Relationship between Self-Assessed Productivity, Gender and Age in Mixed-Mode and Fully Air-Conditioned Offices in Florianópolis, Brazil

Abstract

The quality of the indoor environment and anthropometric parameters influence the users’ productivity in a building. This study aims to evaluate the influence of environmental and anthropometric characteristics on the self-assessed productivity of office building users in the humid subtropical climate of Florianópolis, Brazil. Three hybrid buildings equipped with air-conditioning and natural ventilation and one centrally air-conditioned building were considered. Indoor environmental data were obtained by means of measurements. Electronic questionnaires collected anthropometric characteristics and the perception of productivity by the users. The analysis of the users’ performance was performed using box diagrams. The results showed that women and occupants over 50 years old preferred high temperatures during the use of natural ventilation. When air-conditioning was on, the perception of productivity concerning the operative temperature did not show significant differences between anthropometric variables. It was concluded that the use of air-conditioning reduced the influence of anthropometric characteristics on self-assessed productivity and that people maintained their work performance despite the higher internal temperature variations during natural ventilation operation.

1. Introduction

Productivity, in economic terms, refers to the efficiency in manufacturing processes and services, usually assessed according to the number of people and quantity of resources allocated to a certain activity [1]. Developing a workspace in which productivity is stimulated and favoured is crucial to the output and competitiveness of any venture. By systematically enhancing productivity, lower unemployment rates can be achieved; however, carelessly searching for higher productivity (i.e., high job demands, job insecurity and the use of information and communication technology) may reduce workers’ well-being [2,3]. Paradoxically, lower levels of worker well-being may imply lower productivity; thus, working conditions must be prioritised so that enhancing productivity does not result in a negative impact.

The indoor environmental quality associated with psychosocial aspects and the kind of labour directly impacts the physical and psychological conditions of occupants, which in turn affects the worker’s satisfaction and comfort perception [4]. Thus, analysing environmental conditions and their effects on occupants is essential in shaping a healthy and optimal workspace. Through multiple linear regressions, Chen et al. [5] demonstrated that satisfaction with quality aspects of the internal environment is the most significant predictor of perceived productivity. Kang et al. [6] evaluated the quality of environments in open-plan shared offices in southern China during winter and found a positive correlation between the layout, air quality, thermal environment, lighting, acoustic environment and productivity.

Regarding thermal sensation, according to Cui et al. [7] and Geng et al. [8], productivity is optimised when occupants feel neutral or slightly cold; their performance does not change significantly when these perceptions of the environment are reported. Cui et al. [7] conducted their study in a climate-controlled chamber in China, with temperatures between 22 °C and 32 °C. The experiments of Geng et al. [8] were conducted in controlled environments in Beijing, with temperatures between 16 °C and 28 °C. Similarly, Seppanen et al. [9] reported lower productivity at temperatures above 25 °C, which was defined as a comfortable temperature. According to research by Tarantini et al. [10], productivity is negatively affected in non-neutral comfort conditions, mainly if users feel warm; however, the degree depends on the type of activity performed and other environmental qualities, such as air quality.

Among indoor environmental characteristics, some studies relate productivity to thermal comfort using temperature as an analysis factor [8,11]. According to some authors, the ideal temperature range varies in an interval determined by the activity carried out and personal characteristics, such as gender, age, clothing levels and birthplace [6,12]. In their cross-continental study, Chen et al. [5] found that the second most important determinant of perceived productivity, after satisfaction with quality aspects of the internal environment, is the country of residence. According to the authors, people from different countries hold different beliefs about how the quality of the indoor environment (e.g., lighting and acoustics) can influence productivity. These different beliefs may be due to the characteristics of the building and the workplace context. Schellen et al. [13] developed a study on the perception of thermal comfort, productivity and physiological responses in the face of moderate temperature variations in a climate room in the Netherlands. The authors reported that older people (67 to 73 years) prefer higher temperatures than younger people (22 to 25 years). However, the results of the study showed that temperature variations of 2 °C/h (up or down) in the range of 17 °C to 25 °C do not cause a loss in performance. In a study of artificially ventilated offices in Qatar, India and Japan, Indraganti and Humphreys [14] found that women experienced non-neutral thermal conditions more often than men in cold and hot conditions. This finding is important since thermal sensitivity can imply changes in productivity. Similarly, in an analysis of data surveyed in the literature, Kim et al. [15] reported that women’s satisfaction levels were consistently lower than men’s for all indoor environment quality factors assessed, such as temperature and air quality.

In this context, this study aims to understand how productivity is related to environmental and anthropometric conditions in office buildings in Florianópolis. Given that many studies are conducted in climate chambers or specific buildings, this study also aims to better comprehend the influence of building type (mixed-mode and fully air-conditioned) on productivity, considering the operation mode. This research is justified in understanding how resources can be allocated to optimise productivity, facing environmental issues and, thus, the competitiveness of any initiative. Based on previous studies and methods, adaptations were made to assess the context of the city of Florianópolis to contribute to the region’s energy efficiency literature and subsidise decision making in buildings that aim for a higher performance.

2. Materials and Methods

This research was conducted in the city of Florianópolis, located in the humid subtropical climate of southern Brazil. The average monthly outdoor temperature varies between 17.1 °C and 25.1 °C, while the monthly maximum temperature varies between 21.5 °C and 28.2 °C and the monthly minimum temperature varies between 17.1 °C and 25.1 °C (Figure 1). The annual rainfall index is 1506 mm, with rains evenly distributed throughout the year, resulting in no dry seasons but with predominance in warmer months. The data used in this research were collected between 2014 and 2016 by Rupp [16] in four office buildings: three were mixed-mode buildings, while the fourth was fully air-conditioned. In this study, the mixed-mode buildings were named H1, H2 and H3, while the fully air-conditioned one was named CC.

Table 1. Characteristics of buildings H1, H2, H3 and CC.

Characteristic	Building H1	Building H2	Building H3	Building CC
Total area (m2)	6200	8244	3090	27,432
Building shape	Blocks with rectangular and H shape	Rectangular	Rectangular	Square
Construction type	Reinforced concrete	Reinforced concrete	Reinforced concrete	Reinforced concrete
Walls	Exposed bricks	Plastered and painted beige masonry	Apparent concrete	Apparent concrete
Glass type	Clear	Clear	Clear with film	Tinted glass
External shading	-	Opaque facade element with shutters for ventilation	Recessed glass concerning the façade	Metal louvers with manual control
Internal shading	Most spaces have vertical blinds operated by users	-	Vertical blinds operated by users	-
Air-conditioning	Split	Split	Split	Central
Occupants number	320	350	250	1200
Working hours	7 a.m.–6 p.m.	1 p.m.–7 p.m.	8 a.m.–6 p.m.	7 a.m.–7 p.m.
Season survey	All seasons	Winter	All seasons	All seasons

shows the general characteristics of the buildings.

In all buildings, standard criteria were considered, such as mainly office activities (1.0–1.2 met), the freedom to choose clothing of preference and a disregard of access and transition spaces. In the fully air-conditioned building, it was required that the rooms considered in the research used the HVAC throughout the year and that there were no windows operable by the occupants. The central air-conditioning system has a temperature setpoint equal to 24 °C, and users have no control over the environmental conditions. In the mixed-mode buildings, the HVAC system and the windows were controlled manually by the occupants, according to their preferences.

Outdoor environmental conditions were obtained from a meteorological station owned by INMET (National Institute of Meteorology). Data referred to all seasons. Indoor environmental conditions were collected using microclimate stations, measuring air temperature, globe temperature, air velocity and relative humidity. Using portable anemometers, it was possible to measure the air temperature and air velocity in sites close to windows, portable fans or air-conditioners.

Self-assessed data were collected via electronic questionnaires developed in the Java programming language, available to subjects only during the period of the field experiment. When the electronic questionnaire was run for the first time on the personal computer, it opened a window with options for scheduling the initial appointment. Users were instructed to schedule the start of the questionnaire in the morning at 9:00 a.m. and in the afternoon at 2 p.m. In the questionnaire, first, the users informed their personal and anthropometric data, e.g., gender, age and clothing. After that, users evaluated their productivity and sick building syndrome and suggested actions to improve thermal conditions. Regarding self-assessed productivity, users were asked to estimate how they felt about their productivity at work on the day of the test compared to normal, whether it increased or decreased, using a scale of values from −4 (40% reduction in productivity) to 4 (40% increase in productivity). The different response options regarding the perception of self-evaluated productivity are detailed in

Table 2. Response options regarding the perception of self-evaluated productivity.

Question	Response	Code
Productivity compared to a normal day	40% increase	4
	30% increase	3
	20% increase	2
	10% increase	1
	Neutral	0
	10% reduction	−1
	20% reduction	−2
	30% reduction	−3
	40% reduction	−4

. During the collection period, 1247 responses on perceived productivity were obtained. The productivity data collection was performed through self-assessed productivity, so as not to disturb the participants as they worked normally during the field studies.

In addition to the information collected through the microclimate stations, complementary parameters were calculated based on the collected data. The mean radiant temperature and the operative temperature were obtained through Equations (1) and (2) [18], respectively. The operative temperature considers the heat transfer by radiation between the occupant’s body and the environment and is used to determine the acceptability of the internal conditions of the environment.

$${\mathrm{T}}_{\mathrm{r}}={\left[{\left({\mathrm{T}}_{\mathrm{g}}+273\right)}^4+2.5\times {10}^8·{{\mathrm{V}}_{\mathrm{a}}}^{0.6}·\left({\mathrm{T}}_{\mathrm{g}}-{\mathrm{T}}_{\mathrm{a}}\right)\right]}^{1/4}-273$$

(1)

where ${\mathrm{T}}_{\mathrm{r}}$ is the mean radiant temperature (°C); ${\mathrm{T}}_{\mathrm{g}}$ is the globe temperature (°C); ${\mathrm{V}}_{\mathrm{a}}$ is the air velocity (m/s); and ${\mathrm{T}}_{\mathrm{a}}$ is the air temperature (°C).

$${\mathrm{T}}_{\mathrm{o}}=\mathrm{A}·{\mathrm{T}}_{\mathrm{a}}+\left(1-\mathrm{A}\right)·{\mathrm{T}}_{\mathrm{r}}$$

(2)

where ${\mathrm{T}}_{\mathrm{o}}$ is the operative temperature (°C). A is 0.5 when ${\mathrm{V}}_{\mathrm{a}}$ is lower than 0.2 m/s; 0.6 when ${\mathrm{V}}_{\mathrm{a}}$ is between 0.2 and 0.6 m/s; and 0.7 when ${\mathrm{V}}_{\mathrm{a}}$ is between 0.6 and 1.0 m/s.

The relationships between productivity and operative temperature and productivity and anthropometric variables (gender and age) were analysed using box plots. The box plots divide the data representation into five primary types of information: the minimum value, the first quartile, the median, the third quartile and the maximum value. The chart’s amplitude represents the data dispersion observed through the minimum and maximum values. The central rectangle is where 50% of the values in the data set are concentrated, and the position of the median line in the rectangle represents the symmetry or asymmetry of the distribution.

3. Results and Discussion

3.1. Descriptive Statistics of the Environmental Parameters

Table 3. Descriptive statistics of the environmental parameters of buildings H1, H2, H3 and CC.

Parameter	Ta (°C)	To (°C)	Tr (°C)	Va (m/s)	RH (%)
Building H1 (n = 251)
Average	23.2	23.3	23.3	0.14	61
Standard deviation	1.9	1.9	1.9	0.07	10
Minimum	17.0	17.8	18.2	0.10	34
Maximum	27.0	27.0	27.1	0.55	80
Building H2 (n = 97)
Average	24.0	23.9	24.0	0.18	60
Standard deviation	1.0	0.9	0.8	0.06	9
Minimum	22.0	22.3	21.9	0.10	42
Maximum	25.0	25.5	25.9	0.30	76
Building H3 (n = 588)
Average	24.0	24.2	24.3	0.11	62
Standard deviation	1.2	1.1	1.1	0.07	8
Minimum	20.4	21.6	21.3	0.10	37
Maximum	26.7	26.8	26.9	1.10	83
Building CC (n = 311)
Average	23.3	23.3	23.3	0.12	61
Standard deviation	0.9	0.9	1.1	0.06	8
Minimum	21.5	21.7	21.2	0.10	43
Maximum	26.5	25.9	26.8	1.00	76

Note: n is the number of participants in the survey.

shows the descriptive analysis of the environmental parameters collected by the microclimate stations and parameters calculated according to Section 2.

Generally, there is no notable variation between the environmental parameters of the four buildings. More specifically, concerning air and the operative and radiant temperature, it can be highlighted that buildings H2 and CC showed narrower ranges of values (a smaller data dispersion). In building H2, data collection occurred only during winter; in building CC, air-conditioning is continuously used. In hybrid buildings, wider temperature ranges were observed. As for air velocity, building H2 presented the best-fitted range, followed by buildings H1, CC and H3. Finally, the highest dispersion and the widest range of relative humidity data were found in building H1.

3.2. Relationship between Productivity and Operative Temperature

Figure 2, Figure 3, Figure 4 and Figure 5 show the box plots relating productivity to operative temperature in buildings H1, H2, H3 and CC, respectively.

In building H1, when using air-conditioning, the best productivity evaluations occurred at operative temperatures of 24 °C and 25 °C. At 23 °C, a high dispersion of the productivity votes was observed, encompassing the entire spectrum of increased and decreased productivity, with a slight tendency to increase productivity. The temperature of 26 °C was the only one that showed a predominance of productivity reduction votes. During natural ventilation, the worst productivity evaluations were in the lower temperature range, between 17 °C and 19 °C. From 20 °C, the majority of votes reported an increase in productivity. It is possible to verify that for all the temperatures observed in both situations, the productivity evaluation was better using natural ventilation than air-conditioning, except for 24 °C, whose distribution was identical in both cases.

In building H2, air-conditioning was only used a few times, given the low sampling and limitation of records during the winter. Among the temperatures recorded, 23 °C and 24 °C showed good productivity results, with medians equivalent to a 20% or more productivity increase. At the operative temperature of 25 °C, the productivity evaluation was lower, with votes ranging from 0 to 1. During natural ventilation, all temperature records showed a trend towards increased productivity.

In building H3, when using air-conditioning, an increase in productivity was verified in all recorded temperatures, except for 25 °C, which showed a balance between reductions and increases in productivity (median in the “0” vote). The temperature of 21 °C was the only one in which there were only votes for productivity equivalent to normality and positivity. Temperatures of 22 °C and 23 °C maintained the same median as 21 °C but with a higher dispersion of votes, encompassing even the spectrum of reduced productivity. From this, it is possible to observe a trend of reduced productivity when exceeding the operative temperature of 23 °C. During natural ventilation, the only temperatures representing increased productivity were 21 °C and 26 °C. For the temperature range of 22 °C to 25 °C, the productivity was evaluated more positively with air-conditioning than with natural ventilation.

In building CC, productivity was best evaluated at the lowest temperature, 21 °C, with a 10% productivity increase median. The temperatures of 22 °C and 23 °C showed a balance between positive and negative productivity evaluations, while the two highest temperatures in the interval, 24 °C and 25 °C, tended to increase productivity. It can be seen that, except for the minimum temperature observed in the interval, the increase in temperature increased productivity.

The operative temperatures at which the highest number of productivity votes were recorded differed in each building. In building H1, productivity was better evaluated during intermediate temperature values (24 °C and 25 °C). In H3, productivity was inversely related to temperature increases (more productivity votes between 21 °C and 23 °C). In building CC, the 24 °C and 25 °C temperatures had the best productivity ratings. In H2, due to the low sampling, it is not possible to establish a productivity trend regarding the operative temperature.

In mixed-mode buildings, during the use of air-conditioning, there was a tendency to reduce productivity at temperatures higher than 24–25 °C, a similar trend to that observed by Seppanen et al. [9], who found that for each degree Celsius variation above 25 °C, productivity was reduced by up to 2%. This trend was also observed by Subramanian et al. [19], who established 23.79 °C as the ideal temperature for the work environment. Lipczynska et al. [20] found that the self-reported productivity of office users in Singapore was similar at 23 °C and 26 °C. According to the authors, productivity is maintained at slightly higher temperature values than those found by this research, which may be due to the adaptation of users to warmer climatic conditions.

However, Lan et al. [21] noted no differences between productivity assessments in a study with controlled room temperature in three different situations (23 °C, 26 °C and 27 °C), concluding that, by providing shared control of cooling devices, room temperature can be increased without reducing productivity. The results of this study can be related to the results obtained in productivity evaluations in mixed-mode buildings during natural ventilation, where, even at the highest temperatures, no drop in self-assessed productivity was observed.

3.3. Relationship between Productivity and Anthropometric Variables

3.3.1. Gender

Figure 6, Figure 7, Figure 8 and Figure 9 present the box plots relating productivity to operative temperature and the gender categories according to the operating mode for buildings H1, H2, H3 and CC, respectively.

In building H1, women generally showed better productivity during air-conditioning use, with a positive median at all temperatures except at 26 °C. However, men more often presented median votes referring to neutrality. At 24 °C, the best performance occurred, with a median of vote 1; at 26 °C, the worst performance occurred, with a median between the neutrality vote and the 10% productivity reduction vote.

During natural ventilation, women tend to decrease productivity as the temperature increases from 20 °C to 23 °C. From 24 °C on, the productivity evaluation presented its best performance. For men, the worst performance was recorded during the lowest temperatures (17 °C and 18 °C). At the other recorded temperatures, except 24 °C, the median of the votes was positive.

In building H2, during air-conditioning use, women recorded positive productivity at 23 °C and 24 °C and negative at 25 °C. Men recorded positive productivity at both 23 °C and 24 °C. During natural ventilation, women showed stable performance throughout the temperature range (22 °C to 25 °C). However, men tended to show decreased productivity when the operative temperature increased from 22 °C to 24 °C, concentrating votes near neutrality at 25 °C.

In building H3, when air-conditioning was on, women tended to reduce productivity from 23 °C, resulting in median neutrality at 25 °C. However, men’s productivity showed almost no change between 21 °C and 25 °C. Intermediate temperatures (23 °C and 24 °C) were associated with decreasing productivity, bringing the median close to the neutrality vote. During natural ventilation, women tend to increase productivity from 22 °C to 26 °C, where the median vote was equivalent to a 20% increase in productivity. Conversely, men presented stability in the productivity evaluations from 22 °C to 24 °C, and for higher temperatures, they presented a tendency to decrease productivity.

In building CC, women showed stable performance in the temperature range recorded, with the best evaluation at 24 °C and the worst at 25 °C. However, men performed considerably better at the highest temperatures (24 °C and 25 °C) and worst at 22 °C and 23 °C, with median votes at neutral.

Based on the results, considering natural ventilation, women reported better productivity at higher temperature ranges (around 25 °C). However, the operative temperature had less influence on men’s productivity, which remained almost stable at intermediate temperatures, with a possible tendency for reduced productivity at temperatures higher than 25 °C. These results corroborate with Indraganti and Humphreys [14], who found that women report non-neutral thermal conditions more often than men since the literature suggests that self-reported productivity is higher in thermal-neutral environments. These findings also agree with Chen et al. [5], who noted that women are more likely to consider the quality of the internal environment when assessing productivity. When air-conditioning is on, the results suggest a reduction in women’s productivity at temperatures above 25 °C. However, in this operation mode, the trends were less evident for both genders.

3.3.2. Age

The anthropometric variable age was divided into categories to facilitate data analysis. Rupp et al. [22] proposed the categorisation used here and divided age into two categories: over 50 years and under 50 years. Figure 10, Figure 11, Figure 12 and Figure 13 show the box plots relating productivity to operative temperature and age categories according to the operation modes for buildings H1, H2, H3 and CC, respectively.

In building H1, considering air-conditioning, there is a tendency to evaluate better productivity for intermediate operative temperatures in both age categories. For people under 50 years old, this optimal temperature range occurs between 24 °C and 25 °C, where the votes have a median in vote “1” and encompass the entire range of positive votes. For people over 50 years old, the optimal temperature range occurs at 23 °C and 24 °C, with a median in vote “1” and encompassing the entire positive productivity scale.

With natural ventilation, most recorded temperatures were associated with positive productivity in both age categories. For people over 50 years old, the lowest temperatures were associated with reduced productivity. For temperatures above 20 °C, there is a tendency to increase productivity, except at 23 °C.

In building H2, when air-conditioning was on, people under 50 recorded better productivity during the two lowest recorded temperatures, 23 °C and 24 °C. People over 50 years old recorded only votes referring to a 20% increase in productivity. Considering natural ventilation, people under 50 years old tend to progressively decrease productivity with an increased operative temperature during the entire range of operative temperatures recorded. However, for people over 50 years old, the median productivity (associated with positive votes) presents low variation for the three lowest temperatures recorded. The highest recorded temperature (25 °C) was associated with the best productivity evaluation.

In building H3, considering air-conditioning, people below 50 years old tended to maintain positive productivity levels for operative temperatures between 21 °C and 23 °C. From 23 °C onwards, productivity tended to decrease as the temperature increased until 25 °C. For people over 50 years old, there is a low variation in productivity throughout the recorded temperatures.

With natural ventilation, the temperatures that limited the recording interval (21 °C and 26 °C) were associated with the best productivity evaluation for people under 50. Intermediate temperatures showed productivity variation. For people over 50 years old, all operative temperatures showed a median neutral or positive vote, with no clear tendency.

In building CC, for people under 50 years old, the median of the votes showed low variation. The only operative temperature in which no negative votes were recorded was 21 °C. From this, the increase in temperature was associated with a reduction in the range of negative votes observed, indicating a tendency for productivity to increase as the temperature rises. For people over 50 years old, the highest temperature recorded, 25 °C, showed the best productivity evaluation, with the median in the 30% increase in productivity vote. The lowest temperature recorded, 22 °C, was the one that showed the highest number of votes related to productivity reduction.

With natural ventilation, older users showed a higher self-assessed productivity at higher temperatures, around 25 °C. Considering air-conditioning, the results showed that the productivity of people under 50 could decrease above certain temperatures, with the threshold temperature varying between 23 °C and 25 °C, depending on the building. The results corroborate with Schellen et al. [13], who reported the preference of older people for higher temperatures; however, unlike the study of those authors, in this study, the productivity of users in this study was influenced by the temperature. This difference can be understood given the users’ different countries of residence, which is a determining factor in the perception of productivity [5]. For example, Batiz et al. [23] also found a better performance in thermal comfort conditions in southern Brazil. Furthermore, productivity may be influenced by other variables, e.g., indoor air quality, lighting and acoustic aspects [6,10].

4. Conclusions

This study evaluated the influence of environmental and anthropometric variables (gender and age) on the self-assessed productivity of users in office buildings in Florianópolis, southern Brazil. The analyses counted on 1247 responses from three hybrid buildings, possibly using natural ventilation or air-conditioning, and one building with a central air-conditioning system.

Based on the results about productivity in relation to the operative temperature, it was found that when the air-conditioning was on in hybrid buildings, users tended to maintain constant productivity levels. After a specific operative temperature, their performance decreased. These findings were in line with those reported in [9,20] and agree with [7,8,10] that, in general, productivity is lower in warmer conditions. However, this study found the opposite behaviour in the building with a central air-conditioning system: users tended to be more productive at the higher temperatures recorded. These differences in performance trends according to building type (mixed-mode and fully air-conditioned) may be related to the manual interventions on windows and ventilation systems that occupants can make in hybrid buildings. With natural ventilation, it was not possible to establish a productivity pattern based on temperature, probably due to the possibility of the users’ adaptations and adjustments to environmental conditions. This indicates that people maintained their work performance despite the higher internal temperature variations during natural ventilation.

Different behaviours were observed concerning the relationship between productivity and anthropometric variables. Using natural ventilation, older users showed more positive productivity votes at high temperatures, as did the females. These findings are consistent with trends observed in previous studies on anthropometric characteristics [5,13,14] and reiterate that gender and age significantly impact productivity [24]. However, when air-conditioning was on, the differences in behaviour were small, making it difficult to describe a trend. This leads to the conclusion that using air-conditioning can reduce the influence of anthropometric characteristics on the perception of productivity.

Through a large sample of responses from users in all seasons of the year, this study contributes to understanding user productivity in naturally ventilated and air-conditioned office environments. Despite the research findings, some limitations can be identified, such as disregarding the season and spatial variability in productivity analyses, evaluating productivity through users’ perceptions of their performance and condensing environmental variables into the operative temperature. Therefore, future studies can investigate productivity through cognitive tests (e.g., memory, attention, logic, etc.), considering the analysis of the seasons and the spatial variability in the office environments. Furthermore, the relationship between other environmental variables, such as air speed, and productivity, can be investigated.