1. Introduction
Building ventilation is the main strategy for replacing contaminant indoor air with fresh outdoor air through natural or mechanical means [
1]. Natural ventilation is the intentional movement of air from the outdoors to the indoors typically driven by wind forces or buoyancy effect caused by temperature differences without the use of mechanical systems. These natural driving forces are normally produced through operable windows and pressure differences between spaces. These forces can drive the flow of fresh air through and within a building, moving from a high-pressure to a low-pressure zone. Indeed, natural ventilation is considered an efficient strategy for reducing cooling energy use of buildings in some cases in addition to improving indoor air quality by diluting pollutants and minimizing the impact of bacterial and viral infections.
From an operational standpoint, natural ventilation could be considered a simple system; however, high air change rates can be achieved when specific design techniques are implemented. It is worth noting that buildings operating with mechanical ventilation systems are associated with more Sick Building Syndromes (SBS) such as headache, dizziness, and throat irritation compared to buildings depending merely on natural ventilation systems [
2,
3,
4]. Nevertheless, the extensive use of mechanical ventilation systems and their possible negative impact on the occupants and the environment offered the opportunity to sustainable strategies and passive systems, such as natural ventilation systems, to spread widely as a sustainable supplementary solution [
5]. On the other hand, natural ventilation could be a reliable system to improve IAQ for most buildings, especially for those characterized as high occupancy density buildings, such as educational buildings. For classrooms, natural ventilation is considered as one of the most commonly used sustainable solutions capable of providing high ventilation rates to ensure healthy and productive indoor environment while reducing energy use compared to other ventilation strategies [
6,
7].
IAQ is associated strongly with human health, and is usually defined as the ability to meet human needs [
8]. Human needs have many requirements that must be considered during the designing stage of a building. Recently and after the COVID-19 pandemic, people have been showing a special interest in the health aspect of the indoor environment where air pollution levels often exceed the outdoor environment, while approximately 90% of the people spend most of their time in indoor spaces [
9,
10]. Nevertheless, improving the IAQ of existing classrooms can be challenging if conventional design features are considered. High ventilation rates could be achieved with natural ventilation as stated above; however, there are several factors that could influence the performance of natural ventilation such as outdoor climate conditions, building geometry, and the design of the building envelope [
11]. Recent studies discussed the important role of natural ventilation in improving the quality of the indoor environment. One of the studies investigated the IAQ in eight different schools while evaluating and proposing solutions to improve it through natural ventilation strategies, which showed positive results [
12]. A study conducted by Ma’bdeh et al. [
13] proved that through some architectural modifications to a classroom building, especially in hot climate areas, the IAQ can be enhanced, which in turn improves students’ academic performance and provides a healthy environment for them. The study used computer simulations to analyze the rate of natural ventilation in the classroom building before and after the architectural improvement. The results showed positive indicators and improvement of the rate of natural ventilation after the architectural modifications. Furthermore, natural ventilation systems can directly affect IAQ; however, due to the large number of parameters such as opening types, dimensions, orientations, air distribution, etc., accurate predictions of the performance of natural ventilation can be hard to quantify and control [
14].
Moreover, a number of studies investigated the interaction between human and indoor microbial contamination in indoor spaces [
10,
15,
16]. The transmission and spread of these microbial and viral particles typically happen through airborne means and the interaction with indoor building surfaces. It is worth mentioning that a large number of studies routed the impact of airborne transmission diseases to human health [
17,
18,
19]. Many previous studies indicated that natural ventilation reduces the risks of infection and airborne diseases, and there is recent scientific evidence that supports the validity of the outcomes of these studies [
20]. In fact, the relationship between airborne infection and ventilation systems has been discussed extensively in the literature. Most of these studies have introduced mathematical models to explore the viral dynamics and investigate pathogenic characters of viral infection [
21]. However, the most recognized method used for evaluating the impact of ventilation strategies on viral transmission in buildings is Computational Fluid Dynamics (CFD). In addition, CFD modeling has also been used to monitor microbial spread and airborne cross-infection [
22]. In fact, CFD has long been utilized as an effective and efficient method to accurately analyze air distribution and precisely visualize airflow patterns and contaminant movements in enclosed spaces. For example, a recent numerical study conducted by Wang et al. [
23] explored cross-transmission of pollutants for a single-sided natural ventilation. This study indicated that the risk of infection decreases when increasing the ventilation rate; however, the risk of infection is not quantified in this research study. In another study presented by Lipinski et al. [
24], several ventilation strategies have been analyzed using CFD to examine risk reduction in pathogen transmission in buildings. Furthermore, a recent study investigated the likelihood of COVID-19 airborne transmission for various scenarios of aerosol droplets dilution and transport using CFD along with Monte-Carlo simulations to quantify the exposure time of aerosol for indoor environments [
25]. Moreover, Peng et al. [
26] reviewed several studies that have investigated pathogen transmission through CFD simulations and concluded that natural ventilation is considered the main method to dilute pathogen concentrations. CFD modeling has also been utilized to explore the evolution of coughed droplets for various ventilation patterns in air-conditioned space [
27]. Additionally, airborne contaminant transmission has been evaluated in a hospital and laboratory settings using airflow simulation environment that couples CFD with a multizone network software [
28]. Some research studies utilized other modeling techniques besides CFD such as empirical models and dynamic airflow modeling using EnergyPlus to evaluate indoor pollutants and viral infection transmission in naturally ventilated buildings [
29,
30].
The Wells–Riley model has been used in various recent research studies to estimate the probability of infection in buildings to enhance ventilation techniques in order to minimize the probability of airborne infection. This mathematical model assumes a steady-state quanta generation rate delivered by infected people and a ventilation rate term responsible for diluting the quanta concentration of the virus [
31]. Therefore, according to this model, increasing the ventilation rate will exponentially reduce the probability of infection in buildings. Hence, many research studies explored the influence of ventilation rates on infection transmission in buildings either delivered naturally through windows or delivered by mechanical ventilation systems. For instance, Dai and Zhao [
32] evaluated the relationship between ventilation rates and the probability of infection in buildings using the Wells–Riley model and determined the quantum generation rate value of COVID-19. Additionally, Kurnitski et al. [
33] used the Wells–Riley model to develop a simplified ventilation equation capable of estimating the ventilation rate of particular quanta emission rates in order to reduce infection risk in buildings. On the other hand, mechanical ventilation has also been evaluated to improve ventilation performance to minimize infection risks. A recent study by Sha et al. [
34] presented a modified Wells–Riley model to investigate the effect of mechanical ventilation on airborne viral infection by combining ventilative cooling rate and dilution ventilation rate. Another recent study introduced by Stabile et al. [
35] performed simulation analysis of typical school scenarios to estimate the required mechanical and natural ventilation rates to prevent virus transmission. It is worth noting that limited research studies applied the Wells–Riley model to estimate the probability of infection in the context of multizone building environments. For instance, Yan et al. [
36] estimated viral infection risk of SARS-CoV-2 for a mechanically ventilated multizone building using CONTAM simulation software. Additionally, Alaidroos et al. [
30] investigated the probability of infection of a naturally ventilated historical building using the EnergyPlus Airflow-Network model along with the Wells–Riley model to estimate the probability of infection for various natural ventilation rates.
Indeed, high ventilation rates are most of the time necessary to maintain acceptable indoor air quality as mentioned above; however, high ventilation rates might cause high indoor air velocities that could lead to uncomfortable draught in the occupied spaces [
37]. According to ASHRAE Handbook of Fundamentals [
38], draught is “an undesired local cooling of the human body caused by air movement” and is classified as one of the most annoying factors affecting indoor environmental comfort in ventilated spaces. Therefore, it is crucial to investigate the characteristics of airflow within the indoors and its impact on draught and occupant comfort especially for naturally ventilated spaces where airflow turbulence is commonly occurred due to air velocity fluctuations [
39]. Generally, in ventilated spaces, periodic fluctuation of air velocity commonly exists which causes turbulence. In fact, air turbulence is known to cause draught complaints more than laminar (nonfluctuating) airflow [
40]. Moreover, several studies showed that complaints of draught increase with the increase in turbulence intensity [
37]. This explicit relationship between turbulence intensity and draught complaints is due to the increase in convective heat transfer when turbulence increases [
41].
The earlier studies that investigated the effect of draught on occupants’ discomfort focused on examining the local air movement around an occupant’s head and ankles, which was the first to conclude that indoor discomfort is increased with the increase in indoor air velocity [
42,
43]. However, these studies did not take occupants’ thermal perceptions into account; thus, it was hard to determine the exact influence of draught in contrast to the overall occupant’s comfort. Afterwards, a series of experiments were conducted to analyze the characteristics of air movement and its impact on the overall comfort for occupants who were thermally neutral. Specifically, Fanger and Pedersen [
44] explored the effect of air velocity, periodical fluctuation of airflow, and air temperature on draught discomfort. They confirmed the relationship between air velocity and draught discomfort in addition to finding that a fluctuating airflow leads to more discomfort conditions than a constant airflow. Furthermore, Fanger and Christensen [
45] also found in another experimental study that turbulent airflow significantly influences the sensation of draught. Yet, in this study, the turbulence intensity was kept within the range of 30% to 60%. Therefore, another related study conducted by Fanger et al. [
40] expanded the previous findings by applying three levels of turbulence intensities (low: less than 12%; medium: between 20% and 35%; high: greater than 55%). Hence, turbulence intensity was integrated into the new draught model which has since been adopted in indoor thermal comfort standards such as ASHRAE 55 [
46] and ISO 7730 [
47] to predict the percentage of dissatisfaction from draught.
Few recent studies have discussed the influence of ventilation on draught risk; for instance, Mumovic et al. [
48] revealed that mechanical and hybrid ventilation systems in new secondary schools usually causes draught problems where the draught risk is above 15% most of the time. It is to be noted that the risk of draught should always be less than 15% in the occupied spaces according to the ASHRAE thermal comfort standard [
49]. Another investigation conducted by Deng and Tan [
50] explored the effect of low air temperatures and outdoor wind speed on the draught risk in a naturally ventilated office building. Moreover, Markov et al. [
51] also discussed the possible problems with the draught assessment procedure provided by ISO 7730:2005 [
47] when strong fluctuations in airflow is anticipated in university classrooms. Additionally, Conceição et al. [
52] a conducted a numerical study to predict the draught risk in addition to IAQ and indoor thermal comfort for a classroom with desk-type personalized ventilation system.
As discussed earlier, numerical simulation methods are commonly used to analyze and evaluate ventilation system effectiveness on minimizing airborne infection transmission in buildings. The numerical modeling techniques presented in the literature are used successfully to visualize contaminants distribution and airflow patterns; however, the vast majority of these numerical simulation packages are not capable of estimating the impact of airborne viral transmission on occupant’s infection rate inside buildings. In fact, there is a lack of research studies to predict the probability of infection based on the ventilation rates in naturally ventilated buildings, especially for educational buildings where occupancy density is commonly high. In addition, occupants’ discomfort due to draught caused by high ventilation rates is a common issue in naturally ventilated buildings that has not been extensively studied yet. Hence, optimal ventilation rates that serve the right balance of both viral infection prevention and minimum draught risk is a research gap that should be investigated. Therefore, the contribution of this research study is to provide a list of design recommendations to prevent airborne viral infection while avoiding draught discomfort through a comprehensive analysis of the effectiveness of natural ventilation for educational buildings.
This study will use a classical mathematical model that predicts the probability of infection using several variables such as ventilation rates, duration of exposure to infectors, and the number of source patients. In addition, Fanger’s draught model will be used to estimate the draught risk in this study. A whole building energy simulation tool is utilized to investigate the influence of climate conditions and building design characteristics and to predict hourly ventilation rates which are then fed to the models to analyze the probability of infection and the draught risk for each time step.
2. Materials and Methods
A typical classroom building in King Abdulaziz University–Rabigh branch has been selected as a case study for a thorough investigation of the effectiveness of natural ventilation for educational buildings. Rabigh is located in the western region of the Kingdom of Saudi Arabia on the red sea cost which features a sweltering and muggy days during summer and mild windy days during winter. The average annual wind speed is around 4 m/s while the maximum wind speed reaches up to 7.5 m/s. This prevalent outdoor wind speed commonly promotes desirable natural ventilation rates if the building is well designed to harvest and distribute the air evenly throughout the indoors. The analyses have been performed using EnergyPlus [
53] to simulate natural ventilation using Rabigh’s TMY weather file. The 3D building model and the actual prototype building are shown in
Figure 1. In particular, the Airflow-Network module in EnergyPlus has been utilized to analyze the building’s airflow distribution and ventilation rates. The Airflow-Network model can accurately simulate the airflows of multizone indoors produced from cross ventilation and stack effects, which has been validated through several experimental studies [
54]. The selected classroom building is a three-story building with a gross floor area of 2758 m
2 and an overall window to wall ratio of about 20%. Classrooms are rectangular and their dimensions are around 7 × 9 × 3 m (W × L × H) as illustrated in
Figure 2. The classrooms buildings at KAU-Rabigh have a linear disposition which enables the placement of openings along the longer side of the rooms. This is a typical and representative case study in terms of construction, typology, and integration of environmentally driven design principles, and due to its standard characteristics, it has been selected to assess natural ventilation and spread of viral infection within classrooms.
There are several known means typically employed to improve indoor air quality in buildings. However, this research study focuses on evaluating the effectiveness of natural ventilation in controlling the spread of airborne virus transmission in buildings and to assess draught discomfort caused by high ventilation rates. As stated earlier, previous studies suggest a strong relationship between ventilation rates and concentrations of particulate matter in addition to microbial and virus transmission within buildings. However, ventilation rates, in the case of natural ventilation, are directly affected by specific parameters such as wind speed, building orientation, window to wall ratio, space volume, and window opening fraction; thus, these parameters could also affect the behavior of airborne transmission of viruses and the probability of infection in buildings. Accordingly, the emphasis will be on assessing the impact of ventilation rates produced by natural ventilation on the likelihood of infection of occupants in educational buildings to eventually recommend a list of design guidelines based on selected building parameters. Specifically, general analysis of the performance of natural ventilation is performed throughout the year for the selected educational building including the impact of building orientation and window opening fraction on the air change per hour of the building. Correlation analysis is then implemented for classrooms volume, ventilation rates, window opening fraction (WOF) and ventilation rates. Furthermore, sensitivity analysis is conducted to analyze the impact of selected parameters on the probability of infection (POI) including the number of source patients, class volume, window opening fraction. For easier and efficient analysis, new ratios are introduced such as the classroom volume to the number of students ratio which will be analyzed further against the WOF to explore their impact on the POI. Based on the outlined results, correlations are developed for a selected set of variables and stepwise regression is performed to develop a fit model to predict POI through specific design parameters. Optimization is then performed to find the optimum ventilation rate, classroom volume and WOF/(volume/student) ratio that delivers the lowest POI. Furthermore, draught risk is estimated for three levels of indoor airflow turbulence intensities, and the relationship between the probability of infection, draught risk and ventilation rates is analyzed. Finally, design guidelines for educational buildings are presented to optimize the performance of natural ventilation to prevent airborne transmission of viral infection in classroom spaces while reducing the chances of occupant’s complaints from draught discomfort.