key: cord-0753990-z4pm22wr
authors: Anand, Prashant; Cheong, David; Sekhar, Chandra
title: A review of occupancy-based building energy and IEQ controls and its future post-COVID
date: 2021-09-10
journal: Sci Total Environ
DOI: 10.1016/j.scitotenv.2021.150249
sha: ffe758f786fc626f0c89cdb16ff616dfb978171b
doc_id: 753990
cord_uid: z4pm22wr

Occupancy schedules and density can have a substantial influence on building plug, lighting, and air conditioning energy usage. In recent years, the study related to occupancy and its impact on building energy consumption has gained momentum and is also promoted by ASHRAE as it has created a multi-disciplinary group to encourage a comprehensive study of occupant behaviour in buildings. Past studies suggest that building systems do not consume the same energy and provide similar Indoor Environmental Quality (IEQ) to their designed specifications due to inaccurate assumptions of occupants and their behaviour. Supplying ASHRAE specified minimum required ventilation based on accurate occupancy may lead to significant air-conditioning energy savings. However, the same strategy is not suitable in the current time since minimum required ventilation may not be sufficient to mitigate the SARS-CoV-2 virus spread in confined spaces. High-temperature cooling augmented with elevated air movement across an acceptable range of velocity can maintain the health and comfort of occupants by providing higher ventilation and without an energy penalty. The analysis of the literature highlights strengths, weaknesses, and key observations about the existing occupancy monitoring and occupancy-based building system control methods to help in the direction of future occupancy-based research.

International Energy Agency (IEA) has identified energy efficiency of buildings as one of the main criteria towards limiting CO 2 emissions in the long term [1] . Governments around the world are working actively with IEA on various initiatives and regulations to promote energy efficiency in the building sector. The prime objective of this energy-efficient built environment goal is to save energy as well as to create a healthy and comfortable environment for occupants. Therefore, the aim is to be able to relate the operation of various energy use systems in a building to its occupants' usage patterns, so that a demand-based operational strategy could be conceived. However, this cannot be successfully achieved if accurate information on the interaction between occupants and energy use systems is incomplete or absent during the operation of buildings. Further, the Indoor Environmental Quality (IEQ) of modern buildings is of utmost importance today, as humans spend 87% of their time within an indoor built environment in general and the same has reached as high as 100% as a result of COVID-. 19 identification of occupancy being a primary factor influencing building energy usage, both actively and passively. Then the subsequent section discusses the reported variations in building energy usage due to occupancy. In further sections of the literature review, the current developments in occupancy-based building control to enhance the energy efficiency of plugload, light-load and HVAC-load usage are discussed. Additionally, various occupancy monitoring methods are also reviewed. Further, the potential role of occupancy-based HVAC control in viral load mitigation has been reviewed with respect to current COVID-19 pandemic and any such events in future. Finally, the summary and key observations to form a hypothesis for this study are discussed. A description of multiple sections is presented in Figure 1 . 

The variation in building energy usage due to occupants can be divided into two main categories: 'variations due to occupant active interaction with building systems' and 'variations due to occupant passive interaction with building systems' (see Figure 3 ) [24] . While active interaction is defined by the actions taken by occupants to control the building systems, passive interaction is defined by the factors associated with occupant's presence without any physical action taken by them to control building system.

Journal Pre-proof

The active interactions by occupants are the control of lighting systems (Light loads) to enhance their visual comfort and use of plug loads for electrical equipment. In institutional buildings, occupants may use numerous lighting and electrical equipment, increasing the heat gains of the space and thus subsequently causing an increase in cooling load. Thus, it can be concluded that the active interaction of occupants has a significant impact on building energy usage directly from the use of systems (plug-loads and lighting) and indirectly from the heat load generated by in-use systems. Similarly, the occupant's passive interaction with building systems (e.g. HVAC) influences energy consumption in buildings to a great extent, affecting the building thermal comfort and indoor air quality (IAQ). In this case, although the interactions between occupant and HVAC system is passive, occupants actively influence the energy usage by acting as a movable heat and CO 2 source [25] . In the following subsection, the outcome of several J o u r n a l P r e -p r o o f Journal Pre-proof studies conducted in the past to quantify the occupants active and passive influence over building energy usage, is summarised.

Active interaction of occupants with building energy systems is a complex mechanism, affected by numerous parameters such as number of occupants, behaviour of occupants, type of space, type of work, day and time etc. [26] . Overall, these parameters can be grouped into three resolutions: a) Temporal, b) Spatial and c) Occupancy (See Figure 4 ). Temporal resolution states the scale of time, spatial resolution states the information of space and occupancy resolution states the behaviour of occupants. In general, the sociological or psychological aspects significantly influence occupant behaviour [27] [28] [29] .

To measure the active influence of occupants over building energy usage, accurate information of each resolution is essential. However, due to the high uncertainty of occupant behaviour, it is difficult to capture each energy-usage action and associated contributing factor.

This uncertain occupant behaviour affects the indoor environment and thus is a major reason for large variations in energy consumption. Conversely, indoor conditions (temperature, humidity, lux level, etc.) also affect occupant behaviour which in turn changes the total building energy consumption.

J o u r n a l P r e -p r o o f Journal Pre-proof Figure 4 Resolutions occupancy interaction with energy systems [26] The simpler forms of energy-related occupant behaviour include thermostat regulation, J o u r n a l P r e -p r o o f Journal Pre-proof Studies discussed above acknowledged a significant variation in energy usage by similar building spaces due to varying occupant's behaviour. However, the amount of variation also depends on buildings types [33] [34] [35] [36] [37] . Therefore, an accurate knowledge of occupant behaviour and its variation among different building types is crucial to determine operational strategies for building energy use systems. These strategies should accommodate both occupant behaviour patterns as well as changes in behaviour. To accomplish occupant comfort and well-being requirements, active interaction between occupants and building systems is more significant than the efficiency of systems alone [38] . [46] . Taking the number of wi-fi users as a typical value for the number of occupants, the results showed a variation between 63 and 69% of the total electricity consumption due to variation in occupant density over a period of time for institutional buildings [46] . Mahdavi et al. (2009) investigated the effect of the occupancy ratio using statistical methods for the lighting consumption of office buildings. Based on occupancy ratio, the variation in correlation ranged from 70 to 90%. The energy usage for lighting proves a good correlation benchmark with the occupant counts for most of the building types [47] . Hotel buildings tend to show a good correlation with the consumption of steam for heating during the winter season [48] . However, the correlation between HVAC energy usage and occupancy show a varying result with buildings types. In the case of campus buildings, a weak correlation has been found between occupancy rate and energy usage [46] . This fragile correlation is possibly due to the 'fixed' HVAC operation scheduled by facility managers for campus buildings. Thus, there exists a research opportunity to develop a framework using occupancy load as an input to better optimise HVAC operations. This will lead to energy savings as well as enable a better zone level thermal comfort for the occupants of campus buildings.

Some studies on prediction of cooling load based on occupancy and floor area information are currently available. For example, a study by Kwok (2011) demonstrates an increase in the precision of the cooling load prediction model when real-time occupancy and real-time weather data were used as simulation inputs compared to a fixed occupancy schedule alongside weather input [49] . However, this study used CO 2 concentration for measuring realtime occupancy, which is less accurate for estimating occupancy in large spaces because such spaces end up removing air via return ducts before it is measured by the CO 2 sensors.

To study the dynamic schedule and behaviour of occupants, this section showcases current studies which comprise data collection methods such as occupant observation, occupant

surveys and experiments. The monitoring methods of occupancy can be broadly divided into two key categories: (a) Observational studies; and (b) Experimental studies.

In this type of study, occupants' activities, attendance, and indoor environmental parameters are inactively supervised using sensors. Based on seasonal variation (e.g., summer

and winter) and building component usage type (e.g., windows and blinds), the monitoring period and the selection and placement of occupancy sensors can be determined [50] and Occupant recognition methods in previous studies comprise wearable detectors (e.g., fitbit), motion sensors (e.g., passive infrared and ultrasonic), CO 2 sensors, video cameras, Information Technology Networks (e.g., Cellular data, Wi-Fi etc.) and security-based systems (e.g., GPS) [12, 24, 25, 53, 54] . Among these, the use of motion sensors is the leading technology used for occupancy detection [54] , but is not relevant for the detection of motionless or near motionless occupants, e.g. at an office workstation). However, it can reliably measure the state of occupancy transition such as arrival, departure and long transitional vacancy periods [38] .

Some researchers found that the integration of motion detectors and CO 2 sensors can improve the accuracy of occupancy detection [55] [56] [57] . However, a significant delay is found between occupant presence and increase in CO 2 . So, this process is problematic in zones with several sources of supply and several sinks for CO 2 e.g., an open-plan office. [54] .

In the past, researchers have used cameras to count occupants and to identify every individual.

Using computer vision techniques, this counting and identification can be done with high accuracy [58] . However, the same has been identified as a computationally rigorous method which involves substantial cost and confidentiality implications [54] .

There are numerous researchers who measured the potential of using current technology advancements for occupancy counting and detection, i.e. global positioning system (GPS), cellular data, wireless local area network (WLAN) and Bluetooth [59] [60] [61] [62] [63] [64] . These advancements building management with sophisticated algorithms would be enough for monitoring occupants via wi-fi. However, the uniform distribution of sensors is critical for garnering accurate information, especially for buildings having a large scale (like Institutional buildings).

In the past, researchers have observed occupants to undergo various energy usage behaviours which include switching on/off lights, door/window/blinds operation, thermostatadjustment, and other adaptive measures like use of fan or clothing adjustments. However, a detailed monitoring process was recognized to be costly, time-consuming and likely to have low accuracy [66] . Labour-intensive monitoring approaches (e.g., manual photographing and recording) caused a restriction with regards to sample size and duration. Nevertheless, they were valued for recognizing key behavioural patterns and more importantly motivations behind those behaviours. Numerous researchers have tried to automate this chaotic manual data collection method [51, 67] . For example, a researcher collected occupant interaction from motorized blinds electronically and found that these blinds are used more often than manual blinds [68] . Perhaps, the least laborious procedure for observing blind/window usage is by proximity sensing devices.

Similarly, window/blinds operating behaviour can also be supervised via photographic methods and by questionnaire surveys [36, 69, 70] . However, a limitation with this sensor is that it is unable to differentiate a situation between full and partial opening of windows [71] . In a similar kind of study, researchers have found that the opening of doors are less associated with indoor or outdoor environmental conditions (like change in temperature) [69, 72] .

To monitor thermostat adjustments, integrated sensors or set-point logs have been used in the past [73] [74] [75] . Contemporary digital thermostats have logging abilities and the collected data J o u r n a l P r e -p r o o f can be frequently transferred to a database via the internet. Most of these past studies on thermostat control were based on questionnaire surveys [76] . The use of a fan is the foremost technique for cooling in several buildings, predominantly for spaces where air conditioning is not available or treated as a luxury. It has been examined by numerous researchers using opinion polls or mature plug-level energy data [69, 77, 78] . Clothing levels have no direct influence on energy usage, but it disturbs occupant thermal comfort leading to adaptive actions by occupants.

It cannot be sensed electronically, so observations and surveys are to be used [69, 79, 80 ].

Experimental studies play a vital role in measuring occupant comfort conditions (Thermal and IAQ related) which influence its behaviour. There are several examples on the success and reliability of this method, such as Fanger's thermal comfort model and Daylight

Glare Probability model [81] [82] [83] . In experimental studies, subjective responses and environmental parameters are collected simultaneously within a controlled environment such as a living laboratory. This kind of study is very useful in understanding the physiological and psychological influence over occupant behaviour based on different groups of age, culture, gender, etc. [84] [85] [86] [87] . However, it is uncertain whether generalized models of occupant behaviour can be achieved as the behavioural data of occupants could be specific to the data collection period. For example, numerous behaviour patterns can be infrequent and can occur on daily or monthly basis [88] . So, an adequate number of subjects, measurement duration and frequency are highly important for this kind of study [66, 89] . Most developed occupant behaviour models used a 5-min frequency for data collection [90] .

J o u r n a l P r e -p r o o f

Utmost behaviour is comparatively occasional; for example, switching on the light can be a single time action every day and operation of windows/blinds can be seen on a daily basis, though rarely, and dependent on arrival and departure. Therefore, sampling durations of 15 min to 1 h are usually appropriate and any outliers can be ruled out over a yearly simulation [91] .

Regarding sample sizes, current study related to occupant monitoring generally range from twodigit to some thousand digits of occupancy [51] . As there is significant diversity among individual behaviour, an acceptable sample from the common population is essential [88, 92] .

There are some well-known procedures for defining acceptable sample size, such as power analysis [93] . Other than that, replicating a realistic environment with proper regard to social restraints and dynamics is not very likely. Many other limitations to study occupant behaviour in a living laboratory have been discussed in the past [94, 95] . One researcher has summarized the pros and cons of various occupancy counting systems and concluded computer vision based systems incorporating cameras to be the most accurate [12] .

In summary, occupancy monitoring can be categorised as occupant density, equipment density, and adaptive behaviours. The first and second group can be studied through several occupancy detections and counting methods listed in Table 2 and electrical power meters.

However, price, confidentiality and precision continue to be a challenge. Similarly, monitoring of adaptive behaviour needs monitoring of occupant actions and measurement of environmental variables simultaneously. To get unbiased data, preferably, occupants should not be aware that they are being under observation as this may influence their behaviour [96] . There are organizational and legal restrictions (e.g. personal data protection act) within which this kind of study is to be conducted. Comprehensive procedures on all deliberated features of occupant monitoring is going to be a part of IEA EBC Appendix 66 [38] .

Journal Pre-proof 

In recent years, occupancy-based building system control related research has gained momentum in both academia and industry. Such research has been conducted considering actual and predicted occupancy as an input to the building system control logic to control the operation of plug, lighting and HVAC loads. The efficiency of this kind of control logic is normally based on the level of occupancy information such as their a) presence/absence, b) individual level preference and c) action/behaviour as a group or individual [113] .

Real-time response to occupancy for plug-load control is still a complex process. There are some products available for spaces with few occupants such as residential spaces or individual offices to switch on/off their equipment based on occupancy information [114] .

However, the same is not common for spaces which have varying occupancy level due to the complexity of identifying the presence of equipment users out of the total occupancy. To control plug-load, some researchers have tested methods such as local motion sensors and wearable

Bluetooth devices, to detect the presence/absence of computer users, to allow computer to reboot/sleep [24, [115] [116] [117] . It has been concluded that thorough information related to the needs of occupants is essential for trustworthy plug-load control. There may be a case when occupants have left their computer switched on intentionally to finish their computational work in their absence, so forcefully shutting down their system or keeping the systems in sleep mode in the absence of users may be a loss to their work and is undesirable. Overall, the real-time response to occupancy for plug-load control has a significant potential for further improvement as the same is in its nascent stage.

Journal Pre-proof

The control of plug-load by thorough learning of individual preferences over time has also been explored by research. Most of the research related to learning individual preferences are specific to residential buildings, with focus on a single occupant [118] [119] [120] . However, one research had developed a generic method to learn the needs of individual occupants over a period of time in any type of space, irrespective of occupancy size, though this requires further validation studies [121] .

During the early stages of occupant-centric building control research, it has been realized that predicting appliance usage patterns based on past common patterns is very useful in enabling a predictive control of power supply, especially for household appliances which usually have less variation on a daily basis [122, 123] . However, the study related to the control of plug-load based on their specific behaviours is not widely explored in academia, because of difficulties in accurately detecting activity, due to the limited control present in building systems. There are a few research studies which demonstrate activity based control of plug-loads, based on the monitoring of occupant presence and computer usage [24, 124] .

Realtime lighting control based on occupancy is well-established in commercial buildings, especially in comparison to plug-load control. Illuminance of a space is generally controlled by sensing the motion of occupants using the Passive Infrared (PIR) based motion sensors. This system turns on lighting when motion is detected and switches it off after a fixed time delay from last movement detection. Improvements to this system by optimising the time J o u r n a l P r e -p r o o f Journal Pre-proof delay have been discussed in previous research [125] . Additionally, researchers have also discussed the possibility to further optimize the control of lighting in large open-plan spaces by confining lighting only to areas where occupants are nearby to offer greater energy efficiency [126, 127] . However, the optimization of lighting by confining to occupied zones only and the energy savings associated with this method have not been reported so far for institutional building spaces.

Some control applications such as automated control of lighting or HVAC system have the flexibility to take individual's comfort preferences as input. Such options with control application can provide personalized lighting and air-conditioning. Personalisation in lighting levels have been applied in a previous scenario by recording lighting adjustments made by each occupant [128] .

This system of occupant-based control is based on the assumption that a user's activity and specific behaviour dictates their environmental requirements. One research has lighting control implemented through visual detection of occupant location and activity [129] . Similarly, another research study investigated a method to control the electrical equipment and HVAC system/equipment based on the occupant presence and occupant activity (computer use) [124] . A domestic study demonstrates the use of Infrared cameras and presence sensors to define activity categories. [130] . Faster response controls for lighting and control of appliances has prediction horizon transients in the order of minutes [131] or incorporate methods to predict the next action J o u r n a l P r e -p r o o f Journal Pre-proof in sequence. This system is useful for controlling household appliances, lighting etc, demonstrated in the Adaptive House Adaptive House [122] and MavHome [123] projects, where action sequences and environment conditions around the user are detected and recorded, followed by the system attempting to automate repetitive actions.

In the past, many researchers have discussed the possibilities of HVAC system related energy savings by controlling the cooling requirement of spaces with an accurate presence/absence information of occupants [43, [132] [133] [134] [135] [136] . Most of the reported energy savings are obtained by not providing supply air during unoccupied hours. However, these studies are mainly conducted for a small space that have single or low occupancy, to achieve a controlled test environment. There are a few studies for larger or multi zone spaces which also show significant improvement in energy use using real-time occupancy levels in HVAC system control [137, 138] . However, none of the past studies have been conducted for institutional spaces such as classrooms, which have significant variation in occupancy during operational hours and may require a more complex control strategy. Additionally, to overcome the complexity of data collection, most of the earlier studies are conducted through energy simulations. In this study, this research gap has been addressed by selecting a large floor with varying space typologies and occupancy levels for investigation. Additionally, the methodology developed as a control strategy should be considered one of the main contributions.

J o u r n a l P r e -p r o o f

The control of HVAC system based on occupancy is subject to individual thermal comfort preferences. These control preferences can be obtained using thorough thermal comfort survey [139, 140] , or by monitoring the thermostat adjustments over a period of time [141] . In contrast, preference-based control systems are highly specific to space typologies and occupant density.

However, these studies are mainly conducted for small spaces that have single or low occupancy, to achieve a controlled test environment. There are a few studies for larger or multizone spaces which also show significant improvement in energy use using real-time occupancy levels in HVAC system control.

Application of preference-based control systems is highly reliant on occupant density and space/building typology. Earlier studies are mainly applicable for single occupant residential scenarios [141] , with a few based on generic platforms that can incorporate any type of space and occupancy [121, 142] . Preference-based control systems have been demonstrated to create conflict in comfort between individuals in multi-occupancy spaces. Further research is required to find common ground among conflicting comfort standards for multi-occupancy spaces. One potential future approach could be based on the distinction between tracked occupant groups and occupancy detected by ambient sensors [143] .

A difference between actual energy saving and estimated energy saving from the above mentioned control methods is very common. energy savings for private offices and restrooms respectively. As expected, with a decrease in delay time, energy saving increases. It has been concluded that the energy saving of any space depends on the availability of daylight, space usage, and occupant's density over time.

Floyd (1996) has investigated the energy savings from 56 offices and 72 classrooms by installing PIR and ultrasonic based occupancy sensors compared to manual control [145] . They monitored energy consumption of the spaces studied for a year (6 months with manual control and another 6 months with occupancy-based control). Up to 19% of energy saving possibilities were found for office building and 11% for schools. Additionally, there were several incidences of sensor malfunction which caused an increase in energy consumption in certain spaces. It has been speculated that the increased energy consumption of spaces might have been caused from the false activation of lights during night time. Maniccia et al. (1999) , studied monitored energy use in perimeter and interior offices over a period of 4 months, finding that 43% energy savings can be achieved by using PIR occupancy sensors that had a 30-minute time delay as their setting instead of the conventional office hour (08:00-18:00) scheduling of lights [146] .

A study comparing energy savings among various lighting control options like occupancy sensing, manual dimming, brightness adjustment and daylighting found that 20-26% savings could be achieved by occupancy sensing strategies [147] . Their predicted energy savings from the model, in comparison to continuous occupancy, were 26 .1% with a 20-minute time delay and 33.3% for 5-minute time delay [148] .

For a typical VAV based HVAC system, the required supply air rate normally derives from zone set-point temperature. Pressure-independent Terminal-Box (TB) can control both minimum and maximum required supply air rates regardless of duct pressure. However, in conventional control, a fixed minimum supply air rate is provided to zones to ensure that sufficient ventilation air is delivered to the zone served irrespective of the thermal load [149] .

The minimum airflow rate through a terminal box is normally chosen to prevent poor air mixing, along with ventilation, but is also dependent on the type of diffuser and its sizing [150] . This proposed a control strategy of resetting set points in terminal boxes during unoccupied and lightly occupied hours [151, 152] .

However, there are also reported cases that show small or insufficient improvement in HVAC energy savings by following a proactive control strategy based on occupancy sensing and prediction. Oldewurtel F et al.(2013) found that predictive control for a single-occupant office space did not provide significant improvement over a reactive control system [153, 154] .

Similarly, another study explored various VAV terminal box control strategies which compares J o u r n a l P r e -p r o o f Journal Pre-proof fixed schedules to reactive occupancy based control and simulation based predictive control [135] and experimentation [136] for a small office space and found no significant improvement in energy savings between the reactive and predictive control. Small spaces with low occupancy are not ideal candidates for predictive control systems as found in [155, 156] .

Ventilation pays an important role in mitigation of viral loads in a confined space [3] [4] [5] [6] [157] [158] [159] [160] [161] [162] . However, enhanced ventilation through HVAC systems to mitigate viral load and maintain the health and comfort of occupants can lead to significantly higher energy requirements [18] .

Occupancy-based supply of required ventilation and elevated air movement at an increased zone set-point temperature thermal comfort and perceived air quality can be achieved [21, [163] [164] [165] [166] [167] [168] .

Other investigations found that an operative temperature range of 24-27 °C, with an air velocity of 0.4 m/s, was considered to be thermally acceptable to building occupants in hot-humid climate countries [169, 170] . Some earlier studies also found that better thermal comfort and perceived air quality was observed at both 26°C and 29°C if the zone is supplied with enhanced ventilation and elevated air movement employing ceiling fan [166, 171] . Elevated temperature set point brings significant energy savings reported in several studies [21, [172] [173] [174] [175] [176] [177] . Energy savings in the range of 32-73% among HVAC systems at widened temperatures band could be achieved with fans or personal ventilation [172] . There [164, [179] [180] [181] [182] [183] . Furthermore, whether the simple alleviation of thermal discomfort without considering ventilation affects measured and perceived air quality need to be ascertained [168] [169] [170] [171] .

Different aspects of occupancy-based building system control have been studied through objective specific literature review and from these past research outcomes, it can be hypothesised that occupancy is one of the main drivers behind the performance of building systems. Inaccurate information of occupants may lead to poor IEQ along with energy wastage. Occupancy information can be measured by understanding occupant's energy usage pattern along with their presence/absence. The measure for poor IEQ could be temperature, humidity, air movement and ventilation; and finally, the energy wastage could be determined by measuring the plug, light and cooling loads of the building studied.

A thorough literature review has highlighted various aspects of occupancy-based building system control. Although many studies exist that analyse the potential of occupancy information in controlling building systems to enhance energy efficiency and IEQ, there are specific gaps in these approaches which are highlighted below.

 Although many studies have been done for office and residential spaces, there is a limited study which combines and compares energy use variations among different kind of spaces.  Energy savings by optimizing the occupants energy use behaviour of plug and light loads during occupied hours has not been studied significantly for institutional building spaces with varying occupancy levels. Additionally, there could be a vast potential for energy saving just by confining the lighting to occupied zones for a larger space as the same level of illuminance is not required for entire spaces.

 In the research of cooling load control by optimizing VAV operation, most of the earlier studies are conducted for small space with few occupants to avoid the complexity of data collection from a larger space. Although few researchers have discussed larger/multizone spaces, these studies were mostly based on simulation data which lack actual building operational data.

 In most of the earlier studies, data collection period was quite short (a day to a week).

 The implication of occupancy based VAV operation is limited to energy saving assessment and there is insufficient reported case for its implications over IEQ especially AHU to zone level temperature and humidity control.

 To address this issue, researchers from both industry and academia are working towards occupancy-based Indoor environmental quality (IEQ) control via sensing and associated control of temperature, humidity and CO 2 . However, a single integrated system for multiple sensors sharing a common transceiver platform and networked with key actuators/actuating

systems have yet to be deployed though few initial prototypes have been developed.

J o u r n a l P r e -p r o o f Journal Pre-proof  It is important to recognize the varying IEQ requirement of occupants rather than expecting them to adapt to nearly homogeneous IEQ conditions. For example, a sensor-actuated network can be used to collect individual level CO 2 data that can be used to minimise the spread of COVID-19 infection through demand-controlled ventilation. Further, the proposed network could be used to improve occupants' health by knowing that people need exposure to a 24-hour cycle of light and dark to maintain circadian rhythm.

The literature review highlights that the primary reasons for the discrepancy between the planned and actual performance of a building are due to the assumption that the interaction between occupants and building systems (such as plug, light, and air-conditioning) is inactive or constant. However, in reality, the interaction between building systems and their occupants is dynamic, and the oversimplified understanding of occupancy does not correctly replicate the complex impact of occupancy over building systems. However, most of the occupancy-based VAV operation is limited to energy-saving assessment and there is an insufficient reported case for its implications over IEQ especially AHU to zone level temperature and humidity control.

The reported energy saving is due to the supply of minimum required ventilation. However, in current COVID times, ventilation plays an important role in the mitigation of viral loads in a confined space and enhanced ventilation through HVAC systems to mitigate viral load can lead to significantly higher energy requirements. So, the post-COVID-19 occupancy-based HVAC system control requires the supply of adequate ventilation at an elevated air movement and increased zone set-point temperature to create a synergy between occupant's health, IEQ perception, and building energy efficiency.

J o u r n a l P r e -p r o o f

Energy efficiency is the answer for building a secure and sustainable energy system

The National Human Activity Pattern Survey (NHAPS): A resource for assessing exposure to environmental pollutants

Impact of building ventilation systems and habitual indoor incense burning on SARS-CoV-2 virus transmissions in Middle Eastern countries

The contribution of dry indoor built environment on the spread of Coronavirus: Data from various Indian states

Indoor Air Quality: Rethinking rules of building design strategies in post-pandemic architecture

SARS-CoV-2 presented in the air of an intensive care unit (ICU), Sustain

Indoor air quality improvement in COVID-19 pandemic: Review, Sustain

Energy saving estimation for plug and lighting load using occupancy analysis

IEA EBC annex 53: Total energy use in buildings-Analysis and evaluation methods

Revealing occupancy patterns in an office building through the use of occupancy sensor data

Comparison of different occupancy counting methods for single system-single zone applications

A Probabilistic Model to Predict Household Occupancy Profiles for Home Energy Management Applications

Simulation and validation of indoor temperatures and relative humidity in multi-zone buildings under occupancy conditions using multi-objective calibration

Data driven occupancy information for energy simulation and energy use assessment in residential buildings

Building occupancy and energy consumption: Case studies across building types

Prediction of occupancy level and energy consumption in office building using blind system identification and neural networks

Computation of zone-level ventilation requirement based on actual occupancy, plug and lighting load information

Occupancy-based Zone-level Building Energy Control

Development of a data driven Occupancy-Based building energy and IEQ model

Occupancy-based zonelevel VAV system control implications of thermal comfort, ventilation, indoor air quality and building energy efficiency

Occupancy data at different spatial resolutions: Building energy performance and model calibration

Integrated sensor data processing for occupancy detection in residential buildings

Lam Khee Poh, Design-Build-Operate Energy Information Modeling for Occupant-oriented Predictive Building Control

Occupant behavior and schedule modeling for building energy simulation through office appliance power consumption data mining

Measuring building occupancy using existing network infrastructure

Building performance simulation for design and operation

Factors influencing occupants' blind-control behaviour in a naturally ventilated office building

Occupant control of passive systems: The use of Venetian blinds

Including Occupants in Building Performance Simulation : Integration of an Agent-Based Occupant Behavior Algorithm With Energyplus

Survey on energy consumption of air conditioning in summer in a residential building in Beijing

The effect of occupancy and building characteristics on energy use for space and water heating in Dutch residential stock

A comprehensive analysis of the impact of occupancy parameters in energy simulation of office buildings

Predicted vs. actual energy performance of non-domestic buildings: Using post-occupancy evaluation data to reduce the performance gap

User behavior in whole building simulation

Using results from field surveys to predict the effect of open windows on thermal comfort and energy buildings

The Impact Of The Building Occupant On Energy Modeling Simulations

Occupant behavior modeling for building performance simulation: Current state and future challenges

Robustness of building design with respect to energy related occupant behaviour

Analytical approach of thermosyphon solar domestic hot water system performance

Calibration of building energy simulations with Occupancy and plug-load schedules derived From metered building electricity consumption

The dark side of occupants' behaviour on building energy use

Occupancy-driven energy management for smart building automation

An analysis of the energy and cost savings potential of occupancy sensors for commercial lighting systems

ENERNET: Studying the dynamic relationship between building occupancy and energy consumption

ENERNET: Studying the dynamic relationship between building occupancy and energy consumption

User behavior and energy performance in buildings

Analysis of the rule of influence of hotel occupancy ratio on energy consumption

A study of the importance of occupancy to building cooling load in prediction by intelligent approach

Development of an adaptive window-opening algorithm to predict the thermal comfort, energy use and overheating in buildings

Manually-operated window shade patterns in office buildings: A critical review

Analytical approach of thermosyphon solar domestic hot water system performance

A high-resolution domestic building occupancy model for energy demand simulations

Occupancy detection through an extensive environmental sensor network in an open-plan office building

Real-time occupancy detection using decision trees with multiple sensor types, Procs

Estimating the inputs of gas transport processes in buildings

CO2-Based Occupancy Detection for On-Line Outdoor Air Flow Control

Intelligent environments and active camera networks

A survey of indoor positioning systems for wireless personal networks

Behaviour Profiling with Ambient and Wearable Sensing, 4th Int. Work. Wearable Implant. Body Sens. Networks (BSN

Opportunities To Save Energy and Improve Comfort By Using Wireless

Survey of wireless indoor positioning techniques and systems

Positioning with bluetooth

Localisation of forgotten items using RFID technology

Virtual occupancy sensors for real-time occupancy information in buildings

Window blind occlusion : a pilot study

Building automation and perceived control: A field study on motorized exterior blinds in Dutch offices

The use of shading systems in VDU task offices: A pilot study

On the behaviour and adaptation of office occupants

Interactions with window openings by office occupants

Towards a model of user behaviour regarding the manual control of windows in office buildings

Optimal design method for building energy systems using genetic algorithms

Fiddling with thermostats: energy implications of heating and cooling set point behavior

On the behavioral effects of residential electricity submetering in a heating season

Occupant behavior and successful energy conservation: Findings and implications of behavioral monitoring

How people use thermostats in homes: A review

Characterising occupant behaviour in buildings: towards a stochastic model of occupant use of windows, lights, blinds, heaters and fans

Unsupervised disaggregation of appliances using aggregated consumption data., 1st KDD Work

Dynamic predictive clothing insulation models based on outdoor air and indoor operative temperatures

Weather, clothing and thermal adaptation to indoor climate

Thermal Comfort:Analysis and Applications in Environmental Engineering

Evaluation methods and development of a new glare prediction model for daylight environments with the use of CCD cameras

Van Den Wymelenberg, Evaluating Human Visual Preference and Performance in an Office Environment Using Luminance-based Metrics

Survey of occupant behaviour and control of indoor environment in Danish dwellings

Gender differences in thermal comfort and use of thermostats in everyday thermal environments

The effects of gender, acclimation state, the opportunity to adjust clothing and physical disability on requirements for thermal comfort

Human Factors in Lighting

Monitoring manual control of electric lighting and blinds

Adaptive actions on shading devices in response to local visual stimuli

A Critical review of state-of-the-art energy and comfort related occupant behavior in office buildings Building Simulation

A novel approach for building occupancy simulation

A probabilistic model to predict building occupants'diversity towards their interactions with the building envelope

Research Methods for the Behavioral Sciences: Cengage Learning

External validity is more than skin deep: some answers to criticisms of laboratory experiments

What do laboratory experiments tell us about the real world

Data analysis and stochastic modeling of lighting energy use in large office buildings in China

Radio frequency occupancy sensing load control

Trained-once device-free crowd counting and occupancy estimation using WiFi: A Doppler spectrum based approach

Real-time room occupancy estimation with Bayesian machine learning using a single PIR sensor and microcontroller

A novel occupancy detection solution using low-power IR-FPA based wireless occupancy sensor

Hidden Markov Model for Improved Ultrasound-Based Presence Detection

UltraSense: A Self-Calibrating Ultrasound-Based Room Occupancy Sensing System

IoT-based occupancy monitoring techniques for energy-efficient smart buildings

Space-conditioning control employing image-based detection of occupancy and use

Building occupancy estimation and detection: A review

Sensor-based Occupancy Behavioral Pattern Recognition For Energy And Comfort Management In Intelligent Buildings

Building energy and comfort management through occupant behaviour pattern detection based on a large-scale environmental sensor network

Predicting occupancy counts using physical and statistical Co2-based modeling methodologies

Non-intrusive occupancy sensing in commercial buildings

A Bluetooth based occupancy detection for buildings

Personal thermal comfort models based on physiological parameters measured by wearable sensors

Smart phone based occupancy detection in office buildings

Energy intelligent buildings based on user activity: A survey

A review of occupant-centric building control strategies to reduce building energy use

Exploiting user behaviour for context-aware power management

Simulating the dynamics of occupant behaviour for power management in residential buildings

Design and implementation of smart energy management system for reducing power consumption using ZigBee wireless communication module

Design and installation of an advanced EIB TM fuzzy indoor comfort controller using Matlab TM

The Design and Implementation of a Smart Building Control System

Occupant-oriented mixed-mode EnergyPlus predictive control simulation

An indoor localization system based on artificial neural networks and particle filters applied to intelligent buildings

Lessons from an Adaptive Home

MavHome: an agent-based smart home

An opportunistic activity-sensing approach to save energy in office buildings

Smart occupancy sensors to reduce energy consumption

On the application of wireless sensors and actuators network in existing buildings for occupancy detection and occupancy-driven lighting control

An Approach for More Efficient Energy Consumption Based on Real-Time Situational Awareness

Intelligent Light Control J o u r n a l P r e -p r o o f using Sensor Networks

Vision-based user-centric light control for smart environments

Smart heating systems: Optimizing heating systems by kinetic-awareness

The potential for location-aware power management

Association for Computing Machinery

Alternative strategies for supply air temperature control in office buildings

Experiences with Occupancy based Building Management Systems

Occupancy-based zone-climate control for energyefficient buildings: Complexity vs. performance

Experimental study of occupancy-based control of HVAC zones

Token based scheduling for energy management in building HVAC systems

Occupancy learning-based demand-driven cooling control for office spaces

Co-ordinated management of intelligent pervasive spaces

User-led decentralized thermal comfort driven HVAC operations for improved efficiency in office buildings

Creating an Ambient-Intelligence Environment Using Embedded Agents

User-centric smart buildings for energy sustainable smart cities

iPower: an energy conservation system for intelligent buildings by wireless sensor networks

Field Analysis of Occupancy Sensor Operation: Parameters Affecting Lighting Energy Savings

Measured Field Performance and Energy Savings of Occupancy Sensors: Three Case Studies

Occupant Use of Manual Lighting Controls in Private Offices

Comparison of Control Options in Private Offices in an Advanced Lighting Controls Testbed

On the Prediction of Lighting Energy Savings Achieved by Occupancy Sensors

Fundamentals of HVAC Control Systems (IP)

Sizing VAV boxes

Terminal Box Air Flow Reset: An Effective Operation and Control Strategy for Comfort Improvement and Energy Conservation

Optimization Control Strategies for HVAC Terminal Boxes

Importance of occupancy information for building climate control

Adaptable cooling coil performance during part loads in the tropics-A computational evaluation

Proc. Fourth Int. Conf. Futur. Energy Syst. -e-Energy '13

Model-based controllers for indoor climate control in office buildings -Complexity and performance evaluation

Risk of Coronavirus Disease 2019 Transmission in Train Passengers: an Epidemiological and Modeling Study

Sustainable design of courtyard environment: From the perspectives of airborne diseases control and human health

Evaluation of the exposure risk of SARS-CoV-2 in different hospital environment

An overview on the role of relative humidity in airborne transmission of sars-cov-2 in indoor environments

Assessing and controlling infection risk with Wells-Riley model and spatial flow impact factor (SFIF)

How can airborne transmission of COVID-19 indoors be minimised?

Ventilation rates and health: Multidisciplinary review of the scientific literature

The Acceptable Air Velocity Range for Local Air Movement in The Tropics

Thermal sensations and comfort investigations in transient conditions in tropical office

Thermal comfort and self-reported productivity in an office with ceiling fans in the tropics

Energy saving and improved comfort by increased air movement

Human comfort and perceived air quality in warm and humid environments with ceiling fans

Towards a Brazilian standard for naturally ventilated buildings: guidelines for thermal and air movement acceptability

Development of an adaptive thermal comfort equation for naturally ventilated buildings in hot-humid climates using ASHRAE RP-884 database

Thermal comfort, perceived air quality, and cognitive performance when personally controlled air movement is used by tropically acclimatized persons, Indoor Air

Extending air temperature setpoints: Simulated energy savings and design considerations for new and retrofit buildings

Optimization of room air temperature in stratum-ventilated rooms for both thermal comfort and energy saving

Thermostat strategies impact on energy consumption in residential buildings

The impact of increased cooling setpoint temperature during demand response events on occupant thermal comfort in commercial buildings: A review

A knowledge based approach for selecting energy-aware and comfort-driven HVAC temperature set points

Performance characteristics of a fan filter unit (FFU) in mitigating particulate matter levels in a naturally ventilated classroom during haze conditions

Generic aspects of the airborne spread of human pathogens indoors and emerging air decontamination technologies

Ventilation Efficiencies of Desk-Mounted Task/Ambient Conditioning Systems, Indoor Air

Findings of Personalized Ventilation Studies in a Hot and Humid Climate

Performance of Personalized Ventilation in Conjunction with Mixing and Displacement Ventilation

Human Response to Five Designs of Personalized Ventilation

Personalized ventilation: evaluation of different air terminal devices

☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.J o u r n a l P r e -p r o o f