Research on the multi-objective optimization of energy consumption and indoor environment: a case study of residential structures in hot-summer and cold-winter regions

Balancing the relationship between building energy consumption and the health performance of the indoor environment has emerged as a crucial scientific issue for the sustainable development of residential buildings. Drawing upon existing regulations and standards, this paper employs the building energy use intensity (EUI) as an indicator of energy consumption and systematically selects four key parameters, namely, the thermal - humidity environment, light environment, sound environment, and air quality, as evaluation indices for indoor health and comfort. A comprehensive quantitative evaluation method for the health performance of the indoor environment is developed. Taking a residential building in Nanjing as a case study, a multi - objective optimization model is established based on the energy consumption and health performance of the indoor environment to attain the dual objectives of minimizing the EUI and maximizing the level of health and comfort. The optimization results indicate that the building EUI index is notably reduced by 1.96%–6.41%, and the duration of indoor environmental health and comfort is extended by 13–117 h. Based on these results, optimal solutions highlighting different optimization objectives can be selected for diverse groups of people with varying requirements.

1 Introduction

China stands as a major consumer of energy and emitter of carbon. In 2022, the nation’s primary energy consumption reached 159.39 EJ, accounting for 26.4% of the global total. Concurrently, this energy consumption generated 10.55 billion tons of carbon dioxide, approximately 30% of the world’s total. The construction sector has sustained high levels of energy consumption and carbon emissions over an extended period. In 2020, the total energy consumption throughout the entire construction process in China amounted to 2.27 billion tons of standard coal, accounting for 45.5% of the national total energy consumption. Moreover, the total carbon emissions during this process reached 5.08 billion tons of carbon dioxide, accounting for 50.9% of the national total (China Association of Building Energy Efficiency et al., 2023). Simultaneously, there is a growing emphasis on the quality of life, particularly the health and comfort of the indoor environment (Liu, 2021). This is especially relevant for the elderly and children, who spend a significant amount of time within residential buildings. The indoor environment exerts a long - term influence on their health and comfort. Nevertheless, the enhancement of building health and comfort should not be achieved at the expense of high energy consumption. Therefore, it is imperative to balance the contradiction between the two. Consequently, coordinating the relationship between the energy consumption and the health performance of residential buildings to meet multi - objective requirements is a research - worthy problem.

In recent years, scholars have conducted extensive investigations into the multi-objective optimization of building performance from three perspectives: optimization parameters, optimization objectives, and optimization methods. Concerning optimization parameters, passive energy conservation remains the primary approach for building energy-efficiency measures. Zhou and Hu (2025) analyzed the influence of building orientation, as well as the materials and thickness of the building envelope, on building energy consumption and indoor thermal comfort. They significantly reduced building energy consumption and the discomfort index by optimizing design parameters within a certain range. Liu et al. (2018) analyzed the effects of spatial design factors, including floor height, plane aspect ratio, standard floor area, and horizontal shading length, on building energy consumption. They proposed an energy-saving design scheme for high-rise office buildings in Tianjin. Regarding optimization objectives, research primarily focuses on reducing energy consumption while considering environmental and economic costs as the main goals. Yu et al. (2016) developed a dual-objective optimization model for residential buildings with the goals of reducing annual cooling and heating loads and increasing the proportion of indoor thermal comfort time. This achieved a 47.74% reduction in annual loads and a 3.94% increase in indoor thermal comfort hours. Wang et al. (2022) optimized the design parameters of orientation and shading systems of classrooms in the Nanchang area, with building energy consumption and indoor lighting environment as the optimization targets. The optimized composite shading system can effectively improve the indoor lighting environment and significantly reduce building energy consumption in different orientations. Maiques et al. (2025) aimed to optimize indoor carbon dioxide concentration and ensured indoor air quality by optimizing building orientation and ventilation strategies. Finally, regarding optimization methods, some studies employed the multi-objective optimization plugins Wallacei or Octopus on the Grasshopper platform, while others combined machine learning with optimization algorithms. Based on the Wallacei plugin, Chen (2022) optimized window parameters for residential buildings, targeting operational energy consumption, lifecycle carbon emissions, and indoor health performance. Zhou (2017) optimized the shape, exterior windows, and atriums of office buildings in cold regions with the aim of optimizing natural lighting performance based on the Octopus plugin. Asadi et al. (2014) optimized the envelope material type, solar collector type, and HVAC system parameters of a school building in Portugal through a combination of artificial neural networks and genetic algorithms, with the goals of reducing building energy consumption, renovation cost, and improving indoor thermal comfort.

Existing research on the multi-objective optimization of building performance is often confined to the combination of energy consumption and a single environmental objective, with relatively few studies addressing the comprehensive performance of indoor environments. In this paper, a quantitative evaluation system that conforms to current Chinese standards and represents the comprehensive health performance of indoor environments is established. Based on a residential building in Nanjing, this study develops a building performance optimization model that reduces energy consumption while enhancing the health and comfort index of the indoor environment. It also derives the optimal envelope design solutions for residential buildings in hot-summer and cold-winter regions to meet the needs of different population groups.

2 Evaluation of multi-objective optimization

2.1 Objectives of optimization

The research has chosen energy consumption as a pivotal objective for optimizing building performance, with a specific emphasis on the energy utilized during the operational and usage phases of buildings, encompassing energy consumption for cooling, heating, HVAC systems, lighting, and equipment operation. To mitigate the impact of climatic regional disparities, variations in building types, and differences in building scales on total energy consumption, the study employs energy use intensity (EUI) per unit built-area as the metric for assessment. In green building assessment frameworks, such as the UK’s BREEAM (Building Research Establishment, 2025) and the US’s LEED (U.S. Green Building Council, 2025), EUI is defined as the ratio of a building’s annual final energy consumption to its built-area (or enclosed space volume). Consequently, this study utilizes EUI as a quantitative measure to represent building energy consumption per unit area and establishes the reduction of EUI as a primary optimization goal.

The indoor environment of buildings constitutes a multidimensional integrated system. As outlined in the T/CECS 462-2017 Evaluating Standard for Healthy Housing, the evaluation system for healthy residential buildings comprises six principal indicators, including spatial comfort, air freshness, acoustic environment, lighting quality, and others. Within the spatial comfort indicator, subcategories such as the indoor thermal-humidity environment are included. An examination of existing evaluation standards indicates that various standards consistently recognize the thermal-humidity environment, light environment, acoustic environment, and air quality as core indicators for evaluating indoor environmental performance, albeit with differing emphases. Therefore, this study selects the thermal-humidity environment, light environment, acoustic environment, and air quality as the principal indicators for assessing comprehensive indoor environmental performance. Additionally, the study systematically identifies representative physical parameters corresponding to each principal indicator to serve as secondary evaluation indicators. Details are as follows:

The indoor thermal-humidity environment is a distinct microclimate, shaped by the combined effects of building envelopes and temperature regulation devices such as air conditioning (Lu, 2007). The predicted mean vote (PMV), an internationally recognized quantitative evaluation index, integrates four objective parameters: air temperature, relative humidity, mean radiant temperature, and air flow rate (Fanger, 1970). According to prevailing standards, including GB 50176-2016 Code for Thermal Design of Civil Buildings, the secondary indicators related to the indoor thermal-humidity environment and their health and comfort ranges are presented in Table 1.

Table 1

Table 1. Primary and secondary indicators of indoor environment and respective health and comfort ranges.

The indoor light environment, influenced by both natural daylighting and artificial lighting systems, constitutes a physiological and psychological environment affected by spatial geometry. It encompasses illumination intensity, visual impacts, and color rendering effects. Based on GB 50034-2013 Standard for Lighting Design of Buildings (Ministry of Housing and Urban-Rural Development of the People’s Republic of China, 2013) and related research (Le-Thanh et al., 2022; Mardaljevic et al., 2012), the secondary indicators for the light environment with their health and comfort ranges are identified in Table 1.

The indoor acoustic environment is a comprehensive environment formed by the interaction of various acoustic elements within a building’s interior space. For residential buildings, the primary concern is to prevent noise interference. According to current standards such as GB 50118-2010 Code for Design of Sound Insulation of Civil Buildings (Ministry of Housing and Urban-Rural Development of the People’s Republic of China, 2010), the secondary indicators related to the indoor acoustic environment and their health and comfort ranges are shown in Table 1.

Indoor air quality encompasses physical, chemical, biological, and radioactive indicators pertinent to human health and comfort. The study concentrates on chemical indicators such as concentrations of indoor air pollutants, including CO2, PM2.5, PM10, and TVOC. In accordance with the T/CECS 462-2017 Evaluation Criteria for Healthy Residences, the secondary indicators related to indoor air quality and their health and comfort ranges are presented in Table 1.

The proportion of hours within the specified healthy and comfortable range for each secondary indicator, out of the annual total of 8,760 h, is designated as the Health and Comfort Index (Q) for that particular secondary indicator. In alignment with the weighting distribution outlined in the evaluation criteria for healthy residences (China Association for Engineering Construction Standardization, 2017), this study omits two assessment subitems related to water quality hygiene and health promotion. The weights for the remaining four assessment subitems—space comfort, adequate lighting, quiet environment, and fresh air—are determined through linear interpolation. These weights are subsequently allocated as the primary indicator evaluation weights, as detailed in Table 2. Secondary indicator evaluation weights are distributed equally.

Table 2

Table 2. Weights of primary indicators.

In summary, the formula for calculating the health and comfort index of a building’s indoor environment is expressed in Equation 1:

$Q={\omega }_1{Q}_{\text{thermal}\_\text{humidity}}+{\omega }_2{Q}_{\text{light}}+{\omega }_3{Q}_{\text{sound}}+{\omega }_4{Q}_{air}(1)$

Where:

$Q_{\text{thermal}\_\text{humidity}}=1/4\left(Q_{\mathrm{T}}+Q_{RH}+Q_{\text{MRT}}+Q_{\mathrm{V}}\right)(2)$

$Q_{\text{light}}=1/3\left(Q_{\text{UDI}}+Q_{\text{GA}}+Q_{\text{RA}}\right)(3)$

$Q_{\text{sound}}=Q_{\text{NL}}(4)$

$Q_{\text{air}}=1/4\left(Q_{{CO}_2}+Q_{{PM}_{2.5}}+Q_{{PM}_{10}}+Q_{TVOC}\right)(5)$

Where ω_1 to ω_4 represent the evaluation weights for the thermal-humidity environment, light environment, sound environment, and air quality, respectively, as shown in Table 2. Q in Equations 2–5 signifies the aggregate health and comfort index across all indicators.

2.2 Optimization parameters

The roof and exterior walls play a crucial role in regulating the heat exchange between a building and its external surroundings environment, thereby influencing both energy consumption and indoor environmental conditions. External windows are a critical part of the building envelope and also the most active area for heat exchange and the weakest point in terms of insulation. Energy loss through windows is approximately four times that of walls, five times that of roofs, and twenty times that of floors, accounting for about 40%–50% of the total energy loss through the building envelope. Moreover, the type and design of windows significantly impact indoor environmental performance, including thermal-humidity and lighting conditions. In light view of the optimization objectives, the heat transfer coefficients of the roof and exterior walls, exterior window types, window to wall ratios, and windowsill heights are selected as optimization parameters.

Based on project requirements and in compliance accordance with the DB 32/4066-2021 Design Standard for Thermal Environment and Energy Conservation of Residential Buildings in Jiangsu Province (Jiangsu Provincial Department of Housing and Urban-Rural Development, 2021), three thicknesses of extruded polystyrene (XPS) boards—, namely, 80 mm, 90 mm, and 100 mm—, are chosen as roof insulation layers. This resulting results in corresponding roof heat transfer coefficients of 0.42 W/(m²·K), 0.38 W/(m²·K), and 0.34 W/(m²·K), which define the optimization range for the roof heat transfer coefficient. Three types of rock wool boards with thicknesses of 30 mm, 40 mm, and 50 mm are selected as the exterior wall insulation layer, yielding leading to corresponding exterior wall heat transfer coefficients of 0.72 W/(m²·K), 0.68 W/(m²·K), and 0.64 W/(m²·K), which establish the optimization range for the exterior wall heat transfer coefficient. Ten representative types of exterior windows from Jiangsu’s commonly used window database and the national standard window database, as illustrated shown in Supplementary Appendix Table 1, are selected as the optimization range for exterior window types. In the table, K, SHGC, and VT denote represent the heat transfer coefficient, solar heat gain coefficient, and visible light transmittance, respectively. The optimization range for the north-facing window-to-wall ratio (WWR) is set between 0. The optimization range for the north - facing window - to - wall ratio (WWR) is set from 0.20 and to 0.50, with increments of 0.50, with an increment of 0.05, while the south - facing WWR ranges from 0.30 to 0.60 with the same increment. The optimal range for windowsill height is set between from 0.4 m and to 1.2 m, with increments of 0.2 m, with an increment of 0.1 m. These parameters are presented in detailed in Table 3.

Table 3

Table 3. Design parameters and respective value ranges.

3 Building performances

3.1 Model description

Nanjing, situated at 32.00°N latitude and 118.80°E longitude, falls within China’s hot summer and cold winter zone A. The city experiences an average temperature of 3.1 °C in its coldest month and 28.3 °C in its warmest month. The annual heating degree days (HDD) amount to 1,775 °C·d, and the cooling degree days (CDD) total 176 °C·d. In accordance with the Chinese code for thermal design of civil buildings (Ministry of Housing and Urban-Rural Development of the People’s Republic of China, 2016), the thermal design of buildings in Nanjing must satisfy summer cooling requirements and adequately address winter insulation. This investigation is grounded in a residential project in Nanjing that has been awarded both green three-star and healthy three-star certifications. Building No. 1 from this project has been chosen as the subject of this study. The building, which rises to 18 stories above ground, houses commercial spaces on the first and second floors with respective heights of 4.2 m and 3.6 m. Standard floors, spanning from the fifth to the eighteenth story, each have a floor height of 2.97 m. The architectural rendering is depicted in Figure 1.

Figure 1

Figure 1. Architectural rendering of the residential project.

The study’s computational model was simplified by adopting a representative standard floor as the basis for simulation, with the corresponding Revit model depicted in Figure 2a. The middle level, avoiding top-floor heat loss and ground-floor cold air infiltration effects, was selected as the standard floor, making it more representative. Moreover, due to the north-south symmetry of the building, simulating only 2 south-facing and 2 north-facing units is sufficient to cover all apartment types. Utilizing this Revit model, an enclosed volume was constructed on the Rhino platform to symbolize the residential interior space. This volume was subsequently subdivided into functional zones, encompassing living rooms, bedrooms, studies, kitchens, bathrooms, and stairwells, as illustrated in Figure 2b. Parametric modeling of the enclosed volume was performed using the Grasshopper plugin to produce wall configurations. Exterior walls were outfitted with windows in accordance with the window-to-wall ratio. External window shading devices and doors were subsequently integrated. All elements were then assembled to form a comprehensive Honeybee analysis model.

Figure 2

Figure 2. Single-story computational model. (a) Revit model. (b) Rhino model.

Upon conducting a detailed project analysis and consulting the GB 50176-2016 Code for the thermal design of civil buildings (Ministry of Housing and Urban-Rural Development of the People’s Republic of China, 2016), the compilation of design parameters for the building envelope was undertaken and subsequently summarized in Table 4. These parameters constitute the foundational values for the multi-objective optimization process.

Table 4

Table 4. Design parameters and values of building envelope.

The heat transfer coefficients for the roof and exterior walls were determined by their relationship with thermal resistance R. These opaque structural elements, in conjunction with transparent windows, which are characterized by their heat transfer coefficient, solar heat gain coefficient, and visible light transmittance, were assimilated into a comprehensive construction library. This library, which houses all building envelope design parameters, was subsequently integrated into the Honeybee energy model. Ultimately, a comprehensive construction library, incorporating all envelope design parameters, was compiled and imported into the Honeybee simulation model. In accordance with the DB 32/4066-2021 design standard for thermo-environment and energy conservation in residential buildings, internal heat gains, including those from occupancy, HVAC systems, lighting, and equipment, were configured for various room types. The operational temperature thresholds for the HVAC system were established as follows: the heating set-point is set at 12 °C, and the cooling set-point is set at 28 °C.

3.2 Energy consumption simulation

The aforementioned model is integrated into the HB Annual Loads module to calculate residential building loads, resulting in a total building load intensity of 59.19 kWh/m². This encompasses a cooling load of 22.39 kWh/m², a heating load of 13.50 kWh/m², a lighting load of 10.69 kWh/m², and an equipment load of 12.62 kWh/m². Figure 3 illustrates the monthly disaggregated load distribution, indicating that cooling and heating loads account for more than 60% of the total annual building loads. Subsequently, the HB Model to Osm module is employed for energy consumption simulation, yielding an Energy Use Intensity (EUI) of 59.452 kWh/m². This includes cooling energy of 21.56 kWh/m², heating energy of 13.93 kWh/m², lighting energy of 10.69 kWh/m², and equipment energy of 12.61 kWh/m². These figures exhibit a close correlation with the disaggregated load distribution patterns, thereby affirming the model’s consistency.

Figure 3

Figure 3. Distribution of monthly disaggregated loads.

3.3 Indoor environmental simulation

Upon examining the functional attributes and practicality of the Grasshopper platform, the utilization of the Honeybee Energy and Honeybee Radiance plugins is employed to simulate and analyze the indoor thermal-humidity and lighting environments of architectural structures, respectively. The outcomes from these two domains are amalgamated to facilitate a holistic evaluation of the indoor environment’s performance. The indoor environmental health and comfort index Q serves as a quantifiable metric for indoor environmental objectives.

The annual hourly data for air temperature, mean radiant temperature, and relative humidity in each room are obtained. Throughout the year, the indoor air temperature is maintained within the range of 18 °C–26 °C for 3020 h, constituting 33.19% of the total 8,760 h in a year; the indoor mean radiant temperature is within the range of 20 °C–28 °C for 3483 h, representing 39.04% of the year; and the indoor relative humidity is within the range of 30%–60% for 4296 h, accounting for 44.80% of the year. Utilizing the comprehensive performance evaluation method for indoor environmental quality as outlined above, the indoor thermal-humidity health and comfort index Q is calculated to be 39.01%.

In the simulation of the indoor lighting environment, daylight analysis points are generated with a grid subdivision precision of 0.5 for the purpose of calculating the Useful Daylight Illuminance (UDI), with the measurement plane positioned 0.8 m above the floor level. Figure 4 illustrates the UDI outcomes of the building. By calculating the average UDI values across all rooms, a mean UDI of 72.19% is attained. In the Glare Autonomy (GA) assessment, the analysis points are generated with a grid subdivision precision of 1, with the measurement plane elevated 1.2 m from the floor. Figure 5 presents the glare autonomy results. The outcomes suggest that most areas experience minimal glare, with occasional occurrences near south-facing windows. The average GA across all spaces is 95.33%. By employing the aforementioned comprehensive indoor environmental performance evaluation methodology, the building’s lighting environment health and comfort index is determined to be 83.76%. Following a comprehensive assessment, the integrated healthy comfort index for the indoor light and thermal-humidity environment is established at 59.05%.

Figure 4

Figure 4. Effective daylight illuminance.

Figure 5

Figure 5. Visualization of glare autonomy.

Based on the aforementioned simulation analysis, the Energy Use Intensity (EUI), the indoor environmental health and comfort index Q, the thermal-humidity environmental health and comfort index, $Q_{\text{thermal}\_\text{humidity}}$ the light environmental health and comfort index$Q_{light}$, and the health and comfort indices for each secondary indicator are shown in Table 5.

Table 5

Table 5. Consumption of residential energy and the index of indoor health and comfort.

4 Analysis of multi-objective optimization

4.1 Optimization results

The Wallacei plugin utilizes the NSGA-II genetic algorithm to conduct multi-objective optimization for the model. The optimization targets are the minimization of the building’s Energy Use Intensity (EUI) and the maximization of the indoor health and comfort index Q. The population size is set to 30, with the number of generations set to 40, while the genetic algorithm parameters, including crossover and mutation rates, are maintained at their default settings. Following the execution of 30 × 40 generations of genetic algorithm optimization, a total of 18 Pareto optimal solutions have been ascertained, as depicted in Figure 6. The EUI, indoor environmental health and comfort index Q, $Q_{\text{thermal}\_\text{humidity}}{Q}_{light}$ along with the health and comfort indices of each secondary indicator, and the corresponding architectural design parameter schemes for each optimal solution are elaborated in Supplementary Appendix Tables 2, 3.

Figure 6

Figure 6. Pareto optimal solutions.

Figure 7 illustrates the optimized ranges for EUI, Q, $Q_{\text{thermal}\_\text{humidity}}$ and $Q_{light}$. In terms of energy consumption, all optimized EUI values are lower than the baseline figures, with a minimum of 55.981 kWh/m² and a maximum of 58.284 kWh/m². Compared to the baseline, the EUI values exhibit a reduction ranging from 1.168 to 3.471 kWh/m², which constitutes 1.96%–6.41% of the total energy consumption. With respect to the indoor environment, 15 optimized cases surpass the baseline in terms of indoor health and comfort Q values, while 3 cases fall below, with Q values ranging from 57.87% to 59.89%. The indoor health and comfort Q has improved by 0.15%–1.34%, indicating an increase in the annual time spent within the health and comfort range by 13–117 h. The $Q_{\text{thermal}\_\text{humidity}}$ has improved by up to 40.03%, equating to an additional 89 h annually. The $Q_{light}$ has reached up to 86.60%, corresponding to 215 additional hours annually. Notably, the GA index exhibits only minor improvements, whereas the UDI index demonstrates a substantial enhancement compared to the baseline, reaching up to 76.76%, which translates to an annual increase of approximately 410 h.

Figure 7

Figure 7. Optimized indices. (a) EUI. (b) Indoor environment health and comfort index Q. (c) $Q_{\text{thermal}\_\text{humidity}}.$ (d) $Q_{light}$.

Improving the comfort of indoor environments often necessitates greater energy consumption to sustain optimal levels of temperature, humidity, lighting, and air quality. Through the co-optimization of energy efficiency and indoor environmental objectives, tailored solutions are proposed, taking into account varying optimization priorities. When energy use intensity (EUI) is established as the primary optimization criterion, a comparative analysis of 18 Pareto optimal solutions uncovers distinct performance attributes. Solution 13 exhibits exceptional energy conservation, achieving a 6.5% reduction in EUI (from 59.452 to 55.981 kWh/m²). Nonetheless, this improvement is accompanied by a diminishment $Q_{light}$ in relative humidity (from 84.14% to 82.62%) and air quality (from 58.42% to 57.87%). Conversely, solution 3 effects a 5.3% reduction in EUI while concurrently enhancing overall indoor air quality from 58.42% to 58.57%, along with minor improvements in both $Q_{light}$ and $Q_{\text{thermal}\_\text{humidity}}$. These results indicate that solution 13 is the optimal selection for achieving maximum energy savings, whereas solution 3 demonstrates superior comprehensive performance when balancing energy efficiency with indoor environmental quality. Figures 8, 9 offer comparative visual representations of these optimization outcomes, furnishing empirical evidence to support differentiated decision-making in accordance with specific project requirements.

Figure 8

Figure 8. Comparison of the alterations in the environmental utilization index and the health and comfort index (Q) for solution 13, pre- and post-optimization. (a) EUI. (b) Indoor environment health and comfort index Q.

Figure 9

Figure 9. Comparison of the alterations in the environmental utilization index and the health and comfort index (Q) for solution 3, pre- and post-optimization. (a) EUI. (b) Indoor environment health and comfort index Q.

Upon the primary objective of enhancing the indoor environment, the elimination of impractical solutions reveals that solution 17 optimally reduces energy consumption and simultaneously improves the health and comfort indices of the environment. Specifically, the indoor health and comfort index Q experiences a 2.38% increase, extending the annual duration of comfort by 121 h. In the context of thermal-humidity conditions, the $Q_{\text{thermal}\_\text{humidity}}$ index experiences a 2.62% increase, adding 89.35 h annually to the period where conditions remain within the healthy and comfortable range. Regarding lighting conditions, the $Q_{light}$ index improves by 2.31%, and the effective daylight factor increases by 5.67%, resulting in approximately 357 additional hours annually where the effective daylight factor falls within the healthy and comfortable range. Concurrently, the Energy Use Intensity (EUI) decreases by 2.04%, as depicted in Figure 10. Solution 17 effectively achieves the dual objectives of improving comfort and reducing energy consumption, establishing it as the optimal choice when prioritizing indoor environmental quality.

Figure 10

Figure 10. Comparison of solution 17, pre- and post-optimization. (a) EUI (b) Indoor environment health and comfort index Q (c) $Q_{\text{thermal}\_\text{humidity}}$.

When the primary objective is to enhance indoor thermal-humidity conditions, solution 17 emerges as the optimal strategy. This optimized solution sustains indoor air temperature and mean radiant temperature at levels that are healthy and comfortable, comparable to baseline conditions, while significantly enhancing the relative humidity comfort level. Specifically, the annual duration during which relative humidity remains within the healthy and comfortable range increases by approximately 254 h, as depicted in Figure 10. This solution is particularly advantageous for households with elderly individuals, children, or residents with respiratory conditions, as the extended period of healthy humidity levels aids in reducing respiratory health risks and fosters a safer living environment for these vulnerable groups.

Upon prioritizing the optimization of the indoor lighting environment, Solution 12 emerges as the optimal selection subsequent to the elimination of unsuitable alternatives. The lighting environment level exhibits a substantial enhancement, increasing from 84.14% to 86.30%, which corresponds to an additional 189 h annually wherein the residential lighting conditions persist within the healthy and comfortable spectrum. The glare autonomy index manifests a moderate improvement. However, the effective daylight factor index experiences a notable augmentation, ascending from 72.08% to 76.18%. This is tantamount to approximately 359 additional hours annually that meet the healthy and comfortable standards for the effective daylight factor. Furthermore, the thermal-humidity environment level improves marginally from 39.01% to 39.75%. The solution also accomplishes gains in energy efficiency, with the EUI decreasing from 59.452 kWh/m² to 57.895 kWh/m², as depicted in Figure 11. This solution is particularly apt for user groups with elevated requirements for natural lighting, such as home office workers and students.

Figure 11

Figure 11. Comparison of the alterations in the environmental utilization index and the health and comfort index (Q) for solution 12, pre- and post-optimization. (a) EUI (b) Indoor environment health and comfort index Q.

4.2 Parameter impact analysis

In the context of building envelope design parameters, the optimization process has identified that the optimal roof heat transfer coefficient is 0.34 W/(m²·K), which can be achieved by employing a 100 mm thick extruded polystyrene board for roof insulation. For the exterior walls, the optimal heat transfer coefficient is 0.64 W/(m²·K), attainable through the use of a 50 mm thick rock wool panel for insulation. The optimal window configuration boasts a heat transfer coefficient of 1.5 W/(m²·K), a solar heat gain coefficient of 0.644, and a visible light transmittance of 0.72, specifically realized with aluminum-clad wood casement windows (configured as 5 + 12Ar+5 + 12Ar+5, incorporating warm edge technology). The optimal ranges for key parameters are as follows: a north-facing window-to-wall ratio between 0.25 and 0.50; a south-facing window-to-wall ratio between 0.35 and 0.50; and a windowsill height ranging from 1.0 m to 1.2 m. Collectively, these specifications represent the most efficacious envelope design solutions as discerned through the optimization process.

Figure 12 illustrates the correlation between the thermal transfer coefficients of the roof, exterior walls, and windows with the Energy Use Intensity (EUI) and other pertinent factors $Q_{\text{thermal}\_\text{humidity}}$. The graph demonstrates that lower thermal transfer coefficients correspond with a reduction in building EUI and enhancements in performance $Q_{\text{thermal}\_\text{humidity}}$. This finding further elucidates why the multi-objective optimization process selected a 100 mm thick extruded polystyrene board for roof insulation and a 50 mm thick rock wool panel for exterior wall insulation. Consequently, in the design of roofs and exterior walls, prioritization should be given to thicker insulation panels when economically viable. For windows, the optimal solution involves the use of casement aluminum-clad wood windows with a 5 + 12Ar+5 + 12Ar+5 configuration and warm edge technology, boasting a thermal transfer coefficient of 1.5 W/(m²·K). In window design, whenever budgetary constraints permit, preference should be given to windows with lower thermal transfer coefficients.

Figure 12

Figure 12. Influence of thermal transfer coefficients on energy use intensity $Q_{\text{thermal}\_\text{humidity}}$ (EUI). (a) Roofs. (b) Exterior walls. (c) Windows.

As depicted in Figure 13, the solar heat gain coefficient (SHGC) of windows significantly affects energy consumption and the indoor thermal-humidity environment, thereby influencing the health and comfort levels within. With an increase in SHGC, the building’s energy use intensity (EUI) initially diminishes and subsequently escalates, while the $Q_{\text{thermal}\_\text{humidity}}$ indoor environment progressively enhances. However, when the SHGC attains a value of 0.644, the EUI experiences an unexpected increase rather than a further decline. This anomaly could be ascribed to the impact of additional variables, such as building orientation and shading mechanisms, on the correlation between SHGC and EUI. Consequently, a higher SHGC should be contemplated during the design phase of windows.

Figure 13

Figure 13. Impact of SHGC on EUI and $Q_{\text{thermal}\_\text{humidity}}$.

Table 6 presents the relationship between windowsill height and Energy Use Intensity (EUI), $Q_{\text{thermal}\_\text{humidity}}$, $Q_{\mathrm{U}\text{DI}}$ and $Q_{\text{GA}}$. The data indicates that windowsill height exerts a minimal influence on EUI, $Q_{\text{thermal}\_\text{humidity}}$, and $Q_{\text{GA}}$. Nevertheless, it significantly impacts the $Q_{\mathrm{U}\text{DI}}$, which improves as the windowsill height increases, as depicted in Figure 14. Consequently, within the allowable scope of design standards, appropriately elevating the windowsill height can enhance the Useful Daylight Illuminance (UDI) level.

Table 6

Table 6. EUI and indoor environmental comfort indicators under different windowsill heights.

Figure 14

Figure 14. Influence of windowsill height on $Q_{\mathrm{U}\text{DI}}$.

The visible transmittance (VT) exhibits a primary correlation with illumination $Q_{light}$ levels. In the context of commonly utilized windows in Nanjing, with VT values of 0.62 and 0.72, we preserved other design parameters as constant and conducted calculations $Q_{light}$, as illustrated in Table 7. Comparative analyses reveal that VT has a negligible effect on glare autonomy (GA) levels. To enhance the level of useful daylight illuminance (UDI), it is recommended to select windows with a higher visible light transmittance.

Table 7

Table 7. Indoor $Q_{\text{light}}$ under different visible transmittance of windows.

Impact of south/north-facing WWR on EUI, Q_{thermal_humidity} and Q_UDI is depicted in Figures 15a-c, respectively. Analysis indicates that the north - facing WWR presents a linear correlation with both EUI and $Q_{\text{thermal}\_\text{humidity}}$. Specifically, lower ratios lead to a reduction in EUI and an improvement in $Q_{\text{thermal}\_\text{humidity}}$. In contrast, the south-facing window to wall ratio (WWR) demonstrates a nonlinear relationship with EUI. Optimal energy efficiency is achieved at a WWR of 0.4, where EUI is minimized. Nevertheless, the indoor $Q_{\text{thermal}\_\text{humidity}}$ level shows a positive correlation with the south - facing WWR, with a consistent improvement as the ratio increases.

Figure 15

Figure 15. Impact of south/north-facing window to wall ratio on (a) EUI, (b) Q_{thermal_humidity}, and (c) Q_UDI.

In terms of the lighting environment, the north - facing WWR significantly impacts the useful daylight illuminance (UDI). Higher ratios enhance UDI levels by augmenting the availability of natural daylight, while having a negligible effect on glare autonomy (GA) levels. Conversely, the south - facing WWR has minimal influence on UDI but shows a linear correlation with GA levels. As the WWR increases, the GA level gradually decreases. Therefore, during the design process, the WWR needs to be meticulously optimized to balance the requirements of both the thermal - humidity environment and daylighting performance.

5 Conclusion

This study establishes a quantitative evaluation system to assess the overall performance of indoor building environments. Focusing on the optimization objectives of building energy consumption and the health and comfort of indoor environments, the study employs simulations and multi-objective optimization analyses to evaluate building performance, leading to the following conclusions:

1. Utilizing the standards such as GB/T 50378-2019 “Green Building Evaluation Standard” and T/CECS 462-2017 “Healthy House Evaluation Standard,” this research identifies representative physical parameters associated with four environmental sub-categories: thermal-humidity environment, lighting environment, acoustic environment, and air quality. These parameters act as benchmarks for evaluating the health and comfort levels of a building’s indoor environment. The annual proportion of each indicator within the health and comfort range is defined as its health and comfort index. Consequently, a quantitative evaluation system for the comprehensive performance of a building’s indoor environment is formulated.

2. With objectives centered on energy consumption, indoor environmental health and comfort, indoor thermal-humidity environmental health and comfort, and indoor lighting environmental health and comfort, the study selects building envelope structural design parameters as the primary optimization variables for multi-objective optimization. The aim is to ascertain an optimized design scheme for the envelope structure that achieves reduced energy consumption and heightened health and comfort levels. Utilizing a 30 × 40 genetic algorithm for optimization, 18 Pareto optimal solutions were obtained. Compared to the baseline, the optimized residential energy consumption exhibited a reduction in the EUI value ranging from 1.96% to 6.41%, and the indoor health and comfort level Q increased by 0.15%–1.34%, corresponding to an annual increase of 13–117 h of residential health and comfort.

3. The selection of multi - objective optimization schemes is conducive to making rational decisions on health optimization targets and parameters in accordance with the health requirements and living comfort preferences of the resident population, thus enabling the preferential selection of the optimal solution scheme.

4. Through a quantitative analysis of the relationship between individual design parameters and residential performance objectives, this study identifies the specific influence trends and magnitudes of impact of each design factor on building performance. In the design process of residential buildings in Nanjing, the following specifications should be adopted within the permissible range: roofs and exterior walls with lower heat transfer coefficients, windows with lower heat transfer coefficients coupled with higher solar heat gain coefficients and visible transmittance values, and appropriately increased windowsill heights. For the window-to-wall ratios of south- and north-facing facades, a comprehensive consideration must be made to balance the requirements of both indoor thermal-humidity conditions and daylighting performance.

Data availability statement

The datasets presented in this article are not readily available because No restriction. Requests to access the datasets should be directed to c2phbmRyZXdAMTYzLmNvbQ==.

Author contributions

MZ: Writing – original draft. JS: Writing – review and editing.

Funding

The author(s) declared that financial support was not received for this work and/or its publication.

Conflict of interest

Author MZ was employed by China Railway 22nd Bureau Group Real Estate Development Co., Ltd.

The remaining author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fbuil.2025.1747750/full#supplementary-material

References

Asadi, E., Silva, M. G., Antunes, C. H., Dias, L., and Glicksman, L. (2014). Multi-objective optimization for building retrofit: a model using genetic algorithm and artificial neural network and an application. Energy Build. 81, 444–456. doi:10.1016/j.enbuild.2014.06.009

CrossRef Full Text | Google Scholar

Building Research Establishment (2025). BREEAM new construction. Available online at: https://breeam.com/.

Google Scholar

Chen, J. (2022). Research on multi-objective optimization method of residential window technology aiming at low-carbon and indoor health. Zhejiang province, China: Zhejiang University.

Google Scholar

China Association for Engineering Construction Standardization (2017). Evaluation standard for healthy housing (T/CECS 462-2017). Beijing: China Architecture and Building Press.

Google Scholar

China Association of Building Energy Efficiency, Institute of Urban-Rural Construction and Development, Chongqing University (2023). Research report on building energy consumption and carbon emissions in China (2022). Architecture (02), 57–69.

Google Scholar

Fanger, P. O. (1970). Thermal comfort-analysis and application in environment engineering. Copenhagen: Danish Technology Press.

Google Scholar

Jiangsu Provincial Department of Housing and Urban-Rural Development (2021). Design standard for thermo-environment and energy conservation for residential buildings (DB 32/4066-2021). Nanjing: Jiangsu Science and Technology Press.

Google Scholar

Le-Thanh, L., Nguyen-Thi-Viet, H., Lee, J., and Nguyen-Xuan, H. (2022). Machine learning-based real-time daylight analysis in buildings. J. Build. Eng. 52, 104374. doi:10.1016/j.jobe.2022.104374

CrossRef Full Text | Google Scholar

Liu, J. (2021). The development and current situation of technology residences. Resid. Sci. Technol. 41 (05), 8–23. doi:10.13626/j.cnki.hs.2021.05.002

CrossRef Full Text | Google Scholar

Liu, L., Liu, C., and Wu, D. (2018). Energy saving analysis and optimization of geometric factors of office buildings in tianjin. J. Harbin Inst. Technol. 50 (04), 181–187. doi:10.11918/j.issn.0367-6234.201608092

CrossRef Full Text | Google Scholar

Lu, B. (2007). Study on comprehensive assessment system of indoor environmental quality. J. Environ. Sci. Guide 26 (1), 86–89. doi:10.13623/j.cnki.hkdk.2007.01.027

CrossRef Full Text | Google Scholar

Maiques, M., Tarragona, J., Gangolells, M., and Casals, M. (2025). Energy implications of meeting indoor air quality and thermal comfort standards in mediterranean schools using natural and mechanical ventilation strategies. Energy Build. 328, 115076. doi:10.1016/j.enbuild.2024.115076

CrossRef Full Text | Google Scholar

Mardaljevic, J., Andersen, M., and Roy, N. (2012). “Daylighting metrics: is there a relation between useful daylight illuminance and daylight glare probability?,” in First Building Simulation and Optimization Conference, Loughborough, UK, 189–196.

Google Scholar

Ministry of Housing and Urban-Rural Development of the People's Republic of China (2010). Code for design of sound insulation of civil buildings (GB 2050118-2010). Beijing: China Architecture and Building Press.

Google Scholar

Ministry of Housing and Urban-Rural Development of the People's Republic of China (2013). Standard for lighting design of buildings (GB 50034-2013). Beijing: China Architecture and Building Press.

Google Scholar

Ministry of Housing and Urban-Rural Development of the People's Republic of China (2016). Code for thermal design of civil buildings (GB 50176-2016). Beijing: China Architecture and Building Press.

Google Scholar

U.S. Green Building Council (2025). LEED reference guide for building design and construction. Available online at: https://www.usgbc.org/leed/.

Google Scholar

Wang, Y., Yang, W., and Wang, Q. (2022). Multi-objective parametric optimization of the composite external shading for the classroom based on lighting, energy consumption, and visual comfort. Energy Build. 275, 112441. doi:10.1016/j.enbuild.2022.112441

CrossRef Full Text | Google Scholar

Yu, W., Wang, D., and Li, B. (2016). Multi-objective design method of improving the indoor thermal environment with low energy consumption in residential buildings. J. Civ. Archit. Environ. Eng. 38 (04), 13–19. doi:10.11835/j.issn.1674-4764.2016.04.003

CrossRef Full Text | Google Scholar

Zhou, B. (2017). Research on multi-objective optimization of multilayer office building space in cold region based on daylight performance. China: Harbin Institute of Technology, Heilongjiang province.

Google Scholar

Zhou, B., and Hu, M. (2025). Impact of architectural design elements on the energy efficiency and thermal comfort in neighborhood building ensembles. J. Build. Eng. 100, 111708. doi:10.1016/j.jobe.2024.111708

CrossRef Full Text | Google Scholar