Investigating energy savings when using iot devices in buildings: a case study in the uk

Abstract

This study investigates the balance of emission levels from smart buildings with embedded IoT sensors against energy reduction using Integrated Environmental Solution Virtual Environment software. Findings show that the annual energy consumption from smart buildings was reduced by over 38% with smart HVAC and smart lights. The embodied emission level from the smart building increased by 7% at over 2 kgCO2/m2/yr, a drawback that should be considered during the production of IoT materials used in buildings. This study recommends that real-time monitoring, measurement and analyses are carried out to increase renewables penetration into the energy mix.

Full text

2022 European Conference on Computing in Construction
Ixia, Rhodes, Greece
July 24-26, 2022
INVESTIGATING ENERGY SAVINGS WHEN USING IOT DEVICES IN BUILDINGS: A
CASE STUDY IN THE UK
Adole Onuh Onuh1, Haibo Feng1, Qian Chen2 and Borja Garcia de Soto2
1 Department of Mechanical and Construction Engineering, Northumbria University, Newcastle, UK
2 S.M.A.R.T. Construction Research Group, Division of Engineering, New York University Abu Dhabi (NYUAD),
Experimental Research Building, Saadiyat Island, P.O. Box 129188, Abu Dhabi, United Arab Emirates
Abstract
Smart buildings are described to be efficient in their daily
operation by integrating IoT technologies into the
building systems (e.g., lighting and HVAC). However,
concerns have been raised about how and if they perform
this task. This study investigates the balance of emission
levels from smart buildings with embedded IoT sensors
against energy reduction using Integrated Environmental
Solution Virtual Environment software. Findings show
that the annual energy consumption from smart buildings
was reduced by over 38% with smart HVAC and smart
lights. The embodied emission level from the smart
building increased by 7% at over 2 kgCO2/m2/yr, a
drawback that should be considered during the production
of IoT materials used in buildings. This study
recommends that real-time monitoring, measurement and
analyses are carried out to increase potential renewables
penetration into the energy mix.
Keywords: energy saving; IoT; sensors; smart building;
IES VE simulation.
Introduction
The energy crisis is happening, and it will only worsen
given the current increases in energy demand if no drastic
measures are put in place to improve the existing
production and consumption practices. Among the
different sectors of the economy, the building and
construction sector alone accounts for 38% of global
energy-related Carbon dioxide (CO2) emissions (UNEP,
2020). In 2019, emissions from this sector stood at an all-
time high of 10 gigatonnes of carbon dioxide (GtCO2),
attributed to direct and indirect emissions caused by
increased demands in the operational phase of buildings
and adverse weather changes, which resulted in an
increase of 8.47% in final energy use between 2010 and
2019 (IEA, 2020). The construction industry in the UK
alone contributed more than 13 million (m3) tons of CO2
in 2019, of which buildings and building-related works
contributed 17% (Tiseo, 2021). However, due to the
COVID-19 pandemic, global emissions were reduced by
7% (UNEP, 2020), and will be continuously reduced by
6% each year if we are to halve the target of the present
direct emission by 2030. Another source of concern is the
direct emission from fuels used in domestic buildings for
heating during the winter months, as this accounted for
10% of the UK’s carbon footprint back in 2016 (UKGBC,
2021). Several studies have investigated and made
recommendations on how the built environment can
reduce its embodied and operational emissions rate. They
also mentioned several adaptive measures aided by
technology, from system automation to passive controls
deployed over the last decade. One such adaptive and
dynamic technology is the Internet of things (IoT), which
has shown huge savings (Amaxilatis et al., 2017;
Machorro-Cano et al., 2020; Alsalemi et al., 2022).
The use of IoT technology in buildings has proved
effective, and energy savings are well documented.
Different studies exist that measure and quantify the
emissions levels from using these IoT components for
building operations and maintenance and suggest how to
reduce them. For example, Zhao et al. (2022) proposed an
innovative IoT framework to promote low-energy
building in densely populated urban areas, which was
approved to achieve low energy costs. Tanasiev et al.
(2021) explored detailed IoT solutions to control the
HVAC system and monitor the environmental
performance in a real building, which led to significant
CO2 emission reductions. However, little research has
been done about the potential environmental ramification
of IoT devices in buildings to justify if the energy savings
can equate to the carbon emission levels.
Therefore, the complete picture of excluding the impact is
missing. Because of this gap, the energy savings in terms
of green impact have not been estimated closer to reality
in smart buildings, prompting an urgent need for a more
holistic approach guided by data. The impact of building
emissions embedded with IoT systems should be clearly
known to provide stakeholders with a better image for
decision-making, particularly as it affects the
environment. However, that is missing in the current body
of knowledge. This study seeks to address this gap by
investigating how the emission levels from embedded IoT
devices operating in smart buildings do balance the
energy reduction of the smart building. The investigations
are performed using the Integrated Environmental
Solution Virtual Environment (IES VE) simulation tool.
Literature review
IoT developments in general
IoT technologies have been widely considered to have the
potential to connect objects to the internet, enabling
objects to see, observe, detect, record, learn, make
decisions, and take actions based on data with little or no
human interference (Alsalemi et al., 2022). Therefore,
they improve business workflows by minimizing errors,
and reducing operational costs and waste, while
increasing speed and system efficiency. However, these
improvements naturally come with increased demand,
growth, production, and network connections where IoT
components numbers spike over the roof through the

years. These increasing numbers are likely to raise great
concerns and challenges from the economic, social, and
environmental aspects (Quisbert-Trujillo et al., 2020).
The smart items connected to the internet grew from 6.4
billion in 2016 to 25 billion in 2021, while some
researchers predicted these numbers will run into billions
ranging from 125 to about 160 billion by 2030 globally
(Nižetić et al., 2020). Another issue is the environmental
stress by way of e-waste as numbers soared. Le Brun &
Raskin (2020) discussed the sourcing and the utilization
of rare earth metals required for their production as vital,
especially as it consumes six times the energy used in
plastic or metal processing. Quisbert-Trujillo et al. (2020)
sounded more alarmed about the energy consumed in
manufacturing, the ecotoxicity at the end of life, and the
required components to support accessories that make IoT
systems work, e.g., battery replacement on some of these
battery-powered devices, and the resources for
compressing data to prevent traffic of communication of
radiofrequency. Nižetić et al. (2020) questioned how the
monitoring of the network would occur with no current
standard framework guiding policy on-demand
maintenance, security, and privacy. They also alluded to
an increase in fossil fuel use in production and the low
rate of recycling e-waste, currently at about 20% or less.
Furthermore, the lead content of these e-wastes is
dangerous to life. Finally, the urgent legislature on e-
waste, the need for harmonization of the recycling
process, and improvement in the percentage amount being
recycled each year as annual generation stood at 44 billion
metric tonnes. The wastes generated from the IoT
deployment should be compensated by their benefits in
terms of energy savings and carbon emissions.
IoT in Building Energy Savings
The global energy savings conundrum necessitates
retrofit measures in buildings to reduce energy intensity.
Papadopoulos et al. (2019) divide this into the technical
and operational retrofit. Walls, roof insulation, and high-
performance windows are some examples of technical
retrofits, and may not be economically feasible to replace
in the existing buildings due to the high upfront cost
associated with them. Also, it involves the physical
alteration in the design, something particularly
challenging for the old and historic buildings as is in the
UK. Another challenge is the operational or human-based
retrofit, which refers to actions that occupants and
building managers can take to improve energy
performance by adjusting the HVAC system, reducing
light and equipment usage, and opening windows for
natural ventilation during the summer months. Dong &
Andrews (2009) combined the distribution of IoT sensors
and energy plus simulation tools to achieve 30% energy
savings while maintaining indoor comfort levels for a
room. For it to be efficient, it requires a large number of
network sensors to accurately detect occupant activities,
making it costly and unpractical, especially in large open
offices fitted with more people at once. Papadopoulos et
al. (2019) demonstrated it is possible to have as high as
60% in energy savings for a large office when the HVAC
setpoint ranges between 17.50 °C for heating and 27°C
for cooling during occupied hours without compromising
occupant comfort.
Since there is a lack of empirical studies, whereas related
evidence is needed for evaluating the benefits of IoT
deployment in the HVAC and lighting system, the scope
of this study will be concentrating on some of these
human-based retrofits – HVAC and light systems
embedded with IoT sensors.
Methodology
Simulation Design
Simulation tools are appropriate to calibrate and compare
the performance of different building design options at the
early design stage. This will minimize information losses,
reduce cost, and increase speed, time, and analysis
flexibility. It also provides important data for
stakeholders, particularly regarding the early concept
design stage of buildings.
Different building simulation studies have been
conducted in the past using a host of applications – Design
builder, Energy Plus, IES VE, E-Quest, and Green
building Studio, to name a few. The question does
sometimes arise as to which is best to use during design.
Each comes with something unique to offer, and the
decision on which simulator to use lies with the designer
and depends on certain parameter(s) of interest and the
nature of the analysis required.
The IES VE tool was used for this study because it offers
a wide range of modeling applications within its virtual
environment – SunCast and SunPath for Solar Analysis,
Apache HVAC for thermal comfort, and Radiance and
FlucsDL for lighting and Sensor settings, among others.
Life Cycle Analysis (LCA) was used to estimate the
environmental impacts (the carbon emissions in terms of
kgCO2) throughout the building life cycle. IES VE
version 2021 comes with the added feature of the One-
Click life cycle assessment (LCA) tools within its VE Gia
virtual environment, allowing users to run environmental
impact and energy performance simultaneously, thereby
saving time and reducing error. Therefore, IES VE with
One-Click LCA add-in was used in this study to conduct
the LCA process.
IES VE also uses other plugins to design and analyze
building models. In this work, we created the building
geometry using the ModelIT, which is the central 3D core
application for geometry data input shared by all modules.
The model also includes relevant weather files and
specified zone thermal conditions. ApacheSys and
ApacheHVAC were used to define and integrate the
HVAC system to the model and apply setpoint
temperatures and system flow rates. Daylight and
artificial light analysis and sensor activation were
performed with SunCast and RadianceIES, which would
serve as the input during the following dynamic
simulation run. Then the dynamic simulation engine,
Apache interphase, was used to run all the energy
analyses, and the results were presented on Vista for
interpretation. Both models were directly transferred to

the VE Gia with links to One-Click LCA for the
environmental analysis (IESVE, 2021).
IES VE also uses various inbuilt applications to design
and analyze building models. For example, ModelIT is the
central 3D core application for geometry data shared by
all modules. ApacheSys and ApacheHVAC are used to
define what HVAC system is used and apply setpoint
temperature and flow rates within the system.
RadianceIES is used for daylight and artificial light
simulation, and VE Gia links to One-Click LCA. VistaPro
provides quick access to results with the flexibility to
compare design variables on the same page for easy
analyses from one or more simulations (IESVE, 2021).
This study modeled a 5-story commercial office building
with an underground tunnel and a bridge that connects it
to the adjacent section, housing the server rooms and more
offices, situated at Newcastle, North-eastern part of
England. It is located at longitude 1.690W and latitude
55.040N, and 81m above sea level. The design was
created using the 2013-2016 amended version of the
Building Regulations 2010 Part L1A document, focusing
on the cold-humid climate of the location in mind. The
building design details are presented in Figure 1, Figure 2
and Table 1.
The building comprises offices, meeting rooms, a
restaurant, a café, and ICT system rooms. We will be
simulating two basic models – a conventional and a smart
building model; however, the smart building will be
further broken down into smart HVAC change (i.e.,
HVAC with independent sensors control system) and
smart light sensing (i.e., sensor-controlled artificial light
bulbs). The annual percentile for the heating load was
(99%), while the monthly cooling load percentile was
(10%), giving an outdoor winter design heating
temperature of (-2.700C) and maximum cooling load of
(19.900C db. and 15.800C wb.) because the model was a
thermally heavy classed model.
Figure 1: Back view of the investigated building model
Figure 2: Front view of the investigated building model
Table 1: General building parameters
Building parameter
Details
Building area
4303.14 m2
External wall thermal transmittance
0.2599 W/m2K
Window-wall-ratio
36%
External window thermal transmittance
1.6 W/m2K
Infiltration rate
0.25 ac/h
Orientation
210 degrees
Building height
14.8 m
Service life
30 years
Roof thermal transmittance
0.18 W/m2K
In line with the literature review above, we carried out two
distinct changes and analyzed their effects with the 2020
(present data) and 2050 (future data) weather files
obtained from the University of Northumbria database for
Newcastle. Both weather files have been integrated into
the IES VE tool for easy access and application during the
simulation runs. The isolated changes were independently
applied to the conventional model to determine the effect,
from which comparison will be drawn for present and
predicted future weather data. For this simulation, we
have chosen the 16th – 20th July as the typical summer
week and the 22nd – 26th January as the typical winter
week for further analysis. Also, 1st January – 31st
December is the annual duration. Table 2 shows the
design parameters for the investigated systems, which are
simulated in IES VE simulation software.
Table 2: Design parameters for the investigated systems in the
smart building model
Design Parameters
Investigated
Boiler loads
Apache System –
ApacheHVAC with Sensors
Chiller Loads
Apache System –
ApacheHVAC with Sensors
Light Gains
Non-Dimming – Dimming with
Sensors
Total System Energy
HVAC and Light systems.
CO2 Emissions
Embodied and Operational
Emissions.

Scenarios development
For the energy phase of this analysis, we have developed
four scenarios: Conventional Building, Daylight
Harvesting Model, Apache HVAC Model, and the Smart
Building. The energy analysis covers the boiler, chillers,
and light energy from the artificial bulb present in the
model.
• The Conventional Building system controls are
set to ON or OFF only all day, all year and
following occupants’ daily profiles. The heating
and cooling supplied to the building are not by
passive method but an active Apache System
with timed switches. The systems are single or
multi-split, fan coil, and single room cooling
systems. The lighting gains for this case were
non-existing as the artificial lights were set to be
ON continuous throughout the working hours
and go OFF at the end of the day following the
weekly profile adopted without any dimming
effect, keeping the light energy consumed the
same all through the year. A similar approach
applies to the HVAC systems and other forms of
internal gains - people and equipment, as no
difference occurred due to no change in the
profile. As a result, the internal environment of
the model was only affected by the building
envelope and the applied weather data.
• Apart from the HVAC system in buildings,
another area with huge potential for energy
savings is the lighting system on the radiance
application. We applied the open-loop system
with the IoT light sensors placed at the roof level
pointing downwards to control the dimming
effect of the artificial light bulbs within set
boundaries throughout the day depending on the
occupancy movement and natural daylight
illuminance level within the space per time. The
results obtained were compared against the
conventional building without these sensors
activated to analyze their energy savings for
present-day and predicted future weather files.
• Apache HVAC systems with independent
sensors were integrated into the conventional
building model, replacing the initial Apache
system sensing and controlling dry resultant
temperature and flow rates into spaces. The
Apache HVAC model created consists of the
same three different systems as the Apache
system used for the Conventional building, with
independent sensors replacing the timed
switches. Sensors monitor and control set
temperatures and relative humidity within the
zones to which they have been assigned.
• A smart building is a model integrated with the
Apache HVAC system and artificial light
sensors working in sync to improve energy
savings by sensing and controlling the internal
environment within a space. These systems have
been set up with their respective profiles and
boundaries to keep the internal condition of a
room within defined ranges and not compromise
on occupant thermal comfort and vision.
After calculating the energy savings, both models were
transferred to the One-click-LCA via the VE Gia tool. The
LCA analysis was run for the building using 30 years
(2020-2050) life horizon from cradle to grave. The benefit
of using the One-click-LCA is that it recognizes the model
materials, makes all the necessary adjustments and
assumptions, and pulls the data from its global resources
to calculate the carbon emission levels for each model
with the result presented in a .csv file.
Findings
Total system energy result - Conventional Building
The total energy use intensity (EUI), i.e., the ratio of total
energy to building floor area, increased by 3% from 131.2
KWh/m2/yr to about 135.2 kWh/m2/yr for the 2020 and
2050 weather data, respectively. For both weather files,
July was the dominant month with figures surpassing 60
MWh, while February had the minimum consumption all
year, as shown in Figure 3.
For most periods of the year, the total system’s energy
consumed for the future weather data (2050 data) was
more, except for January, February, and December, for
which the present weather data (2020) was dominant.
Furthermore, the total system energy during the summer
months (May – Sept.) was, on average, 26% more than the
winter months (Nov. – Mar.) by 2020 and 30% more by
2050.
Figure 3: Monthly total systems energy consumption of
Conventional Building (MWh) (For interpretation of the
references to color in all colored figure captions, the reader is
referred to the web version of this paper.)
Total system energy result - Daylight Harvesting
Model
The impact of daylight harvesting on the total system
energy compared to the conventional system is presented
in Figure 4. In this case, the energy savings were minimal
at about 3% per annum. Again, energy savings for the
0
10
20
30
40
50
60
70
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Energy (MWh)
Conv Building Present data.aps
Conv Building Future data.aps

summer months exceeded that of winter, but the winter
period provided some remarkable results with 78% and
73% in ratio to the summer figures even with the
shortened daylight hours during this period. July showed
the biggest savings of 3.3 MWh and 3.2 MWh when
results were compared with the conventional model for
both weather files.
Figure 4: Monthly Total System Energy Daylight Harvesting
Model vs. Conventional Model (MWh)
Total system energy result - ApacheHVAC Model
The total annual system energy of smart building,
compared to the conventional building shown in Figure 5,
decreased by over 55% in both weather files. Similar to
our Conventional Building model, July was the month
with maximum energy consumption at over 28 MWh and
30 MWh respectively for future and present data. The
average annual system consumption was more than
double between both models, with the conventional
model standing at above 47% for both weather files.
Finally, when considering the HVAC model alone, the
mean system energy demand during summer was 1.68
MWh more than the present data, and 0.52 MWh less in
the winter months. This indicates dominant future
summers where temperatures are expected to rise due to
anthropogenic activities.
Figure 5: Monthly Total Energy Conventional Building vs.
ApacheHVAC Model (MWh)
Total system energy result – Smart Building
Annual system energy saving improved by 56.8% and
57.2% for both weather files, respectively, amounting to
over 300 MWh in savings in favor of the smart building,
as shown in Figure 6. Monthly, the smart building’s
energy savings also improved by at least 58% compared
to its corresponding pair in the conventional building. The
maximum monthly energy consumed stood at 28.3 MWh
and 26.5 MWh in July for both weather files. Of the yearly
total, summer alone accounted for more than half of the
smart building’s energy for the future weather data, while
it stood at 47.9% for the present weather data.
Figure 6: Monthly Total System Energy, Smart vs.
Conventional Building (MWh)
0
10
20
30
40
50
60
70
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Energy (MWh)
DayLight Model Dimming On Future data.aps
DayLight Model Dimming On Present data.aps
Conv Building Present data.aps
Conv Building Future data.aps
0
10
20
30
40
50
60
70
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Energy (MWh)
HVAC Model Present data.aps
HVAC Model Future data.aps
Conv Building Present data.aps
Conv Building Future data.aps
0
10
20
30
40
50
60
70
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Energy (MWh)
Smart Building Future data.aps
Smart Building Present data.aps
Conv Building Present data.aps
Conv Building Future data.aps

Carbon emission – Smart vs. Conventional Building
The CO2 emissions of the smart and conventional
buildings were assessed with One-click-LCA from cradle
to grave. The external impacts include benefits and loads
beyond system boundaries. These are energy recovered by
recycling materials and reusable products; hence their
values were subtracted from the total CO2 emissions. The
total CO2 emission for the smart building by the end of 30
years lifetime (1.83 t CO2 e / m2) is 37.1% lower than that
of the conventional building (2.91 t CO2 e / m2), which is
similar to the results provided by Su et al. (2020).
The embodied carbon emissions are 33.12 kg CO2/m2/yr
for the smart building and 30.91 kg CO2/m2/yr for the
conventional building. Based on CO2 classification, the
embodied carbon emissions of both models are presented
in Table.3. Electricity consumption contributes to the
highest portion of the total embodied emissions, which are
93.3% and 95.3% for the smart and conventional
buildings, respectively. The internal walls and non-
bearing structures contribute to the least emission (less
than 0.1%) for both designs. The electricity consumed in
this stage is sometimes referred to as the embodied
energy, i.e., the electricity consumed in material
extraction, manufacture, transportation, and all other
processes in the supply chain during the construction and
end of life phases of both buildings (Su et al., 2020).
Table 3 Embodied Carbon emission between smart and
conventional buildings.
Category
CO2e
emissions -
ton (Smart
Building)
CO2e emissions
- ton
(Conventional
Building)
Electricity use
7377.8
11949.4
Floor slabs,
ceilings, roofing
decks, beams,
and roof
187.7
187.7
External walls
and facade
159.2
178.6
Fuel use
117.9
159.2
Windows and
doors
58.6
58.6
Internal walls
and non-bearing
structures
8.1
8.1
The operational emissions for both designs were
measured by the carbon emissions generated from the fuel
consumption, including heating, cooling, lighting and
appliances. As shown in Figure 7, the life cycle
operational emission of the smart building is 28.15
kgCO2/m2/yr, which is over 65% less than the
conventional buildings. The peak emission for the
conventional building during the summer months reaches
41.9 kgCO2/h, which is over two times higher than the
emissions from the smart building.
Figure 7. Annual Operational Emission – Smart vs.
Conventional building.
Discussion
Energy savings and carbon emissions from buildings are
two key aspects that recent research has focused on
addressing issues of global warming associated with the
built environment. The impact of IoT technologies, e.g.,
sensor-based systems, on energy savings is well
documented and presented in the literature review section.
We have demonstrated the potential of these systems in
improving the energy savings in smart buildings by
enhanced automation in the prediction, monitoring, and
sensing of occupant behaviors and other factors, and
activating systems controls to maintain thermal comfort
of occupants within set limits. The EUI for the smart
model was 135.4kwh/m2, which is about 38% lower than
a conventional building at 219.3kwh/m2. This result
follows the trend observed by Ali Al-janabia et al. (2019)
for buildings of similar size. Annual heating system
energy savings for the smart design stood at over 35% (8
MWh) and over 85% (300 MWh) for cooling system
demands which agrees with the works of Papadopoulos et
al. (2019), who carried out similar studies on building
energy at different locations.
Energy savings from cooling demand more than doubled
those from heating and can be attributed to several
reasons. First, the effect of weather on temperature rise
and shortened winter months in the future makes cooling
demand a dominant parameter for the smart designs and
should be factored into future design decisions. Climatic
conditions also control cloud cover within the external
environment. Secondly, both models were not created to
use the passive cooling method (i.e., Natural ventilation)
but the active method (Apache HVAC). Internal gains
from people, equipment, and lights were very active,
contributing to the increased demand for cooling in the
summer months and resulting in the dry temperatures
exceeding setpoints at peak hours of the day. However, a
0
5000
10000
15000
20000
25000
30000
35000
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
System Carbon Emission (kgCO2)
Time (year)
Smart Building Present data.aps
Conv Building Present data.aps

reverse effect occurred in the winter months as internal
gains played a role in reducing the boiler energy
consumption. Building orientation indirectly affects the
cooling demands, as our designs were south-facing, with
about 5% more glazing in the north than the south zones.
The excessive heat gains from the solar position further
add to the internal gains, which, in turn, can also
contribute to the cooling energy demand of both models
and even may result in overheating. The effect of local
shading was not considered as this is not within the scope
of this work, but we believe it will be of significant effect
in swinging the energy demands depending on the
location and height of the shading object. The same goes
for building insulation. The results of dimming control by
the sensors from both designs kept the minimum
illuminance level within the spaces above 500lux.
IoT devices’ environmental impact was also investigated
to better understand certain issues raised earlier –
particularly the carbon emission from smart buildings.
The embodied emissions from the models showed a slight
increase in the smart building (2.21 kgCO2/m2/yr),
although the bulk materials were the same in both models;
however, we believe this result is justifiable in reality.
Given that IoT devices are incorporated into the building
envelope, it increases the amount of embodied carbon
emissions of the building. Considering the mining and
extraction, transportation, and all the other processes on
the supply chain plus e-waste generated at the end-of-life
of these components are accounted for as discussed in the
literature study. At 33.12 kgCO2/m2/yr, the embodied
carbon emissions in the smart building are 7% more than
that recorded for conventional buildings. At the same
time, the operational carbon emission from the smart
building is 57% less than the value of the conventional
building. The ratio of embodied to operational carbon
emission from both designs stood at 32:68 for
conventional buildings and 46:54 for the smart building,
which falls within the acceptable range for UK buildings.
In the studied models, energy was the major determinant
in both embodied and operational carbon emissions. The
CO2 emission level of the smart building was less
compared to the conventional model, probably because of
the sensor controls present to regulate the operations of
the boiler, chillers, and consumption of the artificial lights
while keeping the internal environment comfortable. To
reduce the emissions of not just CO2 but all greenhouse
gases (GHG) within the built environment, the UK must
continue to push for green energy generation and
utilization. One of the goals of the 2021 United Nations
climate change conference (COP26) that took place in
Glasgow is to “Secure global net-zero by mid-century and
keep global temperature rise of 1.5 degrees within reach”
(COP26, 2021). This will only become a reality through
sincerity of purpose and people-focused business models
integrated into energy generation, transmission, and
utilization. For example, sources of fuel used in
generating these energies are a major concern, especially
in the Newcastle area, where heating during winter is
predominantly by gas boilers. Building professionals,
homeowners, and tenants should embrace the support
from the UK government that advocates replacing boilers
with heat pumps and thermal photovoltaics in residential
and commercial buildings for space heating and daily hot
water supply to help reduce emission rates at home,
offices, and even at the grid.
Conclusions and future work
Several technologies and digital devices have been
deployed in both real and simulated environments to try
and achieve the net-zero target of the UK government by
2050. Systems of interconnected digital devices, objects,
machines, and people (e.g., users, facility managers) with
the ability to transfer and share data over a network,
broadly defined as the Internet of Things (IoT), are used
in buildings to reduce energy consumption and improve
the thermal comfort of an occupant by real-time
monitoring and employing different management
strategies. To better understand the environmental impact
of these IoT technologies – particularly the use of sensors
in buildings, we set out to investigate if the energy savings
balances the carbon emissions on IES VE for a
commercial office building in Newcastle, UK.
The EUI of the smart building was reduced by over 38%
compared to the conventional building with the
deployment of smart HVAC and light sensors used in
controlling dry bulb temperatures and daylight within a
space according to defined occupant profiles. The annual
cooling demands were also reduced by over 300MWh
representing 85% savings during the summer months as
the independent sensors on the smart HVAC sensed and
controlled both flow rates and dry bulb temperatures to
keep occupants cool and comfy all summer within limits.
A similar result was recorded during the winter months
with 35% energy savings.
The artificial light sensors in the smart building showed
great potential savings by controlling dimming proportion
with increased or decrease external illuminance. At
maximum, it reached 72% (1.3KW) energy-saving daily
during the early spring and kept the level of internal
illuminance above 500 lux, which is the minimum
requirement for commercial offices.
The positive ripple effect of these IoT sensor actions is
visible in the operational emissions for the smart building
as boilers, chillers and light bulbs are being regulated,
indirectly controlling the level of fuel and electricity
consumed in the process in line with the occupant profile.
Operational emissions from the smart building are 57.8%
less than that of the conventional building at 66.24
kgCO2/m2/yr.
However, a negative result was recorded for the embodied
carbon emission from the smart building at 33.12
kgCO2/m2/yr, representing 7% more than the embodied
emission level for conventional buildings. It was
attributed to the increased material, energy used in
manufacturing, end-of-life e-waste and transportation
from the addition of sensor components and other devices
to help them work efficiently in the smart building.
Therefore, the result suggests the energy savings from
embedded IoT sensors do, in fact, balance and outperform

their emission levels, but caution must be in place. While
embedded IoT sensors reduce energy utilization and
improve performance and comfort inside smart buildings,
their embodied emission contribution margin should be
considered. If advances in technology increase the level
of smartness of buildings, then the embodied emissions
will become a concern for the future. Simulation results
sometimes do not transcend to reality. Future research
will include real-time monitoring, measurement, and
analyses of the energy savings in smart buildings against
the measured emission level from the building with IoT
sensors. Also, it will include the shading control system
and use of motion detection sensors, as these were part of
the limitations of this study.
References
Al-janabi, A., Kavgic, M., Mohammadzadeh, A., &
Azzouz, A. (2019). Comparison of EnergyPlus and IES
to model a complex university building using three
scenarios: Free-floating, ideal air load system, and
detailed. Journal of Building Engineering, 22, 262–280.
Alsalemi, A., Himeur, Y., Bensaali, F., & Amira, A.
(2022). An innovative edge-based Internet of Energy
solution for promoting energy saving in buildings.
Sustainable Cities and Society, 78, 103571.
Amaxilatis, D., Akrivopoulos, O., Mylonas, G., &
Chatzigiannakis, I. (2017). An IoT-Based Solution for
Monitoring a Fleet of Educational Buildings Focusing
on Energy Efficiency. Sensors, 17(10), 2296.
https://doi.org/10.3390/s17102296
COP26, United Nations Climate Change Conference UK
2021, In Partnership with Italy. (2021) Retrieved from
https://ukcop26.org/cop26-goals/
Dong, B., & Andrews, B. (2009). Sensor-based
occupancy behavioral pattern recognition for energy
and comfort management in intelligent buildings,
Eleventh International IBPSA Conference Glasgow,
Scotland July 27-30, 2009.
IEA (International Energy Agency) (2020), Is cooling the
future of heating?, IEA, Paris. Retrieved from
https://www.iea.org/commentaries/is-cooling-the-
future-of-heating
IESVE (Integrated Environmental Solution Virtual
Environment). (2021). Retrieved from https://distance-
learning.iesve.com/courses/enrolled/431397
Le Brun, G., & Raskin, J.-P. (2020). Material and
manufacturing process selection for electronics eco-
design: Case study on paper-based water quality
sensors. Procedia CIRP, 90, 344–349.
https://doi.org/10.1016/j.procir.2020.02.041
Machorro-Cano, I., Alor-Hernández, G., Paredes-
Valverde, M. A., Rodríguez-Mazahua, L., Sánchez-
Cervantes, J. L., & Olmedo-Aguirre, J. O. (2020).
HEMS-IoT: A Big Data and Machine Learning-Based
Smart Home System for Energy Saving. Energies,
13(5), 1097. https://doi.org/10.3390/en13051097
Nižetić, S., Šolić, P., López-de-Ipiña González-de-
Artaza, D., & Patrono, L. (2020). Internet of Things
(IoT): Opportunities, issues and challenges towards a
smart and sustainable future. Journal of Cleaner
Production, 274, 122877.
https://doi.org/10.1016/j.jclepro.2020.122877
Papadopoulos, S., Kontokosta, C. E., Vlachokostas, A., &
Azar, E. (2019). Rethinking HVAC temperature
setpoints in commercial buildings: The potential for
zero-cost energy savings and comfort improvement in
different climates. Building and Environment, 155,
350–359.
https://doi.org/10.1016/j.buildenv.2019.03.062
Quisbert-Trujillo, E., Ernst, T., Samuel, K. E., Cor, E., &
Monnier, E. (2020). Lifecycle modeling for the eco
design of the Internet of Things. Procedia CIRP, 90,
97–101. https://doi.org/10.1016/j.procir.2020.02.120
Su, X., Tian, S., Shao, X., & Zhao, X. (2020). Embodied
and operational energy and carbon emissions of passive
building in HSCW zone in China: A case study. Energy
and Buildings, 222.
doi:https://10.1016/j.enbuild.2020.110090
Tanasiev, V., Pătru, GC., Rosner, D., Sava, G., Necula,
H., and Badea, A. (2021) Enhancing environmental and
energy monitoring of residential buildings through IoT.
Automation in Construction. 126: 103662.
Tiseo, I. (2021). CO2 emissions from the construction
industry in the UK 1990-2019. Retrieved from
https://www.statista.com/statistics/486106/co2-
emission-from-the-construction-industry-uk/
UKGBC (United Kingdom Green Building Council).
(2021). UKGBC’s vision for a sustainable built
environment is one that mitigates and adapts to climate
change. Retrieved from
https://www.ukgbc.org/climate-change-2/
UNEP (United Nations Environment Programme).
(2020). Building sector emissions hit record high, but
low-carbon pandemic recovery can help transform
sector – UN report. Retrieved from
https://www.unep.org/news-and-stories/press-
release/building-sector-emissions-hit-record-high-low-
carbon-pandemic
Zhao, W., Chen, J., Hai, T., Mohammed, M.N., Yaseen,
Z.M., Yang, X., Zain, J.M., Zhang, R., and Xu, Q.
(2022). Design of low-energy buildings in densely
populated urban areas based on IoT. Energy Reports. 8
4822–4833.
https://doi.org/https://doi.org/10.1016/j.egyr.2022.03.1
39.