A proposal on residential lighting design considering visual requirements, circadian factors and energy performance of lighting

ABSTRACT This study assesses residential areas which have been converted into workplaces and are still used after the Covid-19 in terms of visual, non-visual, and energy performance requirements of lighting. We proposed a lighting design using LED systems with dimmable and tuneable features. Circadian factors in WELL Building Standard are analyzed for compatibility with the current visual requirements. The impact of various design parameters on lighting energy consumption, including daylight availability, lighting schedules, lighting control strategies, and light reflectance value of walls is evaluated through a case study in Turkey. Although the annual lighting energy consumption is higher than it was before the pandemic, building energy simulation results show that the application of LED systems with lighting energy measures can improve lighting energy performance by up to 38%. From the non-visual dimension of light, our data indicate that higher melanopic illuminance and/or colour temperature of light sources are necessary to entrain and sustain the circadian rhythm under overcast sky conditions in winter months. On the other hand, an increase in luminous intensity can lead to glare and higher energy consumption while a higher colour temperature may affect the physiology and psychology of occupants negatively.


Introduction
The COVID-19 pandemic has led many of us to a retreat at home, performing a great deal of daily routines indoors. Meanwhile, the modes of working, teaching, and learning have been widely rethought, hybrid modes of working emerging as a common solution in many workplaces. As residential spaces have emerged to take on working and learning functions with the pandemic, improving indoor environmental quality and energy conservation have become imperative, while ensuring comfort-related design needs. In this context, the design of interior lighting will also be considered in line with these issues.
It is well known that the light does more than just enabling vision. Besides supporting visual perception, the lighting design influences human physiology and psychology through its impacts on our mood, cognition, and the circadian system Dai et al. 2017). Besides visual and non-visual aspects of lighting design, energy performance requirements need to be satisfied. Well-being is upheld by proper lighting design that satisfies visual and psychological comfort (So and Leung 1998;IWBI 2022;CIE S 026/E 2018). Therefore, recent studies have focused on the emerging lighting technologies that present opportunities for different lighting scenarios while promoting psychological well-being, improving user performance, and saving energy Awada et al. 2021).
As shown by Viola et al. (2008), Arendt (2010), Do and Yau (2010), Jarboe, Snyder, and Figueiro (2020), light is effective in generating a non-image-forming, biological response affecting the human circadian clock. The human circadian system is impacted by not only the illuminance at human cornea but also the spectral power distribution, time, duration, and the spatial distribution of light.  state that a lighting system delivering a circadian stimulus (CS) equal to or greater than 0.3 can reduce sleepiness and increase vitality and alertness in office workers. High levels of light reaching the cornea during the daytime will align circadian rhythms with a dayactive and night-inactive (sleeping) pattern, hence increasing alertness during working hours CIE S 026/E 2018;Blume, Garbazza, and Spitschan 2019). Similarly, Sahin and Figueiro (2021) analyzed the stimulating effects of bright white light (>1000 lx at the cornea) and compared with lower levels (30 lx) of short wavelength (blue light), showing that the long wavelength (red light) promoted alertness without suppressing melatonin levels and disturbing the circadian system. Another international building standard of the International WELL Building Institute (IWBI) focuses on the stimulation from light to support circadian entrainment indoors. Yet, no maximum threshold is stated in WELL Building Standard v2 Q4 (2022) which is defined as a system for "measuring, certifying and monitoring the performance of building features that impact health and well-being" (IWBI 2022).
In this study, we aim to assess residential spaces that have been converted into offices and/or study areas during and after the Covid-19 pandemic in terms of their visual, non-visual, and energy performance requirements of lighting. In this regard, we propose a lighting design with LED technology which is compatible with lighting control strategies and the human-centric lighting concept. The impacts on the annual lighting energy consumption of four factors i) the building orientation, ii) the occupancy-related lighting schedules, iii) manual and daylight-linked lighting control strategies, and iv) light reflectance value (LRV) of the walls is analyzed for the selected residential units.

Residential lighting and the Covid-19 pandemic
The Covid-19 pandemic has changed our living, working, and education conditions and habits. Since residences have been used as offices and/or study areas since the start of the pandemic, the prominence of residential lighting in the annual total energy consumption has increased with the elevated occupancy rate. According to the Turkish Energy Market Regulatory Authority's statistics (EPDK 2022), household electricity consumption in the study area increased by 8.2% in the period from the beginning of the pandemic to the controlled relaxation of the Covid-19 containment measures in Turkey, when compared to the same period of the previous year. In the same period, the annual household electricity subscription for households has shown a linear increase of ~0.2% (EPDK 2022). Demand for lighting as well as other appliances plug load is effective in the rise of this electricity load. The impact of "stay home living patterns" on energy consumption in residential buildings including lighting is also reported by Farrow (2020), Hinson (2020), IEA (2020), Krarti and Aldubyan (2021), Wang, Li, and Jiang (2021), Rouleau and Gosselin (2021), Abdeen et al. (2021), andSurahman et al. (2022).
Previous studies have also emphasized that residential lighting becomes the second-largest electricity consumer in buildings with about 20%, accounts for about 10% of the household electricity consumptions in the European area and a significant contributor to energy costs in most industrialized countries (Makaremi et al. 2017;EEA 2014;Baloch et al. 2018). Although there are numerous studies evaluating the visual requirements for user comfort and lighting energy consumption in households, there were no direct measurements of the impact of occupancy patterns during and after the pandemic and working from home on lighting energy. Furthermore, studies that focus on the non-visual effects of lighting in developing countries such as Turkey are limited. Therefore, we aimed to look at residential lighting design in a holistic way.

Residential Lighting Design Criteria
As recommended in EN12464 (2011) Dai et al. (2018a); (2018b), Jarboe, Snyder, and Figueiro (2020) addressing the circadian lighting design principles. The research by Duff, Kelly, and Cuttle (2015) and Aguilar-Carrasco et al. (2021) highlights the importance of sufficient melatonin suppression to promote daily circadian entrainment with an exposure of daylight at least one hour. The literature review by Alkhatatbeh and Asadi (2021) indicates that a light source must be considered together with the circadian effect which is dependent on the melanopic illuminance, spectral power distribution (SPD), and the direction of the distribution of light. Energy efficiency is another important criterion in lighting design. Mardaljevic, Heschong, and Lee (2009), Blomsterberg (2011), Boubekri (2014), and Frascarolo, Martorelli, and Vital (2014) suggest that the first principle is to maximize the benefit from daylight. Singh and Garg (2010), and Yu and Su (2015) argued that the function of a space and its daylight availability determine daytime electric lighting load and energy savings. In this regard, Das and Saikat (2015) analyzed outdoor illuminance and the artificial lighting requirements in residential buildings, analyzing the impact of the location and the orientation of the building including the interior design based on several plan typologies, and occupancy patterns. Zhen et al. (2019) calculated the energy consumption of indoor lighting by optimizing natural lighting in residential buildings in Xi'an, China.
Some authors in the extant literature argued that not only daylight availability and technology but also humans' energy-related behaviour in buildings will be included in energy performance studies to assess whether building energy conservation goals can be achieved (Zhang et al. 2018). Bourgeois, Reinhart, and Macdonald (2006) studied behavioural models and investigated the impact of occupancy-based manual and automated lighting control on the total energy in buildings. The results show that the occupants prefer daylighting rather than artificial lighting can reduce the primary energy consumption by 40%, when compared to the users relying on constant electric lighting. In a recent study by Gerhardsson, Laike, and Johansson (2021), occupants' experiences with their residential lighting in Sweden are explored by a survey. The authors provide information on the background of wasted lighting energy linked to behavioural goals, such as safety, or psychological wellbeing or social needs.
The choice of light source and luminaire as well as the use of lighting control strategies has also a decisive impact on the lighting energy efficiency of buildings over the electricity consumption (Aman et al. 2013;Attia, Hamdy, and Ezzeldin 2017). There have been numerous studies to investigate lighting energy savings by controlling the regulation of lighting power gradually (dimming) or completely (on/off). For example, the literature review by Dubois and Blomsterberg (2011) provides evidence for energy conservation in office buildings by the improvements in electric lighting installation and daylight harvesting strategies. The findings of Frascarolo, Martorelli, and Vital (2014), Makaremi et al. (2019), Van Thillo, and Verbeke and Audenaert (2022) document the benefits of using LED technology with lighting control automation in energy savings.
The light reflectance value (LRV) of surfaces is another design variable that affects lighting energy consumption. Stephen and Coley (2011), Mohelnikova and Hirs (2016), Mangkuto, Rohmah, and Asri (2016), and Singh and Rawal (2011) suggest that the reflective property of interior materials can contribute to achieving the required illuminance, reducing the operating hours of artificial lighting, hence lowering the overall energy consumption.
The lighting design criteria outlined above and discussed in the scope of the study are illustrated in Figure 1.

Description of the study area
The research investigated the residential lighting design criteria depicted in Figure 1 through a case study of multi-storey apartment blocks in Bursa, Turkey. The housing complex meets the minimum energy performance standards in Turkey (BEP 2008) and holds an energy performance certification (EPC) according to Building Energy Performance National Calculation Methodology (BEP-TR 2010). The complex comprises six-storey residential buildings that were constructed on an area of 21,500 sq. m, with two flats on each storey.
The shadow analysis of the apartment blocks for the morning and noon hours on solstices is shown in Figure  2. Block VII is the least favourable building in terms of annual lighting energy; whereas Block II is the best. Considering the contrasting positions of these two blocks, lighting and building energy simulations were run and the results of improvement measures were evaluated for these two buildings. The sample residential units are located on the first floors of these two reference buildings with different obstruction angles and facade orientations. The flats have otherwise equivalent architectural and interior design features.

Material and Method
In this work, we propose a lighting model that considers the combination of visual and non-visual aspects of light together with the energy performance for exemplar residential units. The lighting design criteria in Figure 1 are investigated by computer-based building performance simulations for the reference buildings. Photopic and melanopic illuminance levels were calculated by DIALux Evo, a widely used parametric tool for lighting design and luminotechnical calculation. Annual lighting energy consumption was calculated by DesignBuilder software which is the graphical interface of the EnergyPlus simulation engine developed by the US Department of Energy.

Residential Lighting Design Analysis
In previous research, Manav and Kaymaz (2021) analyzed occupants' preferences for lighting fixtures, lamp type, and colour temperature based on a questionnaire. In the present study, we use data from this survey based on which we model the reference building's lighting design in the DIALux Evo software. The lighting scenario considers dimmable LED lighting fixtures with a minimum of 50 lm/W efficacy, 80 CRI, and tunable white technology. The indoor environments are assessed according to the visual requirements recommended by EN 12464-1 standard (2011), IESNA Illuminating Engineering Society of North America (2011) and Guide F (2012) lighting guides together with the threshold levels for circadian entrainment in the toolbox of IWBI (2019) and WELL standard v2, Q4 (IWBI 2022).
In relation to the indoor visual requirements, the static lighting simulations are performed for the horizontal working planes (calculation grid is located 0.80 m above the ground and 0.5 m away from the room perimeter). All the luminaires in the residential unit are set to maximum (100%) and 60% intensity, respectively. Average illuminance level (E av ), uniformity of lighting (U o : E min/ E av ) and unified glare ratio (UGR) are recorded for the visual tasks with respect to general and task lighting. The light reflectance values for floor, wall, ceiling and furniture are as follows; ρ floor = 0.4, ρ wall = 0.5, ρ ceiling = 0.8 and ρ furniture = 0.6. Table 1 shows the lighting system layout with isolines at the workplane level in a false colour rendering at 100% intensity and also the luminaire specifications. Further technical data about the selected products can be reached at the website of DIALux library (Lumsearch 2022). The following lighting design principles are considered in this study.
The living room is equipped with a dimmable and tunable luminaire above the seating area together with indirect strip lights that are installed as cove lighting. The dining area is also lit by a pendant luminaire with direct-indirect light distribution. One portable floor lamp and LED light strips are located in the living room. Pendant lighting fixtures are utilized for the general lighting of the kitchen and also above the dining table for task lighting. Linear light fixtures are placed under the kitchen cabinet to highlight the countertop. Master bedroom and child rooms are illuminated by

Circadian Lighting Design Analysis
Several metrics for quantifying the non-visual effects of light have been proposed in the literature. Rea et al. (2012) proposed a model that introduces the parameters "circadian stimulus" (CS) and "circadian lighting" (CLA) to characterize light as a stimulus to the biological clock. The circadian stimulus model is based on data that depends on studies about the impact of light on melatonin suppression. This is a scale from 0.1 to 0.7 CS which is recommended by Figueiro, Gonzales, and Pedler (2016), and Rea and Figueiro (2018). They indicate that an exposure of 0.3 or higher CS at the eye for at least one hour in early hours of the day is effective in stimulating the circadian system. Lucas et al. (2014) proposed a metric to quantify the effective irradiance for each of the five photoreceptive inputs which includes two separate measures of the light's effect on the circadian rhythm: i) WELL Building Standard unit Equivalent Melanopic Lux (EML), and ii) the melanopic Equivalent Daylight Illuminance (mEDI) proposed by The International Commission on Illumination (CIE S 026/E 2018).
The entrainment of the circadian rhythm is evaluated in accordance with the recommendations from WELL Building Standard v2 Q4 (2022). WELL recommends the following values and settings for the living environment, work and learning areas: • The measurements will be conducted for the vertical plane in the center of the room at the eye level (h:1.2 m) facing forward to stimulate and sustain the circadian rhythm of users. • During the daytime, 200 or more EML is required in the living environment. • During the nighttime, 50 or less EML will be provided on the vertical plane (h: 0.76) in the living environment. • At 75% or more of workstations, a minimum of 200 EML will be provided during 9:00-13:00 for every day of the year, which may incorporate daylight.

EML or above will be provided at all workstations by electrical lights
• At 75% or more of study desks, a minimum of 125 EML will be provided at least for 4 hours every day of the year, which may incorporate daylight.
In order to assess whether the requirements in the WELL Standard for the specified time intervals in the where L is the simulated illuminance on the working plane and R is the melanopic ratio of the light source. This efficacy (R) is also referred to as "Melanopic to Photopic ratio" (M/P Ratio) in the WELL Building Standard. Figure 3 shows the relative spectral power distribution (SPD), the circadian stimulus, and the visual stimulus of the light sources in the toolbox of IWBI (2019). For the EML calculations, the simulated average illuminance (E av ) is multiplied by the M/P ratios for the 2700 K and 4000 K LED sources and daylight at 6500 K (D65) which are assumed to be 0.45, 0.76, and 1.1, respectively in line with IWBI (2019).

Annual Lighting Energy Performance Analysis
The reference building energy model was generated in the Designbuilder software and the annual energy performance simulations were run in the EnergyPlus engine in accordance with the architectural project and Building Energy Performance National Calculation Methodology (BEP-TR) results.
The window-to-wall (WWR) ratio of the building is 27%. The total solar heat gain coefficient (SHGC) is 0.75 and the visible light transmittance of windows (T vis ) is 0.80. The reference building's occupancy data is based on a previous survey (Manav and Kaymaz 2021) and the lighting power densities are specified according to the selected luminaires in the lighting scenario, as defined in Section 2.1.
In the lighting energy analysis, occupancy-based lighting schedules, daylight-linked lighting control and the light reflectance value (LRV) of walls are evaluated. The average lighting energy use per square meter is compared with the reference buildings for the following scenarios.

Occupancy-based Lighting Schedule Scenarios
Lighting schedules were defined for general lighting operating hours, from 07:00 to 24:00, including weekdays and weekends. We specify these scenarios consistently with the occupancy patterns during and after the pandemic as identified in the research by IEA (2020), Krarti and Aldubyan (2021)  The considered lighting schedule scenarios are as follows: • S 1 -pre-pandemic scenario: Lighting system operates mostly in the morning and during the night in line with previous survey results (Manav and Kaymaz 2021). It is assumed that each residence has at least one working member between 08:00 and 18:00. • S 2 -stay-at-home scenario: The scenario reflects the increased use of residential space during the governmental restrictions and Covid-19 lockdown. • S 3 -post-pandemic scenario: This scenario captures today's hybrid working conditions.
The occupancy-based lighting schedules are visualized in Figure 4, where the occupancy fraction is between 0 and 1 with zero indicating that the lights are off and one indicating they are fully on for the specified room and hour. Lighting schedules regarding the occupancy rates before, during, and after the Covid-19 pandemic. (Continued).

Daylight-linked Lighting Control Scenarios
Daylight availability and lighting control strategy are other major factors that affect lighting energy consumption in buildings. In the study, we consider four lighting control strategies to test the extent and efficacy of lighting energy savings. C 1 scenario: Lights are fully-on due to user presence in a zone. C 2 scenario: Lighting system is controlled by manual switching through according to the lighting schedule. Lamps are switched off when the target general lighting level, i.e., 150 lx in the living room, 300 lx in the kitchen, 100 lx in the bedrooms is achieved on the horizontal working plane by daylight only. C 3 scenario: In the stepped lighting control strategy, the general lighting level can be dimmed by four steps, and finally switched on/off in response to daylight availability by a photo-sensor installed in the center of the room. C 4 scenario: In the linear lighting control strategy, the light level can be dimmed continuously and linearly from maximum electric power to minimum light output as daylight penetration increases. The LEDs are switched off completely when the necessary set point is guaranteed by daylight.

Interior Lighting Simulation and EML Calculation Results
The lighting scenarios with two different dimming levels (electric light intensity: 100% and 60%) and three colour temperature values (CCT: 2700 K, 4000 K, 6500 K) are evaluated with respect to the recommended indoor visual and non-visual lighting requirements. The simulation results for the photopic and melanopic light intensity levels in the living, work, and learning areas are presented in Table 2 under different daylight scenarios (09:00 and 13:00 on the two solstice days at 6500 K). Due to the extended and shifting working hours with the Covid-19 pandemic, simulations were iterated including the nighttime scenario. The simulation results for the indoor visual and circadian lighting requirements at night are given in Table 3. The electric lighting system is assumed to be the same for the two reference residential units with equivalent interior and architectural features.
The static simulation results in DIALux Evo are demonstrated by colour-coding where yellow indicates that the results meet the minimum threshold values, while red/green cells represent the values below/above the threshold. For UGR results, the inverse of the colour scale is applied.
Considering the recommended light level on the task area (EN 12464-1, 2011;IESNA, 2011;CIBSE, 2012), the required minimum average illuminance (E av ) of 150 lx is maintained for the living room (152 lx under 60%); 100 lx is satisfied for the master bedroom (101 lx under 60%) and child room (121 lx under 60%), and 300 lx is achieved for the possible workstations such as the dining table (327 lx under 60%) in the kitchen and the living room (302 lx under 60%), and study desk in the child room (480 lx under 60%) for both electric lights intensity scenarios (Table  3). When the luminaries in the kitchen are dimmed by 60% intensity, 300 lx general lighting on the horizontal working plane is not met by electric lighting (200 lx under 60%). However, the horizontal illuminance requirement is met with the influence of daylight except at 09:00 on the winter solstice (231 lx for SE and 225 lx for N under 60% in Table 2).
The illuminance uniformity (U o ) shall be greater than 0.40 for general lighting and 0.60 for task lighting for working areas (European Standard EN 12464-1 2011). According to the electric lighting simulation results in Table 3, 0.40 is satisfied in all zones; however, 0.60 is not achieved on the task plane of the bedrooms. The results for electric lights at 100% intensity also show that the viewing positions from the bedrooms exceed the UGR limit and causes uncomfortable glare, which should also be considered for the calculation points in the living room. On the other hand, the calculations of UGR measured at 60% intensity indicate that most working planes are below the recommended maximum 19 UGR for general lighting and 22 UGR for task lighting.
The EML calculations in Table 3 demonstrate how the residential lighting system would perform if all LED lighting fixtures were installed with flexibility where CCT can be adjusted between 2700 K and 6500 K together by dimming up and down the electric light intensity. When melanopic light intensity in the living environment of each zone is evaluated, the results in Table 3 show that the minimum requirement of 200 EML on the vertical plane at the eye level of the occupant can rarely be achieved by only electric lights (203 EML in the living room and 190 EML in the kitchen at 6500 K under 100%). Similarly, the same non-visual requirement cannot be satisfied in any of the calculation areas on December 21 st at 09:00 due to the lack of sunlight penetration in Table 2. Despite the influence of diffused daylight on winter solstice and direct daylight on summer solstice, the vertical illuminance is not sufficient for any of the bedrooms in. On the other hand, 200 EML is exceeded in some of the living environments on December 21 st when luminaires are tuned to 6500 K CCT at 100% intensity (268 EML and 249 EML for the SE facing rooms in Block II; 264 EML and 248 EML for the N facing rooms in Block VII in Table 3).
During the nighttime, the melanopic light intensity is above the maximum threshold of 50 EML in most of the residential areas at 100% intensity; yet, this target can be achieved by setting luminaries to lower CCT and/or dimming the light sources after 20:00 (2700 K and 60% intensity scenario in Table 3), as suggested in WELL Building standard (IWBI 2022). Finally, in the context of home-office hybrid working, eye-level vertical illuminance is evaluated as the critical parameter in stimulating the circadian system. The minimum circadian lighting design target of 200 EML is achieved for all the work areas either by incorporating daylight (Table 2) or by changing the M/P ratio from 0.45 to 1.1 at 100% electric light intensity (Table 3). Table 2 represents that a minimum of 125 EML is achieved for at least 4 hours per day (between 9:00 and 13:00) on solstice days for the designated learning areas. However, as seen in Table 3, these task areas do not meet the minimum melanopic requirements on the vertical plane at 60% electric light intensity with luminaires of 2700 K CCT.
Since there is no direct correspondence between horizontal illuminance and vertical illuminance, the light level on the horizontal plane has to be increased significantly to also achieve sufficient EML that complies with the minimum melanopic requirements. Dubois et al. (2015) also pointed out the potential impact of vertical illumination suggesting that lower horizontal lighting levels can be more easily tolerated when vertical surfaces are well-lit. On the other hand, increasing the luminous flux should not cause overly high light levels, glare formation and excessive energy consumption, as discussed in the findings of the review by Alkhatatbeh and Asadi (2021). Since it is a complex task, an integrated approach (both daylight and artificial light with dynamic and tunable features) should be followed to satisfy the visual and non-visual requirements in residential spaces.

Annual Lighting Energy Simulation Results
The proposed residential lighting design is assessed for three different lighting schedules, four different control strategies, and four wall LRV options as defined in Section 2.3. The lighting energy results under the different scenarios are presented for Block II and Block VII in Figure 5.
Due to the increase in the occupancy rate, hence in the daily use of light sources during the Covid-19 pandemic, the annual lighting energy consumption in the stay-at-home scenario (S 2 ) is the highest in all cases. The pre-pandemic (S 1 ) to stay-at-home (S 2 ) percentage increase in lighting energy consumption varies between 24.1% and 67.5% across different cases, depending on building orientation, user behaviour, and lighting control preferences. The increase in lighting energy consumption is in accordance with the findings of Krarti and Aldubyan (2021) and Surahman et al. (2022) that increased electricity demand occurred in domestic households during lockdowns.
The use of electric lighting control systems is effective for energy efficiency through reducing the operating time of the lighting system. When general lighting level is modulated by photo-sensors, a percentage improvement of 4.7-5 and 5.6-8.7 is achieved by stepped-dimming control strategy (C 3 ), when compared to manual control (C 2 ) for Block II and Block VII, respectively. Likewise, linear off dimming control (C 4 ) decreases lighting energy consumption by 6.3-6.8% for Block II and by 8.4-11.3% for Block VII in comparison to the C 2 (manual on/off) scenario. Depending on the lighting schedule, the annual lighting energy consumption in Block VII (N facing living space) is 4.8-8.1% higher than in Block II (SE facing living space). This building orientation-based lighting energy difference decreases with the use of lighting control.
As Frascarolo, Martorelli, and Vital (2014) pointed out in their work, it is important to consider the need for manual control by integrating customizable lighting solutions for user-managed workstations. On the other hand, constant use of artificial lighting may result in higher energy consumption through the over-use of electric lighting and also in discomfort glare during the day. The simulation outcomes reveal substantial differences in energy loads between scenarios where the lights are controlled by photo-sensors and no user is involved (C3, C4) and the scenario where the lights are manually controlled in the fully-on mode (C1). For instance, linear off dimming control in the stay-athome scenario (C 3 -S 2 ) leads to electric savings up to 35.7% compared to the C 1 -S 2 scenario. A similar conclusion was reached in the literature review by Dubois and Blomsterberg (2011) and Chew et al. (2017) that more than 40% energy savings can be achieved by using daylight-linked control systems depending on building orientation, window characteristics, reflectance of interior surfaces and shading factor.
Our findings also support the earlier results by Singh and Rawal (2011), Mohelnikova and Hirs (2016), Mangkuto, Rohmah, and Asri (2016) that changing the LRV of surface colours influences the lighting energy consumption without altering the number of light sources. Depending on the occupancy scenario, the annual lighting energy is decreased by 3.3-3.8% in Block II and 5.1-6.8% in Block VII via changing the LRV from ρ wall = 0.5 to ρ wall = 0.8.

Conclusion
The study proposes a residential lighting design that focuses on indoor visual requirements and lighting energy performance, while stressing the significance of lighting on circadian system. The impacts of several lighting design parameters are assessed through a simulation case study. The interior lighting simulations show that a relatively high melanopic illuminance and/ or M/P ratio of sources are necessary to achieve the EML targets without daylight or under overcast sky conditions in winter months, which is key to stimulating and sustaining the circadian rhythm of occupants working and studying in residential spaces. However, increasing the vertical light intensity also increases the illuminance distribution on the horizontal plane, which causes glare and should be avoided according to European Standard EN 12464-1 (2011). Additionally, changing the M/P ratio with a higher (or lower) CCT influences the space perception, which is effective on the physiology and psychology of occupants. The horizontal illuminance needs to exceed 500 lx in order to reach the recommended targets of 150 EML and 125 EML for work and learning areas, respectively by electric light. However, this requires greater lighting power installation and energy consumption. On the other hand, the values recommended in the WELL building standard can be met across all cases when the daylight is included. Thus, interior lighting design should support the flexibility for the task planes with dimmable and tunable lighting features and a balance of melanopic and photopic illuminance should be achieved for workplaces by incorporating daylight in the lighting design.
Although the annual lighting energy consumptions are higher than the levels before the pandemic, the building energy simulations show that integrating dynamic LED systems with daylight-linked lighting control strategies and walls with higher LRV can improve lighting energy performance by up to 38%. Therefore, we propose lighting solutions that are robust against the changes in living, working, and learning habits expected to be sustained even after the pandemic.
The presented work is part of an ongoing research project investigating user comfort and residential lighting design requirements considering today's living and working conditions underlying the significance of occupancy patterns. From this standpoint, we believe that circadian lighting is a key component of the holistic lighting design approach. Future studies could further explore this scientific field by comparing the results based on computer simulations with on-site measurements. We want to apply our lighting setup in practice with dynamic LED lighting systems. To determine optimal lighting design solutions for user requirements, further post-occupancy investigations are also needed. Since energy efficiency in lighting is the main focus parameter, the association between circadian lighting design and daylight availability will also be addressed by annual climate-based daylight metrics in future research.

Disclosure statement
No potential conflict of interest was reported by the author (s).

Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Notes on contributors
Egemen Kaymaz completed her B.Arch in 2009 and received M.Sc. Degree in Environmental Control and Building Technology in 2012 at Istanbul Technical University. As a Ph.D. candidate in Bursa Uludag University, her research explores design support models, visual comfort, lighting energy performance, and optimization of buildings. Banu Manav is an Interior Architect, with a Ph.D. in architecture/building science. Prof. Dr. Manav is currently the Dean of the Faculty of Art and Design, Kadir Has University, Turkey. She is the delegate of Turkey for (CIE) Division 6 (Photobiology and Photochemistry), a member of AIC Study Group on Environmental Colour Design, and the General Secretary of Turkish National Committee on Illumination. Her research areas include artificial lighting system design, space perception-color-light, light-health, lighting quality, and user comfort-design parameters.