Study of globe temperature relative to air temperature during cognitive activities in information technology laboratories

Abstract

BACKGROUND:

Considering that, environments with information and communication technology innovations, including educational institutions, are providing more interaction among individuals anywhere in the world and contributing to higher learning flexibility, it is necessary to pay extra attention to the radiation dissipated by technological equipment in these environments.

OBJECTIVE:

Investigate whether the behavior of the globe temperature (t_g) in relation to the air temperature (t_a) could affect the performance of students in information technology laboratories (ITLs).

METHODS:

The methodological procedures adopted consisted of the following analyses in six institutions: thermal variables - mean radiant temperature (t_rm) and (t_g-t_a); students’ performance and architectural elements.

RESULTS:

ITL G was the ITL with the highest incidence of thermal radiation, thus a mathematical model was proposed for this sample to determine whether (t_g - t_a) and t_rm are related to overall student performance (D_t). For each increase of one degree in the difference between the globe temperature and the air temperature (t_g-t_a), the students’ performance in the institution G decreased by approximately 29%.

CONCLUSION:

As well as productivity can be altered due to changes in air temperature in air-conditioned teaching environments, in this specific case, if t_g> >t_a, possibly the thermal radiation may interfere with the performance of the people present in the environment technological innovations of communication and information.

Keywords

Thermal radiation globe temperature mean radiant temperature student performance

1. Introduction

Thermal comfort is closely related to the energy consumption in facilities, the productivity of the occupants, and, in the case of schools, student performance and learning [1]. Therefore, the thermal comfort inside classrooms, laboratories, and other learning environments gains significance and is a variable that greatly influences student performance [2]. A study on environmental ergonomics finds evidence that favorable thermal conditions in a classroom promote more comfort, greater safety and better performance. These conditions are presented in the International Standards Organization (ISO) 7730: Ergonomics of the thermal environment [3]. Among environmental comfort variables, temperature and classroom lighting have the greatest influence on students’ learning and are the most common reasons for student complaints [4, 5].

However, the thermal comfort in the heat conservation equation is the result of six parameters: temperature, air humidity, thermal radiation, relative air velocity, personal activities, and clothing [6]. It is emphasized that merging environmental ergonomics with architecture engenders multidisciplinary knowledge where ergonomic aspects are incorporated into the design process in the interest of creating environments that ensure greater attractiveness and comfort [7].

Many researchers have been attracted to this field in recent years because of the growing public discussion about global climate change [8]. According to the report on climate change of the Brazilian Ministry of Science and Technology (2013), it is expected that the mean air temperature in Brazil will be 3 to 6°C higher in 2100 than at the end of the 20th century. These climate changes can raise the mean temperature in all Brazilian regions even more [9] and because of the growing inclusion of information and communication technology innovations (new ICT) in educational institutions that result in an increase in thermal radiation, it is important to investigate how this increase can affect students’ comfort and performance.

According to Yang et al. [10], given the rapidly increasing use of these technologies during class (e.g. computers, projectors) researchers have been motivated to examine the effect of those equipment have both on learning and on the perception of students in the educational environment. However, a consensus regarding their benefit or harm to students has not been reached because some authors advocate that the technology used in the classroom is a positive factor whereas others claim that it has a negative effect on learning. Nevertheless, according to Yang et al. [11], temperature is one of the most significant factors affecting the perception of students regarding comfort in the classroom.

Therefore, considering that environments with new ICT, including educational institutions, are providing more interaction among individuals anywhere in the world and contributing to higher learning flexibility, it is necessary to pay extra attention to the radiation dissipated by the technology in these environments. According to Silva [2], these types of radiation are low in frequency and are transformed into thermal radiation, which, together with the heat emitted by humans and the environment and climatic and personal variables, increases the total radiation in a room.

Research on thermal radiation has received little attention in the literature reviews on thermal comfort research and practices by Rupp et al. [8], De Dear et al. [12], and Van Hoof (2008) [13], which has limited the development of studies in this area. According to research conducted by Halawa et al. (2014) [14], most of the published work attributes the level of thermal comfort of the environments only to dry bulb temperature (air temperature), without taking into account the influence of other comfort variables. In Fanger’s equation (1970) [6], the thermal radiation factor is represented by the mean radiant temperature (t_rm), but its impact on thermal comfort is often neglected.

The globe temperature is one of the variables of the equations for calculating the mean radiant temperature, and it corresponds to the temperature that allows assessment of the thermal radiation level of the surfaces in a certain environment, [15], where a significant difference between the globe temperature and the air temperature indicates an increase in thermal radiation in a work environment. The globe temperature is the temperature also used in the thermal stress indicator Wet-bulb globe temperature (WBGT). WBGT seems to be still used world widely for the evaluation of heat stress conditions and it is recommended by ISO and American Conference of governmental Industrial Hygienists as a screening method [16 –19]. In a study conducted by Vasconcelos (2015) [20], with cadets from the Military Police of Paraíba (PMPB), it was evidenced that in addition to air temperature, the globe temperature also exerted an influence on general cognitive performance. Several studies on thermal comfort have been conducted, but in most of these studies, the mean radiant temperature is assumed to be equal to the air temperature of the room [21, 22].

Mean radiant temperature is one of the most important variables for assessing thermal comfort, particularly under hot and sunny climatic conditions [23]. According to Akimoto et al. [24], an increase in the mean radiant temperature of an internal environment can reduce the productivity of its occupants, considering the significant influence of thermal comfort on their productivity and satisfaction. Halawa et al. [14] provide a critical overview of the impact of the thermal radiation field on thermal comfort and energy control and consumption in buildings. This critical overview shows that the thermal radiation field is an essential parameter of thermal comfort. Arslanoglu and Yigit [11] study thermal comfort inside a climate chamber under the effects of radiation from lighting lamps in shopping centers. Their results show that the heat flux due to radiation from these lamps causes local differences in skin temperature and that these differences cause thermal discomfort. Furthermore, the incidence and increase of thermal radiation in an environment are related to the architectural design. Walikewitz et al. [22] show that different environmental characteristics (such as wall exposure, construction materials, and room and window size) may play fundamental roles in determining the internal climate and the incidence of thermal radiation in an environment; therefore, these characteristics must be considered in the architectural design.

Working and learning environments are undergoing changes related to technological aspects. Today, it is possible to teach classes, participate in lectures, and even ask and answer questions without the restrictions of time and space. The link between wireless communication systems, computers, and projectors allows student-teacher interactions, facilitating communication and increasing learning. However, with the inclusion of new technologies in the educational environment, it is necessary to rethink the quality of buildings and to propose lighting and air conditioning designs that are compatible with the activities performed in them. If on the one hand, these new designs, together with the inclusion of new technologies, are modernizing, then, on the other hand, they may further increase the thermal load in the environment.

Adverse external environmental conditions, which may be related to climatic changes leading to air temperature increases, are another factor that contributes to a higher thermal load. The increase in air temperature in the environments with new ICT may still be related to cardiovascular dysfunctions, and with the reduction of physical and cognitive performance in humans. According to research conducted by Siqueira et al. [25], the cognitive performance of undergraduate students in the field of engineering and technology increased while performing activities in a learning environment with an air temperature of approximately 23.3°C (according to their thermal perception), when students have an initial blood pressure of 93.33 mmHg and a heart rate 60 bpm.

Therefore, it is important to investigate the behavior of the globe temperature relative to the air temperature, with possible effects on the increase in the mean radiant temperature, in classrooms where students perform learning activities with new ICT subject to air temperature variations.

2. Experimental methods

The methods adopted in this study consisted of the following stages: analysis of thermal variables, student performance, analysis of architectural elements, data processing, mathematic modelling, and ethical considerations.

The study was performed in air-conditioned information technology laboratories (ITLs) with new ICT (multimedia and wireless systems, air conditioning and lighting, personal computers, online printers) in the following institutions: (A) the Federal University of Piauí (UFPI), Teresina-Piauí; (B) the Federal University of Santa Catarina (UFSC), Florianopolis - Santa Catarina; (C) the University of Brasília (UnB) - Gama Campus, Brasília - Federal District; (D) the Federal University of the São Francisco Valley (UNIVASF), Petrolina - Pernambuco; (E) the Mathematics and Computer Sciences Institute (ICMC) of the University of São Paulo (USP), São Carlos - São Paulo; (F) the Federal University of Amazonas (UFAM), Manaus - Amazonas; and (G) the Paraíba Military Police Educational Center (MPEC), João Pessoa - Paraíba. The number of participants per institution, location, and information related to the data collection period of three consecutive days are summarized in Table 1. The study was conducted between 2012 and 2016 and that most of the measurements were carried out in 2016.

Table 1
Sample size per institution

Institution Location Sample size Period

A – UFPI Teresina – Piauí 28 students 15–17 SEPT. 2015

B – UFSC Florianopolis - Santa Catarina 19 students 21–23 MAR. 2016

C – UnB Brasília - Federal District 18 students 26–28 APR. 2016

D - UNIVASF Petrolina – Pernambuco 15 students 23–25 MAY 2016

E - ICMC São Carlos - São Paulo 15 students 17–19 NOV. 2014

F - UFAM Manaus – Amazonas 28 students 1–3 AUG. 2016

G - MPEC João Pessoa – Paraíba 40 students 18–20 JULY 2012

Total 163 students

Institution	Location	Sample size	Period
A – UFPI	Teresina – Piauí	28 students	15–17 SEPT. 2015
B – UFSC	Florianopolis - Santa Catarina	19 students	21–23 MAR. 2016
C – UnB	Brasília - Federal District	18 students	26–28 APR. 2016
D - UNIVASF	Petrolina – Pernambuco	15 students	23–25 MAY 2016
E - ICMC	São Carlos - São Paulo	15 students	17–19 NOV. 2014
F - UFAM	Manaus – Amazonas	28 students	1–3 AUG. 2016
G - MPEC	João Pessoa – Paraíba	40 students	18–20 JULY 2012
Total		163 students

Data were obtained from measurements performed in the ITLs, one session per day from 1:00 p.m. to 3:00 p.m. For each ITL, the measurements were performed at three dry bulb air temperatures (t_a) of 20°C, 24°C and 30°C, according to the ASHRAE 55 [26], ISO 7726 [27] and ISO 7730 [28] standards. The temperatures were regulated by the conditioned air. All data were measured continuously during collection at 1-minute intervals to determine whether the thermal conditions of the environment were constant. During the measurement sessions, the activities performed by students included the use of personal computers and/or laptops (Institution G) to access the electronic address (link) for the Battery of Reasoning Tests-5(BPR-5) [29]. The BPR5 is designed to assess overall cognitive functioning and is used by specialists, for instance, as an auxiliary measure for psychodiagnosis, job selection and learning assessment. The instrument comprises five subtests that assess the following skills: Abstract Reasoning (AS), Verbal Reasoning (VR), Spatial Reasoning (SR), Numerical Reasoning (NR) and Mechanical Reasoning (MR). A survey was created using the website Qualtrics.com to apply the instrument and measure the amount of time the participant required to answer each question. The number of hits, the time required to answer and the relationship between these factors was calculated. The tests were randomly distributed; each participant completed one test on each of the study days.

As for the thermal resistance of clothing, the mean was 0.45±0.12. In view of the fact that the students were instructed to wear similar clothing over the three days of data collection. Before the students performed their activities in the laboratories, they were taken to a room with favorable environmental conditions to provide a rest time for stabilization of their bodies, which was verified through the measurement of their arterial pressures using a HEM-7220 automatic arterial pressure monitor and a Polar FT7 heart rate monitor. The procedures for measuring blood pressure complied with the VI Hypertension Guidelines [30].

2.1. Thermal variables

The dry bulb temperature (t_a), the globe temperature (t_g), and the relative humidity (RH) were collected by a BABUC-A microclimate station and a TGD300 thermal stress meter. The instruments met the requirements of the ISO 7726 [27] standard. The air velocity was considered constant V = 0.1 m/s, an estimate for closed environments [14, 27].

The transducers connected to the instruments measured the thermal variables and had the following features:

Aspiration psychrometer: This instrument measures the air temperature or dry bulb temperature (T_a) as well as the wet bulb temperature, T_bu, with a resolution of 0.03 K, an accuracy of±0.13 K, a 90 s response time, and a – 20°C to 60°C measurement range. It provides the RH value OF THE AIR with a 0.1% resolution and accuracies of±0.5% from 70 to 98%,±1% from 40 to 70%, and±2% from 15 to 40%, with a measurement range of 0 to 100%.

Globe thermometer: The globe temperature, dry bulb temperature and air velocity are necessary to obtain the mean radiant temperature. The globe thermometer has a globe painted black, with 0.95 emissivity (□) and an external diameter of 0.15 m, a resolution of 0.03°C, an accuracy of±0.15°C, about 20 to 30 minutes response time, and a measuring range from 10°C to 100°C.

According to the guidelines of the ISO 7726 [27] standard and considering that the environments were uniform in terms of temperature distribution, the microclimate station was installed in the middle of the laboratories at a height of 0.6 m from the floor, once the students were seated. The instruments were programmed to record the measurements every minute from the time the students entered the classroom. For stabilization purposes, the instruments were installed in the laboratories 30 minutes before data collection.

Moreover, to check whether heat exchange occurred by radiation between the students and the environment, the following hypotheses were tested according to the mean radiant temperature equation (equation (1)) from ISO 7726 [27]:

If t_g = t_a, then t_rm = t_g, and there is no heat exchange by radiation.

If t_g > t_a, then t_rm increases proportionally with t_g, which indicates heat exchange by radiation.

If t_g < t_a, then t_rm decreases proportionally with t_g, which indicates that there is no heat exchange by radiation.

trm = \sqrt[4]{{tg}^{4} + \frac{h_{cg}}{ɛ_{g^{σ}}} (t_{g} - t_{a})}

(1)

Where t_rm = the mean radiant temperature; t_g = the globe temperature; $h_{cg} = 6.3 \frac{v^{0.6}}{D^{0.4}}$ , where v = the air velocity and D = the globe diameter; ɛ_g= the emissivity of the black globe (non-dimensional); and σ= the Stefan-Boltzman constant = 5.67×10^-8 W/m².K⁴.

2.2. Student performance

Simultaneously with the thermal measurements in the ITLs, which mainly assessed the behavior of the globe temperature relative to the air temperature (T_g-T_a), the students performed activities related to the (BPR-5) to determine their general cognitive performance during the climatic variations in the ITLs. The BPR-5 consists of five sub-tests. It is an instrument that assesses cognitive abilities and provides estimates of general cognitive functioning and of strengths and weaknesses in five specific areas analyzed by the following five sub-tests: the Abstract Reasoning (AR) Test, the Verbal Reasoning (VR) Test, the Numeric Reasoning (NR) Test, the Spatial Reasoning (SR) Test, and the Mechanical Reasoning (MR) Test. According to the authors, the BPR-5 is used by professionals in psycho-diagnosis, personnel selection, professional orientation, and learning assessment, among other areas; in other words, it is used to check general cognitive functioning [31].

In this study, the BPR-5 was decomposed and reorganized into three summarized tests of the same difficulty level to be applied each day. For this purpose, Siqueira [32] developed a survey in the Qualtrics platform, which enabled the application and measurement of the response time for each question. The total number of right answers varied from 0 to 20, considering that the overall reasoning test score consisted of the sum of the scores of the five sub-tests, each containing four questions, with a value of 1 for each question. Each test battery consisted of four questions for each reasoning test, as shown in Table 2.

Table 2
Reasoning sub-tests based on BPR-5

Test Description

Verbal Reasoning (VR) Grammar and the types of relationships between words. The test consists of four items in which the analogy relationship between a first pair of words must be found and applied in a consistent manner to identify a fourth word among five response options that maintains the same relationship with a third word shown. (Duration: 4 minutes)

Abstract Reasoning (AR) Types and number of modification rule types. The test comprises four items in which it is necessary to find the relationship between the two first terms and apply it to the third term to identify the fourth figure among five response options. (Duration: 8 minutes)

Mechanical Reasoning (MR) Types of knowledge in physics. The test comprises four items consisting of figures that show a problem and a response option. The questions consist of practical problems that involve physical and mechanical contents. The response is the option that provides the best solution to the problem. (Duration: 8 minutes)

Spatial Reasoning (SR) Number of rotation axes, rotation directions, visible faces of a cube, and the presence of visual stimuli on the edges of the cube faces. The test comprises four items with a series of moving tridimensional cubes. The movements can be constant, i.e., always to the right, or alternate, i.e., to the left and upwards. Once the movement is found by analyzing the different faces, the representation of the cube that would follow if the movement identified were applied to the last cube of the series must be chosen among the response options. (Duration: 10 minutes)

Numeric Reasoning (NR) Size of a numeric sequence, mathematical operations used, number of logical sequences used, and the complexity of the logic sequences. The test comprises four items that are linear or alternate series of numbers in which the subject must find the arithmetic relationship of the progression and apply it to identify the last two numbers that would complete the series. (Duration: 10 minutes)

Test	Description
Verbal Reasoning (VR)	Grammar and the types of relationships between words. The test consists of four items in which the analogy relationship between a first pair of words must be found and applied in a consistent manner to identify a fourth word among five response options that maintains the same relationship with a third word shown. (Duration: 4 minutes)
Abstract Reasoning (AR)	Types and number of modification rule types. The test comprises four items in which it is necessary to find the relationship between the two first terms and apply it to the third term to identify the fourth figure among five response options. (Duration: 8 minutes)
Mechanical Reasoning (MR)	Types of knowledge in physics. The test comprises four items consisting of figures that show a problem and a response option. The questions consist of practical problems that involve physical and mechanical contents. The response is the option that provides the best solution to the problem. (Duration: 8 minutes)
Spatial Reasoning (SR)	Number of rotation axes, rotation directions, visible faces of a cube, and the presence of visual stimuli on the edges of the cube faces. The test comprises four items with a series of moving tridimensional cubes. The movements can be constant, i.e., always to the right, or alternate, i.e., to the left and upwards. Once the movement is found by analyzing the different faces, the representation of the cube that would follow if the movement identified were applied to the last cube of the series must be chosen among the response options. (Duration: 10 minutes)
Numeric Reasoning (NR)	Size of a numeric sequence, mathematical operations used, number of logical sequences used, and the complexity of the logic sequences. The test comprises four items that are linear or alternate series of numbers in which the subject must find the arithmetic relationship of the progression and apply it to identify the last two numbers that would complete the series. (Duration: 10 minutes)

Source: Almeida and Primi [20].

Based on this classification and categorization, similar items were randomly distributed among the tests; thus, each student answered one test per day, for a total of three reasoning tests. Performance could be classified into three ranges: below average, with up to 25% correct answers; average, with between 25% and 75% correct answers; and above average, with more than 75% correct answers [31].

2.3. Analysis of architectural elements

A short architectural analysis of the building envelopes and their constituent materials was performed according to the NBR 15220 – Thermal Performance of Buildings standard [33] in addition to analysis of the architectural design following several recommendations by Roriz [34], considering that these factors may be associated with a possible increase in t_g relative to t_a that would cause an increase in t_rm. The architectural analysis was performed to better understand the possible thermal behavior of the building, mainly regarding the differences in the air and globe temperatures with a possible increase in t_rm. It is possible to examine the thermal performance as well as of the architectural design of a building through a brief assessment of the building envelope and its constituent materials because an adequate design together with the correct use of materials may contribute to thermal comfort, with repercussions on general comfort and productivity.

Thus, if the architectural design has more negative than positive aspects in relation to architectural elements and t_g > t_a, it is likely that some architectural element combined with the internal heat sources contributes to an increase in the globe temperature, thus increasing the thermal load of the environment. However, if the architectural design has more positive aspects and t_g > t_a, it is likely that the increase of the thermal radiation in the environment is associated with only internal sources of heat.

Table 3 shows the architectural design recommendations according to Roriz [33].

Table 3
Architectural design recommendations according to Roriz

Implantation Buildings must have an elongated shape, with the largest façades toward the north and the south to reduce the thermal load due to solar radiation. There can be a slight deviation in this orientation to capture the prevailing breezes during the most humid months or to allow more solar heating during cold months.

Space between buildings The distances between buildings must be considerable to allow for free air circulation. As a rule, these distances must be at least five times larger than the height of the buildings. Construction on stilts may also be helpful for this purpose.

Ventilation The design must optimize cross ventilation. For this purpose, it is desired that rooms be arranged in “simple rows” along buildings.

Window size Windows must be large, taking between 40% and 80% of the north and south façades to allow cross-ventilation to the rooms at the level of the occupants’ bodies.

Window position In the north and south façades, to allow cross-ventilation to the rooms at the level of the occupants’ bodies. To optimize air velocity inside rooms, the exit windows must be slightly larger and taller than the entrance widows.

Window protection Rooms must be protected from direct solar radiation, and windows must be carefully protected from rain.

Walls Walls (low thermal inertia) and surfaces should be light colors to reflect solar radiation. Thermal transmittance (U)≤2.8 W/m²°C; delay (φ)≤3 hours, and solar heat factor (F_so)≤4%. Considering F_so = 4.U.α, where U is thermal transmittance, and α is solar radiation absorbance.

Roofing Roofing must be light but with high thermal resistance. This high insulation is important to avoid excessive heating of the roof’s lower face (ceiling). Adopt thermal transmittance (U)≤0.85 W/m²°C; delay≤3 hours; and (F_so)≤3%.

Implantation	Buildings must have an elongated shape, with the largest façades toward the north and the south to reduce the thermal load due to solar radiation. There can be a slight deviation in this orientation to capture the prevailing breezes during the most humid months or to allow more solar heating during cold months.
Space between buildings	The distances between buildings must be considerable to allow for free air circulation. As a rule, these distances must be at least five times larger than the height of the buildings. Construction on stilts may also be helpful for this purpose.
Ventilation	The design must optimize cross ventilation. For this purpose, it is desired that rooms be arranged in “simple rows” along buildings.
Window size	Windows must be large, taking between 40% and 80% of the north and south façades to allow cross-ventilation to the rooms at the level of the occupants’ bodies.
Window position	In the north and south façades, to allow cross-ventilation to the rooms at the level of the occupants’ bodies. To optimize air velocity inside rooms, the exit windows must be slightly larger and taller than the entrance widows.
Window protection	Rooms must be protected from direct solar radiation, and windows must be carefully protected from rain.
Walls	Walls (low thermal inertia) and surfaces should be light colors to reflect solar radiation. Thermal transmittance (U)≤2.8 W/m²°C; delay (φ)≤3 hours, and solar heat factor (F_so)≤4%. Considering F_so = 4.U.α, where U is thermal transmittance, and α is solar radiation absorbance.
Roofing	Roofing must be light but with high thermal resistance. This high insulation is important to avoid excessive heating of the roof’s lower face (ceiling). Adopt thermal transmittance (U)≤0.85 W/m²°C; delay≤3 hours; and (F_so)≤3%.

The NBR 15220 standard [33] makes recommendations for good building thermal performance and provides data on the thermal transmittance (U), thermal capacity (CT), and thermal delay (φ) of walls and roofs. Thus, reference information from NBR 15220 [33] was used to obtain data on side closings and roofing listed in Table 3. These considerations may explain the increases in the mean radiant temperature in the classrooms under study.

2.4. Data processing

The data collected were tabulated in a Microsoft Excel^® spreadsheet and prepared for use in the programs R, SPSS^® and STATISTICA to perform descriptive data analysis, create graphs, establish correlations between the variables, and build the mathematical model. Non-parametric Kruskal-Wallis tests were used to test the hypothesis that the distribution of the parameters was similar over all the days of the experiment. Spearman’s correlation was used to test the correlations between the variables of interest.

2.5. Mathematical modeling

A mathematical model was built using the generalized linear model (GLM) class to investigate the behavior of (t_g - t_a) and t_rm in the cognitive performance (D_t index – overall performance as a function of time) of the students, considering that (t_g - t_a)>0 indicates an increase in t_rm with repercussions on the increase in the internal thermal load and that the effects of these variables on the variability of the students’ cognitive performance must be analyzed.

GLMs correspond to a large class of statistical models defined in terms of a set of independent random variables, in which each variable has a certain exponential distribution [35]. The model has two parts:

Probability distribution of Y(D_t), Y ∼ N(μ, σ²)

The function that links the expected value Y(D_t) with a linear combination of the explanatory variables t_rm and (t_g - t_a), according to equation (2):

$E (Dt) = β_{0} + β_{1} T_{trm} + β_{2} (t_{g} - t_{a})$ (2)

2.6. Ethical considerations

This study followed the Brazilian guidelines (Resolution CNS 466/12 and complements), which state that any study involving human beings must be submitted to a Committee on Ethics in Research. This study was submitted, acknowledged, and approved as ethically adequate under registration number CAAE 57408416.3.0000.5188.

3. Characterization of the rooms and architectural elements

3.1. Information technology laboratory A – UFPI

The ITL of Institution A has an area of approximately 112 m² and 48 computers with LCD monitors are installed. The room temperature is controlled by two split-type air conditioners, and there is artificial and natural lighting. The windows are in the back of the room and covered with 50% transparency film, mirrored on the outside to reduce the entry of solar radiation.

The largest building façade is oriented to the northeast, but the elongated shape of buildings reduces exposure to solar radiation and increases exposure to the prevailing winds. The building has two floors, and the laboratory under study is on the west side of the ground floor; thus, it receives solar radiation on the northwest wall in the afternoons. The windows, located in the southwest façade of the building, have no protection elements; thus, they receive direct solar radiation. The window area is 10.43 m², or 41.68% of the laboratory. The most common construction materials in the ITL are concrete and masonry, and the ceramic tile roof is supported by a wood structure. Because of its porosity, this material helps reduce the flow of heat into the building.

3.2. Information technology laboratory B – UFSC

The ITL of institution B has an area of approximately 100 m² with 25 computers with LCD monitors. The glass windows are located in the back of the room. The upper parts of the windows are tinted and have curtains. The room temperature is controlled by two split-type air conditioners, and there is artificial and natural lighting.

The building has three floors and its façades are practically the same size. The laboratory is located on the first floor, and the windows are in the southwest façade with an area of 10.35 m², or 47% of the laboratory. There is vegetation that helps cool the air entering the building. The use of curtains, an internal solar protection element, was observed, although they do not impede the entrance of direct solar radiation which is then obstructed by the curtain after crossing the glass and reflected as heat that remains inside the building. The predominant construction materials of the ITL are concrete and masonry, and the roof is constructed of fiber-cement tiles that contribute to heat gain inside the building because of their high thermal transmittance and solar heat factor.

3.3. Information technology laboratory C – UnB

This ITL has an area of approximately 186.93 m² with 80 computers with LCD monitors. There are glass windows on the side of the laboratory. The room temperature is controlled by two split-type air conditioners, and there is artificial and natural lighting.

The largest façade is oriented to the northeast, and the building has a central courtyard. The building has two floors, and the laboratory is on the east side of the ground floor; thus, it does not receive direct solar radiation in the afternoons. The windows are located in the southeast façade with an area of 33.82 m², or 59.22% of the laboratory. They have brises-soleils as external solar protection elements to reduce the heat flow to the inside of the building. The main construction materials used in the ITL are concrete and masonry, and the roof is constructed of fiber-cement tiles.

3.4. Information technology laboratory D – UNIVASF

This ITL has an area of approximately 100.24 m² with 24 computers with LCD monitors. The laboratory has a total of five air conditioners; three of them are split-type units and the other two are for ceiling installation.

The largest façade is oriented to the northwest, which, together with the elongated shape of the building, increases the incidence of solar radiation. The building has three floors, and the laboratory is located on the north side of the first floor, receiving solar radiation on the northwest façade in the afternoons. The windows are made of matte glass (some are covered with sheets of dark paper) and are located on the northwest façade, having an area of 15.21 m², or 43.89% of the laboratory. There is no concern with protecting the windows, which are very hot; it is uncomfortable to stay close to them. The pillars external to the building help protect the windows, but they are ineffective in avoiding a large heat gain due to direct solar radiation. The main construction materials used are concrete and masonry, and the roof is constructed of fiber-cement tiles, which contribute to increasing the heat gain because of their high thermal transmittance and solar heat factor.

3.5. Information technology laboratory E – ICMC

This ITL has an area of approximately 61.68 m² with 40 computers with LCD monitors. The windows are made of glass and located on the side of the laboratory. The room temperature is controlled by two split-type air conditioners, and there is artificial and natural lighting.

The building has an elongated shape, and its largest façade is oriented to the north, which reduces exposure to solar radiation. The windows have an area of 7 m², or 40.79% of the laboratory area. The vegetation in the surroundings helps cool the airflow that enters the building and protects against solar radiation. The predominant construction materials are concrete and masonry, and the roof is constructed of fiber-cement, although its solar heat factor is reduced because it is not blackened. The side closings are composed of PVC partition panels.

3.6. Information technology laboratory F – UFAM

This ITL has an area of approximately 69.42 m² with 30 computers with LCD monitors. The laboratory has two split-type air conditioners installed in the north wall and wood and glass windows in the north and south façades.

The largest façade of the building is oriented in the north-south direction, and the elongated shape of the building reduces exposure to solar radiation and increases the effect of the prevailing winds on the largest façades. The windows are large and located on the north and south façades, with areas of 17.85 m², or 61.78% of the laboratory area, and 14.5 m², or 42.37% of the laboratory area, respectively. The gardens around the building help cool the airflow toward the building and protect it against solar radiation. The predominant construction materials are steel, concrete, and masonry. The roof is constructed of fiber-cement tiles and is independent of the structure, which helps protect the building because of the air circulation in the sides and the cushion of ventilated air between the fiber-cement tiles and the concrete ceiling.

3.7. Information technology laboratory G – MPEC

This ITL has an area of approximately 76.38 m². In it, the temperature is controlled by two split-type air conditioners, and it has artificial lighting and some natural light that enters through small crevices in the windows. The windows are made of wood and glass and oriented to the south, with no view to the outside. The north wall receives solar radiation. The students used their own laptops during data collection.

The largest façades are oriented in the north-south direction, which, together with the elongated shape of the building, reduces exposure to solar radiation. The building that houses the laboratory is far from other buildings in the area, which allows the passage of air currents. The ITL is located in the west side of the building; thus, it receives solar radiation in the afternoons, although vegetation and a low external wall provide protection to this façade. The windows are in the south façade, with an area of 3.81 m², or 20.19% of the laboratory area. The predominant construction materials are concrete and masonry. The roof is constructed of fiber-cement tiles on top of the rooms’ slab and is supported by hollow elements, which creates a cushion of ventilated air between the tiles and the concrete ceiling.

4. Results

4.1. Analysis of the thermal variables

Table 4 shows the mean values of the thermal variables measured in the three data collection sessions for each ITL.

The analysis of the heat exchanged due to radiation between the environment and the individuals in each session per ITL was performed using the data in Table 4 and equation (1). The results are shown in Table 5.

Table 4
Mean thermal conditions by information technology laboratory (ITL)

Variable ITL - A ITL - B ITL – C ITL - D ITL - E ITL - F ITL - G

T_a (°C) Session 20 °C 20.07±0.01 22.64±0.01 24.53±0.01 22.43±0.00 23.09±0.07 23.58±0.12 20.28±0.30

Session 24 °C 22.95±0.01 24.21±0.03 24.58±0.01 23.04±0.01 23.30±0.03 25.93±0.12 23.00±0.20

Session 30 °C 33.74±0.01 29.08±0.04 24.79±0.01 28.52±0.04 28.80±0.01 30.85±0.12 29.50±0.44

T_g (°C) Session 20 °C 21.03±0.01 23.28±0.01 24.37±0.01 23.27±0.00 23.72±0.03 22.94±0.03 22.13±0.08

Session 24 °C 24.12±0.01 24.72±0.02 24.47±0.01 24.04±0.02 23.94±0.01 25.33±0.01 25.24±0.31

Session 30 °C 33.55±0.17 28.72±0.04 24.48±0.05 28.64±0.05 27.78±0.05 29.92±0.00 29.59±0.59

T_rm (°C) Session 20 °C 21.66±0.01 23.66±0.01 24.27±0.01 23.71±0.01 24.10±0.06 22.55±0.07 23.26±0.15

Session 24 °C 24.82±0.01 25.02±0.02 24.41±0.00 24.64±0.03 24.33±0.02 24.98±0.06 26.56±0.38

Session 30 °C 33.46±0.17 28.51±0.04 24.50±0.01 28.70±0.05 27.19±0.08 29.39±0.06 29.64±1.16

T_g- T_a(°C) Session 20 °C 0.97±0.00 0.64±0.01 –0.11±0.01 0.83±0.00 0.63±0.07 –0.64±0.11 1.86±0.26

Session 24 °C 1.17±0.00 0.51±0.02 –0.16±0.01 1.00±0.02 0.65±0.03 –0.60±0.11 2.25±0.17

Session 30 °C –0.17±0.01 –0.36±0.00 –0.32±0.00 0.12±0.01 –1.01±0.04 –0.93±0.11 0.09±1.03

RH (%) Session 20 °C 69.20±0.16 65.51±0.05 66.12±0.07 64.95±0.16 48.16±0.73 51.33±1.04 43.59±1.64

Session 24 °C 67.57±0.24 66.67±0.15 58.04±0.21 64.53±0.03 53.90±0.59 55.36±1.16 49.27±2.53

Session 30 °C 64.63±0.86 82.67±0.33 71.70±0.18 71.83±0.16 45.37±0.33 69.30±0.47 62.87±9.31

Variable		ITL - A	ITL - B	ITL – C	ITL - D	ITL - E	ITL - F	ITL - G
T_a (°C)	Session 20 °C	20.07±0.01	22.64±0.01	24.53±0.01	22.43±0.00	23.09±0.07	23.58±0.12	20.28±0.30
	Session 24 °C	22.95±0.01	24.21±0.03	24.58±0.01	23.04±0.01	23.30±0.03	25.93±0.12	23.00±0.20
	Session 30 °C	33.74±0.01	29.08±0.04	24.79±0.01	28.52±0.04	28.80±0.01	30.85±0.12	29.50±0.44
T_g (°C)	Session 20 °C	21.03±0.01	23.28±0.01	24.37±0.01	23.27±0.00	23.72±0.03	22.94±0.03	22.13±0.08
	Session 24 °C	24.12±0.01	24.72±0.02	24.47±0.01	24.04±0.02	23.94±0.01	25.33±0.01	25.24±0.31
	Session 30 °C	33.55±0.17	28.72±0.04	24.48±0.05	28.64±0.05	27.78±0.05	29.92±0.00	29.59±0.59
T_rm (°C)	Session 20 °C	21.66±0.01	23.66±0.01	24.27±0.01	23.71±0.01	24.10±0.06	22.55±0.07	23.26±0.15
	Session 24 °C	24.82±0.01	25.02±0.02	24.41±0.00	24.64±0.03	24.33±0.02	24.98±0.06	26.56±0.38
	Session 30 °C	33.46±0.17	28.51±0.04	24.50±0.01	28.70±0.05	27.19±0.08	29.39±0.06	29.64±1.16
T_g- T_a(°C)	Session 20 °C	0.97±0.00	0.64±0.01	–0.11±0.01	0.83±0.00	0.63±0.07	–0.64±0.11	1.86±0.26
	Session 24 °C	1.17±0.00	0.51±0.02	–0.16±0.01	1.00±0.02	0.65±0.03	–0.60±0.11	2.25±0.17
	Session 30 °C	–0.17±0.01	–0.36±0.00	–0.32±0.00	0.12±0.01	–1.01±0.04	–0.93±0.11	0.09±1.03
RH (%)	Session 20 °C	69.20±0.16	65.51±0.05	66.12±0.07	64.95±0.16	48.16±0.73	51.33±1.04	43.59±1.64
	Session 24 °C	67.57±0.24	66.67±0.15	58.04±0.21	64.53±0.03	53.90±0.59	55.36±1.16	49.27±2.53
	Session 30 °C	64.63±0.86	82.67±0.33	71.70±0.18	71.83±0.16	45.37±0.33	69.30±0.47	62.87±9.31

Table 5

Summary of the heat exchange due to radiation by session and by information technology laboratory (ITL)

ITL	SESSION 20°C	SESSION 24°C	SESSION 30°C
A	T_g > T_a	T_g > T_a	T_g < T_a
	0.97 °C±0.0	1.17 °C±0.0	–0.17 °C±0.0
	There was heat exchange due to radiation	There was heat exchange due to radiation	There was no heat exchange due to radiation
B	T_g > T_a	T_g > T_a	T_g < T_a
	0.64 °C±0.0	0.51 °C±0.0	–0.36 °C±0.0
	There was heat exchange due to radiation	There was heat exchange due to radiation	There was no heat exchange due to radiation
C	T_g < T_a	T_g < T_a	T_g < T_a
	–0.11 °C±0.0	–0.16 °C±0.0	–0.32 °C±0.0
	There was no heat exchange due to radiation	There was no heat exchange due to radiation	There was no heat exchange due to radiation
D	T_g > T_a	T_g > T_a	T_g > T_a
	0.83 °C±0.0	1.00 °C±0.0	0.12 °C±0.0
	There was heat exchange due to radiation	There was heat exchange due to radiation	There was heat exchange due to radiation
E	T_g > T_a	T_g > T_a	T_g < T_a
	0.63 °C±0.0	0.65 °C±0.0	–1.00 °C±0.0
	There was heat exchange due to radiation	There was heat exchange due to radiation	There was no heat exchange due to radiation
F	T_g < T_a	T_g < T_a	T_g < T_a
	–0.64 °C±0.0	–0.60 °C±0.0	–0.93 °C±0.0
	There was no heat exchange due to radiation	There was no heat exchange due to radiation	There was no heat exchange due to radiation
G	T_g > T_a	T_g > T_a	T_g > T_a
	1.86 °C±0.2	2.25 °C±0.0	0.09 °C±1.0
	There was heat exchange due to radiation	There was heat exchange due to radiation	There was heat exchange due to radiation

The highest amount of heat exchange due to thermal radiation occurred in ITL G, in the city of João Pessoa, which may have been due to the number of laptops (40) in the room, considering that they emit more radiation than personal computers [2], and/or due to the architectural characteristics of the room that could have resulted in a high incidence of thermal radiation. There was no heat exchange due to radiation in ITLs C (Brasília) and F (Manaus). These results could be explained by a low outside temperature (22°C in Brasília) or by the presence of architectural elements or construction materials that favor the non-emission of thermal radiation to the environment (ITL F).

To better visualize these findings, Fig. 1 shows that the increase in globe temperature relative to air temperature was more accentuated in ITL G in the 20°C, 24°C and 30°C sessions, when the increase in t_g changed over time. Therefore the heat exchange in the laboratory of institution G was the highest, with the mean radiant temperature being higher than the globe temperature by 2.25°C $\frac{h_{cg}}{ɛ_{g σ}}$ at any given time.

Fig.1

Globe temperature (t_g) and air temperature (t_a) for each information technology laboratory.

4.2. Analysis of student performance

Table 6 shows the analysis of the students’ cognitive performance for each ITL according to the classification by Almeida and Primi [31] detailed above in the Experimental Methods section.

Table 6
Analysis of student performance according to information technology laboratory (ITL)

Cognitive ability ITL – A ITL – B ITL – C ITL – D

20 °C 24 °C 30 °C 20 °C 24 °C 30 °C 20 °C 24 °C 30 °C 20 °C 24 °C 30 °C

Below average 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0%

Average 85.2% 69.2% 77.8% 58.8% 73.7% 60% 38.9% 52.9% 66.7% 73.3% 35.3% 50%

Above average 14.8% 30.8% 22.2% 41.2% 26.3% 40% 61.1% 47.1% 33.3% 26.7% 64.7% 50%

Cognitive ability ITL – E ITL – F ITL – G

20 °C 24 °C 30 °C 20 °C 24 °C 30 °C 20 °C 24 °C 30 °C

Below average 0% 0% 0% 3.6% 0% 0% 0% 0% 0%

Average 60% 53.3% 27% 85.7% 85.7% 92% 50% 33.3% 26.6%

Above average 40% 46.7% 73% 10.7% 14.3% 8% 50% 66.7% 73.4%

Cognitive ability	ITL – A	ITL – B	ITL – C	ITL – D
Below average	0%	0%	0%	0%	0%	0%	0%	0%	0%	0%	0%	0%
Average	85.2%	69.2%	77.8%	58.8%	73.7%	60%	38.9%	52.9%	66.7%	73.3%	35.3%	50%
Above average	14.8%	30.8%	22.2%	41.2%	26.3%	40%	61.1%	47.1%	33.3%	26.7%	64.7%	50%
Cognitive ability	ITL – E	ITL – F	ITL – G
	20 °C	24 °C	30 °C	20 °C	24 °C	30 °C	20 °C	24 °C	30 °C
Below average	0%	0%	0%	3.6%	0%	0%	0%	0%	0%
Average	60%	53.3%	27%	85.7%	85.7%	92%	50%	33.3%	26.6%
Above average	40%	46.7%	73%	10.7%	14.3%	8%	50%	66.7%	73.4%

ITL - A = UFPI; ITL - B = UFSC; ITL - C = UnB; ITL - D = UNIVASF; ITL - E = USP; ITL - F = UFAM; ITL - G = MPEC.

The results show that the students did not perform below average in the reasoning tests in any ITL, except for laboratory F in the session at 20°C, when approximately 4% of the students had an unsatisfactory performance. The results for ITLs E and G are more satisfactory on average than those of the other laboratories. ITL F has a less satisfactory result on an average than the other institutions.

4.3. Analysis of architectural elements

Heat exchange between the outside and the inside of a building occurs through the building envelope. The envelope can be divided into two parts: opaque and transparent closing elements. The main difference between the two is the ability to transmit solar radiation to the inside [36] and for this reason, the geographical orientation of the building and its windows, solar protections, and space organization are important [37], in addition to their constituent materials, considering that they have a direct relationship with the thermal performance of buildings. Therefore, the thermal transmittance (U) of each construction material, which represents the heat flux transmitted per unit of area and unit of temperature difference, must be considered [25].

Table 7 shows the architectural analysis of each ITL following the methodology defined for this study.

Table 7
Architectural analysis of each information technology laboratory (ITL)

Building Implantation / Space between buildings Ventilation Window position, size and protection Walls Roofing

ITL – A –Largest façade oriented to the northeast– Buildings of elongated shape, far from each other, which reduces exposure to radiation and exposes the largest façades to the prevailing winds– The building has two floors, and the laboratory is located on the west side of the ground floor; thus, it receives solar radiation in the northwest wall in the afternoons –No cross-ventilation inside the laboratory; windows are located in one façade only –Windows in the southwest façade with an area of 10.43 m², or 41.68% of the laboratory– The windows are not protected. The external pillars of the building help protect them, although this protection is ineffective –The walls are made of 6-holed plastered bricks with a thermal transmittance of 2.48 W/m².°C– The external cladding is composed of small bricks (α= 0.5) with a solar heat factor of 4.96%– The thermal delay of the wall is 3.3 hours –Roofing is composed of ceramic tiles with wood structure. Considering a roof of this type with a mixed slab (12 cm thickness), the thermal transmittance is 1.92 W/m².°C– The solar factor is 3.84% (α= 0.5)– The thermal delay of ceramic tile roofing with a mixed slab ceiling is 3.6 hours

ITL - B –The building’s shape is not elongated– The façades have practically the same size– The building has three floors, and the laboratory is on the first floor –There is no cross-ventilation inside the laboratory, windows located on one façade only –Windows in the southwest façade with an area of 10.35 m², or 47% of the laboratory– Use of curtains as internal solar protection elements –The external walls are made of solid exposed bricks, with a thermal transmittance of 3.70 W/m².°C; the internal walls are made of 6-holed plastered bricks with a 2.48 W/m².°C thermal transmittance– Considering that the external wall is made of exposed bricks (α= 0.5), its solar heat factor is 4.96%–The thermal delays of the solid brick and of the hollow plastered walls are 2.4 and 3.3 hours, respectively –The thermal transmittance of a fiber-cement roof is 1.93 W/m².°C–The solar heat factor is 5.4% (α= 0.7, considering that the fiber-cement tiles are blackened)–The thermal delay of fiber-cement roofing with a mixed slab ceiling is 3.6 hours

ITL – C –The largest façade is oriented to the northeast.–The building has a central courtyard and two floors, and the laboratory is on the east side of the ground floor; thus, it does not receive direct solar radiation in the afternoons –There is no cross-ventilation inside the laboratory, windows located on one façade only –Windows are in the southeast of the building with an area of 33.83 m², or 59.22% of the laboratory–Brises-soleils as external solar protection elements –The walls are composed of 6-holed plastered bricks with a thermal transmittance of 2.48 W/m².°C–Considering that the walls are painted white (α= 0.2), their solar heat factor is 1.5%–The thermal delay of the wall is 3.3 hours –The thermal transmittance of a fiber-cement roof is 1.93 W/m².°C–Solar heat factor of 1.5% (α= 0.2 because the fiber-cement tiles are white)–The thermal delay of fiber-cement roofing with a mixed slab ceiling is 3.6 hours

ITL – D –The largest façade of the building is oriented to the northwest –There is no cross-ventilation inside the laboratory; windows are located on one façade only –Windows are in the northwest façade of the laboratory with an area of 15.21 m², or 43.89% of the laboratory –The walls are composed of 6-holed plastered bricks with a thermal transmittance of 2.48 W/m².°C –The thermal transmittance of a fiber-cement roof is 1.93 W/m².°C

The building has an elongated shape, which increases the incident solar radiation because of its orientation There are no window protections. The external pillars of the building help protect them, although this protection is ineffective Considering that the walls are painted yellow (α= 0.3), their solar heat factor is 3% The solar heat factor is 5.4% (α= 0.7, considering that the fiber-cement tiles are blackened)

The building has three floors, and the laboratory is on the north side of the first floor; thus, it receives solar radiation in the northwest wall in the afternoons. –The windows receive direct solar radiation The thermal delay of the wall is 3.3 hours The thermal delay of a fiber-cement roof with a mixed slab ceiling is 3.6 hours

ITL - E –The building has an elongated shape, and it is oriented to the north, which reduces exposure to solar radiation –There is no cross ventilation inside the laboratory; windows are located on one façade only –The windows have an area of 7 m², or 40.79% of the laboratory –The side walls are composed of PVC partition panels –The thermal transmittance of a fiber-cement roof is 1.93 W/m².°C–Solar heat factor of 1.5% (α= 0.2 because the fiber-cement tiles are white)–The thermal delay of a fiber-cement roof with a mixed slab ceiling is 3.6 hours

–Area of the laboratory: 61.68 m² Brises-soleils proveide external solar protection

ITL – F –The largest façade is oriented in the north-south direction –There is cross ventilation; the building has windows on the north and south façades –The windows are large and located on both the north and south façades, with an area of 17.85 m², or 61.78% of the laboratory in the north façade and an area of 14.5 m², or 42.37% of the laboratory, in the south façade –The walls are made of 8-holed bricks with a thermal transmittance of 2.24 W/m².°C –The thermal transmittance of a fiber-cement roof is 1.93 W/m².°C

The buildings have an elongated shape and are far from each other, which reduces exposure to solar radiation and increases the exposure of the largest façades to prevailing winds –Large eaves protect against direct solar radiation The external cladding of the wall is made of beige ceramic tiles with α= 0.3; therefore, the solar heat factor is 2.7% Solar heat factor of 1.5% (α= 0.2 because the fiber-cement tiles are white)

The building has a ground floor only However, the thermal delay of the walls is 3.7 hours The thermal delay of a fiber-cement roof with a mixed slab ceiling is 3.6 hours–The slab has thermal insulation made of cork, which reduces the thermal transmittance and improves thermal performance

ITL - G –The largest façades are in the north-south direction, and the building has an elongated shape; both features contribute to reducing exposure to solar radiation –There is no cross-ventilation inside the laboratory; windows are located on one façade only –The windows are on the south façade, with an area of 3.81 m², or 20.19% in that façade of the laboratory –The walls are composed of 6-holed plastered bricks with a thermal transmittance of 2.48 W/m².°C –The thermal transmittance of a fiber-cement roof is 1.93 W/m².°C

–The building is far from other surrounding buildings, which improves air circulation Considering that the walls are painted white (α= 0.2), their solar heat factor is 1.5% Solar heat factor of 1.5% (α= 0.2 because the fiber-cement tiles are white)

Building with ground floor only. The laboratory is on the west side; thus, it receives solar radiation in the afternoons, although the presence of vegetation and a low wall on the outside protects the building from solar radiation The thermal delay of the wall is 3.3 hours The thermal delay of fiber-cement roofing with a mixed slab ceiling is 3.6 hours

Building	Implantation / Space between buildings	Ventilation	Window position, size and protection	Walls	Roofing
ITL – A	–Largest façade oriented to the northeast– Buildings of elongated shape, far from each other, which reduces exposure to radiation and exposes the largest façades to the prevailing winds– The building has two floors, and the laboratory is located on the west side of the ground floor; thus, it receives solar radiation in the northwest wall in the afternoons	–No cross-ventilation inside the laboratory; windows are located in one façade only	–Windows in the southwest façade with an area of 10.43 m², or 41.68% of the laboratory– The windows are not protected. The external pillars of the building help protect them, although this protection is ineffective	–The walls are made of 6-holed plastered bricks with a thermal transmittance of 2.48 W/m².°C– The external cladding is composed of small bricks (α= 0.5) with a solar heat factor of 4.96%– The thermal delay of the wall is 3.3 hours	–Roofing is composed of ceramic tiles with wood structure. Considering a roof of this type with a mixed slab (12 cm thickness), the thermal transmittance is 1.92 W/m².°C– The solar factor is 3.84% (α= 0.5)– The thermal delay of ceramic tile roofing with a mixed slab ceiling is 3.6 hours
ITL - B	–The building’s shape is not elongated– The façades have practically the same size– The building has three floors, and the laboratory is on the first floor	–There is no cross-ventilation inside the laboratory, windows located on one façade only	–Windows in the southwest façade with an area of 10.35 m², or 47% of the laboratory– Use of curtains as internal solar protection elements	–The external walls are made of solid exposed bricks, with a thermal transmittance of 3.70 W/m².°C; the internal walls are made of 6-holed plastered bricks with a 2.48 W/m².°C thermal transmittance– Considering that the external wall is made of exposed bricks (α= 0.5), its solar heat factor is 4.96%–The thermal delays of the solid brick and of the hollow plastered walls are 2.4 and 3.3 hours, respectively	–The thermal transmittance of a fiber-cement roof is 1.93 W/m².°C–The solar heat factor is 5.4% (α= 0.7, considering that the fiber-cement tiles are blackened)–The thermal delay of fiber-cement roofing with a mixed slab ceiling is 3.6 hours
ITL – C	–The largest façade is oriented to the northeast.–The building has a central courtyard and two floors, and the laboratory is on the east side of the ground floor; thus, it does not receive direct solar radiation in the afternoons	–There is no cross-ventilation inside the laboratory, windows located on one façade only	–Windows are in the southeast of the building with an area of 33.83 m², or 59.22% of the laboratory–Brises-soleils as external solar protection elements	–The walls are composed of 6-holed plastered bricks with a thermal transmittance of 2.48 W/m².°C–Considering that the walls are painted white (α= 0.2), their solar heat factor is 1.5%–The thermal delay of the wall is 3.3 hours	–The thermal transmittance of a fiber-cement roof is 1.93 W/m².°C–Solar heat factor of 1.5% (α= 0.2 because the fiber-cement tiles are white)–The thermal delay of fiber-cement roofing with a mixed slab ceiling is 3.6 hours


ITL – D	–The largest façade of the building is oriented to the northwest	–There is no cross-ventilation inside the laboratory; windows are located on one façade only	–Windows are in the northwest façade of the laboratory with an area of 15.21 m², or 43.89% of the laboratory	–The walls are composed of 6-holed plastered bricks with a thermal transmittance of 2.48 W/m².°C	–The thermal transmittance of a fiber-cement roof is 1.93 W/m².°C
	The building has an elongated shape, which increases the incident solar radiation because of its orientation	There are no window protections. The external pillars of the building help protect them, although this protection is ineffective	Considering that the walls are painted yellow (α= 0.3), their solar heat factor is 3%	The solar heat factor is 5.4% (α= 0.7, considering that the fiber-cement tiles are blackened)
	The building has three floors, and the laboratory is on the north side of the first floor; thus, it receives solar radiation in the northwest wall in the afternoons.	–The windows receive direct solar radiation	The thermal delay of the wall is 3.3 hours	The thermal delay of a fiber-cement roof with a mixed slab ceiling is 3.6 hours
ITL - E	–The building has an elongated shape, and it is oriented to the north, which reduces exposure to solar radiation	–There is no cross ventilation inside the laboratory; windows are located on one façade only	–The windows have an area of 7 m², or 40.79% of the laboratory	–The side walls are composed of PVC partition panels	–The thermal transmittance of a fiber-cement roof is 1.93 W/m².°C–Solar heat factor of 1.5% (α= 0.2 because the fiber-cement tiles are white)–The thermal delay of a fiber-cement roof with a mixed slab ceiling is 3.6 hours
	–Area of the laboratory: 61.68 m²	Brises-soleils proveide external solar protection
ITL – F	–The largest façade is oriented in the north-south direction	–There is cross ventilation; the building has windows on the north and south façades	–The windows are large and located on both the north and south façades, with an area of 17.85 m², or 61.78% of the laboratory in the north façade and an area of 14.5 m², or 42.37% of the laboratory, in the south façade	–The walls are made of 8-holed bricks with a thermal transmittance of 2.24 W/m².°C	–The thermal transmittance of a fiber-cement roof is 1.93 W/m².°C
	The buildings have an elongated shape and are far from each other, which reduces exposure to solar radiation and increases the exposure of the largest façades to prevailing winds	–Large eaves protect against direct solar radiation	The external cladding of the wall is made of beige ceramic tiles with α= 0.3; therefore, the solar heat factor is 2.7%	Solar heat factor of 1.5% (α= 0.2 because the fiber-cement tiles are white)
	The building has a ground floor only	However, the thermal delay of the walls is 3.7 hours	The thermal delay of a fiber-cement roof with a mixed slab ceiling is 3.6 hours–The slab has thermal insulation made of cork, which reduces the thermal transmittance and improves thermal performance

ITL - G	–The largest façades are in the north-south direction, and the building has an elongated shape; both features contribute to reducing exposure to solar radiation	–There is no cross-ventilation inside the laboratory; windows are located on one façade only	–The windows are on the south façade, with an area of 3.81 m², or 20.19% in that façade of the laboratory	–The walls are composed of 6-holed plastered bricks with a thermal transmittance of 2.48 W/m².°C	–The thermal transmittance of a fiber-cement roof is 1.93 W/m².°C
	–The building is far from other surrounding buildings, which improves air circulation	Considering that the walls are painted white (α= 0.2), their solar heat factor is 1.5%	Solar heat factor of 1.5% (α= 0.2 because the fiber-cement tiles are white)
	Building with ground floor only. The laboratory is on the west side; thus, it receives solar radiation in the afternoons, although the presence of vegetation and a low wall on the outside protects the building from solar radiation	The thermal delay of the wall is 3.3 hours	The thermal delay of fiber-cement roofing with a mixed slab ceiling is 3.6 hours

ITL - A = UFPI; ITL - B = UFSC; ITL - C = UnB; ITL - D = UNIVASF; ITL - E = USP; ITL - F = UFAM; ITL - G = MPEC.

The general result (Table 8) for each ITL was obtained from the architectural analysis. With this result, it was possible to verify whether the architectural design and the constituent materials of the ITLs had more positive than negative aspects, in other words, whether these factors contributed to the incidence of radiation from external sources. Accordingly, it could be observed whether the external heat sources contribute to the increase in thermal radiation in the environment.

Table 8

General result of the architectural analysis for each information technology laboratory (ITL)

ITL	Overall Result	Project Aspects
A	This laboratory is in the least favorable location in the building and receives solar radiation on one of its walls throughout the entire afternoon. The same occurs with the windows; because no protective elements are installed over the windows, they receive direct solar radiation, which contributes to the entry of thermal radiation.	NEGATIVE
B	The windows are large. Two different materials are used in the external walls, and the exposed bricks have high thermal transmittance and a high solar factor. The use of curtains suggests that the entrance of solar radiation through the windows is not favorable, considering that they are in the southwest façade.	NEGATIVE
C	The laboratory is in the east side of the building; thus, it does not receive solar radiation in the afternoons. The position of the windows is also favorable, and the solar protection helps reduce the incidence of radiation during the mornings. The solar heat factor of the fiber-cement roof is lower than those of the other laboratories because the slab is not blackened due to the low humidity in the city.	POSITIVE
D	The building is highly exposed to solar radiation because of its incorrect orientation, and the windows of the laboratory are in the northwest side; thus, they receive direct solar radiation, which constitutes the main source of external heat into the building. The use of external solar protection would be extremely important in this building, but there is none, and the use of dark sheets of paper on the inside of the windows attempts to reduce the discomfort caused by the heat passing through the windows. The solar heat factor of the fiber-cement roofing is higher than that of the other laboratories because it is blackened.	NEGATIVE
E	The north-south orientation reduces exposure to solar radiation. The protection provided by the brise-soleil and the vegetation also contributes to reducing the incidence of solar radiation. The solar heat factor of the fiber-cement roofing is low compared to the other locations because it is not blackened.	POSITIVE
F	The building is oriented in the north-south direction, which reduces exposure to solar radiation. In addition, the fiber-cement tiles and the cork insulation are well maintained. The windows are protected from direct solar radiation by large eaves and natural ventilation, with windows in both the north and south façades, which was considered in the building’s design.	POSITIVE
G	The north-south orientation minimizes exposure to solar radiation. The laboratory is on the west side of the building, although this façade is protected by vegetation and a low external wall. The construction materials have good thermal performance. Although the thermal transmittance of fiber-cement tiles is high, the ventilated air cushion between the roofing and the concrete ceiling reduces the heat flow to the room.	POSITIVE

The results of these analyzes showed that, in institutions A, B and D, projects had more negative aspects, thus finding that some element could be contributing to the exchange of heat by radiation in the internal environment, in addition to the internal heat sources (news ICT). Institutions C, E, F and G presented a positive result to the architectural project implicating that most of this heat exchange was probably coming only from internal sources of heat.

4.4. Mathematical modeling

The results of the heat exchanges due to radiation and of the architectural analysis showed that ITL G, the Paraíba Military Police Academy Educational Center in João Pessoa, was the ITL with the highest incidence of thermal radiation. Thus, a mathematical model was proposed to determine whether (t_g - t_a) and t_rm were related to overall student performance (D_t) in this ITL G.

It was observed that the data distribution of D_t was close to an exponential distribution (Fig. 2), which allowed the use of a model from the GLM class. The application of the Shapiro-Wilk test of normality to the dependent variable showed that the distribution was significantly different from a normal distribution (for a level of significance equal to 0.05). Therefore, the adequate fitted regression model was an inverse Gaussian model with a log link function, according to equation (3). $μ = exp (β_{1} . trm + β_{2} . (tg - ta))$ (3)

Fig.2

Frequency distribution of D_t – Sample G.

In addition, the presence of systematic deviations in the model was verified through an analysis of the adequacy of the distribution, the variance function, and the presence of inconsistent influential points. This analytical procedure of systematic deviations was repeated until there were no more inconsistent influential points. Every time an inconsistent influential point was removed, a new model was generated, and each systematic deviation was analyzed again. The fit of the final model without these points resulted in the estimates shown in Table 9.

Table 9

Estimated final model coefficients

Coefficient	Estimate	Standard Deviation	t value	Pr (> \|t\|)
T_rm	–0.0726	0.0063	–11.421	5.04e-15
(T_g-T_a)	–0.341	0.0850	–4.017	0.0002

These analyses confirm the model’s consistency. Thus, based on the estimates shown in Table 9, substituting coefficients β1 and β2 in the equation (3) results in the equation (4): $Dt = e^{- 0.073 Trm - 0.341 (tg - ta)}$ (4)

It is important to note that in Table 9, the p-values are <0.05 but the pseudo R² is low (0.1548). However, regarding the fit of GLMs, Cordeiro and Demétrio [38] highlight that, in fact, a favorable result of the likelihood ratio test (for a certain level of significance) can be considered evidence that a model has a reasonable fit to the data. Thus, considering that the p-value of the likelihood ratio test is <0.001, equation (3) is well fitted.

Based on model (4), for a constant mean radiant temperature (t_rm), student performance is reduced by approximately 29% for each 1°C increase in the difference between the globe temperature and the air temperature (t_g - t_a). This statement is confirmed in Fig. 3 given that by setting three values for the mean radiant temperature (23°C, 25°C and 28°C) and varying (t_g - t_a) between 1°C and 4°C, student performance tends to decrease with the increase in (t_g - t_a).

Fig.3

Performance as a function of T_g - T_a .

5. Discussions

The globe temperature was higher than the air temperature in laboratories A and B on the first two days of data collection; therefore, there was heat exchange due to radiation. Considering that in the ITLs of both institutions the project and its constituent elements were considered negative (Table 8), both the internal and external thermal conditions could be responsible for the difference between the globe and the air temperatures. In the case of ITL D, the globe temperature was higher than the air temperature on the three data collection days. As the project and the constituent materials of this laboratory were also considered negative, there were internal (new ICT) and external factors that resulted in the increase in thermal radiation in the environment. The temperature of the internal surface of the external wall of buildings without adequate insulation and materials are affected by the external conditions, particularly solar radiation. Therefore, the temperature of the internal surface of these walls increases considerably, which increases the thermal radiation in the room.

Heat exchange due to radiation was observed in ITL E. Considering that the results of the architectural analysis of this project indicated that there were more positive than negative aspects, the implication is that most of the heat exchanged likely originated from internal heat sources. In the cases of laboratories C and F, all globe temperatures collected were lower than the air temperature, indicating that there was no heat exchange due to radiation. This finding, together with the good performance of the materials, indicated that the projects had more positive characteristics in both ITLs. Therefore, there was no incidence of thermal radiation originating in the internal and external environments of these ITLs.

ITL G had the highest heat exchange due to radiation, given that t_g > t_a. However, the results of the architectural analysis indicated that the project had more positive aspects. Its north-south orientation minimized exposure to solar radiation and the good thermal performance of the construction materials reduced the incidence of outside radiation. According to Atmaca et al. [39], when a building’s envelope contains adequate insulation and materials, only small internal and external heat gains occur due to solar radiation, considering that the temperature of the internal surface of the walls helps protect from external environmental conditions. Therefore, in this specific case, the number of laptops (40 units) and other heat sources in the ITL of Institution G most likely combined to increase the globe temperature, considering that the mean radiant temperature was higher than the globe temperature by 2.25°C $\frac{h_{cg}}{ɛ_{g σ}}$ (according to equation 1 and the data in the Table 5).

It was found, through the mathematical model, that with each increase of a degree in the difference between the globe temperature and the air temperature (T_g-T_a), the students’ cognitive performance in the computer lab of institution G decreased by around 29%. Therefore, it is likely that in an air-conditioned room with different heat sources such as new ICT, students, walls, and surfaces, where t_g > t_a, these factors may influence the heat exchange due to radiation between the students and their environment.

6. Limitations

With the development of modern society, it is observed that individuals are in most of their time in indoor or closed environments, which has become a concern of several researchers in the search to investigate the conditions of environmental comfort during the activities of workers in the their workspaces. One of the concerns in closed environments is the thermal conditions, and if not controlled can generate a certain effort, thus overloading the human organism and, consequently, interfering in the performance.

In this work, students perform their cognitive activities in information technology laboratories located in the different Brazilian regions. Thus, the climatic differences in Brazil provide different acclimations, so the students participating in this research are acclimated to the climatic specificities of their respective regions, which may interfere in the determination and analytical interpretation of comfort with reflection in the underestimation of some results of the cognitive tests performed students during the collection of thermal data.

Another limitation of the study was about the architectural analysis. When analyzing the relationships between the globe temperature and the air temperature in a constructed environment, especially when there are several sources of heat in this environment, it is important to evaluate the thermal characteristics of materials and the elements and constructive components of the building, because for some building is essential to obtain knowledge about the transfer of heat from the external environment.

In this sense, in the specific article, only a succinct architectural analysis was carried out in accordance with the precepts of NBR 15220 - Thermal Performance of Buildings (2003) to have better understand the possible thermal behavior of buildings, especially regarding differences in air temperature and globe temperature with possible increase of T_rm.

7. Conclusions

Several factors contribute to increasing thermal radiation in a work environment. They include (1) the external thermal conditions – climatic change; (2) the architectural design; (3) the building envelope materials; (4) the communication and information technology equipment installed in the room; and (5) the generation of body heat. Thus all these parameters can contribute to the increase of thermal radiation in an air-conditioned environment. Therefore, urban and human aspects must be considered when heat exchange due to radiation and the environment is investigated [40].

All of these parameters may contribute to increasing thermal radiation in air-conditioned environments, which in this study were ITLs at Brazilian educational institutions where the thermal radiation was investigated through the behavior of the globe temperature relative to the air temperature and by observing the variation in the mean radiant temperature while students performed cognitive tasks. It was found that on the three consecutive days when the air temperature was 20, 24, and 30°C, institutions C and F were the only ITLs in which there was no heat exchange due to radiation between students and the environment, which did occur in institutions A, B, D, E, and G. The heat exchange in the laboratory of institution G was the highest, with the mean radiant temperature being higher than the globe temperature by 2.25°C $\frac{h_{cg}}{ɛ_{g σ}}$ at any given time.

The possible causes were assessed and it was found from the analysis of the architectural elements that the design of the building containing laboratory G had more positive than negative aspects, considering the north-south orientation and the good thermal performance of the construction materials. Therefore, the internal heat sources in the laboratory may have led to an increase in thermal radiation, given the substantial numbers of students and laptops. To confirm these findings, the mathematical model for this ITL showed that for each 1°C increase in the difference between the globe temperature and the air temperature, the performance of the students in the ITL decreased by approximately 29%. This result demonstrates that a higher mean radiant temperature results in a higher thermal radiation in this laboratory and this increase can affect student performance. In this way, as well as productivity can be altered due to changes in air temperature in air-conditioned teaching environments, as already proven by international publications, in this specific case, if t_g> >t_a, possibly the thermal radiation may interfere with the performance of the people present in the environment technological innovations of communication and information.

Conflict of interest

None to report.

References

Wong

, Mui

, Cheung

. Bayesian thermal comfort model. Building and Environment. 2014;82:171–9.

Silva

. Análise sobre conforto térmico e desempenho nos ambientes de ensino com video display terminals (VDT): Estudos de multicasos no Nordeste do Brasil e Norte de Portugal. Projeto de pesquisa submetido ao Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), vinculado ao MCTI/CNPq, 2013.

Cichowicz

, Lewandowska

. The analysis of selected parameters of thermal comfort in the classrooms using CFD technique. World Scientific News. 2017;73(1):72–9.

Muzammil, Mohammad

et al. Design of a work station and its evaluation under the influence of noise and illumination for an assembly task. Work. 2011;39(1):3–14.

Smith

. Designing learning environments to promote student learning: Ergonomics in all but name. Work. 2013;44:39–60.

Fanger

Thermal comfort - analysis and applications in environmental engineering. United States: McGraw-Hill, 1972. pp. 244.

Niemeyer, Labaki. Environmental Perception in Open Public Spaces: Application of a Participatory Methodology. Educational Research Journal. 2015;1:1–9.

Rupp

, Vásquez

, Lamberts

. A review of human thermal comfort in the built environment. Energy and Buildings. 2015;105:178–205.

Pivetta

. Extremos do clima. Pesquisa FAPESP. 2013;210:16–21.

10.

Yang

, Becerik-Gerber

, Mino

. A study on student perceptions of higher education classrooms: Impact of classroom attributes on student satisfaction and performance. Build Environ. 2013;70:171–88.

11.

Arslanoglu

, Yigit

. Experimental and theoretical investigation of the effect of radiation heat flux on human thermal comfort. Energy and Buildings. 2016;113:23–9.

12.

De Dear

, Akimoto

, Areansn

, Brager

, Candido

, Cheong

KWD

. Progress in thermal comfort research over the last twenty years. Indoor Air. 2013;23(6): 442–61.

13.

Van Hoof

. Forty years of Fanger’s model of thermal comfort: Comfort for all? Indoor Air. 2008;18:182–201.

14.

Halawa

, Hoof

, Soebarto

. The impacts of the thermal radiation field on thermal comfort, energy consumption and control—A critical overview. Renewable and Sustainable Energy Reviews. 2014;37:907–18.

15.

Coutinho

. Proteção contra o calor. In: Mattos

U. A. de O.

Másculo

F. S.

(Eds.); Higiene e segurança do trabalho. Rio de Janeiro: Elsevier 2011, pp. 209–231.

16.

Budd

. Wet-bulb globe temperature (WBGT) - its history and its limitations. J Sci Med Sport. 2007;11:20–32.

17.

Oliveira

AVM

, Gaspar

, Raimundo

, Quintela

. On the Measurement of Globe Temperatures: Analysis of the Influence of Different Parameters. Extreme Physiology & Medicine. 2015;4:A14. doi: 10.1186/2046-7648-4-S1-A14

18.

d’Ambrosio Alfano

, Malchaire

, Palella

, Riccio

. WBGT index revisited after 60 years of use. Ann Occup Hyg. 2014;58(8):955–70. doi: 10.1093/annhyg/meu050

19.

Quintela

, Oliveira

AVM

, Cardoso

, Gaspar

, Raimundo

. Assessment of the performance of globe thermometers under different environmental conditions. Occupational Safety and Hygiene II, Taylor & Francis: London, Eds: Arezes PM, Baptista JS, Barroso MP, Carneiro P, Cordeiro P, Costa N, Melo RB, Miguel AS, Perestrelo G, 2014, 523–528. ISBN 978-1-138-00144-2.

20.

Vasconcelos

PEM

, DaSilva

, Coutinho

. Relationship between thermal variables and performance. A study with students of the Military Police Academy of the State of Paraíba Publishing company: Novas Edições Academicas, 2015, ISBN: 978-3-639-83019-4.

21.

Langner

, Scherber

, Endlicher

. Indoor heat stress: An assessment of human bioclimate using the UTCI in different buildings in Berlin. Die Erd. 2013;144:260–73.

22.

Walikewitz

, Janicke

, Langner

, Meier

, Endlicher

. The difference between the mean radiant temperature and the air temperature within indoor environments: A case study during summer conditions. Building and Environment. 2015;84:151–61.

23.

Alfano

FRA

, Isola

, Palella

. On the measurement of the mean radiant temperature and its influence on the indoor thermal environment assessment. Building and Environment. 2013;63:79–88.

24.

Akimoto

, Tanabe

, Yanai

, Sasaki

. Thermal comfort and productivity-Evaluation of workplace environment in a task conditioned office. Building and Environment. 2010;45:45–50.

25.

Siqueira

JCF

, Da Silva

, Coutinho

, Rodrigues

. Analysis of air temperature changes on blood pressure and heart rate and performance of undergraduate students. Work 2017;57:54–43.

26.

ASHRAE. ANSI/ASHRAE Standard 55-2010. Thermal environmental conditions Refrigerating and Air-Conditioning Engineers, Inc.; 2010.

27.

ISO 7726: Ergonomics of the thermal environment - Instruments for measuring physical quantities. Geneva. 2002.

28.

ISO 7730: Ergonomics of the thermal environment. Analytical determination and interpretation of thermal comfort using calculation of the PMV and PPD indices and local thermal comfort criteria, Geneva. 2005.

29.

Baumgartl

VDO

, Nascimento

. A Bateria de Provas de Raciocínio (BPR-5) aplicada a um contexto organizacional. Psico-USF. 2004;9(1):1–10.

30.

Sbc, S. B

; Sbh, S. B De

; Sbn, S. B De

N. VI

. Diretizes Brasileiras de Hipertensão Arterial. Arquivo Brasileiro de Cardiologia. 2010;95:1–51.

31.

Almeida

, Primi

. Perfis de capacidades cognitivas na bateria deprovas de raciocínio (BPR-5): Capacidades cognitivas eBPR-5. Psicologia Escolar e Educacional. 2004;8(2): 135–44.

32.

Siqueira

JCF

. Avaliação de parámetros cardiovasculares e desempenho cognitivo de estudantes de ambientes inteligentes de ensino submetidos á variação de temperatura do ar. Dissertação apresentada ao Programa de Pós-Graduação em Engenharia de Produção do Centro de Tecnologia da Universidade Federal da Parâiba. 2015.

33.

Associação Brasileira de Normas Técnicas - ABNT. NBR 15220 - Desempenho térmico de edifiçãcoes. Rio de Janeiro: ABNT, set. 2003.

34.

Roriz M. Conforto Termico em Edificações: Um Modelo Matemático e uma Aplicação 1996. pp. 185. Tese (Doutorado) - Faculdade de Arquitetura e Urbanismo, Universidade de São Paulo, São Paulo, 1996.

35.

Dobson

, Barnett

. An introduction to generalized linear Models. 3ª Edicao. CRC Press .Taylor & Francis Group. 2008.

36.

Lamberts

, Ordenes

and Guths

Transferencia de calor na envolvente da Edificacao. Universidade Federal de Santa Catarina, Florianopolis. 2008.

37.

Oliveira

EAS

. Análise da Relação entre os Parámetros Arquitetônicos, o Conforto Térmico e a Produtividade em Escritórios com Ventilação Natural. 115 p. Dissertação (Mestrado em Engenharia de Produção) - Universidade Tecnológica Federal do Paraná. Ponta Grossa, 2015.

38.

Cordeiro

, Demétrio

CGB

. Modelos Lineares Generalizados e Extensoes. Piracicaba-SP, 2008.

39.

Atmaca

, Kaynakli

, Yigit

. Effects of radiant temperature on thermal comfort. Building and Environment. 2007;42:3210–20.

40.

Park

. Human-urban radiation exchange simulation model. A Dissertation Submitted in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in the Department of Geography, University of Victoria, 2011, pp. 220.