Optimization of window design in hospital rooms for effective access to daylight

Optimization of Window Design in Hospital Rooms for Effective Access to Daylight Paula Rodríguez1, * , Miguel A. Campano1, Samuel Domínguez-Amarillo1, and Ignacio J. Acosta1 1University of Seville/Department of Building Construction I, Seville, Spain Abstract. Proper access to natural light entails a multitude of consequences for human beings, making it a highly significant aspect within the hospital setting. In consequence, it is imperative to undertake an appropriate optimization of windows in architectural design to reduce energy consumption and mitigate environmental impact, while concurrently enhancing the well-being of occupants. The aim of this study is to quantify the relative effectiveness in terms of energy consumption and natural lighting of hospital room windows, analyzing how a set of key design variables ‒size, proportion, position, and orientation‒ influence in one of the primary lighting dynamic metrics, Daylight Autonomy (DA). The results indicate that it is recommended to prioritize horizontal window designs over floor-to-ceiling alternatives, allocate a minimum area of 1.20m² to the south (greater on the north side), favor central positioning on the facade, and emphasize a southern orientation for optimal illumination. 1 Introduction 1.1 Background Lighting design for hospital bedrooms environments must serve two primary objectives: the optimization of conditions for the execution of specific tasks and the creation of an atmosphere conducive to patient comfort [1–3]. Furthermore, it is imperative to achieve these aims while prioritizing the highest attainable level of energy efficiency [4, 5]. The paramount significance of access to natural light warrants accentuation, given its capacity to curtail the requirement for electric lighting [6], concurrently enhancing health [7, 8] and well-being [9–11]. Consequently, the determination of an appropriate parameter value for this indicator within interior spaces should be taken into consideration during the architectural design process [12–14]. In Spain, buildings account for 20% of electricity consumption [15], of which 28% is attributed to electric lighting. Therefore, the utilization of predictive computational models, * Corresponding author: [email protected] © The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0 (https://creativecommons.org/licenses/by/4.0/). E3S Web of Conferences 487, 02002 (2024) ICREE 2023 https://doi.org/10.1051/e3sconf/202448702002 analyzed with dynamic metric for proper window optimization [16–18], holds the potential to significantly reduce electricity consumption [19]. Traditionally, the sizing of windows in architecture was mainly driven for aesthetic considerations, disregarding quantitative criteria for optimization and energy use. However, advancements in calculation methods and the introduction of novel dynamic metrics for natural lighting have paved the way [20, 21], in recent years, a new way for enhancing window design [16], [22, 23] and lighting control strategies [21], [23–29]. Among these dynamic metrics, Daylight Autonomy (DA) stands out, allowing for greater precision in lighting calculations because it incorporates parameters such as orientation, location and sky types throughout the year [30]. Thus, this metric analyses the behavior of a given space through the annual percentage of hours of use when a minimum illuminance threshold is met by daylight alone. Consequently, utilizing DA makes it possible to effectively tailor the placement and size of windows to ensure an adequate influx of natural light, thus minimizing the need for electric lighting during daylight hours. This approach not only contributes significantly to the reduction of energy usage [31], but also enhances the visual comfort and well-being of occupants [2], [13]. It is worth noting that natural light is the most suitable source for promoting a good circadian entrainment for occupants, since the chrono-regulation of the human being has evolved according to the variable spectra of daylight [32]. Finally, dynamic metrics, such as DA, also align with sustainability goals and promoting energy-efficient practices in healthcare facility design. 1.2 Objectives The main aim of this study is to employ an established methodology [17] that utilizes the calculation of the DA metric for the assessment and enhancement of window design in hospital rooms located in Mediterranean regions. The goal is to improve the working conditions of healthcare personnel and enhance patient comfort, providing a sufficient light to carry out the task, while the circadian stimulus is promoted by a proper use of daylight. In the context of a case study, a hospital room model is situated in Seville and subjected to calculations using the Climate Studio tool. This analysis explores the impact of various variables on window design, encompassing size, proportion, position, and orientation. 2 Methodology 2.1 Calculation Protocol Fig. 1 represents the flux diagram that leads the presented research. Firstly, 3 different window shapes are selected -square, horizontal and verticalin accordance with 3 different sizes ‒12%, 22% and 32%‒. The selected openings are facing North or South, depending on the calculation model. 2 E3S Web of Conferences 487, 02002 (2024) ICREE 2023 https://doi.org/10.1051/e3sconf/202448702002 Fig. 1. Flux diagram of the calculation protocol. 2.2 Case study definition The model under study is an existing hospital room 1.80 m wide by 4.0 m deep by 2.2 m high, with the window located in one of its smaller surface walls. Window size is a variable defined as a ratio of surface in the façade. The inner surfaces have the photometric characteristics described in Table 1, considering them as diffuse reflectors: Table 1. Surface characteristics of materials used. Materials Reflectance Solar Transmittance Floor 0.60 - Walls 0.80 - Ceiling 0.80 - Window 0.90 (joinery) 0.70 (glass) The location of the virtual room chosen to perform the analysis is Seville, Spain, with Latitude 37.42°, Longitude 5.40° and mainly clears skies. Its interior is discretised by an overlaid grid of points at 0.5 m x 0.5 m intervals, positioned at a height of 0.76 m above the floor level [33], simulating what would be a plane for patient recognition, as shown in Fig. 2. From this grid, the central longitudinal section is taken for analysis. 3 E3S Web of Conferences 487, 02002 (2024) ICREE 2023 https://doi.org/10.1051/e3sconf/202448702002 Fig. 2. Hospital room model. 2.3 Selection of variables and calculation models definition The design of the window located on the exterior facade is modified according to the following variables described in Table 2: Table 2. Selected variables. Window size Window proportion Window position Window orientation 12% Horizontal Centered North 22% Vertical Descentered South 32% Combining these variables generates 16 calculation models, as shown in Fig. 3 and Table 3. 12HD 12HC 22HC 32HC 12VD 12VC 22VC 32VC Fig. 3. Calculation models. 4 E3S Web of Conferences 487, 02002 (2024) ICREE 2023 https://doi.org/10.1051/e3sconf/202448702002 Table 3. Hypotheses of calculation. Model Window/ facade Window surface (m²) Window proportion Window position Window orientation 12HDN 12% 0.66 Horizontal Descentered North 12HCN 12% 0.66 Horizontal Centered North 22HCN 22% 1.21 Horizontal Centered North 32HCN 32% 1.76 Horizontal Centered North 12VDN 12% 0.66 Vertical Descentered North 12VCN 12% 0.66 Vertical Centered North 22VCN 22% 1.21 Vertical Centered North 32VCN 32% 1.76 Vertical Centered North 12HDS 12% 0.66 Horizontal Descentered South 12HCS 12% 0.66 Horizontal Centered South 22HCS 22% 1.21 Horizontal Centered South 32HCS 32% 1.76 Horizontal Centered South 12VDS 12% 0.66 Vertical Descentered South 12VCS 12% 0.66 Vertical Centered South 22VCS 22% 1.21 Vertical Centered South 32VCS 32% 1.76 Vertical Centered South 2.4 Selecting the calculation metrics Daylight Autonomy (DA) is one of the most widely used dynamic metric. This concept was proposed in 1989 by the Association Suisse des Electriciens [34] and subsequently redefined by Reinhart [35]. It is defined as "the percentage of occupied time in a space during which the interior natural light reaches the level of illuminance necessary for conducting required tasks" [35].Therefore, the higher the value of DA, the less time spent relying on electric lighting to compensate for lighting deficiencies. This metric is based on numerous sky models, requiring knowledge of the climatic data of the building's location and performing intricate calculations. This requires the use of lighting simulation programs, which calculate the value for each point. DA is mathematically defined by the following equation: 5 E3S Web of Conferences 487, 02002 (2024) ICREE 2023 https://doi.org/10.1051/e3sconf/202448702002 Where ti is the occupied time in the year, wfi is the weighting factor which depends on the illuminance threshold, ED is the daylight illuminance measured at a given point under real sky conditions, and ER is the illuminance threshold. 2.5 User requirements The illuminance threshold (ER) for the DA calculation of this study is established at 500 lux, a recommended threshold for reading and patient recognition [1]. The occupancy period with daylight hours is defined from 8:00 to 17:00 from Monday to Friday, without breaks, including Daylight Saving Time (DST). 2.6 Selecting the calculation and validation tool The selected tool for developing the simulations is Climate Studio for Rhino, a dynamic lighting calculating software based on the Radiance calculation engine created by C. Reinhart [35, 36]. Rhinoceros 7.0 has been used to create the 3D model, as it is the interface that supports Climate Studio. Table 4 shows the calculation parameters used by this programme in this study. Table 4. Parameters of calculation. Parameters Value Parameters Value Ambient Bounces 7 Ambient Divisions 1500 Ambient Super-samples 100 Ambient Resolution 300 Ambient Accuracy 0.05 Limit Reflection 10 Specular Threshold 0.0000 Specular Jitter 1.0000 Limit Weight 0.0040 Direct Jitter 0.0000 Direct Sampling 0.2000 This tool, along with its calculation engine Radiance [37], has been previously validated in various studies [38, 39], both through the CIE Test Cases [40–45] and by comparing it with annual lighting values obtained from monitoring an existing test cell located in Seville, designed for emulating a bedroom [20], [46, 47]. 6 E3S Web of Conferences 487, 02002 (2024) ICREE 2023 https://doi.org/10.1051/e3sconf/202448702002 3 Calculating and analysis of result The values obtained in each model for Daylight Autonomy (DA) are shown in Table 5, north-oriented, and Table 6, south-oriented. It should be noted that sun shading devices or remote obstructions have not been considered when running the simulations. Table 5. Values of DA for north-oriented models in the central axis. distance from facade (m) 12HDN 12HCN 22HCN 32HCN 12VDN 12VCN 22VCN 32VCN 0.61 0.66 0.88 0.92 0.93 0.28 0.76 0.89 0.92 1.12 0.51 0.73 0.88 0.91 0.25 0.40 0.83 0.89 1.63 0.30 0.44 0.79 0.88 0.02 0.11 0.73 0.85 2.14 0.05 0.11 0.73 0.83 0.00 0.00 0.61 0.78 2.65 0.00 0.00 0.57 0.80 0.00 0.00 0.50 0.77 3.16 0.00 0.00 0.47 0.75 0.00 0.00 0.45 0.71 3.67 0.00 0.00 0.45 0.73 0.00 0.00 0.34 0.69 Table 6. Values of DA for south-oriented models in the central axis. distance from facade (m) 12HDN 12HCN 22HCN 32HCN 12VDN 12VCN 22VCN 32VCN 0.61 0.86 0.90 0.93 0.93 0.78 0.86 0.90 0.92 1.12 0.82 0.83 0.90 0.92 0.75 0.77 0.86 0.90 1.63 0.77 0.77 0.86 0.89 0.69 0.72 0.83 0.87 2.14 0.69 0.76 0.84 0.87 0.64 0.69 0.79 0.85 2.65 0.67 0.65 0.81 0.85 0.52 0.57 0.77 0.84 3.16 0.65 0.59 0.76 0.84 0.50 0.51 0.75 0.83 3.67 0.53 0.53 0.76 0.82 0.48 0.50 0.71 0.82 Models oriented towards the north yield lower values compared to those oriented towards the south, exhibiting an increase ranging between 1% and 4%. In cases where the window area is enlarged, a corresponding rise in the value of this indicator has been observed. This enhancement is noticeable both at the position nearest to the window and at the rear of the room, showcasing an increase ranging from 12% to 35%. Thus, with window-to-wall ratios of 12% facing south (1.2 m²) and 22% facing north, it becomes possible to achieve DA values above 50% throughout virtually the entire central axis of the space under study. The proportion of the window also influences the outcomes. In north- 7 E3S Web of Conferences 487, 02002 (2024) ICREE 2023 https://doi.org/10.1051/e3sconf/202448702002 oriented models, a horizontal window configuration results in improved light ingress by 3% to 38%, while in south-oriented models, the enhancement ranges from 2% to 16%. Fig. 4 and Fig. 5 show the evolution of each model as the end of the room is reached, enabling a comparison of each calculation model and how the selected variables impact the final value of DA. Fig. 4. Evolution of DA by depth. North-oriented models. Fig. 5. Evolution of DA by depth. South-oriented models. These graphs reflect the relevance of orientation in daylighting, as well as which variable has a significant impact in each orientation. North-oriented models (Fig. 4) generally exhibit lower DA values, both at the point closest to the window and at the back of the room, sometimes even falling below the threshold. Conversely, south-oriented models (Fig. 5) yield more consistent values throughout the room. 8 E3S Web of Conferences 487, 02002 (2024) ICREE 2023 https://doi.org/10.1051/e3sconf/202448702002 3.1 Study of limitations Considering the existing constraints of this study, it is recommended that forthcoming investigations be undertaken. These studies should prioritize the augmentation of the number of globally representative locations to enhance the applicability of the findings. Furthermore, there is a need to broaden the range of room typologies, occupancy schedules, and lighting thresholds. It would also be beneficial to introduce factors such as remote obstructions to assess their impacts on the daylight access. Moreover, an important aspect not addressed is the evaluation of the maximum window-to-wall ratio. Consequently, it would be of interest to conduct a comprehensive multi-parameter analysis incorporating metrics related to glare and thermal energy consumption. This analysis would provide insights into determining the optimal window- to-wall ratio. Lastly, the deployment of a lighting monitoring system in existing building scenarios holds the potential to furnish empirical data for the purpose of validating the simulationderived outcomes. 4 Conclusions The comparison of the different hypotheses of calculation allows us to understand which variables most influence the obtained results. South-oriented models yield higher values than north-oriented models, by a range of 1 - 4%. With a larger window area, it has been observed that this indicator's value increases, both at the point closest to the window and at the back of the room, by 12 - 35%. The proportion of the window also influences the results, with a horizontal window improving light access by 3 - 38% in north-oriented models and by 2 - 16% in southoriented models. This study concludes the following recommendations regarding window design: • Prioritize horizontal designs over floor-to-ceiling alternatives. • A minimum area of 1.20m² to the south (greater on the north side). • Favor central positioning on the façade rather than off-centre. • Emphasize a southern orientation than northern for optimal illumination. Acknowledgements The results presented were funded by the "Agencia Estatal de Investigación" (AEI) and the "Ministerio de Ciencia e Innovación" (MCIN) of the Government of Spain through the research projects CHRONOLIGHT: Biodynamic wide spectrum lighting for biological chronoregulation and pathogens neutralization in hospital facilities (Ref PID2020- 117563RB-I00) and NEUROLIGHT: Efficient design for biodynamic lighting to promote the circadian rhythm in hospital facilities (Ref PDC2021-120807-I00). References 1. Comité Español de IIuminación., Instituto para la Diversificación y Ahorro de la Energía, and G. de España. Ministerio de Ciencia y Tecnología, Guía Técnica de Eficiencia Energética en Iluminación. Hospitales y Centros de Atención Primaria. Madrid (Spain): Instituto para la Diversificación y Ahorro de la Energía, (2001) 9 E3S Web of Conferences 487, 02002 (2024) ICREE 2023 https://doi.org/10.1051/e3sconf/202448702002