**Analysis of the E**ff**ect of Using External Venetian Blinds on the Thermal Comfort of Users of Highly Glazed O**ffi**ce Rooms in a Transition Season of Temperate Climate—Case Study**

## **Małgorzata Fedorczak-Cisak \*, Katarzyna Nowak and Marcin Furtak**

Faculty of Civil Engineering, Cracow University of Technology, 31-155 Cracow, Poland; knowak@pk.edu.pl (K.N.); mfurtak@pk.edu.pl (M.F.)

**\*** Correspondence: mfedorczak-cisak@pk.edu.pl; Tel.: +48-696-046-050

Received: 24 October 2019; Accepted: 18 December 2019; Published: 23 December 2019

**Abstract:** Improving the energy efficiency of buildings is among the most urgent social development tasks due to the scale of energy consumption in this industry. At the same time, it is essential to meet high requirements for indoor environmental quality and thermal comfort. The issue of overheating is most often analysed in summer but it also occurs in transition seasons, when the cooling systems do not operate. The paper attempts to evaluate the effectiveness of external mobile shading elements on the microclimate of rooms with large glazed areas in the transition season. Passive solutions, such as shading elements, which limit the increase of indoor temperature, do not always allow the acquisition and maintenance of comfortable solutions for the duration of the season, as demonstrated by the authors. Temporary cooling of the rooms may be necessary to maintain comfortable conditions for the users, or other solutions should be devised to improve comfort (e.g., reduction of clothing insulation characteristics). The novelty of the study consists in the analysis of comfort in a "nearly zero energy consumption" building (NZEB) during a period not analyzed by other scientists. This is a transition period during which heating/cooling systems do not operate. The research task set by the authors involved the assessment of the possibility to reduce office space overheating in the transition season (spring) by using external shading equipment in rooms with large glazed areas. An additional research task aimed at checking the extent to which user behaviour, such as reduction in clothing insulation characteristics, can improve comfort in overheated rooms. The results of the tests reveal that the difference in the ambient air temperature between a room with external venetian blinds and an identical room with no venetian blinds in the transition season, i.e., from 27 March to 6 April 2017, ranged from 12.3 to 2.1 ◦C. The use of a shading system (external venetian blinds positioned at an angle of 45◦) reduced the number of discomfort hours by 92% (during working hours) compared to the room without external venetian blinds. A reduction in the thermal insulation of the clothes worn by people working in the room with no venetian blinds helped to reduce the number of discomfort hours by 31%.

**Keywords:** thermal comfort; overheating; transition seasons

## **1. Introduction**

The policy of the European Union obliges member states to introduce a new standard of nearly zero energy buildings [1,2]. The implementation of the directive on the energy characteristics of buildings translates onto requirements for buildings set out in the Technical and Building Conditions [3]. Passive buildings have also become more popular in Europe [4,5]. Both standards of buildings are characterised by very low energy demand. Designing such buildings requires wide knowledge of the

discipline [6]. A number of newly designed office and public buildings are characterised by large glazed areas. Despite selecting the correct insulation characteristics for the building structure elements, excess glazing generates large and undesired solar gain especially in summer. As a result of such design solutions, uncomfortable working conditions are created. Therefore, it seems necessary to estimate thermal comfort conditions at each work station in the period of time when overheating occurs in the building. Proposals for system (need for cooling), architectural and building solutions aimed at the reduction of overheating should result from such analyses. The issue of creating and maintaining indoor conditions comfortable for users has already been described in the literature [7,8]. Fanger proposed that the subjective thermal comfort sensation of the users of rooms should be identified with PMV (predicted mean vote) and PPD (predicted percentage of dissatisfied) assessment indices. The method developed by Fanger was incorporated into European standards concerning the thermal comfort of rooms [9–11]. The need to reduce energy consumption for heating and cooling purposes, which has been increasing recently, encourages designers and researchers to carry out more exact analyses for ensuring thermal comfort. Studies are directed towards improving the comfort model [12,13] through a number of experimental tests in laboratories, as well as in real-life objects [14,15]. Many researchers include thermal comfort into indoor environment quality models, additionally taking into account acoustic and lighting comfort and proposing modifications [16–19].

Summing up, low energy consumption must not be the only design criterion. The indoor microclimate is a combined effect of design, building and use of specific rooms. Human activity, clothing insulation and environmental parameters, such as ambient air temperature, mean radiation temperature, air flow rate and relative humidity, affect thermal comfort. Thermal comfort assessment is based on such indices as PMV and PPD.

The paper describes the influence of external shading elements on thermal comfort in premises with large glazed areas. Energy-efficient windows, which are nowadays regularly installed in buildings, are reaching the point where, on an annual basis, heat gains through a window exceed losses. The architecture of contemporary public and office buildings is characterised by highly glazed facades. Even in temperate climate zones, which is where Poland is situated, the solution generates intensive solar gain in the rooms, especially in the southern and western façades. The study of indoor comfort and air quality has been undertaken by a number of research teams in different climate zones [20–22]. The studies have covered different categories of buildings: residential, office, exhibition and entertainment facilities [23–25]. The scope of the studies analysed in the paper has been limited to office premises in buildings with glazed facades. Overheating is a major problem, which occurs in this type of premises. A number of papers [26–28] quote algorithms and guidelines pertaining to the glazing area size depending on material and construction solutions, and location. Creative vision and visual effect tend to be more important for architecture designs. Designers and investors often neglect researchers' recommendations to include glazed elements in small areas of the façades. That is why the purpose of this paper is to evaluate the effectiveness of external mobile shading elements for rooms with large glazed areas. The results presented in [29] showed that the use of shading devices is of great importance in cities, where temperatures and the intensity of solar radiation are high, but it also helps to reduce the demand for cooling in locations of temperate maritime climate. The authors of papers [30,31] analysed the impact of selected shading devices on the heat and lighting characteristics of rooms. Many of the studies are purely simulations. They are usually carried out for the summer or winter periods [32]. In transition seasons in temperate climates there are more cloudless days with a high value of direct solar radiation intensity. The issue of the overheating of rooms in these seasons is not often analysed in the literature.

In Poland's climate zone, energy for heating purposes in buildings with well-insulated structure is supplied from mid-October to mid-March. This is the period when it may be necessary to cool rooms on sunny days. Typically, the cooling system operates from May to September. Simulation analyses and design calculations for the period are used as the basis for selecting the parameters of the outer casing of building partitions and installation systems.

Researchers studying thermal comfort in buildings typically focus on the summer and winter periods when heating and cooling systems are turned on. The novelty of the authors of this article is the analysis of the transition period when the heating or cooling systems of the room do not work. An additional value is the analysis of thermal comfort in a building with "almost zero energy consumption" (NZEB). NZEB buildings comply with the requirements of Directive 2010/31/EU and will be a standard in Europe from 1 January 2021. The article makes a valuable contribution to the design and use of this type of buildings, so that, in addition to the low energy consumption required by regulations, also achieving comfort of use.

Studies on thermal comfort in this article are connected with a transition period in the climate in Poland (spring). The analyzed time is the turn of March and April. At this time, at noon, the solar altitude of the Sun is 38.04◦ to 49◦. This not-too-large angle of the Sun's height allows for intense penetration of solar radiation into the interior of the room. During this period, heating and cooling systems usually do not work. Most solar energy for the analyzed location is generated in July, June and May. Most thermal comfort tests in buildings are related to summer (cooling) or winter (heating) periods. In the transition periods (spring, autumn) it is assumed for a standard construction to avoid heating or cooling the rooms.

The NZEB building was used for analysis, which by definition should be characterized by good insulation of external partitions, adequate protection against overheating, and a correspondingly low value of energy used for heating and cooling. The authors wanted to show that with modern buildings, with a well-insulated block, strongly glazed passive solutions, limiting the temperature increase in the room, such as shading elements, even for a transitional period may not be sufficient to obtain and maintain comfortable conditions throughout the period, which the article showed the authors. The rooms analyzed were characterized by a large ratio of the glazed area to the floor area of 5.63. The tested rooms are an example of the current solutions of office buildings, characterized by high heat capacity of ceilings, light internal walls and large glazed surfaces.

Avoiding the use of external shading devices due to investment costs results in obtaining thermal conditions uncomfortable for users or the necessity to cool rooms by up to 10 K.

An additional goal of the research presented in the article was to indicate that in the case of objects with a large area of external glazing, the need for cooling should be included in the calculation of energy demand also for spring months. This, of course, is associated with an increase in energy demand indicators for the building and an increase in operating costs.

Tests were carried out in an experimental laboratory building, which serves a public function. The building in which the tests were carried out is situated in the centre of Cracow, among city centre structures. It is a passive technology building, which meets the requirements for nearly zero energy buildings (NZEB) in Poland. It is also an experimental building, designed for performing energy efficiency tests.

The research questions posed by the authors for the studied case are the following:


#### **2. Materials and Methods**

Tests on thermal comfort were carried out in an experimental building of the Krakow University of Technology—Małopolskie Laboratorium Budownictwa Energooszcz ˛ednego (MLBE—Energy-Efficient Building Laboratory of Lesser Poland). The MLBE building is dedicated to such experiments. The tests were performed in rooms with the same area 37.2 m2, on the second and third floors. The location of rooms P1.06 and P2.04 is shown in Figure 1. The external façade of the selected MLBE rooms was southward and westward-oriented. This is a glazed façade with a total area of 26.20 m<sup>2</sup> in each room. Room P2.04 had external mobile shading venetian blinds fitted. Room P1.06 had no shading system (Figure 1).

**Figure 1.** Location of test room P2.04 (with external venetian blinds) and P1.06 (with no venetian blinds).

The rooms selected for the tests represent typical office premises in office buildings of contemporary design.

The MLBE building was designed to meet the requirements of thermal protection for passive buildings [4]. The experimental MLBE building fulfils the Polish requirements for NZEBs presented in [3]. The thermal insulation parameters of the MLBE building and of the external partitions of passive buildings, and the Polish thermal insulation requirements included in [3] are presented in Table 1.


**Table 1.** Heat-transfer coefficient U W/(m2·K) for Energy-Efficient Building Laboratory of Lesser Poland (MLBE) building, requirements according to Polish regulations [3] and requirements for passive buildings [4].

The experimental MLBE building, where tests were performed for the purpose of the study, was divided into 14 independent heating and cooling zones. This means that every zone can be heated or cooled independently. The MLBE building's control system was used to stabilise the conditions in the rooms adjacent to the test rooms. During the experiment, the control system was designed so that a temperature of +20 ◦C was maintained in the rooms adjacent to the test rooms.

The thermal comfort test presented in the paper was carried out for the transition season when the heating/cooling systems in the rooms were not in operation. The representative period lasted for 11 representative days, between 27 March and 6 April 2017. The representative period was selected with regard to the greatest number of cloudless days with direct solar radiation.

The air was supplied by an Air Handling Unit (AHU) with a recuperator with 90% heat recovery (Figure 2a). The quantity of the air supplied by a system of intake ventilators (Figure 2b) was constant and amounted to 25 m3 per hour (assuming that only one user was present). The supplied air temperature was +18 ◦C.

**Figure 2.** (**a**) View of the Air Handling Unit (AHU) with a recuperator with 90% efficiency. (**b**) The view of the intake vent grille in room P1.06; the grille is located in the central part of the ceiling.

The test was carried out using two sets of sensors for thermal comfort tests. The set of sensors placed in room P2.04 is shown in Figure 3a,b. The analyzed building is an office building. The building in which the tests were carried out is a building that meets the requirements of the building "with almost zero energy demand" (NZEB). Buildings of this type will be standard in European Union countries from 2021. NZEB buildings are characterized by high wall insulation and high tightness of the building envelope. The building is equipped with a heating and cooling system, the power of which has been calculated assuming that the system works in the summer and winter season. The system was not expected to work during the transition period. Most charts present round the clock measurements, but the detailed analysis focuses on working hours. The rooms analyzed are small office rooms with an area of about 37 m2. The research assumed that each room is used by one person between 8.00 a.m. and 8.00 p.m.

The parameters of the sensors are presented in Table 2.

**Figure 3.** (**a**) View of microclimate meter. (**b**) Description of sensors.



Figure 4 presents the arrangement of the test equipment in rooms P1.06 and P2.04. The microclimate meter is placed at the intended workstation.

**Figure 4.** Arrangement of sensors in rooms P1.06 and P2.04.

The tests were carried out using measuring equipment that meets the requirements of the PN–EN 7726 [33] standard. The measuring equipment (Figure 3) is a microclimate meter.

The measured parameters included:


The data were collected every 10 min.

The data from the sensors are given in Table 2.

On the basis of measurements, thermal comfort parameters PMV and PPD were calculated from formulas [9,34,35].

External environmental parameters were collected from weather stations located at the laboratory's south elevation (Figure 5). The weather station recorded the following data (Figure 6):


**Figure 5.** Weather station on the southern façade.

**Figure 6.** Description of sensors in the weather station.

The scope of measurement and accuracy of the sensors in the weather station are presented in Table 3.



## **3. Results**

Sunny days with minor cloud cover dominated in the period selected for the analysis, i.e., from 27 March to 6 April 2017. As a result of minor cloud cover, the air temperature during the day was fairly high but low during the night. Figure 7 presents the waveform of locally measured values (weather station on the southern façade) of outdoor air temperature and total solar radiation intensity. In Figures 7–17 the hours of use of the rooms (8 a.m.–8 p.m.) are marked with a rectangle.

**Figure 7.** Waveform of outdoor air temperature and total solar radiation between 27 March and 6 April 2017. Measurements were carried out by means of a weather station on the southern façade.

No heating or cooling systems were operated in the test rooms, as was described in the "Materials and Methods" section. The rooms were not used in the reference period and, as such, no internal heat gains were generated. The only external façade of rooms P1.06 and P2.04 is glazed. Other walls were adjacent to the rooms, where there were constant temperature conditions (+20 ◦C).

## *3.1. Room P1.06*

Figure 8 presents the wavelength of outdoor and indoor temperatures in room P1.06—with no venetian blinds installed.

**Figure 8.** Waveform of outdoor and indoor temperatures in room P1.06—with no venetian blinds installed.

The duration of maximum and minimum air temperature values in room P1.06 corresponds to the wavelength of the maximum and minimum outdoor temperatures. This is shifted by about two hours against the maximum values of solar light intensity for the southern façade.

The ambient air temperature in the analysed room reaches very high values. From 23.25 ◦C at night, to 35.13 ◦C in the early afternoon. Despite low outdoor temperatures at night (decreasing to ca. 5 ◦C), intensive solar radiation during the day (max. value 672 W/m2) contributes to such a temperature wavelength. Another factor which affects temperature distribution in the room is the continuous air supply at a constant temperature of 18◦C and the stabilised temperature conditions in the adjacent rooms (P1.04, P1.07 and P1.10).

Figure 9 presents indoor temperature (*Ti*), mean radiation temperature (*Tr*) and operating temperature *To* in room P1.06. The *Tr* value was identified based on the recorded temperature measurements of a blackened sphere. Significant differences, which sometimes exceed 20%, can be observed.

**Figure 9.** Ambient air temperature (*Ti* ◦C) and infrared radiation temperature (TR ◦C) in room P1.06.

Differences between *Ti*, *TR* and, consequently, the operating temperature *To* are clearly noticeable during the hours of intensive solar radiation. Sample wavelengths of the temperature values for the selected day are presented in Figure 10. At high indoor temperature that day in room P1.06 (26.9 ◦C–35.6 ◦C), the operating temperature reached the maximum value of 38.3 ◦C.

**Figure 10.** Values of ambient air temperature *Ti*, mean temperature of a blackened sphere TR and resultant operating temperature *To* (Room P1.06) on 2 April 2017.

## *3.2. Room P2.04*

Room P2.04 is equipped with a system of external sun shading devices—venetian blinds—with panels that can be positioned at any angle. For the entire period of the experiment, the venetian blinds were positioned at an angle of 45◦. Figure 11 shows the values of total solar radiation on the southern façade, indoor air temperature in rooms P1.05 and P2.04 and external temperature.

**Figure 11.** Wavelength of external radiation and indoor temperature in room P2.04—with external sun shading devices.

Figure 12 presents a wavelength of outdoor and ambient air temperatures in room P2.04.

**Figure 12.** Wavelength of outdoor and indoor temperatures in room P2.04—with external sun-shading devices.

Ambient air temperature in room P2.04 was highly stabilised and ranged from 20.13 ◦C to 23.80 ◦C in the analysed period. This result is mainly caused by the significant reduction in solar gain owing to the system of external venetian blinds. The external panels were positioned at a 45◦ angle towards the windowpane. The nearly unchanged temperature values in room P2.04 were also related to the constant air supply temperature (18 ◦C) and the stabilised temperature conditions in the adjacent rooms (P2.03, P2.05 and P2.08). Solid reinforced concrete floors with a high heat capacity, limiting the room at the top and bottom, were another factor which alleviated temperature fluctuations.

The maximum temperature in room P2.04 was normally recorded in late afternoon hours. The time difference between the value of the greatest solar radiation intensity and the highest temperature in room P2.04 was up to six hours. This resulted from the position (inclination) of the panels of the external venetian blinds, which protected the room from direct solar radiation gain contrary to room P1.06, which was not equipped with a system of sun shading devices.

As a result of using external sun shading devices in room P2.04, the ambient air temperature distribution and the value of the mean radiation temperature were stable, which contributed to a favourable distribution of operating temperature (Figure 13).

**Figure 13.** Wavelength of indoor air temperature changes in room P2.04 (*Ti*) and mean radiation temperature (TR).

In this case, contrary to room P1.06, the values of ambient air temperature *Ti*, radiation *TR* and *To* were similar, even during intensive solar radiation hours. The recorded differences do not exceed 0.25 ◦C.

The identified mean radiation temperature value for the room with external venetian blinds is lower than the ambient air temperature recorded during that time. The opposite occurs in room P1.06 (Figures 9 and 10), where solar radiation penetration was strong. The mean surface radiation temperature reaches much higher values than ambient air temperature.

Figure 14 presents a wavelength of outdoor and ambient air temperatures in rooms P1.06 (with no venetian blinds) and P2.04 (with external shading devices).

**Figure 14.** Wavelength of outdoor and ambient air temperatures in rooms P1.06 (with no venetian blinds) and P2.04 (with external shading devices). Visible data gaps are caused by technical problems with the measuring equipment. The sensors had to be submitted for technical inspection.

## *3.3. Comparison of Results*

When identifying thermal comfort indices PMV and PPD, it was assumed for both rooms that their users worked in clothes characteristic for the winter/spring season. The thermal insulation values of the clothing, Clo, were adopted based on standard [9], which applies to people doing office work. The Clo and Met values adopted for analysis are shown in Table 4.

**Table 4.** Clo and Met values adopted for thermal comfort analyses for the transition season between 27 March and 6 April 2017.


The results of analysis and measurements of PMV and PPD comfort indices for both rooms are presented in Figures 15 and 16.

**Figure 15.** Values of predicted mean vote (PMV) index in rooms P1.06 and P2.04. Visible data gaps are caused by technical problems with the measuring equipment. The sensors had to be submitted for technical inspection.

**Figure 16.** Values of predicted percentage of dissatisfied (PPD) index in rooms P1.06 and P2.04.

**Figure 17.** (**a**) PMV relationship (assumed value: +0.5 [-]) P1.06. (27.03.2017); (**b**) P1.06. (28.03.2017); (**c**) P1.06. (29.03.2017); (**d**) P1.06. (30.03.2017).

#### **4. Discussion**

The wavelengths of the values of the PMV and PPD indices are presented in Figures 15 and 16. They clearly suggest the high efficiency of the external shading devices. The values of the PMV index in the room with closed venetian blinds ranged from −0.75 to +0.17. The percentage of dissatisfied people amounted to 5–17%. This means that the conditions in the room both during the day and at night were nearly perfect. The number of thermal comfort hours in room P2.04 (external venetian blinds installed) in the analysed period was 23, with only nine hours between 8.00 a.m. and 10.00 p.m. These hours can be treated as the office hours, which means that users do not stay in the building beyond them. A large part of the energy consumed by office buildings is used for cooling. This energy can be significantly reduced owing to shading devices. Many researchers have demonstrated the efficiency of such measures through simulations [32,36]. A simulation study in southern Italy, presented in [37], showed the possibility of reducing energy consumption for cooling purposes in highly glazed office rooms with external roller blinds by nearly 50%, compared to rooms with no external roller blinds installed. In the room marked as P1.06 (with no external sun-shading devices), conditions regarded as comfortable lasted for only 16 h. For the rest of the time, temperatures in the range of 27 ◦C and 35 ◦C were unacceptable for the users. The maximum PMV value amounts to 3.48 at 100% of dissatisfied people (PPD = 100%). A number of researchers has studied the behaviour of users of rooms in order to improve ambient air quality, thermal comfort and building performance efficiency (performance improvement). The opening of windows is among such behaviours [38] and [39]. Studies of thermal comfort [40], presented by the team from Harbin, covered transition seasons, in addition to the winter heating season. The behaviours of people aimed at adjusting their thermal comfort were analysed for the spring season, also when heat distribution systems were not in operation. The users of office and residential premises improved their comfort by opening windows, changing the insulation characteristics of their clothing and by drinking more water. Windows were not opened in the rooms of the MLBE building analysed in the paper. The authors analysed the extent to which a change in the insulation characteristics of clothes (Clo = 1.0 assumed in the tests) may contribute to the thermal

comfort experience of the users under the conditions that occurred in room P1.06. The Clo value at which the thermal comfort of users would become neutral was calculated. The results are presented in Figure 17a–d. For selected days (27–30 March), the figures present the Clo value (blue bars) at which the users of the room would have to reduce the insulation characteristics of their clothes to reach an acceptable level of thermal comfort experience. For the analysed office building, the PMV should range from −0.5 to +0.5.

Analysing the results presented for sunny days (Figure 17a–c), it can be concluded that despite highly uncomfortable (unfavourable) thermal conditions in the room, user behaviour can improve the comfort experienced. Such improvement is possible in the pre-noon and afternoon hours. At noon and in the early afternoon, no clothing reduction helped to reach comfortable conditions.

On a cloudy day (30 March) with lower solar radiation intensity, comfortable conditions can be reached by slightly changing the insulation characteristics of clothing (clo) to the values of 0.62–0.92 (Figure 17d). Only the calculations for the period between 14:00 and 16:00 revealed a significant decrease in the clothing insulation characteristics due to the momentary operation of the sun (Figure 11).

An analysis of the selected period between 27 March and 6 April 2017 revealed that when the users of the rooms with no external venetian blinds changed their clothing insulation characteristics, the number of discomfort hours reduced from 117 to 81 (during the hours of use of the rooms).

## **5. Conclusions**

The building in which the tests were carried out meets the requirements of the building "with almost zero energy demand" (NZEB). Buildings of this type will be standard in European Union countries from 2021. NZEB buildings are characterized by high wall insulation and high tightness of the building envelope. The building is equipped with a heating and cooling system, the power of which has been calculated assuming that the system works in the summer and winter season. The system was not expected to work during the transition period. Studies have shown that in the case of large glazed facades, overheating also occurs during the transition period, especially during cloudless sunny days. Such a building's response in the analyzed time may contribute to the need to cool the rooms, and thus to increase operating costs.

Highly glazed façades in office rooms may generate significant internal thermal gains even in the transition season (spring). External venetian blinds used in the reference room helped to maintain comfortable thermal conditions (−0.5 < PMV < +0.5). This means that the implementation of such systems helps to reduce or even eliminate energy consumption for cooling purposes.

The aforementioned thesis is confirmed by the thermal comfort conditions observed during the test in the room with no venetian blinds. Despite the same geometry, orientation, use and equivalent schedule of ventilation equipment operation, the thermal comfort conditions obtained (0.20 < PMV < 3.70) varied significantly from those in the room with venetian blinds.

Another conclusion drawn based on the experiment conducted concerns the possibility of the users improving their own thermal comfort experience. Replacing clothes recommended for the heating season (Clo = 1.0) with summer office clothes (0.25 < Clo < 1.0) helped to reduce the number of thermal discomfort hours of employees by 30%. Such actions can reduce the number of operating hours of the cooling system from 12 h to even less than 4 h a day. This was a case study carried out on chilly but sunny days of the transition season.

**Author Contributions:** Conceptualization, M.F.-C., M.F.; methodology, M.F.-C., K.N.; software, M.F.-C., K.N.; validation, M.F.-C., K.N.; formal analysis, M.F.-C., K.N.; investigation, M.F.-C., K.N.; resources, M.F.-C., M.F., K.N.; data curation, M.F.-C., K.N.; writing-original draft preparation, M.F.-C., K.N.; writing-review and editing, M.F.-C., K.N.; visualization, M.F.-C., K.N.; supervision, M.F.-C., K.N.; project administration, M.F.-C., K.N.; funding acquisition, M.F.-C., M.F. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** The authors thanks Aleksander Panek, Michał Piasecki and Agnieszka Lechowska for substantive support.

**Conflicts of Interest:** The authors declare no conflict of interest.

## **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Review* **Future Design Approaches for Energy Poverty: Users Profiling and Services for No-Vulnerable Condition**

## **Andrea Boeri \*, Valentina Gianfrate, Saveria Olga Murielle Boulanger and Martina Massari**

Architecture Department, University of Bologna, 40136 Bologna, Italy; valentina.gianfrate@unibo.it (V.G.); saveria.boulanger@unibo.it (S.O.M.B.); m.massari@unibo.it (M.M)

**\*** Correspondence: andrea.boeri@unibo.it

Received: 30 January 2020; Accepted: 21 April 2020; Published: 24 April 2020

**Abstract:** Analyzing data from the Energy Poverty Observatory in Europe, it emerges that more than 50 million households in the EU live in energy poverty (people that cannot heat their homes during winter; cannot make their homes comfortable during the summer; pay their energy bills late). Research studies realized in the last 20 years highlight that making energy demand efficient and effective is the more significant and socially important the more it is able to involve users who are unable to sustain energy demand. The evolution of the research sees a narrowing of the field of investigation by focusing on the user dimension of energy poverty, stressing the role of citizens not only as consumer but also as producers of solutions to tackle energy poverty, real energy communities of agents. The paper aims to provide a systematic literature review highlighting the major findings of the topic, investigating the relationship between spatial and social issues, and looking at the state of energy poverty by addressing the profiling of users and consequently of services useful to overcome their current vulnerable condition. The paper is structured in two core sections. The first one gives the results of a systematic literature review on the energy/fuel poverty topic, the second one deepens the role of communities and individuals need, crucial in defining new design approaches for supportive solutions to tackle energy poverty.

**Keywords:** energy poverty; vulnerable users; energy communities; energy poverty metrics

## **1. Introduction**

Energy poverty is gaining attention from European [1,2] global [3,4] policies and research paths [5,6] for more than a decade. The concept identifies a situation where a family or individual does not achieve an adequate level of essential energy services often due to a cascade of conditions, including low income and high energy expenditure caused by poor energy efficiency of housing [7].

It is a complex and often fragmented problem crossing several disciplines [8,9] affecting from 50 to 150 million people who are unable to pay for the primary energy services—such as heating, cooling, lighting, travel, and electricity—that are needed to ensure a decent standard of living [10].

Based on pioneering studies [11,12], a household could be affected by energy poverty due to low income, high energy costs, and energy inefficient dwellings. In concrete terms, this means that vulnerable citizens [13] do not have access to energy services or that the use of these energy services compromises their ability to access other basic services. The Covenant of Mayors 2018 highlights that energy poverty can have serious consequences for health, well-being, social inclusion, and quality of life [14]. According to the report conducted by the International Energy Agency [15], energy poverty includes not only lack of access to modern energy services, but also the reliability of these services and concerns in the affordability of access. Households affected by energy poverty, experience inadequate levels of some essential energy services such as lighting, heating/cooling, use of household appliances, transport, and much more.

Data from the Energy Poverty Observatory show that, in Europe, more than 50 million households alone live in energy poverty conditions, and in particular:


In Italy, the various existing indicators estimate that Italian households in energy poverty are between 2.2 and 4.3 million. About 9.4 million individuals [16–18] can be considered energy poor and vulnerable, as they are unable to sustain energy costs to maintain an adequate level of comfort. For this reason, energy poverty has been seriously taken into account by several policy areas and agendas concerning social, economic, political, environmental, health [19], and climate. In particular, the European Commission's "Clean Energy for All European Citizens" [20] legislative package is providing measures to strengthen the position of the consumers, by providing rules that will give them more flexibility while also protecting them. The package focuses also on self-determination of the consumer, who is allowed to take its own decisions on "how to produce, store, sell or share energy" [20].

The European Commission (EC) has long been committed in tackling the problem, reinforcing its engagement with the establishment of the EU Energy Poverty Observatory in January 2018—to measure, monitor, and share knowledge and good practices to manage energy poverty—and the growing number of initiatives funded and promoted on the subject. The Covenant of Mayors for 2030, in addition to "taking action to mitigate climate change and adapt to its inevitable effects, requires signatories to commit to providing access to secure, sustainable and affordable energy for all. Covenant signatories can improve the quality of life of their citizens and create a fairer and more inclusive society through the reduction of energy poverty" [14].

The challenge of energy poverty has also been included in the United Nations' Agenda 2030 as one of the actions foreseen in Sustainable Development Objective 7 "Ensuring access to affordable, reliable, sustainable, and modern energy systems for all". These are just the last global orientation tackling a phenomenon that is emerging as a consequence of long-lasting macro-level [21] trends, such as the economic crisis, the obsolescence of the building stocks, the lack of qualitative data and knowledge about fuel poverty penetration, the heterogeneity of the phenomenon.

In 2009, EU GDP fell by 4.2% compared to 2008 and there was a sharp increase in unemployment. The economic and financial crisis that started in 2007 occurred against a background of falling wages for European workers. In addition, there was an increase in domestic energy prices, due to the EU's dependence on imported sources. From the end-user perspective, evidence suggest that the price of domestic energy in the EU has steadily increased since the mid-1990s, gradually reducing the purchasing power of households [22]. In the case of Italy for example, the increase in prices, while consumption was substantially stable, has led to an increase in energy expenditure, whose incidence on the total rose from 4.7% in 2007 to 5.1% in 2017. This created a worrying prospect for the most vulnerable households.

Furthermore, most Southern Europe buildings were built in the last century without (or with few) wall insulation, leading to inadequate winter performances. Not to mention the general low thermal transmittance values of surfaces, influencing thermal lag and, consequently, building summer performances. In fact, as assessed by researchers [23,24], fuel poverty is strongly linked with the thermal characteristics of buildings and fuel poor people are not only income poor. Thus, the topic is complex as it involves not only the necessity to create and boost national or local policies, but also the need to find methods for target identification, data sharing, and profiled services.

Energy transition and the resulting emergence of energy poverty take place against the background of marked differences between EU Member States. The geographical-spatial component is becoming increasingly important [25,26], together with the political-programmatic characterization of each state, and within them. For example, Italy is reducing its national energy-clean measures due to recent

austerity policies but also to its political instability; this leaves room for the individual regions or even municipality to take over and produce operative but fragmented policies; Germany, on the other hand, is proposing a transformation of the energy sector with large-scale measures of energy efficiency and renewable energy solutions.

A growing amount of studies [13,27–29] highlight that making energy demand more efficient is all the more significant and socially important the more it is able to involve users who are unable to sustain energy demand. Therefore, contemporary research paths and European policy documents are narrowing the field of investigation by focusing on the user dimension of energy poverty, stressing the role of citizens not only as consumer, but also as producers of solutions to tackle energy poverty and real energy communities of agents.

Following these premises, the paper aims to read the concept in a European dimension not only by providing a five-years systematic literature review highlighting the most recent findings of the topic but also providing a qualitative analysis related to methods of measurement, mitigation measures, and users and vulnerabilities. In particular, the decision to focus mainly on the last five years was led by the intention to detect the most innovative and recent approaches to fuel poverty and to discover if the scientific debate is proceeding toward the inclusion of profiling studies.

In fact, the point of view of the relationship between the state of buildings and the state of poverty is investigated by addressing one of the determinants in the resolution of energy poverty: the profiling of users and consequently of services useful to overcome their current condition of poverty.

In order to answer to those research questions, the paper is structured in two core sections. The first one gives the results of a systematic literature review on the energy/fuel poverty topic having the objective to understand the main and more recent lines of development in the last five years; the second one performs a research aiming to deepens the role of communities, of vulnerabilities, and the potentialities of profiling services to address the problem. Finally, the conclusion recalls the main findings of the paper, assess some field limitations of the papers, and draft future research paths.

## **2. Research Methodology**

The research underlying this paper is twofold, and it follows the structure of the paper. The first part of the paper follows in fact the methodology of the systematic literature review, implying the selection of a database, a timeframe, and specific queries, with the objective to analyze the most recent lines of development on the fuel and energy poverty topic.

The second part follows the methodology of a literature review of papers appeared in the most important international journal, without covering a specific timeframe and including both open access and not-open access papers. This review included the concepts of energy poverty measures, users and vulnerabilities and measures of mitigation. This second part of the research has the objective to give a more qualitative analysis of some of the main interesting aspects of the topic.

The systematic literature review has been performed in December 2019–January 2020 by using the Web of Science (WOS) database and by selecting open access and papers accessible from university libraries and online databases [30]. As a limitation to the field of investigation, the literature review has been done only considering papers written in English, as it is commonly recognized as the scientific international language and belonging to the European context. The results of this systematic review are showed in paragraph 3.

In the repositories, we searched works of the last five years (2015–2020) and we entered the following queries:


The totality of the previous papers has been browsed and filtered following some rules as follows:


From these filters, 118 paper remained and were analyzed in depth. The analysis covered the following aspects:


## **3. Analysis of Articles on European Energy Poverty: A Systematic Literature Review of the Most Recent Approaches**

As introduced, energy and fuel poverty is a growing theme for European researchers. In the past years, in particular, the topic sees an increase in interest from several research fields. Not only the social science is interested in the topic but also architecture, medicine, and policies studies. On the one hand, in fact, the problem is more and more highlighted by policy-makers and professionals, and, on the other, more knowledge is produced, allowing more people searching innovative solutions.

On the semantic point of view, it is possible to observe how the two locutions of "fuel poverty" and "energy poverty" are generally considered as synonyms. Both, in fact, tend to include water, electricity, gas, and other fuels. A third locution appears frequently in the search: "energy justice" which can be considered a more general topic, not only working on technical solutions, but also including socio-political aspects.

On the chronological point of view, the systematic literature review (Figure 1) shows that almost the 35% of the total number of analyzed papers was written in 2019. Even with this predominance, the topic has been quite well investigated homogeneously in the last five years, with a small increase in 2017. An additional query on the WOS database about the keyword "energy poverty"—starting from 1995—showed how the topic has increased year by year since 1999, with a boost starting from 2010 (Figure 2), when the awareness on climate change and the role of building retrofitting in improving households conditions increased.

It is during these years that one of the first policies has been developed in the UK. In particular, the first UK policy is relevant as it sets a threshold below which a household can be considered fuel poor according to Boardman studies [31] (to spend more than 10% of the household income on energy in the home). Additionally, as described by Dubois and Meier [21], in 2009, a European project (EPEE) analyzed that, in the EU, fuel poverty involved between 50 and 125 millions of people and they put the attention to the connection between people vulnerability (especially in relation with elderly, disabled people and single-parent families) and the quality of buildings, highlighting how fuel poverty is more likely to happen in cold damp properties with insufficient heating system and insulation.

**Figure 1.** The image shows the distribution of the papers in the last few years.

**Figure 2.** The image shows the growth of the number of papers from 2000 to 2019.

Considering the last five years, the selected papers can be clustered according to major recurring themes. In fact, 10 major categories have been found, as follows:


Before going deep into the identification of the main trends per each theme, it is important to notice that the majority of the selected paper apply their research and analysis on specific case studies. In fact, only a few papers, mainly related to the theory and method category, are untied from a specific case study. Furthermore, 3.58% of papers are related to a specific application. This finding is interesting as it shows how the topic is strictly linked with local contexts. As, in fact, argued by Dubois, "fuel poverty is strongly linked to the characteristics of housing [ ... ] indeed fuel poverty often results from a combination of low incomes and high energy needs" ([23], p. 107).

Among 58% of research applied to case studies, the high majority (51%) is referred to UK case studies, as visible in Figure 3. In fact, in Europe, the United Kingdom has been one of the first countries implementing specific national policies to solve it [2]. Thus, despite most part of the research seeming to come from the UK, several countries are strengthening their effort in addressing the topic, according to their individual policy framework. Secondly, in Spain (19%), there is a growing attention for the topic, while the rest of applications are mostly equally spread across Europe, with few highlights in Greece, Hungary, Ireland, Poland, and Portugal. Excluding the UK, most applications are concentrated on Southern Europe (39% of papers). This might be related to the effects of the economic crisis hitting harder in Southern Europe countries, provoking cascade negative effects on the construction industry—in charge of the management and improvement of the building's performances—and on the economic possibilities of the most fragile sector of the population. Furthermore, most Southern European buildings were built in the last century without (or with few) wall insulation, leading to inadequate winter performances. Not to mention the general low thermal transmittance values of surfaces, influencing thermal lag and, consequently, building summer performances In fact, data provided by the European Energy Poverty Observatory, in 2018, about the share of population not able to keep their home adequately warm, confirm how the worst situations are mainly concentrated in Southern and Eastern Europe with Bulgaria recording 33.7% of population in this situation, followed by Lithuania with the 27.9%, Greece (22.7%), Cyprus (21.9%), Italy 14.0%. In addition, data about arrears on utility bills seem to confirm this evidence: Spain 35.8%, Bulgaria 30.1%, Croatia 17.5%, Romania 14.4%, and Slovenia 12.5%. Even if these indicators are not sufficient to completely show the fuel poverty situation in Europe [32], they still seem interesting as they confirm the presence of emergencies in those regions and thus the necessity of increasing a knowledge and service creation effort for mitigating it.

**Figure 3.** The graphic in the image represents the geographic distribution of the papers analyzed.

Assuming that a significant part of papers relies on local analysis, the thematic point of view is also interesting for the objectives of this research. In fact, given the 10 categories identified, it is interesting to note that most of them are spread in three major topics: methods and indicators (26%), theory (18%), and policies (16%) (Figure 4). This fact shows that the European debate about the topic seems to still be concentrated on answering research questions such as: "how can we measure the phenomenon?", "what is the extent of it?", "how can the fuel/energy poverty be defined?", "what does energy/fuel poverty mean?", "how can we increase our knowledge about it?". In particular, the identification of methods and indicators seems to be the most explored theme.

As described in several research works [33–35], the most used methodology for understanding the extension of the topic and for analyzing its characteristics in a specific context is to combine quantitative data from statistics (local, national, and European) with qualitative data. If the first category is more easy to find, the second one entails a deeper and closer relation with the local context. In fact, qualitative data are almost always collected through interviews and focus groups or with forms of participation usually adopted in social science disciplines. An increasing amount of studies is then focusing on the spatial aspects of the phenomenon, linking the role of the urban space with the poverty [36–39]. In fact, there are several aspects linked with the spatial approach: on the one hand, there are social connections at the neighborhood dimension that can mitigate energy poverty by creating chains of solidarity and support; on the other hand, the homogeneous district often has similar characteristics in terms of building performance and energy necessities. Thus, implementing a spatial approach into the analysis of energy poverty can give new insights both on the design/technical and social perspectives.

Finally, another interesting method for improving knowledge on energy poverty and, together, studying new strategies is strictly connected with BIM innovations. As described in Zhang et al. [40], in fact, the Building Information Technology can be successfully used also for energy poverty by adding socio-economic information to the model. This allows for investigating retrofitting solutions, among a panel of interventions, also considering poverty-related information.

Among the papers centered on methodologies, an important number of them proposes new metrics of measurement in terms of indicators or indexes. A deeper analysis of these is provided in the following paragraphs.

The second category where papers are concentrated is the theoretical one. As a matter of fact, the topic is still under analysis especially for what concerns the link between energy poverty and energy justice. Attention is in fact related to the ethical aspects of energy and to the necessity to guarantee its access to everybody [40]. However, the majority of theory-related papers are concentrated between 2015 and 2017, with less development in the last two years, where, instead, the new topic of energy citizenship arises. With energy citizenship intended, there is participation of customers in energy production not only as passive actors of the market but as prosumers and active participants [41]. However, the link between energy poverty and energy citizenship is still under-explored, and it constitutes an interesting line of research for the future of the topic.

The systematic literature review showed some red lines and some current trends of the international debate about fuel and energy poverty. The two core themes that seem to be highly relevant for the extent of this research are mainly two: metrics for increasing knowledge on the phenomenon, which is the base for then addressing the problem with efficient solutions, and profiling as a way to study solutions (products and services) specifically targeted to different categories of people.

## **4. A Qualitative Analysis on Three Core Aspects of Energy Poverty: Methods of Measurement, Users and Vulnerabilities, Mitigation Measures**

#### *4.1. Measuring Energy Poverty: Critiques to Standard Methods*

The current multiplicity of measurement methods and the growing research interest in providing alternative metrics is a critical point. Data incompatibility, different database and data sources, different measurement units, and different scales of reference are only some of the limitations to a national/international metric for the phenomena.

A review of the most recent proposals of metrics for energy poverty highlights clearly the fragilities of the current metrics while providing possible solutions. The UK is a pioneer in research into assessing the extent of fuel poverty, but France and Germany have also developed a considerable amount of research [42,43] into the issue of assessing and identifying who is most vulnerable.

Moving forward from the initial indication of less than 10% of income as an indicator of energy poverty [31], EPOV has been providing and gathering a series of data concerning energy poverty and, more importantly, it has been collecting different metrics. This data stream has been providing more evidence-based approaches to energy poverty [12], but also a higher awareness of the multi-dimension of the problem and of its metrics. Many authors [44,45] recall the critical issues arising from geographical diversity (different countries require different mixes of policies and measures to address energy poverty) and of scale (macro level and policies and micro level to implement tailored-based measures). To this is added the lack of a common definition, one of the main causes of the insufficient policy measures adopted so far [46]. In the light of this complexity, several attempts have been made to develop a number of effective measurement strategies, covering not only diversified domestic energy uses, but also the habits of different segments of the population.

The amount of variables and indicators to be aggregated to measure energy poverty—and eventually to provide a profile of the consumer—is linked to the different interpretations and definitions of such topic [44,47]. The indicators provided have a varied nature (e.g., self-reported experiences, income, energy expenditure, building features) and rely on secondary dimension linked to energy poverty determinants (e.g., energy market, climate, cost of living). This opens up a criticality in terms of research on the topic, of political framing (related to welfare or housing or energy), and of possible actions to tackle it. A further challenge for research entails finding a key for its reading that is multi-disciplinary, related to specific context, user-cantered but at the same time that provides practical, economic, and feasible solutions to be applied short-term.

As reviewed by many articles [45,47,48], the European Commission has been suggesting since 2010 a series of consensual indicators deriving from existing data surveys. With a consensual approach to energy poverty measurement, Pye et al. [28] suggest "a pragmatic approach" tailored according to each member state according to their best available data. Nevertheless, this attempt seems insufficient to deal with the lack of correlation among the definition of energy poverty, the clear identification of the users and the indicators to monitor it. To address this multiplicity, Rademaekers et al. [49] recommend using an array of indicators in combination, such as: "high share of energy expenditure in income: part of population with share of energy expenditure in income more than twice the national median (EPOV, 2010 HBS); hidden energy poverty: part of population whose absolute energy expenditure is below half the national median (EPOV, 2010); inability to keep home adequately warm: based on self-reported thermal discomfort (Eurostat, 2016 SILC); arrears on utility bills: based on households' self-reported inability to pay utility bills on time in the last 12 months (Eurostat, 2016 SILC)" [8]. This multi-dimension of indicators is enabled by the "Statistics on Income and Living Conditions" (Eurostat-SILC) that provides a set of proxy indicators used to compare energy poverty within the EU. A multi-indicators approach is also suggested by Herrero [48] who calls for methods that might capture as many variables as possible, not only in terms of domestic energy services measurement but also intersecting social, behavioral, demographic, etc. The author warns that a single-indicator might reduce the vastity and multi-disciplinarily of the issue and therefore of the possibility to tackle it. He urges the use of several indicators able "to capture the diversity of experiences and intensities" [48] of the issue.

These methodologies might be useful to frame energy poverty in a specific moment and place. Nevertheless, the phenomenon is subject to fluctuations and variations related to primary and secondary factors. Furthermore, data are referred either to the production sectors or to the type of energy carrier or source, such as, for example, the production or distribution of electricity, the need for electricity or heat. Moreover, the data may come from different sources: data collected and managed by institutional entities, authorities or other public or private research bodies, or open data. The different databases collect and return the information following their own criteria, in different format and aggregation. This kind of database on energy needs or consumption for the various sectors are not directly comparable with other databases, e.g., for the construction and real estate sector or the demographic sector. In other words, there is no direct correlation between energy needs and consumption and their identification and location at urban level.

For this reason, Rademaekers et al. [49] provide an additional set of indicators (demographic factors, energy prices, income, kind of household, heating system, supply choice, building efficiency and building stock, policy intervention) to be applied to cross-identify it.

This short review of the measurements methodologies shows a large effort in the quantification of the phenomena, in spite of a discursive and narrative approach, based more on quality. Despite showing a common denominator in the observation of the individual user, the approaches lack a narrative, explorative, and descriptive dimension and risk to exclude some categories and variables from the analysis. The user role remains underexplored; nevertheless, its key importance is widely mentioned, using different labels and evoking concepts. This reading key is addressed in the following chapter.

## *4.2. The User Dimension of Energy Poverty*

To tackle energy poverty implies a change of paradigm in the everyday lifestyles and a different configuration of the urban socio-technical systems, linked to energy as a social necessity [50]. These systems (infrastructural, electric, logistic, waste, etc.), in fact, require a complete review in the way they are organised, distributed, and managed. Both their technical features and their social roles and responsibilities [50] need to change in order to provide the basis for the support of the transition towards a more efficient and sustainable urban system. In the research on energy poverty, the key role of users has been largely recalled by literature. Energy saving linked to user behavior is deemed as significant as that deriving from technology [51]. Two main orientations towards the users are highlighted: target users as "vulnerable customers" and pro-active users indicated as "energy communities" and "energy citizenship".

The concept of vulnerable customers can include income levels, the available share of energy expenditure, energy efficiency of households, dependence on electrical equipment for health reasons, age, among other criteria. Pye et al. [28] categorized the different interpretation of the concept according to each Member State, highlighting the recurrence of a user as a receipt of social welfare, while in France, Sweden, and Italy, it reflects the relationship between low income and high expenditure. Recent studies include vulnerable users, low-income families with children, elderly people in nursing homes, rented persons, and those with an unstable employment situation [52,53]. Directive 2019/944 of 5 June 2019 concerning common rules for the internal market in electricity and amending Directive 2012/27/EU, in Article 28 identifies users as "vulnerable customers", stating that "member States shall take appropriate measures to protect customers and shall in particular ensure that vulnerable customers are adequately protected". Around this concept, a working group of the European Commission was created, following the 2015 Citizen Energy Forum: the Vulnerable Consumer Working Group (VCWG), in charge of exploring the potential for common approaches across the EU to vulnerability and energy poverty users' definitions and policies. This concept, however, seems to narrow the field of application to mere economic customers' communities, identifying them with economic/financial indicators and moving away from the concept of energy as a necessity and a human right [54], but rather towards energy as a commodity. An attempt to overcome this risk is to face the issue extending the idea of energy to the concept of citizenship meant as collective ownership and exercise of rights.

Designing a different role for the users, it is possible to imagine a proactive community, including vulnerable groups, that progressively adopts effective solutions, strategies, and management policies to face collectively the energy poverty issues.

The concept of energy citizenship [34,55–57] entails an "approach [that] argues for the social necessity of public engagement and participation in processes of policy-making and planning, driven by principles of local empowerment" [55]. This definition urges the citizens to be involved not only in a personal change of their behavior—by reducing their energy consumption expenditures (EU Directive 2018/2012)—but also in being agents and providers of solutions to address (part of) the wider energy poverty problem. In this view, users are called to "shape new routines and enact system change" [58]; therefore, they seem to be autonomously drawn to behave in order to improve their condition while at the same time set an example for a deep change toward a more efficient energy consumption and use. Ultimately, they should be politically influential [56], hence acting as social and political actors [34,35].

Many authors argued that this interpretation of "user as energy citizen" tends to "ignore crucial questions of unequal agency and access to resources" [50] identifying the citizens only as those who can purchase and spend in a market regime, causing exclusion issues and failing to include the same categories affected by energy poverty issues, and the vulnerable customers themselves. Bues and Gailing [59] stress the significative role of economic power dynamics in citizens' engagement, which limit the access to technology and economic/financial resources, as well as to knowledge, information and data. Lennon et al. [50] propose to "re-conceptualize energy citizenship, moving away from individualist and economistic perspectives and locating it within collective contexts of engagement". This approach seems to recall what has been commonly defined as energy communities, another evocative label used to frame not only the attention to the user, but its pro-active role in the management and improvement of sustainable behaviors around energy.

The energy communities are introduced in the Regulation (EU) 2018/1999 (11/12/2018), in particular in the art. 20 "Integrated Communications on renewable energy" as "the subjects to be promoted and facilitated for the self-consumption of renewable energy" (letter b paragraph 7). This definition followed the growing decentralization and distribution energy systems [33,60] as a precondition likely to produce a new category of actors engaged in the energy system with new active roles.

In the preamble of Regulation 2018/1999, it is pointed out that "[ ... ] member States should assess the number of households in energy poverty, taking into account the domestic energy services needed to ensure a basic standard of living in their national context, existing social policy and other relevant policies, as well as the Commission's indicative guidelines on related indicators, including geographical dispersion, which are based on a common approach to energy poverty".

It is clear that the attention to the user (as vulnerable customer) and its pro-active involvement as agent–energy community, energy citizenship—is key for tackling energy poverty issues in Europe. Nevertheless, some authors [34] highlight the lack of practical indication of measures or stress the limitation in these general labels. In fact, these approaches tend to overlook the qualitative identification criteria of such agents, their behavioral patterns, their possibility to access resources (in terms of knowledge, information, as well as financial ones) together with the connection with the characteristics of the buildings in which they live and the correlation with their demographic and financial situation. This opens up a reflection on accessibility, to information and knowledge. European cities and metropolis particularly benefit from the presence of knowledge structures, the density and easiness of connection and access to information and economic sources. The territorial reality in which the cities fit, however, presents phenomena of increasing social and political polarization, depopulation of internal territories and the removal of disadvantaged social classes in favor of a few categories that hold most of the knowledge capital. In this scenario, it appears that the majority of the poorer energy communities may be left out from the decision-making epicenters, as well as from the possibility of political participation and representation.

It is widely known that energy poverty emerges as determined by three factors: income, energy cost, and building characteristics. Following the above-mentioned strategies that focus on the role of the user as consumer and producers of solutions, the challenge is to adopt a new design approach, able to combine spatial issues with new services, new enabling technologies, and new supportive management tools.

## *4.3. Mitigation Strategies from Literature*

Participation is one of the ways in which energy poverty can be mitigated. Some studies indeed focus on the role of community participation in improving the knowledge on energy and on energy services. As an example, Martiskainen et al. [61,62] describe the organization of Energy Cafés, a format for letting communities engaging with energy locally. In particular, the community organize pop-up temporary events where information and advice on energy are provided to other persons. As assessed in the conclusion of the paper, "Energy Cafés open up for various forms of advocacy, highlighting a broken link between the expectations of the energy markets and energy practices in the home" [62].

Participation is also intended as the involvement of other actors besides end-users (which usually are residents). Several studies [57,63] focus in fact on the benefits of involving in a coordinated analysis and problem-solving other categories of persons, such as owners and tenants, social workers, and healthcare practitioners.

Policy-related papers (16% of the total) are mainly focused on the analysis of current policies in different European countries. Especially addressed to analyze the UK situation, some others compare different countries among them.

Profiling-related papers (12%) are of particular interest as they recognize how energy poverty can be different in relation with different people living in a household. In fact, there are several categories of people potentially more exposed to energy poverty, but also energy poverty can affect them differently (also with different health effects). In relation with the literature review, actually, the categories of persons more studied in relation with energy poverty are students, elders and young. However, up to now, it seems that research is concentrated in identifying the effects of fuel poverty on them instead of defining solutions specifically designed for these categories. This last point doesn't seem to be present in most of the papers, but it constitutes an interesting topic to be further developed.

From a technical point of view, 10% of analyzed papers focus on technical and technological solutions to fuel and energy poverty. Among them, most are based on finding solutions to produce more energy from renewable sources [64–68]. In particular, Donaldson et al. [68] propose the use of urban brownfields for producing energy targeted to poor households. Some research is then focused on the implementation of storage systems and lithium batteries together with energy production [64,65]. Very few papers focus on retrofitting solutions as a way to reduce and mitigate energy poverty. Often, researchers argue that deep retrofitting solutions are costly and they usually need a temporary relocation of tenants. However, as in Aranda et al. [69], some solutions are proposed, such as working on the building envelopes, on lighting systems and on energy generation. Nevertheless, less disruptive and low-cost retrofitting solutions (also not linked to solving energy poverty) have been studied by architectural and engineering researchers for years, but they do not seem to constitute the unique response to fuel poverty because, as many authors argue, fuel and energy poverty also involves social, political, and economic concerns [70–74].

#### **5. Discussion and Conclusions**

The systematic literature review showed how currently it is possible to recognize some red lines in which the European research on fuel and energy poverty is mostly aligned.

At first, most research is still investigating the knowledge about the topic, in terms of definitions, semantic analysis, and theories, but also in terms of collecting and analyzing data. For the matter of data collection, it seems that the current research is mostly aligned to merge quantitative data coming from official sources (local, national, or EU) with qualitative data coming from different forms of participatory approaches (from interviews, focus groups, and observation).

A high amount of effort is put into defining the most suitable indexes and metrics to describe the phenomenon. In terms of technical solutions, most of the literature is aligned in working mainly on RES, storage, and less in retrofitting solutions. However, the role of technical solutions doesn't seem to be enough to mitigate the problem, as poverty has deep social, political, and economic roots that need to be understood.

For that reason, several studies are now proposing the role of urban actors and solidarity chains at the neighborhood level as a way to complement, with soft solutions, the harder component of retrofitting.

Most studies, then, are concentrated in the United Kingdom, as one of the first countries highly engaged in addressing the issue, even if some other countries are showing an increasing amount of attention, especially Spain and, in general, the Southern European countries.

Despite the presence of a high interest in the topic, some barriers and limitations can still be found, as well as potentially interesting new lines of research.

The limitations and barriers have been distinguished in three categories: economic, social, and technic/technological.

Concerning the economic issues, the high cost of the energy mitigation measures of the single or block of buildings is still one of the major challenges. As, for example, indicated in the Report on the state of energy poverty in Italy (2018), climate change affects household energy demand, exerting upward pressure on energy prices, whose incidence on the total rose from 4.7 per cent in 2007 to 5.1 per cent in 2017. The percentage incidence of energy expenditure is higher for less well-off households (those with a lower than average expenditure). The condition of these households has worsened over the last decade: in 2007, 20% of the least affluent households spent about 6% of total spending on lighting, heating, cooking, and cooling domestic environments, and, ten years later, this share increased by about half a percentage point (while it was stable or even decreasing for other households). In fact, electricity prices have long been burdened by systemic charges linked to the support of renewable energy and the mechanism for pricing emissions within the EU; in the future, the same will happen for other energy sources that will be subject either to some form of carbon tax or to other forms of restrictions on their use (such as the use of coal in electricity generation).

According to many predictions [22,45], European cost of electricity will rise further in the coming decades, due to the energy transition, the prosecution of the economic crisis and a serious de-carbonization struggle that will lead to a further increase in energy prices.

A second economic issue regards the difficulties in systemically eradicating the issue with subsidies and economic support measure. As pointed out by Bouzarovski [54], in fact, subsidies and measures that avoid disruption as a result of repeated arrear payments temporarily alleviate the energy poverty of households, but do not solve the problem at all.

The social aspects concern understanding and awareness, understood as self-awareness or understanding of the phenomenon of the population regarding energy and energy efficiency. Certain conditions such as energy poverty are not understood as emergency conditions by the population. While people who are in the condition of Absolute Poverty are identified and aware and have access to social services in the territory (Municipality, Caritas, etc.), the condition of Energy Poverty is not recognized.

Many financial schemes and social benefits have been developed to support the installation of measures that reduce energy consumption [36], but they are not well exploited, as low-income households cannot save the necessary funds to cover their initial expenses and generally have difficulty in obtaining a loan.

The condition of Energy Poverty involves a series of actions by subjects that can be "measured" in social/sociological terms, such as:


Indicators of social aspects should be oriented to the identification of the above-mentioned conditions. Nevertheless, vulnerable groups of citizens are not easily categorized because their individual needs, knowledge, culture, or living conditions cannot always be simply labelled. Therefore, knowledge transfer on energy use and energy efficiency advice should be targeted to people's specific social situations. The variety of stress factors and living conditions makes knowledge transfer difficult.

The technical/technological limitations do not specifically concern knowledge or development limitations, but rather expensive applications of them. Household efficiency measures are particularly relevant for reducing energy poverty, but much can also be done in the direction of energy savings in other areas. Low cost solutions [8] might be a temporary answer to tackle emergencies (e.g., monitoring systems, room control temperature, devices for the production of DHW—Domestic Hot Water, selection of energy-efficient installations and equipment, electrical storage batteries). Additionally, as the major signs of energy poverty within people are difficultly detectable (e.g., low indoor temperature), it seems that the current progressive transition to smart cities, and the use of digital connected devices in households—such as smart thermostats—can be of great help for gaining a deeper insight also on the energy poverty topic.

Among the solutions proposed to support the increase of energy efficiency in housing are related to:


Another interesting aspect, emerging from the literature, is the necessity to link data with the spatial dimension by using the technologies of BIM and GIS. The use of such software can in fact be helpful in understanding the urban dynamics on energy poverty and, thus, in finding specific solutions to each context. However, it is necessary to understand the availability of qualitative data from the public and energy providers and to boost their transition to open source aggregated datasets.

In conclusion, it seems that the problem needs an integrated and holistic approach, able to combine the technical issues about building performances, enabling technologies, with a deeper understanding of social connections, economic conditions (also in terms of investigating why people are in economic difficulties). A complete understanding of people behavior and awareness can help in the characterization of profiles to face the several challenges connected to energy poverty.

Behaviors and local cultural factors can drive basic energy use practices [75]: the factors and their relationships that influence consumer behavior and practices are dynamic, highly dependent on human elements, change over time, and influence consumer behavior. For these regions, the process of consumption practices becomes unpredictable and detached from any recognizable pattern [76]. Investigating the size and value of end-user profiles and behaviors before the design phase [77] helps to identify the best combination of technical and social measures, through the adoption of user-friendly energy efficiency systems that could reduce the energy poverty, facilitate the use by tenants and increase their environmental and energy awareness.

The design challenge is to contribute to the necessary framework made of technical and social solutions, policies for energy justice, and participatory processes to understand the energy equity dimension in terms of accessibility and affordability [39].

## *Limitations and Further Research*

The paper has some limitations due to its choices and methodology. At first, the decision to exclude non-English written papers resulted, obviously, in a limitation in the number of analyzed findings. Especially for what concern locally specific case studies, some interesting approaches were probably not included into the analysis for that reason. However, we believe that the general analysis and the major findings of the paper have not been penalized by this choice.

The paper identified some interesting trends that could be deepened in future research and, in particular, the following:


**Author Contributions:** Conceptualization, A.B. and V.G.; methodology, A.B. and V.G.; formal analysis, S.O.M.B. and M.M.; investigation, S.O.M.B. and M.M.; resources, S.O.M.B. and M.M.; data curation, S.O.M.B. and V.G.; writing—original draft preparation, S.O.M.B. and M.M.; writing—review and editing, A.B. and V.G.; visualization, S.O.M.B.; supervision, A.B. and V.G.; project administration, A.B and S.O.M.B.; funding acquisition, A.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Review* **Outdoor Wellbeing and Quality of Life: A Scientific Literature Review on Thermal Comfort**

## **Ernesto Antonini \*, Vincenzo Vodola, Jacopo Gaspari and Michaela De Giglio**

Department di Architecture—DA, University of Bologna, Viale Risorgimento 2, 40136 Bologna, Italy; vincenzo.vodola2@unibo.it (V.V.); jacopo.gaspari@unibo.it (J.G.); michaela.degiglio@unibo.it (M.D.G.)

**\*** Correspondence: ernesto.antonini@unibo.it

Received: 17 March 2020; Accepted: 9 April 2020; Published: 21 April 2020

**Abstract:** While indoor comfort represents a widely investigated research topic with relation to sustainable development and energy-demand reduction in the built environment, outdoor comfort remains an open field of study, especially with reference to the impacts of climate change and the quality of life for inhabitants, particularly in urban contexts. Despite the relevant efforts spent in the last few decades to advance the understanding of phenomena and the knowledge in this specific field, which obtained much evidence for the topic's relevance, a comprehensive picture of the studies, as well as a classification of the interconnected subjects and outcomes, is still lacking. This paper reports the outcomes of a literature review aimed at screening the available resources dealing with outdoor thermal comfort, in order to provide a state-of-the-art review that identifies the main topics focused by the researchers, as well as the barriers in defining suitable indexes for assessing thermal comfort in outdoor environments. Although several accurate models and software are available to quantify outdoor human comfort, the evocated state of mind of the final user still remains at the core of this uncertain process.

**Keywords:** outdoor thermal comfort; human thermal perception; thermal comfort assessment; quality of life

## **1. Introduction**

## *1.1. Review Contest and Boundaries*

While indoor conditions have been the main concern for research on user comfort since the second half of 20th century, assessing outdoor comfort has emerged as a challenging field during the last few decades. Three main phenomena have pushed towards this change:


Scientists worldwide have thus focused their attention on this topic, making available a wide range of tools and methods to assess human thermal outdoor comfort in different climatic contexts. Over 100 biometeorological and thermal stress indexes [10] have been developed, adopting different approaches and rationales, aiming at linking the microclimatic conditions to the perceived sensations.

Since the available knowledge on human thermal perception and related evaluation protocols were mainly the ones previously developed for interior spaces and other confined spaces, the assessment of outdoor conditions initially refers to these patterns [11].

In fact, human thermal comfort and its assessment were studied since the beginning of the 20th Century, when the first simplified models were developed [12]. The two node model applied thermodynamics principles to energy exchanges between the human body and its thermal environment [13] for the first time during the 1930s, but it is only from the 1960s onwards that researchers were able to analyze the main climatic parameters connected to the perception of thermal comfort (e.g., air temperature, radiant temperature, air humidity, air flow velocity) when the first climate chambers were made available [14]. The cornerstone studies of Givoni [15] and Fanger [16] led in the following years to the identification of new parameters that are currently considered essential elements in the contemporary assessment of thermal comfort. The advances in the physics of heat exchange knowledge gained during the 1980s and the increasing availability of computer tools to support the research activity allowed relevant progress on the understanding of the human thermal environment [17–22] and the formulation of indexes based on body heat exchange [11].

In order to model human thermal comfort in outdoor environments, solar radiation was first added to the set of climatic variables in use for indoor spaces [23,24]. Olgyay assumed that solar radiation must be combined to the effects of other climatic elements, to draft a "bioclimatic chart" for the outdoor conditions [23].

Further studies have shown that outdoor thermal comfort is a more complex notion and a multilayered condition, which is very difficult to properly describe as a whole by considering biometeorological factors only [25]. Although the thermal state appears as very influential among the many factors shaping the quality of outdoor spaces, a wide range of additional social and physical aspects, however, were identified as relevant, especially those linked to behavioral variables [26].

Nonetheless, the issue remains open, especially regarding the assessment of the human variables influenced—including cultural, behavioral and psychological factors—on the perception of the environment's physical conditions [10].

The efforts spent in the last few decades to advance the understanding of these phenomena provide evidence of the topic's relevance, even if a comprehensive picture of available studies is still lacking, as well as a classification of the interconnected subjects and outcomes. A systemic overview of the available knowledge could be therefore a useful tool for identifying the different research trends and classifying their objectives, approaches, results and implications.

This paper reports the outcome of a literature review aimed at screening the available resources dealing with outdoor thermal comfort, in order to provide a state of the art that identifies the main topics focused on by the researchers, as well as the barriers in defining suitable indexes and approaches for thermal comfort assessment in outdoor environments.

## *1.2. Theoretical Background*

The International Organization for Standardization (ISO) has released a series of international regulations for the evaluation of thermal comfort; ISO 13731:2001 defines physical quantities and provides a reference for terminology and symbols to adopt for standards on ergonomics of the thermal environment [27], while ISO 7726:1998 identifies the means and instruments for measuring the physical quantities involved [28]. ISO 7730:2005 provides an analytical determination and interpretation of thermal comfort using calculation of the Predictive Mean Vote (PMV) and Percentage of Person Dissatisfied (PPD) indexes and local collected data [29]. Although the model was developed for indoor comfort assessment, it may be adapted for outdoor spaces by adding the radiative exchange values [30]. This ISO Standard also includes annexes providing comprehensive databases for the metabolic rates of different human activities and the thermal insulation values of clothing ensembles. Moreover, ISO 7243:2017 enables the estimation of the worker heat stress by Wet Bulb Globe Temperature Index (WBGT) [31], which can be applied for both indoor and outdoor work environments.

In addition to ISO standards, other regulations such as American National Standards Institute (ANSI) / American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) Standard 55-2017 and European Standards (EN) 15251-2017 specify calculation and evaluation methods for thermal comfort. However, they are mostly addressed to indoor environments, and focused on those parameters affecting the energy performance of buildings [32].

The awareness of the topic's broad latitude has prompted some authors to focus on the possible gaps in criteria and assumptions currently in use to map the factors influencing the perceived outdoor comfort and wellbeing sensation, starting from the definition of thermal comfort itself.

ASHRAE defines thermal comfort as a state of mind that expresses satisfaction with the thermal surroundings [33], which means that human thermal comfort refers to a subjective sensation, different from one subject to another [34]. Some studies argue that this definition may appear rather vague [10]: it does not specify what that state of mind is (in terms of perception, feeling, etc.) and it does not provide any indication of how to relate this mental state into something that can be measured, nor which variables could be involved [35]. Thus, this is still an open issue from different points of view, although the definition is intended to be the most general as possible, to provide a common understanding, thus leaving each study the responsibility to state the assumptions (and limitations) in their own premises.

Additionally, when the assessment of this mental state has to be investigated, referring to the outdoor environment, the relationship between the human body and a large set of spatial and temporal variables must be also considered. Theoretically, this gap could be filled by adapting for the outdoor comfort assessment the same methodologies and indexes developed to evaluate indoor comfort. However, several authors discussed this position as unsuitable, arguing that the theoretical models developed for describing thermoregulation functions within the indoor environment are not adequate to feature the outdoor thermal comfort conditions [34,36]. This is mainly because of the outdoor environment's greater complexity, and its temporal and spatial variability [34]. Thus, the need is acknowledged for empirical data from field surveys on the subjective human perception of outdoor wellbeing, which should enable investigation of thermal comfort in open spaces from a broader and more realistic perspective [34].

In order to make the reading easier, Table 1 provides the nomenclature of the main terms and acronyms reported in the paper, as well as Table 2, which summarizes the main thermal comfort indexes.




**Table 2.** Nomenclature of the main thermal comfort indexes cited.

## **2. Methodology**

Although overall human comfort involves several environmental agents acting simultaneously, including air quality and thermal, acoustic, and lighting factors [34,37], this study was limited to the thermal wellbeing, since it plays a crucial role in affecting the comfort perception in outdoor environments.

The review was based on a systematic search for peer-reviewed papers published within the last twenty years. The aim of the review was to identify the most relevant trends in studies dealing with outdoor human thermal comfort. Five main search engines were used: Science Direct, Google Scholar, Scopus, Web of Science (WOS) and Researchgate. The following keywords were used for the preliminary retrieval of papers from the sources:


This allowed a first selection based on the paper title and abstract. Additionally, papers referenced within the selected articles were considered as secondary sources, thus embedded in the second step of the review process.

#### *Review Process and Outcomes*

The outcome of the first search round provided more than 25,000 results, of which 1059 were retrieved from the Science Direct database, 16,600 from the Google Scholar search engine, 710 from the Scopus database, 594 from Web of Science and 6160 from Researchgate. Duplications were deleted in a second step of the process and the results were also filtered using a combination of the proposed keywords. The results distribution from Science Direct, reported in Figure 1, indicates a growing interest in the second decade that can be certainly associated to new drivers, such as the effects of climate change and related heat waves, but also to the development of web-based solutions to share knowledge and studies that facilitated the communication and exchange among the scientific community. Furthermore, it must be noted that the increasing demand for scientific publications on the topic within the academic circuits, for both research purposes and career advancement, may have influenced the numeric growth of studies, as well as their availability in scientific journals.

**Figure 1.** Results retrieved from the ScienceDirect search engine (keys: publication after 1999; text string "outdoor thermal comfort" in paper abstract and/or title).

The selection results were then refined according to the predefined set of keywords; a total of 855 sources were found at the end of the filtering process, including journal articles, book chapters, reviews and peer-reviewed conference papers. The sources matching with three or more keywords (out of six) were considered as having highly relevant contents. They were then shortlisted and their full text downloaded; 146 significant outputs were identified by selecting those focusing on the relationships between outdoor thermal comfort and microclimatic variables.

The sources referenced by the selected papers were also explored, thus increasing to 236 the final number of the surveyed articles. Figure 2 displays the incidence per year (Figure 2a) and the breakdown by issues addressed (Figure 2b).

**Figure 2.** Distribution by year (**a**) and by topic (**b**) of the final 236 sources considered for the review purposes.

*Energies* **2020**, *13*, 2079

The analysis of the 236 final resources, as shown in Figure 2b, pointed out that studies on outdoor comfort could be divided into four main groups:


Accordingly, this review adopts the same structure, reporting the findings grouped in three sections, corresponding to the main topics to which scientific efforts were devoted toward identifying both effective methodologies and main limitations for the full comprehension of the subject. In addition, a brief introduction to standards and regulations is provided in the theoretical background paragraph of the introduction.

From the literature analysis, field studies on outdoor thermal comfort were carried out around the world in the last 20 years. Givoni et al. discussed methodological issues and deepens problems in outdoor comfort research based in Japan and Israel [164], as well as in China [74], where other studies were also conducted [51,54,57]. Other studies were performed with reference to the following geographical areas: Canada [207], Argentina [138], Sweden [147,175,218], Portugal [131], United Kingdom [187,209], Italy [214], Morocco [132], Emirates [227], Egypt [118], Malaysia [192], Bangladesh [141], Australia [142], New Zealand [208]. Among the largest research projects, the most extensive was RUROS: Rediscovering the Urban Realm and Open Spaces [37], which included field surveys carried out in seven European cities: Athens, Thessaloniki, Milan, Fribourg, Kassel, Cambridge, Sheffield.

## **3. Results**

## *3.1. Mathematical Models and Indexes*

Several complex thermal indexes were developed to date, describing and quantifying the thermal environment of humans and the energy fluxes between the human body and the surrounding environment. De Freitas and Grigorieva [79,80] carried out a three-stage study, providing a comprehensive register of 165 indexes suitable for the purpose, which were subsequently grouped and classified. Thus, they observed that indexes are almost designed for a specific application, so the choice depends on the context in which the index is used, as well as on the availability of the data needed to quantify it. They also found that the best performing indexes are those based on the body/atmosphere energy balance, which, however, are those needing more complicated calculation routines and more detailed input data. What emerges as an additional and more serious drawback is that body–atmosphere energy balance indexes are often based on numerical models that were not validated. This leads to the conclusions that there is not an overall best index [79] and the use of a standardized human body could introduce errors, since the characteristics of the human body vary individually [80].

A review about models and standards is provided by Coccolo et al. [157] who analyzed a number of outdoor human comfort models and the related physical variables as well as the applicability with reference to the climate, with reference to the research goals, dividing the models in the following three groups:


A selection of case studies about the different models since 2000 is organized in a graph according to the Köppen climate classification by Coccolo et al [157]. Temperate climate emerges as the most studied condition, followed by those referring to arid, cold and tropical climates, while very few studies were performed for polar environments. The outcomes of this research leads to the following conclusions:


From the literature analysis, it emerges that PMV and PET are the most widely used indexes [49,50,59,61,62,67,71,75,77,80,85,86,88,89,92,93,95,98,109,110,112,114,118,128,131,133,138,141, 142,148,151,154]. However, PMV was elaborated for indoor environments by observing people sitting in climate chambers; it does not consider the dynamic adaptive response of the human body and instead references a punctual static situation. Thus, its use in outdoor environments may often give misleading results. Cheng et al. [74] clearly demonstrate that PMV generally overestimates the thermal sensation in summer and underestimates in winter.

Developed for the outdoor environment, PET is a procedure often adopted, with results that are well correlated with onsite monitoring and questionnaires [49,77,89,114,131,133,141,142,148]. The PET index can be used as an alternative to PMV; however, the main drawback is that it can underestimate the effect of latent heat fluxes and overestimate radiant heat flows [74]. By comparing the outdoor thermal environment and the thermal sensation of pedestrians in two different Chinese cities, and through the thermal unacceptability percentage and PET indexes, Yang et al. [54] observed that the thermal unacceptability expressed by people was different according to the city, despite the similar outdoor thermal conditions.

In arid climates, ITS results show high correlation with field studies [196,200], while SET [52,142,191] and OUT SET\* [133,148] appear to be more applied in and reliable for temperate climates. Being more scientifically updated, UTCI is instead the only index that was applied to all climates [59,84,99], ensuring a good matching between onsite measures and simulations [70,133,157]. PE and WCI, describing thermal sensations from comfort to extreme cold stress are preferable in cold climates. ITS, H, HI and WGBT make available detailed thermal scales for hot sensations, often neglecting the cold ones [157].

From the research carried out so far in order to understand the primary models used, it seems that there are apparently a large number of indexes available for the assessment of outdoor comfort, despite each of them presenting some drawback with different levels of error or approximation.

Although energy balance based indexes are widely used, the main limitation lies in the steady-state condition that does not fully reflect people rarely experience thermal equilibrium in outdoor environments [100].

A positive note is that the scientific community seems to be aware of this limit and much effort was spent to understand the human adaptation capacity as well [25].

## *3.2. Human Thermal Perception and Thermal Adaptation*

The term adaptation can be broadly defined as the gradual decrease of the organism's response under an iterated exposure to a stimulus, thanks to the effects of all the actions deployed by the subject to make it better suited to survive in such an environment. In the context of thermal comfort, this can involve all the processes that people perform to reduce the gap between the environmental conditions and their requirements [194]. In other words, whenever metabolic activity or environmental conditions change, the human body tends to adapt itself to those changes. According to Nikolopoulou and Steemers [194], the adaptation basically occurs in three different ways:


A large number of studies were done in the last twenty years that aimed to incorporate the human dimensions into comfort assessment methods, performed both through climate chambers and by direct field surveys. Some of these studies investigated the possible adaptation from a thermophysiological perspective [95,174], others focused on the parameters that determine the human perception of comfort [81,142,194,208].

Some links between human thermoregulation mechanisms and thermal environment conditions were established by running tests in climate chambers [149,195]. However, whether these results can be transferred to the behavior of people in external environments is still an open question, since all the aspects that influence adaptation actions in real contexts are highly complex to reproduce (providing wide temperature ranges, checking for human physical and behavioral changes, measuring temperatures of people' skin and core to evaluate thermoregulation features etc.).

Thus, the assessment of the human thermal sensation must consider the environmental stimuli, which are dynamic and perceived subjectively. The stimuli are dynamic due to the human adaptation to external climatic conditions, which is a progressive process influenced by various adaptive factors. It is subjectively perceived, since the human perception of thermal comfort is not always nor univocally dependent only on objective biometeorological conditions. This means that the individual attitude towards outdoor space is not only determined by the state of the body, but also by the state of the mind. This suggests that the ideal framework for thermal comfort assessment should work on at least four levels: physical, physiological, psychological and social/behavioral [7].

Therefore, by considering simultaneously all the factors (whether objective or subjective) influencing human thermal perception, it is possible to obtain an evaluation of outdoor comfort as coincident as possible with reality. Each of these factors can be estimated or calculated through different approaches (measurements, modeling, field interviews and observations). Thus, working on the four levels of evaluation of human outdoor thermal comfort, it should make it possible to connect the external microclimatic conditions with the perception of people who use a certain space at a specific temporal moment.

In other words, this framework should allow a linkage of "climatic knowledge" with "human knowledge" [7].

"Neuroarchitecture" seems to be opening a new research field that is able to drive some advances in this direction, combining neuroscience and architecture to better understand how space is perceived by the human brain [159,160,165]. The outcome from neuroarchitecture studies could thus enhance the effectiveness of the design of the built environment, providing a better knowledge of the relationship between the humans and their spatial wellbeing [157].

Coburn et al. [202] investigated how neuroarchitecture could mature into an experimental science by outlining the related challenges ahead and identifying the priority need for a specific framework to guide research. To date, however, relatively little work has been done on architecture neuroscience, and further studies are needed.

Therefore, the final suggestion is to apply a multidisciplinary approach, including both studies on physical phenomena and human psychology. For this reason, it is highly recommended that shared principles and definitions be included in the common framework.

## *3.3. Software and Predictive Tools*

Givoni et al. [164] addressed the need for prediction tools in order to support designers in understanding the effect of a change in a climatic element that influences people's outdoor comfort. In fact, the availability of simulation and scenario-testing tools within an assessment framework is crucial, as they provide a platform for both the integration of knowledge from various perspectives and the comparisons of different design options.

Currently, tools for simulating virtual scenarios are becoming increasingly available and updated, allowing reproduction of even complex environmental contexts. The literature review performed identified the most used software: ENVI-Met, RayMan, SOLWEIG and the UTCI calculator (Figure 3). While ENVI-Met is based on computational fluid dynamics (CFD) and thermodynamics, RayMan and SOLWEIG are basically 3D radiation models.

**Figure 3.** The most used software and tools outlined from the literature review.

ENVI-Met is a tool to simulate outdoor space microclimate by quantifying energy and mass exchanges, wind turbulence, vegetation effect on the outdoor conditions, bioclimatology data and pollution scattering. It is based on four interrelated systems: soil, vegetation, atmosphere and buildings. Outdoor microclimate is described by air temperature, Mean Radiant Temperature (MRT), wind speed and direction, short- and long-wave radiation from a single building to an entire city [236]. It was used by Acero et al. to evaluate the differences in thermal comfort comparison models and onsite measures in four different locations [58]. The study points out that some deviations may occur in the ENVI-Met output; however, the tool provides useful and quite reliable outcomes (e.g., comparison of urban planning scenarios during typical meteorological conditions). Nevertheless, limitations must be clearly outlined in order to avoid misleading results. ENVI-Met is often used while interacting with other tools and plugins. Through the postprocessing tool called BioMet, it is possible to determine thermal comfort according to PMV, PET, UTCI [215] and MRT. Additionally, thanks to a generative algorithm called ENVI-BUG Software, it is possible to combine ENVI-Met, Rhinoceros, Grasshopper and LadyBug. Fabbri et al. effectively adapted it to obtain a 3D output, achieving a simplified method for displaying results and making them easier to read for nonexpert users [230].

Developed at the University of Freiburg, RayMan is a diagnostic microscale radiation model able to calculate radiation fluxes and thermophysiologically indexes, such as PMV, PET, SET\* [222], UTCI, PT [220] and MRT. It is mainly used to compare the effect of multiple planning scenarios in different situations from micro to regional scales [220]. It allows the use of several input data such fish-eye pictures or obstacle files to obtain additional outputs like shade and sunshine duration, as well as the possibility to run long-term data sets. The major drawback is that the model cannot calculate air temperature, air humidity and calculate or adjust wind speed. These gaps are often bridged by preparing the data in the input files or running simulations with different wind speeds [222].

SOLWEIG (SOlar and LongWave Environmental Irradiance Geometry) enables quantification of PET, UTCI and MRT within complex urban settings as described by Lindberg et al. [218]. It applies the theory of radiative fluxes and mean radiant temperature, the main limitation is that it takes only building geometry into account, while vegetation is not considered when mean radiant temperature is calculated.

The UTCI calculator allows determination of a pedestrian's thermal comfort according to the Universal Thermal Climate Index [117]. Abdel-Ghany et al. demonstrated its application in combination with RayMan to evaluate UTCI index in arid climatic conditions [128], concluding that the model can be used successfully in arid environments to evaluate the thermal sensation, with the heat stress outcome very close to the PET index.

These environmental modeling tools can provide a better understanding of climatic conditions and a mean to effectively assess human thermal comfort outdoors, helping town planners and decision makers to compare and test several design alternatives in terms of attractiveness and effectiveness [232]. In addition, the development of such tools and software can solve the limitation of the different methodologies used in research.

#### **4. Discussion**

The literature review in this paper lists the main available resources dealing with outdoor thermal comfort issues. Its aims to provide a state-of-the-art review to identify the main topics on which current research focuses, and what the main barriers are that limit the identification of successful indexes for the assessment of thermal comfort in the outdoor environment, especially in urban contexts [150,170].

The main outcome is that thermal comfort in the outdoor environment is a complex issue with multiple layers and that the human state of the mind plays a key role in influencing peoples' perception on space and its utilization.

The literature review performed points out that many indexes and approaches have been developed, however they are specifically addressed to meet particular contexts with relation to specific variables. A relevant drawback deals with the limited consideration of the dynamic adaptive response of the human body, since the most frequently adopted indexes focus on the energy balance between the human body and the environment. More recent indexes seem to pay more attention to human perception and behavior, but their application is still limited and therefore a discussion on the potential outcomes is still hard and may be somehow misleading. The main suggestion is to focus on a limited and possibly shared number of procedures that adequately consider the human thermal perception and thermal adaptation.

It must be noted, despite some attempts to integrate different disciplines on the same topic, that most studies are organized by adopting a silos approach.

Moreover, many studies reveal that different people experience the environment in a different way. The human response to a physical stimulus is not in direct relationship to its intensity, but depends on the "information" that people have for a particular situation, and on the associated psychological factors influencing the thermal perception of a space and the changes occurring in it [194]. If physiological acclimatization is not sufficient to meet a comfort status, physical adaptation will be introduced to adjust oneself to the environment or alter the environment to his needs (such as altering clothing levels, modifying posture and position, or even changing metabolic heat with the consumption of hot or cool drinks).

Since the agents acting on this mechanism belong to at least four different but interconnected patterns (physical, psychological, physiological and social/behavioral), the ideal framework for thermal comfort assessment should work on all of them.

No effective methods are available today that include the human dependent factors within the outdoor thermal comfort models. However, the scientific community seems to become increasingly aware of the importance of the psychological and social/behavioral factors, as indicated by the number of recent studies exploring these fields. Integrating the physical energy balance and the human variables would allow for creation of more complete and thus more effective models, drastically improving the reliability of their results.

Despite the complexity of the above interrelations, these topics should be approached at design level. More effective simulation and modeling tools could be developed within that framework, providing designers and decision makers with a means to better achieve the thermal outdoor comfort target by integrating the factors related to climatic conditions and those belonging to the people's sensitivity to environmental stimuli; these tools could be wisely adopted in urban design. In this way, shopkeepers could be the first group to realize the benefit of such cool oases in a hot environment, and finally it would be possible to extend the positive impacts not only to the environmental domain of cities but also to the economic domain.

The highly advisable trend sketched by this scenario cannot hide the fact that shortcomings are still evident in research, concerning both the tools and the goals. Concerning tools, the systematization of all knowledge belonging to physical, physiological and psychological studies seems to be the only way forward today for the identification of an effective and shared method to evaluate the achievement of thermal comfort in outdoor spaces. Regarding goals, deeper interdisciplinary studies are needed to support with evidence the assumption that the more intense use of outdoor space will benefit the economic and social life of the city.

## **5. Conclusions**

The literature review performed points out that many different interrelated issues drive the research on outdoor thermal comfort. The topic emerges as worthy to be further investigated, especially in relation to social and behavioral implications, requiring to possibly adopt a transdisciplinary approach. Despite the effort spent to create accurate models and software to properly assess outdoor human comfort, the evocated state of mind of the final user still remains at the core of the uncertain process. The field studies, including surveys and tests with different categories of users, highlight the need to refine the available indexes to better reflect the real perception of the human body in different conditions.

Therefore, each user has to clearly understand the basic equations of the software or tool chosen, in order to select the one that best suits the needs for the research purpose and the application

context. This often generates some uncertainty that makes a comparative approach among the different outcomes more difficult.

Some relevant shortcomings can be currently detected and listed as follows, with the aim to address and prioritize the research efforts for further improving an understanding of outdoor human comfort.

Assuming that an effective assessment requires consideration of physical, physiological, psychological and behavioral levels, the detected lack of shared principles and definitions within a common framework reduces the possibility to take into account the multiple and interconnected nature of phenomena. Accordingly, the definition of a comprehensive and stable framework represents a top priority:


The use of reliable predictive and simulation tools will support decision makers, designers and planners to better realize the potential impacts of their decision and strategies when transforming the built environment. Finally, a multidisciplinary approach, including studies on physical phenomena, human psychology and architecture, is highly recommended, assuming people's wellbeing is the ultimate goal.

**Author Contributions:** Conceptualization, E.A. and J.G.; methodology, V.V., E.A. and J.G.; formal analysis, V.V.; investigation, V.V.; resources, M.D.G. and V.V.; writing—original draft preparation, V.V.; writing—review and editing, E.A. and J.G.; visualization, V.V.; supervision, E.A. and J.G. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

## **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Review*
