Natural Ventilation Technique of uNVeF in Urban Residential Unit Through a Case Study

Abstract

The present study was motivated by the need to enhance indoor air quality and reduce airborne disease transmission in dense urban environments where high-rise residential buildings face challenges in achieving effective natural ventilation. The problem lies in the lack of scalable and convenient tools to optimize natural ventilation rate, particularly in urban settings with varying building heights. To address this, the scientific technique developed with an innovative metric, the urbanized natural ventilation effectiveness factor (uNVeF), integrates regression analysis of wind direction, velocity, air change rate per hour (ACH), window configurations, and building height to quantify ventilation efficiency. By employing a field measurement methodology, the measurements were conducted across 25 window-opening scenarios in a 13.9 m² residential unit on the 35/F of a Hong Kong public housing building, supplemented by the Hellman Exponential Law with a site-specific friction coefficient (0.2907, R² = 0.9232) to estimate the lower floor natural ventilation rate. The results confirm compliance with Hong Kong’s statutory 1.5 ACH requirement (Practice Note for Authorized Persons, Registered Structural Engineers, and Registered Geotechnical Engineers) and achieving a peak ACH at a uNVeF of 0.953 with 75% window opening. The results also revealed that lower floors can maintain 1.5 ACH with adjusted window configurations. Using the Wells–Riley model, the estimation results indicated significant airborne disease infection risk reductions of 96.1% at 35/F and 93.4% at 1/F compared to the 1.5 ACH baseline which demonstrates a strong correlation between ACH, uNVeF and infection risks. The uNVeF framework offers a practical approach to optimize natural ventilation and provides actionable guidelines, together with future research on the scope of validity to refine this technique for residents and developers. The implications in the building industry include setting up sustainable design standards, enhancing public health resilience, supporting policy frameworks for energy-efficient urban planning, and potentially driving innovation in high-rise residential construction and retrofitting globally.

1. Introduction

Natural ventilation is recognized to benefit human health, enhance indoor air quality (IAQ), and reduce the risk of airborne disease transmission, in particular COVID-19 [1,2,3]. In 2015, world leaders committed to attain 17 Sustainable Development Goals (SDGs) and to achieve specified key targets and performance by 2030 [4]. One of the core indeces, SDG 3, focuses on good health and well-being, emphasizing the needs of managing and reducing the risk of health which can be contributed by natural ventilation.

Understanding the metrics of natural ventilation effectiveness is crucial for effectively strengthening the infection risk reduction and supporting sustainable development within urban environments in Hong Kong [5]. However, existing knowledge about how window-opening configurations influence the natural ventilation effectiveness and reduction in airborne disease infection risk is limited. In this research, a case study of a studio flat in Hong Kong was selected to evaluate the ventilation performance across different window-opening scenarios and quantify the COVID-19 infection risk. By integrating uNVeF with empirical equations, this research provides insight for the architects and engineers to optimize the design and planning of new buildings. These findings contribute to the broader field of urban science and global sustainability goals.

Effective ventilation has been studied extensively. A recent bibliometric analysis focusing on thermal comfort and IAQ examined 14,179 case studies from 2020 to 2024 with keywords of residential thermal comfort, outdoor thermal environments, and IAQ’s health impacts [2]. Their findings highlighted the trend and global interest in integrating improved natural ventilation effectiveness techniques to promote healthier indoor environments in high-density urban cities.

In order to improve thermal comfort and IAQ for urban high-rise buildings in Hong Kong, the Hong Kong Building Department set up a local regulation, namely the APP-130 Lighting and Ventilation Requirement for Habitable Room and Kitchen in Domestic Buildings. It is required that habitable rooms and kitchens in domestic buildings should achieve minimum 1.5 ACH by natural means [6]. This standard ensures that newly constructed public housing estates comply with this ventilation requirement. Compliance with local regulations further indicates the importance of natural ventilation in Hong Kong’s building design.

However, achieving this standard in practice is complicated due to multiple influencing factors. Chen et al. (2018) reviewed that natural ventilation is challenging in practical design due to climate, ambient air quality, airflow, wind direction, floor plan, building orientation, urban density, and outdoor environmental conditions [7]. These factors are required to be considered, and the most critical element is the window-opening configuration. The present research addresses this element by systematically evaluating how window-opening configurations impact natural ventilation performance in a high-density urban setting.

In order to assess the effect of window-opening strategies against natural ventilation effectiveness, prior research provides valuable methodologies adaptable to Hong Kong’s residential buildings. Liu & Lee (2020) have conducted walk-through surveys and site measurements to identify the optimum window-opening degree on the natural ventilation performance of residential buildings in Hong Kong. The results show that the optimal range for the window-opening degree is between 60% and 90% [8]. Also, their methodology provided an equation for calculating the effective window-opening area is applicable to present studies.

Recent studies have discovered that ACH is commonly used to quantify the effectiveness of indoor natural ventilation. Various works for studying natural ventilation focused on calculating the ACH have been demonstrated over the past year. Zhou et al. (2021) summarized different empirical models to calculate the air flow rate driven by wind-driven natural ventilation and commented that these models are popular and have been proven effective to a certain extent [9].

Choi & Song (2020) studied three indoor ventilation rate measurement methods. They are inlet air velocity-based, pressure difference-based, and reference wind velocity-based. They concluded that inlet air velocity-based measurement of the ventilation rate is shown to be convenient and simple for experimental use [10]. Sawachi et al. (2004) also conducted experiments using an anemometer by inlet air velocity method to determine the air flow at window opening and ventilation rate under cross-ventilation [11].

Practically, the wind speed at the higher floor level of the building is usually stronger than the lower floor because of the pressure difference between warm air and cold air. In order to verify the compliance of natural ventilation at the lower floor level, Wu et al. (2022) suggested using Hellman Exponential Law Equation to determine the wind velocity at the desired height. The equation correlates two wind velocities and their measurement altitudes with an exponential function of the friction coefficient [12]. The frictional coefficient varies by location, function of land, and different landscape type. In the present research, regression analysis is used to estimate a unique frictional coefficient for a particular façade surface of the case study building.

Additionally, natural ventilation can influence the control of airborne diseases. There are numerous types of airborne diseases that emerged in the past, and it is anticipated that new diseases and variants will arise in the future. A scientific method is needed to estimate the infection risk and identify how natural ventilation can minimize the risk. In public housing estates where infected occupants may host guests, increasing the natural ventilation rate can significantly lower the risk of cross-infection. Escombe et al. (2007) and Dai et al. (2023) revealed that the Wells–Riley model can be used to predict the probability of infection in indoor spaces and can further estimate the reduction in infection risk achieved through the improved natural ventilation rate [13,14]. Several studies provided data and methodologies to estimate the risk of airborne disease transmission. Guo et al. (2021) reported that the Wells–Riley equation is based on the concept of a ‘quantum of infection’. The probability of a susceptible person being infected by a quantum within a space is expressed as the exponential relationship between the parameters of number of infectors, quanta generation rate, pulmonary ventilation rate, exposure time, and air ventilation rate [15].

To apply the Wells–Riley model, experimental data for pulmonary ventilation and quanta generation rates are essential in order to complete the Wells–Riley equation. The outbreak of new diseases, in particular COVID-19, has promoted extensive research that enable easy collection of relevant data. Xia et al. (2020) reported the tidal volume of Chinese patients suffering from COVID-19 [16], and the International Commission on Radiological Protection (1994) provided comprehensive figures regarding the average weight of humans based on different ages, genders, and ethnic groups [17]. Liu et al. (2020) reported the breathing rate of patients with COVID-19 in China [18]. Also, Buonanno et al. (2020) and Shen et al. (2021) provided backward calculation values of quanta generation rate for COVID-19 [19,20]. These data facilitated the calculation of COVID-19 infection risk using the Well–Riley model.

Moreover, technology transfer (TT) is critical for accelerating development in promoting natural ventilation design strategies, distributing natural ventilation measurement experience, and creating positive influence in sustainability to enhance indoor thermal comfort and IAQ in urban societies. Craiut et al. (2022) reviewed global TT mechanisms, emphasizing their role in applying sustainable technologies, such as natural ventilation systems, in university-industry collaborations which support SDG development globally and locally for enhancing environmental and health benefits [21]. Craiut et al. (2022) also indicated that TT enables the transfer of academic research into practical applications, while the Triple Helix framework fosters synergy effect that aligns research with industry needs and government policies [21]. This collaborative model ensures that innovative techniques are effectively implemented, improving occupants’ well-being and supporting sustainable urban development.

The existing literature has explored methods in studying the influence of window-opening degree on natural ventilation performance in high-rise residential buildings, and many studies investigated the probability of COVID-19 infection risk using Wells–Riley model. However, there is little existing investigation examining the correlation between the degree of window opening, natural ventilation rate, COVID-19 infection risk, and the technology transfer to practical implementation. To address the research gap, the present study aims to provide methods to measure the natural ventilation rate, develop novel uNVeF metric for classifying natural ventilation performance, estimate COVID-19 infection risk across different window-opening scenarios, and utilize the TT framework to transfer academic case studies into sustainable building design solutions. These efforts can collectively promote healthy and sustainable indoor environments in urban cities around the world.

This paper commences by providing procedures and methods used in this case study and then presents the findings and data analysis which are used to explore the above research aims. The significance of natural ventilation rates to COVID-19 infection risk and implications of the findings are then discussed.

2. Methodology

In conjunction with addressing the research gap, the framework of the study is illustrated in Figure 1. The collected field measurement data, incorporating the uNVeF technique, were analyzed to determine the effectiveness of natural ventilation and estimate COVID-19 infection risk across 25 window-opening scenarios. All equations used are described in Section 2.1, Section 2.2 and Section 2.3.

This case study examined a Hong Kong public housing estate development that attained the highest level of BEAM Plus certification due to its sustainable design in building orientation and wind corridor implementation for facilitating and enhancing natural ventilation.

The development covers approximately 13.65 hectares. It consists of 7 buildings that are 40 stories high. The buildings are situated in a small town area surrounded by low-density local villages and relatively urban residential unit terrain as illustrated in Figure 2.

The selected urban residential unit has a floor area of 13.9 m² with a headroom of 2.55 m, and it is located on 35/F as shown in Figure 3. It is a studio urban residential unit with one living area, one washroom, and one kitchen as shown in Figure 4. The windows are side hung type as illustrated in Figure 5. The total overall operable window area is 1.77 m², and the operable window-to-wall ratio is 0.15.

The selection of the representative residential unit is based on 4 criteria. Firstly, 46% of the population resides in public housing estates in Hong Kong [22]; therefore, a public housing estate was chosen to represent the residential buildings in the present study.

Secondly, the selected building block, highlighted in Figure 2, is surrounded by other high-rise buildings and has the most unfavorable wind environment in terms of availability of natural ventilation. It was therefore ideal for the present study.

Thirdly, Lin et al. (2025) reported that the average building height in Hong Kong exceeds 100 m [23]. The selected 35th floor, which is about 101.5 m above the ground floor level, matched this characteristic height. Therefore, it was selected for analysis as a reflection of typical high-rise building conditions.

2.1. Intake Air Velocity Method

Site measurements were conducted on 18 and 19 March 2025. The outdoor temperature is on average 20 °C. The nearby Hong Kong Observatory weather station is located at Ta Kwu Ling which is 2.9 km away from the building. According to the data from this weather station, the prevailing winds are, in general, from east to southeast throughout the year at a speed of 6–8 m/s.

A series of window-opening scenarios was developed to measure the ventilation rates. Air velocity was measured at 500 mm away from the air inlets of each window to minimize turbulent effects. Four side-hung-type windows were under investigation. Two measurement points were taken for each side-hung-type window under varying effective window-opening areas. These data were used to calculate the ACH. The optimal natural ventilation configuration with the highest ACH was identified.

Table 1. Window-opening scenarios for measurement.

Scenario	Living Room Window 1 (%)	Living Room Window 2 (%)	Living Room Window 3 (%)	Kitchen Window (%)
1	100	100	100	50
2	100	100	100	75
3	100	100	100	100
4	100	100	100	0
5	75	75	75	50
6	75	75	75	75
7	75	75	75	100
8	75	75	75	0
9	50	50	50	50
10	50	50	50	75
11	50	50	50	100
12	50	50	50	0
13	0	0	0	50
14	0	0	0	75
15	0	0	0	100
16	0	0	0	0
17	50	0	0	0
18	75	0	0	0
19	100	0	0	0
20	0	50	0	0
21	0	75	0	0
22	0	100	0	0
23	0	0	50	0
24	0	0	75	0
25	0	0	100	0

outlines 25 window-opening scenarios to measure inlet air velocity and indoor air quality in the selected urban residential unit. In each scenario, the combination involved window-opening degrees of 0%, 50%, 75% or 100% for 3 windows located in the living room and 1 window located in the kitchen.

Table 2. Measuring instruments.

Name	Brand	Model	Accuracy	Measurement Intervals
Hot Wire Anemometer	Testo	425	±(0.03 m/s + 5% of measured value)	0.01 m/s
Temperature and Humidity Data Logger	Testo	175 H1	±0.4 °C (−20 to +55 °C) ± 1 Digit & ±0.03%RH/K ± 1 Digit	0.1 °C 0.1%RH
CO2 Meter	Testo	535	±(50 ppm + 5% of measured value)	1 ppm
PM2.5 Meter	Aeroqual	S500	±0.005 mg/m3 + 15%	0.001 mg/m3

lists the measuring instruments used to measure inlet air velocity, room temperature, relative humidity, CO₂ concentration, and PM_2.5 concentration.

Using the effective window-opening area equation (Equation (1)) suggested by Liu & Lee (2020), the natural ventilation rate can be determined for each scenario for comparison [8].

$$\mathrm{Effective\; window\; opening\; area}{A}_{eff}=L^2\times \sin \mathsf{\theta }+H\sin \mathsf{\theta } \cos \mathsf{\theta }+\left(L-L\mathit{cos} \theta \right)\times H$$

(1)

where

L is the length of the window (m);

H is the height of the window (m);

$\mathsf{\theta }$ is the window-opening angle in degrees.

2.2. Adjustment of Inlet Air Velocity for Different Floor Levels

Due to the difficulty of accessing private urban residential units for on-site measurement, the Hellman Exponential Law Equation (Equation (2)) was employed to estimate the inlet air velocity at lower floor levels.

$$v_2=v_1\times {\left({\displaystyle \frac{Z_1}{Z_2}}\right)}^{\alpha }$$

(2)

where

$v_2$ is the wind velocity at the desired height (m/s);

$v_1$ is the reference velocity (m/s);

$Z_1$ is the reference height (m);

$Z_2$ is the desired height (m);

$\alpha$ is the friction coefficient.

After determining the value of the friction coefficient by regression analysis, the wind velocity and inlet air velocity for urban residential units at different floor heights can be estimated.

2.3. Wells–Riley Model

The Wells–Riley model (Equation (3)) was utilized to estimate the probability of airborne infection risk in a scenario where a guest visits an occupant known to be infected with an airborne disease. This approach assesses the effectiveness of natural ventilation under different conditions in reducing public health risks associated with airborne diseases.

$$P=1-\exp \left({\displaystyle \frac{-Iqpt}{Q}}\right)$$

(3)

where

P is the probability of infection;

I is the number of infectors;

q is the quanta generation rate (quanta/h);

p is the pulmonary ventilation rate for a person (m³/h);

t is the exposure time interval (h);

Q is the room ventilation rate (m³/h).

Using this equation, the probability of infection can be estimated based on the actual room ventilation rate.

3. Data Collection and Analysis

The measurement results for average wind speed, room temperature, CO₂ concentration, PM_2.5 concentration, relative humidity, calculated ACH, and effective window-opening area in different scenarios are tabulated in

Table 3. Parameter data set in different scenarios.

Scenario	Average Inlet Air Velocity (m/s)	Room Temperature (°C)	CO2 Concentration (ppm)	PM2.5 Concentration (ppm)	Relative Humidity (%)	Total ${\mathit{A}}_{\mathit{e}\mathit{f}\mathit{f}}$ (m2)
1	0.093	17.2	411	0.016	50.4	2.69
2	0.086	17.0	418	0.018	50.2	2.78
3	0.188	17.3	423	0.015	50.1	2.80
4	0.135	17.9	436	0.016	50.1	2.35
5	0.363	17.5	418	0.016	43.1	2.58
6	0.383	17.8	409	0.015	37.6	2.67
7	0.289	17.8	409	0.016	37.9	2.69
8	0.159	17.9	423	0.019	48.6	2.24
9	0.193	18.0	409	0.015	35.2	2.13
10	0.241	17.9	406	0.021	37.1	2.22
11	0.151	17.9	410	0.015	38.6	2.24
12	0.040	18.2	438	0.016	35.8	1.79
13	0.040	18.2	495	0.015	34.9	0.34
14	0.064	18.2	535	0.015	35.2	0.43
15	0.068	18.3	543	0.016	37.9	0.45
16	0	18.4	548	0.014	61.6	0
17	0.031	18.3	578	0.015	38.6	0.65
18	0.061	18.4	574	0.017	38.2	0.81
19	0.058	18.4	562	0.015	38.5	0.85
20	0.036	18.4	549	0.016	38.0	0.65
21	0.036	18.5	507	0.016	38.5	0.81
22	0.036	18.5	503	0.017	37.4	0.85
23	0.027	18.5	520	0.016	38.1	0.49
24	0.031	18.5	538	0.016	38.1	0.61
25	0.048	18.5	559	0.015	38.5	0.64

.

3.1. Correlation of Window-Opening Degree with ACH

As the window size and window-opening degree vary in different scenarios, the overall window-opening degree for a series of windows is defined by a dimensionless uNVeF for comparison. It is the ratio of the actual window-opening area to the total window-opening area when fully open, ranging from 0 to 1.

$$\mathrm{uNVeF}\left(\mathrm{dimensionless}\right)={\displaystyle \frac{\sum A_{eff}}{\sum A_{eff,fullyopen}}}$$

(4)

where $A_{eff,fullyopen}$ is the total window-opening area of a series of windows in the fully opened condition.

By using Equation (4), the calculated uNVeF in different scenarios is tabulated in

Table 4. Measured inlet air velocity and ACH at 35/F.

Scenario	Living Room Window 1 (%)	Living Room Window 2 (%)	Living Room Window 3 (%)	Kitchen Window (%)	Average Inlet Air Velocity	ACH	uNVeF
					(m/s)
1	100	100	100	50	0.093	27.2	0.961
2	100	100	100	75	0.086	25.3	0.992
3	100	100	100	100	0.188	56.2	1
4	100	100	100	0	0.135	33.2	0.838
5	75	75	75	50	0.363	81.7	0.922
6	75	75	75	75	0.383	91.0	0.953
7	75	75	75	100	0.289	78.5	0.961
8	75	7	75	0	0.159	49.2	0.799
9	50	50	50	50	0.193	38.7	0.761
10	50	50	50	75	0.241	50.1	0.792
11	50	50	50	100	0.151	33.5	0.799
12	50	50	50	0	0.040	27.0	0.638
13	0	0	0	50	0.040	5.6	0.123
14	0	0	0	75	0.064	11.2	0.154
15	0	0	0	100	0.068	12.5	0.162
16	0	0	0	0	0	0	0
17	50	0	0	0	0.031	8.2	0.232
18	75	0	0	0	0.061	20.0	0.291
19	100	0	0	0	0.058	19.9	0.305
20	0	50	0	0	0.036	9.6	0.232
21	0	75	0	0	0.036	12.0	0.291
22	0	100	0	0	0.036	5.3	0.305
23	0	0	50	0	0.027	7.7	0.174
24	0	0	75	0	0.031	12.5	0.218
25	0	0	100	0	0.048	8.3	0.228

. The relationship between uNVeF and ACH is then studied by regression analysis.

The curve below in Figure 6 identifies the significance level of the influence of uNVeF on ACH. The regression model uses a 2nd-order polynomial equation. R² = 0.8469 shows a high correlation between uNVeF and ACH.

At lower uNVeFs, a small increase in window-opening degree from the fully closed state can greatly improve ACH as indicated by the steepest slope of the curve. Also, ACH was strongly influenced by uNVeF in this region.

At higher uNVeFs, the correlation level is moderately significant, and there is a slight decrease in ACH at the highest uNVeF, suggesting that more external factors such as turbulent flow may affect the natural ventilation rate.

The optimum window-opening scenario appeared when uNVeF is 0.953 (all windows in the living room and kitchen opened at 75%) which gives a result of 91 ACH.

3.2. Adjustment on Inlet Air Velocity

To compute the result using Equation (2), the friction coefficient factor $\alpha$ specific for this case study should be first determined.

Based on the data from the nearby Ta Kwu Ling weather station, the minimum ambient temperature is 8 °C, the maximum ambient temperature is 24 °C, the anemometer elevation is 13 m ($Z_1$), the wind speed measured is 9 km/h ${(v}_1)$ at 1000; and the wind direction is southeast on 19 March 2025.

To verify the compliance of natural ventilation at lower floor levels and derive the friction coefficient of $\alpha$, the wind velocity at public corridors facing the same direction as the selected urban residential unit at different floor levels was measured and tabulated in

Table 5. Measured wind velocity at corridor facing northeast.

Floor	Outside Wind Velocity (v2) (m/s)	Floor Level from Ground (Z2) (m)
40/F	1.90	115.5
35/F	1.75	101.5
30/F	1.68	87.5
25/F	1.53	73.5
20/F	1.49	59.5
15/F	1.14	45.5
10/F	1.09	31.5
5/F	1.14	17.5

. It shows that the wind velocity decreases linearly as the building floor level decreases.

The outside wind velocity at different floor levels was input into $v_2$ of Equation (2). The scatter graph in Figure 7 visualizes the relationship between wind velocity factor $v_2/v_1$ and floor-level factor $Z_2/Z_1$. Power function was used for regression analysis. The result R² = 0.9232 shows a strong correlation between wind velocity $v_2/v_1$ and floor level $Z_2/Z_1$. Therefore, the inlet wind velocity adjustment equation is derived as follows:

$$v_2/v_1=0.757{\left(Z_2/Z_1\right)}^{0.2907}$$

(5)

The site-specific friction coefficient is determined to be $\alpha$ = 0.2907.

The inlet air velocity for each window facing the same direction as the selected urban residential unit at any floor level can be estimated separately using Equation (5).

For the best ventilation scenario at 1/F (lowest floor) in the same building, inlet air velocity is estimated as shown in

Table 6. Summary of adjusted inlet air velocity $\left(v_2\right)$ for the best ventilation scenario.

Best Ventilation Scenario	Inlet Air Velocity at 35/F (v1) (m/s)	Inlet Air Velocity at 1/F (v2) (m/s)
Window 1	0.285	0.096
Window 2	0.155	0.052
Window 3	0.383	0.129
Window 4	0.708	0.239

.

For the worst ventilation scenario at 1/F (lowest floor) in the same building, inlet air velocity is estimated as shown in

Table 7. Summary of adjusted inlet air velocity $\left(v_2\right)$ for the worst ventilation scenario.

Worst Ventilation Scenario	Inlet Air Velocity at 35/F (v1) (m/s)	Inlet Air Velocity at 1/F (v2) (m/s)
Window 1	0	0
Window 2	0	0
Window 3	0.108	0.036
Window 4	0	0

.

The result of ACH can be calculated by Equation (6). The ACH is obtained from the summation of the air flow rate of windows 1 to 4.

$$\begin{gathered} ACH=\sum \left({v_{2.windown}\times A}_{eff.windown}\right)\times 60\times 60÷roomvolume \\ (wheren=\mathrm{window}\mathrm{number}1to4) \end{gathered}$$

(6)

By repeating the calculation steps for another scenario, the calculation result was tabulated in

Table 8. Estimated inlet air velocity and ACH at 1/F (lowest floor).

Scenario	Living Room Window 1 (%)	Living Room Window 2 (%)	Living Room Window 3 (%)	Kitchen Window (%)	Inlet Air Velocity	ACH	uNVeF
					(m/s)
1	100	100	100	50	0.031	9.2	0.961
2	100	100	100	75	0.029	8.5	0.992
3	100	100	100	100	0.063	19.0	1
4	100	100	100	0	0.046	11.2	0.838
5	75	75	75	50	0.122	27.6	0.922
6	75	75	75	75	0.129	30.7	0.953
7	75	75	75	100	0.098	26.5	0.961
8	75	75	75	0	0.054	16.6	0.799
9	50	50	50	50	0.065	13.1	0.761
10	50	50	50	75	0.081	16.9	0.792
11	50	50	50	100	0.051	11.3	0.799
12	50	50	50	0	0.013	9.1	0.638
13	0	0	0	50	0.013	1.9	0.123
14	0	0	0	75	0.022	3.8	0.154
15	0	0	0	100	0.023	4.2	0.162
16	0	0	0	0	0	0	# undefined
17	50	0	0	0	0.010	2.8	0.232
18	75	0	0	0	0.021	6.8	0.291
19	100	0	0	0	0.020	6.7	0.305
20	0	50	0	0	0.012	3.2	0.232
21	0	75	0	0	0.012	4.0	0.291
22	0	100	0	0	0.012	4.2	0.305
23	0	0	50	0	0.009	1.8	0.174
24	0	0	75	0	0.010	2.6	0.218
25	0	0	100	0	0.016	4.2	0.228

^# “undefine” of uNVeF in Scenario 16: all windows are closed; zero of window area, air velocity, ACH in Table 1, Table 3 and Table 4.

. It found that all scenarios at 1/F (lowest floor) can achieve the minimum 1.5 ACH requirement.

The results of the best and worst ventilation scenarios are summarized in

Table 9. Summary of best and worst ventilation scenarios.

	Living Room Window Opening (%)	Kitchen Window Opening (%)	ACH
Best ventilation scenario at 35/F	75	75	91.0
Best ventilation scenario at 1/F	75	7	30.7
Worst ventilation scenario at 35/F	0	50	5.3
Worst ventilation scenario at 1/F	0	50	1.8

. It shows that the natural ventilation rate achieved at higher floor levels is significantly larger than at lower floor levels.

3.3. Estimation on COVID-19 Infection Risk

The Wells–Riley model is applied to estimate the COVID-19 infection risk in a selected urban residential unit. The parameter values from

Table 10. Values of input parameters.

Input Parameter	Value
Number of infectors (I)	1
Quanta generation rate (q)	61 quanta/h
Pulmonary ventilation rate (p)	0.8496 m3/h
Exposure duration (t)	1 h

were input into Equation (3) to calculate the probability of COVID-19 infection risk.

For the quanta generation rate, Buonanno et al. (2020) and Shen et al. (2021) reviewed a COVID-19 outbreak event at a restaurant in Guangzhou, China and suggested that the quanta generation rate can be backward calculated using the Wells–Riley model when the ventilation settings of the outbreak are known. Based on this approach, the quanta generation rate of COVID-19 virus particles while the infector is seated and speaking is taken as 61 quanta/h [19,20].

For the pulmonary ventilation rate, Xia et al. (2020) reported that the tidal volume of Chinese patients suffering COVID-19 is 8 mL/kg. Also, ICRP (1994) indicated an average weight of 59 kg for Chinese male adults. And Liu, S. et al. (2020) determined that the breathing rate of COVID-19 patients in China is 30 breaths per min. Therefore, the pulmonary ventilation rate is calculated to be 0.8496 m³/h [16,17,18].

Using Equation (3), the probability of COVID-19 infection risk when a healthy person visits a COVID-19 infector for 1 h was assessed. When the infector is seated and speaking in an urban residential unit with 1.5 ACH natural ventilation rate, the probability of COVID-19 infection risk for a healthy visitor not wearing a mask is 62.2%. The infection risk % values under different scenarios at 35/F and 1/F (lowest floor) are tabulated in

Table 11. Probability of COVID-19 infection risk at different ACH at 35/F.

Scenario	Inlet Air Velocity	ACH	uNVeF	Q	Infection Risk
	(m/s)			(m3/h)	(%)
1	0.093	27.2	0.961	621.5	5.2
2	0.086	25.3	0.992	568.3	5.6
3	0.188	56.2	1	1260.8	2.6
4	0.135	33.2	0.838	752.9	4.3
5	0.363	81.7	0.922	2049.3	1.8
6	0.383	91.0	0.953	2134.5	1.6
7	0.289	78.5	0.961	1807.8	1.8
8	0.159	49.2	0.799	1154.3	2.9
9	0.193	38.7	0.761	1133.0	3.7
10	0.241	50.1	0.792	1374.5	2.9
11	0.151	33.5	0.799	941.2	4.3
12	0.040	27.0	0.638	799.1	5.3
13	0.040	5.6	0.123	159.8	23.0
14	0.064	11.2	0.154	255.7	12.2
15	0.068	12.5	0.162	273.5	11.0
16	0	0	0	0	## Undefined
17	0.031	8.2	0.232	238.0	16.2
18	0.061	20.0	0.291	461.7	7.0
19	0.058	19.9	0.305	440.4	7.1
20	0.036	9.6	0.232	277.0	14.2
21	0.036	12.0	0.291	277.0	11.5
22	0.036	12.3	0.305	273.5	11.1
23	0.027	5.3	0.174	166.9	24.0
24	0.031	7.7	0.218	191.8	17.2
25	0.048	12.5	0.228	294.8	11.0

^## “undefine” of risk in Scenario 16: all windows are closed; zero of window area, air velocity, ACH in Table 1, Table 3 and Table 4.

and

Table 12. Probability of COVID-19 infection risk at different ACH at 1/F (lowest floor).

Scenario	Inlet Air Velocity	ACH	uNVeF	Q	Infection Risk
	(m/s)			(m3/h)	(%)
1	0.031	9.2	0.961	305.4	14.7
2	0.029	8.5	0.992	291.2	15.7
3	0.063	19.0	1	639.3	7.4
4	0.046	11.2	0.838	387.1	12.2
5	0.122	27.6	0.922	1136.5	5.2
6	0.129	30.7	0.953	1239.5	4.6
7	0.098	26.5	0.961	944.7	5.4
8	0.054	16.6	0.799	433.3	8.4
9	0.065	13.1	0.761	500.8	10.6
10	0.081	16.9	0.792	650.0	8.3
11	0.051	11.3	0.799	412	12.1
12	0.013	9.1	0.638	88.8	14.8
13	0.013	1.9	0.123	17.8	53.9
14	0.022	3.8	0.154	32.0	32.1
15	0.023	4.2	0.162	35.5	29.2
16	0	0.0	0	0	## Undefined
17	0.010	2.8	0.232	24.9	40.9
18	0.021	6.8	0.291	60.4	19.4
19	0.020	6.7	0.305	60.4	19.5
20	0.012	3.2	0.232	28.4	36.4
21	0.012	4.0	0.291	35.5	30.3
22	0.012	4.2	0.305	35.5	29.1
23	0.009	1.8	0.174	17.8	55.6
24	0.010	2.6	0.218	21.3	42.8
25	0.016	4.2	0.228	39.1	29.3

^## “undefine” of risk in Scenario 16: all windows are closed; zero of window area, air velocity, ACH in Table 1, Table 3 and Table 4.

.

In Table 11 and Table 12, when the uNVeF was at 0.953, the maximum natural ventilation of 91 ACH and 30.7 ACH was found. The COVID-19 infection risk results are 1.6% on 35/F and 4.6% on 1/F (lowest floor), respectively. Compared to the baseline 1.5 ACH natural ventilation rate, which is 62.2%, the infection probability can be significantly reduced by 97.4% and 92.5%, respectively.

The scatter chart in Figure 8 was plotted to illustrate the relationship between uNVeF and COVID-19 infection risk in the urban residential unit at high floor levels and low floor levels. An exponential equation was applied for regression analysis. The results R² = 0.7815 (35/F) and 0.8012 (1/F (lowest floor)) show high correlation between uNVeF and infection risk.

The higher the natural ventilation rate, uNVeF corresponded to lower COVID-19 infection risks. The infection risks are within 20% for both cases, with the uNVeF higher than 0.7. At a lower uNVeF, the infection risks at low floor levels exhibit significantly higher infection risks than at high floor levels in the building.

4. Discussion

4.1. Summary of Findings

The present study systematically conducted field measurements for the natural ventilation performance of a 13.9 m² urban residential unit in a Hong Kong public housing estate by introducing a novel metric: the urbanized natural ventilation effectiveness factor (uNVeF). The ACH achieved by natural ventilation was examined using the inlet air velocity method under 25 window-opening scenarios. The results show that the natural ventilation rate fully complied with the Building Department’s statutory requirement of 1.5 ACH. Notably, the highest ACH was obtained when all windows in the living room and kitchen were opened at 75% with the corresponding uNVeF at 0.953. However, a decrease in ACH was observed at 100% window opening. Moreover, a site-specific friction coefficient of 0.2907 was derived from the Hellman Exponential Law. Using this coefficient, the ACH of another floor level can be easily estimated. Furthermore, using the Wells–Riley model, the present study found a significant reduction in COVID-19 infection risk of 97.4% at upper floor level and 92.5% at lower floor level compared to the 1.5 ACH baseline condition. Finally, the uNVeF metric was further correlated with the infection risk of airborne diseases, accounting for COVID-19 as an example. Adopting the uNVeF standard of 0.7 to 0.95 can enhance natural ventilation, public health, and sustainable development in urban residential buildings.

4.2. Interpretation of Results

The findings validate the performance of natural ventilation in complying with the statutory ACH requirements and establish uNVeF as an innovative framework for conveniently assessing natural ventilation efficiency in high-rise urban settings. The peak ACH at a uNVeF of 0.953 distinctly reflects optimal airflow dynamics at a particular effective window-opening area and is further supported by a high correlation (R² = 0.8469), indicating that there is a relatively good predictive relationship between uNVeF, ACH and infection risk. Notably, the unexpected decrease in ACH at full window opening likely results from complex turbulent airflow interactions. This phenomenon deserves further in-depth exploration into external factors such as building orientation, microclimate conditions, and urban morphology. Moreover, the method of utilizing the Hellman Exponential Law incorporating a site-specific friction coefficient facilitates a quick estimation of ventilation rate, particularly for the worst natural ventilation scenarios at lower floors of urban buildings where there was marginal compliance of ACH at lower levels due to reduced wind velocity. Critically, the substantial reduction in infection risk at higher uNVeF values underscores the role of natural ventilation as a low-energy solution for mitigating airborne disease transmission and offering scalable public health benefits in dense urban environments which aligns with global sustainability goals. These insights collectively highlight the transformative potential of uNVeF as a convenient tool for optimizing ventilation strategies and enhancing urban resilience.

4.3. Comparative Analysis

The observed optimal window-opening range of 60–90%, corresponding to a uNVeF of 0.7–0.95, aligns closely with Liu & Lee (2020) who reported peak ventilation efficiency within this range in Hong Kong residential buildings using CFD simulations [8]. Furthermore, the substantial reductions in infection risk observed in the present study strongly corroborate the findings by Escombe et al. (2007) and Dai et al. (2023) who demonstrated natural ventilation performance in mitigating airborne disease transmission [13,14]. However, their studies focus on general indoor environments rather than the high-rise urban residential units. In contrast, the present research relies on field measurements of inlet air velocity method which significantly enhances its practical relevance in classifying the level of infection risk in high-rise urban residential buildings as it typically requires fewer resources and is less time-consuming.

4.4. Sensitivity of Infection Risk Estimation

A sensitivity analysis was conducted to assess the impact of variations in quanta generation rate (q) and pulmonary ventilation rate (p) on COVID-19 infection risk estimates derived from the Wells–Riley model to address potential limitations of fixed parameter values.

The present baseline model employed a quanta generation rate representing a seated and speaking infector and a pulmonary ventilation rate based on resting Chinese male adults. However, these parameters may differ significantly with increased activity levels or more transmissible COVID-19 variants. To account for the variability, the pulmonary ventilation rate was adjusted to 1.8 m³/h, reflecting moderate exercise [17], and the quanta generation rate was increased to 100 quanta/h which estimated the variants of Delta and Omicron [14]. The analysis evaluated two scenarios: the baseline condition (ACH = 1.5) and the optimal ventilation condition (uNVeF = 0.953, ACH = 91 at 35/F, ACH = 30.7 at 1/F). Other parameters including exposure time (1 h) and number of infectors (1) remained unchanged as per Equation (3).

With original parameters, the infection risk was calculated at 62.2%. Increasing the pulmonary ventilation rate to 1.8 m³/h elevated the risk to 84.3%, due to greater inhalation of infectious quanta during exercise. Raising the quanta generation rate to 100 quanta/h increased the risk to 76.5% because of higher viral load emission. Combining both adjustments (q = 100 quanta/h, p = 1.8 m³/h) yielded an infection risk of 94.7%. These outcomes highlight the mutual effect of elevated activity and variant transmissibility in a low natural ventilation rate situation.

At 35/F (ACH = 91), the baseline infection risk was 1.6%. With p = 1.8 m³/h, the risk increased to 3.4%. With q = 100 quanta/h, it increased to 2.6%. And with both p and q increased, it reached 5.5%. At 1/F (ACH = 30.7), the baseline risk was 4.6%, escalating to 9.7%, 7.5%, and 15.8% for the respective variations. These findings indicate that high ventilation rates substantially reduce infection risk, but the benefits are attenuated under conditions of increased pulmonary rates or quanta emissions, particularly at lower floors with constrained ACH.

4.5. Thermal Comfort and Infection Control

Natural ventilation impacts both infection risk and thermal comfort which is essential for occupant well-being. The Predicted Mean Vote model defines thermal comfort as a function of air temperature, radiant temperature, ventilation rate, and humidity [2]. In the present study, measurements were conducted in March 2025 at an average outdoor temperature of 20 °C which is a comfortable condition. However, higher natural ventilation rates introduce high wind velocity may potentially affect room temperature and cause thermal discomfort. Additionally, reduced glazing shading allows more solar radiation to penetrate into the rooms and may potentially increase ventilation demands and affect thermal comfort. Bungău (2024) highlights the need for further research to optimize mixed-mode ventilation systems, ensuring they effectively balance infection control with thermal comfort [2].

4.6. Built Environment and Health

The relationship between built environment characteristics and occupant health is critical, as highlighted by Bungau (2024). A recent review emphasized that inadequate natural ventilation is correlated to increased exposure to allergens, pollutants, and respiratory diseases [24]. The present research demonstrates that optimizing window configurations (uNVeF = 0.7–0.95) significantly reduces infection risk, directly improving health benefits, and can be further linked to reducing indoor pollutants in urban residential buildings. The research findings support sustainable building environments by providing evidence aligning natural ventilation optimization to IAQ improvements and supporting the design of healthy urban residences, particularly in public housing estates.

4.7. Technology Transfer (TT) in Urban Buildings

The uNVeF metric developed in the present study provides a practical, scalable, and low-cost tool for optimizing natural ventilation with significant implications for public health and sustainable environment in high-rise urban residential buildings. The optimal uNVeF (0.7–0.95) offers actionable guidelines for architects and engineers to design window configurations and operations to maximize natural ventilation, enhance indoor air quality (IAQ), and mitigate airborne disease transmission risks, such as COVID-19. These findings align with global sustainability goals (SDG 3) which promote healthier indoor environments in urban cities [4].

The practical application of the uNVeF metric is significantly enhanced through TT mechanisms as outlined in the Triple Helix model. It emphasizes collaboration among academic institutions, industry, and government to accelerate innovation and implementation of sustainable technologies [21]. By integrating uNVeF into building design processes, stakeholders can bridge the gap between academic research and practical applications. For instance, academic institutions can provide evidence-based methodologies, such as the field measurement techniques and uNVeF framework developed in this study, to guide industry practices. Industry stakeholders including investors, developers, and construction companies can adopt these findings to design buildings with optimized window configurations to facilitate compliance with ventilation standards. Government can support this process by revising building codes and sustainability certifications, such as BEAM Plus, to incorporate uNVeF as one of the standard metrics for natural ventilation performance.

The Triple Helix framework aligns research with industry needs and policy frameworks to ensure effective implementation of innovative techniques [21]. For example, integrating uNVeF into building codes could drive innovation in passive design strategies, reducing energy consumption and aligning with global energy efficiency goals. This collaborative model also supports knowledge exchange through global research networks particularly involving regions like China and the USA who are leaders in sustainable construction research [21]. The low-cost, field-based measurement technique used in present research requires minimal measuring equipment. This enhances its scalability for widespread adoption in urban planning and makes it easily accessible for both new developments and retrofitting projects in public housing estates.

Furthermore, the uNVeF framework empowers residents and caregivers by providing clear guidelines for optimizing natural ventilation during home-based self-quarantine scenarios. In TT frameworks, residents and caregivers should be educated on the benefits of natural ventilation and understand the present findings in sharing workshops, training programs, and policy briefings for enhancing the culture of low-infection-risk practices. These efforts collectively contribute to creating resilient and healthy urban environments that align with the broader objectives of urban science and global sustainability.

4.8. Research Limitation

The proposed uNVeF metric as presented in this study is tailored to the characteristics of the selected urban residential unit. Consequently, its applicability is currently limited to this particular case study and should not be directly generalized to other units or developments without further validation. Despite this limitation, the metric retains significant practical value as its low-cost and field-based measurements make it easily accessible for architects, engineers, building managers, and even laymen to assess and optimize natural ventilation in buildings.

While the uNVeF technique demonstrates promising results within the confines of the studied unit, the methodology has several limitations that warrant discussion.

First of all, the inlet air velocity method is sensitive to external factors such as wind conditions, turbulence, building orientation, and surrounding urban structures that may introduce variability in measurements. Also, the reported natural ventilation rates represent the peak measurement under optimal wind conditions only and without fully accounting for the dynamic changes in wind patterns in dense urban environments with complex airflow interactions. Therefore, cautious interpretation is advised when applying these site-specific numerical results to other buildings.

Additionally, the Hellman Exponential Law used to estimate natural ventilation rates at lower floors involved a self-determined and site-specific frictional coefficient. The coefficient may not be applicable to other building facades of urban buildings with different wind exposure, terrain, and structural characteristics. For accurate application, a new frictional coefficient should be determined on a case-by-case basis.

Moreover, since the field measurements were conducted in spring season, they do not reflect seasonal changes in temperature, wind velocity, and wind direction that may influence the natural ventilation performance. This restricts the applicability of the findings across different times of the year. More field measurements are suggested to be conducted across all seasons to capture these variations to enhance the applicability of uNVeF.

Furthermore, the uNVeF metric was developed based on measurements from a single residential unit with a northeast-facing orientation. The generalizability of the units with different orientations, floor plans, or building designs was limited. A database established for a comprehensive study of uNVeF performance metrics with ACH across various urban contexts would enable robust statistical analysis, improve generalizability, and enhance prediction accuracy.

Lastly, the Wells–Riley model used for infection risk estimation assumes uniform mixing of air inside the high-rise residential unit that may potentially oversimplify the realistic airflow patterns where localized air stagnation may occur. In addition, the infection risk estimation relied on fixed parameters of quanta generation rate and pulmonary ventilation rate may not account for the variations in different ages, genders, and activity levels or more transmissible virus variants. These variations could be eliminated when the variable data specific to different conditions are available.

4.9. Recommendations for Future Research

The authors understand that urban ventilation is influenced by site-specific factors such as building density, canopy effects, and microclimate variations. However, the uNVeF framework is flexible for adaptation to other buildings by recalibrating the friction coefficient and conducting site-specific measurements. For example, future studies could apply the methodology to buildings with different orientations or in other urban settings, using localized weather data to adjust the Hellman Exponential Law. The strong correlation between uNVeF, ACH and infection risk suggests that the metric can be a robust indicator of ventilation performance across similar high-rise urban environments, provided that site-specific adjustments are made.

To enhance the uNVeF framework, future research can expand the framework by incorporating inlet air velocity measurements over a full year to capture seasonal variations in temperature and wind patterns. This would improve the applicability of the uNVeF technique across different climatic conditions. Also, future studies could conduct more testing across diverse building types and urban contexts. The data of these measurements can be stored in a database for further comprehensive analysis and application.

Additionally, by integrating uNVeF with real-time environmental monitoring systems, its applicability to complex urban settings can be enhanced while maintaining its low-cost and scalable nature. This could provide more accurate predictions and broader practical relevance.

4.10. Scope of Validity and Practical Relevance

The uNVeF metric and its associated findings are primarily validated for a northeast-facing high-rise residential unit under specific climatic conditions. Its scope of validity is thus limited to similar buildings in dense urban environments with comparable wind patterns, orientations, and window configurations. The variable site-specific friction coefficient may apply to buildings with different facade characteristics or surrounding urban contexts.

To achieve broader practical relevance, the uNVeF technique could be refined and validated using CFD simulations across at least 10% of the residential units in the development, representing a range of orientations and configurations. This would involve generating a more robust data set, enabling calibration, and potentially generalizing the technique. Through this extended validation process, the technique could be adapted to accommodate a wider range of design scenarios, enhancing its applicability in real-world planning and design.

Additionally, alternative approaches such as full-scale field measurements, wind tunnel testing, or coupled energy and airflow simulations could provide further insights and improve prediction accuracy. While these methods are more resource-intensive, they can quantify the uncertainty of the current approach, estimated at ±10% for key ventilation performance metrics based on preliminary sensitivity analyses.

In summary, the current results should be viewed as a demonstration of the uNVeF technique within a specific context. Its generalization to broader applications will require further simulation, validation, and refinement.

5. Conclusions

This case study establishes the urbanized natural ventilation effectiveness factor (uNVeF) as a low-cost, field-based, and innovative technique for enhancing natural ventilation in high-rise urban residential units. By integrating regression analysis of wind velocity, window-opening configurations, and building height, the study demonstrates that optimized window settings consistently surpass statutory ventilation requirements. The uNVeF metric identifies window-opening configurations that are capable of achieving peak ventilation efficiency, offering actionable guidance for architects and residents.

The Wells–Riley model highlights the public health benefits of optimized natural ventilation, predicts significant reductions in airborne disease transmission risk, and classifies the level of indoor infection risk. These findings magnify the role of natural ventilation as a low-energy solution for mitigating infection risks in urban environments and contributing to global sustainability objectives.

The application of uNVeF technique bridges academic research with industry and government through the Triple Helix model.

Throughout this case study, it is recommended that future research integrate more accurate factors of the uNVeF technique for general application into building codes and urban planning, fostering resilient, healthy, and sustainable urban environments.