Developing an Optimized Energy-Efficient Sustainable Building Design Model in an Arid and Semi-Arid Region: A Genetic Algorithm Approach

Abstract

The building sector is a major contributor to resource consumption, energy use, and greenhouse gas emissions. Sustainable architecture offers a solution, leveraging Building Energy Modeling (BEM) for early-stage design optimization. This study explores the use of genetic algorithms for optimizing sustainable design strategies holistically. A comprehensive analysis and optimization model was developed using genetic algorithms to individually optimize various sustainable strategies. The optimized strategies were then applied to a pre-existing building in Kabul City, a region facing significant environmental challenges. To enhance accuracy, this study integrated energy simulations with Computational Fluid Dynamics (CFD). This research combines genetic algorithms with energy simulation and CFD analysis to optimize building design for a specific climate. Furthermore, it validates the optimized strategies through a real-world case study building. Optimizing the Window-to-Wall Ratio (WWR) and shading devices based on solar exposure significantly improved the building’s energy performance. South (S)-facing single windows and specific combinations of opposing and adjacent windows emerged as optimal configurations. The strategic optimization of building component materials led to substantial energy savings: a 58.6% reduction in window energy loss, 78.3% in wall loss, and 69.5% in roof loss. Additionally, the optimized pre-existing building achieved a 48.1% reduction in cooling demand, a 97.5% reduction in heating demand, and an overall energy reduction of 84.4%. Improved natural ventilation and controlled solar gain led to a 72.2% reduction in peak-month CO₂ emissions. While this study focused on applicable passive design strategies, the integration of advanced technologies like Phase Change Materials (PCMs), kinetic shading devices, and renewable energy systems can further improve building performance and contribute to achieving net-zero buildings.

1. Introduction

The rapid pace of urbanization and population growth has placed immense pressure on our planet’s ecosystems, leading to a host of environmental challenges, including global warming, air pollution, ozone depletion, water pollution, resource depletion, deforestation, soil degradation, waste accumulation, and biodiversity loss [1]. The building sector, in particular, has emerged as a major contributor to these issues due to its significant consumption of resources, energy, and materials, as well as its generation of waste and greenhouse gas emissions [2,3]. With nearly half of the world’s population now residing in urban areas [4], the building sector’s role in exacerbating environmental problems has become increasingly critical. It is estimated that the sector accounts for approximately 40% of global resource use, 30% of total energy consumption, and 30% of carbon emissions [5,6,7,8,9,10]. Modern buildings alone are responsible for over 40% of global energy consumption and 30% of annual greenhouse gas emissions [11]. Kabul City, the capital of Afghanistan, is a prime example of a city grappling with the adverse effects of urbanization. The city’s arid and semi-arid climate, characterized by extreme temperature fluctuations, necessitates significant energy consumption for heating and cooling residential buildings [12,13]. Coupled with poor air quality and overpopulation, Kabul has become one of the world’s most polluted cities [14].

Building elements, such as layout, glazing type, shape, façade patterns, and shading strategies, significantly influence a building’s energy performance [13]. Given these factors, sustainable architecture has emerged as a critical trend in the 21st century, driven by environmental concerns and the need for more sustainable building technologies and practices [15,16]. Numerous design strategies, standards, and assessment methods have been developed to promote sustainable building design at various scales [5,17,18,19]. However, a gap persists between theoretical approaches and practical implementation. To bridge this gap and achieve optimal sustainable and energy-efficient building designs, simulation-based optimization methods have gained prominence [20,21,22]. Building Energy Modeling (BEM) has become a powerful tool for supporting sustainable building design by providing opportunities for environmentally friendly design options early in the design stage [23,24,25,26]. Parametric and optimization simulation methods, specifically, offer significant potential for developing high-performance buildings [27,28]. By incorporating these advanced techniques, designers can make informed decisions and implement optimized design strategies to achieve sustainable and energy-efficient buildings.

Genetic algorithms, though relatively new to architectural design optimization, have emerged as a promising tool for addressing complex design challenges [29]. Inspired by Darwinian evolution and Mendelian genetics, these algorithms employ a population-based approach to search for optimal solutions [30]. The process begins with the generation of an initial population of random solutions, represented as chromosomes. Each chromosome encodes a specific design configuration and its fitness is evaluated based on predefined objective functions. Through iterative cycles of selection, crossover, and mutation, the algorithm evolves the population, favoring solutions with higher fitness. Genetic algorithms have found widespread application in various fields, including pattern recognition, image processing, neural networks, and optimal control [31]. With the advancement of computational power, their potential to revolutionize architectural design optimization is significant.

Traditional methods often fall short in addressing the complexity of architectural design problems. Genetic algorithms offer a powerful solution by balancing multiple objectives and optimizing design solutions. Various studies have explored the application of genetic algorithms in architectural design for diverse purposes, including online optimization [32] and green building design [33]. Jearsiripongkul et al. [34] utilized optimization algorithms to optimize building energy consumption and hybrid energy system production. Zhao et al. [35] applied urban Building Energy Modeling to evaluate the impact of various retrofit strategies, such as windows, air conditioners, lighting, and rooftop photovoltaics. Nielsen [36] employed dynamic programming to optimize building design based on thermal load, daylight availability, cost, and usable area. Hauglustaine and Azar [37] optimized building envelopes considering factors like code compliance, energy consumption, and cost. Mehraban et al. [38] focused on optimizing energy performance in hot climates by analyzing key building features such as walls, windows, and roofs. Miao et al. [39] optimized a range of building parameters, including setpoints, air change rates, shading devices, and window properties. Kar et al. [40] optimized energy efficiency, construction cost, and daylight based on the Window-to-Wall Ratio and solar heat gain coefficient. These studies collectively demonstrate the effectiveness of genetic algorithms in handling complex, non-differentiable functions and achieving significant improvements in optimization results compared to baseline scenarios.

Research suggests that integrating optimal form, layout, and sustainable design strategies, renowned for their climate-responsive design principles, can significantly reduce annual energy consumption in contemporary building design. While numerous studies have employed the Non-Dominated Sorting Genetic Algorithm (NSGA-II) to optimize building layout, construction materials, and design elements, these investigations have often focused on specific building types or a limited subset of design strategies. For instance, studies might consider factors like the Window-to-Wall Ratio (WWR), shading strategies, orientation, or material choices, but rarely in an integrated and holistic manner. Moreover, a comprehensive analysis of each strategy’s individual impact on a single test room remains scarce. Existing research predominantly relies on single optimization and simulation methods, which may limit the accuracy and reliability of the results. DesignBuilder, a powerful tool capable of accurate simulations and integration with genetic algorithms, has been largely overlooked. Additionally, a framework for sustainable building design tailored to specific contexts, such as the highly polluted city of Kabul, is notably absent.

This study aimed to develop a comprehensive analysis and optimization model for sustainable building design strategies, applicable globally. Genetic algorithms were employed to individually analyze and optimize each strategy, leading to a holistic understanding of their optimal forms. These optimized strategies were subsequently applied to a pre-existing building to assess their real-world impact. Furthermore, the study integrated simulation and Computational Fluid Dynamics (CFD) methods with genetic algorithms to enhance the accuracy and applicability of the optimization process. This integrated approach allowed for a more precise evaluation of the proposed strategies and their potential to improve building performance. Finally, the study aimed to provide a holistic sustainable design framework for Kabul City, a region grappling with significant environmental challenges and a lack of sustainable building practices. By applying the optimized strategies to a case study building, this research offers valuable insights for researchers, policymakers, and practitioners to develop sustainable design guidelines and create buildings that are not only comfortable, economically viable, and locally responsive but also contribute to addressing contemporary environmental challenges.

Study Area

Kabul, Afghanistan’s capital and largest city, is situated at approximately 34°31′31″ north latitude and 69°10′42″ east longitude in the eastern part of the country. The city experiences a distinct arid to semi-arid climate, characterized by hot summers and cold winters [12]. This climate is marked by four well-defined seasons throughout the year. These seasons are clearly defined as follows: spring, occurring in March, April, and May; summer, extending from June to August; fall, covering September, October, and November; and winter, which takes place in December, January, and February. Temperatures can fluctuate significantly, ranging from lows of −10 °C in winter to highs of 40 °C in summer [12,41]. According to [42,43], prevailing wind patterns shift seasonally. From May 3 to October 6, winds generally blow from north to south, while from October 6 to May 3, the dominant wind direction is west to east. The climatic conditions of Kabul City, as determined by DesignBuilder analysis using local EnergyPlus Weather File (EPW) weather data and corroborated by literature and official websites, are illustrated in Figure 1. The city exhibits distinct seasonal variations, experiencing a hot period from May to September and a cold period from October to April. The hottest months occur during summer, while the coldest are in winter. Wind speeds, a crucial factor for natural ventilation and passive cooling, typically range from 2.19 to 4.02 m per second, with occasional peaks exceeding 14 m/s. Solar radiation, which plays a significant role in passive heating and cooling, exhibits peak values of up to 0.8 kW/m² during the hot period and up to 1 kW/m² during the cold period, particularly in winter. To optimize building energy performance, strategies should be implemented to mitigate excessive solar heat gain during the hot period, such as shading devices, and to maximize solar gain during the cold period, such as appropriate window orientation, glazing selection, and shading strategies.

Kabul’s urban development is primarily concentrated on the valley floor of the Kabul River, at an elevation of approximately 1800 m above sea level [44,45]. The city’s topography is diverse, with approximately 56% of its territory composed of mountains and rugged terrain and 38% consisting of flatter land. In recent years, Kabul has emerged as one of Asia’s fastest-growing cities [46]. It is divided into 22 districts, encompassing a land area of 1049 square kilometers and an estimated 396,095 dwelling units [47,48]. The rapid urbanization of both formal and informal settlements in Afghanistan, particularly in Kabul, has resulted in increased pressure on natural resources. Unplanned urban growth, especially in informal settlements, poses significant long-term risks to human health and the environment [48]. Projections indicate that by 2060, approximately 50% of the Afghan population will be urban dwellers [49]. Kabul’s housing stock is diverse, encompassing three primary types: regular detached houses (26%), apartments (1%), and irregular houses (70%) [13]. The latter, prevalent in informal settlements, often lack adequate infrastructure. Even in formal areas, newly constructed housing may not be optimally designed for local climatic conditions and cultural preferences [49]. Energy consumption patterns in Kabul are influenced by seasonal variations. During winter, residents frequently rely on coal or wood for heating, contributing to air pollution and health issues [50]. In warmer months, electricity and natural gas are commonly used for residential needs [42]. However, in colder periods, households often turn to wood, coal, charcoal, and natural gas as primary energy sources.

Urban growth and buildings’ energy consumption are the primary drivers of air pollution in Kabul, with pollutants such as NO₂, SO₂, and CO reaching alarming levels. The city’s rapid population growth and poor air quality have placed it among the world’s most polluted urban areas, alongside cities like New Delhi and Beijing [14]. The burning of coal, wood, gas, and biomass, primarily for heating during winter, is a significant source of these pollutants in Kabul [50]. Concentrations of NO₂, CO, and SO₂ typically decrease from winter to summer. The lack of sustainable building design and energy-efficient practices in Kabul exacerbates the problem. Many buildings are not designed to optimize energy use, leading to an increased reliance on fossil fuels and subsequent air pollution. To mitigate air pollution in Kabul, it is essential to adopt energy-efficient building design and construction practices. Additionally, the integration of renewable energy sources can significantly reduce reliance on fossil fuels and improve air quality.

2. Materials and Methods

Prioritizing well-designed strategies, such as optimizing building orientation, incorporating effective façade elements, implementing shading devices, selecting appropriate materials, and utilizing efficient heating and cooling systems, can significantly contribute to achieving energy-efficient and sustainable buildings. By focusing on these strategies, it is possible to attain sustainable building practices while maximizing energy efficiency. In this study, DesignBuilder v7 was utilized to analyze sustainable building strategies based on the local context of Kabul City. To individually analyze and optimize each strategy, a test room was subjected to a genetic algorithm. For each strategy, specific objectives and variables were defined to guide the optimization process. After determining the optimal outcomes for each strategy, these were integrated into an existing building model to simulate their energy performance, CO₂ emissions, and thermal comfort within the proposed comprehensive model for the study area. To further improve the accuracy and readability of the design model, a comparative Computational Fluid Dynamics (CFD) analysis was conducted on the existing building and its optimized form. This analysis evaluated the impact of the optimized strategies on building thermal comfort, indoor airflow patterns, and energy efficiency. A detailed flowchart, shown in Figure 2, illustrates the step-by-step process of this study’s methodology.

The novelty of this study lies in the combination of these strategies and the integration of research methods such as a genetic algorithm, annual energy simulation, and CFD analysis. Finally, to evaluate the impact of these strategies on building sustainability and validate the accuracy of the analyses, all optimized strategies were implemented on a pre-existing building and compared to its original form.

2.1. Strategies for Designing Energy-Efficient and Sustainable Buildings

2.1.1. Windows Glazing

The optimization of glazing systems significantly enhances thermal insulation, thereby reducing heat transfer and improving energy efficiency. The market offers a diverse range of glazing options, each with distinct properties influenced by factors such as color, material, reflectivity, and thickness. Glazing systems can be categorized into single, double, or triple-pane configurations, which substantially impact heat loss and energy conservation. This study focused on the most common glazing types in the study area, summarized in

Table 1. Common glazing types and their key characteristics for energy performance analysis.

Glaz Type	Description	Type of Pane			Thickness (mm)		Gap (mm)	Insulation		U-Value (W/m2K)
		Single Pane	Double Pane	Triple Pane	3	6	13	Air	Argon
Type 1	Sgl Clr or Green or Grey 3 mm	✓			✓					5.894
Type 2	Sgl Clr or Green or Grey 6 mm	✓				✓				5.778
Type3	Dbl Clr or Green or Grey 3 mm/13 mm Air		✓		✓		✓	✓		2.716
Type 4	Dbl Clr or Green or Grey 3 mm/13 mm Arg		✓		✓		✓		✓	2.556
Type 5	Trp Clr 3 mm/13 mm Air			✓	✓		✓	✓		1.757
Type 6	Trp Clr 3 mm/13 mm Arg			✓	✓		✓		✓	1.620

.

In this study, window glazing was considered the primary variable, with six common glazing types (options) selected for genetic algorithm analysis, as depicted in Table 1. The objective of the genetic algorithm was to minimize the heating and cooling demand by identifying the most optimal glazing type from Table 1.

2.1.2. Windows Orientation

Building envelopes, particularly window openings, significantly impact energy performance, accounting for 20–40% of energy loss [51]. Optimal window orientation is crucial for achieving sustainable design and thermal comfort across diverse climates [52]. The primary objective of orientation is to harness natural ventilation and solar radiation strategically [53]. Ideal alignment is typically east–west, maximizing solar gain in winter and minimizing it in summer [54]. This study investigated the influence of window configurations on energy efficiency through simulations. Single-sided orientations (north, east, south, and west) and dual-sided orientations (north–south, east–west, north–east, east–south, south–west, and west–north) were analyzed. Genetic algorithms were used to optimize the Window-to-Wall Ratios (WWRs) and Vertical Shading Angles (VSAs) for each window configuration within a test room, assuming optimal glazing.

Window-to-Wall Ratio (WWR) and Shading Strategies

The WWR, the proportion of window area to total wall area, is a crucial factor in sustainable building design. Optimizing the WWR enhances solar gain, natural ventilation, and daylighting, consequently lowering energy consumption and costs. Moreover, effective shading strategies, such as overhangs, louvers, fins, blinds, and combined systems, are crucial for controlling solar heat gain and improving energy efficiency. The size of shading devices should be optimized to balance solar access in winter and shading in summer [55,56]. S-facing openings typically require horizontal shading, while east- and west-facing openings benefit from vertical or combined shading. The Vertical Shading Angle (VSA) and Horizontal Shading Angle (HSA) are key parameters for designing effective shading systems. The VSA is the angle formed between the sun’s rays and the horizontal plane of a building’s facade. The HSA is the angle between the sun’s rays and the vertical plane of a building’s facade. VSA is primarily used to assess the effectiveness of horizontal shading devices like overhangs and louvers in blocking sunlight. HSA, on the other hand, is used to evaluate the effectiveness of vertical shading devices such as fins, blinds, and overlaps in blocking sunlight, primarily considering the upper half of the window. This study utilized genetic algorithms integrated within DesignBuilder v7 to optimize variables such as the VSA between 10° and 90° and the WWR between 10% and 60%. The optimization was conducted for various window configurations, including single-sided, opposing, and adjacent orientations, considering optimal glazing properties. A range of shading devices, including no shading, overhangs, louvers, fins, blinds, and combined systems with different configurations, were analyzed to identify the optimal solution for each scenario based on the optimized VSA.

2.1.3. Construction Material

This study examined the impact of building envelope materials on the energy performance of buildings in Kabul City. The primary focus was optimizing the thermal performance of walls, roofs, and ground floors to reduce heat transfer and enhance energy efficiency. A baseline test room with a single south-facing window was established, incorporating an optimized glazing type, WWR, and shading strategy. Subsequently, a genetic algorithm optimization analysis was conducted to identify the most energy-efficient construction materials for the walls, roof, and floor. The wall, roof, and ground floor components were considered as variables for the genetic algorithm. Various types or options for each component were evaluated to identify the optimal combination for minimizing the heating and cooling demand.

Walls: The predominant wall types used in Kabul City, as documented in [57], were reviewed alongside commonly used options.

Table 2. Common wall types and their key characteristics for energy performance analysis.

Wall Type	Description
	Outside Plaster (150 mm)	EPS (Outside) (mm)			Mud-Brick (mm)	Brick—Baked (mm)	Aerated Brick (mm)	Aerated-Con Block (mm)	Brick—Reinforced x 2 (mm)	Concrete Cast Dense x 2 (mm)	EPS (Inside) (mm)			Earth or Mud Wall (mm)	Stone-Basalt (mm)	Inside Plaster (mm)	Total U-Value (W/m2·k)	Total R-Value (m2·k/W)
	15	5	10	15	320	320	320	300	110	30	5	10	15	800	500	15
Type 1	✓					✓										✓	1.649	0.606
Type 2	✓				✓											✓	1.483	0.674
Type 3	✓						✓									✓	0.771	1.297
Type 4	✓							✓								✓	0.676	1.480
Type 5	✓					✓					✓					✓	0.539	1.856
Type 6	✓	✓				✓										✓	0.539	1.856
Type 7	✓					✓						✓				✓	0.322	3.106
Type 8	✓		✓			✓										✓	0.322	3.106
Type 9	✓					✓							✓			✓	0.230	4.356
Type 10	✓			✓		✓										✓	0.230	4.356
Type 11	✓								✓				✓			✓	0.239	4.180
Type 12	✓									✓			✓			✓	0.249	4.015
Type 13	✓													✓		✓	0.565	1.771
Type 14	✓														✓	✓	2.679	0.373

summarizes these wall types for the genetic algorithm, including their layer composition, thermal transmittance (U-value), and thermal resistance (R-value). The U-value reflects the rate of heat transfer through the wall, while the R-value indicates its ability to resist heat flow.

Roofs: A variety of roof types exist, differentiated by material composition and construction methods. This study investigated the prevalent roof types in Kabul City to identify the most energy-efficient option for the local climate.

Table 3. Common roof types and their key characteristics for energy performance analysis.

Roof Type	Description
	Waterproof (mm)	Bitumen Sheet (mm)	Mortar (mm)	Cement Mortar (mm)	EPS (mm)			Slab (Co-Aerated) (mm)	Slab (Co-Reinforced) (mm)	Plaster (mm)	Soil (mm)	Reed Thatch (mm)	Plywood (mm)	Timber Joists (d/mm)	Total U-Value (W/m2·k)	Total R-Value (m2·k/W)
	5	5	50	50	5	10	15	120	120	15	50	200	15	150
Type 1		✓		✓					✓	✓					3.083	0.324
Type 2	✓		✓					✓		✓					1.003	0.997
Type 3		✓		✓	✓				✓	✓					0.635	1.574
Type 4		✓		✓		✓			✓	✓					0.354	2.824
Type 5		✓		✓			✓		✓	✓					0.254	4.074
Type 6		✓		✓							✓	✓	✓	✓	0.378	2.644

provides an overview of these common roof types, including their thermal properties.

Ground Floors: Unlike upper floors that primarily function as partitions, ground floors directly interact with the ground and can influence heat loss and energy demand.

Table 4. Common ground floor types and their key characteristics for energy performance analysis.

Floor Type	Description
	Carpet (mm)	Tiles (mm)	Ceramic (mm)	Mortar (mm)	Cement Mortar (mm)	EPS (mm)			Cast Concrete (mm)	Co-Compacted (mm)	Gravel (mm)	Soil (mm)	Total U-Value (W/m2·k)	Total R-Value (m2·k/W)
	10	20	20	30	30	5	10	15	100	100	200	500
Type 1	✓	✓		✓					✓		✓	✓	0.925	1.081
Type 2	✓		✓		✓					✓	✓	✓	0.697	1.435
Type 3	✓		✓		✓	✓				✓	✓	✓	0.372	2.685
Type 4	✓		✓		✓		✓			✓	✓	✓	0.254	3.935
Type 5	✓		✓		✓			✓		✓	✓	✓	0.193	5.185

details the thermal properties of various ground floor types commonly found in the study area.

Although interior partition walls and floor slabs between conditioned spaces are often neglected in energy simulations due to their perceived minimal impact on heat transfer, this study investigated their potential influence on both energy and acoustic performance. Internal partitions were modeled similarly to baseline (Type 1) external walls, comprising two layers of plaster and a 250 mm-thick brick infill. Floor slabs were modeled as non-insulated roofs constructed with ceramic tiles, mortar, reinforced concrete slabs, and plaster.

2.2. Building Energy Modeling (BEM)

Building Energy Modeling (BEM) is a valuable tool for optimizing various aspects of building design, including window characteristics, building form and thermal mass, envelope design, daylighting analysis, water harvesting, energy modeling, renewable energy integration, sustainable material selection, and site and logistics management [58]. BEM enables architects to investigate and evaluate design decisions early in the design process, leading to more informed and optimized designs [59,60]. Parametric and optimization techniques are powerful methods for maximizing building performance [27]. DesignBuilder v7, a comprehensive BEM related software, was selected for this study due to its ability to conduct whole-building energy simulations and CFD analysis and its integration of genetic algorithms for optimization. This tool facilitated the analysis and optimization of the strategies outlined in this study, adapting them to the specific context of the study area.

To effectively simulate and optimize building strategies for Kabul City, a representative test room was designed based on typical residential building characteristics.

Table 5. Characterization of a test room based on local context that utilized for simulation and optimization of strategies within the DesignBuilder software.

Simulation Parameters	Test Room	Description
Site	Location	Kabul, Afghanistan (34.522362 N & 69.097524 E)
	Hourly weather data	EPW file from Kabul International Airport weather station
Test room	Room type	Residential
	Test room area	6 m × 6 m
	Test room height	3 m
Construction material	Wall	Wall Type 1
	Roof	Roof Type 1
	Floor	Floor Type 1
	Glazing	Single-pane, clear glass with 6 mm thickness
Opening	Openable areas	50%
	Framing	Wooden window frame
	Operation	Hot months based on control set points
Cooling	System seasonal CoP	COP 3
	Device	Split air conditioner
	Fuel	Electricity
	Distribution method	Distribute cool air directly
	Cooling set point	24 °C
	Operation	During hot period of the year (May to September)
Heating	System seasonal CoP	η = 0.8
	Device	Gas heater
	Fuel	Natural gas
	Distribution method	Radiation
	Heating set point	22 °C
	Operation	During cold period of the year (October to April)
Ventilation	Natural ventilation	Based on minimum temperature set points
Infiltration	Crack template	Good/infiltration 0.000060 kg/s·m @ 1 Pa
Lighting	Template	LED

presents significant information regarding the test room configurations and simulation parameters, which were derived from common existing buildings in the study area to accurately represent local conditions.

DesignBuilder, a robust simulation tool, offers a diverse array of precise HVAC system templates. These templates are highly adaptable to various research contexts and can be fine-tuned to match the specific characteristics of the study area. For this particular study, meticulous care was taken to select and modify the HVAC system template to accurately represent the typical residential buildings found in the region. Kabul’s arid to semi-arid climate, with its hot summers and cold winters, was carefully considered in the model. Moreover, the geographic location and building materials were chosen to accurately represent local conditions and building practices.

2.2.1. Genetic Algorithm

DesignBuilder is a powerful software tool capable of analyzing and optimizing various building energy efficiency strategies. Beyond its core functions of energy simulation and Computational Fluid Dynamics (CFD) analysis, DesignBuilder offers advanced optimization capabilities. By defining up to two objectives and ten key variables with multiple options, the software can systematically simulate and optimize these variables to achieve the desired outcomes. The optimization process relies on a genetic algorithm, illustrated in Figure 2. This algorithm iteratively evaluates different variable combinations against the defined objectives, gradually refining the solution. The process terminates when a predefined maximum number of generations is reached or a satisfactory solution is found. The optimization process often yields multiple optimal and suboptimal solutions, each with its own set of characteristics and trade-offs between the two objectives. While the optimal solution for one objective may not be ideal for the other, the final selection should aim to strike a balance between both. This decision-making process typically involves expert judgment and consideration of the specific project requirements.

In this study, to assess the impact of various strategies, each strategy was independently implemented in a test room. For each strategy, specific objectives and variables with corresponding options were defined. The primary objective was to minimize the heating and cooling demand. The strategies, serving as variables in the genetic algorithm analysis, were considered with distinct relative options. To achieve more reliable and accurate results, the optimization process using the genetic algorithm was divided into three stages:

• The initial stage focused on window glazing as the variable, analyzed individually in a test room using a genetic algorithm with 14 options encompassing six different glazing types with varying colors and thicknesses, as detailed in Table 1.
• The second stage considered window configurations with two variables, including WWR and VSA for shading. Ten distinct window configurations were considered individually: four single-sided scenarios, two opposing scenarios, and four adjacent scenarios. For each window scenario, the WWR variable had 11 options ranging from 10% to 60% in 5% increments and the VSA variable had 17 options ranging from 10° to 90° in 5° increments.
• The third stage focused on construction materials, examining three key variables: wall materials, roofing materials, and ground floor materials. Fourteen options were considered for walls (detailed in Table 2), six for roofs (detailed in Table 3), and five for ground floors (detailed in Table 4).

Following each implementation, an optimization process was conducted to identify the most efficient and effective combination of parameters within that strategy. By iteratively implementing, analyzing, and optimizing individual strategies, a comprehensive understanding of their impact on energy consumption, thermal comfort, and environmental performance was gained. This approach enabled informed decision-making for the design and optimization of energy-efficient buildings in Kabul City.

2.2.2. Annual Energy Simulation

Subsequently, to assess the specific impact of the optimized strategies on a real-world building, a comprehensive annual energy simulation analysis was conducted from January 1st to December 31st. All identified optimal strategies were implemented into a pre-existing building model within DesignBuilder to provide detailed insights into energy demand, CO₂ emissions, and natural ventilation potential. These results offer valuable information regarding the reliability of the study and the effectiveness of the proposed model and optimized strategies.

2.2.3. Computational Fluid Dynamics (CFD)

Finally, to assess the effectiveness and impact of optimized strategies on indoor air temperature fluctuations and thermal comfort during natural ventilation, comprehensive Computational Fluid Dynamics (CFD) simulations were performed on both the existing and optimized buildings. These analyses focused on specific periods characterized by suitable natural ventilation rates, wind speeds, and indoor–outdoor temperature differentials. To ensure consistent conditions, the summer simulation was conducted on June 27th at 11 p.m. This specific time was selected based on annual energy simulations, which identified conditions with suitable natural ventilation rates and air temperatures for accurate CFD analysis.

Both the pre-existing and optimized building designs were initially modeled in DesignBuilder, incorporating local climate data and typical residential building characteristics to prepare for subsequent Computational Fluid Dynamics (CFD) analysis. Boundary conditions for both models were derived from hourly energy simulations conducted throughout the year. This ensured the selection of appropriate timeframes for CFD analysis and the establishment of realistic boundary conditions. Local EPW weather data were integrated into DesignBuilder to accurately represent site-specific conditions. Based on the simulation results, a timeframe with a predominant 6.5 m/s wind speed from the north side was selected for the CFD analysis. The surface temperatures of interior components ranged from 27 °C to 32 °C, with an average indoor temperature of 29.5 °C and an outdoor temperature of 25 °C. Additionally, the airflow balance, including inflow and outflow through windows, and associated pressure differentials varied for each window due to their specific orientation and configuration

Moreover, accurate grid generation and turbulence modeling are essential for reliable CFD simulations. In this study, DesignBuilder v7 was used to generate a high-quality mesh for the test room with a grid spacing of 0.3 m and a merge tolerance of 0.03 m. The mesh quality was assessed by examining grid statistics and the maximum aspect ratio, which was found to be below 5, indicating suitability for the current study. Regarding turbulence modeling, DesignBuilder offers a range of options, including constant effective viscosity and k-epsilon models. Given the focus on natural ventilation with particulate flow, the k-epsilon turbulence model was selected due to its widespread use and validation in similar studies [22,60]. This model is well-suited for capturing the complex flow patterns and pollutant transport associated with natural ventilation.

Furthermore, the grid resolution significantly impacts the accuracy of CFD simulations. To ensure grid independence, the mesh was systematically refined by reducing the grid spacing to 0.15 m while maintaining consistent grid line merging tolerances. This refinement led to an approximately 50% increase in mesh cells. A comparative analysis of results from both mesh configurations revealed negligible differences in boundary conditions and CFD outcomes. This suggests that the initial mesh configuration was adequately accurate and well-adjusted.

In summary, the selection of a robust turbulence model, coupled with the accurate boundary conditions derived from DesignBuilder simulations using local EPW weather data and the validation of mesh quality through DesignBuilder’s statistical analysis and maximum aspect ratio checks, significantly improved the alignment of CFD results with real-world conditions. Furthermore, conducting a grid independence study by refining mesh spacing assessed the accuracy and reliability of the CFD results.

3. Results

The DesignBuilder v7 software employed a genetic algorithm to optimize sustainable building design strategies for a test room in Kabul City. The primary objective was to minimize both the heating and cooling loads. Given the region’s arid and semi-arid climate, characterized by hot summers and cold winters, heating and cooling demands often conflict. Therefore, selecting the optimal glazing solution requires careful consideration to balance both energy needs and achieve an overall energy-efficient design. To identify the most suitable glazing type, the software iteratively simulated 1414 generations, considering 14 variables across six glazing types (variables) listed in Table 1. Four optimal glazing types were identified and are highlighted in Figure 3.

Among these options, triple-pane glass with argon gas insulation (Glazing Type 6) emerged as the most effective solution, with 3386 kWh of heating demand and 1464 kWh of cooling demand. This configuration strikes a balance between energy efficiency and other performance factors.

3.1. Single-Side Windows

In the case of single-sided windows, the analysis was conducted in a test room with optimized glazing (Glazing Type 6). This study considered variables such as a WWR range of 10% to 60% and a VSA range of 10° to 90°. Due to the reduced solar exposure on the north (N) facade, shading strategies were not considered for this orientation. Instead, the focus was on optimizing the WWR to maximize daylighting and thermal comfort while minimizing energy consumption. After 2553 iterations, 11 optimal WWR configurations were identified, as highlighted in Figure 4. Among these, a 60% WWR with a heating demand of 5354 kWh and a cooling demand of 1260 kWh emerged as the most balanced and energy-efficient solution for the north-facing window.

For the east (E) facade, 286 iterations identified 52 optimal configurations, as depicted by the red points in Figure 5. Among these, a 60% WWR coupled with an overhanging shade at a 65-degree VSA emerged as the most balanced and energy-efficient solution, with 5077 kWh heating demand and 1392 kWh cooling demand.

The optimization process was repeated for the south (S)-facing window, yielding 38 optimal combinations of WWR and shading devices after 120 iterations, as illustrated in Figure 6. The most effective configuration was found to be a 60% WWR with an overhanging shade at an 80-degree VSA, offering the best balance of energy efficiency with 3729 kWh heating demand and 1282 kWh cooling demand.

Similarly, for west-facing windows, 326 iterations were conducted to identify the optimal WWR and shading strategies. As shown in Figure 7, 53 optimal combinations were found. Among these, a 50% WWR with an overhanging shade at a 55-degree VSA, resulting in a heating demand of 4970 kWh and a cooling demand of 1570 kWh, emerged as the most balanced and energy-efficient solution.

As a result, within single-side window configurations, the S-facing test room emerged as the most energy-efficient, balancing heating and cooling demands with the lowest total energy consumption. This makes it the optimal choice for various room types. In contrast, the N-facing test room, with its lowest cooling demand and highest heating demand, exhibits the highest total energy consumption, rendering it suitable only for summer-use rooms.

3.2. Opposing Windows

An analysis considering similar variables was conducted on a test room with optimized glazing to assess the impact of opposing window configurations (NS and EW). For the NS orientation, after 4748 iterations, 43 optimal combinations were identified, as shown in Figure 8. The most balanced and energy-efficient solution was a 60% WWR with an overlapping overhang at a 75-degree VSA for the S-facing window and a 10% WWR without shading for the N-facing window. This configuration resulted in a heating demand of 3910 kWh and a cooling demand of 1251 kWh.

For the EW window orientation, 1880 iterations identified 163 optimal configurations, visualized as red points in Figure 9. The most balanced and energy-efficient solution involved a 30% WWR with an overlapping overhang (65° VSA) for the W-facing window and a 60% WWR with an overhang (70° VSA) for the E-facing window. This configuration minimized energy consumption, resulting in a heating demand of 4668 kWh and a cooling demand of 1893 kWh.

Comparing the two opposing window configurations, the NS orientation demonstrated a lower total energy demand, encompassing both heating and cooling, compared to the EW orientation. This reduced energy consumption can be attributed to the significant solar gain from the S orientation and the potential for natural ventilation from the N side, influenced by the prevailing wind direction in the study area.

3.3. Adjacent Windows

An analysis of adjacent window configurations, based on similar variables, was conducted in a test room with optimized glazing. For the NE window combination, 536 iterations revealed 63 optimal configurations, as shown in Figure 10. The most balanced and energy-efficient solution was a 60% WWR with an overlapping overhang at a 75-degree VSA for the E-facing window and a 10% WWR without shading for the N-facing window. This setup resulted in a heating demand of 4942 kWh and a cooling demand of 1550 kWh.

Similarly, for adjacent E- and S-facing windows, 1692 iterations were conducted to identify 140 optimal combinations, as shown in Figure 11. The most balanced and energy-efficient solution involved a 10% WWR with an overhang (55° VSA) for the E-facing window and a 60% WWR with an overhang (85° VSA) for the S-facing window. This configuration resulted in a heating demand of 3520 kWh and a cooling demand of 1423 kWh.

For adjacent S and W window orientations, the optimization algorithm iterated 1402 times (Figure 12) to identify 125 optimal configurations of WWR and shading devices. The most effective configuration involved a 60% WWR with an overlapping overhang (VSA 85°) for the S-facing window and a 10% WWR with louvres (VSA 50°) for the W-facing window. This configuration resulted in a heating demand of 3512 kWh and a cooling demand of 1461 kWh.

The optimization process was repeated for adjacent windows facing W and N. The optimization algorithm performed 699 iterations, as shown in Figure 13, to identify 82 optimal configurations of WWR and shading devices. The most effective configuration involved a 60% WWR with louvres at a 50-degree VSA for the W-facing window and a 10% WWR with no shading for the N-facing window. This configuration resulted in a heating demand of 4929 kWh and a cooling demand of 1630 kWh.

Consequently, comparing adjacent configurations, the ES and SW orientations exhibited lower energy demands for both heating and cooling, with similar results. These configurations emerged as optimal and energy-efficient options due to the potential for solar gain from the S side and partial solar gain from the E and W sides.

3.4. Envelope Configurations

The DesignBuilder v7 program, utilizing a genetic algorithm, was employed to simulate and optimize building components or variables, including walls, the roof, and the ground floor, with various material configurations. These analyses were conducted in an S-facing test room, where the glazing type, WWR, and VSA for shading devices were optimized. Three variables, including the wall, roof, and ground floor materials, with respective options detailed in Table 2, Table 3, and Table 4, were considered to identify the optimal choice within each category. After 810 iterations, 33 optimal solutions were identified for walls, roofs, and ground floor, highlighted in red in Figure 14. From these 33 optimal solutions, the combination of wall Type 10, roof Type 5, and ground floor Type 2, as depicted in Table 2, Table 3 and Table 4, was selected as the most balanced and effective choice.

Walls are fundamental architectural elements that serve multiple purposes. They define interior and exterior spaces, support structural loads, provide security, protect against weather elements, and contribute significantly to thermal insulation and energy efficiency. As walls constitute a major portion of building envelopes, they are a primary source of thermal loss. This study investigated various wall types, listed in Table 2. Among these, Wall Type 10, constructed with baked brick, plaster on both sides, and 0.15 mm Expanded Polystyrene (EPS) insulation on the exterior, demonstrated superior performance in reducing thermal loss compared to other configurations.

However, the walls are not the sole element significantly impacting building thermal loss and energy efficiency. Roofs and ground floors also play fundamental roles in building design and energy efficiency. Therefore, these elements, combined with walls, were included as variables in the genetic algorithm analysis. Roofs, functionally similar to walls, emerged as the second most important element in building thermal loss and energy efficiency. Among the various roof options in Table 3, Roof Type 5, featuring a Bitumen sheet, Cement mortar, Slab (Co-Reinforced), Plaster 150 mm, and EPS insulation, demonstrated superior performance in reducing thermal loss compared to other configurations.

The ground floor, which plays a crucial role in building design, was analyzed with several options listed in Table 4. Among these, ground floor Type 2, without insulation, featuring carpet, ceramic, cement mortar, reinforced concrete (slab), and gravel demonstrated superior performance in reducing thermal loss compared to other configurations. This is attributed to the reduced thermal difference between the ground and indoor environments. Moreover, the ground’s natural temperature fluctuations, warmer in winter and cooler in summer, support building energy efficiency. Insulation materials can sometimes negatively impact this performance. Consequently, the findings suggest that walls and roofs benefit significantly from insulation, whereas the ground floor exhibited optimal performance without insulation.

4. Retrofitting and Optimization of a Pre-Existing Building

The predominant housing type in Afghanistan is a building with a private yard, where over 95% of the population resides [61]. This research focuses on the most common housing type with a private yard in Kabul City. Each room in these houses has individual windows, providing independent outdoor access. However, due to limited room orientations, some rooms lack courtyard views, potentially impacting their quality. To mitigate this, more important rooms, such as the living room and guest room, are strategically positioned on the north- or courtyard-facing side. Secondary spaces, like toilets, stairs, and kitchens, are oriented differently based on environmental considerations. The specific building chosen for this study was constructed in 1980. Comprehensive access to information about its existing structure facilitated its selection as a test model for analysis and optimization. This accessibility also aids in accurately creating a building model, simulating real-world conditions, and validating simulation results against actual data. The existing building, with its precise specifications, has been meticulously modeled within the DesignBuilder environment, as illustrated in

Table 6. The most common residential building type in the study area was modeled within the DesignBuilder environment, including both as a pre-existing and optimized building.

	Pre-Existing Building	Optimized Building
Building illustration (Axonometric)
First floor plan
Ground floor plan

.

DesignBuilder accurately models the building, incorporating its complete set of specifications and existing conditions. As shown in Table 6, the model mirrors the actual building’s details within the study area. The building, designed by Kabul City Municipality, features a modern morphology similar to the test room specifications in Table 5, except for infiltration, which is considered medium. Simulations of the existing building revealed substantial energy losses and gains through various building elements, coupled with significant heating and cooling demands, as depicted in Figure 15. In Figure 15 and Figure 16, negative values in the Heat Balance indicate energy loss, while positive values denote energy gain from different building components. System Loads are represented by negative values for cooling demand and positive values for heating demand.

Additionally, a simulation was performed on the existing building after incorporating optimized strategies, as illustrated in Figure 16. The results revealed that optimizing building component materials, glazing, WWR, and shading strategies (VSAs) significantly reduced both heating and cooling demands and minimized energy losses through building components, as shown in Figure 16.

The findings highlighted the substantial impact of optimized window configurations, shading devices, and construction materials on the building’s energy performance. Notably, the optimized building (Figure 16) exhibited significantly reduced cooling demands in summer and heating demands in winter compared to the pre-existing building (Figure 15). During late spring and early autumn, when outdoor temperatures are moderate, the pre-existing building still required significant energy for both heating and cooling due to its inefficient design. This building missed opportunities to leverage solar gain for heating and natural ventilation for cooling, while also suffering from significant thermal losses through its uninsulated envelope. These factors contributed to the building’s overall energy inefficiency and sustainability issues. In contrast, the optimized building effectively utilized solar gain for heating and natural ventilation for cooling through strategic design choices. Additionally, its optimized envelope significantly reduced thermal losses, resulting in neutral months (late spring and early autumn), with minimal energy demand for heating or cooling. Even during non-neutral months, the optimized building exhibited substantial reductions in both heating and cooling loads.

Simulation Validation

The accuracy of the simulation process was assessed by comparing the simulated electricity consumption of the existing building with its actual, real-world consumption. Real-world electricity consumption data, obtained from official electricity bills for the period 2022–2023, served as a reliable benchmark for evaluation. This approach is supported by previous studies [62,63]. Additionally, this approach was selected due to the significant accessibility of real-world electricity consumption data from an existing building that closely aligned with simulation results. This building, illustrated in Table 6, was designed by Kabul City Municipality and features a contemporary architectural style. Its construction and simulation parameters adhere to local residential building characteristics and closely match the specifications outlined for the test room in Table 5.

The electricity consumption data for both the pre-existing and simulated building, as modeled in Design Builder, encompass the total energy expenditure for cooling, lighting, and electrical appliances. Based on the simulation results presented in

Table 7. Comparative analysis of actual, simulated, and optimized electricity consumption in a pre-existing building.

Seasons	Months	Existing Building (Actual)	Simulation Results for the Existing Building	Simulation Results for the Optimized Building
Winter	December	448	126	109
	January	521	126	109
	February	538	114	98
Spring	March	200	126	109
	April	150	122	105
	May	156	337	230
Summer	June	201	768	395
	July	164	1168	608
	August	158	996	550
Autumn	September	160	563	417
	October	150	126	109
	November	200	122	105
Total		3046	4694	2944

, the estimated annual electricity consumption of the existing building was 4694 kWh. By implementing optimized strategies and material selections, electricity consumption could be reduced to 2944 kWh, representing a 37.2% reduction. However, the actual annual electricity consumption, as recorded by the Afghanistan electricity board, was significantly lower at 3046 kWh. It is important to note that electricity bills in Afghanistan are issued in Persian and follow the Hijri calendar. To facilitate comparison with the simulation results, these records were converted to the Gregorian calendar, as detailed in Table 7.

The simulated electricity consumption values obtained using DesignBuilder are approximately 1.54 times higher than the actual recorded values. This significant discrepancy is primarily due to frequent and prolonged power outages in the study area, which can last up to 15 h per day during peak consumption periods, such as summer days and winter nights [64,65,66,67]. Consequently, residents often rely on alternative energy sources, including private generators, wood, and traditional domestic methods. Additionally, during winter, electricity is sometimes used alongside gas and wood for heating, as evidenced by the increased actual electricity consumption data during winter in Table 7. However, it was impossible to precisely define and adjust the exact local conditions within the simulation environment. Therefore, to validate the simulation results and address the discrepancy, an approximation approach, suggested by previous studies [67], was employed to better reflect local conditions.

Given that summer experiences 12 h of daytime outages (June–August), while winter sees nighttime outages coupled with increased demand for supplementary heating (December–February), a 0.5 coefficient was applied exclusively to the summer months. This adjustment aims to mitigate the disparity in power consumption error between seasons. By applying the 0.5 factor only to summer and considering power outages in both seasons, the model balances the effects of peak daytime outages in summer against nighttime outages and elevated heating demands in winter. Applying this factor to the summer electricity consumption and summing it with the initial simulated values for the other months resulted in a revised annual electricity consumption estimate of 3228 kWh. This revised value closely aligns with the actual electricity bills, showing a discrepancy of only 5.9%. This level of error is well within the acceptable range of ±15% commonly used to evaluate the reliability of simulation results and the software tools employed [68,69].

In this case study, the initial simulated values were not comparable due to significant local discrepancies that could not be modeled in DesignBuilder. Therefore, the simulated values were revised using an approximation, resulting in an error of 5.9%, which is within the permissible limit. Additionally, some errors were difficult to consider exactly similar to the existing building conditions, such as the exact timing of electricity outages, variations in the number and behaviors of occupants, and assumed factors in DesignBuilder like real building elements and heat transfer factors (U values), which may differ from the assumed values.

5. Discussion

Numerous strategies aimed at promoting sustainability and energy efficiency in building design are widely employed globally. Several studies have explored similar strategies, often employing genetic algorithms for optimization. For instance, research has utilized genetic algorithms to optimize energy consumption and thermal comfort by adjusting building layout, orientation, WWR, and construction materials [31]. Other studies have applied genetic algorithms to optimize the number, placement, and support structure of parametric façades [70] and to optimize room dimensions, WWR, solar heat gain coefficient, and visible light transmittance [71]. Additionally, genetic algorithms have been used to optimize building orientation, material selection, window design, and ventilation ratios based on life cycle energy consumption and green evaluation metrics [72].

This study focuses on the analysis and optimization of common and practical building sustainability strategies. By prioritizing optimized design strategies through genetic algorithm techniques, including building orientation, façade components, envelope design, shading strategies, and material selection, significant improvements in energy efficiency and sustainability can be achieved. The DesignBuilder v7 software was employed to analyze sustainable building strategies tailored to specific local contexts. To mitigate the complexity and potential inaccuracies associated with applying multiple strategies simultaneously, a test room was used to individually analyze and optimize each strategy. By isolating and optimizing each strategy, the optimal outcomes were determined. These optimized strategies were then integrated into an existing building model to assess their collective performance within the proposed comprehensive framework. Furthermore, CFD analysis was conducted to enhance this study’s clarity and to evaluate the impact of optimized strategies on indoor air temperature, natural ventilation, and thermal comfort. The novel approach of this study involves the optimization of individual strategies using genetic algorithms, followed by their integration into a comprehensive building simulation model. This methodology provides valuable insights into the effectiveness of various sustainability strategies and offers a systematic approach to achieving energy-efficient and sustainable building design.

The initial phase of simulations and optimizations focused on various glazing types. Trp Clr 3 mm/13 mm Arg emerged as the most energy-efficient option, while Sgl Gey 6 mm proved to be the least efficient, leading to significant energy losses throughout the year. This highlights the substantial impact of glazing composition and material properties on energy loss, solar gain, and heat loss through windows. Further analysis, depicted in Figure 17, revealed that optimizing the WWRs and shading devices across different window configurations yielded significant insights. S-facing single windows and those with S-facing orientations were identified as the most energy-efficient choices. However, optimal WWRs and shading devices are crucial to maximize their benefits. Conversely, N-facing single windows and their corresponding configurations exhibited higher energy consumption. While N-facing windows received the lowest solar gain, making them less energy-efficient for overall building performance, they could be suitable for specific applications, such as summer rooms with low cooling demands. E-facing windows received significant morning solar gain, impacting overall building energy. W-facing windows received high afternoon solar gain, significantly impacting energy consumption. Careful consideration of WWRs and shading devices is essential to harness their potential benefits.

Materials play a pivotal role in achieving sustainable and energy-efficient building designs. Insulated building components, such as walls, roofs, and floors, are crucial for effective thermal insulation. By reducing heat transfer, these components significantly minimize the need for excessive heating or cooling demands, leading to lower energy consumption and a reduced environmental footprint. The analysis strongly suggests that plastered brick walls with 0.15 m of EPS insulation offer the best balance of energy efficiency and sustainability. This combination effectively reduces heat loss and gain, optimizing thermal performance. In contrast, plastered brick masonry and stone masonry without insulation are significantly less energy-efficient and resource-intensive. For roofing systems, concrete structures with 0.15 m of EPS insulation emerged as the most energy-efficient option. This configuration provides superior thermal performance, reducing heat loss and gain. Conversely, uninsulated concrete roof structures are less effective in maintaining thermal comfort. Regarding flooring, uninsulated floors with carpeting were found to be the most suitable choice. This combination offers a balance of thermal comfort, providing warmth in winter and a cool surface in summer. By carefully selecting materials and incorporating insulation, it is possible to design buildings that are both energy-efficient and environmentally friendly.

Consequently, this study proposes a significant design model for the study area, incorporating the most common and effective sustainable building strategies. These strategies, including optimized glazing types, WWRs, shading techniques, and construction materials, were derived from a genetic algorithm analysis of individual strategies. The optimized characteristics of each strategy are provided in

Table 8. Optimized strategies derived from a comprehensive analysis of each strategy using a genetic algorithm in a test room.

Optimized Strategies	Single-Sided Windows				Opposing Windows				Adjacent Windows
Orientation	N	E	S	W	NS		EW		NE		ES		SW		WN
					N	S	E	W	N	E	E	S	S	W	W	N
WWR (%)	60	60	60	50	10	60	60	30	10	60	10	60	60	10	60	10
VSA (degree)	0	65	80	55	0	75	70	65	0	75	55	85	85	50	50	0
Window glazing	Trp Clr 3 mm/13 mm Arg (glazing type 6)
Wall	Outside plaster 15 mm + EPS 150 mm + brick 320 mm + inside plaster 15 mm (wall Type 10)
Roof	Bitumen sheet 5 mm + cement mortar 50 mm + EPS 150 mm + slab (co-reinforced) 120 mm + plaster 150 mm (roof type 5)
Ground floor	Carpet 5 mm + ceramic 10 mm + cement mortar 30 mm + concrete 100 mm + gravel 200 mm (ground floor type 2)
Partition wall	Plaster 15 mm + brick 220 mm + plaster 15 mm
Internal floor (floor slab)	Carpet 5 mm + ceramic 10 mm + cement mortar 30 mm + slab (co-reinforced) 120 mm + plaster 150 mm

.

This study investigated the energy performance of a pre-existing building through simulation analysis and optimization. A detailed building model, incorporating the proposed strategies outlined in Table 8, was developed to assess the baseline and optimized configurations. The primary objective was to quantify the potential energy savings achievable through strategic interventions. The results indicate a significant reduction in energy loss through windows, primarily attributed to the optimized glazing type, WWR, and shading devices aligned with appropriate VSA. Compared to the baseline configuration (single-glazed, unoptimized WWR, and shading), the optimized design resulted in an 11,701 kWh decrease in annual window heat loss. Similarly, the optimization of wall and roof construction materials led to substantial energy savings. The adoption of plastered brick walls with 0.15 m EPS insulation reduced wall heat loss by 14,601 kWh annually, compared to the baseline uninsulated walls. Likewise, the incorporation of 0.15 m EPS insulation in the roof reduced roof heat loss by 6666 kWh annually. The analysis suggests that, for the specific conditions of this study, the ground floor without insulation represents an appropriate design choice. The comparative analysis of energy loss and gain from both existing and optimized building components is presented in Figure 18.

By optimizing the WWR and implementing effective shading devices, the building’s natural ventilation rate is significantly improved. This results in an average increase of 0.6 Air Changes per Hour (ac/h) hour during hot periods, leading to a more comfortable indoor environment and reduced reliance on mechanical cooling systems. Additionally, the optimized shading strategies effectively mitigate solar heat gain, contributing to an annual energy reduction of up to 12,434 kWh. This strategic control of solar radiation and enhanced natural ventilation contributes to efficient heating in winter and cooling in summer, as illustrated in Figure 19.

During the peak summer months (June to August), solar heat gain in the pre-existing building was substantial, reaching 8587 kWh. By implementing optimization strategies, this value was significantly reduced to 4737 kWh. This reduction directly contributed to a decrease in cooling demand. Conversely, during the coldest winter months (December to February), solar heat gain in the pre-existing building was 10,797 kWh, which was reduced to 8316 kWh in the optimized building. While this reduction slightly increased heating demand, the arid and semi-arid climate of the study area mitigated the negative impact of reduced solar gain during winter. However, the combined effect of these optimizations resulted in a substantial reduction in the building’s total energy demand. Overall, in its optimized form, the building significantly reduces both cooling and heating demands compared to the baseline configuration. Specifically, cooling demand is reduced by 4641 kWh during hot months, while heating demand is slashed by 26,650 kWh during cold months, as supported by [39]. As a result, the overall annual energy demand of the building is reduced by a substantial 31,190 kWh, as illustrated in Figure 20.

The optimization of pre-existing buildings significantly contributes to reducing CO₂ emissions, improving natural ventilation, and enhancing occupant comfort. As illustrated in Figure 21, the optimized building exhibits substantial reductions in CO₂ emissions throughout the year due to decreased energy consumption. Notably, the peak CO₂ emissions of the existing building, reaching 1800 kg, were reduced to 500 kg in the optimized building, signifying a 1300 kg reduction.

To further analyze the impact of air fluctuations on thermal comfort, CFD simulations were conducted. As shown in Figure 22, the pre-existing building, with its single-side ventilation, suboptimal WWR, and shading strategies, exhibited uneven temperature distribution and limited thermal comfort. The temperature distribution ranged from 25.8 °C near the windows to a higher temperature of up to 27.5 °C in most areas of the building, especially on the first floor. Air velocities were measured to range from 0.5 to 1.8 m/s, with higher velocities observed near windows. These conditions underscored the shortcomings of energy-efficient design strategies in existing buildings, particularly in relation to natural ventilation. This deficiency negatively impacted both thermal comfort and energy demand.

In contrast, the optimized building, with its well-designed cross-ventilation, Window-to-Wall Ratio (WWR), and shading strategies, exhibited a more uniform temperature distribution and improved thermal comfort, as illustrated in Figure 23. The temperature distribution throughout the building was more consistent, with most rooms maintaining temperatures between 25 °C and 25.5 °C, similar to the outdoor temperature. Additionally, air velocities were generally higher, ranging from 0.8 to 3.0 m/s, indicating suitable air movement in most areas of the building. These factors significantly contributed to the increased thermal comfort and reduced energy demand.

In addition, to assess occupant comfort in both the existing and optimized buildings, Predicted Mean Vote (PMV) analysis was conducted using CFD simulations. PMV, a key indicator derived from the Fanger model, predicts the overall thermal comfort for a large population under specific environmental conditions. A PMV value of +3 indicates very hot conditions, while −3 indicates very cold conditions. This study focused on the hottest summer period to analyze PMV values for both buildings. As shown in Figure 24, PMV values in the existing building ranged from −0.33 to 1.16, indicating significant variations in comfort levels within different rooms. Rooms with single-side ventilation, in particular, exhibited higher temperatures, potentially leading to discomfort for occupants.

However, after implementing the optimized strategies, the thermal comfort within the building significantly improved. As depicted in Figure 25, the PMV values in the optimized building range from −0.14 to 0.23, indicating a more suitable and nearly neutral comfort level for occupants in all rooms.

By implementing the optimized strategies outlined in this study, it is possible to significantly reduce heating and cooling loads, potentially leading to zero-energy buildings. This achievement not only optimizes energy consumption but also enhances indoor environmental comfort, reduces CO₂ emissions, and promotes material conservation. The application of these strategies is crucial for advancing the energy efficiency and sustainability of buildings.

Additionally, this study emphasizes the potential of local and passive design strategies to be applied holistically without the need for additional economic investments or advanced technologies. However, it is essential to consider incorporating advanced technologies and strategies, such as Phase Change Materials, kinetic shading devices, and renewable energy systems, into sustainable building design for more efficient performance. Therefore, integrating these modern technologies in future studies, achieving zero-energy buildings, and significantly contributing to environmental sustainability becomes a viable and impactful goal.

6. Conclusions

The findings of this study clearly demonstrate the significant potential of genetic algorithms and simulation-based analysis in optimizing building design strategies to achieve energy-efficient and sustainable buildings. By carefully optimizing building components, such as WWR, shading devices, glazing types, and construction materials for walls, roofs, and ground floors, substantial reductions in energy loss and gain can be realized.

In the arid and semi-arid climate of the study area, this research underscores the substantial influence of optimized WWR and shading device configurations on achieving energy-efficient building designs tailored to specific site conditions. The optimal WWR and VSA for shading devices vary depending on window orientation and local climatic factors. In this particular case, S-facing orientations outperformed others in both single-side and dual-side configurations. These orientations required higher WWRs and VSAs while benefiting from partial shading strategies. E and W orientations across all window configurations necessitated significant shading solutions, particularly during peak solar radiation periods, leading to lower VSAs. Conversely, N orientations, experiencing minimal solar gain, did not require shading devices. Overall, among all window configurations, those featuring S-facing windows, especially in adjacent configurations, emerged as the optimal choice for energy-efficient building designs in this specific climatic context.

This study highlights the superior insulating properties of triple-pane glazing filled with argon gas compared to conventional single-pane and other multi-pane configurations. Furthermore, incorporating EPS insulation, particularly for walls and roofs, significantly improved building thermal performance and reduced energy loss. However, omitting insulation in internal partition walls, floor slabs, and the ground floor proved to be an effective strategy for optimizing building envelopes without compromising energy performance. The implementation of these optimized strategies and materials on a pre-existing building with single-pane glazing and uninsulated external walls resulted in significant energy savings. Window energy loss was reduced by 58.6%, wall energy loss by 78.3%, and roof energy loss by 69.5%.

Consequently, by comparing the unoptimized pre-existing building with the optimized building outlined in this study, cooling demand decreased by 48.1%, and heating demand reduced by a significant 97.5%. Overall, the optimized building achieved an impressive 84.4% reduction in total energy demand. This led to a 72.2% reduction in CO₂ emissions during peak demand months due to the substantial decrease in energy consumption. Additionally, the optimized building resulted in an average 6% increase in natural ventilation and nearly neutral comfort levels for occupants in all rooms. These findings emphasize the significance of employing advanced optimization techniques in building design to realize substantial energy savings and environmental benefits.