Parametric Optimization and Assessment of Modern Heritage Shading Screen for a Mid-Rise Building in Arid Climate: Modernizing Traditional Designs

Abstract

The construction domain in the Middle East region has experienced significant growth in recent years. In the United Arab Emirates (UAE), for instance, the number of newly constructed areas with mixed-use development buildings is increasing yearly. Given the region’s harsh climatic conditions, characterized by an extremely hot climate, designing sustainable and energy-efficient buildings is crucial. Under these conditions, shading is a primary strategy. This research explores how parametric design and optimization based on genetic algorithms (GAs) can improve shading structures to reduce solar radiation and lower cooling energy consumption. It is focused on parametric design and optimization of window shading structures, either for retrofitting existing buildings or for new construction. Advanced tools are essential to achieve this goal, as they enable designers to incorporate various architectural features that enhance energy efficiency. The parametric structures are meant to increase the shaded parts by reducing the amount of solar radiation on building facades, reducing the amount of energy consumption for air conditioning and improving overall energy efficiency. The methodology follows the following steps: selection of the case study, weather analysis, modeling and simulation, parametrization process (maximize shaded area and reduce total solar radiation), results, and analysis. The modeling, simulation, and parametrization are completed using Rhino/GH. While the research focuses on a mid-rise building in Abu Dhabi as a case study, the approach can be applied more broadly to buildings in hot climates. Due to excessive solar radiation in arid climate regions, shading of fenestration is a primary focus. The findings show that the GA-based optimized shading system, based on the cumulative radiation, decreased the total radiation amount by 19% and reduced cooling energy use by 26.2% for the case study. This demonstrates that parametric design can contribute to more sustainable and energy-efficient architecture.

1. Introduction

Energy efficiency in buildings is a global concern. Worldwide, buildings are responsible for over one-third of global energy consumption and associated emissions, making their optimization critical for sustainability [1]. Energy efficiency is particularly critical in the Gulf Cooperation Council (GCC) countries due to the extreme climate and rapid urbanization. The building sector in these countries accounts for a substantial share of energy consumption, with cooling demands driving high electricity usage [2].

The UAE buildings consume approximately 90% of the country’s total electricity [3]. Further, the residential building stock in Abu Dhabi, the capital city of UAE, recorded the highest energy consumption rates. Abu Dhabi buildings’ cooling loads are higher than other cities in the UAE [4]. Energy demands are a major concern related to the architecture industry in the UAE. Existing residential and commercial buildings account for 40% of the total energy consumed by the existing stocks as electricity [3]. Due to the extremely hot weather conditions in the UAE, HVAC equipment generates the highest electrical load, which accounts for 40% of the overall electrical load throughout the year and up to 60% of the summer months’ peak electrical load [5]. According to a study, residential buildings comprise the dominant part of the building sector in the UAE and account for the most significant energy usage [6]. The residential sector consumes nearly a quarter of all global electricity, in addition to the GCC countries’ residential sectors, which are among the most energy-intensive in the world [7]. According to Al Rais, cooling demand accounts for approximately 50% of total electricity consumption and 70% of the peak electricity demand in Abu Dhabi. Abu Dhabi has one of the world’s highest per capita electricity consumption rates. Much of this consumption is attributed to air conditioning (AC) in the residential sector [8]. In the UAE, the cooling loads account for 47% of the total residential building electricity demand, and this can increase to above 60% during the summer peak, and “residential buildings in the UAE consume 18.3 TWh electric energy for cooling” [9]. Building envelopes’ exposure to solar radiation causes high air temperatures inside the structures [10]. Building envelope characteristics greatly affect heat gain in buildings.

This research explores how parametric design and optimization can improve shading structures to reduce solar radiation and lower cooling energy consumption.

1.1. Parametric Architecture

Parametric architecture design is a design approach in which aspects like building elements and technical components are formed by algorithmic procedures rather than directly by manipulation. Parameters and guidelines in this method define the link between the design goal and design performance [11]. Parametric modeling is a “Process of developing a computer model or describing a design problem. This representation is based on relationships between objects controlled by variables. Making changes to the variables results in alternative models. Selection of a solution is then based on some criteria related to performance, ease of construction, budget requirements, user needs, aesthetics, or a combination of these” [12].

Parametric design is based on an algorithmic approach, the algorithmic output is edited by the continuous variation in the input parameters [13]. This means that an algorithm is a container of variable instructions. On its own, without input, it is considered only information. When the algorithm is fed up with input parameters, it transforms the parameters according to its instructions and generates new data known as the output. The application of computational tools in architectural design implies the use of parametric relationships, autonomous processes, and algorithms to generate designs that require little human interaction [14,15]. Parametric design generates relationships and connections among every component of the design. When one of them is modified, the other elements adapt to the modification as well, generally by automatically changing parameters or related values, as in a system of equations [16].

1.1.1. Parametric Optimization

Optimization techniques have become essential in designing energy-efficient buildings by helping architects and engineers fine-tune design parameters to reduce energy consumption. Researchers can evaluate different design options by combining simulation tools with optimization algorithms and find the most effective solutions [17]. Among these methods, evolutionary algorithms like genetic algorithms (GAs) have proven highly effective in solving complex design challenges, such as improving shading systems, window placement, and facade performance to lower cooling demands [18]. Inspired by natural selection, these algorithms explore multiple design possibilities, making them particularly valuable in energy-efficient architecture [19]. A novel technique integrating energy simulation and optimization evolved. This approach has been given new names, such as “Computational optimization” [20], “Simulation-based optimization” [21], “Building performance optimization”, “Performance-driven design” [22], etc. Energy-efficiency-based design optimization for buildings is obviously a growing approach that is being actively researched. This technique uses optimization algorithms to develop new designs based on simulation findings and user-specified design goals [23].

1.1.2. Parametric Optimization Significant and Relevant

Architecture and building design are methods of creation that consider functionality, aesthetic principles, and structural stability [24]. Determining the forms, sizes, and orientation of the various elements of a built environment affects the total cost of construction, illumination use, energy consumption, heating and cooling loads, shading efficiency, acoustics, efficient accessibility, and solar gains, among many other features [25]. Considering this, determining the best forms, orientation, and size of built environment elements is essential in the design process since the outcomes affect the built environment’s life cycle. In such a scenario, it is critical to formulate design components and parameters that consider the broad range of context parameters. These challenges have made it advantageous to use computational design optimization methods.

1.2. Relevant Studies

Recent advances in computational tools for design applications, along with ideas from artificial intelligence, have created new opportunities for computers to guide and actively engage in design processes. Caldas and Norford [26] proposed a tool that leverages the principles of generative and goal-oriented design to assist architects in generating and evaluating specific features of a solution to maximize the final configuration’s behavior. The study’s specific issue is the location and size of windows in an office building. Genetic algorithms (GAs) are used to generate and seek optimal design solutions for a building’s lighting and thermal performance. The results demonstrated that utilizing GA-based optimization significantly enhanced daylight access, while reducing thermal loads. Similarly, Tsangrassoulis et al. [27] investigated a genetic algorithm to optimize opening areas, glazing characteristics, and shading arrangements, seeking minimum energy consumption, with the optimized properties showing a measurable reduction in cooling energy demand. Rapone [18] attempted to optimize the sizes for the window elements and shading devices to minimize CO₂ emissions, considering the heating, cooling, and artificial lighting loads and utilizing single- and multi-objective optimization. To minimize the artificial lighting load, cooling load, heat gain, and glare, as well as to maximize the daylight metrics as objective functions, the author utilizes the standard genetic algorithm in the MATLAB (2010; R2010a) software environment; the results show that a comprehensive approach to facade design can yield a significant reduction in energy consumption and emissions. Rakha and Nassar [28] used a GA in optimizing ceiling geometry in terms of natural lighting provision. Ceiling geometry optimization significantly improved daylight penetration. While their focus was on ceiling geometry, the results highlight the broader potential of computational design tools to optimize building components for passive energy solutions.

Several studies have explored integrating parametric design, building simulation modeling, and genetic algorithms (GAs) to optimize energy performance in architecture. Fang and Cho [29] developed an optimization process that assessed daylighting and energy performance in small office buildings. Their findings demonstrated that optimized designs improved daylighting performance by 38.7%, 31.6%, and 28.8%, while energy consumption decreased by 20.2%, 18.5%, and 17.9%. This study utilized Rhinoceros, Grasshopper, and their plug-ins Ladybug and Honeybee for daylight and energy modeling. Turrin et al. [30] utilized a GA in combination with parametric modeling to identify the optimum parametric design alternatives for a roof structure based on its structural performance. The results demonstrated that GA-based optimization effectively enhanced roof stability while minimizing material usage, leading to more efficient and cost-effective architectural solutions. GA-driven form-finding, in their study, supports the validity of using similar computational strategies in optimizing shading geometries for energy-efficient design. Similarly, Zhang et al. [31] designed a free-form construction to receive more solar radiation by optimizing the shape. Rhinoceros and Grasshopper were utilized for the free-form building model. The optimization was conducted using a multi-objective genetic algorithm considering maximizing space efficiency, solar radiation gain, and minimizing the shape coefficient. The optimization process resulted in building forms that improved thermal performance, reduced material waste, and enhanced energy efficiency. Tuhus-Dubrow and Krarti [32] utilized MATLAB (2008; R2008a) to develop a simulation optimization tool. The tool is utilized to optimize the building shape and envelope features. The simulation–optimization tool was developed by a combination of a genetic algorithm and the energy simulation engine DOE-2, to select the optimal values of a comprehensive list of parameters associated with the envelope to minimize energy use for a residential building. Their findings showed that GA-based optimization effectively reduced total energy consumption by refining key parameters such as building orientation, window-to-wall ratios, and insulation levels, demonstrating that GAs can be successfully used to enhance envelope performance. A study by Manzan and Pinto [33] aimed to optimize the shading device design of an office building, considering heating, cooling, and lighting energy consumption. Their study was conducted utilizing the mode FRONTIER generic optimization platform. Their study demonstrated that optimized shading systems significantly reduced heating, cooling, and lighting energy demand, reinforcing the role of parametric shading solutions in improving building energy performance. Evins et al. [34] attempted to optimize building solar gain. The tool used for the optimization process is the genetic algorithm in the MATLAB toolbox. Their study confirmed that GA-driven design approaches can generate high-performance solutions that reduce energy consumption. Glassman and Reinhart [35] optimized facade parametric design utilizing Grasshopper linked with Energy Plus for simulation. The authors highlight that combining parametric design software with simulation programs provides the opportunity to create energy-efficient building designs and components. Karaman et al. [36] optimized daylighting through the facade design utilizing an evolutionary algorithm in the Grasshopper tool set; their study showed that parametric optimization could significantly enhance indoor daylight quality while maintaining energy efficiency. This supports the current research, as it demonstrates that GA-driven parametric design can successfully optimize shading elements for improved energy performance and occupant comfort.

The reviewed studies show that parametric design and genetic algorithms are highly effective in making buildings more energy efficient. They highlight how optimized shading systems can improve daylight use, reduce cooling needs, and lower energy consumption. A key takeaway is that balancing factors like daylight, heat, and comfort leads to more innovative and more sustainable designs. This reinforces the idea that parametric shading can be crucial in reducing solar radiation and energy use, especially in hot climates, making evolutionary algorithms a valuable tool in modern architecture.

2. Methodology

The methodology is based on a case study investigation and follows the steps below, where the case study is described, followed by a weather analysis that considers the climate conditions that impact this research. The modeling and simulation enter into the tool’s initial work environment, where the design is defined, and the parameters are created. The results are then explained based on all the previous analyses.

2.1. Case Study Selection

2.1.1. Abu Dhabi City Climate Profile

The UAE climate is a hot and arid climate due to the high sunlight and long daytime hours. Summertime runs from April to September, with heat in coastal cities reaching 50 °C and humidity readings often exceeding 90%. The average temperature in summer in Abu Dhabi ranges from 28 to 36 degrees Celsius, while the average winter temperature ranges from 17 to 27 degrees Celsius. From June to September, temperatures typically exceed comfort levels [8].

The UAE’s extreme climate makes building energy efficiency and environmental sustainability difficult. The temperature (a wet bulb design temperature of 30.6 °C and a dry bulb design temperature of 45.0 °C) and humidity (80% in winter and 70% in summer), combined with the high solar irradiance (the annual average global horizontal irradiance is over 20 MJ/m²/day), necessitate the use of air conditioning to maintain acceptable indoor comfort levels [37].

2.1.2. Selected Building Analysis

This study refers to Abu Dhabi’s downtown mid-rise building stock that was built before 2011, when the guidelines of the Estidama rating system began to be applied [38]. The selected case study building was chosen mainly due to the availability of stable data to achieve the study’s objective. The case study, Building A, is a mid-rise residential structure with five residential standard levels, a roof floor, a mezzanine floor, and a ground floor that contains entrance and commercial shops. The total area of the building is 3191 m². The building’s height is 23.85 m. It is in the Al Markaziyah area in downtown Abu Dhabi. Figure 1 illustrates the case study’s typical floor plan, southwest elevation.

One unit from the case study building is simulated in terms of energy consumption in the form of electricity. In addition to solar radiation simulation for the full building envelope, these simulations facilitate achieving the objectives and targets of this research. The apartment AP1 area is 91 m², consisting of 2 bedrooms, 2 bathrooms, a living room, and a kitchen. The apartment is located in the southwest part of Building A. The bedrooms and living room windows are located on the southwest elevation with a 42% window-to-wall ratio, while the kitchen window is located on the northwest elevation with a 9% window-to-wall ratio. Figure 2 illustrates the apartment AP1 floor plan.

Table 1. Case study building details.

Technical Details	Value
Area	91 m2.
WWR	Window to wall ratio in the south elevation is 46%
Construction Details	External wall: 20 mm plaster outside + 200 mm concrete block + 20 mm plaster inside U-value: 2.6 W/m2 K Windows: Single panel glass U-value: 5.3 W/m2 K Internal walls: 100 mm concrete + 20 mm plaster inside and outside Floor: 200 mm concrete slab, 50 mm sand, and 10 mm ceramic tiles
HVAC Details	VRV central air conditioning system, Coefficient of Performance (COP) = 2.02 Cooling average set point = 22 °C Ventilation system ACH: 1.8 ACH
Lighting	Lighting power density (LPD) = 1.37 W/m2
Electrical Appliance	Lowest power density = 9.17 W/m2 Highest power density = 21 W/m2
Occupancy Details	Occupancy Ratio 0.04 people/m2
Location and Climate Details	Abu Dhabi, UAE; Arid Climate. (EPW file: ARE_AZ_Abu.Dhabi-Bateen.Exec.AP.412160_TMYx.2007) 2021.epw

shows the case study building details.

2.2. Case Study Modeling and Simulations

The simulation was conducted utilizing Rhinoceros 7.0 software with the Grasshopper plug-in. Based on open-source VP languages, Grasshopper is used to generate complex geometrical forms, parametric designs, and environmental and performance simulations of various features.

2.2.1. Optimization Tool Selection

Special-purpose optimization platforms like Grasshopper optimization plug-ins require relatively less computer coding [33]. Based on this, the optimization tool for Grasshopper, which is based on the genetic algorithm, was selected to be the optimization tool. Grasshopper for Rhinoceros was selected to be the parametric design tool. However, Grasshopper software 1.3.0 interacts with the Open Studio package, which includes the Energy Plus simulation engine. Energy Plus was utilized for simulations in this study, and Grasshopper optimization plug-ins for Galapagos were utilized for optimization.

The Galapagos plug-in for Grasshopper is an evolutionary solver that finds answers to complicated design challenges [39].

2.2.2. Design of the Parametric Structure

The parametric pattern designed for this research is inspired by traditional embroidery on traditional tents in the UAE. The inspired lines applied on the grid were created based on the traditional Islamic pattern drawing rules (radial around point, symmetry, rotation, purification, mirror, and repetition), as shown in Figure 3.

Parametric shading based on Islamic patterns was chosen to be the shading device type to be applied and optimized in this study, among the other shading devices, as it is an advanced shading method that achieves both environmental and social sustainability by incorporating the identities and cultural elements of the building design in the Arab and Islamic region. Parametric shading have roots in the regional architecture represented by Mashrabiya, Orosi, and Moshabak. Islamic architecture used those sustainability-based elements to deal with environmental constraints in various areas and climates throughout the Islamic world [40,41]. The previously mentioned elements were designed based on Islamic geometric patterns. Islamic geometric patterns are elements with a strong dependence on geometric shapes and mathematical relations [42,43]. The pattern can be considered an Islamic pattern if it is derived from Arabic calligraphy or traditional patterns [42]. Geometric patterns have four main characteristics: interlacing, symmetry, unboundedness, and flow [43].

The parametric design was created using a fundamental unit inspired by the traditional needlework of the Arabic tent. The unit was mirrored on a different axis and repeated on the x-axis and y-axis, and then the pattern was modeled as shown in Figure 4.

Grasshopper for Rhino was utilized to model the parametric pattern using a visual script. The script was created with a defined range of parameters for the pattern to serve research purposes. The depth of the pattern (Pd), repetition of the units on the x-axis (n) and y-axis (m), and area of the perforation controlled by the offsets of the panel (of.1, of.2) settled in the visual script as variables with defined ranges, so they could be changed during the optimization prosses until reaching the optimum value for each to maximize the shaded area and reduce the total solar radiation on the elevation and, accordingly, reduce cooling loads. Figure 5 visualizes pattern parameters, and Figure 6 shows the way it was created and controlled using Grasshopper visual script.

The parametric script applied to the case study of this research, Building A, initially with random values of the parameters. A radiation analysis was completed to explore the general effect of the parametric pattern on solar radiation. The total yearly radiation on the building envelope before including the parametric pattern as calculated by the total radiation analyses script was 455,822.90 kWh, while the total radiation on the building after applying the parametric pattern with random values for the parameters on the typical floors of Building A at the southwest facade was calculated as 403,722.30 kWh. The reduction in the total radiation is 12%. The results are illustrated in Figure 7 and Figure 8.

For the optimization, according to the literature, the parameters of the parametric shading devices are similar to the ones in this research. The structure parameters that affect solar radiation and energy consumption are the following: The perforation ratio, defined as follows: “Perforation Ratio (PR): The percentage of the perforation opening area to the whole area of the screen” [44]. The depth ratio, defined as follows: “Depth Ratio (DR): The ratio between the depth and the width of each perforation opening” [44]. The gap width between the building face and the shading device. This study will focus on the PP and DR. At the same time, the gap width between the building faced and the shading device will be constant at 100 cm based on the recommendation of the Abu Dhabi municipality to enable facade maintenance and avoid heat traps. The target value of this research is to reduce energy consumption by a ratio close to 30%, as it is the highest ratio of reduction found in the reviewed literature related to shading devices like the one optimized in this study.

Table 2. Recommended practice based on the literature review.

Recommendation	Resources
(1) The design should use light colors on facades with high solar exposure, to increase solar reflectivity and reduce solar heat gain.	[45]
(2) The materials, textures, and colors used for the fenestration elements (windows, doors, openings) shall create a subtle contrast and visual richness in the facade.	[45]
(3) The materials used shall prevent glare and reflection.	[45]
(4) Responding to the local cultural values, such as the valuation of privacy for certain indoor spaces.
(5) Parameters of the shading device:    - The gap width between the building face and the shading device: 40 cm–100 cm (recommended to be 1 m to enable maintenance and to avoid trapped heat)    - The perforation ratio: between 50 and 70% to provide acceptable rates of daylighting and visual connection. “Perforation Ratio (PR): The percentage of the perforation opening area to the whole area of the screen” [44].    - Depth ratio: 0.7–1.25    The impact was insignificant with increased depths (starting at 40 cm). “Depth Ratio (DR): The ratio between the depth and the width of each perforation opening” [44].	[45,46,47]
	[44,46,47]
	[44,46]
(6) The perforation ratio for the traditional parametric shading (Mashrabiah) applied on the privet residence was between 8% and 40%, starting with a small ratio on the ground floor, increasing on the upper floor, and possibly reaching 60% on the upper floors. For public buildings, the average perforation ratio was 60%.	[40]
(7) The external fixed perforated parametric shadings could achieve energy savings of up to 30% of the total energy consumption in the west and south orientations.	[48]
(8) “Best practice in the UAE shows its most energy-efficient buildings consuming 110–160 kWh/m2/year”.	[49]

summarizes the recommended practice for shading devices identical to the one targeted by this study based on the literature review. For the variable sliders, which control the PP and DP, a range of minimum and maximum values was defined based on the literature review to ensure adequate daylighting values and a comfortable visual connection. As shown in Table 2.

2.2.3. Optimization Script

In most cases, an optimization process requires two types of input: variables and objective functions. Variables are the values that control the properties of the building design, while the objective function is a mathematical expression that defines the optimization goal. It quantifies the performance or quality of a solution by mapping input variables (design parameters or decision variables) to a single scalar value, representing either a cost, benefit, or performance measure.

The objective functions in this study are the building performance metrics that are typically calculated by simulation tools [17]. Figure 9 illustrates the basic framework for the optimization tool in this research; x represents the parametric pattern metrics (variables), y represents the targeted building details set, and finally, z represents the objective functions. However, to maximize the benefit of the parametric structure, in reducing the total radiation on the building, we utilized, respectively, the cooling electricity demand, which is one of the objectives of this research, an optimization script created using a Grasshopper plug-in for Rhino and the Galapagos evolutionary optimization solver for Grasshopper, along with the radiation, energy consumption, and parametric pattern scripts that were developed during the previous stages of this research.

The Galapagos plug-in has an input node called the genome, which receives the set of variables for the parametric design to be differentiated and solved. In this study, these variables are represented by parametric pattern metrics. In addition to that, the Galapagos plug-in interface has another node called fitness, which receives the objective function for maximization or minimization; Figure 9 illustrates the plug-in interface.

In the previous section, the variable parameters were defined to be PP and DP based on the literature review. To define the objective function, three different trials for creating the optimization VP script were conducted in this research by changing the objective function in each trial as follows:

• Optimization scenario with annual cooling loads as the objective function.
  The first trial was creating the optimization script based on the energy prediction model. The scripts include the energy consumption simulation script, which was developed in other research related to this study. In the energy simulation model, the data are fed into the Honeybee (HB) tools to create an HB model with all the needed details. Then, the HB model transfers the model to recognizable data by Energy Plus to run the simulation and obtain the results. In the optimization case, the energy model is connected to the parametric structure pattern through the shading device tool from HB, while the optimization plug-in Galapagos is connected to the total energy consumption, which is executed using the HB Openstudio tool. The fitness of Galapagos is connected to the energy model’s total value, while the genomes are connected to the shading device variable.
• Optimization scenario with cumulative radiation as the objective function.
  This type of analysis is a method to calculate the total annual radiation for the building, taking into consideration the specific characteristics of the building material and the window-to-wall ratio.
• Optimization scenario with solar radiation as the objective function.
  This type of method calculates the total annual radiation for the building without taking into consideration the specific characteristics of the building material. The goal of the scripts is to minimize the objective function. The eligibility for each script was tested by running the script using Building A model inputs and apartment Ap1.

3. Results

3.1. Results of the Optimization Scenarios

3.1.1. Annual Cooling Load-Based Optimization Scenario

Running this script trial shows low performance and demonstrates that it is a difficult method to implement. This is due to the very long time it takes to generate the solutions using this script. Another trial for this scenario was conducted; in trial 2, the simulation period was limited to 1 month instead of the annual period; however, the optimization process stopped before reaching the optimum value.

3.1.2. Cumulative Radiation Analysis-Based Optimization Scenario

The optimization script created based on this objective function gives adequate results. The running trial on Building A generated 100 different design solutions for the parametric pattern, simulated them, and demonstrated the optimum design.

Comparing the radiation analysis results for Building A before adding the parametric pattern and after adding the optimized based on the cumulative radiation pattern on the southwest elevation decreased the total radiation amount by 19%. The PR for the optimum parametric structure produced by this scenario was 63%, while the DR was 0.9.

3.1.3. Radiation Analysis-Based Optimization Scenario

The optimization script created based on this objective function running a trial on Building A generated 100 design solutions for the parametric pattern, simulated them, and proposed the optimum design. Figure 10 shows the script.

Comparing the radiation analysis results for Building A before adding the parametric pattern and adding the optimized radiation-based pattern decreased the total radiation amount by 17.8%. The PR for the optimum parametric solution was 65.3%, and the DR was 0.87.

Table 3. Optimization script creation trials.

Objective Function-Based Optimization Script	Script Eligibility	Application in Building A Test Results	Input Data
Annual Cooling Load	Script test took 14.45 h to generate 3 design solutions. No optimum value resulted in the first trial. For the second trial, 36 h to generate 34 solutions, and the system stopped before reaching the optimum solution.	No optimum result was generated.	Building location Orientation Construction details Occupancy ratio Lighting ration Appliance Geometry Area Thermal zones EPW Airtightness
Annual Cumulative Radiation	Script test took 3.40 h to generate 100 design solutions and pick the optimum among them.	The optimum design pattern reduces the cumulative radiation by 19%.	Geometry Construction details Window-to-wall ratio EPW Location Orientation Context
Annual Radiation	The Script test took 1.50 h to generate 100 design solutions and pick the optimum among them.	The optimum design pattern reduces the radiation by 17.8%.	Geometry EPW Location Orientation Context

summarizes the results of the optimization script creation approach trials.

Based on Table 3, three final optimization scripts were created; the first one was based on the annual cumulative radiation, eligible to be used in the cases where the building dataset is available and easy to access, while the second script, which was created based on the annual radiation, is eligible to be used in the cases where the dataset is limited and difficult to access. The last one, the energy simulation-based script, needs detailed input in addition to higher-specification devices to run the optimization in an adequate time frame.

3.2. Optimization Layers

Since the data for the case study is available, the cumulative radiation-based script was applied to optimize the pattern for reducing the solar radiation on the building envelope in different layers:

3.2.1. Optimization Layer 1

Optimization of pattern metrics based on the cumulative radiation. As previously mentioned, the optimized solution based on a cumulative radiation pattern decreased the total radiation amount by 19%. In contrast, applying the resulting optimized pattern to Building A decreased the annual energy consumption for one apartment by 26.2%, as the loads decreased from 28,191.8 kWh to 20,777.7 kWh.

3.2.2. Optimization Layer 2

Optimization of the pattern metrics to achieve adequate daylighting and visual connection levels. Referring to Table 2, a perforation ratio of 50–70% is recommended to provide adequate daylighting and visual connection. To ensure the proper daylight and visual connection levels are achieved, the maximum and minimum range for the parametric pattern panel offsets metric sets between 3 and 7, as shown in Figure 11.

3.2.3. Optimization Layer 3 Material Selection

Lightweight aluminum with high reflectivity and light neutral color coating is selected as a material choice for the parametric structure, as it is lightweight, recyclable, and available at affordable prices. The material chosen refers to Table 2.

Figure 12 shows samples from the optimization of the parametric pattern results after setting the maximum and minimum ranges for the parametric pattern metrics in the cumulative radiation-based model.

3.3. Impact of the Optimized Shading Structure on Solar Radiation Reduction

The application of the optimized parametric shading structure significantly reduced the total annual solar radiation on the building facade. The comparison between pre- and post-optimization radiation levels is summarized in

Table 4. Solar radiation reduction.

Scenario	Total Solar Radiation (kWh)	Reduction (%)
Baseline (no shading structure)	455,822.90 kWh	—
Random parametric pattern (initial test design)	403,722.30 kWh	12%
Optimized shading structure (final design)	368,211.50 kWh	19%

, highlighting a 19% reduction in solar radiation.

The optimized shading pattern effectively reduced direct solar exposure on the southwest facade, the most critical orientation in hot climates. This decrease in solar heat gain directly influences the cooling demand of indoor spaces, as detailed in the following section.

3.4. Energy Consumption Reduction by the Optimized Shading

The annual cooling energy consumption of a single apartment (AP1) was analyzed before and after applying the optimized shading structure.

Table 5. Energy consumption reduction.

Scenario	Annual Cooling Energy Consumption (kWh)	Reduction (%)
Baseline (no shading structure)	28,191.8 kWh	—
Optimized shading structure (final design)	20,777.7 kWh	26.2%

presents the results, showing a 26.2% reduction in cooling energy demand, demonstrating the shading structure’s effectiveness in improving building energy efficiency.

The reduction in energy consumption translates into lower operational costs for air conditioning and contributes to sustainability goals by reducing carbon emissions associated with electricity use. Given that HVAC systems account for 47–60% [9] of electricity demand in UAE residential buildings, even moderate reductions have a significant impact on energy efficiency.

In summary, the final script for the genetic algorithm-based advanced optimization set is as follows:

(a). The script’s input parameters are defined as the targeted building dataset, including the building location, geometry, orientation, context, wall-to-window ratio, building envelope materials, and EPW file. The advanced script receives data using HB model tools.

(b). Variable parameters: Parametric pattern metrics, which are set to maximum and minimum values that ensure a 40–70% perforation on the final optimized pattern to ensure adequate values for daylighting and visual connectivity. The advanced tool receives the metrics through the genome node on the Galapagos evolutionary solver.

(c). The objective function for minimization: The annual cumulative radiation on the building. The tool receives the objective function value through the fitness node on the Galapagos evolutionary solver.

This study confirms that parametric shading structures optimized using evolutionary algorithms can significantly reduce energy consumption in hot climates. Compared to traditional fixed shading devices, the AI-driven parametric approach allows for greater adaptability by generating over 100 design solutions and selecting the optimal one.

The final optimized perforation ratio (PR) of 63% and depth ratio (DR) of 0.9 align with best-practice recommendations (PR: 50–70%) [44,46,47], ensuring both energy efficiency and visual comfort. The methodology demonstrates that shading structures can be customized based on climatic and urban conditions, improving adaptability for different locations.

4. Discussion

This study elucidates the significant impact of parametric optimization on enhancing energy efficiency in buildings situated in harsh climates, specifically in Abu Dhabi. By leveraging advanced computational tools such as Rhino and Grasshopper, the research developed and optimized a Mashrabiya shading screen that reduced solar radiation exposure on building facades by 19%. This reduction in solar gain corresponded to a substantial decrease in energy consumption for air conditioning, culminating in a 26.2% reduction in annual energy use for a single apartment.

The study used evolutionary algorithms, namely, genetic algorithms, to explore the complex design environment and find the best solutions that balance aesthetic, functional, and environmental performance criteria. The combination of parametric development and optimization tools allowed for a more comprehensive and effective investigation of design alternatives, outperforming the old trial-and-error method. This study emphasizes the potential of computational design methodologies in addressing the critical demand for sustainable construction solutions in energy-intensive locations.

Furthermore, the results emphasize the importance of considering local climatic conditions and building performance metrics in the design process, ensuring that the proposed solutions are contextually appropriate and environmentally beneficial. The developed optimization tool script is particularly relevant to the local industry, providing structured decision support for architects and other stakeholders seeking to decrease energy use through retrofitting existing structures.

The study’s findings demonstrate that creative sustainable facade designs can be produced by efficiently using parametric workflow and genetic algorithm optimization, which are related to the research in [35], as the authors show that combining parametric design software with simulation programs seems to present a potential to create building forms and components that are energy efficient.

In contrast to the optimization method used in [50], where the author suggests six different solutions and selects the best one, the AI-based method developed and used in this study allows for the generation of over 100 solutions, simulates them, and determines the best one. Thanks to AI tools and techniques, numerous solutions that would not be achievable with traditional methods can be generated.

The PR for the optimized parametric structure for Building A was 63%, according to the optimized solutions from this investigation. In order to achieve sufficient daylighting and visual connection, the PR should be between 50% and 70%, according to recommendations and findings from various studies [46,47,51]. The solutions produced by this study fell within the suggested range.

Applying the optimal solution to Building A results in a reduction in energy consumption, consistent with the findings of [52], which indicated that fixed perforated shading devices could successfully cut overall energy consumption by 30%. Returning to this study’s findings, the observed decrease for Ap1 was 26.2%.

In architectural design, which integrates aesthetic values, functionality, and structural stability, considering the shapes, direction, and sizes of various built environment components has a significant effect on building artificial light use, costs, energy use, cooling and heating demands, shading performance, functional accessibility, acoustics, and solar gains. Consequently, optimizing these elements is crucial, as it affects the life cycle of the built environment. This study highlights the necessity of developing design elements and parameters while considering various contextual factors. Given the design tasks that challenge the human intellect, computational decision support systems emerge as a feasible answer. Computational optimization, particularly multi-objective optimization, has successfully handled important design complexity issues in engineering and, more recently, architecture.

Additionally, adding this shading structure has a psychological impact on the citizens. Connecting modern architecture with heritage architecture is a primary goal of the Abu Dhabi municipality standards and codes. Furthermore, as per a new study, updated thermal sensations of occupants impact their mood states [53].

The parametric shading structure was designed to balance functional efficiency and aesthetic appeal by drawing inspiration from traditional Mashrabiya patterns. These patterns provide a culturally significant visual identity, aligning modern architecture with heritage elements. To maintain a balance between design elegance and energy efficiency, this study constrained the perforation ratio (PR) between 50% and 70%, ensuring sufficient daylighting while reducing overheating risks. Additionally, depth ratio (DR) constraints were set to prevent excessive material use, which could negatively impact facade integration.

The fabrication and installation of custom parametric shading structures can incur higher initial costs compared to standard fixed shading elements. However, these costs must be weighed against the long-term energy savings achieved through cooling load reductions. Moreover, the 26.2% reduction in cooling demand translates into lower electricity bills, potentially offsetting the initial investment over a building’s operational life.

The structural integrity of the parametric shading system was a critical consideration in this study. The lightweight aluminum structure ensures minimal additional load on existing facades. The design adheres to Abu Dhabi Municipality facade regulations, maintaining a 1 m gap for ventilation and maintenance access.

5. Conclusions

The optimized solution based on cumulative radiation patterns decreased the total radiation amount by 19%. At the same time, applying the resulting optimized pattern to Building A decreased the annual energy consumption for one apartment by 26.2%.

The optimal parametric shading structure attempts to increase the shadowed area to reduce solar radiation on building facades, save energy for air conditioning, and improve energy efficiency. To attain this purpose, mixed qualitative, quantitative, and experimental methodologies were used throughout the study to gather, define, and analyze the necessary data and tools. The tools and resources were defined after researching the relevant literature. Evolutionary computation was the way of applying AI in this study through the genetic algorithm for the optimization process; GH for Rhinoceros software was used to generate the parametric pattern; HB and LB plug-in toolsets were utilized to input the case study details and connect the case study model to the Energy Plus engine, which was utilized in this study to simulate the energy consumption; the LB tool was used to create a solar radiation script and an energy simulation script in coloration with the Energy Plus engine. The Galapagos tool was used for the optimization process, based on the results of the previous stages of this research. The final script for the advanced optimization toolset is as follows:

The script’s input parameters are defined as the targeted building dataset, which includes the building location, geometry, orientation, context, wall-to-window ratio, building enveloping materials, and EPW file. The advanced tool receives data using the HB model’s tool.

Variable parameters: Parametric pattern metrics, which are set to maximum and minimum values that ensure a 40–70% perforation on the final optimized pattern to ensure adequate values for daylighting and visual connectivity. The advanced tool receives the metrics through the genome node on the Galapagos evolutionary solver.

The objective function for minimization: The annual cumulative radiation on the building. The tool receives objective function values through the fitness node on the Galapagos evolutionary solver.

Future research could extend the scope of case studies to include various building types in different climates, enhancing the generalizability of findings. Exploring the integration of optimized shading structures with renewable energy systems, like solar panels, could further improve sustainability. Additionally, evaluating the impact on indoor environmental quality and occupant health would provide a general view of the benefits. User preferences can be one of the considered parameters. The economic feasibility of each pattern is the subject of future studies. In addition, we suggest developing dynamic shading systems that adapt to real-time conditions and offer more significant energy savings and comfort. Furthermore, we recommend conducting lifecycle assessments of shading structures, which would ensure environmental benefits throughout the building’s lifespan, and enhancing simulation tools for more accurate environmental data and performance metrics that could lead to better-informed design decisions.

While this research focused on a mid-rise residential building in Abu Dhabi, future studies could enhance the research by extending the analysis to various climatic conditions, such as hot/humid, temperate, and cold climates. Parametric simulations for different climate zones could help refine shading parameters (e.g., the perforation ratio, depth ratio, and gap width) to optimize performance based on local solar radiation, seasonal temperature variations, and humidity levels. Adaptive shading systems with smart materials could improve energy efficiency by adjusting to cooling needs in summer and passive heating in winter. Additionally, implementing physical prototypes in diverse locations would validate the simulation results, ensuring practical applicability across different environments.