
different sections but mostly on control systems and by Vienna, Austria.
reviewing the final version. Coleman, M.J., K.N. Irvine, M. Lemon and L. Shao,
2013. Promoting behaviour change through
Ethics personalized energy feedback in offices. Build. Res.
Inf., 41: 637-651.
This article is original and contains unpublished DOI: 10.1080/09613218.2013.808958
material. The corresponding author confirms that all of Dascalaki, E. and M. Santamouris, 2002. On the potential
the other authors have read and approved the manuscript of retrofitting scenarios for offices. Build. Environ., 37:
and no ethical issues involved. 557-567. DOI: 10.1016/S0360-1323(02)00002-1
Doulos, L., A. Tsangrassoulis and F. Topalis, 2008.
References Quantifying energy savings in daylight responsive
AEFEETS, 2010. Real Prospects for Energy Efficiency systems: The role of dimming electronic ballasts.
in the United States. 1st Edn., National Academies Energy Build., 40: 36-50.
Press, ISBN-10: 0309156866, pp: 352. DOI: 10.1016/[Link].2007.01.019
Baker, S.C., 2013. A bright future: Market Dubois, M.C. and Å. Blomsterberg, 2011. Energy saving
transformation in texas lighting efficiency incentive potential and strategies for electric lighting in future
programs. Energy Eng., 110: 35-36. North European, low energy office buildings: A
Berrutto, V., M. Fontoynont and P. Avouac-Bastie, literature review. Energy Build., 43: 2572-2582.
1997. Importance of wall luminance on users' DOI: 10.1016/[Link].2011.07.001
satisfaction: Pilot study on 73 office workers. Eames, M., T. Dixon, T. May and M. Hunt, 2013. City
Proceedings of the 8th European Lighting futures: Exploring urban retrofit and sustainable
Conference, Amsterdam, The Netherlands. transitions. Building Res. Inf., 41: 504-516.
BBP, 2010. Low carbon retrofit toolkit: A roadmap to DOI: 10.1080/09613218.2013.805063
success. London, UK Better Buildings Partnership. Edmonds, I.R. and P.J. Greenup, 2002. Daylighting in
Boyano, A., P. Hernandez and O. Wolf, 2013. Energy the tropics. Solar Energy, 73: 111-121.
demands and potential savings in European office DOI: 10.1016/S0038-092X(02)00039-7
buildings: Case studies based on energy plus Ehling, K., 2000. Tageslichtsysteme: Lichttechnische
simulations. Energy Build., 65: 19-28. Bewertung und Wirtschaftlichkeit. 1st Edn., VDI-
DOI: 10.1016/[Link].2013.05.039 Verlag, Berlin, ISBN-10: 318316504X, pp: 117.
Boyce, P.R., N.H. Eklund and S.N. Simpson, 2000. Ehrlich, C., K. Papamichael, J. Lai and K. Revzan, 2002.
Individual lighting control: Task performance, mood
A method for simulating the performance of
and illuminance. J. Illuminating Eng. Society, 29:
299-311. DOI: 10.1080/00994480.2000.10748488 photosensor-based lighting controls. Energy Build.,
Boyce, P.R., J.A. Veitch, G.R. Newsham, C.C. Jones 34: 883-889.
and J.M. Heerwagen et al., 2006. Occupant use of DOI: 10.1016/S0378-7788(02)00064-6
switching and dimming controls in offices. Enkvist, P.A., T. Nauclér and J. Rosander, 2007. A Cost
Lighting Res. Technol., 38: 358-378. Curve for Greenhouse Gas Reduction. 1st Edn.,
DOI: 10.1177/1477153506070994 McKinsey Quarterly, pp: 1-7.
Chidiac, S.E., E. Catania, E. Morofsky and S. Foo, 2011. Erhorn-Kluttig, H., H. Erhorn and S. Wössner, 2004.
Effectiveness of single and multiple energy retrofit Exemplary retrofit concepts in Europe: Energy-
measures on the energy consumption of office efficient public buildings (results from IEA/EU-
buildings. Energy, 36: 5037-5052. projects).
DOI: 10.1016/[Link].2011.05.050 Escuyer, S. and M. Fontoynont, 2001. Lighting controls:
Choi, A.S. and M.K. Sung, 2000. Development of a A field study of office workers' reactions. Lighting
daylight responsive dimming system and Res. Technol., 33: 77-96.
preliminary evaluation of system performance. DOI: 10.1177/136578280103300202
Build. Environ., 35: 663-676.
Fostervold, K.I., P.J. Larsen, E. Lillelien, T. Mjøs and
DOI: 10.1016/S0360-1323(99)00055-4
Chow, S.K., D.H. Li, E.W. Lee and J.C. Lam, 2013. M.O. Berg, 2010. Energy efficient lighting control
Analysis and prediction of daylighting and energy systems: Consequences for lighting quality,
performance in atrium spaces using daylight-linked environment, health and human factors. Proceedings
lighting controls. Applied Energy, 112: 1016-1024. of the CIE Lighting Quality and Energy Efficiency,
DOI: 10.1016/[Link].2012.12.033 (CLQ’ 10), Vienna, Austria, pp: 368-375.

■■
Marie-Claude Dubois et al. / Energy Research Journal 2015, ■ (■): ■■.■■
DOI: 10.3844/erjsp.2015.■■.■■

Fotios, S.A. and C. Cheal, 2007. Lighting for subsidiary IEA, 2000. Daylight in Buildings: A Source Book on
streets: Investigation of lamps of different SPD, Part Daylighting Systems and Components. A Report of
2 - Brightness. Lighting Res. Technol., 29: 233-252. IEA SHC Task 21, ECBCS Annex 29. 1st Edn.,
DOI: 10.1177/1477153507080805 International Energy Agency, Washington.
Galasiu, A.D., M.R. Atif and R.A. MacDonald, 2004. IEA, 2006. Light's Labour's Lost: Policies for Energy-
Impact of window blinds on daylight-linked dimming Efficient Lighting. 1st Edn., OECD/IEA, Paris,
and automatic on/off lighting controls. Solar Energy, pp: 558.
76: 523-544. DOI: 10.1016/[Link].2003.12.007 Ihm, P., A. Nemri and M. Krarti, 2009. Estimation of
Galasiu, A.D., G.R. Newsham, C. Suvagau and D.M. lighting energy savings from daylighting. Build.
Sander, 2007. Energy saving lighting control Environ., 44: 509-514.
systems for open-plan offices: A field study. J. DOI: 10.1016/[Link].2008.04.016
Illuminating Eng. Society North Am., 4: 7-29. Itani, T., N. Ghaddar and K. Ghali, 2013. Strategies for
DOI: 10.1582/LEUKOS.2007.04.01.001 reducing energy consumption in existing office
Garg, V. and N. Bansal, 2000. Smart occupancy sensors buildings. Int. J. Sustainable Energy, 32: 259-275.
to reduce energy consumption. Energy Buildings, DOI: 10.1080/14786451.2011.622765
32: 81-87. DOI: 10.1016/S0378-7788(99)00040-7 Jennings, J.D., F.M. Rubinstein, D. DiBartolomeo and S.L.
Gasparovsky, D. and J. Raditschova, 2013. Lighting Blanc, 2000. Comparison of control options in private
properties and efficiency of luminaires exceeding offices in an advanced lighting controls testbed
their lifetime. Proceedings of CIE Centenary submitted to the offices. Illuminating Eng. Society, 29:
Conference, Paris, France, pp. 317-326. 39-60. DOI: 10.1080/00994480.2000.10748316
Gentile, N., H. Håkansson and M.C. Dubois, 2012. Juslén, H., M. Wouters and A. Tenner, 2007. The influence
Lighting control systems in individual offices at of controllable task-lighting on productivity: A field
high latitude: measurement of lighting conditions study in a factory. Applied Ergonomics, 38: 39-44.
and electricity savings. Proceedings of the DOI: 10.1016/[Link].2006.01.005
International Conference on Energy Efficiency in Kobav, M.B. and G. Bizjak, 2010. Long term study:
Commercial Buildings, (IEECB '12), Frankfurt, Energy savings obtained with use of daylight sensor
Germany, pp: 333-344 and dimming ballasts. Proceedings of the CIE
Gentile, N., T. Laike and M.C. Dubois, 2013. Lighting Lighting Quality and Energy Efficiency, (CLQ’ 10),
control systems in peripheral offices rooms at high Vienna, Austria, pp: 618-621.
latitude: Measurements of electricity savings and Koyle, B. and K. Papamichael, 2010. Dual-loop
users' preferences. Energy Procedia, 57: 1987-1996. photosensor control systems: Reliable, cost-effective
DOI: 10.1016/[Link].2014.10.063 lighting control for skylight applications.
Granderson, J. and A. Agogino, 2006. Intelligent office Proceedings of the ACEEE Summer Study on
lighting: Demand-responsive conditioning and Energy Efficiency in Buildings, (ASSE’ 10), Pacific
increased user satisfaction. J. Illuminating Grove, California, pp: 157-166.
Engineering Society North Am. J. Illuminating Eng. Krarti, M., P.M. Erickson and T.C. Hillman, 2005. A
Society North Am., 2: 185-198. simplified method to estimate energy savings of
Gugliermetti, F. and F. Bisegna, 2005. A model study of artificial lighting use from daylighting. Build.
light control systems operating with Electrochromic Environ., 40: 747-754.
Windows. Lighting Res. Technol., 37: 3-20. DOI: 10.1016/[Link].2004.08.007
DOI: 10.1191/1365782805li123oa Labayrade, R. and P. Avouac, 2013. Visual quality
Guo, X., D.K. Tiller, G.P. Henze and C.E. Waters, 2010. assessment of LED spots in comparison to low-
The performance of occupancy-based lighting voltage halogen spots. Proceedings of Centenary
control systems: A review. Lighting Res. Technol., Conference, (CIE’ 13), Paris, France, pp: 348-377.
42: 415-431. DOI: 10.1177/1477153510376225 Lee, E.S. and S.E. Selkowitz, 2006. The New York
times headquarters daylighting mockup:
Hanselaer, P., C. Lootens, W.R. Ryckaert, G. Deconinck
Monitored performance of the daylighting control
and P. Rombauts, 2007. Power density targets for
system. Energy Build., 38: 914-929.
efficient lighting of interior task areas. Lighting Res. DOI: 10.1016/[Link].2006.03.019
Technol., 39: 171-184. Lehman, B., A.J. Wilkins, S.M. Berman, M.E.
DOI: 10.1177/1365782807076737 Poplawski and N.J. Miller, 2011. Proposing
Hestnes, A.G. and N.U. Kofoed, 2002. Effective retrofitting measures of flicker in the low frequencies for
scenarios for energy efficiency and comfort: Results of lighting applications. Proceedings of the Energy
the design and evaluation activities within the Conversion Congress and Exposition, Sept. 17-22,
office project. Building Environ., 37: 569-574. IEEE Xplore Press, Phoenix, pp: 2865-2872.
DOI: 10.1016/S0360-1323(02)00003-3 DOI: 10.1109/ECCE.2011.6064154

■■
Marie-Claude Dubois et al. / Energy Research Journal 2015, ■ (■): ■■.■■
DOI: 10.3844/erjsp.2015.■■.■■

Li, D.H., K.L. Cheung, S.L. Wong and T.N. Lam, 2010. NC, 2012. Life-cycle assessment of energy and
An analysis of energy-efficient light fittings and environmental impacts of LED lighting products,
lighting controls. Applied Energy, 87: 558-567. part I: Review of the life-cycle energy consumption
DOI: 10.1016/[Link].2009.07.002 of incandescent, compact fluorescent and LED
Li, D.H., T.N. Lam and S.L. Wong, 2006. Lighting lamps. Department Energy.
and energy performance for an office using high Newsham, G.R., S. Mancini and R.G. Marchand, 2008.
frequency dimming controls. Energy Conversion Detection and acceptance of demand-responsive
Manage., 47: 1133-1145. lighting in offices with and without daylighting.
DOI: 10.1016/[Link].2005.06.016 LEUKOS: J. Illuminating Eng. Society North Am., 4:
Li, D. and J. Lam, 2003. An investigation of daylighting 139-156. DOI: 10.1582/LEUKOS.2007.004.03.001
performance and energy saving in daylit corridor. Onaygil, S. and O. Guler, 2003. Determination of the
Energy Build., 35: 356-373. energy saving by daylight responsive lighting
DOI: 10.1016/S0378-7788(02)00107-X control system with an example from Istanbul.
Love, J., 1998. Manual switching patterns in private Building Environ., 38: 973-977.
offices. Lighting Res. Technol., 30: 45-50. DOI: 10.1016/S0360-1323(03)00034-9
DOI: 10.1177/096032719803000107 Osterhaus, W., 2014. Daylight and electric light
Mahdavi, A., A. Mohammadi, E. Kabir and L. monitoring at horsens hall in denmark. Internal
Lambeva, 2008. Occupants’ operation of lighting report IEA Task 50 Advanced Lighting Solution for
and shading systems in office buildings. J. Build. Retrofitting Buildings.
Performance Simulation, 1: 57-65. Plympton, P., S. Conway and K. Epstein, 2000.
DOI: 10.1080/19401490801906502 Daylighting in schools: Improving student
Mahlia, T., H. Abdul Razak and M. Nursahida, 2011. performance and health at a price schools can
Life cycle cost analysis and payback period of afford. National Renewable Energy Laboratory
lighting retrofit at the University of Malaya. (NREL), Golden, Colorado.
Renewable Sustainable Energy Rev., 15: 1125-1132. Polimeni, J., K. Mayumi, M. Giampietro and B. Alcott,
DOI: 10.1016/[Link].2010.10.014 2007. The Jevons Paradox and the Myth of
Masoso, O. and L. Grobler, 2010. The dark side of Resource Efficiency Improvements. 1st Edn.,
occupants behaviour on building energy use. Routhledge, ISBN-10: 1844074625, pp: 200.
Energy Build., 42: 173-177. Poplawski, M.E. and N.M. Miller, 2013. Flicker in solid-
DOI: 10.1016/[Link].2009.08.009 state lighting: Measurement techniques and proposed
Moore, T., D.J. Carter and A.I. Slater, 2001. A reporting and application criteria. Proceedings of the
comparative study of user opinion in offices with and CIE Centenary Conference, (CC’ 13), Paris, France.
without individually controlled lighting. Proceedings Porritt, J., S. Tulej and S.A. Mucklejohn, 2013. The
of the 9th European Lighting Conference Lux rebound effect-an overview of the implications for
Europa, (ELC’ 01), Reykjavik, Iceland. lighting energy. Proceedings of the CIE Centenary
Moore, T., D.J. Carter and A.I. Slater, 2002. User Conference, Paris, France.
attitudes toward occupant controlled office lighting. Rea, M.S., 1996. Essay by invitation. Lighting Design
Lighting Res. Technol., 34: 207-219. Application, 26: 15-16.
DOI: 10.1191/1365782802lt048oa Rea, M.S., 2010. Energy-efficient lighting. Energy
Moore, T., D.J. Carter and A.I. Slater, 2003. Long-term efficient lighting in a human perspective.
patterns of use of occupant controlled office Proccedings of conference CEEBEL (Centrum för
lighting. Lighting Res. Technol., 35: 43-57. energieffektiv belysning), Katrineholm, Sweden.
DOI: 10.1191/1477153503li061oa Rea, M.S., J.D. Bullough and Y. Akashi, 2009. Several
Motta Cabrera, D.F. and H. Zareipour, 2013. Data views of metal halide and high-pressure sodium
association mining for identifying lighting energy lighting for outdoor applications. Lighting Res.
Technol., 41: 297-320.
waste patterns in educational institutes. Energy Buildi.,
DOI: 10.1177/1477153509102342
62 : 210-216. DOI : 10.1016/[Link].2013.02.049
Richman, E.E., A.L. Dittmer and J.M. Keller, 1995.
Mucklejohn, S.A., A.L. Whittaker and J. Gore, 2013.
Field analysis of occupancy sensor operation:
Unrevealing efficacy, maintenance and lighting Parameters affecting lighting energy savings.
energy for the end user. Proceedings of the CIE Illuminating Eng. Society North Am.
Centenary Conference, (CC’ 13), Paris, France. Roisin, B., M. Bodart, A. Deneyer and P. D’Herdt,
Nair, M.G., K. Ramamurthy and A.R. Ganesan, 2013. 2008. Lighting energy savings in offices using
Classification of indoor daylight enhancement different control systems and their real consumption.
systems. Lighting Res. Technol., 46: 245-267. Energy Build., 40: 514-523.
DOI: 10.1177/1477153513483299 DOI: 10.1016/[Link].2007.04.006

■■
 Marie-Claude Dubois et al. / Energy Research Journal 2015, ■ (■): ■■.■■
DOI: 10.3844/erjsp.2015.■■.■■

Rubinstein, F., N. Colak, J. Jennings and D. Neils, 2003. Tsao, J.Y. and P. Waide, 2010. The world’s apetite for
Analyzing occupancy profiles from a lighting light: Empirical data and trends spanning three
controls field study. Proceedings of CIE Session. centuries and six continents. LEUKOS: J.
San Diego, California. Illuminating Eng. Society North Am., 6: 259-281.
Ryckaert, W.R., I.A. Roelants, M. van Gils, G. Durinck U.S. Department of Energy, 2010. DOE solid-state
and S. Forment et al., 2011. Performance of LED lighting caliper program. Summary of results:
linear replacement lamps. CIE, 2011. Proceedings of Round 11 of product Testing. Pacific Northwest
the 27th Session of the CIE. Sun City, South Africa. National Laboratory.
Sanati, L. and M. Utzinger, 2013. The effect of Veitch, J.A. and G.R. Newsham, 2000. Preferred luminous
window shading design on occupant use of blinds conditions in open-plan offices: Research and
and electric lighting. Building Environ., 64: 67-76. practice recommendations. Lighting Res.
DOI: 10.1016/[Link].2013.02.013 Technol., 32: 199-212.
Saunders, H.D. and J.Y. Tsao, 2012. Rebound effects for DOI: 10.1177/096032710003200404
lighting. Energy Policy, 49: 477-478. Wen, Y.J. and A.M. Agogino, 2011. Personalized
DOI: 10.1016/[Link].2012.06.050 dynamic design of networked lighting for energy-
Schuler, M., 1995. Building simulation in application: efficiency in open-plan offices. Energy Build., 43:
Developing concepts for low energy buildings 1919-1924. DOI: 10.1016/[Link].2011.03.036
through co-operation between architects and Viitanen, J., J. Lehtovaara, E. Tetri and L. Halonen,
engineer. Proceedings of the Solar World 2013. User preferences in office lighting: A case
Congress, International Solar Energy Society study comparing LED and T5 lighting. LEUKOS: J.
(ISES), Harare, Zimbabwe. Illuminating Eng. Society North Am., 9: 261-290.
Simpson, R., 2003. Lighting Control: Technology and Williams, A., B. Atkinson, K. Garbesi, E. Page and F.
Applications. Focal Press, Oxford. Rubinstein et al., 2012. Lighting controls in
ISBN-13: 978-0240515663. commercial buildings. J. Illuminating Eng. Society
Spyropoulos, G.N. and C.A. Balaras, 2011. Energy North Am., 8: 161-180.
consumption and the potential of energy savings in Vogel, K., 2012. Strategies for lighting retrofits. LD+A,
Hellenic office buildings used as bank branches-A The magazine of the Illuminating Engineering.
case study. Energy Build., 43: 770-778.
Voss, K., G. Löhnert, S. Herkel, A. Wagner and M.
DOI: 10.1016/[Link].2010.12.015
Wambsganß, 2006. Bürogebäude mit Zukunft:
Stafford, A., C. Gorse and L. Shao, 2011. The Retrofit
Konzepte-Analysen-Erfahrungen. 2nd Edn.,
Challenge: Delivering low carbon buildings. Centre
for Low Carbon Futures, Leeds, UK. Solarpraxis, Berlin, Germany.
Säter, M., 2010. Psychological, physiological and visual Zhenjun, M., P. Cooper, D. Daly and L. Ledo, 2012.
responses to electromagnetic radiation in natural and Existing building retrofits: Methodology and
artificial light. Proceedings of CIE Lighting Quality state-of-the-art. Energy Buildi., 55: 889-902.
and Energy Efficiency (LQEE’ 10), Vienna, Austria, DOI: 10.1016/[Link].2012.08.018
pp: 422-425. Zmeureanu, R. and C. Peragine, 1999. Evaluation of
To, D.W., L.K. Sing, T.M. Chung and C.S. Leung, interactions between lighting and HVAC systems in
2002. Potential energy saving for a side-lit room a large commercial building. Energy Conversion
using daylight-linked fluorescent lamp Manage., 40: 1229-1236.
installations. Lighting Res. Technol., 34: 121-132. DOI: 10.1016/S0196-8904(99)00011-4
DOI: 10.1191/1365782802li038oa

■■

View publication stats