Simulation-Based Multi-Objective Optimization for Building Retrofits in Iran: Addressing Energy Consumption, Emissions, Comfort, and Indoor Air Quality Considering Climate Change

Abstract

Climate change poses significant challenges to energy efficiency and occupant comfort in residential buildings. This study introduces a simulation-based multi-objective optimization approach for architectural design, aimed at addressing these challenges and enhancing environmental sustainability. Utilizing EnergyPlus for energy simulations and jEPlus to identify objective functions and design parameters, the research employed the NSGA-II algorithm through jEPlus + EA for multi-objective optimization. A Morris sensitivity analysis assessed the impact of 25 design variables—including heating and cooling setpoints, air infiltration rates, insulation types, window selections, airflow rates, and HVAC systems—on key objective functions. Applied to a residential building in Sari, Iran, the study analyzed various climate change scenarios to minimize five main objectives: primary energy consumption, greenhouse gas emissions, indoor air quality, predicted percentage of dissatisfied, and visual discomfort hours. The weighted sum method was used to select optimal solutions from the Pareto front. Results demonstrated that the recommended energy retrofit strategies could reduce primary energy consumption by up to 60%, greenhouse gas emissions by 60%, predicted thermal dissatisfaction by 65%, and visual discomfort hours by 83%, while also achieving indoor air quality levels that meet ASHRAE recommended standards. However, the implementation of these energy-efficient solutions may require careful consideration of trade-offs in design decisions when addressing climate change challenges.

1. Introduction

The 2030 Agenda and the Sustainable Development Goals (SDGs) represent a global commitment to promoting sustainable development and addressing critical challenges such as climate change, energy efficiency, and environmental sustainability. Residential buildings play a pivotal role in achieving these global energy objectives, as they account for a substantial portion of energy consumption and carbon emissions worldwide. To align with the SDGs, it is essential to implement advanced simulation tools in the building sector to enhance energy efficiency and mitigate environmental impacts effectively. By modeling and analyzing various dimensions of residential structures—such as energy consumption, thermal performance, and indoor environmental quality—designers and engineers can rigorously evaluate multiple design options, systems, and materials. This proactive evaluation facilitates the identification of optimal energy-efficient solutions prior to construction or retrofitting, ultimately leading to enhanced building performance and improved energy management. Retrofitting existing residential buildings is a vital strategy for advancing energy efficiency and sustainability, particularly in the context of the SDGs. This process involves upgrading or modifying building elements and systems to decrease energy consumption while enhancing overall performance. Key retrofit strategies include improving insulation, upgrading HVAC systems, installing energy-efficient lighting, and integrating renewable energy technologies. These measures not only reduce energy demand but also significantly lower carbon emissions by decreasing reliance on fossil fuels. The integration of simulation tools with retrofitting initiatives enables a targeted approach to identifying specific areas for improvement, evaluating potential energy savings, and understanding the environmental benefits of various retrofit measures. This comprehensive methodology supports strategic decision-making by prioritizing interventions that yield the most significant impact on energy efficiency and carbon reduction, thus contributing to the broader goals set by the 2030 Agenda. Overall, the synergy of simulations and retrofitting represents a robust framework for advancing the sustainability of the residential building sector, paving the way for a more environmentally responsible future in line with the SDGs.

The study conducted by Pan et al. underscores the critical role of building performance simulation (BPS) and building energy modeling (BEM) in achieving energy-efficient building designs, operations, and retrofits. It reviews the current state of research, identifies emerging trends, and explores future opportunities for utilizing BEM across various stages and scales of building projects, while offering recommendations for further advancements in the field [1]. In their assessment, Naderi et al. emphasizes the optimization of architectural specifications and control parameters for smart shading systems. Their findings reveal considerable potential for energy savings and improved occupant comfort through the proposed optimization approach [2]. The research carried out by Baghoolizadeh et al. on indoor air quality in the residential sector is of critical importance. This study investigates various design factors and applies optimization techniques to concurrently lower CO₂ concentrations and pollutant levels, while enhancing thermal comfort for building occupants. The results indicate significant improvements in these parameters [3]. The exploration conducted by Nateghi et al. offers a thorough methodology for optimizing the energy demands of multistory hotels. Their findings indicate significant energy savings achieved through the optimization of building parameters. Additionally, they propose reinvesting the resulting cost savings into renewable energy systems to improve overall sustainability [4]. The investigation performed by Tavakolan et al. highlights significant challenges associated with energy retrofitting, particularly in Iran, where distinctive energy pricing policies obstruct the feasibility of these projects. This research presents a simulation-based multi-objective optimization framework and underscores the necessity for policy reforms to improve the economic viability of energy retrofit initiatives [5]. The study undertaken by Ziaee and Vakilinezhad offers crucial insights for designers in selecting the optimal properties of light shelves and identifying the best design options for urban settings such as Tehran and Sari [6]. The analysis completed by Dokhanian et al. on energy, environmental, and economic considerations throughout the life cycle of educational buildings is crucial. Their research on shading strategies for an educational building in Sabzevar City identifies concrete overhang shading as the most effective solution, demonstrating the effectiveness of simple overhangs and light-shelf structures made from concrete and wood materials [7]. The research conducted by Naderi et al. assesses the design parameters of direct evaporative cooling (DEC) systems in four diverse climates within Iran. It evaluates the impacts of cooling pad saturation efficiency, airflow rate, and thermostat set points on electricity and water consumption, utilizing EnergyPlus simulations for a standard residential apartment [8]. The research organized by Mostafazadeh et al. aimed at enhancing building energy performance for sustainability and climate change mitigation. In Iran, retrofitting is complicated by subsidized energy prices and conflicting objectives. This study introduces prNSGA-III, a modified NSGA-III algorithm that employs parallel computing and a results-saving archive for efficiency. The methodology combines EnergyPlus with MATLAB for a comprehensive analysis of active, passive, water conservation, and renewable strategies, optimizing environmental performance and minimizing thermal discomfort and life cycle costs in a residential building in Tehran while considering future climate changes and energy price fluctuations [9]. According to the assessment directed by Alimohamadi and Jahangir, given the substantial energy consumption of residential buildings, targeted interventions are essential to mitigate their environmental and economic impacts. In Iran, challenges related to energy calculations and pricing hinder retrofitting efforts. This research integrates MATLAB and EnergyPlus to optimize three objective functions—energy consumption, thermal comfort, and net present values—using the NSGA-II optimization algorithm. Key decision variables include HVAC systems, set temperatures, insulation, and window types. The study also analyzes the effects of energy inflation on these objectives, revealing that many optimal solutions may be unprofitable under current pricing, with the highest profitability reaching [10]. The evaluation performed by Karimi et al. introduces an innovative optimization framework that combines Bayesian optimization, XGBoost algorithms, and multi-objective genetic algorithms to enhance the performance of non-residential buildings, focusing on total energy (TE), indoor overheating degree (IOD), and predicted percentage dissatisfied (PPD) across historical and future scenarios. By utilizing IOD as a key performance indicator, the framework provides a comprehensive assessment of thermal discomfort and energy efficiency, addressing limitations of traditional metrics. Key design factors, such as window-to-wall ratio and outdoor temperature, are identified as significant influences on performance metrics, underscoring the need for effective management of building envelopes and HVAC systems in response to evolving climate conditions [11]. The exploration conducted by Karimi et al. analyzes a decade of research (2010–2020) on urban heat islands and their mitigation strategies, based on 91 studies. It highlights the predominant focus on humid subtropical, hot summer Mediterranean, and temperate oceanic climates, emphasizing the critical role of urban parks, trees, and green roofs in reducing urban heat. The findings suggest a balanced distribution of research across various mitigation measures, indicating that funding levels do not significantly impact the effectiveness of these strategies [12].

This study presents a comprehensive framework for energy retrofit projects specifically designed for buildings in Iran, particularly in the humid subtropical regions bordering the Caspian Sea. While previous research as follows

Table 1. The number of studies conducted on multi-objective optimization in the building sector.

Authors	Objective Functions	Design Variables	Application	Geo. Regions	Ref.
Mostafazadeh et al. (2023)	Discomfort hours, environmental performance, life cycle cost	Envelope insulations, HVAC systems, window properties, PV solar cells, domestic faucets	Residential buildings	Iran	[9]
Tavakloan et.al (2022)	Primary energy consumption, net present value, discounted payback period (NSGA-II)	Envelope insulations, HVAC systems, window properties, PV solar cells	Residential buildings	Iran	[5]
Naderi et al. (2020)	Energy consumption, thermal and visual discomfort	Shading control strategy, temperature set-point, solar irradiance set-points, load set-points	Office buildings	Iran	[2]
Mauro et al. (2015)	LCC, energy consumption (sensitivity analysis)	Envelope characteristics (wall, roof, and window), HVAC systems, renewable energy	Office buildings	South Italy	[13]
Duash et al. (2016)	Heating and cooling energy consumption, thermal comfort (sensitivity analysis)	Envelope characteristics (wall, roof, and window), HVAC systems	Residential buildings	United States	[14]
Lin et.al (2022)	Energy Consumption, Thermal comfort (ANN + Whale Optimization Algorithm)	Air condition and mechanical ventilation (ACMV) characteristics	Residential buildings	Taiwan	[15]
Chegari et.al (2021)	Annual weighted average discomfort degree hours, annual thermal energy demand (ANN + NSGA-II)	Building envelope, ACH, glazing	Residential buildings	Morocco	[16]
Jung et.al (2021)	Building energy demand, life cycle cost, life cycle assessment (ANN + NSGA-II)	Building envelope, ACH, WWR, number of occupants, window properties	Residential buildings	Korea	[17]
Wu (2021)	Energy consumption, thermal comfort (NSGA-II)	Primary airflow rate, chilled water flow rate	Sample room	Singapore	[18]
Bre et al. (2020)	Energy demand, thermal comfort (ANN + NSGA-II)	Building azimuth, building envelopes, air mass flow, window properties	Residential buildings	Argentina	[19]
Ebrahimi-Moghadam et.al (2020)	Total energy, predicted percentage of dissatisfied (NSGA-II)	Shelf properties	Residential buildings	Mashhad, Iran	[20]
Yao et al. (2018)	Energy demand, thermal comfort (parametric analysis)	Building azimuth, wall properties, WWR, window properties, ACH	Residential buildings	Chongqing, Shanghai, Changsha	[21]
Gou et.al (2018)	Building energy demand, comfort time Ratio (ANN + NSGA-II)	Building orientation, window properties, building envelope, wall insulations	Residential buildings	Shanghai	[22]
Li et.al (2017)	Life cycle costing, carbon dioxide emissions, the amount of thermal discomfort throughout the year (C+++ NSGA-II)	Building envelope, building azimuth, WWR	Residential buildings	Nanjing	[23]
Bre et al. (2017)	Energy demand, thermal comfort (GA + Morris screening method for sensitivity analysis)	Building azimuth, building envelopes, window and door properties	Residential buildings	Argentine	[24]
Wang et al. (2024)	Building energy consumption, thermal comfort (support vector regression non-dominated genetic algorithm-II (SVR-NSGA-II))	Parameters of the building envelope are as follows: the exterior wall U-value, roof U-value, exterior wall U-value, solar heat gain coefficient value (SHGC), and south, north, east, west window-to-wall ratios	Office buildings	Chongqing	[25]

, has focused on limited energy efficiency measures and climate impacts, this study addresses the integrated effects of climate change on the combined use of active, passive, and renewable strategies to enhance energy performance, occupant comfort, and indoor air quality. Despite the urgent need for adaptation to climate change, there is a notable lack of studies on its implications for the residential building sector in these unique climatic contexts. This research aims to fill this gap by examining how projected climate scenarios influence building design strategies in humid subtropical regions.

2. Methodology

The proposed methodology is organized into four phases, which are explained in the following sections and illustrated in Figure 1:

Phase 1 involves the development of the initial building model using input data obtained from comprehensive audits, followed by energy simulations to calibrate the baseline model effectively.

In Phase 2, objective functions are defined, and parametric energy retrofit measures (ERMs) are integrated into the building model to evaluate potential enhancements.

Phase 3 employs the NSGA-II algorithm, which iteratively refines the combinations of ERMs, resulting in a set of Pareto optimal solutions that balance trade-offs among multiple objectives.

Finally, Phase 4 implements a multi-criteria decision-making approach using the weighted sum method to derive the final retrofit strategies, tailored according to the preferences of stakeholders.

The research process involves the following steps in more detail:

• Building architecture modeling: architecture modeling is carried out using the existing building plan and architectural drawings by SketchUp to create an accurate representation of the building.
• Collecting data: Data collection is crucial when assessing a building’s energy performance and planning for retrofitting projects. Each building has unique characteristics that influence its energy consumption, and gathering accurate data is essential for meaningful analysis and recommendations. These data can be categorized into the following types: building layout and general information, occupancy information, construction details, lighting, and specifications for mechanical and electrical systems.
• Building energy simulation: EnergyPlus is used to perform detailed building energy simulations. These simulations capture the energy performance of the building under various conditions, creating a baseline EnergyPlus model of the building studied.
• Calibration of building energy model: The calibration of building energy models is a critical process aimed at aligning simulated energy consumption predictions with actual energy usage, primarily to enhance accuracy and reliability in energy performance assessments. The calibration process involves comparing simulation results against actual data from energy bills and adjusting the model to reduce discrepancies. The energy simulation outcomes are compared with actual energy consumption data sourced from energy bills to validate the model. If discrepancies arise between the results, modifications to the model are necessary. This iterative process continues until the model is calibrated to an acceptable level of accuracy and accurately reflects real-world conditions. While acceptable calibration thresholds can vary, a CV(RMSE) of less than 15% and an NMBE closer to 0% (between −5% to +5%) are typically considered good practice [26].
• Identifying the weaknesses of the model and energy waste: The energy performance results are analyzed to identify elements with significant energy waste, highlighting areas with high potential for improvement. This information informs us of the subsequent steps in defining decision variables.
• Parametric study: a comprehensive parametric study is conducted using EnergyPlus to gain insight into how various factors influence building energy performance, facilitating informed decisions for design or retrofitting projects.
• Energy retrofit measures (ERMs): This section introduces various energy retrofit measures (ERMs) that are designed to enhance building performance. The alternatives for each decision variable are suggested based on factors such as market availability, stakeholders’ requirements, climatic conditions, and the characteristics of the building. A comprehensive set of ERMs—encompassing passive, active, and renewable strategies—is examined with the dual objectives of minimizing primary energy consumption and maximizing environmental performance by reducing carbon emissions. The most significant ERMs investigated in this study include the following:

  ▪

  Building envelope: enhancements in thermal insulation for floor and roof, and replacement of exterior door and windows (passive measures),

  ▪

  Controlling infiltration: improving window/door airtightness (passive measures),

  ▪

  Building physics: optimizing north window-to-wall ratio, installing overhangs and fins (passive measures),

  ▪

  Daylighting controls: implementing stepped, continuous, continuous-off lighting controls (passive measures),

  ▪

  Window shading control: applying shading control strategies (passive measures),

  ▪

  Intelligent control systems: utilizing an energy management system (EMS) EnergyPlus to improve indoor air quality (active measures),

  ▪

  HVAC systems: adjusting heating and cooling setpoint and replacing existing systems (active measures),

  ▪

  Renewable energy sources: installing photovoltaic (PV) panels (renewable measures).

• Scenario development: Two scenarios are explored in relation to future climate change. The first scenario assesses the case study under current climate conditions, while the second scenario considers the potential impacts of climate change. A simulation-based multi-objective optimization (SBMO) framework is employed for each scenario, as varying weather files influence the solutions along the Pareto front. Scenario 1—current climate conditions: This scenario utilizes present weather data for the building performance simulation process. It relies on baseline data that reflect current climate conditions, including temperature, precipitation patterns, and seasonal variations, with observations grounded in existing environmental conditions and practices. Scenario 2—climate change projections: In this scenario, the CCWorldWeatherGen software V1.9, developed by Jentsch et al. [27], is utilized to generate EnergyPlus Weather (EPW) and Typical Meteorological Year Data (TMY2) weather files for the A2 emission scenario (medium-high). This approach estimates future energy consumption and greenhouse gas emissions for the years 2050 and 2080, based on the HadCM3 (Hadley Centre Coupled Model version 3). Objective functions and decision-making: jEPlus is used to determine the objective functions and decision-making parameters for the optimization. These functions represent the performance metrics to be minimized, and the decision-making parameters define the design variables to be optimized.
• Sensitivity analysis: Morris sensitivity analysis with jEPlus + EA is conducted to understand how the input variables affect the objective functions. This analysis helps identify the most influential design variables.
• Multi-objective optimization: The NSGA-II evolutionary algorithm, integrated with jEPlus + EA, is employed for multi-objective optimization. NSGA-II iteratively evolves a population of candidate solutions, applying genetic operators such as selection, crossover, and mutation to explore the design space and find optimal or near-optimal solutions.
• Evaluation and selection: The optimization process generates a set of Pareto solutions, representing trade-offs between the multiple objectives. The weighted sum method is used to select the final solution from the Pareto front, considering the specific preferences or priorities of the decision-maker. The weighted sum method is a simple and commonly used approach in multi-objective optimization to select the final solution from a set of Pareto optimal solutions.

The approach integrates detailed architectural modeling, energy simulation, sensitivity analysis, and optimization techniques to support sustainable building design and decision-making. Understanding both scenarios can ultimately aid in engaging various stakeholders, from governments to local communities, in taking collaborative action toward climate-resilient futures.

3. Objective Functions

3.1. Environmental Objective Functions

The environmental objective function encompasses various metrics aimed at optimizing building performance while minimizing negative impacts on the environment and enhancing occupant comfort.

3.1.1. Energy Consumption

In the presented framework for assessing the environmental impacts of building energy retrofits, primary energy consumption (PEC) is employed as a key metric, as preferred in the recast of the Energy Performance of Buildings Directive (EPBD). PEC is calculated by converting net electricity and natural gas demands into primary energy consumption values using established primary energy factors relevant to the local context. Each type of energy source is assigned to a primary energy factor (PEF) or conversion factor that reflects the energy loss that occurs in the conversion and distribution processes. These factors are usually determined by local regulations or energy efficiency standards. Monthly energy demand figures are derived from dynamic simulations conducted with EnergyPlus software V24.1, which models a building’s energy use under various conditions. Notably, this framework incorporates rooftop photovoltaic (PV) panels as a retrofitting measure. The energy generated by these PV panels is subtracted from overall energy consumption, resulting in a reduced net demand on the electrical grid [13,28]. This holistic approach underscores the importance of integrating renewable energy sources and enhances the overall sustainability assessment of building retrofits. To convert site energy to source energy in EnergyPlus, we need to use specific conversion factors that account for the energy losses during production, transmission, and delivery. For Iran, the conversion factors can vary based on the efficiency of the energy infrastructure. The typical conversion factors used globally, which can be adapted for Iran, are as follows:

• Electricity: The site-to-source conversion factor for electricity is generally around 3.15 [29]. This means that for every unit of electricity consumed onsite, approximately 3.15 units of energy were used at the source to generate and deliver that electricity.
• Natural gas: The site-to-source conversion factor for natural gas is typically around 1.05 [29]. This reflects the relatively lower losses in the production and delivery of natural gas compared to electricity.

These factors can be adjusted based on the specific efficiencies of Iran’s energy infrastructure.

3.1.2. Greenhouse Gas Emissions

Greenhouse gases (GHGs) are gases in the Earth’s atmosphere that trap heat. They play a critical role in regulating the planet’s temperature through the greenhouse effect. Understanding greenhouse gases and their emissions is crucial for addressing climate change and its associated risks. Reducing GHG emissions can help mitigate climate impacts and contribute to a more sustainable future [30,31]. CO₂-eq, or carbon dioxide equivalent, is a standard unit for measuring the impact of various greenhouse gases (GHGs) in terms of their global warming potential (GWP). It expresses how much warming gas causes compared to CO₂ over a specific time frame, typically 100 years. Organizations and individuals use CO₂-eq to measure and report their overall carbon footprint, helping to understand their contributions to climate change [32,33]. In Figure 2, the proportion of greenhouse gas emissions is presented as a percentage.

Climate Change Impacts

Increasing concentrations of GHGs from human activities enhance the greenhouse effect, leading to global warming and climate change. Some of the impacts include the following:

• Rising temperatures: increased average temperatures and more frequent heatwaves.
• Severe weather events: more frequent and intense hurricanes, droughts, and floods.
• Melting ice and rising sea levels: Arctic and Antarctic ice melt contributes to rising sea levels, threatening coastal areas.
• Ecosystem disruption: changing habitats affecting biodiversity, with some species struggling to adapt to rapid changes [34].

Emission Factors

In the context of climate change and environmental policy, understanding the emission factors associated with greenhouse gases is crucial for devising effective mitigation strategies. In Iran, the building sector serves as a significant contributor to carbon dioxide (CO₂), methane (CH₄), and nitrous oxide (N₂O) emissions. The numerical values of these emission factors are pivotal in determining the overall emissions from this sector, thereby informing policy decisions and sustainability efforts. Emission factors are essential for estimating greenhouse gas emissions from different fuel types, electricity generation, and specific activities. In Iran, a typical set of emission factors can be derived from national inventories and available studies on emissions from different sectors:

• Electricity generation: in Iran, the emission factor for electricity generation is approximately 565 g of CO₂ per kWh [35].
• Natural gas: the average emission factor for natural gas in Iran is approximately 53.06 kg CO₂ per MMBtu [36] (181.07 g CO₂ per kWh).

Iran’s emissions primarily come from the following sectors:

• Household energy: residential areas significantly contribute to CO₂ emissions, with estimates suggesting that households accounted for approximately 23.4% of total CO₂ emissions in the country [37]. This is connected to energy consumption patterns, where fossil fuels predominately supply energy.
• Industrial sector: this sector is also a major contributor, responsible for around 24.1% of CO₂ emissions [37].
• Transportation: accounts for about 21.2% of CO₂ emissions, highlighting the impact of vehicle fuel consumption [37].

The calculation of carbon dioxide equivalent is as follows:

$${CO}_{2}e=EnergyConsumption\times EmissionFactor$$

(1)

To calculate CO₂ equivalent emissions (CO₂e) in Iran, one needs to utilize specific emission factors that define the greenhouse gas emissions associated with various sources and activities. These factors help in converting the emissions of different greenhouse gases into a common unit, allowing for an aggregation of their impacts. In Iran, the emission factors for methane (CH₄) and nitrous oxide (N₂O) associated with electricity and natural gas consumption are significant for assessing greenhouse gas emissions. Emission factors in g/MJ:

Electricity generation:

• Methane (CH₄) emission factor: 1.4 g CH₄ per MJ of electricity produced [38].
• Nitrous oxide (N₂O) emission factor: 0.03 g N₂O per MJ of electricity produced [38].
• Natural gas consumption:
• Methane (CH₄) emission factor: 5 g CH₄ per MMBtu of natural gas. With a conversion of 1 MMBtu ≈ 1.055 GJ, this equates to approximately 4.73 g CH₄ per MJ [39].
• Nitrous oxide (N₂O) emission factor: 0.1 g N₂O per MMBtu of natural gas, which is about 0.094 g N₂O per MJ [39].

The emission factors indicate a higher methane emission associated with natural gas compared to electricity. In summary, these values illustrate the importance of transitioning to cleaner energy sources to mitigate greenhouse gas emissions in Iran.

Global Warming Potentials

The GWP is a key metric developed by the Intergovernmental Panel on Climate Change (IPCC) that quantifies the warming impact of greenhouse gases compared to CO₂ over a specified time horizon, commonly 100 years. When calculating CO₂e, it is essential to also consider the global warming potential (GWP) of other gases.

• Methane (CH₄), with a GWP of 28 (over 100 years), has a significant impact when converted to CO₂e, which is crucial in comprehensive emissions accounting [36].

$${CO}_2{e}_{{CH}_4}=Massof{CH}_4\left(Metrictonnees\right)\times GWPof{CH}_4\left(28\right)$$

(2)

• Nitrous oxide (N₂O): The GWP for N₂O is typically around 298 for a 100-year time horizon [40]. This means that one unit of N₂O is equivalent to 298 units of CO₂ in terms of its potential impact on warming over the next century.

$${CO}_2{e}_{N_{2}O}=Massof{N}_{2}O\left(Metrictonnes\right)\times GWPof{N}_{2}O\left(298\right)$$

(3)

$$Total{CO}_{2}e={CO}_2{e}_{electricity}+{CO}_2{e}_{NaturalGas}+{CO}_2{e}_{{CH}_4}+{CO}_2{e}_{N_{2}O}$$

(4)

3.1.3. Indoor Air Quality Objective Function

The indoor air quality (IAQ) objective function focuses on quantifying and optimizing factors that influence air quality within indoor environments. Key considerations for IAQ include:

• Pollutant levels: monitoring concentrations of CO₂ and other harmful substances is crucial for ensuring occupant health and productivity.
• Ventilation rates: adequate ventilation is essential to dilute indoor pollutants and maintain air quality.
• Humidity control: keeping relative humidity within an optimal range (30–60%) to prevent mold growth and enhance comfort.

By utilizing this objective function, we aim to create healthier indoor environments that promote occupant well-being.

3.1.4. Thermal Comfort Objective Function

This method is applicable in environments where occupants engage in activities resulting in average metabolic rates between 1.0 and 2.0 met, and where the clothing worn offers a thermal insulation of 1.5 Clo or less. The ASHRAE thermal sensation scale, designed to quantify individuals’ perceptions of temperature, is defined as follows: +3 for hot, +2 for warm, +1 for slightly warm, 0 for neutral, –1 for slightly cool, –2 for cool, and –3 for cold. The predicted mean vote (PMV) model uses principles of heat balance to correlate six key factors influencing thermal comfort with the average responses of individuals on the scale. The predicted percentage dissatisfied (PPD) index is connected to the PMV, as shown in Figure 3.

$$PMV=f\left(I_{cl},M,t_a,t_{mrt},P_a,v\right)$$

(5)

$$PPD=100-95\exp \left(-0.03353{PMV}^4-0.2179{PMV}^2\right)$$

(6)

This index is derived from the premise that participants who rate their comfort as +2, +3, –2, or –3 are considered dissatisfied, and it simplifies the PPD to be symmetrical around the neutral PMV point.

Table 2. A suitable thermal environment for overall comfort [41].

PPD	PMV Range
<10	−0.5 < PMV Range < +0.5

specifies the recommended ranges for PPD and PMV for typical settings. The comfort zone is defined as the range of combinations of the six thermal comfort factors where the PMV remains within the recommended limits outlined in Table 2. The PMV model is calculated based on the relevant air temperature and mean radiant temperature, as well as considering the metabolic rate, clothing insulation, air speed, and humidity. If the PMV value produced by the model falls within the suggested range, it indicates that the environmental conditions are within the comfort zone [41].

3.1.5. Visual Comfort Objective Function

Visual comfort is crucial for occupant well-being and productivity in built environments, with glare being a significant concern among its various influences. EnergyPlus, an advanced building energy simulation software, provides robust capabilities for assessing glare and other visual comfort metrics. One key output is the Daylighting Reference Point 1 Glare Index Setpoint Exceeded Time, which helps evaluate visual comfort in daylighting spaces. By analyzing this exceeded time, designers and researchers can pinpoint periods of unacceptable glare, informing design decisions such as adjusting window sizes, incorporating shading devices, or reconfiguring layouts. The glare index quantitatively measures glare at a specific reference point, often derived from metrics like the Unified Glare Rating (UGR), Daylight Glare Probability (DGP), and Daylighting Glare Index (DGI). EnergyPlus calculates glare index values based on luminance levels at this reference point, considering factors like window placement, shading strategies, and surface reflectance. According to

Table 3. Glare regions and their related glare index [44].

Zone	Region	DGI	UGR
Discomfort zone	intolerable	>28	>28
	just intolerable	28	28
	uncomfortable	26	25
	just uncomfortable	24	22
Comfort zone	acceptable	22	19
	just acceptable	20	16
	noticeable	18	13
	just perceptible	16	10

, the setpoint indicates a predefined threshold for acceptable glare levels, determined by established design standards such as DGI and UGR. The Exceed Time reflects the duration during which the glare index surpasses this threshold, providing essential information for assessing compliance with visual comfort standards and guiding architectural improvements. By integrating glare assessment tools, EnergyPlus facilitates a comprehensive evaluation of visual comfort, contributing to healthier and more productive indoor environments.

$$DGI=10{\mathit{log}}_{10}⌈0.478{\sum }_{i=1}^n\left({\displaystyle \frac{L_{si}^{1.6}{\Omega }_{si}^{0.8}}{L_b+0.07{\Omega }_{si}{L}_{win}{P}_i^{1.6}}}\right)⌉$$

(7)

where L_si, L_b, and L_win are the luminance of the glare source, the background, and the window in “cd/m²”, respectively. In addition, Ω_si is the solid angle subtended by the glare source from the occupants’ point of view, modified by Guth’s position index, i.e., P_i. n is the number of glare sources [42]. For residential and office buildings, the maximum allowable threshold for DGI is 22 [43].

4. Case Study

4.1. Characteristics of the Case Study Building

The optimization of objective functions is being implemented for a typical residential unit located on the last floor of a five-story reinforced concrete apartment building constructed in 2017 in Sari, Iran. This region has a humid subtropical climate characterized by mild conditions without a dry season and hot summers. The results of this study may serve as a useful guideline for buildings with similar functionalities, climatic conditions, construction technologies, and HVAC systems. This building is adjacent to neighboring units to the east and south while being exposed to external environmental factors to the north and west. Heat transfer occurs between this building and the adjacent units during the spring and summer months, with recorded indoor temperatures of 26 °C and 24 °C, respectively. In contrast, during the fall and winter seasons, temperatures of 26 °C and 28 °C indicate ongoing thermal interaction with the adjacent units. To understand the building’s features, occupancy types, and energy usage trends, data were gathered through surveys, building inspections, and energy bills. The occupancy pattern for this residential unit, which accommodates three individuals, is as follows: From Saturday to Wednesday, the unit is fully occupied until 6:30 a.m., after which it remains unoccupied until 3:00 p.m. Between 3:00 p.m. and 5:00 p.m., 60% of the residents are present, and full occupancy resumes until midnight. On Thursday, there is full occupancy until 6:30 a.m., followed by a period of unoccupancy until 2:00 p.m., after which full occupancy again occurs until midnight. Finally, all residents are fully present throughout the day on Fridays and holidays. This schedule illustrates a combination of full occupancy and designated periods of absence. The exterior walls consist of clay blocks with a thickness of 35 cm, while the interior walls are 28 cm thick, finished with 1.6 cm of gypsum board. The windows are fitted with double-glazing glass, characterized by a U-value of 2.670 W/m²K, a solar heat gain coefficient (SHGC) of 0.703, and a visible transmittance of 0.781. The main characteristics of the building and its exterior fenestration are summarized in

Table 4. Characterization of exterior fenestration.

SUB-SURFACES	Construction	Glass Area [m2]	Frame Area [m2]	Divider Area [m2]	Glass U-Factor [W/m2-K]	Glass SHGC	Glass Visible Trans.	Frame Conductance [W/m2-K]	Divider Conductance [W/m2-K]
SUB SURFACE 4-Hall	DBLCLR6MM/13MMAIR	0.52	0.32	1.21	2.670	0.703	0.781	4.000	4.000
SUB SURFACE 1-Hall	DBLCLR6MM/13MMAIR	8.90	0.69	1.58	2.670	0.703	0.781	4.000	4.000
SUB SURFACE 2-Kitchen	DBLCLR6MM/13MMAIR	1.21	0.24	0.000	2.670	0.703	0.781	1.800	0.000
SUB SURFACE 10-Bedroom 1	DBLCLR6MM/13MMAIR	3.09	0.38	0.000	2.670	0.703	0.781	1.800	0.000
SUB SURFACE 11-Bedroom 2	DBLCLR6MM/13MMAIR	3.21	0.38	0.000	2.670	0.703	0.781	1.800	0.000

. Additionally, Figure 4 illustrates the monthly temperature range in Sari.

Infiltration refers to the unintended flow of outdoor air into a thermal zone and is primarily caused by the opening and closing of exterior doors, cracks around windows, and minor gaps in building elements. Per the information provided by

Table 5. Characterization of the baseline building.

Location	Coordinates	Floor Area (m2)	U-Value (W/m2K)			Heating and Cooling Systems	Lighting System	Air Infiltration Rate
			External Walls	Roof	Double-Glazing Windows
Sari, Iran	{N 36°3′}, {E 53°2′}	79.35	0.52 (Façade), 0.50, 0.51, 0.53 (Kitchen), 0.49 (Bathroom)	0.44	2.67	Natural gas boiler (η = 0.85) Duct Split (EER = 3.13 and COP = 3.53)	Light Emitting Diode (LEDs) Lamp	1.9 Ac/h

Heating degree days (baseline 18 °C): 1382.00, heating setpoint: 20 °C; cooling degree days (baseline 10 °C): 3017.00, cooling setpoint: 28 °C; setback temperature for heating: 18 °C, mean outdoor annual relative humidity: 79.91%; setback temperature for cooling: 29 °C; maximum dry-bulb temperature: 36.8 °C occurs on Jul 29; minimum dry-bulb temperature: −2.1 °C occurs on 15 February; climate: humid subtropical (mild without dry season, hot summer, lat. 20–35° N); ASHRAE climate zone: 3A, ASHRAE description: warm-humid, elevation (m) above sea level: 69; source: Meteonorm 8.2.0

, in this building, the infiltration rate measures 1.9 air changes per hour (Ach) for the thermal zones, including the hall, kitchen, and bedrooms. Air infiltration significantly contributes to energy loss, making building sealing a critical component of energy efficiency. The effectiveness of sealing measures is influenced by the number and size of air leakage paths, and improving airtightness is considered a decision variable in the optimization process that aims to reduce the infiltration rate through the application of air barrier strips. Ventilation in the building is facilitated through two distinct systems: a mechanical intake ventilation system in the toilet, which operates at a fan pressure rise of 70 Pa with a total efficiency of 0.9, and natural ventilation in the bathroom with a flow rate of 0.00944 m³/s per person. Additionally, there are two operable windows—one in the hall with an opening area of 0.7 m² and another in the kitchen with an opening area of 0.9 m²—alongside two glass doors in the bedrooms, each with an opening area of 1 m². These openings are utilized for natural ventilation and thermal comfort during seasonal transition days when the mechanical ventilation system is offline. Cross-zone mixing occurs between adjacent zones, facilitating equal air exchange among pairs (Hall and Bedroom 1, Hall and Bedroom 2, Hall and Kitchen, Hall and Toilet, Hall and Bathroom). This cross-mixing significantly affects the energy balance within both zones. The heating system consists of hot water radiators supplied by a natural gas boiler that operates with an efficiency of 85%. The building’s cooling system comprises a split air conditioner, which has an Energy Efficiency Ratio (EER) of 3.13 and a Coefficient of Performance (COP) of 3.53. Energy for the building is sourced primarily from natural gas and grid electricity. The cooling system, lighting fixtures, and electrical appliances, including a refrigerator and television, draw electricity, while natural gas is utilized for space heating via hot water radiators, domestic hot water (DHW) supply, and cooking. The calculation of carbon emissions associated with gas heating and electrical cooling is summarized in

Table 6. Emissions intensity from electricity and natural gas generation in Iran.

Existing Fuel Resource Name	Units	Electricity	Natural Gas
Source Energy Factor	J/J	3.15	1.05
CO2 emission factor	g/MJ	156.94	50.3
CH4 emission factor	g/MJ	1.4	4.73
N2O emission factor	g/MJ	0.03	0.094

.

Figure 4. the monthly ranges of dry-bulb and wet-bulb temperatures, along with relative humidity in Sari, Iran, illustrate the city's climatic variations throughout the year. These patterns indicate significant fluctuations in temperature and humidity, which can impact both the local ecosystem and daily life.

Table 7. Baseline performance data.

Net Source Energy (Primary Energy Consumption) [kWh]		CO2 Equivalent [kg]	Zone Air CO2 Concentration [ppm]	PPD [%]	Daylighting Reference Point Glare Index Setpoint Exceeded Time [h]
Electricity (Site) [kWh]	Natural Gas (Site) [kWh]
3814.23	19,711.67
32,712.09		17,771.39	718.10	27.40	215.95

offers a comprehensive overview of the baseline performance data, which is essential for understanding energy consumption patterns and their associated environmental impacts. This table highlights key metrics related to energy usage, specifically focusing on both electricity and natural gas consumption, as well as the resulting carbon dioxide (CO₂) emissions. By analyzing this data, stakeholders can gain valuable insights into the operational efficiency of the system and identify opportunities for enhancing sustainability. Furthermore, the information regarding CO₂ concentrations and the predicted percentage of dissatisfied occupants underscores the importance of assessing indoor air quality and thermal comfort.

The average annual electricity and natural gas consumption from 2021 to 2023 were 3814.23 kWh and 19,711.67 kWh, respectively, as obtained from building utility bills. Figure 5 and Figure 6 illustrate and compare the energy simulation results with actual consumption data. According to ASHRAE Guideline 14-2014 [45], a model is considered calibrated if the normalized mean bias error (NMBE) is less than 5% and the coefficient of variation of the root means square error (CV(RMSE)) is less than 15% when using monthly data. For hourly data, the thresholds are NMBE < 10% and CV(RMSE) < 30%. The CV(RMSE) represents the positive average of the squared errors divided by the actual meaning and can be interpreted as the percentage error between simulated and measured data. NMBE, on the other hand, reflects the signed error relative to the mean, indicating the percentage bias—positive when the simulation underestimates (NMBE > 0) and negative when it overestimates (NMBE < 0) the actual data during the evaluation period. For electricity and natural gas consumption, the values for NMBE and CV(RMSE) are −0.97%, 12.90%, 0.35%, and 14.50%, respectively, indicating a strong correlation between the actual and simulated consumption results.

$$NMBE={\displaystyle \frac{\sum (V_{actual}-V_{Modeled})}{\left(N-1\right)\times Mean\left(V_{actual}\right)}}\times 100\%$$

(8)

$$CV\left(RMSE\right)={\displaystyle \frac{\sqrt[2]{\frac{\sum {\left(V_{actual}-V_{Modeled}\right)}^2}{N-1}}}{Mean\left(V_{actual}\right)}}\times 100\%$$

(9)

where $V_{actual}=$ parameter’s measured or metered value for each time step (e.g., month), $V_{modeled}=$ parameter’s estimated or modeled value for each time step, $N=$ number of time steps being analyzed during period of evaluation.

Table 8. Space usage type.

Thermal Zones	Space Area [m2]	Regularly Occupied Area [m2]	Unconditioned Area [m2]	Typical Hours/Week in Operation [h/Week]
Thermal Zone 1—Toilet	1.82	1.82	0.00	120.04
Thermal Zone 2—Hall	48.04	48.04	0.00	120.04
Thermal Zone 3—Kitchen	6.04	6.04	0.00	120.04
Thermal Zone 4—Bathroom	3.10	3.10	0.00	120.04
Thermal Zone 5—Bedroom 1	11.38	11.38	0.00	120.04
Thermal Zone 6—Bedroom 2	8.98	8.98	0.00	120.04
Totals	79.35	79.35	0.00	-

outlines the details of a residential unit featuring six thermal zones. It specifies key information, including the area, regularly occupied area, unconditioned area, and typical hours of operation per week.

According to the diagram presented in Figure 7, several key factors are contributing significantly to energy loss in the existing building, with specific heat removal values highlighted. The window heat removal accounts for a substantial loss of −2740.8 kWh, indicating a considerable amount of thermal energy being exchanged through glazing systems. Additionally, interzone air transfer contributes to heat loss with a value of −490.79 kWh, reflecting the inefficiencies in air movement between different thermal zones. Infiltration is another critical area, showing a significant heat removal of −17,830.7 kWh, which emphasizes the impact of uncontrolled air leaks that allow outside air to enter the conditioned spaces, thereby disrupting thermal comfort and increasing energy demand. Furthermore, opaque surface conduction and other forms of heat transfer contribute an additional −2841.78 kWh to the overall energy loss. Collectively, these values highlight the areas with the highest energy losses, suggesting that targeted energy retrofit measures in these domains are essential for minimizing energy waste, enhancing the building’s overall energy efficiency, and promoting sustainable building practices. Addressing these factors could lead to significant reductions in energy consumption and operational costs, while simultaneously improving indoor comfort levels.

4.2. Characteristics of the HVAC of the Case Study Building

The PTAC system serves an integral function within various building designs, particularly in environments where localized heating and cooling are crucial. The incorporation of both a DX cooling coil and a diverse range of heating options (electric, gas, or water-based) optimizes the climate control capabilities of the unit, enhancing comfort and energy efficiency in indoor settings. Moreover, the outdoor air mixer implemented in this configuration is pivotal for integrating fresh outdoor air into the conditioned space, thereby improving indoor air quality (IAQ). Such systems are particularly beneficial in densely populated environments where cross-contamination of air can be a concern. The balance achieved between cooling and heating loads, guided by precise controls, facilitates an energy-efficient operation, conforming to modern environmental standards and expectations for reduced carbon footprints, essential in the pursuit of sustainable building practices [46]. The design scheme aligns with the contemporary directives for energy performance in buildings, addressing both the thermal comfort of occupants and the operational costs for facility managers. As emphasized in the guidelines established by the European Union, a focus on energy efficiency, particularly in HVAC systems, is paramount as we move towards achieving energy-neutral buildings in the context of future urban development [47]. Figure 8 schematic: the operational diagram representing a packaged terminal air conditioner with its associated components, depicting the integration of the outdoor air mixer and heating/cooling coils.

4.3. Psychrometric Chart

The weather data for a typical year in Sari have been imported into Climate Consultant 6.0 for analysis. Figure 8 presents a psychrometric chart that offers a detailed overview of the thermal environment throughout the year, facilitating informed decisions regarding building design and HVAC system operation. The chart indicates that the blue box represents the hours during which the climate is comfortable without any interventions, comprising only 5.5% of the year. This highlights the necessity for the building to implement measures to enhance occupant comfort. Given Sari’s classification as a humid subtropical region, it must account for both heating and cooling requirements across different seasons. As illustrated in Figure 9, the demand for active heating constitutes 26.8%, while active cooling accounts for 19.7%. Consequently, to achieve energy efficiency through passive design, maximizing solar energy utilization and improving the properties of the building’s outer envelope are essential strategies. This facilitates analysis of how weather affects indoor conditions and highlights potential strategies for optimizing comfort and energy efficiency. The figure allows stakeholders to visualize the relationship between temperature and humidity, essential for assessing indoor air quality, comfort levels, and system control strategies.

4.4. Parametric Study with EnergyPlus

In this parametric study, the primary objective is to minimize two critical performance functions: indoor air quality (IAQ) and thermal comfort. This optimization will be conducted using EnergyPlus, which incorporates various control mechanisms such as thermostat settings for maintaining thermal comfort, humidistats for humidity regulation, and an energy management system (EMS) for overall operational efficiency. Both IAQ and thermal comfort will be managed to ensure they remain within predefined acceptable ranges essential for occupant health and comfort. Furthermore, optimization techniques will be applied within the jEPlus + EA framework to minimize three other main objective functions, culminating in the goal of simultaneously optimizing all five target functions. To facilitate this comprehensive multi-objective optimization process, systematic modifications will be implemented in the EnergyPlus input file based on the parameters detailed in

Table 9. Parametric study with EnergyPlus before optimization by jEPlus + EA.

Design Variables for Parametric Study	Initial Value	Parametric Value
Window shading Control strategies	On if schedule allows	On night if low outdoor temperature and on day if cooling (dominant effect on cooling and heating load)
Air velocity	0.05 (fall and winter), 0.2 (spring and summer)	0.1 (fall and winter), 0.5 (spring and summer)
Coil cooling DX	Single speed	Variable speed
Split air conditioner system	Existing split air conditioner system (COP = 3.53, EER = 3.13)	Split air conditioner system + inverter (COP = 4.3, EER = 3.81)
Headered pumps	Single speed	Variable speed
Boiler type and efficiency	Hot water boiler (η = 0.85)	Condensing hot water boiler (η = 0.95)
Boiler flow mode	Constant flow	Leaving setpoint modulated
Zone control: thermostat: thermal comfort	−1.2 < PMV range < +1.2 20% < PPD < 30%	−0.5 < PMV range < +0.5 PPD < 10%
Indoor air quality	PPM < 800	PPM < 700 (The standard level of indoor air quality; ASHRAE recommends CO2 levels not exceeding 700 ppm above outdoor levels (ASHRAE-62.1 [48]) by energy management system (EMS) and zone cross-mixing EnergyPlus
	- (Zone mean annual air relative humidity = 52.49%) -	Zone control: humidistat (Zone mean annual air relative humidity = 47.04%) Zone predicted moisture load to humidifying setpoint moisture transfer rate = −0.00331, [kgWater/s] Zone predicted moisture load to dehumidifying setpoint moisture transfer rate = 0.001928 [kgWater/s]
Solar panel	-	Polycrystalline Area = 10, 15, 20, 25 m2

, thereby adequately preparing the simulation model for the application of a genetic metaheuristic algorithm. This algorithm, known for its adaptability in solving complex multi-dimensional optimization problems, will explore potential solutions and their performance relative to the defined objectives, ultimately identifying configurations that balance energy efficiency with enhanced indoor environmental quality.

According to

Table 10. Indoor air quality with energy management system (EMS) EnergyPlus.

Ventilation Profile
Indoor air quality with energy management system (EMS) EnergyPlus
IF Occupancy Sensor > 0 && ((CO2 Sensor_Hall >= 600ppm && CO2 Sensor_Hall < 800ppm) || (CO2 Sensor_Kitchen >= 600 ppm && CO2 Sensor_Kitchen < 800ppm) || (CO2 Sensor_Bedroom1 >= 600 ppm && CO2 Sensor_Bedroom1 < 800ppm) || (CO2 Sensor_Bedroom2 >= 600 ppm && CO2 Sensor_Bedroom2 < 800ppm)), SET A1 = 0.25 opening area,
ELSEIF Occupancy Sensor > 0 && ((CO2 Sensor_Hall >= 800ppm && CO2 Sensor_Hall < 1000ppm) || (CO2 Sensor_Kitchen >= 800ppm && CO2 Sensor_Kitchen < 1000 ppm) || (CO2 Sensor_Bedroom1 >= 800 ppm && CO2 Sensor_Bedroom1 < 1000 ppm) || (CO2 Sensor_Bedroom2 >= 800 ppm && CO2 Sensor_Bedroom2 < 1000 ppm)), SET A1 = 0.5 opening area,
ELSEIF Occupancy Sensor > 0 && ((CO2 Sensor_Hall >= 1000 ppm) || (CO2 Sensor_Kitchen >= 1000 ppm) || (CO2 Sensor_Bedroom1 >= 1000 ppm) || (CO2 Sensor_Bedroom2 >= 1000 ppm)), SET A1 = 1 opening area, ELSE, SET A1 = 0 opening area, ENDIF;

, the operational procedure of the energy management system (EMS) in this case is designed to regulate indoor air quality by managing carbon dioxide (CO₂) concentrations in thermal zones. The protocol is initiated when occupants are present in these zones, ensuring that the system is responsive to actual occupancy and corresponding air quality needs. The procedure outlines a tiered response based on CO₂ concentration levels. When the CO₂ concentration exceeds 600 parts per million (ppm) but remains below 800 ppm, the EMS will open 25% of the total area of windows and glass door openings. This partial opening allows for the introduction of fresh air into space, effectively diluting the accumulated CO₂ and thereby improving indoor air quality without excessive energy consumption. If the concentration rises further, exceeding 800 ppm but remaining under 1000 ppm, the EMS response escalates; it will open 50% of the area of the openings. This increased ventilation rate addresses the heightened levels of CO₂ more aggressively, ensuring that air quality remains within healthier thresholds for occupants. In scenarios where CO₂ concentrations surpass 1000 ppm, a critical threshold indicating significant air quality degradation, the EMS will fully open all window and glass door openings. This full ventilation strategy is implemented to quickly reduce CO₂ levels and improve air quality, safeguarding occupant health and comfort. The overarching goal of this operational procedure is to maintain optimal indoor air quality through proactive management of CO₂ levels, thereby improving occupant well-being and productivity. By utilizing the EnergyPlus energy management system, this approach not only enhances the comfort levels within the building but also demonstrates energy efficiency by modulating ventilation based on real-time conditions, optimizing both energy consumption and air quality simultaneously. This responsive system underscores the importance of integrating sustainable practices in building management, promoting healthier indoor environments while adhering to energy conservation principles.

Table 11. Results of parametric study with EnergyPlus before optimization by jEPlus + EA.

Net Source Energy (Primary Energy Consumption) [kWh]		CO2 Equivalent [kg]	Zone Air CO2 Concentration [ppm]	PPD [%]	Daylighting Reference Point Glare Index Setpoint Exceeded Time [h]
Electricity (Site) [kWh]	Natural Gas (Site) [kWh]
3996.14	27,815.39
24,974.56		23,147.30	650.14	9.77	347.19

presents the results of the parametric study conducted using EnergyPlus prior to optimization with jEPplus + EA. As noted, the objective of maintaining PPD and indoor air quality has been successfully achieved at this stage, with both metrics remaining at acceptable levels. Specifically, the percentage of occupants dissatisfied with thermal comfort is kept below 10%, indicating a high level of comfort within space. Furthermore, the CO₂ concentration is maintained below 700 ppm, which is essential for ensuring a healthy indoor environment. These results demonstrate effective design strategies and system performance in promoting occupant well-being and comfort.

4.5. Sensitivity Analysis

Sensitivity analysis is a critical technique used to evaluate how variations in independent variables influence a specified dependent variable, while adhering to a defined set of assumptions. Among the various methods available, the Morris method is particularly esteemed for its efficacy in conducting sensitivity analyses on complex models with multiple input factors. This approach generates two key metrics: Mean (μ), which represents the average of the absolute values of the elementary effects, thereby indicating the overall significance of each input factor; and Standard Deviation (σ), which measures the variability of the elementary effects and offers insights into the non-linear behaviors and interactions among variables. By utilizing these metrics, the Morris method facilitates a more comprehensive understanding of the sensitivity of model outputs to changes in input parameters [49,50]. In accordance with Figure 10, the sensitivity analysis revealed that ten design variables—HVAC systems, heating and cooling setpoint, airflow rate, air infiltration rate, air velocity, roof and floor insulation, and window construction for both north/west and south orientations—most significantly impact the objective functions.

4.6. Characteristics of Design Variables

Table 12. Details of PV panels investigated.

Area (m2)	Tilt Angle (°C)	Efficiency (%)	Type	Location
10, 15, 20, 25	36.64	13.2	Polycrystalline	Roof, south oriented

and

Table 13. Electric Loads Satisfied.

Solar Panel Properties	Electricity [kWh]	Percent Electricity [%]
Photovoltaic Power	5448.481	143.05
Power Conversion	−108.97	−2.9
Total On-Site Electric Sources	5339.511	140.19
Electricity Coming from Utility	2711.379	71.19
Surplus Electricity Going to Utility	4242.166	111.38
Net Electricity from Utility	−1530.79	−40.2
Total On-Site and Utility Electric Sources	3808.725	100.00
Total Electricity End Uses	3808.725	100.00

summarize the properties of solar panels and their electricity production. Table 12 provides an overview of the details of the photovoltaic (PV) panels investigated. The panels, with areas of 10, 15, 20, and 25 m², are input into the software for parametric analysis. A tilt angle of 36.64° is used to optimize sunlight exposure, and the panels exhibit an efficiency of 13.2%, indicating the proportion of solar energy converted into usable electricity. These polycrystalline panels are installed on a roof, specifically south oriented, to maximize solar gain. The study aims to select the configuration that yields the highest electricity production, ensuring optimal energy output for the given parameters. According to Table 13 the photovoltaic system generates a total of 5448.481 kWh of electricity, representing 143.05% of the expected output. After accounting for power conversion losses, the total on-site electricity produced is 5339.511 kWh, which meets 140.19% of the energy demand.

The study identifies and evaluates a total of 25 design variables, as presented in

Table 14. Characteristics of design variables for investigated energy retrofit measures (ERMs).

Design Variables	Units	Ranges	Initial Values
Heating Setpoint	°C	Double [18, 28]	20
Cooling Setpoint	°C	Double [28, 32]	28
Air Infiltration Rate	1/h	Double [0.1, 1.9]	1.9
Clothing Level for Spring	Clo	Double {0.61, 0.67, 0.72, 0.74}	0.72
Clothing Level for Summer	Clo	Double {0.36, 0.54, 0.57}	0.54
Clothing Level for Fall	Clo	Double {0.89, 0.96, 1.01}	0.89
Clothing Level for Winter	Clo	Double {1.10, 1.14, 1.30, 1.37}	1.10
Air Velocity	m/s	Double [0.1, 0.5]	0.05 (Fall, winter), 0.2 (spring, summer)
Air Flow Rate	m3/s	Double {0.00944, 0.01888, 0.02832, 0.03776, 0.0472, 0.05664, 0.06608, 0.07552, 0.08496, 0.0944}	0.00944
Roof Insulation	-	Discrete {ExistingRoof, RoofwithVegetation, RoofwithInsulationCellularGlass25mm, RoofwithInsulationCellularGlass50mm, RoofwithInsulationCellularGlass75mm, RoofwithInsulationGlassfiber25mm, RoofwithInsulationGlassfiber50mm, RoofwithInsulationGlassfiber75mm, RoofwithInsulationExpandedrubber(rigid)25mm, RoofwithInsulationXPS25mm, RoofwithInsulationXPS50mm, RoofwithInsulationXPS75mm, RoofwithInsulationBatt89mm, RoofwithInsulationBatt154mm, RoofwithInsulationBatt244mm, RoofwithInsulationBoard25mm, RoofwithInsulationBoard50mm, RoofwithInsulationBoard75mm, RoofwithInsulationExpandedPerlite25mm}	ExistingRoof
Floor Insulation	-	Discrete {ExistingFloor, FloorwithInsulationCellularglass25mm, FloorwithInsulationCellularglass50mm, FloorwithInsulationCellularglass75mm, FloorwithInsulationGlassfiber25mm, FloorwithInsulationGlassfiber50mm, FloorwithInsulationGlassfiber75mm, FloorwithInsulationExpandedrubber(rigid)25mm, FloorwithInsulationXPS25mm, FloorwithInsulationXPS50mm, FloorwithInsulationXPS75mm, FloorwithInsulationBatt89mm, FloorwithInsulationBatt154mm, FloorwithInsulationBatt244mm, FloorwithInsulationBoard25mm, FloorwithInsulationBoard50mm, FloorwithInsulationBoard75mm, FloorwithInsulationExpandedPerlite25mm}	ExistingFloor
Window Construction (North, West)	-	Discrete {DblClr6mm/13mmAir, DblBronze6mm/13mmAir, DblGreen6mm/13mmAir, DblGrey6mm/13mmAir, DblBlue6mm/13mmAir, DblRef-A-LTint6mm/13mmAir, DblRef-A-MTint6mm/13mmAir, DblRef-A-HTint6mm/13mmAir, DblLoE(e2 = 0.1)Tint6mm/13mmAir, DblLoESpecSelTint6mm/13mmAir, DblElecAbsBleached6mm/13mmAir, DblElecAbsColored6mm/13mmAir, DblElecRefBleached6mm/13mmAir, DblElecRefColored6mm/13mmAir, DblLoEElecAbsBleached6mm/13mmAir, DblLoEElecAbsColored6mm/13mmAir, DblLoEElecRefBleached6mm/13mmAir, DblLoEElecRefColored6mm/13mmAir, TrpLoEFilm(33)Bronze6mm/13mmAir, TrpLoEFilm(44)Bronze6mm/13mmAir, TrpLoEFilm(55)Bronze6mm/13mmAir, TrpLoEFilm(66)Bronze6mm/13mmAir, QuadrupleLoEFilms(88)3mm/8mmKrypton}	DblClr6mm/13mmAir
Window Construction (South)	-	Discrete {DblClr6mm/13mmAir, DblClrLowIron5mm/13mmAir, DblLoE(e2 = 0.2)Clr6mm/13mmAir, DblLoE(e2 = 0.1)Clr6mm/13mmAir, DblLoESpecSelClr6mm/13mmAir, DblElecAbsBleached6mm/13mmAir, TrpLoEFilm(55)Clr6mm/13mAir, TrpLoEFilm(66)Clr6mm/13mmAir}	DblClr6mm/13mmAir
Window Material Gas	-	Discrete {AIR13MM, ARGON13MM, KRYPTON13MM, XENON13MM}	AIR13MM
Window Height (O.K.B)	m	Double [0.20, 1.40] and {0.17}	0.17
Overhang Depth	m	Double [0.1, 0.5] and {0.01}	0.01
Tilt Angle Overhang	°C	Integer {30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100, 105, 110, 115, 120}	90
Fin Depth 1 (Left Depth)	m	Double [0.1, 0.5] and {0.01}	0.01
Fin Depth 2 (Right Depth)	m	Double [0.1, 0.5] and {0.01}	0.01
Tilt Angle Fin	°C	Integer [30, 120]	90
Window Material Shade	-	Discrete {HIGHREFLECT-LOWTRANSSHADE, MEDIUMREFLECT-MEDIUMTRANSSHADE, MEDIUMREFLECT-LOWTRANSSHADE, LOWREFLECT-HIGHTRANSSHADE, LOWREFLECT-MEDIUMTRANSSHADE, LOWREFLECT-LOWTRANSSHADE}	MEDIUMREFLECT-MEDIUMTRANSSHADE
Lighting Control Type	-	Discrete {Continuous, Stepped, ContinuousOff}	Stepped
Solar Absorptance	-	Double {0.9, 0.91,0.94, 0.89, 0.86, 0.85, 0.84, 0.82, 0.8, 0.79, 0.77, 0.71, 0.69, 0.59, 0.57, 0.54, 0.53, 0.51, 0.45, 0.4, 0.3}	0.7
Exterior Door	-	Discrete {G0413mmwood, G0525mmwood, G0650mmwood, G07100mmwood}	G0650mmwood
HVAC Systems	-	Discrete {ConstantVolumeChillerBoiler, DualDuct, FanCoil, FanCoil-DOAS, PackagedVAV, PTAC, PTAC-DOAS, PTHP, PTHP-DOAS, UnitaryHeatPump:AirToAir, VAVWater-CooledChillers, VAVAir-CooledChillers, VRF, VRF-DOAS, WaterToAirHeatPump}	PTAC

. Among these variables, several key factors play a critical role in influencing the energy performance and overall efficiency of the building. These factors include the heating and cooling setpoints, which determine the indoor temperature and directly affect energy consumption; the air infiltration rate, which measures the uncontrolled exchange of outside air with indoor air and impacts heating and cooling loads; and the types and thicknesses of thermal insulation used in both the floor and roof, which are crucial for minimizing heat loss or gain. Additionally, the construction of windows is significant, as it influences thermal performance and light infiltration. Other important variables include the air flow rate, essential for maintaining indoor air quality and comfort, and the type of HVAC (heating, ventilation, and air conditioning) system employed, which affects energy use, operational efficiency, and occupant satisfaction. By systematically analyzing these design variables, the study aims to identify optimal strategies for improving building energy efficiency.

Figure 11, Figure 12, Figure 13, Figure 14, Figure 15 and Figure 16 in the case study illustrate various aspects of the baseline building model, providing a comprehensive visualization of its design and performance. Figure 11 and Figure 12 shows the overall case study building, highlighting its architectural features and layout. Figure 13 presents a detailed rendering by thermal zones, which is critical for understanding temperature distribution and comfort levels within space. Figure 14 focuses on construction types, illustrating different materials used in the building, which impact energy efficiency and sustainability. Figure 15 depicts the boundary conditions, essential for assessing the thermal interactions between the building and its environment. Finally, Figure 16 illustrates the integration of solar panels, emphasizing the building’s commitment to renewable energy and enhancing its overall energy performance. Together, these figures provide valuable insights that support the analysis and findings presented in this research.

4.7. Simulation of Energy-Related Measures Using NSGA-II

The total number of available combinations for the introduced Energy-Related Measures (ERMs) is approximately 1.41 × 10²⁴, illustrating an extensive search space. Given that each EnergyPlus simulation requires approximately three minutes to complete, simulating all possible combinations of ERMs would be impractical due to the immense computational time involved. However, the application of the Non-dominated Sorting Genetic Algorithm II (NSGA-II) significantly reduces the required computation time. Drawing on findings from previous studies [51,52], the parameters for the NSGA-II were optimized, with a crossover rate of 100%, a mutation rate of 20%, and a tournament size of 5, which collectively contribute to balanced computational efficiency and reliable results. The maximum number of generations was set to 100, serving as the stopping criterion for the NSGA-II process. Utilizing parallel computing on an Intel Xeon^® processor (2.80 GHz, 12 MB cache, and 12 cores), the total elapsed computational time was approximately 7 h and 20 min. This setup demonstrates the model’s capacity to intelligently explore the solution space while efficiently managing computational resources.

5. Results and Discussion

The optimization process yields a diverse set of results, with each point on the Pareto front representing an optimal solution. To identify the most suitable solution from these Pareto front points, this study employs the weighted sum method as a decision-making approach, as expressed in Equation (8).

$$f_{ws}\left(x\right)={\sum }_{i=1}^5{a}_i{\displaystyle \frac{f_i\left(x\right)-f_i{\left(x\right)}^{min}}{{f_i\left(x\right)}^{max}-{f_i\left(x\right)}^{min}}}$$

(10)

Two scenarios are examined concerning future climate change. The first scenario assesses the case study under current climate conditions, while the second scenario considers the potential impacts of climate change.

5.1. Scenario 1. Present Climate Condition

In this scenario, the building performance is assessed under current climate conditions using data from the Meteonorm 8.2.0.epw file for Sari City. The analysis examines different configurations of building materials, insulation types, and mechanical systems to evaluate their effects on energy efficiency, CO₂ equivalent emissions, indoor environmental quality, and thermal comfort.

Figure 17 illustrates the Pareto front for Scenario 1 under present climate conditions, showcasing the relationship between indoor air quality (IAQ), primary energy consumption (EC), and CO₂ emissions. The red and blue points on the 3D plot represent different solutions, with the red points indicating more optimal combinations. As seen in the graph, there is a noticeable trade-off where increased IAQ levels correlate with higher energy consumption and CO₂ emissions. However, the curve’s flattening suggests that beyond a certain point, improvements in IAQ come with diminishing returns regarding energy efficiency. This analysis emphasizes the need to balance air quality with energy performance and emissions, guiding stakeholders toward sustainable retrofit strategies.

According to Figure 18, X-axis (primary EC): the horizontal axis represents primary energy consumption, ranging from approximately 20,000 kWh to 200,000 kWh. Y-axis (IAQ in ppm): the vertical axis measures indoor air quality, expected to vary from below 500 ppm up to around 1000 ppm or higher. The optimal IAQ range is indicated as being between 500 and 700 ppm. The analysis emphasizes the importance of selecting or designing HVAC systems that effectively balance energy consumption with optimal indoor air quality (IAQ) performance. Particularly, systems that maintain IAQ within the ideal range can enhance user satisfaction. Conversely, systems with high energy consumption do not achieve good IAQ may require redesign or optimization to improve performance and address user dissatisfaction. Understanding the relationship between energy consumption and IAQ is crucial for informing market trends, guiding manufacturers in developing technologies that prioritize both efficiency and air quality. The scatter plot demonstrates a clear correlation between primary energy consumption and IAQ, highlighting that maintaining IAQ levels between 500 and 700 ppm is critical for optimal performance. Systems such as “FanCoil”, “PTHP”, “PTHP-DOAS”, “VRF”, and “VRF-DOAS” effectively manage to achieve this balance while consuming less energy. Stakeholders should concentrate on technologies that reduce energy costs while ensuring high indoor air quality, which will lead to better user satisfaction and compliance with health standards.

Based on Figure 19, X-axis (primary EC): this axis represents the primary energy consumption of HVAC systems, ranging roughly from 20,000 to 200,000 kWh. Y-axis (PPD %): the Y-axis indicates the predicted percentage of dissatisfaction among users due to system performance, with higher values representing greater predicted dissatisfaction. The scatter plot illustrates the relationship between primary energy consumption (EC) in kilowatt-hours (kWh) and the predicted percentage of dissatisfaction (PPD) across various HVAC systems, with colors distinguishing different system types and highlighting performance trends. Notably, systems such as “PTAC”, “Unitary Heat Pump”, and “PTHP-DOAS” fall within the lower energy consumption range and correspond to low dissatisfaction rates, indicating satisfactory user performance. The plot reveals that increased energy consumption does not inherently guarantee greater user satisfaction. Thus, it is crucial to prioritize systems that efficiently utilize energy while maintaining low levels of predicted dissatisfaction. By understanding these dissatisfaction levels, stakeholders can make informed selections of HVAC systems that better align with user needs. Additionally, focusing on high-energy consumption systems that also exhibit high predicted dissatisfaction may uncover opportunities for redesign or technological advancements. This analysis underscores that higher energy consumption is not synonymous with lower dissatisfaction, emphasizing the importance of achieving a balance between performance and energy efficiency. Future research should identify specific characteristics of systems that effectively reduce predicted dissatisfaction to enhance design and operational strategies in HVAC solutions.

Referring to Figure 20, the boxplot effectively illustrates the primary energy consumption associated with different HVAC systems. The variations in energy usage highlight the significance of selecting systems that meet energy consumption objectives, efficiency benchmarks, and overall operational costs. Systems such as “VRF”, “FanCoil”, “PTHP”, and “PTAC” demonstrate comparatively lower energy consumption, ranging from approximately 15,000 to 25,000 kWh. These options may be regarded as more efficient choices in terms of energy use and emissions, making them potentially more suitable for environmentally conscious projects.

As shown in Figure 21, this bubble chart provides a comprehensive view of how various HVAC systems perform regarding their energy consumption and associated with CO₂ emissions. Identifying systems that offer the best balance between these two factors can guide decisions towards more efficient and environmentally friendly HVAC solutions. Further analysis could explore the economic aspects and operational efficiencies associated with these systems. Systems like “VRF”, “FanCoil”, and “PTAC” exhibit lower energy consumption (around 15,000–25,000 kWh) and lower CO₂ emissions. These systems could be viewed as more efficient choices for energy use and emissions, making them potentially more favorable for implementation in eco-conscious projects.

Figure 22 presents a comprehensive visualization through parallel coordinates, illustrating the intricate relationships among various parameters (P1 to P25) and performance metrics.

Figure 23 showcases the selected energy retrofit measures (ERMs) for optimal solutions under Scenario 1, reflecting present climate conditions. The parallel coordinates visualization highlights how each ERM interacts with key performance indicators, represented on the axes from P1 to P25. The purple lines indicate the best solutions among the available options, revealing significant fluctuations and demonstrating the varying effectiveness of each ERM.

Figure 24 depicts the distribution of HVAC systems in the Pareto front for the first scenario. The pie chart illustrates that the VRF-DOAS system dominates with 24%, followed closely by PTAC at 19% and PTAC-DOAS also at 19%. Other notable systems include PTHP-DOAS at 12% and VRF at 14%. In contrast, systems like Constant Volume Chiller Boiler, Unitary Heat Pump, and Packaged VAV exhibit minimal representation, with 0%, 3%, and 1%, respectively. This distribution highlights a clear preference for versatile and efficient systems, emphasizing their critical role in achieving energy efficiency and sustainability objectives in current climate conditions. The dominance of VRF-DOAS reflects its effectiveness in optimizing indoor climate control while minimizing energy consumption.

5.2. Scenario 2. Considering Climate Change for the Mid and Late Century (2050 and 2080)

5.2.1. Scenario 2. Considering Climate Change (2050)

In this scenario, the building performance is evaluated under projected climate conditions for the year 2050, considering anticipated changes in temperature, humidity, and other environmental factors due to climate change. The analysis aims to understand how these changes will affect energy consumption, indoor air quality, thermal comfort, and overall building performance.

• Utilizing updated meteorological data, such as those from the Meteonorm 8.2.0.epw file, the analysis incorporates expected increases in average temperatures and variations in precipitation patterns. These data are critical for simulating future energy demands and thermal comfort levels.
• Increased cooling loads: As temperatures rise, the demand for cooling is expected to increase significantly. Simulations indicate that buildings may require up to 20–30% more energy for cooling compared to present conditions. This necessitates a reevaluation of HVAC systems to ensure they can handle the increased loads efficiently.

The second scenario highlights the significant challenges posed by climate change on building performance by 2050. It emphasizes the need for proactive measures in design and retrofitting to ensure energy efficiency and occupant comfort in a changing climate. By focusing on advanced materials, innovative HVAC solutions, and effective shading strategies, buildings can be better equipped to handle the impacts of climate change while promoting sustainability and resilience. Further research and simulation will be essential to refine these strategies and adapt to future conditions effectively.

Figure 25 presents the Pareto front for Scenario 2, reflecting the impact of climate change in 2050. The 3D plot illustrates the balance between indoor air quality (IAQ), primary energy consumption (EC), and CO₂ emissions. The red points indicate optimal solutions, demonstrating a clear trade-off where improvements in IAQ are associated with increases in both energy consumption and CO₂ emissions. As the number of selected solutions increases towards the upper left area of the graph, it becomes apparent that achieving high indoor air quality may necessitate higher energy use, complicating sustainability efforts. The distribution of the points indicates a shift in performance metrics due to climate change, emphasizing the need for innovative strategies to enhance energy efficiency while maintaining satisfactory indoor conditions. This analysis underlines the challenges of navigating environmental impacts while optimizing HVAC systems in future scenarios.

Figure 26 illustrates the selected energy retrofit measures (ERMs) for optimal solutions in Scenario 2, focusing on the effects of climate change in 2050. The parallel coordinates plot highlights how each ERM correlates with key performance indicators, represented from P1 to P25. The magenta lines signify the best performing solutions, revealing significant variations across different measures. Specific ERMs, such as PTAC and VRF-DOAS, show marked effectiveness in achieving desired outcomes, while others contribute less optimally. The clustering of lines indicates that certain combinations of ERMs yield better performance in terms of energy consumption and indoor air quality, underscoring the complexities involved in retrofitting for climate resilience. This analysis emphasizes the importance of selecting optimal ERMs to address the challenges posed by climate change while enhancing energy efficiency and environmental sustainability.

As indicated by Figure 27, X-axis (primary EC): the horizontal axis represents primary energy consumption, ranging from approximately 20,000 kWh to 200,000 kWh. Y-axis (IAQ in ppm): the vertical axis measures indoor air quality (IAQ), expected to range from below 500 ppm to around 1000 ppm or higher, with an optimal IAQ range of 500 to 700 ppm. The analysis highlights the critical need for HVAC systems that effectively balance energy consumption with optimal IAQ performance. Systems that maintain IAQ within the ideal range enhance user satisfaction, while those with high energy consumption and poor IAQ may require redesign or optimization. Understanding the relationship between energy consumption and IAQ is vital for guiding market trends and helping manufacturers prioritize both efficiency and air quality. The scatter plot shows a clear correlation between primary energy consumption and IAQ, emphasizing that maintaining IAQ levels between 500 and 700 ppm is essential for optimal performance. Systems such as “FanCoil”, “FanCoil-DOAS”, “PTAC”, “PTAC-DOAS”, “VRF”, and “VRF-DOAS” effectively achieve this balance with lower energy consumption. Stakeholders should focus on technologies that reduce energy costs while ensuring high indoor air quality, leading to greater user satisfaction and adherence to health standards.

As illustrated in Figure 28, X-axis (primary EC): this axis represents the primary energy consumption of HVAC systems, ranging from approximately 20,000 to 200,000 kWh. Y-axis (PPD %): the Y-axis indicates the predicted percentage of user dissatisfaction due to system performance, with higher values reflecting greater dissatisfaction. The scatter plot illustrates the relationship between primary energy consumption (EC) in kilowatt-hours and the predicted percentage of dissatisfaction (PPD) across various HVAC systems, using color coding to distinguish different system types and highlight performance trends. Notably, systems such as “PTAC”, “PTAC-DOAS”, “PTHP”, “PTHP-DOAS”, “Unitary Heat Pump”, and “VRF” fall within the lower energy consumption range while corresponding to low dissatisfaction rates, indicating satisfactory user performance. The plot demonstrates that increased energy consumption does not guarantee greater user satisfaction, underscoring the need to prioritize systems that efficiently utilize energy while maintaining low predicted dissatisfaction. By understanding these dissatisfaction levels, stakeholders can make informed choices about HVAC systems that align better with user needs. Additionally, examining high-energy consumption systems that also show high predicted dissatisfaction may reveal opportunities for redesign or technological advancements. This analysis emphasizes that higher energy consumption does not equate to lower dissatisfaction, highlighting the importance of balancing performance and energy efficiency. Future research should focus on identifying specific characteristics of systems that effectively minimize predicted dissatisfaction to enhance design and operational strategies in HVAC solutions.

In accordance with Figure 29, the boxplot effectively illustrates the CO₂-equivalent emissions associated with different HVAC systems. Systems such as “PTHP”, “Unitary Heat Pump”, “VRF”, and “PTAC” show relatively lower CO₂-eq emissions, ranging from approximately 5000 to 15,000 kg. These options can be regarded as more efficient choices in terms of energy use and emissions, making them particularly suitable for environmentally conscious projects.

Per Figure 30, bubble chart provides a comprehensive view of how various HVAC systems perform regarding their energy consumption and associated with CO₂ emissions. Identifying systems that offer the best balance between these two factors can guide decisions towards more efficient and environmentally friendly HVAC solutions. Further analysis could explore the economic aspects and operational efficiencies associated with these systems. Systems like “VRF”, “FanCoil”, “PTHP”, and “PTAC” exhibit lower energy consumption (around 10,000–30,000 kWh) and lower CO₂ emissions. These systems could be viewed as more efficient choices for energy use and emissions, making them potentially more favorable for implementation in eco-conscious projects.

Figure 31 illustrates the distribution of HVAC systems present in the Pareto front for the second scenario, targeting the year 2050. The pie chart clearly shows the relative contributions of various HVAC technologies, with PTAC-DOAS representing the largest segment at 28%. Other notable systems include VRF-DOAS at 19% and PTHP-DOAS at 10%. Meanwhile, systems like FanCoil, Unitary Heat Pump, and Packaged VAV contribute minimally, with percentages below 10%. This distribution highlights a preference for systems that effectively balance performance and energy efficiency, indicating a strategic direction towards optimized HVAC solutions in future climate conditions. The dominance of PTAC-DOAS underlines its importance in achieving sustainability goals while addressing indoor climate needs.

5.2.2. Scenario 2. Considering Climate Change (2080)

In this section, the building performance under projected climate conditions for the year 2080 is evaluated. This scenario builds upon the findings from the 2050 analysis, incorporating further anticipated changes in climate variables such as temperature, humidity, and extreme weather events.

As outlined in Figure 32, the Pareto front for Scenario 2, focused on climate change in 2080, is illustrated in the provided 3D plot, which the trade-offs between indoor air quality (IAQ), primary energy consumption (EC), and CO₂ emissions. The red points on the graph indicate solutions that optimize these parameters, showing an inverse relationship between energy consumption and CO₂ emissions, while maintaining acceptable IAQ levels. As the graph progresses, it becomes evident that improved energy efficiency can lead to lower emissions without significantly compromising air quality. This analysis highlights the importance of identifying optimal energy retrofit measures that balance these critical factors, ensuring both environmental sustainability and health standards.

The chosen energy retrofit measures (ERMs) for the optimal solutions in Scenario 2, concerning climate change in 2080, are highlighted in Figure 33, which illustrates various energy performance metrics. Notably, the graph reveals fluctuations in efficiency and effectiveness values, emphasizing which ERMs yield the highest benefits in terms of energy savings and emissions reduction. The selected measures not only address immediate energy demands but also align with future sustainability goals, making them crucial for adapting to the anticipated climatic shifts. Overall, the analysis suggests a strategic integration of ERMs that collectively enhance resilience against climate-related impacts.

As presented in Figure 34, X-axis (primary EC): the horizontal axis represents primary energy consumption, ranging from approximately 20,000 kWh to 200,000 kWh. Y-axis (IAQ in ppm): the vertical axis measures indoor air quality (IAQ), expected to vary from below 500 ppm to around 1000 ppm or higher, with an optimal range of 500 to 700 ppm. This analysis underscores the importance of selecting or designing HVAC systems that balance energy consumption with optimal IAQ performance. Systems that maintain IAQ within the ideal range enhance user satisfaction, while those with high energy consumption that fail to achieve good IAQ may require redesign or optimization to address performance issues and user dissatisfaction. Understanding the relationship between energy consumption and IAQ is crucial for informing market trends and guiding manufacturers in developing technologies that prioritize both efficiency and air quality. The scatter plot reveals a clear correlation between primary energy consumption and IAQ, emphasizing that maintaining IAQ levels between 500 and 700 ppm is vital for optimal performance. Systems such as “FanCoil”, “FanCoil-DOAS”, “VRF”, and “VRF-DOAS” effectively achieve this balance while consuming less energy. Stakeholders should focus on technologies that reduce energy costs while ensuring high indoor air quality, leading to increased user satisfaction and compliance with health standards.

In reference to Figure 35, X-axis (primary EC): this axis represents the primary energy consumption of HVAC systems, ranging from approximately 20,000 to 200,000 kWh. Y-axis (PPD %): the Y-axis indicates the predicted percentage of user dissatisfaction due to system performance, with higher values representing greater dissatisfaction. The scatter plot illustrates the relationship between primary energy consumption (EC) in kilowatt-hours and the predicted percentage of dissatisfaction (PPD) across various HVAC systems, with color coding highlighting different system types and performance trends. Notable systems such as “PTAC”, “PTHP”, “VRF”, and “VRF-DOAS” fall within the lower energy consumption range and correspond to low dissatisfaction rates, indicating satisfactory user performance. The plot reveals that increased energy consumption does not guarantee greater user satisfaction, emphasizing the need to prioritize systems that efficiently utilize energy while maintaining low dissatisfaction levels. Understanding these dissatisfaction levels enables stakeholders to make informed selections of HVAC systems that better align with user needs. Additionally, examining high-energy consumption systems with high predicted dissatisfaction may reveal opportunities for redesigning or technological advancements. This analysis reinforces that higher energy consumption does not equate to lower dissatisfaction, highlighting the importance of balancing performance and energy efficiency. Future research should focus on identifying specific characteristics of systems that effectively reduce predicted dissatisfaction to improve design and operational strategies in HVAC solutions.

As depicted in Figure 36, boxplot illustrates the primary energy consumption associated with various HVAC systems. The variation in energy usage underscores the importance of selecting systems that align with energy consumption goals, efficiency benchmarks, and overall operational costs. Systems such as “VRF”, “FanCoil”, “PTHP”, and “PTAC” demonstrate comparatively lower energy consumption, ranging from approximately 15,000 to 25,000 kWh. These options are considered more efficient in terms of energy use and emissions, making them suitable choices for environmentally conscious projects.

In alignment with Figure 37. bubble chart offers a visual comparison of selected HVAC systems based on their energy consumption and CO₂ equivalent emissions, thereby elucidating their environmental impact. Specifically, the analysis highlights those five systems—Packaged Terminal Air Conditioners (PTACs), Fan Coil Units (FanCoils), variable refrigerant flow systems (VRFs), Packaged Terminal Heat Pumps (PTHPs), and Unitary Heat Pumps—demonstrate significantly lower energy use and carbon dioxide emissions compared to other systems. This analysis underscores the critical importance of selecting energy-efficient HVAC systems as a strategy for reducing carbon footprints and promoting sustainability. By prioritizing the deployment of PTACs, FanCoils, VRFs, PTHPs, and Unitary Heat Pumps, policymakers and industry stakeholders can make informed decisions that contribute to global climate goals.

The data presented in the pie chart in Figure 38 reveal the dominance of various HVAC systems within the dataset. The Fan Coil system, holding the largest market share at 22%, underscores its effectiveness and efficiency in heating and cooling applications, making it a popular choice in the HVAC landscape. Following closely, the PTAC-DOAS (19%) highlights a growing trend towards systems that integrate outdoor air, enhancing indoor air quality. Variable refrigerant flow (VRF) systems account for 14%, reflecting their energy-efficient operation and adaptability across different environments. In the moderate representation category, both PTHP-DOAS (11%) and Fan Coil-DOAS (9%) indicate a tendency to combine technologies for optimal HVAC performance. The Packaged Terminal Heat Pump (PTHP) has a notable share of 7%, showcasing its utility in specific settings. Conversely, systems such as VRF-DAS (9%), Unitary Heat Pump (2%), PTAC (4%), and Packaged VAV (3%) exhibit lower prevalence, suggesting they are less common in certain applications or regions. Notably, the absence of VAV Water-Cooled Chillers (0%) points to either a lack of implementation in the projects studied or a potential shift toward more efficient alternatives.

Analysis of

Table 15. Optimal ERMs for Scenario 1—current climate conditions, Scenario 2—climate change (2050), and Scenario 2—climate change (2080).

Design Variables	Scenario 1—Current Climate	Scenario 2—Climate Change (2050)	Scenario 2—Climate Change (2080)
Heating Setpoint	20	24	22
Cooling Setpoint	28	28	30
Air Infiltration Rate	0.35	1.85	1.55
Clothing Level for Spring	0.74	0.72	0.67
Clothing Level for Summer	0.36	0.36	0.57
Clothing Level for Fall	0.89	1.01	1.01
Clothing Level for Winter	1.1	1.37	1.1
Air Velocity	0.3	0.4	0.4
Air Flow Rate	0.01888	0.00944	0.00944
Roof Insulation	RoofwithInsulationGlassfiber75mm	RoofwithInsulationBoard75mm	RoofwithInsulationXPS25mm
Floor Insulation	FloorwithInsulationXPS50mm	FloorwithInsulationXPS25mm	FloorwithInsulationGlassfiber50mm
Window Construction (North, West)	DblLoE(e2 = 0.1)Tint6mm	DblElecRefColored6mm	TrpLoEFilm(33)Bronze6mm
Window Construction (South)	DblLoE(e2 = 0.2)Clr6mm	DblLoE(e2 = 0.2)Clr6mm	TrpLoEFilm(55)Clr6mm
Window Material Gas	XENON13MM	ARGON13MM	XENON13MM
Window Height (O.K.B)	1.05	0.3	0.5
Overhang Depth	0.25	0.45	0.2
Tilt Angle Overhang	100	120	70
Fin Depth 1 (Left Depth)	0.2	0.5	0.15
Fin Depth 2 (Right Depth)	0.25	0.25	0.5
Tilt Angle Fin	75	100	105
Window Material Shade	LOWREFLECT-MEDIUMTRANSSHADE	LOWREFLECT-HIGHTRANSSHADE	LOWREFLECT-MEDIUMTRANSSHADE
Lighting Control Type	Continuous	ContinuousOff	ContinuousOff
Solar Absorptance	0.91	0.79	0.57
Exterior Door	G0525mmwood	G0650mmwood	G0525mmwood
HVAC Systems	PTHP-DOAS	PTHP	VRF

focuses on key parameters such as heating and cooling setpoints, clothing levels, insulation types, window constructions, and HVAC systems as follows:

• The heating setpoint is adjusted from 20 °C in the current climate to 24 °C in the 2050 scenario, reflecting a necessary increase to maintain thermal comfort as external temperatures rise. In the 2080 scenario, the heating setpoint is slightly reduced to 22 °C, indicating a potential adaptation to changing climate conditions. Conversely, the cooling setpoint remains constant at 28 °C for the 2050 scenario but increases to 30 °C in the 2080 scenario. This trend highlights the anticipated rise in cooling demand due to higher ambient temperatures, indicating a need for adjustments in cooling strategies and energy consumption to maintain comfortable indoor conditions.
• The suggested clothing levels for spring are as follows: 0.74 Clo for sweatpants and a long-sleeve sweatshirt, 0.72 Clo for long-sleeve coveralls and a t-shirt, and 0.67 Clo for a knee-length skirt, long-sleeve shirt, and full slip, corresponding to Scenario 1, Scenario 2 (2050), and Scenario 2 (2080), respectively. For summer, the recommendations include 0.36 Clo for walking shorts and a short-sleeve shirt in both Scenario 1 and Scenario 2 (2050), and 0.57 Clo for trousers and a short-sleeve shirt in Scenario 2 (2080). In the fall, suggested clothing levels are 0.89 Clo for overalls, a long-sleeve shirt, and a t-shirt and 1.01 Clo for two long-sleeve sweaters and a t-shirt, corresponding to Scenario 1, Scenario 2 (2050), and Scenario 2 (2080), respectively. For winter, the recommendations are 1.10 Clo for a knee-length skirt, long-sleeve shirt, half-slip, and long-sleeve sweater in Scenario 1 and Scenario 2 (2080), and 1.37 Clo for insulated coveralls and long-sleeve thermal underwear tops and bottoms in Scenario 2 (2050).
• Insulation strategies evolve significantly across scenarios. Roof insulation transitions from glass fiber (75 mm) in the present climate to board insulation (75 mm) in 2050, and finally to XPS insulation (25 mm) in 2080. This progression indicates a growing recognition of the need for enhanced thermal resistance in response to rising temperatures. Similarly, window constructions vary, with the north and west-facing windows transitioning from double low-emissivity (LoE) glass to more advanced glazing systems, such as triple low-emissivity (LoE) film (33) bronze 6 mm in the 2080 scenario. These changes reflect an adaptive strategy to minimize solar heat gain and improve energy efficiency.
• The HVAC systems also adapt to changing climate conditions. The current climate utilizes a packaged terminal heat pump with dedicated outdoor air systems (PTHP-DOAS), while the 2050 scenario shifts to a standard PTHP system. The variable refrigerant flow (VRF) system in the 2080 scenario suggests a strategic move to enhance indoor air quality and energy efficiency as external temperatures rise and ventilation needs increase.

Table 16. Results of optimal solutions and objective function values.

Different Scenarios	Net Source Energy (Primary Energy Consumption) [kWh]	CO2 Equivalent [kg]	Zone Air CO2 Concentration [ppm]	PPD [%]	Daylighting Reference Point Glare Index Setpoint Exceeded Time [h]
Scenario 1—current climate	23,504.84	7707.92	678.37	9.47	99.73
Scenario 2—climate change (2050)	19,659.19	7010.57	662.92	9.81	72.77
Scenario 2—climate change (2080)	13,066.08	8301.65	678.71	11.71	35.02

presents the optimal solutions derived from the multi-objective optimization process, along with the corresponding values for the objective functions. These results highlight the most effective strategies for achieving energy efficiency, reducing carbon footprint, improving indoor environmental quality, and achieving thermal and visual comfort in building retrofits.

As shown in Figure 39, baseline energy consumption is represented on the far left. It serves as a reference point for evaluating the performance of the scenarios. The analysis of Figure 38 reveals that the implementation of energy performance strategies in both scenarios has led to improved energy efficiency compared to baseline measurements. It highlights the importance of considering both site and source energy in evaluating overall performance and sustainability.

As shown in Figure 40, in the base model, the primary energy consumption is 32,712.09 kWh. After optimization, the primary energy consumption levels are as follows: 23,504.84 kWh for Scenario 1 (present climate), 19,659.19 kWh for Scenario 2 (2050, climate change), and 13,066.08 kWh for Scenario 2 (2080, climate change).

As shown in Figure 41, in the base model, the primary energy consumption per total building area is 412.24 kWh/m². After optimization, the primary energy consumption levels are as follows: 296.21 kWh/m² for Scenario 1 (present climate), 247.75 kWh/m² for Scenario 2 (2050, climate change), and 164.66 kWh/m² for Scenario 2 (2080, climate change).

6. Conclusions

This research presents a simulation-based structure designed to optimize the environmental aspects of building energy retrofits in response to climate change, with an emphasis on occupant comfort and health outcomes. The efficiency of the Non-dominated Sorting Genetic Algorithm II (NSGA-II) optimization algorithm is significantly improved through a parallel computing architecture, allowing for simultaneous simulations on multi-core processors. This methodology includes a results-saving archive to reduce redundancy and lessen computational demands, resulting in a notable decrease in processing time. By considering both environmental impacts and social factors, this framework aids in the advancement of more sustainable and resilient building practices. The optimization process combines EnergyPlus, a comprehensive building energy simulation engine, with NSGA-II, implemented within the jEPlus + EA environment. The framework evaluates a combination of active, passive, and renewable strategies aimed at maximizing environmental performance, enhancing indoor air quality, and minimizing hours of thermal and visual discomfort. It is applied to a residential case study in Sari, Iran, assessing two scenarios related to future climate change. The results imply the following conclusions for the investigated case study:

• The proposed energy retrofit measures (ERMs) have the potential to significantly improve building performance relative to baseline conditions. These measures can reduce primary energy consumption by up to 60% and decrease CO₂-equivalent emissions by as much as 60%. Furthermore, optimal implementation of these solutions can lead to a 65% reduction in predicted thermal dissatisfaction and an 83% reduction in visual discomfort hours, achieving indoor air quality levels that meet ASHRAE recommended standards.
• Climate change significantly impacts building performance by changing external thermal loads, which in turn affects heating and cooling requirements. Rising outdoor temperatures greatly increase cooling demand during the summer months while simultaneously reducing heating demand during the winter. Since heating demand accounts for a considerable segment of the total energy consumption in the case study building, global warming results in decreased primary energy consumption and lower CO₂-equivalent emissions.
• Air infiltration rates exhibit a significant increase from 0.35 ACH (air changes per hour) in the present climate to 1.85 ACH in the 2050 scenario, before stabilizing at 1.55 ACH in the 2080 scenario. This increase may be attributed to the anticipated deterioration of building envelopes under more extreme weather conditions, leading to higher energy losses and increased heating and cooling loads. The implications of this trend highlight the importance of enhancing building airtightness and implementing effective air sealing strategies to mitigate energy loss.
• In parametric studies utilizing the parametrics group within EnergyPlus, among the 24 window shading control strategies available, the strategy labeled “OnNightIfLowOutdoorTempAndOnDayIfCooling” emerges as the most effective in terms of its significant impact on cooling and heating loads. This strategy operates by activating shading at night when the outdoor air temperature falls below a predetermined setpoint. Additionally, if the cooling rate of the zone was non-zero during the preceding time step, shading is also implemented during the day. The application of shading both at night and during the day is contingent upon a specified schedule, if applicable.
• The EnergyPlus energy management system (EMS) was effectively utilized to maintain carbon dioxide concentrations within the six thermal zones between 600 and 700 ppm. This specified range aligns with the recommendations set forth by the ANSI/ASHRAE Standard 62.1, which emphasizes the importance of adequate indoor air quality for occupant health and comfort. Maintaining carbon dioxide levels within this range is crucial, as elevated concentration can lead to discomfort, reduced cognitive function, and other adverse health effects. The successful management of indoor air quality through the strategic implementation of the EMS demonstrates the potential for advanced building management systems to facilitate compliance with recognized environmental standards, thereby enhancing the overall performance and sustainability of the indoor environment. This approach not only ensures regulatory compliance but also contributes to the well-being of occupants by providing a healthier and more comfortable living or working space.
• For both scenarios analyzed, the implementation of energy-efficient measures is strongly recommended. These measures include the installation of extruded polystyrene (XPS) and fiberglass thermal insulation with a thickness ranging from 25 to 75 mm for the flooring, as well as the application of fiberglass, board, and XPS insulation with similar thickness specifications for the roofs. This approach is anticipated to significantly enhance the thermal performance of the building envelope, leading to improved energy efficiency and reduced overall energy consumption. For south-facing windows, it is recommended to use double low-emissivity clear glass (e2 = 0.1 to 0.2) for both Scenario 1 and Scenario 2 (2050), and triple low-emissivity film (55) clear 6mm for Scenario 2 (2080), with 13mm spaces filled with xenon for Scenario 1 and argon and xenon for Scenario 2. For north- and west-facing windows, double low-emissivity tinted glass (e2 = 0.1) is advisable for Scenario 1, while Scenario 2 (2050) should feature double electrochromic reflective colored 6mm glass, and Scenario 2 (2080) should use triple low-emissivity film (33) bronze 6mm, with 13mm spaces filled with xenon for Scenario 1 and argon and xenon for Scenario 2.
• For HVAC systems, it is recommended to install Packaged Terminal Heat Pumps (PTHPs) with inverter technology (COP = 4.3, EER = 3.81) alongside a dedicated outdoor air system (supply fan total efficiency = 0.9) and variable refrigerant flow (VRF) system (gross rated cooling COP = 4.3 and gross rated heating COP = 4.6). Additionally, the installation of 25 m² of photovoltaic panels is suggested. Temperature controls should be set at 28 °C and 30 °C for cooling, and 20 °C, 24 °C, and 22 °C for heating. However, if economic considerations are a primary concern, renovating the windows and HVAC systems may not be advisable.
• The analysis of clothing levels reveals a decrease in thermal comfort adaptation over time. For spring, clothing levels decrease from 0.74 in the present climate to 0.72 in 2050 and further to 0.67 in 2080. Similar trends are observed in summer and fall, with clothing levels in summer remaining constant at 0.36 before increasing to 0.57 in 2080. This suggests a potential shift in occupant behavior in response to changing climatic conditions, which may affect indoor thermal comfort and energy consumption patterns.
• The north-windows-to-wall ratio for both scenarios are 41.34% for Scenario 1, 60.72% for Scenario 2 in 2050, and 55.55% for Scenario 2 in 2080. The fluctuating percentages over the timeframes suggest an adaptive approach to building design, responding to evolving climate conditions and energy efficiency standards.
• In Scenario 1, the overhead lights are set to Continuous Lighting Control, which allows them to gradually and linearly dim from maximum electric power and light output to a minimum level as daylight illuminance increases; the lights remain at the minimum output even with further increases in daylight. In Scenario 2, set to continuous-off, the lights behave similarly but switch off completely when they reach the minimum dimming point, rather than remaining illuminated.
• The solar absorptance values for the external walls and roof in both scenarios are as follows: 0.91 for classic bronze, 0.79 for slate gray (redwood, teal), and 0.57 for rawhide. These values play a significant role in energy efficiency. Surfaces with high solar absorptance contribute to increased heat gain inside buildings, which can affect cooling loads, while surfaces with lower absorptance help maintain cooler indoor temperatures by reflecting more sunlight. Therefore, based on the analysis obtained for both scenarios, it is recommended to use materials with solar absorptance values above 0.5.
• In all optimal solutions for both scenarios, the installation of photovoltaic (PV) panels is consistently chosen. This trend is primarily due to Iran’s favorable climatic conditions, characterized by high solar radiation, along with the government’s commitment to purchasing generated solar electricity at rates higher than those of grid electricity.

The findings highlight the urgent need for collaborative action among diverse stakeholders, governments, architects, local communities, and researchers to cultivate a shared commitment to climate-resilient futures. Engaging these groups can facilitate the adoption of innovative design practices and retrofitting solutions that not only mitigate environmental impact but also enhance the quality of life for residents. This research provides a solid foundation for future studies and initiatives aimed at creating sustainable, energy-efficient, and comfortable living spaces that can adapt to the evolving challenges of climate change. In Iran, the existing building technology frequently relies on outdated and inefficient systems, resulting in elevated energy consumption. Enhancements in building design and energy efficiency present opportunities for significant reductions in emissions; however, the adoption of renewable energy sources remains limited. Rapid population growth and urbanization in major Iranian cities further exacerbate energy demand, leading to increased emissions as residential and commercial buildings strain the existing energy grid. The proposed approach, characterized by a robust metaheuristic algorithm and parallel processing capabilities, offers advanced computational methods for addressing complex building energy performance optimization challenges. Future studies will apply this framework to diverse case studies that encompass a range of construction technologies, functionalities, and climatic conditions. This comprehensive analysis aims to establish nationwide guidelines for energy retrofit projects to achieve environmentally friendly buildings. Furthermore, the adaptability of the proposed model will support the design of nearly zero-energy, zero-carbon future, and climate-resilient buildings, which is particularly crucial given the high rate of new construction in developing countries in the Middle East compared to their developed counterparts. Implementing this method at the district level will provide valuable insights for policymakers in planning energy retrofits for neighborhoods.