An Energy-Efficient Approach for Thermal Comfort and Sleep Quality in Subtropical Bedrooms

Abstract

This study conducted a within-subject study to assess sleeping environmental comfort, acceptance, and self-reported sleep quality in air-conditioned and mixed-mode ventilated bedrooms in a subtropical region during the summer. A wide thermal comfort temperature range of 22.2 °C to 28.2 °C was observed, with slightly warmer thermal sensation at higher temperatures but no significant differences in sleep quality or environmental comfort acceptance within this range. Subjects adapted to warmer sleeping conditions by choosing lighter clothing and bedding insulation. Energy simulations indicated a reduction in the percentage of nights requiring cooling from 65% to 23% by increasing the set-point temperature from 22 °C to 28 °C, resulting in a potential 95% savings in cooling energy. This study advocates for an economical and energy-efficient approach to enhance sleeping thermal comfort while reducing cooling energy usage. These findings offer valuable insights for improved residential building design and optimized cooling energy management practices, especially in light of intensified climate change and the imperative for behavioral changes to promote building sustainability.

1. Introduction

Sleep, an essential physiological process occupying a substantial portion of human daily routine [1], is pivotal for bodily restoration, memory consolidation, and biological regulation [2]. Adequate and quality sleep is linked to improved cognitive performance, an enhanced immune system, a reduced risk of chronic diseases, and better emotional regulation [3]. Therefore, achieving optimal sleep is crucial for overall health and well-being.

Thermal comfort during sleep involves an interaction between the body’s thermoregulatory mechanisms and environmental conditions of the sleep environment. During sleep, the core body temperature (CBT) naturally fluctuates, signaling sleep onset as it declines and awakening as it rises [4]. The hypothalamus, the brain’s thermoregulatory center, is crucial in maintaining this optimal temperature range by adjusting physiological responses such as sweating and peripheral vasodilation [5]. Various environmental conditions can significantly influence these thermoregulatory processes and thus disrupt sleep. For instance, significant heat exposure from ambient temperature and bedding insulation can impact slow-wave sleep (SWS) and rapid eye movement (REM), triggering behavioral thermoregulation such as adjusting bed covers or changing body position [6,7]. Conversely, under low or neutral air temperatures, air turbulence around the body has been shown to affect sleep quality detrimentally [8]. Therefore, bedroom thermal conditions are crucial in determining sleeping thermal comfort, potentially influencing sleep quality and subsequent day-to-day productivity [9].

Thermal comfort and sensation vary considerably depending on the indoor environmental control methods. In regions such as Hong Kong and Japan, characterized by humid subtropical climates prone to extreme heat, indoor environmental control strategies present unique challenges. While air conditioning effectively moderates temperature and humidity, studies suggest that setting indoor temperatures in the pre-sleep waking state may not align optimally with sleeping conditions, as a higher neutral temperature is usually preferred during sleep [10,11]. Notably, the tendency towards overcooling can disrupt sleep stages and cause thermal discomfort [12,13]. Alternatively, adopting mixed-mode or natural ventilation, despite offering less control over thermal conditions and may result in thermal conditions that fall outside the standard thermal comfort limits, has demonstrated increased occupant acceptance of these warmer environments [14,15]. While adaptive thermal behaviors in waking states are well-documented in naturally ventilated settings [16], their thermal adaptation during sleep remains unproven. Therefore, a comparative analysis exploring the effects of air-conditioned versus naturally ventilated environments on sleep quality and thermal comfort parameters is necessary.

Cooling demand in buildings is a significant contributor to energy consumption. In the European Union (EU), energy-related greenhouse gas (GHG) emissions from buildings account for one-third of total emissions, with the residential sector responsible for one-quarter of final energy use [17,18]. The adoption of air conditioning is projected to increase significantly due to rising temperatures from climate change and growing income levels [19]. Excessive use of air conditioning negatively impacts urban thermal comfort, raising street temperatures and worsening the urban heat island effect [20,21]. There is an urgent need to mitigate the harm inflicted on the urban environment as part of the broader effort to combat climate change. Solutions include advancing cooling technologies and promoting sustainable alternatives, such as mechanical fans, to reduce dependence on air conditioning [22].

This study conducted a within-subject examination of sleeping environmental comfort, acceptance, and self-reported sleep quality in air-conditioned and mixed-mode ventilated bedrooms. It sought to address the research question regarding the influence of ventilation on sleeping thermal comfort, by determining the thermal comfort levels in bedroom environments during summer and estimating the energy-saving potential of adopting higher set-point temperatures within the acceptable thermal range. It is the first study to explore thermal adaptation during sleep induced by behavioral changes. The findings provide insights for enhanced residential building design and optimized cooling energy utilization practices, which is instrumental in reducing long-term carbon emissions and promoting sustainability within the built environment.

2. Materials and Methods

2.1. Participants

In total, 15 healthy adults (10 males, 5 females) aged 16–61 were recruited for this study. All participants were local Hong Kong residents living in public apartments with window-type air conditioners. They were non-smokers, had no sleep disorders, and had average body mass indices (BMI) ranging from 18.7 to 23.5. During the experimental days, the participants were instructed to refrain from engaging in intense physical activities such as exercising, consuming alcoholic or caffeinated beverages within 8 h of bedtime, and obtaining at least six hours of sleep. The participants were also healthy during the days of the experiment, without taking any medication that could induce or regulate their sleep. The participant selection process aimed to minimize potential bias related to individual demographics, health conditions, or substances that could affect sleep quality.

2.2. Experimental Procedure

This study required each participant to complete a two-night sleep experiment, with trials conducted on consecutive nights. The participants experimented with their home environments to maintain ecological validity and minimize the effects of an unfamiliar sleeping environment on sleep quality. One night, the participants slept in their beds, and an air conditioner cooled the bedroom as mechanical ventilation. The other night, a mechanical fan was used for cooling instead of the air conditioner as mixed-mode ventilation. The two ventilation methods were selected due to their prevalent use for cooling purposes in subtropical residential settings. Each method is effective for cooling and has its own advantages and disadvantages. The order of air conditioner and fan use was randomized to mitigate the sequence effect, with half of the participants using the air conditioner on the first night and the other half using the fan first.

Before bedtime, the participants recorded when they went to bed and noted the noise level in the bedroom environment measured by a sound level meter (Lutron SL-4023, Taiwan, China). They were instructed to activate the thermal environment monitoring devices adjacent to their sleeping positions. Throughout the night, air temperature, radiant temperature, and relative humidity were measured using a heat index wet bulb globe temperature (WBGT) meter (Lutron WBGT-2010SD, Taiwan, China), with readings taken every five minutes. The air velocity was assessed using an air velocity transducer (TSI 8455, Shoreview, MN, USA) with a 1-min sampling interval. The participants were free to choose their sleepwear and bedding (i.e., type of bed covering and cover percentage) as they usually would, given the perceived environment. No disturbances were introduced during the sleep period.

Immediately upon waking, the participants completed a questionnaire recalling their previous night’s sleeping experience. This record included assessments of thermal sensation using the ASHRAE 7-point thermal sensation scale (i.e., thermal sensation vote (TSV)) [23], perceived air quality on a 5-point scale (i.e., air quality vote), and perceived aural, visual, and overall environmental comfort scored from 0 to 100. The participants also answered dichotomous yes/no questions about their satisfaction with the different environmental aspects and the overall bedroom environment. In addition to environmental sensation and acceptance, self-reported sleep quality was assessed using a modified Pittsburgh Sleep Quality Index (PSQI) survey that evaluates sleep quality, latency, disturbance, ease of awakening, and sleep sufficiency [9,24].

2.3. Predicted Mean Vote (PMV)

The predicted mean vote (PMV) was calculated to estimate the predicted thermal sensation and dissatisfaction based on the measured bedroom environmental conditions and the participants’ bedding systems. The PMV was determined using Fanger’s established model as shown in Equation (1), where M is the metabolic rate, W is the effective mechanical power, p_a is the water vapor partial pressure, t_a is the air temperature, f_cl is the clothing surface area factor, t_cl is the clothing surface temperature, t_r is the mean radiant temperature, and h_c is the convective heat transfer coefficient.

$$\begin{array}{cc}\hfill \mathrm{P}\mathrm{M}\mathrm{V}=[0.303 \times & \exp (-0.036M)+0.028]\hfill \\ & \times \{\left(M-W\right)-3.05\times {10}^{-3}\times \left[5733-6.99\times \left(M-W\right)-p_a\right]\hfill \\ & - 0.42\times \left[\left(M-W\right)-58.15\right]-1.7\times {10}^{-5}\times M\times \left(5867-p_a\right)\hfill \\ & - 0.0014\times M\times \left(34-t_a\right)-3.96\times {10}^{-8}\hfill \\ & - f_{cl}\times \left[{\left(t_{cl}+273\right)}^4-{\left(t_r+273\right)}^4\right]-f_{cl}\times h_c\times \left(t_{cl}-t_a\right)\}\hfill \end{array}$$

(1)

A metabolic rate of 0.7 MET was assumed for a sleeping person, and the clothing insulation values of the bedding systems were estimated based on prior research [25]. For simplicity, the operative temperature was calculated as the average of the measured air temperature and radiant temperature.

2.4. Statistical Analysis

The study collected repeated measures from the same participants under two treatment conditions: one with an air conditioner and the other with a fan. Due to the non-parametric nature of the data, appropriate statistical tests were selected for analysis. The Friedman Test was utilized to examine categorical variables as sample groups relevant to the controlled conditions, including the participant’s ratings on thermal sensation, air quality votes, and sleep quality, all measured on an ordinal scale. The Mann–Whitney U test assessed aural, visual, and overall environmental comfort scores as independent sample groups. The McNemar’s Test was applied to analyze binary environmental satisfaction data from the participant’s yes or no responses. The Wilcoxon Signed-Rank Test was used for continuous data from measurement, such as environmental conditions. These non-parametric tests were considered suitable and more robust due to the characteristics of the data, such as a small sample size and non-parametric variables, making them an appropriate analytical approach for this study.

2.5. Bedroom Cooling Energy Simulation

The building energy simulation program EnergyPlus (version 24.1.0) was used to model the cooling energy consumption in a bedroom setting to estimate the potential annual energy savings in cooling by adjusting the set-point temperature. The bedroom layout was derived from the standardized New Cruciform public housing floorplan provided by the Hong Kong Housing Authority. Notably, all participants in this study resided in public housing units with this specific floor plan.

The bedroom’s dimensions were 3.81 m (L) × 2.05 m (W) × 2.7 m (H), with two internal walls, two external walls, and two windows at different orientations. For simplification, only the bedroom was simulated. The internal walls, the ceiling, and the floor were treated as adiabatic, and the bedroom was fully sealed without any air infiltration. The U-values for the external walls and windows were set at 2.88 W/m²K and 5.66 W/m²K, respectively. Please refer to

Table 1. Building materials adopted for cooling energy simulation.

	Building Materials
Characteristics	Mosaic Tile	Heavyweight Concrete	Gypsum Board	Clear Glass
Location	External wall	External wall	External wall	Windows
Roughness	Medium smooth	Medium rough	Medium rough	N/A
Thickness (m)	0.005	0.235	0.013	0.006
Conductivity (W/mK)	0.87	1.95	0.25	0.9
Density (kg/m3)	2400	2240	800	N/A
Specific heat (J/kgK)	850	900	1090	N/A

for the characteristics of the building materials used in the simulation.

The simulation involved one occupant sleeping inside the bedroom from 23:00 to 07:00 the next day for an entire year. It was assumed that the sleeping individual generated 40 W of heat within the bedroom. Regional climate data collected from 2007 to 2021 were employed to define the outside air properties affecting the bedroom environment. The simulation did not include energy usage in the bedroom outside the designated sleeping hours. The impact on cooling energy usage was assessed by varying the indoor temperature set-points within a comfortable range. In the simulation, the air conditioner operated for the indoor mean air temperature exceeded the set-point. This analysis allows for a systematic evaluation of the trade-off between indoor sleeping thermal comfort and energy efficiency.

3. Results and Discussion

3.1. Experiment Data

Thermal parameters, sleep characteristics, and environmental sensations and acceptances were statistically compared from the two experimental nights. The results are presented in

Table 2. Thermal parameters, sleep characteristics, and environmental sensation and acceptance collected from the two experimental nights.

	Mechanical Fan	Air Conditioner	p-Value 1
Environmental condition
Background noise level (dBA)	42.6	44.0	0.65
Outdoor air temperature (°C)	26.4	26.8	0.19
Indoor operative temperature (°C)	27.2	24.4	<0.01
Indoor relative humidity (%)	72.0	60.1	<0.01
Indoor air velocity (m/s)	0.30	0.20	0.16
Sleep characteristics
Coverage percentage (%)	62.4	85.5	<0.01
Bedding insulation (clo)	1.84	2.80	<0.01
Sleep duration (min)	507	522	0.33
Self-reported sleep quality score	13.1	12.8	0.80
Environmental comfort and acceptance
Predicted mean vote (PMV)	0.75	0.55	0.25
Thermal sensation vote (TSV)	0.60	−0.20	0.02
Thermal acceptance	1	1	1
Air quality vote	4.33	4.33	1
Air quality acceptance	1	1	1
Aural score	80.5	80	0.70
Aural acceptance	1	0.93	0.32
Visual score	84.5	83.3	0.65
Visual acceptance	1	1	1
Overall environmental comfort score	84.5	82.7	0.55
Overall environmental comfort acceptance	1	1	1

¹ p-value < 0.05 are highlighted in bold.

. It was observed that using a mechanical fan for bedroom cooling led to significantly higher operative temperatures and relative humidity than using an air conditioner. Under this cooling condition, the participants tended to opt for thinner bedding covers and a substantially lower coverage percentage, decreasing overall clothing insulation by approximately 1 clo. This adaptive behavior compensated for the thermal environment differences under the two cooling situations in response to a warmer bedroom environment. Consequently, the predicted mean vote (PMV), based on Fanger’s model and the chosen clothing insulation, indicated that the thermal comfort experienced by occupants cooled by either the air conditioner or the fan was equivalent.

Subjectively, the participants reported a slightly higher thermal sensation when cooled by the mechanical fan with statistical significance. On average, they indicated a slightly cool to neutral thermal sensation vote (TSV) when cooled by the air conditioner and a neutral to slightly warm TSV when cooled by the fan. Other environmental aspects and overall environmental comfort were comparable between the two cooling environments. Despite the variation in the sleeping environment due to different cooling devices, the sleep duration and sleep quality were statistically similar.

3.2. Thermal Environment and Sensation

The effect of using a mechanical fan and an air conditioner as cooling devices for sleep in bedroom thermal conditions, sleeping thermal sensation, and sleep quality were evaluated. Given no significant difference between the range and mean of outdoor air temperatures on the two consecutive nights, the operative temperature measured in air-conditioned bedrooms had a narrower interquartile range and a lower median than those measured in bedrooms cooled by a mechanical fan. Similarly, due to the air conditioner’s cooling process, which removes moisture from the air, indoor relative humidity in air-conditioned bedrooms was also lower. Figure 1 shows the indoor thermal conditions of the bedrooms.

Responding to a hotter bedroom environment, the participants selected lower overall bedding insulation for their sleep. Figure 2 illustrates a linear relationship between operative temperature and bedding insulation, suggesting an adaptive behavioral approach to restoring sleeping thermal comfort Other involuntary physiological mechanisms, such as posture adjustment, vasodilation, sweating, and sleep cycles, also play a role in regulating body temperature. The consistent self-reported sleep quality scores observed between the two control methods suggest that there was likely no significant sleep disruption caused by intense physiological response (e.g., excessive sweating).

The plot in Figure 3, comparing predicted mean vote (PMV) against thermal sensation vote (TSV), indicates that Fanger’s thermal comfort model may be inadequate in predicting individuals’ thermal sensation during sleep. However, a moderately well-fitted linear relationship between indoor operative temperature and TSV suggests that operative temperature can reliably indicate perceived thermal sensation among participants.

The observed relationship between operative temperature and thermal sensation vote (TSV) in this study aligns with the findings of a chamber experiment conducted by Lan et al. [24]. Their research evaluated the effects of three different summer air temperatures (23 °C, 26 °C, and 30 °C) on sleeping thermal comfort and sleep quality. The similarity in the relationship between operative temperature and TSV, except for a slight downward shift indicating that individuals perceive colder sensations at the same temperatures, supports the validity of our results.

In contrast, a prior study on sleeping thermal comfort conducted by Tsang et al. [9] in Hong Kong during winter suggested a higher sensitivity to varying operative temperatures, with a TSV range of 22 °C to 24.5 °C. Additionally, their study identified a lower neutral temperature than the 25.3 °C observed in our current study. Notably, as indicated in

Table 3. Summary of indoor operative temperature, clothing value and neutral temperature from literatures.

Reference	Indoor Operative Temperature (°C)	Clothing Value (clo)	Neutral Temperature (°C)
Current study	22.3–28.6	1.24–3.55	25.3
Tsang et al. [9]	20.1–26.5	1.84–4.56	23.1
Lan et al. [24]	23, 26, 30	1.64	26
Kim and Kum [27]	22, 26, 30	Not mentioned	26
Pan et al. [28]	17, 20, 23	4.21	23
Wang et al. [29]	11.6–20.6	3.55	18.3
Wang et al. [26]	16, 18, 20, 22, 26	2.3–3.62	20

, the neutral temperature appears to be influenced by thermal adaptivity behaviors such as adjustments in coverage percentage [26]. Overall, this study indicates that the thermal comfort range, characterized by a TSV between −1 and 1, corresponds to an indoor operative temperature range of 22.2 °C to 28.2 °C. It is noteworthy that none of the participants expressed thermal dissatisfaction throughout the observable range of operative temperature, suggesting that an operative temperature of up to 28.6 °C was still considered thermally acceptable for sleep.

3.3. Demographic Difference

Gender differences in the selection of bedding insulation in various thermal environments were examined, as depicted in Figure 4. Male participants showed a more significant reduction in clothing insulation selection as environments became hotter, achieved mainly by lowering coverage percentage. Conversely, female participants completed a partial reduction in clothing value through thinner quilts and decreased coverage percentage.

To further investigate the correlation between the selection of bedding insulation and demographic characteristics, Pearson product–moment correlation coefficients were calculated and are presented in

Table 4. Pearson correlation analysis of demographic characteristics and the relative change in coverage percentage and clothing value selection.

Demographic Characteristics	Relative Change in Coverage Percentage		Relative Change in Clothing Value
	Pearson Coefficient	p-Value 1	Pearson Coefficient	p-Value 1
Age	0.55	0.03	0.60	0.02
BMI	0.25	0.37	0.23	0.40
Weight	0.29	0.30	0.16	0.58
Height	0.25	0.38	0.06	0.84

¹ p-value < 0.05 are highlighted in bold.

. The analysis did not reveal any significant correlations between demographic factors such as weight, height, and BMI with the change in clothing selection. However, a significant moderately positive correlation was found between age and the relative shift in coverage percentage and clothing value. This finding contradicts existing literature on the decreasing sensitivity to changing thermal parameters with age during waking hours [30]. In addition, previous studies have indicated a decrease in axillary temperatures upon retiring and waking with age [31]. The positive correlation between age and the relative change in clothing insulation selection could potentially be attributed to the significant moderately positive correlation between age and BMI (Pearson coefficient: 0.63; p-value: 0.01). It has been previously suggested that a higher BMI may lead to heightened sensitivity to elevated temperatures [32].

While BMI is calculated from weight and height, the basal metabolic rate (BMR) takes into account additional factors such as age, gender, weight, and height. In order to assess the collective impact of these demographic characteristics on thermal comfort and sensation, BMR was calculated for the participants using a previously established equation [33]. Pearson correlation analysis revealed a moderate negative relationship between BMR and clothing value (Pearson coefficient: −0.40; p-value: 0.03). This indicates that participants with a higher BMR tended to select lower clothing value for sleep. However, no significant relationships were observed between BMI, other demographic characteristics, and clothing value selection. These findings suggest that BMR may offer a more comprehensive assessment of the interaction between these characteristics when selecting bedding insulation.

3.4. Sleeping Thermal Environment and Sleep Quality

The self-reported sleep quality under two different bedroom cooling conditions was assessed. Overall, the participants reported slightly better scores for sleep latency (Q1), thermal comfort (Q6 and Q7), sleep quality (Q2–5, Q8, Q9, and Q13), and ease of awakening (Q10) when the bedroom was cooled by a mechanical fan, which had a higher bedroom temperature than those cooled by an air conditioner. Conversely, sleep sufficiency slightly improved when using an air conditioner (Q11 and Q12).

Despite these observed variations, as indicated in

Table 5. Statistical summary of self-reported sleep quality.

	Average Cumulative Score
	Sleep Latency	Thermal Comfort	Sleep Quality	Sleep Sufficiency	Ease of Awakening
Environmental condition
Mechanical fan	0.87	2	7.6	1.8	0.87
Air conditioner	0.8	1.93	7.3	1.9	0.8
p-value	0.8	0.8	0.3	0.61	0.8
Groups of thermal sensation
TSV = 0	0.85	1.95	7.35	1.8	0.85
TSV ≠ 0	0.8	2	7.7	2	0.8
p-value	0.84	0.84	0.36	0.39	0.84
Population with a change in thermal sensation in the two thermal conditions
Cooler sensation	0.78	1.9	7.44	1.89	0.67
Warmer sensation	0.89	2	7.89	1.89	0.78
p-value	0.74	0.74	0.18	1	0.74

, the pairwise Friedman Test did not reveal any statistically significant differences in sleep quality between the two thermal conditions. The Mann–Whitney U Test also suggested no significant difference in sleep quality between the group with a neutral TSV and the other TSV categories.

For participants who experienced a change in thermal sensation, often from a neutral (0) sensation to a warm (+1) sensation, improvements in sleep latency, thermal comfort, sleep quality, and ease of awakening were observed. Although the Friedman Test did not indicate a significant change in self-reported sleep quality, these findings highlight the positive impact of the warm thermal sensation on various aspects of sleep.

It is noteworthy that the results should be interpreted with caution, as this study involved a sample size of 15 participants (N = 15) providing self-reported data, which may only reflect the responses of a specific population. In addition, the research primarily focused on temperature and ventilation control methods, while humidity and air quality were not addressed in the analysis.

3.5. Annual Energy-Saving Potential in Bedrooms

Regional climate data from Hong Kong indicates that night-time air temperatures can reach 26 °C in April and 30 °C in June, with peaks exceeding 32 °C at night. Throughout the summer, ambient temperatures frequently exceed 26 °C for most nights, posing challenges in creating a comfortable sleeping environment.

While air conditioning effectively cools bedrooms, suboptimal set-point temperatures can compromise thermal comfort, lead to energy wastage, and potentially disrupt sleep and impact sleep quality [10,11,12,13].

This study suggests that a thermal range of 22.2 °C to 28.2 °C is deemed acceptable for sleep, with slightly improved sleep quality at higher temperatures. Figure 5a illustrates the number of days requiring air conditioning for bedroom cooling. Simulation results indicate that about 65% of nights necessitate cooling to sustain a 22 °C bedroom temperature. Adjusting the set-point to 25 °C reduces the need for cooling to half, with only around 23% of nights requiring cooling at 28 °C.

Figure 5b shows the cooling energy consumption of the bedroom at various set-point temperatures. Raising the set-point temperature from 22 °C to 23 °C saves approximately 109 kWh of energy annually. The energy-saving potential diminishes as the set-point rises, with setting the temperature at 25 °C resulting in a 62% reduction in energy consumption compared to 22 °C and a further increase to 28 °C consuming only 5% of the energy at 22 °C. Maintaining a lower indoor temperature requires a significant increase in cooling energy consumption due to prolonged operational periods throughout the year, as well as the substantial temperature differential between indoor and outdoor air during hot summer nights.

According to the simulation results, a notable decrease in cooling energy consumption can be efficiently attained by increasing the bedroom set-point temperature. Conversely, alternative ventilation methods, such as mechanical fans, could enhance thermal comfort and sensation while reducing overall cooling energy usage [34]. These energy savings are particularly pronounced in larger apartments with high cooling demands. In temperate climates, where cooler nights are more frequent, acknowledging the adaptive thermal range could help optimize cooling energy use during summer nights, minimizing excessive consumption.

For sustaining the urban environment, reducing air conditioning use could mitigate the increase in urban heat and cooling degree days (CDD), thereby discouraging excessive cooling consumption in buildings and breaking the feedback loop [35]. Efforts should be focused on assisting residents in decreasing their reliance on air conditioning to alleviate urban thermal effects. A potential challenge in realizing energy-saving potential includes habitual behavior in residential settings, where individuals rely on their usual air conditioning settings to ensure thermal comfort, often overlooking the excessive energy consumption involved. Cultural influences also play a role, as people may prioritize immediate thermal comfort without sufficient awareness of the environmental impact of their cooling choices. Providing indications of electricity consumption and associated carbon emissions from air conditioning could enhance awareness and encourage more sustainable practices. This study aims to motivate the adoption of a sustainable approach to air conditioning operation, optimizing energy efficiency while effectively meeting residential cooling needs.

4. Conclusions

Sleeping thermal comfort is a significant research area that demands attention, given that residential cooling energy consumption is primarily attributed to air conditioning during sleep in subtropical regions. Through a within-subject analysis, this research evaluated sleeping environmental comfort, acceptance, and self-reported sleep quality of 15 participants (N = 15) in air-conditioned and mixed-mode ventilated bedrooms during summer. This study revealed a broad spectrum of thermal comfort temperatures ranging from 22.2 °C to 28.2 °C among the study subjects. In the sleeping environment of higher indoor temperatures, subjects adjusted to the conditions by opting for lighter clothing and bedding insulation. Notably, no significant disparities were observed in sleep quality, thermal, or environmental acceptance between the two ventilated environments.

EnergyPlus simulations showed that a maximum of 95% of the energy consumption could be reduced by increasing the set-point from 22 °C to 28 °C, indicating the energy-saving potential achievable by adopting higher set-point temperatures in the bedroom within this acceptable thermal comfort range.

This study identifies the substantial potential for energy savings in residential buildings during sleep, a subject often underestimated. By adopting an adaptive behavioral approach, this study’s findings suggest that it is feasible to maintain thermal comfort within acceptable ranges while significantly reducing reliance on air conditioning. This adaptive strategy enhances energy efficiency and promotes sustainable and comfortable living environments. Objective physiological measurement could be incorporated in future research to examine in-depth the correlation between behavioral adaptation and temperature variations during sleep.