Towards Climate-Resilient Dwellings: A Comparative Analysis of Passive and Active Retrofit Solutions in Aging Central European Housing Stock

Abstract

This article evaluates the effectiveness of various energy retrofit solutions—both passive and active—for reducing energy demand and improving indoor thermal conditions in apartments of typical multifamily buildings in Central Europe, considering current and future climate conditions. This study combines computer-based co-simulations (EnergyPlus and CONTAM) with in situ thermal measurements to identify challenges in maintaining indoor thermal conditions and to support model validation. Key indicators include the number of thermal discomfort hours and heating and cooling demand. The evaluated strategies include passive measures (wall insulation, green or reflective roofs, roller blinds, solar protective glazing) and active solutions such as mechanical cooling. The comfort operative temperature range of the adaptive model is adopted as a measure of thermal comfort and the energy demand in individual apartments as a measure of energy efficiency. The simulation results showed that solar protective glazing combined with a reflective roof reduced thermal discomfort hours by up to 95%, while modern windows alone decreased them by 90% and lowered heating demand by 18%. In contrast, typical passive solutions such as internal blinds or balconies were significantly less effective, reducing discomfort hours by only 11–42%. These findings highlight that, while no single retrofit measure is universally optimal, well-selected passive or hybrid strategies can substantially improve summer comfort, maintain winter efficiency, and reduce long-term reliance on energy-intensive cooling systems in aging multifamily housing.

1. Introduction

In the face of ongoing climate change, the energy retrofit of residential buildings is becoming not only a matter of energy efficiency and cost savings, but also one of residents’ comfort and health. With rising average temperatures and the increasing frequency of extreme weather events (such as heatwaves), the challenges associated with building modernization are taking on a new dimension. Traditional energy retrofits have primarily focused on reducing heat loss and improving energy efficiency during the winter season. While many buildings now have insulated facades, reducing heating use, global warming has led to apartment overheating during summers [1,2,3,4,5]. Furthermore, insufficient ventilation [6,7] and the behavior of residents, such as the improper use of windows or blinds [8,9], can exacerbate the risk of overheating. Many older residential buildings, particularly in Central and Eastern Europe, were not designed with high temperatures and increased heat loads in mind. Transforming these structures into more heat-resilient infrastructures can be both costly and technically challenging, due to structural or urban planning limitations, for example. This necessitates a thorough analysis of how new solutions can be implemented in existing buildings, such as appropriate solar insulation, shading systems, or passive ventilative cooling, which were not always considered in older designs.

The problem of the overheating of dwellings covers various climatic areas, not only the warmest ones [10]. The report of the Ministry of Housing, Communities and Local Government [11] lists, among others, Germany and France, where air-conditioned buildings are not standard. The literature clearly indicates the benefits of having buildings with a more efficient thermal envelope; but, relying solely on retrofit strategies based on building insulation can yield unexpected results if the risk of overheating is not taken into account [12,13]. Unfortunately, heatwaves are expected to become more frequent and with higher temperatures as global warming continues [14]. To address this problem, building retrofits in the temperate climatic conditions of Europe should balance winter heat loss reduction with passive cooling measures in summer [15,16,17].

1.1. Literature Review

Numerous studies have demonstrated the potential of passive strategies in reducing energy demand and mitigating summer overheating in existing buildings. These strategies include external insulation, shading devices, reflective surfaces, night ventilation, and improvements in window design [10,16,18,19,20,21,22,23,24,25,26]. In their conclusions, the authors highlight the crucial importance of properly selecting window glazing, given its substantial impact on energy efficiency. Several simulation-based analyses confirm that combining passive solutions—such as insulation and shading—with hybrid systems that integrate passive techniques and active air conditioning can significantly reduce energy use and improve indoor human comfort in both hot and temperate climates [23,24,25,26,27,28,29,30,31,32,33,34,35]. The authors acknowledge that ventilative cooling, particularly night-time ventilation when integrated with other passive strategies, can result in a near-complete reduction in thermal discomfort. The benefits of ventilative cooling have also been widely recognized and highlighted within the framework of the International Energy Agency’s Annex 62 [36]. Recent reports from international bodies such as UNEP (United Nations Environment Programme) [37] and GlobalABC [38] also emphasize the importance and feasibility of implementing passive cooling during retrofit interventions, particularly in urban housing.

Despite the availability of passive technologies, one of the most critical and often overlooked aspects is user behavior. Studies show that occupant actions—especially window operation—can dramatically influence both cooling potential and overall building performance [39,40,41,42]. Behaviorally driven ventilation effectiveness varies widely, and probabilistic models have been proposed to account for this variability [43,44,45]. So far, no universal model for modeling the behavior of building users has been developed. This may be particularly difficult for residential buildings, where the users are people of different ages who have very different preferences regarding indoor conditions and, consequently, very different settings related to heating, ventilation, and air conditioning systems. Although the ASHRAE Global Occupant Behavior Database [46,47] provides a useful international reference, its coverage of window-opening behavior—especially for passive cooling purposes in residential buildings—is limited, both geographically and seasonally. As a result, tailored or localized modeling approaches are still needed to realistically capture human interaction with natural ventilation systems.

This variability in occupant behavior becomes even more significant when considered in the context of climate change, which is projected to exacerbate the risk of summer overheating and thermal discomfort—even in buildings that have undergone energy retrofits [48,49,50,51,52,53]. For example Escandón et al. [51] predicted an increase of 35% in the thermal discomfort hours in summer for multifamily buildings constructed in Spain. Similar conclusions were reached by Moazami et al. [52], who calculated the energy use of 16 standard ASHRAE buildings under different climate scenarios for Geneva. Recent multi-city projections of cooling degree days in seven European countries indicate that this indicator may roughly double by 2050—with the largest relative increases occurring in northern locations—highlighting the need to anticipate changes to the building stock and shifts in occupant behavior when planning retrofit strategies for European climates [54]. While passive methods such as night ventilation and solar shading can delay or mitigate the need for mechanical cooling, their effectiveness under future climate scenarios—and in relation to user behavior—remains insufficiently studied [55].

1.2. Research Gap, Aim, and Scientific Novelty

Buildings in Central and Eastern Europe, particularly older multifamily housing, face significant challenges in maintaining thermal comfort with low energy use under changing climatic conditions. In Poland, residential buildings account for a substantial share of the country’s total energy use, reaching 20.2% in 2021 [56], which is similar to the EU average. Of the 6.8 million residential buildings [57] (14.6 million apartments), 68% are in multifamily structures, mostly built in the 1960s–1970s with prefabricated concrete [58]. Thermally upgrading this aging building stock has become a strategic priority, not only for energy savings but also for ensuring acceptable indoor thermal environments.

Despite growing interest in methods of reducing room overheating, many areas are still not sufficiently researched. There is a lack of systematic studies on the synergistic effect of combining passive and active approaches. There is a gap in the analysis of how the effectiveness of these methods changes with predicted climate changes, such as increasing average temperatures or intensifying heatwaves, especially for naturally ventilated buildings. Most simulation studies assume constant air exchange, which greatly simplifies the problem and does not take into account the variability of the infiltrating air flow, especially in the case of the probabilistic behavior of people in terms of opening windows. The influence of wind and the stack effect are practically omitted in building performance simulation due to the difficulty of modeling gravity ducts in buildings, which affects the accuracy of the assessment of instantaneous heat loads, especially in multi-story buildings. A significant part of the research is focused on the design of new buildings that are resistant to climate change. Meanwhile, the issues of energy-efficient passive renovation methods of existing buildings are relatively poorly developed. This is important because most of the global building stock consists of older structures that require adaptation. Solving these problems can significantly contribute to sustainable building development and enhance energy efficiency in the context of climate change.

Apartments on the top floor of multifamily buildings are more exposed to the influence of the external environment than other apartments due to contact with the roof, especially when the roof is not insulated. Therefore, this research aims to assess renovation solutions that improve thermal conditions in such apartments, taking into account the probabilistic behavior of residents in operating windows in Polish buildings with natural ventilation.

This study answers two key questions: (1) How effective are different passive cooling energy retrofit solutions—individually, in combination, and together with mechanical cooling—in controlling the indoor thermal climate and saving energy in the current and future conditions of central Europe? (2) Will passive solutions provide thermal comfort conditions in dwellings in the future? The effectiveness of several energy retrofit measures—including shading, envelope insulation, and solar reflectance—is assessed using the building performance simulation method. First, the temperature measurements are taken in existing buildings to determine the problem of maintaining the desired thermal conditions to ensure thermal comfort for residents. Then, a multivariant thermal assessment of apartments for the entire year is performed on a building model covering the top and penultimate floors based on co-simulations using EnergyPlus and CONTAM programs.

The main contributions and innovative aspects of this study are summarized as follows:

• Dynamic thermal assessment with ventilation modeling: A novel method is developed to integrate mass and energy transfer calculations, enabling the determination of variable ventilation airflow at each simulation step. This allows for precise evaluation of instantaneous human thermal comfort and air cooling potential throughout the year. The approach addresses a significant research gap in assessing naturally ventilated buildings, especially those with gravity ventilation ducts, under real-time internal and external thermal loads.
• Behavior-integrated energy modeling: By incorporating realistic occupant behavior, particularly window opening patterns, this study reveals how human actions influence the performance of HVAC systems and passive cooling techniques. This behavioral aspect enhances the realism and applicability of simulation results in residential settings.
• Comparative evaluation of passive and active cooling strategies: This study compares passive, active, and hybrid cooling methods, both individually and in combination, identifying the most effective solutions for maintaining indoor thermal comfort in the face of climate warming. This comparative framework is especially relevant for retrofitting older buildings in Central and Eastern Europe.

Together, these innovations provide a comprehensive and practical framework for improving the thermal performance of aging residential infrastructures in a changing climate.

2. Methods

2.1. Research Object

One segment of a multifamily building with five residential floors, built in the 1960s and located in southern Poland, was selected for this study. Each floor contains three apartments with areas ranging from 43 to 63 m². Each apartment consists of two main rooms, kitchen and bathroom with toilet. Six apartments on the two upper floors were taken into account (Figure 1).

The apartments have windows facing north-east and south-west. In its current form, the building has a reinforced concrete structure (prefabricated walls and nonventilated reinforced concrete roof) and has been partially renovated (only insulation of external walls). The building has typical double-glazed windows (argon gas 90% in the chamber) with a solar heat gain coefficient of 0.64 and a heat transfer coefficient U = 1.1 W/m²K. The building is equipped with a central heating system with radiators. Fresh air enters the building through leaks in windows and doors; after mixing with air in the rooms, it is removed through ventilation grilles connected to gravity chimneys. Each apartment is equipped with two gravity chimneys; one is located in the kitchen, the other in the bathroom. This is a typical ventilation system solution in residential buildings in Central and Eastern Europe.

2.2. Locations and Climate Scenarios

The Central Europe location selected for analysis, which corresponds to one of the major cities in southern Poland, was Katowice. This location is characterized by a temperate transitional climate (Dfb class according to the Köppen–Geiger classification [59]). For the simulations, the typical meteorological year (TMY) [60] and future climate were calculated based on global warming forecasts for 2050 (Figure 2). In the TMY, the temperature changed from −18.7 °C to 31.0 °C during the year, with an average value of 8.1 °C. The predicted climate data were calculated from the A2 emissions scenario (one of the most popular scenarios in the literature [61]). For the climate, the temperature ranged from −13.9 °C to 37.6 °C, with an average annual temperature of 11 °C.

2.3. Measurements In Situ

The measurements had two aims: to present the problem of overheating in apartments and to validate the thermal model. For a period of 8 months in 2021, the indoor air temperature and relative humidity were recorded (with a 5 min time step) in two apartments (A11 and A12). Data loggers AR236.B (APAR Control, Raszyn, Poland) with an accuracy of ±0.3 °C for temperature and ±2%RH for humidity were used. In each apartment, there were two recorders from January to August to check the thermal conditions in different periods of the year. For the analysis, the average value of temperature in each apartment (at each recording step) was calculated. During the measurements, the residents used the rooms freely, including manual operation of windows and blinds. Historical climate data were imported from the nearest weather station of the Institute of Meteorology and Water Management–State Research Institute [62].

2.4. Building Thermal Model and Simulation

The thermal calculation using the multizone model of the selected building fragment was performed using EnergyPlus 9.4 (US Department of Energy, Washington, DC, USA) [63] connected with CONTAM 3.4 (National Institute of Standards and Technology, Gaithersburg, MD, USA) [64] simulation programs (Figure 3). In this research, due to the lack of a gravity chimney model in the air flow network (AFN) module of EnergyPlus and available in CONTAM and the simplicity of modeling in the CONTAM program, it was decided to model the natural ventilation in this program. This connection was made using the functional mock-up unit (FMI) standard [65].

The apartments on the top two floors were modeled (Figure 1). Each apartment was treated as a separate thermal zone, which was used as an open space due to its small size. This simplification had little impact on the results because, in reality, internal doors in apartments are usually open, allowing air to flow freely between rooms. Internal walls were considered as additional thermal mass inside the apartments. This model took into account uncontrolled airflows, penetrating through openings and leaks in the envelope of the building. The following airflows of natural (gravitational) ventilation were balanced: flow through gaps in windows, flow through an open or tilted window, and flow through gravitational ventilation ducts. It was assumed that the large balcony window with dimensions of 62 by 192 cm could be opened or tilted, and the small window with dimensions of 62 by 97 cm could be tilted. The model was described in detail in ref. [7]. The input data for the thermal model used in the case studies are listed in

Table 1. Input data for the thermal model.

Model Component	Input Value	Operating Time	Comment
Heating	Indoor temperature setpoints: apartments: 21 °C; staircase: 16 °C	From September to May	According to EN 16798-1:2019 standard [66]
Cooling	Indoor temperature setpoints: 26 °C (only apartments)	All year; only in cases with a cooling system	According to EN 16798-1:2019 standard [66]
Occupants	126 W per person during the day, 73 W per person during the night (sleeping)	According to the individual schedule in each apartment (Figure A1), there was always at least one person in the apartment	According to ASHRAE-55 standard [67]
Lighting	Apartments 3.5 W/m2	Day: turned on when the lighting intensity is lower than 250 lm/m2 Night: turned off	LED lamps and typical lighting intensity for residential premises were adopted
Equipment	Electric hob: 500 W; fridge: 150 W or 250 W; computer: 100 W; and TV: 175 W	According to the individual schedule in each apartment (Figure A2)	According to the typical home equipment power
Infiltration	Airflow calculated in each time step for each zone	All year	One-way flow using POWERLAW model
Opening windows	Variable airflow calculated in each time step for each zone	According to schedule	Two-way flow model (single opening)
Mechanical ventilation	Constant airflow in each time step for each zone	All year; only in cases with a cooling system	According to EN 16798-1:2019 standard [66]
Window blinds	Internal blinds with a solar transmittance of 0.4 and a solar reflectance of 0.4	ON-OFF operating. ON mode: the operative indoor temperature exceeds the comfort temperature by 1.5 K and the perpendicular to the window solar radiation exceeds 150 W/m2	Probability of blinds being closed and opened: 0.5

.

The key part of this research was to properly model people’s behavior regarding window opening. In reality, this activity is uncontrolled and depends on the individual preferences of the residents. In this study, a stochastic window opening control model was created, taking into account the behavior of the residents, which depended on the outdoor temperature, indoor comfort temperature, wind speed, and air change rates. It proposed a probabilistic approach in which the window settings could change during the day (with probabilities of 0.5) and at night (0.25), with specific time and degree restrictions for opening and closing windows. The key assumptions included the following: (i) initial constraints: window opening depended on the wind speed, the difference in indoor and outdoor temperatures, and the comfort temperature (windows could only be opened if the wind speed was low, the ambient temperature was lower than the indoor temperature, and the indoor operative temperature was above the comfort level); (ii) window opening levels: tilting the balcony window, tilting the balcony window and a small window, fully opening the balcony window, and tilting the small window, (iii) time constraints: changing the window opening level was possible during the day at hourly intervals; (iv) additional rules: constraints were introduced to avoid draughts and excessive cooling of the rooms. The limit values of input parameters (e.g., wind speed and temperature differences) at which the window opening was changed were optimized using genetic algorithms. The parameters were selected to minimize the number of hours of thermal discomfort with limited air change rates. The window opening controller was integrated with the EMS module of EnergyPlus. The window setting was changed, taking into account the stochastic nature of human behavior in controlling the windows. Each retrofitting scenario was simulated ten times, with variations in results arising from the applied stochastic model. The final results used in the analysis represent the average across all simulations for each case. A detailed description of the model and optimization can be found in our previous paper [7].

2.5. Thermal Model Validation

Validation of simulation results was carried out based on indoor temperature measurements in two apartments, A11 and A12, in the summer period (June–August). During this period, the heating system was not working and the indoor temperature resulted only from the heat balance in the apartments. The real use of the rooms was not recorded. Therefore, for validation, the indoor heat gains and the window opening schedule were calibrated to match the temperature variation courses. The model also included historical climate data.

The two indicators given in the ASHRAE guide [68] were used to assess the compliance of the simulation model: the normalized mean bias error (NMBE) and the coefficient of variation of root mean squared error CV(RMSE), which are calculated from

$$\mathrm{N}\mathrm{M}\mathrm{B}\mathrm{E}={\displaystyle \frac{\sum _{i=1}^n(M_i-S_i)}{(n-p)·\overline{M_i}}}\times 100,$$

(1)

$$\mathrm{C}\mathrm{V}(\mathrm{R}\mathrm{M}\mathrm{S}\mathrm{E})={\displaystyle \frac{1}{\overline{M_i}}}\sqrt{{\displaystyle \frac{\sum _{i=1}^n{(M_i-S_i)}^2}{n-p}}}\times 100.$$

(2)

where M_i is the measured value, S_i is the simulated value, n is the number of compared values, $\overline{M_i}$ is the mean of the measurement values, and p is the number of adjustable model parameters. The ASHRAE guide suggests these as maximum errors for energy use in buildings, but some researchers also take them into account when comparing indoor temperatures.

Measured indoor temperature data, originally recorded with a 5 min time step, were averaged over three consecutive intervals to match the 15 min time step used in the simulation results for validation purposes (Figure 4). Empirical verification regarding indoor temperature makes the results of the simulation calculations reliable, accurate, and valid for use in this study. The values of the NMBE and CV(RMSE) indicators are small: NMBE is 0.5% and CV(RMSE) is 2% in apartment A11 and 3% in apartment A12. Correlation coefficients R are very high and range from 0.92 in A12 to 0.97 in A11.

2.6. Energy Retrofit Improvements and Evaluation Criteria

The analysis was carried out for three main groups of improvements—building insulation, solar radiation control, and mechanical cooling (Figure 5)—which were divided into nineteen detailed variants of solutions affecting thermal conditions and energy demand in the building (

Table 2. Considered cases and solutions to improve thermal conditions in apartments.

Improvement		Case
		1	2	3 *	4	5	6 **	7	8	9	10	11	12	13	14 *	15 **	16	17	18	19
Building insulation	Walls insulated		✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
	Roof insulated			✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
Natural ventilation	All windows closed					✓
	Opening windows	✓	✓	✓	✓		✓	✓	✓	✓		✓			✓	✓	✓	✓	✓
Solar radiation control	Internal blinds	✓	✓	✓		✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
	Balcony						✓									✓
	Green roof							✓									✓
	Reflective roof								✓			✓	✓					✓
	Solar protective glazing									✓		✓		✓					✓
Mechanical cooling	Air conditioners, mechanical ventilation										✓		✓	✓						✓
Climate	Current (TMY)	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
	Future 2050														✓	✓	✓	✓	✓	✓

* Base models for current and future climates. ** Includes two sub-cases a and b with different balcony depths.

). In the first part of the analysis, the influence of the envelope insulation was checked. Next, simulations were performed using passive solar control, i.e., balconies, green roofs, and a roof covered with a membrane reflecting solar radiation, alongside a reflective roof and windows with better parameters from the point of view of solar protection. Finally, active thermal condition control, i.e., a mechanical cooling system, was simulated. Some combinations of improvements were also calculated, such as solar protective glazing and a cool roof, mechanical cooling and solar protective glazing, and mechanical cooling and a cool roof.

Six cases (in blue line in Figure 5) were simulated using both current climate and future predicted climate data. In all cases (except for two), the rooms were aired by opening windows and internal roller blinds were turned on. In order to demonstrate the validity of using ventilation by opening windows and blinds on windows, additional simulations were carried out without these elements. All the simulations were performed with a 15 min time step for the whole year.

For cases without mechanical cooling, the number of thermal discomfort hours throughout the year and the indicators of the heating demand per square meter of the floor were evaluated. For cases with mechanical cooling, it was assumed that the comfort conditions were met, but this criterion was not assessed while the heating and cooling energy were evaluated. The annual energy demand in individual apartments was adopted as an energy efficiency index, expressed per square meter of floor area. In turn, an adaptive model based on the recommended ranges of indoor operative temperature according to the EN 16798-1:2019 standard [66] was adopted as a measure of human thermal comfort. In this study, the limits of the comfort temperature were calculated for category II of the thermal environment, i.e.,

$${To}_{comfort}=0.33\times Tout+18.8\pm 3.0,$$

(3)

where Tout is the weighted mean of the previous 7-day daily mean outdoor air temperature (°C); the limits only apply when the running mean outdoor temperature is greater than 10 °C. Below this threshold—which applies during the colder seasons of the year—the adaptive comfort model assumes a constant of the operative comfort temperature of 20 °C for category II indoor environments. This condition was maintained by the heating system. In EnergyPlus, the operative temperature is calculated as the average of the indoor air temperature and the mean radiant temperature of the zone. The “zone-average” method was used to calculate the mean radiant temperature as a weighted average. This method assumes that the person is at the center of the zone. This model is commonly used in the thermal simulation of residential buildings where the location of the occupants changes dynamically.

2.6.1. Building Insulation

Three cases were considered as follows:

• Completely insulated building (used as a base model for comparison): insulation of the flat roof with 20 cm of mineral wool (U = 0.146 W/m²K) and the existing condition of the external walls insulated by 15 cm of mineral wool (U = 0.196 W/m²K);
• Partly insulated building (the building in its current state): external wall insulated with 15 cm of mineral wool (U = 0.196 W/m²K) and the flat roof uninsulated (U = 0.632 W/m²K);
• Uninsulated building (original construction condition from the 1960s): the external wall has only old insulation made with 5 cm of mineral wool between the structural layers of the wall (U = 0.750 W/m²K) and flat roof uninsulated (U = 0.632 W/m²K).

2.6.2. Solar Radiation Control

Two variants with balconies of 120 and 160 cm depth for balcony doors were considered (Figure 6). The width of the balcony for apartments A1, A3, A11, and A13 was the width of the external walls, i.e., 4.30 m. For apartments A2 and A12, the width of the balcony was 2.60 m, which was approximately 1/3 of the width of the external wall. For apartments A11–A13, a slab acting as a roof was also planned.

When designing the green roof structure, the main assumption was to select parameters that would best reduce indoor temperature in the summer. The main parameter considered is the leaf area index (LAI), which means the ratio of leaf area to ground area. The higher the parameter value, the greater the plant density in relation to the ground area in which it is located. The LAI = 5 was assumed aligning with Mahmoodzadeh [69]. The model took into account the height of the plants (400 mm) and the thickness of the substrate (150 mm).

Another analyzed solution that affected the thermal conditions in apartments was the use of a cool roof. Such a solution is used in buildings to reduce energy demand by using reflective coatings (with a high solar reflectance coefficient). In this study, a layer with a solar energy absorption coefficient of 0.22 was assumed.

As one of the variants, Pilkington solar control windows were tested, consisting of Pilkington Suncool 70/35 glass on the outside, Argon gas (90%), Pilkington Optifloat Clear glass on the inside, Argon gas (90%), and Pilkington Optitherm SI3 glass on the inside (solar heat gain coefficient of 0.35) [70].

2.6.3. Mechanical Cooling and Ventilation

Split air conditioners with a cooling air temperature setpoint of 26 °C were used for mechanical cooling. Air conditioners are one of the most commonly used solutions to improve thermal conditions in residential buildings in the summer. Due to the mechanical cooling system, the windows were not opened by residents. The calculation model took into account the use of constant air volume ventilation with heat recovery with an efficiency of 70% to provide the required amount of fresh air. The system provided constant airflow throughout the year, which was 126 m³/h for each apartment, according to the EN 16798-1:2019 standard [66] (design air flow rates by room and building type—two main rooms in the dwelling, kitchen and bathroom).

3. Results

3.1. Assessment of Indoor Thermal Conditions Based on In Situ Measurements

The recorded values of temperature and relative humidity are presented in Figure 7. The apartment A12 was characterized by greater load variability, which resulted in greater instantaneous fluctuations in the indoor temperature. The same annual trend can be seen in both apartments. In the winter–spring period, the indoor temperature was in the range of 20–23 °C. During this period, the thermal conditions of the apartments were shaped by the heating system, which in Poland generally operates from September to May. From June, the indoor temperature increased above 26 °C, with values reaching as high as 30 °C. During the recording, residents could open windows and use sunshades according to their preferences, and these activities were not recorded. Despite this, the temperature above 26 °C occurred for 49% and 62% of the summertime in apartments A12 and A11, respectively. Relatively high humidity was recorded in apartment A12 in winter, which was the result of using a humidifier. In summer, relative humidity in both apartments was in the range of 40 to 60% (with minor exceedances up to a maximum of 70%), thus meeting the recommended requirements of the EN 16798-1:2019 standard [66].

It should be noted that the actual use of the rooms was not systematically recorded during the monitoring period. The absence of detailed information on occupant behavior (e.g., window opening patterns, shading use, occupancy schedules) introduces some uncertainty in interpreting the measured data and limits the ability to fully correlate observed thermal conditions with specific user actions.

3.2. Numerical Analysis

The following symbols were adopted for apartments in the analyses: on the penultimate floor, A1, A2, and A3; on the top floor, A11, A12, and A13 (Figure 1).

3.2.1. Impact of External Envelope Insulation (Cases 1 to 3)

Insulating the external walls reduced heating demand by an average of 42%, with the largest drop (75%) in corner apartment A3 without direct roof exposure (Figure 8a). Adding roof insulation brought further reductions, particularly for top-floor units (e.g., −26 kWh/m² in A13); this was a 10-fold reduction. The reduction in heat demand was also visible on the lower floor, which does not touch the roof. This was caused by the increase in the indoor temperature on the top floor in the transitional period (i.e., fall and spring) and, therefore, additional heat gains in the apartments below were present. It can also be seen that, after the full insulation of the building envelope, the annual heat demand was small and amounted to a maximum of 5.5 kWh/m² in apartment A12.

However, wall insulation alone slightly worsened the summer comfort of the residents (due to higher heat accumulation) by 11–29% in apartments on the penultimate floor and by 3–6% on the upper floor (Figure 8b). On the other hand, the reduction in the heat transfer coefficient of the flat roof (after insulation) had a positive effect on the thermal condition in the building, increasing the period of thermal comfort compared with the building before the renovation. For example, for apartment A11, the number of hours of thermal discomfort was reduced by about 110 h per year, which is half of these hours in the same apartment but with an uninsulated roof. The beneficial effect was not influenced by the insulation of the partition itself (because this procedure worsens the thermal conditions) but by the ventilation of the building by opening windows, which was more frequent and intensive due to higher indoor temperatures. The effect was also visible in the apartments on the lower floor, but it was, on average, only 18% (in the range of 8–30%).

3.2.2. Impact of Interior Blinds on Windows (Case 4)

Internal blinds had negligible influence on heating demand (<2%) but reduced annual discomfort hours by 11–19% (Figure 9). The effect was more pronounced on the penultimate floor (over 17% reduction) than the top floor (13%), reflecting differences in solar exposure. The heating demand effect was poor because, in winter, the blinds were very rarely used.

Internal blinds may be a temporary solution and, as climate changes, their impact on improving indoor conditions will decrease [7]. Due to the construction of the blinds on the inside of the window, they provide little insulation against solar radiation and external gains, i.e., heat penetrates the interior of the room.

3.2.3. Impact of Additional Ventilation by Open Windows (Cases 3 and 5)

Keeping the windows closed reduced the heating demand by 27% in apartment A13 to 37% in apartment A02, but caused extreme summer overheating, with the number of discomfort hours reaching 4000 h, which is almost half of the year. Window ventilation increased air change rates nine-fold and reduced discomfort by over 90% (Figure 10). With closed windows, the air change rate oscillated around the value of approximately 0.25 h⁻¹, which, for rooms with a volume of 130 to 190 m³, provided a fresh airflow in the amount of 33 to 48 m³/h. This airflow is too small and does not meet hygiene standards [66].

The effect of occupant behavior on window operation was also examined with respect to air exchange and thermal conditions. The maximum and minimum results from 10 simulations of case 3 were compared. Since the daily probability of opening or closing windows was set at 0.5, any user “errors” (e.g., not opening or closing a window when needed) could occur at different times of the year in each simulation. This variability significantly impacted indoor thermal conditions. While the average annual air change rate remained virtually unchanged, the number of discomfort hours varied by 34 to 66 h between simulations (Figure 10). In the case of apartment A11, this difference amounted to nearly half of the average discomfort hours across all simulations, highlighting the substantial influence of occupant behavior on indoor environmental quality. The impact of the ventilation system on thermal conditions and the heating demand for apartments is described in detail our previous study [7].

3.2.4. Impact of Additional Solar Radiation Control (Cases 6 to 9)

Solar radiation control strategies had varied effects (Figure 11). Balconies negligibly increased annual heating demand (1–2% due to lower solar gains in winter) but significantly reduced summer discomfort. For example, in apartment A12, a 160 cm deep balcony reduced discomfort hours by 86 h (31% reduction), while a 120 cm balcony reduced them by 67 h. The highest percentage reduction in discomfort hours—up to 42%—was observed in apartments A1–A11 with the deeper balconies. An increase in balcony depth by 40 cm resulted in an average improvement of 21 h in thermal comfort.

The green roof yielded a 20% reduction in heating demand on the top floor and 7% on the penultimate floor due to inter-floor thermal interaction. The top-floor apartments also experienced the largest reduction in discomfort hours—up to 60%—while the average improvement on the floor below reached 14%. These results indicate that cooler top-floor temperatures contribute to reduced thermal loads in the apartments below, supporting the use of green roofs as a sustainable alternative to mechanical cooling. Similarly, reflective roofs provided up to 57% discomfort reduction on the top floor but increased heating demand in winter (up to +13%).

Installing solar-protective glazing with low U-values reduced heating demand by 18% on average, with greater savings (by 7 percentage points) on the penultimate floor due to a higher window-to-surface ratio. The smallest savings were observed in apartments A2 (12%) and A12 (7%) due to their central location and limited glazing exposure. Nevertheless, this solution provided the greatest thermal comfort improvement: an average 90% reduction in discomfort hours on the penultimate floor and 85% on the top floor. In apartment A1, discomfort hours were eliminated entirely; in A11, they were reduced to 8 h. These apartments with fewer occupants (only one to two persons, see Figure A1), and thus lower internal gains, experienced the largest proportional benefits from solar gain reduction.

3.2.5. Mechanical Ventilation and Cooling (Case 10)

The model that includes mechanical cooling and mechanical ventilation is characterized by increased heating demand compared with a building without such systems (Figure 11). The increase in heat demand was a consequence of ensuring thermal comfort conditions and providing a larger amount of fresh air required by hygiene standards. The heat demand increased in most apartments by about 35–60% for apartments on the penultimate floor and almost 1.5 times for apartments on the top floor. In this case, the heat for ventilation increased despite the use of heat recovery. In apartment A12, the increase was small, only 28% and, in apartment A2, a decrease in heat demand was even observed. In the case of the middle apartments, which are most exposed to overheating, the windows were opened most often in the base model, resulting in the highest heat demand index in the base variant. Therefore, the increase in the heat demand for ventilation was no longer significant in this case and, in apartment A2, the heat demand for ventilation was even lower in the case of mechanical ventilation. More on this topic is provided elsewhere [7].

3.2.6. Combination of Passive and Active Solutions (Cases 11 to 13)

Typically, during renovation, several major works are carried out with the aim of modernizing the building. To make the best use of the solutions presented in this study, several variants were combined in different ways, thus creating their combinations. Figure 12 compares heating and cooling demands in buildings with such solutions as sun protection windows with a reflective roof, mechanical cooling with a reflective roof, and mechanical cooling with new windows. The solar protective glazing was the most advantageous variant so far in terms of ensuring thermal comfort to residents. Combining it with a reflective roof gave an even better effect, especially on the top floor, where the maximum number of hours of discomfort was only 19 h in apartment A12 (without the reflective roof, it was 50 h). Heating demand increased, on average, in apartments by 9% compared with the variant with only a sun protection window.

In apartments with a mechanical cooling system, replacing the window with a sun-proof version was a good solution. It caused a decrease in the demand for cooling by 27% on average in apartments, but also a decrease in the heating demand by 8%. In the case of combining mechanical cooling and a reflective roof, a decrease in cooling demand was achieved by 13% (with the main effect on the top floor), but the demand for heat increased by 7% on average in apartments. However, this is a much cheaper solution.

3.2.7. Future Climate Conditions (Cases 14 to 19)

In the projected warmer climate, the heating demand in the apartments of the base building drastically decreased in some apartments by as much as 6 times (Figure 13). The average heat demand index was very low and did not exceed 1 kWh/m². It can be concluded that, in the future, in well-insulated buildings, the heating cost will be negligible. Unfortunately, increasing outdoor temperature will deepen the problem of thermal discomfort for residents. For example, for apartment A11, the number of thermal discomfort hours increased by as much as 780 h per year and, for apartment A12, by 900 h/year, i.e., an eight- and four-fold increase, respectively. The worst conditions (similar to the standard climate) were recorded in apartments A2 and A12. The unfavorable conditions lasted there for more than 13% of the year. Considering that this was mainly the summer period from May to September, this was more than 30% of this period.

The use of shading in the form of balconies provided a similar insignificant effect on the heat demand, as in the case of the standard climate. However, comparing the effect of the balcony in the standard and the warm climate on the thermal conditions, the results showed a decrease in the effect with the increase in the average global temperature. The improvement in thermal conditions was more than half as small in the climate of 2050.

For the warm climate, a greater effect of using green and reflective roofs was observed to reduce heat demand. For the upper floor, using a green roof resulted in a 28% decrease in heat demand. Similar to the standard climate, the smallest effect of green and reflective roofs on reduced heat demand was observed for apartments A2 and A12. The effect of using a green roof and a reflective roof on reducing the number of thermal discomfort hours was twice that of 2050.

The use of new windows with solar protective glazing reduced the heat demand for the apartments. The heat demand decreased by 20% on average, which was very similar to the standard climate. The smallest decrease in heating demand was achieved for apartments A2 and A12, located in the middle of the segment. The use of new windows also increased the thermal conditions in the warmer climate. However, this effect was much smaller than in the standard climate and improved thermal conditions by 53% on average. The results on the individual trends in the number of thermal discomfort hours were the same as for the standard climate, and the least significant for middle apartments with windows facing the southwest.

The cooling demand for the climate of 2050 was twice as high as in the standard climate, and the increase was very even across all apartments. The values of heat demand in the warmer climate were low and accounted for 2% to 15% of the cooling demand for individual rooms. On average, heat demand represented 7% of cooling demand in the warmer climate. The presented analysis reflects the expected energy use of air conditioning in apartments required to ensure thermal comfort, without implementing additional solutions to reduce cooling demand.

4. Discussion

Thermal comfort was not assessed in the field measurements of this study (this analysis was carried out based on numerical calculations); however, measurements showed that overheating of dwellings in summer is a problem not only in warm climates—for example, as shown by Murtyasa et al. [71] in a terraced house in Malaysia—but even in temperate climates. Similar indoor temperature ranges (from 23 °C to 28 °C) were recorded in living spaces in the UK climate during the heatwave in June 2018 [1]. Even higher indoor temperatures (exceeding 30 °C) were reported in the same climate by Gupta et al. [28] in 2019 and by Zahiri and Gupta in 2022 [72]. Overheating was also observed during measurement campaigns in Spain [73] and Finland in summer of 2020 [74] and 2021 [75]. Overheating may be defined differently depending on the region, but there is a general agreement that exceeding 26–27 °C is problematic [31].

The study reaffirms previous findings [71,76] that highly insulated envelopes—while beneficial in winter—can trap heat in summer and worsen thermal conditions in buildings. Similar conclusions were drawn by Fallah et al. [77], who emphasized that insulation retrofitting, especially in hot climates, can increase cooling energy demand. D’Agostino et al. [78] also advised caution with excessive insulation in buildings with high internal heat loads. A systematic review by Hu et al. [79] showed that passive techniques can reduce indoor temperature by 2 °C on average, cut cooling loads by 30%, and increase thermal comfort hours by 23% in warm climates. The results of this study confirmed that such effects can also be relevant in temperate climates.

In the context of thermal retrofitting, improvements should target the weakest thermal elements, e.g., walls, windows, and especially roofs, which accumulate heat due to solar exposure [20]. Green roofs and solar protective windows significantly reduce thermal loads. However, their applicability may be constrained by structural limitations or winter performance trade-offs, e.g., reflective roofs increase heating demand due to reduced solar gains in winter compared with conventional or green roofs [20].

In the projected 2050 warmer climate, the relative effectiveness of individual retrofitting measures remains similar but the magnitude of their impact changes. The average energy savings from selected passive measures (green roofs and solar windows) increases from 16% in standard climate to 18% in warmer conditions. As ambient temperatures rise, annual heating demand drops—by up to 72% in some simulated scenarios—thus extending the payback periods of insulation-focused retrofits.

Importantly, warmer climates significantly increase thermal discomfort. Simulations predict a four-fold rise in discomfort hours in multifamily apartments. In extreme cases, some studies predict that acceptable thermal conditions may be met less than 5% of the time annually [71]. Among evaluated passive strategies, solar protective glazing has the strongest effect, followed by balconies and canopies. However, their effectiveness varies by apartment layout and location. For top-floor apartments, green or cool roofs can reduce discomfort hours by over 50%, consistent with findings from Spain [20] and sub-Saharan Africa [27].

Given the time horizon of this study and the declining relevance of historical climate baselines, future-oriented scenarios were adopted. Notably, 2023 was the hottest year on record [80], underlining the urgency of adaptation strategies, especially for durable prefabricated concrete buildings that still have long operational lifespans [81].

However, some limitations must be acknowledged. First, climate variability and uncertainty present inherent challenges in long-term simulations. Although representative climate data for 2050 were used, such projections are based on specific scenarios and assumptions. Real-world climate evolution—especially the frequency, intensity, and duration of heatwaves—may differ significantly from modeled scenarios. Therefore, future studies should consider multiple representative concentration pathways (RCPs) to capture a range of possible outcomes. Second, occupant behavior modeling, particularly window operation, remains a major source of uncertainty. In this study, a stochastic behavioral model was employed to reflect probable user interaction with windows. Nevertheless, behavioral responses can vary greatly across demographics, cultures, and daily routines. While the use of probabilistic patterns improves realism compared with deterministic schedules, the complexity of human decision making cannot be fully captured by simulations. Additionally, the calibration of the model relies on the available literature data, which may not fully reflect local behavioral patterns. Future research should therefore include field studies to validate and refine behavioral assumptions. Moreover, no indoor comfort survey was conducted in parallel with measurements, which limited the direct linkage between subjective comfort and simulated metrics. Incorporating qualitative occupant feedback in future studies would enrich the interpretability of results. Finally, while this study presents a comparative analysis of passive and active strategies, economic aspects (e.g., cost–benefit analysis) and real-world implementation feasibility were not covered in depth. From an economic perspective, the feasibility of implementing passive and hybrid strategies can vary significantly across solutions. Mechanical cooling systems, especially when paired with decentralized mechanical ventilation, tend to incur the highest installation costs. While they ensure thermal comfort, their long-term energy use and maintenance requirements raise concerns about affordability and sustainability. Similarly, the addition of balconies, though architecturally beneficial, typically involves higher construction costs and offers only moderate improvements in indoor thermal conditions. By contrast, solar protective glazing can provide the greatest reduction in discomfort hours at a similar investment cost to balconies, suggesting a more favorable cost-to-effectiveness ratio. The most economically advantageous strategies may be cool or green roofs, which are up to four times cheaper than solar protective windows. However, their effectiveness diminishes on lower floors, limiting their overall impact in multi-story buildings. These considerations suggest that passive strategies should be prioritized not only for their environmental benefits but also for their potential economic viability, especially in large-scale retrofitting programs targeting aging multifamily housing stock.

Despite these limitations, the findings contribute to a better understanding of the potential and constraints of passive retrofitting strategies under changing climate conditions. Given the risks associated with increasing air-conditioning reliance [82,83], prioritizing passive and hybrid solutions [28,71] remains essential for sustainable adaptation in the building sector.

5. Conclusions

The passive methods, suitable for the wide applications presented in this study, seem to be able to significantly reduce the demand for mechanical cooling on a large scale of retrofitted multifamily buildings in the context of climate change.

The main conclusions and recommendations are as follows:

• Insulation of the walls and roof significantly reduces heat demand, for example, a decrease of 42–75% depending on the apartment. Unfortunately, at the same time, thermal conditions worsen in summer due to the increase in heat accumulation in the partitions of the building.
• Internal blinds reduce the thermal discomfort number hours by only 11–19%, and their effect on reducing heat demand is negligible.
• The balconies reduce the overheating of the apartments in the summer, reducing the number of discomfort hours by 27–42%; they slightly increase the demand for heat in the winter (by 1 to 2%).
• A green roof reduces heat demand by 20% on the top floor and improves thermal conditions.
• A reflective roof effectively reduces the number of thermal discomfort hours in apartments on the top floor but increases the heat demand in the winter.
• Modern windows with solar protective glazing reduce the period of human thermal discomfort by an average of 90% and reduce the heat demand by 18%.
• Mechanical cooling effectively eliminates thermal discomfort but significantly increases energy demand, especially for mechanical ventilation.
• In future climate conditions (in 2050), the heat demand will decrease by 72%, but the number of thermal discomfort hours will increase by up to four times.
• Solutions such as balconies, green roofs, or modern windows lose some of their efficiency in warmer climates.
• The most effective combination is solar protective glazing and reflective roof, which reduces the number of thermal discomfort hours by 95%.
• Mechanical cooling should be avoided as a primary solution; instead, more passive measures are recommended to improve thermal comfort.

This study covered only one building; however, the building selected is a typical multifamily residential structure. Such buildings were constructed on a large scale between the 1960s and 1980s to address housing shortages caused by urbanization and population growth in Central and Eastern European countries, including Poland. Apartment blocks from this period were characterized by standardized design, construction technology, and functional layout. As a result, they serve as typical and easily comparable research subjects for energy retrofit analyses. A significant portion of the population in Poland and the region still resides in these dwellings. Therefore, energy retrofitting of this type of housing can bring substantial benefits to a large number of residents. In summary, the multifamily building from the 1960s is not only a symbol of the era of urban development but also a typical example of construction that requires adaptation to contemporary climatic and energy challenges. The results presented have practical implications for energy auditors, designers, urban planners, and decision makers, supporting the development of energy-efficient strategies for improving thermal conditions in aging residential buildings. These recommendations can be summarized as follows:

• Revise building regulations and energy retrofit programs to address both winter heat retention and summer overheating risks, especially in the context of a warming climate.
• Promote passive and hybrid solutions—such as solar protective glazing, green roofs, external shading, and reflective surfaces—as preferable alternatives to mechanical cooling, which increases energy demand and peak loads.
• Include overheating risk and indoor human thermal comfort as standard criteria in energy audits and public funding schemes for building retrofits.
• Prioritize retrofit support for post-war multifamily apartment blocks, which represent a large portion of the housing stock in Central and Eastern Europe and are particularly vulnerable to thermal discomfort.
• Encourage the use of dynamic simulations in design and evaluation processes to better reflect future climatic scenarios and adaptive responses.

Future Research

In this study, only the part of the building that includes the apartments most exposed to the impact of external conditions was analyzed. To expand this research, an analysis of the heat demand of the entire building is planned, which allows one to determine costs and the payback time of the investment. When a specific heat source is taken into account, it will also be possible to determine the positive or negative impact of the proposed solutions on the external environment. Additionally, future work may include a subjective assessment of occupants’ thermal comfort and behavioral patterns through questionnaires. Such surveys would require the engagement of a large number of residents to provide reliable and representative results, and could serve as a valuable complement to the simulation-based findings presented in this study.