Enhancing thermal comfort assessment: A sensitivity study of PMV-PPD and adaptive models in an Algerian reference hotel across different climate zones

Abstract

This study investigated the impact of design parameters on indoor thermal comfort in hotel buildings, using both PMV-PPD and adaptive models. The aim is to provide guidance for improvement in the energy performance and thermal comfort of the accommodation structures, providing analysis extended to the eight Algerian climate zones. The adopted methodology was applied to a reference building hotel in Ghardaïa, Algeria. The energy model was validated through in-situ measurements and then the analysis was extended through EnergyPlus dynamic simulations to assess the influence of 13 design parameters using sensitivity analysis. Results reveal that design temperatures on the hottest days ranged from 34.4°C to 39.3°C, despite external temperatures reaching 45°C. The PMV-PPD model was most affected by space density, ventilation rate, infiltration, ventilation schedules, roof design and azimuth. Conversely, the adaptive model was influenced primarily by ventilation rate, window-to-wall ratio, lighting, space density, ventilation schedules, glazing and azimuth. This study underscores the potential for optimising hotels’ design to enhance thermal comfort and has broader relevance for regions with similar climates to Algeria. The findings offer insights for creating more comfortable indoor environments and fostering sustainable building practices in challenging climates.

Keywords

Thermal comfort predicted mean vote and predicted percentage dissatisfied model adaptive model sensitivity analysis dynamic simulation climate zones of Algeria

Introduction

Within the European Union, the building sector accounts for 40%–45% of total energy consumption.^1–3 The International Energy Agency (IEA) reports that buildings, including those related to construction, consume over 30% of the world’s total energy.⁴ In Algeria, the building sector, comprising both residential and tertiary structures, is responsible for 43% of the country’s total energy consumption. Of this, 34% is allocated to the residential sector, while the tertiary sector consumes 9%, with most of it directed towards space heating and cooling. Hotel buildings account for 3.5% of the energy consumption in this sector.⁵ The assessment of design parameters related to thermal behaviour is a crucial aspect of developing new buildings and retrofitting existing ones.

A fundamental aspect in the evaluation of indoor comfort in building design pertains to whether the spaces are mechanically conditioned (MC) or naturally ventilated (NV). ASHRAE Standard 55-2023⁶ outlines two primary approaches for comfort analysis: the Predicted Mean Vote and Predicted Percentage Dissatisfied (PMV-PPD) model and the adaptive model. The PMV-PPD model considers six parameters, including metabolic rate, clothing insulation, air temperature, radiant temperature, air speed and humidity. Conversely, the adaptive model relies on two parameters: operative temperature and prevailing mean outdoor air temperature. The choice between these models is closely linked to the type of building, with the adaptive model being suitable for NV and not mechanically conditioned spaces and the PMV-PPD model for MC spaces. Globally, the significance of addressing the thermal performance of both MC and NV hotel buildings is undeniable.

Software-based dynamic simulation tools have witnessed rapid advancements in modelling future scenarios, climate changes, technological developments and urbanisation. Therefore, a comprehensive understanding of model parameter variations, along with their sensitivity and uncertainty, is essential for establishing benchmarks and performance boundaries in building design. In addition to assessing annual thermal comfort hours and energy demand, these tools are employed for life cycle assessments and cost projections. Sensitivity and uncertainty analysis techniques are crucial for identifying influential parameters, reducing the number of iterations and streamlining the design process while maintaining accuracy.⁷

Numerous studies have emphasised the significance of designing buildings that are well-adapted to local climates.^8–10 Prior studies have explored thermal comfort in hotels in diverse climates, such as India, utilising adaptive comfort models and in-situ measurements.¹¹ Another study employed dynamic simulations to assess the PMV-PPD comfort model for an Italian reference hotel.¹²

In Algeria, the expanding number of hotels reflects the government’s focus on tourism. Nevertheless, the complexity of investigating hotel buildings lies in their diversity in terms of size, location and type. While previous studies have examined various building models using parameters such as insulation, window-to-wall ratio (WWR), ventilation and airtightness, there remains a gap in the measurement of thermal behaviour and energy performance specifically in Algerian hotel buildings. Both NV and MC types require in-depth exploration.^13,14

Hence, this paper seeks to evaluate the sensitivity of thermal performance in a large hotel building by utilising local sensitivity analysis (LSA) with the Sensitivity Index (SI). LSA was chosen in this study over global sensitivity analysis (GSA) due to its reduced computational demands and shorter processing times. The analysis explored various parameters, encompassing building orientation, envelope, equipment, occupancy and natural ventilation schemes. The study particularly investigated how different climate types in Algeria influence the sensitivity of various output results and parameters. Furthermore, the study delved into the differences in the ability of NV and MC buildings to adapt to these climatic variations. The results offer architects and engineers’ insight into beneficial or optimal solutions for hotels in retrofitting or developing new buildings with similar geometric and thermal characteristics. A comprehensive investigation into each design strategy can serve as a valuable tool for advancing energy efficiency and cost-effectiveness, forming the foundation for energy policies and benchmarking, benefiting both the public and private sectors.¹⁵

The paper significantly advances the understanding of building thermal comfort and energy efficiency, particularly in the context of hotel buildings in Algeria. The comprehensive approach adopted in this study, employing both PMV-PPD and adaptive models, would allow for the systematic evaluation of the impact of various design parameters on indoor thermal comfort. By conducting sensitivity analysis across diverse climatic conditions, the study generates novel insights into factors influencing discomfort hours, thereby providing valuable guidance for architects, engineers and policymakers involved in building design and energy efficiency initiatives. Furthermore, the focus on the Algerian context adds novelty to the research landscape, offering empirical data and recommendations that can inform the design and retrofitting of hotel buildings not only in Algeria but also in regions with similar climates globally. Overall, the paper’s methodology, empirical findings and practical recommendations contribute significantly to advancing knowledge in the field of building thermal comfort assessment and sustainable design practices.

The study aims to achieve the following objectives:

1. To investigate the impact of design parameters on indoor thermal comfort in hotel buildings, using both PMV-PPD and adaptive models.

2. To apply the adopted methodology to evaluate the indoor thermal environment control to a reference building hotel. The assessments were conducted across the eight climate zones of Algeria using in-situ measurements and dynamic simulations.

3. To assess discomfort hours by exploring parametric variations of 13 design parameters for both the PMV-PPD and adaptive models across the eight climate zones of Algeria.

4. To identify optimal design parameters for both NV and MC spaces through sensitivity analysis (SA).

Methods

The methodology of this study involved several key steps as depicted in Figure 1 providing a visual aid to facilitate an understanding of the methodology.

Figure 1.

Framework of the study.

The starting point was to try to answer the question: How to enhance the thermal performance and thermal indoor comfort in hotels located in hot-dry and temperate climates through passive strategies? Consequently, to provide a foundational understanding of the research, the adopted approach commenced with a comprehensive literature review, delving into the subject of sensitivity analysis and prior parametric variation studies finalised to thermal comfort and performance in buildings and in particular in hotels. Furthermore, the analysis of the topography and climate characteristics of the study area was conducted to contextualise the subsequent investigation inherent in the thermal behaviour of the hotel buildings. The core objective of the research was to scrutinise the sensitivity of 13 distinct design parameters used in hotels across the eight climate zones of Algeria, evaluating indoor thermal comfort through both PMV-PPD and adaptive models. After the evaluation of the hotel building stock in Algeria, a reference building was selected to apply the methodological framework that comprised field measurements and dynamic simulations, which were subsequently analysed to glean valuable insights. Field measurements were executed to evaluate the indoor thermal environment during the peak of the summer season (specifically, the end of July and the beginning of August). Following this, dynamic simulations were performed, calibrated to the reference case, to delve into the thermal behaviour of the examined scenario.

A parametric analysis was employed, comparing various design scenarios encompassing azimuth, equipment power, external wall material, glazing, infiltration, internal wall material, lighting, roof material, shading, space density, ventilation schedule, ventilation rate and WWR. Each of these variables was examined about its thermal performance concerning discomfort hours over the year. In this way the study was geared towards a comprehensive examination of design parameters across the eight climate zones of Algeria, to assess the feasibility of implementing these parameters and, when appropriate, to identify potential enhancements in design.

However, the study demonstrates originality in both approach and subject matter. Beginning with a thorough literature review on sensitivity analysis and prior studies related to thermal comfort in hotel buildings, the research integrated topography and climate analysis of the study area, providing contextual depth. The core objective of scrutinising 13 design parameters across Algeria’s climate zones using PMV-PPD and adaptive models was to fill a significant gap in the literature. Employing field measurements and dynamic simulations calibrated to a reference case, the study provided empirical data and insights into indoor thermal environments. The parametric analysis, covering design scenarios from azimuth to ventilation rate, would contribute to the understanding of thermal performance and provided practical design optimisations tailored to diverse climates. Overall, the methodology’s originality lies in its holistic approach, sensitivity analysis and contribution to advancing knowledge in building thermal comfort assessment and passive design strategies, particularly in Algerian hotel buildings.

Literature review

‘Thermal comfort in hotel buildings’: A literature review

Few studies in literature deal with thermal comfort in hotel buildings, investigating the theme from different points of view.

Abdulaali et al.¹⁶ acknowledged the positive contribution of the tourism and hotel industry to a country’s socioeconomic development, reviewing five parameters of indoor environmental quality (air quality, thermal comfort, lighting, visual and acoustic) and their interplay with human comfort.

Wafi, Ismail and Ahmed¹⁷ conducted a comprehensive investigation involving over 900 respondents in a hostel in Malaysia. A questionnaire survey was complemented by the monitoring and daily measurement of climatic factors, which provided insights into occupants’ indoor comfort awareness.

Zineddine, Belakehal and Bennadji¹⁸ evaluated the impact of orientation on the thermal environment in Fernand Pouillon’s architecture in Biskra. Using ECOTECT simulations, it examines how sun and wind orientation affect indoor temperatures, showing that Pouillon’s orientation choices effectively reduce temperature variations. Bohdanowicz and Martinac¹⁹ explored the impact of thermal comfort standards based on ASHRAE Standard 55/92 on energy consumption in the hotel industry. They also investigated the potential of the adaptive approach to reduce energy usage. They concluded that the adaptive approach to thermal comfort management has the potential to become a sensible option even in hotels.

Pinto et al.²⁰ discussed the potential impacts of climate change and heat waves on the tourism sector and hotel buildings’ infrastructure. They underlined the importance of air conditioning and the resulting increased energy demand to ensure guests’ thermal comfort. The study explored tourists’ perspectives on thermal comfort and air-conditioning systems and compared them with in-situ measurements during the research period.

Borowski, Zwolińska and Czerwiński²¹ conducted in-situ measurements in Polish hotel buildings, assessing the acceptability of indoor conditions using the PMV-PPD comfort model. They found that temperature and humidity were generally maintained at satisfactory levels, although CO₂ concentration occasionally led to discomfort.

Regarding the application of passive strategies, Naderian and Heidari²² demonstrated the significance of porches in enhancing thermal comfort in historical hotel buildings in Iran, combining measurements and simulations.

Moreover, Vujošević and Krstić-Furundžić²³ examined the influence of atrium-type hotel buildings in Belgrade on space heating and cooling energy demand. They adopted a simulation approach using the EnergyPlus engine to identify the optimal model for hotels with atriums.

Another study²⁴ explored energy savings and costs, merging external devices with a self-shading envelope for a multi-storey hotel in Saudi Arabia’s hot-humid climate. Utilising DesignBuilder for modelling, the strategy achieved a 20.5% annual energy reduction with a 2-year payback period. In contrast, an alternative option without shading provided only a 5% savings with an impractical 84-year payback. The study highlighted the cost-effective and impactful nature of the proposed shading strategy in energy-efficient design.

Finally, in the study by Acosta et al.,²⁵ the application of a predictive controller improved the thermal and energy performance in two city hotels. The study compared historical electrical consumption data with simulation results.

Sensitivity analysis and parametric simulation

Sensitivity analysis

SA is a widely used approach in the domains of building energy modelling and simulation. It aims to understand model behaviour and identify input parameters that significantly affect output variations. SA comprises two main categories: LSA and GSA. A screening-based method, which falls under GSA, is designed to determine the significance of parameters while minimising the output variance. This approach allows the evaluation of complex models with numerous input parameters at a lower computational cost and in less time. Other GSA methods, such as variance-based and regression-based methods, require more computational resources compared to screening-based methods.²⁶

LSA, which employs derivative-based methods and produces a single sensitivity index, is suitable for determining linear models but may fall short in the case of non-linear models. It does not provide information about parameter shape or quantify correlated parameters. It treats each parameter with uniformly distributed variation and calculates the mean while determining extreme variations. The Morris method, which utilises elementary effect measures, is partly global in its approach but still employs the one-at-a-time (OAT) variable technique. However, it cannot determine parameter correlations found in other GSA methods, such as Sobol and FAST.^26–28

Studies suggest that screening-based methods, like the Morris method, yield results similar to those of variance-based methods (Sobol method) when ranking parameters in building energy modelling and simulation. SA is valuable for identifying parameters that can be omitted in further optimisation studies. The variance-based method accommodates interaction effects and is suitable for continuous variables but may not be ideal for discrete variables.²⁹

Zeferina et al.³⁰ conducted GSA (Morris and Sobol) of 14 parameters related to cooling demand output applied to buildings across six global climates. This study revealed that the ventilation rate had the most significant influence.

Numerous studies have advocated for the incorporation of vernacular architecture design parameters. Hamed³¹ established a connection between architectural identity and energy efficiency, emphasising the remarkable energy conservation potential of architectural identity. Nguyen et al.³² explored vernacular architecture and climate-responsive design in hot-humid climates through several parameters, including envelope and ventilation.

SA is a practical tool for studying building performance, especially during the early stages when parameter modification is still feasible.³³

Sarrazin, Pianosi and Wagener³⁴ highlighted the importance of two critical decisions in GSA: sampling size and input parameter thresholds. The choice of the SA method is contingent on the model outputs. Morris is often considered a reliable method with low computational requirements and provides a global evaluation of both continuous and discrete variables.

Rodriguez et al.³⁵ found that obtaining reliable outputs from SA requires a large number of input parameters, and the uncertainty of inputs can significantly impact outputs. Detailed models’ robustness is often utilised to obtain precise SA results.

Yu et al.³⁶ applied sensitivity indices and LSA across four cities in China with hot summers and cold winters, identifying the most influential envelope parameters in cooling and heating season energy performance. The shading coefficient and WWR were influential during the cooling season, while the wall heat transfer coefficient and shape coefficient played significant roles during the heating season.

The present study adopted the LSA method due to the project’s scale and the numerous input parameters and evaluations across eight climate zones in Algeria for both NV and MC types.

Parametric simulation

Thermal comfort is often described either as a steady state or as an adaptive behaviour that adjusts to varying climates.^6,37,38 The literature review reveals that numerous studies have explored building parameters’ effects on indoor thermal environments through parametric variations, but there is not an extended literature focused on defining the most influential parameters for hotels, especially across multiple climate zones with a unified model.

The study³⁹ evaluated 63 retrofit combinations for a late nineteenth-century historical Victorian house, focusing on internal wall insulation, glazing upgrades and airtightness improvements. Assessing five indicators: energy reduction, cost, payback period, space volume and thermal comfort, the recommended retrofit of £144.71/m² achieved 51.8% energy reduction with an 18-year payback. EnergyPlus was used for dynamic thermal comfort evaluation.

A study⁴⁰ introduced a simulation-based model for optimising green building retrofits, addressing the challenge of balancing energy efficiency and thermal comfort. Using a response surface method, critical parameters were identified for alternative plans aligned with green building standards. Validated on a reference building in Wuhan, the optimal combination of parameters achieved 4% energy savings while improving comfort, providing valuable insights for efficient and comfortable green building development.

Alwetaishi⁴¹ conducted a comprehensive study on the WWR, particularly focusing on its parametric variations. In his investigation, he found that in hot regions, orientations towards the east and south were less favourable for thermal comfort. Specifically, he recommended maintaining a 10% WWR to optimise occupants’ thermal comfort.

Nguyen and Reiter⁴² employed parametric study and model calibration to improve thermal comfort in low-cost apartments utilising natural ventilation. They also evaluated the effectiveness of passive strategies in hot-humid climates.

Yao et al.⁴³ explored various parameters influencing thermal comfort and energy conservation through parametric variation simulations in severe hot and cold climates in China. Yang et al.⁴⁴ investigated design parameters affecting building-integrated photovoltaic and thermal double-skin facades, which impact indoor thermal comfort and energy efficiency across three different Australian climate zones using LSA and OAT analysis.

However, the majority of these studies focused primarily on design parameters within temperate or hot and humid climates. There is a noticeable gap in research regarding design choices specifically tailored for severe hot-dry climates, characterised by extended summer periods and harsh environmental conditions. The significance of revisiting design parameters’ thermal behaviour becomes apparent when considering such extreme conditions. This underscores the potential for integrating in-situ measurements, dynamic simulation and sensitivity analysis tools into sustainable building practices.

In this framework, the study aimed to bridge insights from the literature review with practical application by conducting a sensitivity-based analysis of PMV-PPD and adaptive models in an Algerian reference hotel across diverse climate zones. By evaluating the sensitivity of various design parameters, it sought to enhance understanding and optimise thermal comfort in hotel buildings, contributing to the broader body of research on this topic. Leveraging SA and parametric simulation methods, the study endeavoured to identify influential design parameters and optimise thermal comfort across diverse climatic conditions, thereby advancing the understanding of building performance optimisation in hospitality environments.

The work aligns with previous research in the literature by focusing on optimising thermal comfort in hotel buildings, particularly in response to the increasing significance of indoor environmental quality parameters highlighted,¹⁶ and the necessity to address occupants’ comfort awareness as emphasised by Wafi et al.¹⁷ Additionally, the findings complement the exploration of thermal comfort standards and energy consumption in the hotel industry,¹⁹ offering insights into potential strategies for reducing energy consumption while maintaining optimal thermal comfort levels. Furthermore, the investigation extended the understanding of indoor conditions assessed using the PMV-PPD comfort model, as demonstrated by Borowski et al.,²¹ by conducting SA across multiple climate zones. While previous studies^22,23 have explored specific passive strategies and their influence on thermal comfort and energy demand in hotel buildings, this research has broadened the scope by examining a comprehensive range of design parameters affecting thermal comfort in diverse climatic conditions. Through SA and parametric simulation, it provides significant understanding regarding effective design strategies applicable in various contexts, contributing to the broader goals of sustainability and occupant comfort in the hospitality sector.

Case study presentation and model validation

Location and climate

The Algerian landscape is characterised by three distinctive regions defined by their geographical location, natural conditions and cultural heritage⁴⁵: the Mediterranean coast region, the high plateaus and the southern desert region. These regions encompass Algeria’s 48 provinces, spanning latitudes between 18°N and 37°N and longitudes between 8°W and 12°E. The vast Algerian desert covers approximately 84% of the country’s total land area. Algeria is positioned north of the Mediterranean Sea, east of Tunisia and Libya, south of Mali and Niger and west of Morocco. The country’s climate transitions from temperate and humid along the northern coast to temperate and arid in the high plateaus, ultimately reaching a hot-dry climate in the Algerian desert, characterised by significant seasonal and daily temperature fluctuations. The reference case is situated in Ghardaïa province (No. 47), as depicted in Figure 2. Figure 3(a) and (b) illustrates the summer and winter climate zones in Algeria, as defined by the Algerian Technical Regulation Document (DTR C3-T).⁴⁶

Figure 2.

Geographic regions and provinces in Algeria. Source: Li et al.⁴⁵

Figure 3.

Climate zones in Algeria: summer (a) and winter (b) classifications. Source: DTR C3-T.⁴⁶

Ghardaïa

Ghardaïa is a city located within the M'Zab region, famous for its unique architectural and cultural heritage. The M'Zab region comprises five towns, including El-Atteuf, Bou-Noura, Ghardaïa, Beni-Isguen and Melika, all situated on a limestone plateau divided by the M'Zab Valley. Located approximately 600 km to the south of Algiers, these towns are characterised by their historical significance, cohesive communities, town planning and territorial organisation. Each town’s foundational structure is a fortified cluster of buildings staggered with each other, centred around a central mosque and minaret. Ghardaïa is surrounded by mountains, bounded by Berriane to the north, Bou-Noura to the east and south and Dhayet Bendhahoua to the west. Geographically, Ghardaïa is situated between 32.47° and 32.51° north latitude and 3.62° and 3.69° east longitude, with an altitude of 494 m. The geological formation in the region is predominantly mountainous desert, with the Oued M'Zab serving as the primary valley.⁴⁷ Ghardaïa’s climate is classified as hot-dry, as per the climate zones of Algeria (DTR C3-T). It experiences a cold northwest monsoon from December to April and a hot southeast monsoon from June to October, often followed by sandstorms from the desert. Ghardaïa’s climate is characterised by hot summers (May to September) with temperatures ranging from 15°C to 45°C and cold winters (November to February) with temperatures between 0°C and 25°C. Rainfall is scarce and almost negligible throughout the year. The prolonged periods of high temperatures during the long summer months (5 months) contribute to thermal discomfort. Table 1 provides a summary of Ghardaïa’s monthly normal climate conditions, with temperature variations ranging from 0°C to 45°C and relative humidity (RH) fluctuating between 16% and 43% (see Figure 4). Wind velocity varies between 3.4 m/s and 5.1 m/s (see Figure 5).

Table 1.

Monthly normal climate data for Ghardaïa, Algeria. Source: Meteonorm V8.0.0.13161.⁴⁸

S. No.	Station	January	February	March	April	May	June	July	August	September	October	November	December
1	Average maximum temperature (°C)	16.5	17.5	21.9	27.2	30.1	39.6	42.2	41	36.2	29.1	21.3	19.7
2	Average minimum temperature (°C)	4.6	5.2	9.4	12.4	17.4	24.4	27.7	27.5	22.8	16.1	10	7.6
3	Average mean temperature (°C)	10.4	11.4	15.5	20.2	24	31.7	34.8	34.5	29.7	22.4	15.7	13.6
4	Average relative humidity (%)	43	38	37	30	27	16	17	21	29	36	38	43
5	Average wind velocity (m/s)	4.9	5.1	4.8	4.6	4.7	5	4.2	3.5	3.9	3.4	4.8	4.6
6	Average sunshine hour per day (h)	7.4	8	8.8	9.1	9.8	10.3	10.4	9.3	8.6	8.2	7.4	7.3
7	Total rain fall (mm)	1	6	7	2	1	0	1	4	3	3	3	2

Figure 4.

Annual temperature and RH profiles in Ghardaïa, Algeria. Source: Meteonorm V8.0.0.13161.⁴⁸

Figure 5.

Annual wind velocity and sunshine hours per day in Ghardaïa, Algeria. Source: Meteonorm V8.0.0.13161.⁴⁸

M’Zab Hotel, by Fernand Pouillon

Pouillon’s career, took a positive turn after Algeria gained independence in 1962.⁴⁹ The Algerian government extended an invitation to him, tasking him with the design and supervision of multiple hotels and tourist infrastructure projects throughout the country. Thus, it is possible to choose one of them as a representative of the hotels’ building stock in the country. The study specifically focused on the M’Zab Hotel as a representative example of hotel construction due to its significance and location in the south of Algeria, where a harsh hot-dry climate poses the most challenging conditions. Originally known as ‘Rosthémide’, the M’Zab Hotel, designed by Pouillon, was constructed between 1967 and 1971. With a capacity of 600 bed-places, the hotel was built on the site of an ancient fort ruin, preserving the powerful and enclosed architectural character. The linear structure is harmoniously integrated into the hilly terrain near Ghardaïa, artfully blurring the lines between new and old elements. The seamless fusion of the hotel with its surroundings ensures that visitors experience a sense of protection and balance as they transition from enclosed spaces to expansive vistas of the vast landscape. The hotel exudes an atmosphere reminiscent of a fortified palace rather than a modern hotel. Pouillon’s unique architectural style, characterised by classic and multicultural influences, exemplifies his ability to adapt to the climate, program, physical context and cultural setting, reflecting a regionalist approach with a modern sensibility. During the study, the hotel underwent renovation work, including the restoration of demolished areas, the repainting of walls and various improvements to lighting, air-conditioning and heating systems. The hotel was unoccupied during this period, devoid of users and equipment,^50,51 as shown in Figure 6(a)–(d) and Figure 7(a)–(g).

Figure 6.

External views of the hotel during the renovation period (a, b, c and d).

Figure 7.

Internal views of the hotel during the renovation period (a, b, c, d, e, f and g).

The specific conditions of the chosen Algerian reference hotel, M'Zab Hotel, were selected to fill a research gap in the understanding of thermal comfort assessment in hotel buildings across diverse climate zones. Situated in Ghardaïa, Algeria, the hotel exemplifies the challenges posed by the region’s hot-dry climate. This extreme climate presents an ideal case study for evaluating thermal comfort strategies. Additionally, the M’Zab Hotel’s architectural significance and integration with the local landscape provide valuable insights into how building design can impact occupant comfort in harsh environmental conditions. Through a sensitivity-based study across eight climate zones, the research aimed to identify optimal thermal comfort solutions tailored to both MC and NV hotel buildings, thereby addressing a critical gap in the literature on hospitality building performance optimisation.

Measurements and dynamic simulations

Dynamic simulation

EnergyPlus/DesignBuilder was employed as a simulation tool for evaluating energy requirements, CO₂ emissions and occupant comfort within a building.⁵² This software encompasses a 3D modelling component and modules for thermal/energy analysis, daylight levels and airflow assessments.

Simulation process

The simulation process encompassed the following key steps:

Climate data

Meteorological data for Ghardaïa were employed, providing hourly data on parameters such as temperature, humidity, solar radiation, cloud cover, wind speed and wind direction. For the analysis of discomfort hours, a reference year, derived as the average of several years (2004–2018), was used for simulations. A (.epw) file containing detailed weather information for all 8760 h of the year was imported.⁵³

Model of the case study

Given the hotel’s considerable size, a specific section, the eastern wing (see Figure 8), was chosen for modelling. This wing consists of three rows of rooms (northern and southern, each with three storeys, and western, with one storey), surrounding a patio. The eastern wall features arches. Each room consists of a bedroom, bathroom, closet and an attic balcony opening onto the patio. Windows are situated in the second loading wall, which faces north. The southern facade is consistently protected by corridor spaces. Each room’s area (bedroom + bathroom) varies from 30 to 70 m². Each unit was modelled with multiple zones (11 zones in total on the first level, 8 on the second, 5 on the third, 2 on the fourth and 1 on the fifth). Windows open at night if the indoor temperature exceeds 22°C, allowing for night flushing ventilation. Lighting was controlled automatically according to software requirements, as shown in Figure 9(a) and (b) and Figure 10(a)–(e).

Figure 8.

Hotel roof plan and the modelled eastern section.

Figure 9.

Case study model: studied units (a) and (b).

Figure 10.

Case study model: (a) measured floor (highlighted in red) and (b) created zones and floor plans (27 zones in total).

Thermal characteristics

Table 2 provides thermal characteristics, including the thermal conductivity, density and specific heat capacity of envelope materials.

Table 2.

Thermophysical properties of materials used in the study.

Material	Density (kg/m³)	Thermal capacity (J/kg.K)	Conductivity (W/m.K)	Thickness (m)	Transmittance (W/m²K)
First balcony wall composition
Lime-sand render	1100–1800	936–1080	0.5–0.87	0.03	2.426
Hollow concrete block	1300	1080	1.1	0.15
Plaster	750–1000	936	0.35	0.025
Second load-bearing wall composition
Plaster	750–1000	936	0.35	0.025	1.341
Hollow concrete block	1300	1080	1.1	0.15
Air gap (R = 0.16 m²K/W)				0.1
Hollow concrete block	1300	1080	1.1	0.15
Plaster	750-1000	936	0.35	0.025
Roof composition
Granite tile	2200	936	2.1	0.025	1.964
Cement layer	2200	1080	1.4	0.025
Hollow concrete blocks slab	1450	1080	1.45	0.2
Plaster	750-1000	936	0.35	0.025
Floor composition
Granite tile	2200	936	2.1	0.025	1.611
Cement layer	2200	1080	1.4	0.025
Floating slab	2200	936	1.3	0.15
Backfill	1560	1000	0.85	0.2
Single clear glazing				0.004	5.871

Heat exchange resistance of interior and exterior surfaces is 0.14 m²K/W in wall and 0.22 m²K/W in floor. Source: Algerian technical regulation document (DTR C3-T).⁴⁶

Standard internal gains from occupants were defined, and air changes per hour (ACH) were calculated, accounting for infiltration and air renewal based on the DTR C3-T equations for both summer and winter conditions.⁴⁶

ACHs for envelope’s infiltration were 0.1177 (equation (1)) and 0.055 (equations (2) and (3)) for summer and winter, respectively. The ACH for air renewal of natural ventilation (minimum air renewal exigence of indoor space) was 0.1 (equation (4)) and 0.6 (equations (5) and (6)) for summer and winter, respectively. These rates would influence thermal gain or loss of indoor air, by dividing (Q) over (V), whereas V is the building’s volume equal to 44120.5 m³. Equation (1) was used to determine ACH values:⁴⁶

• Infiltration (summer)

{q v}_{\inf} = \sum ({q v o}_{\inf, i} \times S_{o u v, i}) [\frac{m^{3}}{h}]

(1)

Whereas

{q v}_{\inf}

is the flow rate by wind infiltration (taking into account only the openings located in walls facing the winds),

{q v o}_{\inf, i}

is the opening

(i)

infiltration rate (for windows

({q v o}_{\inf, i} = 14.5

m³/h.m²) and for doors

({q v o}_{\inf, i} = 4.5

m³/h.m²),

S_{o u v, i}

is the opening

(i)

surface (m²).

• Infiltration rate of wind (winter) is defined by equation (2) and the air permeability of the wall ( $P_{p i})$ is defined by equation (3)

Q_{s} = \sum (P_{p i} \times e_{v i}) [\frac{m^{3}}{h}]

(2)

P_{p i} = \sum (P_{o j} \times A_{j}) [\frac{m^{3}}{h}] w h e n (Δ P = 1 P_{a})

(3)

Whereas

Q_{s}

is the flow rate by wind infiltration (considering all envelope’s openings),

P_{p i}

is the wall

(i)

air permeability (m³/h),

e_{v i}

is the wall

(i)

wind exposure coefficient

(e_{v i} = 2.71),

which was calculated based on the opening’s height from the ground and region’s environmental conditions (near the sea, mountains, rural environment, urban environment, etc.),

P_{o j}

is the opening

(j)

surface air permeability (for windows

(P_{o j} = 4)

and for doors

(P_{o j} = 6)

(m³/h.m²)) and

A_{j}

is the opening

(j)

surface (m²).

• Air renewal (summer) is defined by equation (4)

Q_{v} = \sum (Q_{v m i n} \times n_{p}) [\frac{m^{3}}{h}]

(4)

Whereas

Q_{v}

is the specific ventilation rate,

Q_{v m i n}

is the minimum extract flow rate (for rooms of less than three occupants

(Q_{v m i n} = 30)

and for bathrooms,

(Q_{v m i n} = 15)

(m³/h)) and

n_{p}

is the number of pieces.

• Air renewal (winter) is defined by equations (5) and (6)

Q_{v} = Max [0.6 \times V_{h}, Q_{v r é f}] [\frac{m^{3}}{h}]

(5)

Q_{v r é f} = \frac{5 Q_{v m i n} + Q_{v m a x}}{6} [\frac{m^{3}}{h}]

(6)

Whereas

Q_{v}

is the specific ventilation rate,

V_{h}

is the volume of living spaces (m³),

Q_{v r é f}

is the reference extract flow,

Q_{v m i n}

is the minimum extract flow

(Q_{v m i n} = 1040)

(m³/h) and

Q_{v m a x}

is the maximum extract flow

(Q_{v m a x} = 210)

(m³/h).

Model’s validation

Field measurements

Field measurements were conducted using a Testo 480 instrument.⁵⁴ This instrument was employed to gather detailed information about the indoor thermal environment, facilitating the calibration of the simulation model. Subsequently, the validated model would be capable of predicting discomfort hours over extended periods and various climatic conditions.

The measurements included air temperature, RH, air velocity and wall surface temperature. The instruments used for these tests met the ISO standard 7726-2002 requirements.⁵⁵ Detailed information on instruments is shown in Table 3.

Table 3.

Detailed information of instruments.

Description	Parameter measured	Range	Accuracy
Digital temperature and humidity recorder (Ø 12 mm)	Air temperature	−20 to +70°C	±0.2°C (+15 to +30°C)
	Air temperature	−20 to +70°C	±0.5°C (remaining range)
	Relative humidity	0 to 100% RH	±(1.0% RH + 0.7 % of mv) (0 to 90% RH)
	Relative humidity	0 to 100% RH	±(1.4% RH + 0.7 % of mv) (90 to 100% RH)
Surface probe (Ø 4 mm)	Surface temperature of wall and roof	−60 to +1000°C	±1.5°C (−40 to +375°C)
Surface probe (Ø 4 mm)	Surface temperature of wall and roof	−60 to +1000°C	±0.4% of reading (+375 to +1000°C)
Vane/temperature probe (Ø 16 mm)	Air velocity	+0.6 to +50 m/s	±(0.2 m/s + 1 % of mv) (0.6 to +40 m/s)
Vane/temperature probe (Ø 16 mm)	Air velocity	+0.6 to +50 m/s	±(0.2 m/s + 2 % of mv) (40.1 to +50 m/s)

*mv = measurement value. Source: Testo® official website.⁵⁴

A specific indoor location within the hotel’s rooms was selected for these in-situ measurements. The indoor thermal environment was measured at a central point, positioned at a height of 1.7 m, which corresponds to the standing position of individuals, as recommended by ISO 7726⁵⁵ and ASHRAE 55.⁶ Outdoor environmental parameters were obtained from a location shielded from direct solar radiation and rain.

Field measurements of air temperature, RH, air velocity and wall surface temperature were conducted with an hourly timestep from 8:00 a.m. continuing throughout the day. The selection of the timestep was made to ensure a balance between capturing detailed data and practical feasibility. The measurement period was chosen to correspond with the hottest period of summer, encompassing three consecutive days in early August (from 06-08-2019 to 08-08-2019). These selected days were representative of the broader summer period, characterised by clear sky and temperatures typical of hot summer weather. The measurements were carried out under natural conditions, with windows open and no active air conditioning in use. The measurement site was situated on the ground floor, oriented towards the northeast and featured a window overlooking one of the hotel’s two large patios (see Figure 11(a) and (b)).

Figure 11.

Case study and instrument: (a) hotel room layout and window design facing the balcony and patio – measurement location and (b) used instrument (Testo 480).⁵⁴

Model’s validation

To validate the model and ensure its capability to describe thermal behaviour over extended periods, comparisons between simulated and measured values were performed. The validation process considered two key metrics: normalised mean bias error (NMBE) (equation (7)) and the coefficient of variation of the root mean square error (CVRMSE) (equation (8)). The model was considered calibrated when NMBE was less than ±10% and CVRMSE was less than 30%.⁵⁶

N M B E = \frac{\sum_{i = 1}^{n} (y_{i} - {\hat{y}}_{i})}{(n - p) \cdot \bar{y}} \cdot 100

(7)

C V R M S E = \frac{\sqrt{\frac{\sum_{i = 1}^{n} {(y_{i} - {\hat{y}}_{i})}^{2}}{(n - p)}}}{\bar{y}} \cdot 100

(8)

where

y_{i}

is the actual measurement for period

i

{\hat{y}}_{i}

is the simulated value for period

i

\bar{y}

is the average measurement and

n

is the number of measured periods and

p

= 1.⁵⁶

The validation results for operative temperature and RH indicated that the model’s predictions closely matched the actual measurements, with NMBE values within acceptable limits. Figure 12(a) and (b) illustrates these validation results.

Figure 12.

Model validation results for operative temperature and RH.

Thermal behaviour (free running case)

After model validation, thermal behaviour was analysed throughout the entire year. The results for the hottest month in summer (30 days from mid-July to mid-August) are presented in Figure 13(a) and (b). These results depict variations in daily temperature degrees, with mean temperatures fluctuating between 33.6°C and 42.2°C, and occasional external temperatures reaching 45.4°C. The findings suggest that ground floor spaces, especially bedrooms, exhibit better adaptability to external heat conditions compared to other floors, due to their direct exposure to the external environment.

Figure 13.

Typical hottest month temperature extremes in Ghardaïa.

The parametric variations

The study aimed to compare the impact of various climate-responsive design strategies. The data were derived from the ASHRAE 55 standard values (metabolic rate is 1.00 for men, 0.85 for women, 0.75 for children, or an average value if there is a mix of sizes). For clothing insulation, a 0.5 Clo level was used in summer and a 1.0 Clo level was used in winter. Thirteen parameters were evaluated, and multiple scenarios were analysed using the LSA with OAT technique. These parameters included azimuth orientation, equipment power density, external wall types, glazing types, infiltration rates, internal wall types, general lighting power density, roof types, shading devices, space density, natural ventilation schemes (schedule and ACH) and WWR. The selection of these parameters was based on their known influence on building thermal behaviour, and the ranges of their variations were determined using existing literature and preliminary studies (Table 4). The baseline scenario, reflecting the reference case’s actual conditions without any modifications, served as the reference for assessing the model’s behaviour across different climatic conditions.^37,57 Table 5 and Table 6 describe the various continuous and discrete parameters and their characteristics along with the number of cases indicated by the number of simulation runs or the instances of parametric variation corresponding to the respective parameter. Table 7 presents the thermophysical properties of different materials considered in the parametric variations (external walls, internal walls, glazing, roofs and insulation material).

Table 4.

The adopted design parameters from existing literature review.

N°	Parameter	References
x1	Azimuth	Nguyen,³⁷ Yao et al.⁴³
x2	Equipment power	Zeferina et al.,³⁰ Nguyen,³⁷ Yang et al.⁴⁴
x3	Ext. walls	Zeferina et al.,³⁰ Yu et al.,³⁶ Nguyen,³⁷ Li et al.,⁴⁰ Nguyen and Reiter,⁴² Yao et al.,⁴³ Yang et al.⁴⁴
x4	Glazing	Zeferina et al.,³⁰ Heiselberg et al.,³³ Yu et al.,³⁶ Nguyen,³⁷ Qu et al.,³⁹ Nguyen and Reiter.⁴²
x5	Infiltration	Zeferina et al.,³⁰ Nguyen,³⁷ Qu et al.³⁹
x6	Int. walls	Nguyen,³⁷ Qu et al.,³⁹ Li et al.⁴⁰
x7	Lighting	Zeferina et al.,³⁰ Nguyen,³⁷ Yang et al.⁴⁴
x8	Roofs	Zeferina et al.,³⁰ Yu et al.,³⁶ Nguyen,³⁷ Li et al.,⁴⁰ Yang et al.⁴⁴
x9	Shading	Heiselberg et al.,³³ Yu et al.,³⁶ Nguyen,³⁷ Nguyen and Reiter,⁴² Yao et al.⁴³
x10	Space density	Zeferina et al.,³⁰ Nguyen,³⁷ Yang et al.⁴⁴
x11	Ventilation schedule	Nguyen.³⁷
x12	Ventilation rate	Zeferina et al.,³⁰ Nguyen,³⁷ Yao et al.⁴³
x13	WWR	Yu et al.,³⁶ Nguyen,³⁷ Li et al.,⁴⁰ Alwetaishi,⁴¹ Nguyen and Reiter,⁴² Yao et al.⁴³

Table 5.

Continuous design parameters and their characteristics.

N°	Parameter	Unit	Base case value	Min. value	Max. value	Number of cases
x1	Azimuth	Degrees (°)	0° (northern east-southern west)	0°	360°	8
x2	Equipment power	W/m²	5 W/m²	0 W/m²	20 W/m²	6
x5	Infiltration	ACH	∼0.09 ACH	0 ACH	1.4 ACH	15
x7	Lighting	W/m² per 100 lx	5 W/m²	0 W/m²	15 W/m²	6
x10	Space density	People/m²	0 people/m²	0 people/m²	0.6 people/m²	7
x12	Ventilation rate	ACH	0.6 ACH	0 ACH	12 ACH	13
x13	WWR	Percent (%)	∼60 %	10 %	90 %	5

Table 6.

Discrete design parameters and their design choices.

N°	Parameter	Design choices	Base case choice	Number of cases
x3	Ext. walls	Heterogeneous walls with different thicknesses, air gap or insulation	(P + W + A + W + P), W [15 cm], A [10 cm] (W) hollow concrete block – case study wall	55
		- One layer wall = P + W + P
		W [0.1 m, 0.3 m, 0.5 m]
		- One layer wall (with external insulation) = P + I + W+P
		W [0.2 m, 0.4 m], I [0.05 m, 0.1 m]
		- One layer wall (with internal insulation) = P + W + I+P
		W [0.2 m, 0.4 m], I [0.05 m, 0.1 m]
		- Double layers wall (with air gap or insulation) = P + W + A(I)+W + P
		W [0.1 m, 0.2 m], A(I) [0.05 m, 0.1 m]
		(P) Plaster^a
		(W) Wall’s main materials
		- Hollow brick wall
		- Adobe brick wall
		- Lime-sand stone wall
		- Hollow concrete block – case study wall
		- Hollow concrete block wall (I) insulation (expanded polystyrene)
		(A) Air gap
x4	Glazing	Three glazing choices	Simple glazing (single clear layer of 4 mm)	3
		- Simple glazing (single clear layer of 4 mm)
		- Double glazing (double clear layers of 3 mm for each and air gap of 6 mm)
		- Double glazing +krypton gas (double layers of SageGlass Climaplus 6 mm for each and Krypton gas of 12 mm)
x6	Int. walls	Heterogeneous walls with different thicknesses	(P + W + P), W [0.15 m] (W) hollow concrete block – case study wall	11
		- One layer wall = P + W + P
		W [0.1 m, 0.3 m] (P) Plaster^a
		(W) Wall’s main materials
		- Hollow brick wall
		- Adobe brick wall
		- Lime-sand stone wall
		- Hollow concrete block – case study wall
		-Hollow concrete block wall
x8	Roofs	Heterogeneous roofs with different thicknesses and insulation	(G + C + R + P), R [0.2 m] (R) Hollow concrete block slab – case study	36
		- One layer roof = G + C + R+P
		R [0.1 m, 0.2 m, 0.3 m]
		- One layer (with external insulation) = G + C + I+R + P
		R [0.1 m, 0.2 m, 0.3 cm], I [0.05 m, 0.1 m, 0.15 m]
		- One layer (with internal insulation) = G + C + R+I + P
		R [0.1 m, 0.2 m, 0.3 m], I [0.05 m, 0.1 m, 0.15 m] (G) Granite tile^b
		(C) Cement layer^c
		(R) Roof’s main materials
		- Hollow polystyrene block slab
		- Hollow brick block with earth material slab
		- Hollow brick block slab
		- Hollow concrete block slab – case study (P) Plaster^a
		(I) Insulation (expanded polystyrene)
x9	Shading	Shading devices with different sizes or no-shading	Different choices (actual design)	10
		- No shading [0 m]
		- Horizontal overhangs [0.5 m, 1 m, 1.5 m]
		- Vertical sidefins [0.5 m, 1 m, 1.5 m]
		- Overhangs + sidefins [0.5 m, 1 m, 1.5 m]
x11	Ventilation schedule	Five ventilation schemes	On 7/24	5
		- On 7/24
		- Off 7/24
		- On 7/24 summer and mild season
		- On day time summer and mild season
		- Off night time summer and mild season

^aPlaster thickness is fixed for all cases (0.025 m).

^bGranite tile thickness is fixed for all cases (0.025 m).

^cCement layer thickness is fixed for all cases (0.025 m).

Table 7.

Thermophysical properties of proposed materials used in parametric variations.

Material	Density (kg/m³)	Thermal capacity (J/kg.K)	Conductivity (W/m.K)	Thickness range (m)
(x3) Ext. walls (W) and (x6) int. walls (W)
Hollow brick	900	936	0.48	[0.1, 0.5] for ext. walls
Adobe brick	2567.6	1075.1	0.817
Lime-sand stone	1982	1910	0.93
Hollow concrete block – case study	1300	1080	1.1	[0.1, 0.3] for int. walls
Hollow concrete block	2300	1000	1.75	[0.1, 0.3] for int. walls
(x4) Glazing
Simple glazing (U = 5.871 W/m²K)				[0.004]
Double glazing (U = 3.591 W/m²K)				[0.003] for each layer
Double glazing + krypton gas (U = 1.267 W/m²K)				[0.006] for each layer
(x8) Roofs (R)
Hollow polystyrene block slab	830	1290	0.22	[0.1, 0.3]
Hollow brick block with earth material slab	1450	984	0.7
Hollow brick block slab	1150	984	1
Hollow concrete block slab – case study	1450	1080	1.45
Insulation material (I)
Expanded polystyrene	27	1404	0.038	[0.05, 0.1] for ext. walls
Expanded polystyrene	27	1404	0.038	[0.05, 0.15] for roofs

Heat exchange resistance of interior and exterior surfaces is 0.14 m²K/W in wall and 0.22 m²K/W in floor. Source: Algerian technical regulation document (DTR C3-T).⁴⁶

Results

Results are presented, focusing on the SA of various parameters in both the PMV-PPD and adaptive models. The discomfort hours data were based on whether the humidity ratio and operative temperature fall within the regions defined by ASHRAE Standard 55-2023 for the PMV-PPD model and whether the operative temperature is within the region specified in ASHRAE Standard 55-2023 for the adaptive model. The impact of 13 selected parameters on discomfort hours for both models was evaluated using the SI, calculated by using equation (9)³³:

S I = \frac{E_{\max} - E_{\min}}{E_{\max}} \times 100 %

(9)

where

E_{\max}

and

E_{\min}

are the maximum and minimum output values (i.e. discomfort hours for both PMV-PPD and adaptive models), respectively. SI expresses the percentage difference between the maximum and minimum values, representing the range of the output when a design parameter is explored across its entire variation.

Subsequently, the SIs were ranked, with the most influential parameter receiving the top rank (1) and the least influential on the lowest rank (13) based on SI percentages (Table 8).

Table 8.

Sensitivity indices (SIs) and parameter rankings for each climate zone.

Climate zone	N°	Parameter	PMV-PPD model		Adaptive model
Climate zone	N°	Parameter	Sensitivity index (SI)	Rank^a	Sensitivity index (SI)	Rank^a
Algiers/A	x1	Azimuth	2.4905	7	10.0714	5
	x2	Equipment power	3.3894	6	1.9441	12
	x3	Ext. walls	2.0500	9	3.8541	9
	x4	Glazing	2.1028	8	8.8256	7
	x5	Infiltration	21.2045	3	3.1390	11
	x6	Int. walls	0.8308	13	1.5172	13
	x7	Lighting	1.6335	10	14.0139	3
	x8	Roofs	4.8945	5	7.2504	8
	x9	Shading	1.5048	11	3.6211	10
	x10	Space density	49.3928	1	13.7815	4
	x11	Ventilation schedule	8.9407	4	8.8903	6
	x12	Ventilation rate	23.3781	2	34.8930	1
	x13	WWR	0.8713	12	17.7483	2
Saïda/B	x1	Azimuth	2.5025	6	10.4707	5
	x2	Equipment power	1.2214	10	3.3721	12
	x3	Ext. walls	1.9344	8	3.9751	11
	x4	Glazing	1.9801	7	6.2260	8
	x5	Infiltration	20.2558	3	5.2276	10
	x6	Int. walls	0.5192	12	1.4861	13
	x7	Lighting	0.9854	11	14.8262	3
	x8	Roofs	6.5279	5	5.5010	9
	x9	Shading	0.1815	13	8.8716	6
	x10	Space density	49.9103	1	13.6262	4
	x11	Ventilation schedule	10.8249	4	8.7374	7
	x12	Ventilation rate	29.9237	2	37.2996	1
	x13	WWR	1.4352	9	16.8750	2
Guelma/B1 and B2	x1	Azimuth	3.1565	6	9.0423	6
	x2	Equipment power	3.0106	7	1.0993	13
	x3	Ext. walls	1.4785	10	4.4254	10
	x4	Glazing	2.4568	8	7.5589	7
	x5	Infiltration	20.6255	3	3.7280	11
	x6	Int. walls	0.6762	13	1.3563	12
	x7	Lighting	1.5600	9	15.6578	3
	x8	Roofs	5.9097	5	5.4538	9
	x9	Shading	0.9340	12	6.6068	8
	x10	Space density	49.5875	1	14.3747	4
	x11	Ventilation schedule	11.6096	4	9.6829	5
	x12	Ventilation rate	30.6030	2	39.0075	1
	x13	WWR	0.9536	11	17.6470	2
Tiaret/C	x1	Azimuth	1.5474	8	7.2641	7
	x2	Equipment power	0.1796	13	2.7142	11
	x3	Ext. walls	1.9822	7	3.9865	10
	x4	Glazing	1.2320	10	7.0731	8
	x5	Infiltration	17.8588	3	1.4520	13
	x6	Int. walls	0.2565	12	2.5169	12
	x7	Lighting	1.3747	9	13.5900	3
	x8	Roofs	5.5669	5	7.5096	6
	x9	Shading	0.4331	11	6.4990	9
	x10	Space density	49.5630	1	11.9409	4
	x11	Ventilation schedule	10.0919	4	8.7614	5
	x12	Ventilation rate	26.1259	2	32.4311	1
	x13	WWR	2.0754	6	17.4443	2
Béchar/D	x1	Azimuth	5.5944	6	10.3867	6
	x2	Equipment power	2.2858	9	1.0262	12
	x3	Ext. walls	2.1030	10	4.1420	9
	x4	Glazing	2.0284	11	10.6870	5
	x5	Infiltration	16.4905	3	0.6069	13
	x6	Int. walls	0.4729	13	2.2539	11
	x7	Lighting	2.8341	7	15.9376	3
	x8	Roofs	5.7247	5	4.5710	8
	x9	Shading	0.7824	12	4.0404	10
	x10	Space density	49.7406	1	12.0056	4
	x11	Ventilation schedule	11.5595	4	7.8254	7
	x12	Ventilation rate	27.4420	2	30.7609	1
	x13	WWR	2.6462	8	19.0963	2
Ghardaïa (El-Goléa)/E	x1	Azimuth	4.4159	6	10.7758	6
	x2	Equipment power	1.3589	11	0.8333	13
	x3	Ext. walls	1.4380	10	3.5530	10
	x4	Glazing	1.7107	9	11.7283	5
	x5	Infiltration	15.8839	3	1.3650	12
	x6	Int. walls	0.5589	13	2.2943	11
	x7	Lighting	2.0580	8	17.4931	3
	x8	Roofs	5.1077	5	3.8128	9
	x9	Shading	1.3113	12	5.9887	8
	x10	Space density	50.0591	1	14.4492	4
	x11	Ventilation schedule	11.2433	4	7.9483	7
	x12	Ventilation rate	26.1534	2	31.1277	1
	x13	WWR	3.0655	7	20.3459	2
Illizi (Djanet)/E1	x1	Azimuth	5.2782	5	8.9514	6
	x2	Equipment power	0.3485	13	5.8443	10
	x3	Ext. walls	1.8746	9	3.6665	12
	x4	Glazing	1.8279	10	8.5050	7
	x5	Infiltration	14.8925	3	6.5261	9
	x6	Int. walls	0.5095	12	0.5566	13
	x7	Lighting	1.9286	8	16.3128	3
	x8	Roofs	3.0645	7	3.9060	11
	x9	Shading	0.6099	11	9.5765	5
	x10	Space density	49.6571	1	15.0435	4
	x11	Ventilation schedule	12.7772	4	7.6452	8
	x12	Ventilation rate	30.2555	2	30.4976	1
	x13	WWR	4.8224	6	17.9035	2
Ouargla (Hassi Messaoud)/F	x1	Azimuth	5.7613	5	11.2563	6
	x2	Equipment power	1.3410	11	1.9683	12
	x3	Ext. walls	2.0477	9	3.4424	10
	x4	Glazing	1.5822	10	11.9377	5
	x5	Infiltration	15.5967	3	1.2809	13
	x6	Int. walls	0.3945	13	2.8752	11
	x7	Lighting	2.5906	8	16.0456	3
	x8	Roofs	4.3855	6	4.5281	9
	x9	Shading	1.1914	12	4.5762	8
	x10	Space density	49.7177	1	13.2352	4
	x11	Ventilation schedule	11.5824	4	8.7818	7
	x12	Ventilation rate	26.7397	2	30.8781	1
	x13	WWR	2.7225	7	20.7024	2

^a1 indicates the most effective parameter and 13 indicates the lowest one.

Results of the PMV-PPD model

Analysing the range of outputs for design parameters revealed that variables such as azimuth, equipment power, external walls, glazing, internal walls, lighting, shading and WWR showed similar potential across the eight cities, with median discomfort hours around of 3800 h per year. Discomfort hours median could decrease to 3600 h per year for the roof parameter, go below 3500 h per year for the infiltration parameter and reach around 3300 h per year for the ventilation rate parameter. In cases of space density and ventilation schedules, the median discomfort hours were increased to more than 4000 h per year (Figure 14).

Figure 14.

SA results for the PMV-PPD model.

Outliers were observed, indicating increased discomfort hours for the infiltration and ventilation rate parameters and decreased discomfort hours for the space density parameter. This behaviour was due to the significant range of these parameters that affected the output, as reflected more clearly in the SI.

The position of the base case within the range of each parameter indicated the level of design optimisation. When the base case was at the top of the parameter’s range, it indicated the need for further optimisation. When the base case was within the range near the median, it meant that the design met moderate conditions. Finally, if the base case was at the bottom of the range, it indicated that the parameter was well-optimised to reduce discomfort hours. Across all eight cities, the base case or reference case design was consistently positioned at the upper limit of the range, implying that the design was suboptimal and could benefit from further optimisation to reduce discomfort hours (Figure 15).

Figure 15.

SA results for the PMV-PPD model across different cities.

Results of the adaptive model

Exploring the range of outputs for design parameters in the adaptive model revealed that parameters such as azimuth, equipment power, external walls, infiltration, internal walls, lighting, roofs, shading, space density and ventilation schedule exhibited similar potential across the eight cities, with median discomfort hours around 2500 h per year. The median discomfort hours could decrease to 2300 h per year for the infiltration parameter, go below 2000 h per year for the ventilation rate parameter and reach around 2700 h per year for the WWR parameter (Figure 16).

Figure 16.

SA results for the adaptive model across different parameters.

Outliers were observed, indicating increased discomfort hours for parameters such as space density, ventilation rate and WWR. This phenomenon was again related to the substantial range of these parameters affecting the output, as emphasised in the SI (Figure 17).

Figure 17.

Comparative results of the adaptive model across different cities.

The position of the base case within the range of each parameter in the adaptive model indicated the level of design optimisation. The results differed across the eight cities, with the base case being located at the top or bottom of boxplots for various parameters. This means that the design could be optimised more.

Sensitivity index results

The SIs showed relatively consistent rankings across the eight cities. In the PMV-PPD model, space density was the most influential parameter, closely followed by ventilation rate, infiltration, ventilation schedule, roofs and azimuth. The remaining parameters (WWR, lighting, equipment power, shading, external walls, glazing and internal walls) had relatively similar impacts on discomfort hours.

In the adaptive model, ventilation rate was identified as the most influential parameter, followed by WWR, lighting, space density, ventilation schedule, glazing and azimuth. The other parameters (roofs, shading, external walls, infiltration, equipment power and internal walls) showed relatively similar impacts on discomfort hours.

Table 8 provides detailed information on SI percentages and the ranking of each parameter across the eight cities.

Discussion and implications

The study’s primary conclusion underscores that modifying the studied design parameters can be a highly effective approach to reduce discomfort hours in harsh hot climates and create a more acceptable indoor environment for both MC and NV buildings across various climate zones (Figure 18(a) and (b)).

Figure 18.

SI of design parameters for PMV-PPD and adaptive models.

A key aspect to consider is achieving a balance between summer and winter needs in the design process. This balance should be maintained through passive design solutions. For example, during the summertime, solar radiation must be effectively controlled through shading to prevent excess heat gain. Such considerations are critical for the year-round estimation achieved through software simulations.

The parametric variation method employed in this study revealed the varying importance of design parameters across different climate zones. In the PMV-PPD model, the parameters with the most significant impact nationwide, as indicated by the SI and simulations, include space density, ventilation rate, infiltration, ventilation schedule, roofs and azimuth. These parameters are particularly influential due to the PMV-PPD model’s sensitivity to user metabolism and airspeed.

In contrast, the adaptive model highlights ventilation rate as the most influential parameter across the country, emphasising the importance of natural ventilation in expanding the comfort zone. Additionally, WWR, lighting, space density, ventilation schedule, glazing and azimuth were identified as key parameters influencing discomfort hours.

Compared to similar previous studies in the Malaysian context⁵⁸ that evaluated free-running ventilation strategies in specific residential building, a key difference emerges regarding the effectiveness of passive cooling strategies. The Malaysian study indicated that free-running ventilation alone may not achieve indoor comfort temperatures without mixed-mode ventilation. Another study in Malaysian context also⁵⁹ investigated affordable retrofitting methods for achieving thermal comfort in a hot-humid climate residential setting. It demonstrated the effectiveness of using high-density polyethylene (HDPE) nets as roof covers to reduce convective heat flux and improve compliance with thermal comfort standards. The study suggested that low-cost retrofitting methods can effectively enhance indoor comfort in residential buildings in hot-humid climates. Together, these studies offer complementary perspectives on addressing thermal comfort concerns providing practical solutions for retrofitting residential buildings. Another study⁶⁰ on thermal comfort in residential buildings examined comfort temperature ranges and the correlation between indoor and outdoor temperatures. It highlighted the variation in comfort temperatures based on climate and operation modes and discussed the potential for energy savings through adaptation measures such as roof retrofitting, building orientation and shading design. While our study has provided insights specific to architectural design and discomfort hours reduction in hotel buildings, this review offered a broader perspective on comfort temperature variations and energy-saving strategies applicable to residential settings. These results are comparable with the outcomes derived from the current work for Algerian hotels because they both emphasise the importance of passive design strategies and the potential for low-cost retrofitting methods to enhance indoor comfort in different building types and environmental contexts.

Psychometric chart predictions and recommended strategies should be explored and compared with the results obtained in this study. Further investigations should also address SI and uncertainties.

To address the limitations of achieving full thermal comfort passively through design optimisation with LSA and parametric variations, future research should explore active solutions and conduct a GSA using genetic algorithms for optimisation. Moreover, adopting an evaluation system or benchmarking for hotels in different climate regions in Algeria is recommended. Certain parameters can be applied in new constructions, while others can be applied during the refurbishment of existing buildings.

This study serves as a call to action, aiming to draw the attention of researchers, practitioners, decision-makers and architects to the challenges of hotel building design in various Algerian climates and regions with similar climates worldwide. The research encourages a methodological approach to explore the effectiveness of applying design parameters in reducing discomfort hours in both MC and NV buildings.

The approach presented in this study has the potential to significantly reduce discomfort hours in the early design stage and during retrofit projects in diverse Algerian climates and similar climate zones globally. By focusing on the preservation of local resources and cultural techniques, it is essential to ensure the consistent application of strategies for reducing discomfort hours within Algeria’s climatic conditions. This study advocates for future research in the optimisation of indoor environments for both MC (PMV-PPD comfort model) and NV (adaptive comfort model) buildings.

Conclusion

In this study, a comprehensive assessment of the sensitivity of 13 key parameters was conducted and their impact on annual discomfort hours in a reference hotel was discussed. This case study involved an in-depth investigation of a hotel through in-situ measurements and a calibrated dynamic simulation model. Both the PMV-PPD and adaptive comfort models were adopted to assess their performance across the eight climate zones in Algeria. The 13 design parameters were explored for their impact on discomfort hours, using parametric variation and simulations. The findings underscore the potential to optimise design passively, enabling the regulation of thermal performance in hotels located in diverse climates.

The study demonstrates that passive strategies can effectively enhance indoor thermal comfort across various climate zones. The ventilation rate parameter emerges as a crucial design element for both comfort models, while internal walls are identified as having a relatively low impact on the overall outcome. It serves as a real-life test case that has been meticulously examined through LSA in different Algerian climate regions. The study’s application extends to the categorisation of hotels based on discomfort hours, thereby aiding in assessing energy demand for both cooling and heating.

This research offers a generalisable approach to harness the passive design potential for improving indoor thermal comfort in both MC and NV buildings. It provides insights for further exploration and application in the hotel sector and other similar contexts worldwide.

Footnotes

Author contributions

All authors contributed equally in the preparation of this manuscript.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Ahmed Kaihoul

Leila Sriti

References

Balaras

Gaglia

Georgopoulou

Mirasgedis

Sarafidis

Lalas

. European residential buildings and empirical assessment of the Hellenic building stock, energy consumption, emissions and potential energy savings. Build Environ 2007; 42(3): 1298–1314.

Fayaz

Kari

. Comparison of energy conservation building codes of Iran, Turkey, Germany, China, ISO 9164 and EN 832. Appl Energy 2009; 86(10): 1949–1955.

Cao

Dai

Liu

. Building energy-consumption status worldwide and the state-of-the-art technologies for zero-energy buildings during the past decade. Energy Build 2016; 128: 198–213.

International Energy Agency . Tracking buildings. Paris: International Energy Agency, 2020. https://www.iea.org/reports/tracking-buildings-2020 (Accessed 1 March 2021).

L’Agence Nationale pour la Promotion et la Rationalisation de l’Utilisation de l’Énergie (APRUE) . La consommation énergétique finale. Algeria: APRUE, 2017, pp. 2–3, https://www.aprue.org.dz/publications.html (Accessed 5 December 2020).

ASHRAE standard 55-2023: thermal environmental conditions for human occupancy. Atlanta, GA: ASHRAE, Inc., 2023.

Nguyen

A-T

Reiter

. A performance comparison of sensitivity analysis methods for building energy models. Build Simulat 2015; 8: 651–664.

Kaihoul

Sriti

. Towards public identity and climate awareness architecture. In: Cities’ identity through architecture and arts. London: Routledge, 2018, pp. pp129–137.

Kaihoul

Sriti

. The cohesive urban social fabric between local architectural identity and climatic response. International conference arquitectonics network: mind, land and society, barcelona, 29-31 May, 1 June 2018: final papers. GIRAS. Barcelona: Universitat Politècnica de Catalunya, 2018.

10.

Amraoui

Sriti

Di Turi

Ruggiero

Kaihoul

. Exploring building’s envelope thermal behavior of the neo-vernacular residential architecture in a hot and dry climate region of Algeria. Build Simulat 2021; 14: 1567–1584.

11.

Faheem

Bhandari

Tadepalli

. Adaptive thermal comfort in naturally ventilated hostels of warm and humid climatic region, Tiruchirappalli, India. Energy and Built Environment 2023; 4(5): 530–542.

12.

Buso

Becchio

Corgnati

. NZEB, cost-and comfort-optimal retrofit solutions for an Italian Reference Hotel. Energy Proc 2017; 140: 217–230.

13.

Kaihoul

Sriti

Amraoui

Di Turi

Ruggiero

. The effect of climate-responsive design on thermal and energy performance: a simulation based study in the hot-dry Algerian South region. J Build Eng 2021; 43: 103023.

14.

Ali-Toudert

Weidhaus

. Numerical assessment and optimization of a low-energy residential building for Mediterranean and Saharan climates using a pilot project in Algeria. Renew Energy 2017; 101: 327–346.

15.

Nguyen

Rockwood

. Developing an energy benchmarking system for hotel buildings using the statistical method and the simulation-based approach. Journal of Green Building 2019; 14(3): 1–22.

16.

Abdulaali

Marlia

Ismar

MSU

Nurul

UMN

Mahmood

. A review on green hotel rating tools, indoor environmental quality (IEQ) and human comfort. Int. J. Adv. Sci. Technol 2020; 29(3): 128–157.

17.

Wafi

SRS

Ismail

Ahmed

. A case study of the climate factor on thermal comfort for hostel occupants in universiti sains Malaysia (USM), Penang, Malaysia. J Sustain Dev 2011; 4(5): 50–61.

18.

Zineddine

Belakehal

Bennadji

. Evaluation of the thermal environment in architecture hotel of Fernand Pouillon. Case of the Ziban Hotel in Biskra. Journal of Applied Engineering Science and Technology 2018; 4(2): 143–148.

19.

Bohdanowicz

Martinac

. Thermal comfort and energy saving in the hotel industry. AMS 15th conference on biometeorology and aerobiology and 16th international congress on biometeorology. Kansas: Environmental Science, Business, Engineering, 2002, 16A. 8.

20.

Pinto

Machado

Matias

Saraiva

Santos

. Thermal Comfort in Hotels: relation between customers’ opinions and empirical data gathered in 2015 summer. Interdisciplinarity in Social and Human Sciences 2016; 5: 511.

21.

Borowski

Zwolińska

Czerwiński

. An experimental study of thermal comfort and indoor air quality—a case study of a hotel building. Energies 2022; 15(6): 2026.

22.

Naderian

Heidari

. Hotel design in the historical context of Yazd with emphasis on the role of the porch in thermal comfort. Journal of Urban Development and Architecture-Environment Identity (JUDA-EI) 2022; 3(10): 1115. DOI: 10.22034/(jupa-ei).2022.346127.1115.

23.

Vujošević

Krstić-Furundžić

. The influence of atrium on energy performance of hotel building. Energy Build 2017; 156: 140–150.

24.

Alhuwayil

Abdul Mujeebu

Algarny

AMM

. Impact of external shading strategy on energy performance of multi-story hotel building in hot-humid climate. Energy 2019; 169: 1166–1174.

25.

Acosta

González

Zamarreno

Alvarez

. Energy savings and guaranteed thermal comfort in hotel rooms through nonlinear model predictive controllers. Energy Build 2016; 129: 59–68.

26.

Kristensen

Petersen

. Choosing the appropriate sensitivity analysis method for building energy model-based investigations. Energy Build 2016; 130: 166–176.

27.

Morris

. Factorial sampling plans for preliminary computational experiments. Technometrics 1991; 33(2): 161–174.

28.

Sanchez

Lacarriere

Musy

Bourges

. Application of sensitivity analysis in building energy simulations: combining first-and second-order elementary effects methods. Energy Build 2014; 68: 741–750.

29.

Tian

. A review of sensitivity analysis methods in building energy analysis. Renewable and sustainable Energy Reviews 2013; 20: 411–419.

30.

Zeferina

. Sensitivity analysis of cooling demand applied to a large office building. Energy Build 2021; 235: 110703.

31.

El-Dessuky

. Harmonization between architectural identity and energy efficiency in residential sector (case of North-West coast of Egypt). Ain Shams Eng J 2018; 9(4): 2701–2708.

32.

Nguyen

Tran

Reiter

. An investigation on climate responsive design strategies of vernacular housing in Vietnam. Build Environ 2011; 46: 2088–2106.

33.

Heiselberg

Brohus

Hesselholt

Rasmussen

Seinre

Thomas

. Application of sensitivity analysis in design of sustainable buildings. Renew Energy 2009; 34(9): 2030–2036.

34.

Sarrazin

Pianosi

Wagener

. Global sensitivity analysis of environmental models: convergence and validation. Environ Model Software 2016; 79: 135–152.

35.

Rodríguez

Andrés

Muñoz

López

JMC

Zhang

. Uncertainties and sensitivity analysis in building energy simulation using macroparameters. Energy Build 2013; 67: 79–87.

36.

Tian

Yang

Wang

. Sensitivity analysis of energy performance for high-rise residential envelope in hot summer and cold winter zone of China. Energy Build 2013; 64: 264–274.

37.

Nguyen

. Sustainable housing in Vietnam: climate responsive design strategies to optimize thermal comfort. PhD thesis. Liège: Université de Liège, 2013.

38.

Nguyen

Singh

Reiter

. An adaptive thermal comfort model for hot humid South-East Asia. Build Environ 2012; 56: 291–300.

39.

Chen

Wang

Calautit

Riffat

Cui

. Comprehensive energy, economic and thermal comfort assessments for the passive energy retrofit of historical buildings-A case study of a late nineteenth-century Victorian house renovation in the UK. Energy 2021; 220: 119646.

40.

Zhang

. Optimizing energy efficiency and thermal comfort in building green retrofit. Energy 2021; 237: 121509.

41.

Alwetaishi

. Impact of glazing to wall ratio in various climatic regions: a case study. Journal of King Saud University-Engineering Sciences 2019; 31(1): 6–18.

42.

Nguyen

ATRS

. An investigation on thermal performance of a low cost apartment in hot humid climate of Danang. Energy Build 2012; 47: 237–246.

43.

Yao

Costanzo

Zhang

. The effect of passive measures on thermal comfort and energy conservation. A case study of the hot summer and cold winter climate in the Yangtze River region. J Build Eng 2018; 15: 298–310.

44.

Yang

Fiorito

Prasad

Sproul

Cannavale

. A sensitivity analysis of design parameters of BIPV/T-DSF in relation to building energy and thermal comfort performances. J Build Eng 2021; 41: 102426.

45.

Zhao

Han

Faouzi

Washaya

Zhang

Jin

. Evaluating Algeria’s social and economic development using a series of night-time light images between 1992 to 2012. Int J Rem Sens 2018; 39(23): 9228–9248.

46.

Centre National d’Études et de Recherches Intégrées du Bâtiment (CNERIB) . Document technique réglementaire - DTR C3-T: réglementation thermique du bâtiment. Alger: CNERIB, 2011, https://www.cnerib.edu.dz/espace.html (Accessed 13 December 2020).

47.

Cataldi

Abdelhamid

Selva

. The town of Ghardaia in M'zab, Algeria: between tradition and modernity. Traditional Dwellings and Settlements Review 1996; 1: 63–74.

48.

Meteonorm . Meteonorm V8.0.0.13161 software for reliable data sources, 2020. https://meteonorm.com/en/download (Accessed 3 September 2020).

49.

Zineddine

. La Sensorialité dans l'Architecture de Fernand Pouillon en Algérie Indépendante. Biskra, Algeria: Université Mohamed Khider-Biskra, 2019.

50.

Dubord

. Fernand Pouillon. Paris-Milan: Eclecta moniteur, 1986.

51.

Maachi Maïza

. La Composition Architecturale dans l’Œuvre Algérienne de Fernand Pouillon, Cas du Sud-Ouest Algérien. Master's thesis. Béchar, Algeria: Centre Universitaire Béchar, 2002.

52.

DesignBuilder . Software for sustainable building design for engineers, 2002. https://designbuilder.co.uk/software/for-engineers (Accessed 5 January 2021).

53.

Climate.OneBuilding.Org . “Repository of free climate data for building performance simulation.” https://climate.onebuilding.org/default.html (2020, Accessed 13 December 2020).

54.

Testo-480 . Portable and stationary measurement solutions, 1957. https://www.testo.com/en-TH/testo-480/p/0563-4800 (Accessed 22 September 2022).

55.

ISO 7726:2002 . Ergonomics of the thermal environment - instruments for measuring physical quantities. Geneva: International Organization for Standardization, 2002.

56.

ASHRAE guideline 14-2014: measurement of energy and demand savings. Atlanta, GA: ASHRAE, Inc., 2014.

57.

Petersen

Kristensen

Knudsen

. Prerequisites for reliable sensitivity analysis of a high fidelity building energy model. Energy Build 2019; 183: 1–16.

58.

Tuck

Zaki

Hagishima

Rijal

Zakaria

Yakub

. Effectiveness of free running passive cooling strategies for indoor thermal environments: example from a two-storey corner terrace house in Malaysia. Build Environ 2019; 160: 106214.

59.

Tuck

Zaki

Hagishima

Rijal

Yakub

. Affordable retrofitting methods to achieve thermal comfort for a terrace house in Malaysia with a hot–humid climate. Energy Build 2020; 223: 110072.

60.

Aqilah

Rijal

Zaki

. A review of thermal comfort in residential buildings: comfort threads and energy saving potential. Energies 2022; 15(23): 9012.