Energy performance of office buildings in Santiago, Chile: results of unregulated and high solar radiation context

Abstract

In Chile, non-residential buildings are not subject to regulations in regard of energy consumption or indoor environmental quality. With the exception of urban aspects related to volume and sun obstructions, other aspects such as energy performance or façade performance are left to the market's ability to define without any standards or models of adaptation to the country's climate and energy conditions. Santiago, the capital, concentrates most of office buildings, where this development has turned critical due to the widespread use of architectural models based on highly transparent façades, associated with economic success and trans-nationalization, but little consideration of Santiago's temperate-warm climate. This paper presents the results of a currently ongoing measurement campaign considering nine representative buildings selected by types according to the characteristics of its architecture and façade systems. Measurements considered energy consumption, indoor environment quality and user satisfaction. The results obtained in the spring and summer, critical periods of the year in Santiago's climatic context, will be presented. The conclusions of the study show results in two areas regarding the role that architecture plays in the energy performance of buildings: the results obtained when only market mechanisms regulate building performance; and the consequences of this for user comfort

Full text

Energy performance of office buildings in Santiago, Chile: results of
unregulated and high solar radiation context
Claudio Vásquez Z. (1), Felipe Encinas P. (1), Renato D’Alençon C. (2)
(1) Escuela de Arquitectura Pontificia Universidad Católica de Chile, Centro de Desarrollo Urbano
Sustentable (CEDEUS)
(1) El Comendador 1916, Santiago, Chile, clvasque@uc.cl
(2) Technische Universität Berlin, Fakultät VI Planen Bauen Umwelt
Abstract
In Chile, non-residential buildings are not subject to regulations in regard of energy consumption or indoor
environmental quality. With the exception of urban aspects related to volume and sun obstructions, other
aspects such as energy performance or façade performance are left to the market's ability to define without
any standards or models of adaptation to the country’s climate and energy conditions. Santiago, the capital,
concentrates most of office buildings, where this development has turned critical due to the widespread use
of architectural models based on highly transparent façades, associated with economic success and trans-
nationalization, but little consideration of Santiago’s temperate-warm climate.
This paper presents the results of a currently ongoing measurement campaign considering nine
representative buildings selected by types according to the characteristics of its architecture and façade
systems. Measurements considered energy consumption, indoor environment quality and user satisfaction.
The results obtained in the spring and summer, critical periods of the year in Santiago’s climatic context, will
be presented. The conclusions of the study show results in two areas regarding the role that architecture
plays in the energy performance of buildings: the results obtained when only market mechanisms regulate
building performance; and the consequences of this for user comfort
Keywords: Energy consumption, unregulated market, Energy Performance, Office Building, Cluster analysis.
1. Introduction
In Chile, commercial and industrial buildings have no regulations in regard of energy consumption and indoor
environmental quality. Only urban aspects related to the volume that can be built as a ratio to the plot and
basic sun obstruction measures are prescribed in the building code. Other architecture-related aspects, such
as the overall energy performance (demand, consumption, efficiency) or the performance of the envelope
and façade systems (% transparency, U-Value, infiltration) are left unregulated.
In practice, it is left to the market's ability for self-regulation to define standards and models of adaptation to
the country’s climate and energy constraints. In Santiago, the capital, concentrating most of the office
buildings stock and urban development, this economic regulation model has made crisis due to the general
adoption of an architectural model based on high transparency of façades, a widespread symbol associated
with corporate success and trans-nationalization of the economy (fig 1a).
This phenomenon has over a decade of development and only between 2005 and 2013 it has caused the
city to accumulate 205 buildings that together account for more than 750,000 m2 of glazed surface,
distributed according to the chart (Fig. 1b) which shows that the sections of greater transparency concentrate
two-thirds of the buildings in the period [1]. Moreover, it has been established that a 23% of the existing
window surfaces of available uses any sunscreen; however, only for a 5 % it is possible to identify criteria in
their use that are efficient from the energy point of view.

Figure 1: a) Recently built office buildings in Santiago’s bussiness district. Photograph courtesy of Pro-Chile b) Segment
presence Window-to-wall ratio for office buildings. Period 2005 -2013
The climate in the central region of Chile -where Santiago sits- is actually very different from that in the
regions of origin of these architectural models. According to Köppen’s climate classification, Santiago is a
temperate-warm climate with winter rainfall and a long dry season. The figure (fig.2) shows the regions of the
planet with the same climate as the one in Santiago: the Iberian Peninsula, Italy, Middle East, North Africa
and Australia. All of these are areas where the prevailing landscape is rather dry, as opposed to Santiago,
where the availability of fresh water from the Andes has made possible to artificially maintain a green
landscape that does not strictly correspond to its climate. However, the city is exposed to a high sun
irradiation, some 285 W/m2 as an average and 1000 W/m2 as a summer peak.
Figure 2: Regions around the word with the same climate of Santiago, Csb (temperate warm with winter rain), according
to Köppen’s classification [2].
This work focuses on the characterization of energy consumption and indoor environmental quality resulting
from the application of the described building model in the climate of Santiago, in a context where the
regulation does not establish obligations or minimum standards for energy efficiency and thermal or visual
comfort for users.

2. Previous Works
A first measurement campaign was conducted in office buildings in Santiago, between 2011 and 2012 [3]
focusing on establishing a profile for energy consumption and indoor environment quality of transparent
façades buildings representing the stock built between 2005 and 2010. Some of the key findings of that work
are described below:
- Energy consumption of buildings is not correlated with the weather conditions outside, and in
particular it is not reflecting sunlight or temperature changes.
- 43% of interviewed users declared not to achieve thermal comfort in summer; out of this, 26% said
they perceived an excessive cold. Measurements showed that 36% of the working hours is actually
uncomfortable and about 30% of the time lack of comfort is due to overcooling. In winter, the same
occurs in similar proportions, but the cause is overheating.
- Regarding lighting comfort, about 50% of the users who work in areas which have no visual contact
to the outside says not having achieved thermal comfort. Furthermore, over 80% of users declared
using artificial light throughout the day. This is because high glare forces users to have sun protection
systems in all directions, thus leading to maintaining artificial lighting on during the day.
The conclusion was that the analysed office buildings work with a very low performance due to issues
associated with both its architectural characteristics and to problems in the management of their conditioning
systems. This prompted the development of a new study aimed at establishing the performance of the
complete built stock, without discriminating by façade type. This paper is based on the results of the first
phase of this new project.
3. Environmental and Energy Regulations for Office Buildings in Chile
No comprehensive set of regulations or standards for environmental comfort and energy efficiency exist in
Chile. The prescriptions of the Building Code, Ordenanza General de Urbanismo y Construcciones,
O.G.U.C. [4] are limited to U-Value requirements for envelope elements in houses. However, these
prescriptions are not applicable to office or commercial building façades or glazed façades. In addition,
O.G.U.C. contains no definition of standards for thermal comfort or energy consumption, nor do the
Standards of the National Institute for Standards, INN.
Other non-binding definitions do exist in public instruments such as the “Standardized Terms of Reference
with parameters of Energy Efficiency and Environmental Comfort” [5], used by the Department of
Architecture of the Ministry of Public Works; or in specific regulations, such as the D.S. N° 594 “Regulation
about Basic Sanitary and Environmental Conditions in working Places” [6]. Given the dispersion of these
definitions, standards remain to be sought for in international references. A summary of a comparison of
parameters and standards for thermal comfort, lighting comfort and energy consumption between Chile and
European Union is elaborated below:
3.1 Thermal Comfort
The Chilean Ministry of Public Works establishes parameters for the contracting of design and construction
of new public buildings [5], including adaptive thermal comfort in a range from 20°C (winter minimum) to
26°C (summer maximum) for office buildings. In comparison, the European standard EN 15251: 2007
establishes that design values for the indoor temperature for heating load and cooling load calculations shall
be specified at national level. However, it recommends (Annex A) default values recommended for countries
that might not have defined them, for a normal level of comfort expectation (Case II) of 20°C (winter
minimum) to 26°C (summer maximum).
3.2 Energy Consumption
Chile has no energy!related building codes for commercial or public sector buildings. Parameter
recommendations do exist by the Ministry of Public Works [7]: Heating and Cooling Energy Demand in
kWh/m² year; but no performance standards or references are given. In the EU, the Energy Performance of
Buildings Directive [8] introduced the nearly Zero Energy Buildings (nZEB) goal, with no harmonized nZEB
definition throughout the EU. Nearly all countries have now adopted a national methodology, which sets

performance-based requirements for new buildings using several parameters.
The table below (Table 1) summarizes a comparison of the parameters and standards discussed in this
section for thermal comfort, and energy consumption relevant for office buildings. They are compiled from
the relevant regulations, standards, parameters and recommendations existing in Chile and the EN
Standards.
Parameter
Chilean Parameters, Standards and other
Definitions
European Union Parameters and Standards
Range
Unit
Standard
Values
Definition by
Range
Unit
Standard
Values
Definition
by
Heating Energy
Demand, DEC
n/a
kW-h/m² yr
n/d
M.O.P.I.C.,
2008: p 08
Austria
kW-h/m² yr
22,75
Local
regulation.
Source:
BPIE
2011
Switzerland
kW-h/m² yr
46,00
Cooling Energy
Demand, DER
n/a
kW-h/m² yr
n/d
n/a
kW-h/m² yr
n/a
Final Energy
Demand
n/d
kW-h/m² yr
n/d
n/a
Czech Republic
kW-h/m² yr
179,00
Local
regulation.
Source:
BPIE
2011
n/d
kW-h/m² yr
n/d
Portugal
kW-h/m² yr
122,00
Primary Energy
Demand
n/d
kW-h/m² yr
n/d
n/a
Portugal
kW-h/m² yr
407,00
Local
regulation.
Source:
BPIE
2011
n/d
kW-h/m² yr
n/d
Germany Existing
Buildings
kW-h/m² yr
226,00
n/d
kW-h/m² yr
n/d
Germany new
Buildings
kW-h/m² yr
172,00
Operative Design
Temperature for
Conditioning
Equipment
T min
Winter at
55% RH
°C
20
M.O.P., 2012;
G.A.T N° 7,
p. 5
Minimum for
heating (winter
season), ~ 1,0 clo
°C
20
EN
15251:
2007
Annex 1
T max
Summer at
55% RH
°C
25
Maximum for
cooling (summer
season), ~ 0,5 clo
°C
26
Table 1: Comparison of Chilean energy and environmental comfort parameters and those in the EU. Based on [5] [6] [7]
[9] [10] [11]
In the subsequent discussion, these parameters can be compared to the data obtained from the
measurement campaign conducted in regard of comfort prescriptions and not in regard of energy, since the
established levels correspond to energy demand and the measured levels to energy consumption. The
fragmentation of the Chilean parameters and standards compared to foreign regulations, however, puts to
the fore not only the need for further regulation of standards but also for a reference to values that can be
attained and thus prescribed. The question to be addressed in this work is then: given a set of parameters in
regard of thermal comfort and energy demand, what are the current levels obtained in office buildings regard
of thermal comfort and energy consumption?
4. Method
The main study comprised office buildings built in Santiago between 2005 and 2011, a total of 105 cases.
Due to availability or quality of the available information, it was necessary to exclude 14 cases, with a final
consolidated database of 91 case studies. The sample reaches a 99% confidence level and error margin of
3.83%. The Database (DB) was constructed using information obtained on site (interviews and surveys) and
cadastral information obtained from the archives of the Municipal Departments of Works. Because of the
heterogeneity of the information obtained for energy consumption -only 38.4% of cases studied had available
energy billing information- statistical techniques were applied to complete the missing data. Using a multiple
linear regression calculated using some of the continuous variables in the DB, an equation was obtained
based on the goodness-of-fit indicators used by each model, which allowed us to obtain an estimate for the
energy consumption of the 91 buildings in the DB. As a result of this study, five clusters or families of
buildings were defined, on which this paper focuses. One of them turned out to be a unique, completely
different from the rest because of its size.
The clusters were defined based on the various variables in the DB, for which the statistical classification

method called Cluster Analysis was applied. Cluster Analysis is used to generate instance types or groups
with a similarity that can be quantified and measured. To perform cluster analysis, it is necessary to correct
the interdependencies that may exist between the variables and the non-equivalence of the metrics used. In
our case, it is necessary because some variables are expressed, for example, in units of energy (kW) or area
(m2), and others as dichotomous responses (yes-no). The new variables obtained by this process are called
components, which summarize the information contained in the original variables.
Figure 3: Dendogram, a or graphical representation of the hierarchical cluster analysis, showing the successive stages of
the classification
Based on the components resulting from a Principal Component Analysis (PCA), a hierarchical cluster
analysis was conducted, which consists of performing the classification according to a process that responds
to a tree structure, where each stage produces a new branch. Although there is no rule to define the
appropriate number of types it is possible to establish a criterion through the interpretation of a dendrogram,
or graphical representation of the hierarchical cluster analysis, showing the successive stages of the
classification. Figure 3 shows the dendrogram obtained here, which is cut into five clusters concurrency
considering the relative size of the clusters is balanced with the sectioning used, represented in red color.
Table 5 presents the continuous variables with descriptive indicators showing data dispersion, and also the
occurrence of continuous and categorical variables where each is expressed.
Continuous Variables
Robust Mean (Estimator – M. Huber)
Cluster 1
Cluster 2
Cluster 3*
Cluster 4
Cluster 5
N=22
N=31
N=1
N=25
N=11
Building Total Area
36191,9
10253,1
128663,1
9576,8
16313,4
Office Area
20342,0
5897,8
89625,0
5674,3
7639,0
Number of Stories
22
11
52
10
12
Office Façade Area
11075,2
4076,5
32422,9
3705,1
5003,0
Opening Area
7804,1
2051,1
29975,0
1516,5
3058,1
Typical Floor Net Area
1019,7
611,9
1366,9
633,2
715,8
Note (*): Since this is a single case, the calculation of the Robust Mean was dismissed. Individual case
data (Titanium Building) was then considered
Categorical Variables
Median and Mode
Cluster 1
Cluster 2
Cluster 3
Cluster 4
Cluster 5
N=22
N=31
N=1
N=25
N=11
Light Façade
Yes
Yes
Yes
No
No
Opaque Sun Protection
No
No
No
No
Yes
Table 2: Continuous (above) and categorical (below) variables used to define the case studies clusters for analysis
This paper presents the results of the first year on-site measurements of two cases per cluster, which were
selected based on the centroids identified from the qualities of the façades registered in a data base. All
cases considered in this research were statistically selected to represent consistently one of the five clusters

in which was divided the building database compiled during the first year of the project. The measurements
include energy consumption, indoor environment quality, and user satisfaction. Based on the information
thus collected the following indicators were elaborated based on [12] and [13].
4.1 Thermal Comfort Indicators
In regard of thermal comfort: Predicted Mean Vote (PMV); Predicted Percentage of Dissatisfied (PPD);
Comfort Hours Index (CHI); User Satisfaction Index (USI).
4.1.1 PMV: Predicted Mean Vote (+3 to -3)
Indicator reflecting the influence of the physical and physiological variables on thermal comfort. Its aim is to
predict the evaluation of a specific group of people about their thermal environment, taking into account
environmental parameters, metabolic activity and type of commonly used clothing. PMV allows to evaluate
the work spaces in reference to standards and to the rest of the cases analyzed in this study. This index
uses a scale ranging from -3 (predicted cold perception) to +3 (predicted heat perception). The 0 value
implies a confortable situation is predicted.
4.1.2 PPD: Predicted Percentage of Dissatisfied (%)
Indicator derived from the PMV, PPD which aims at quantifying the percentage of "dissatisfied" persons in
relation to specific environmental conditions.
4.1.3 CHI: Comfort Hours Index (%)
It is the analysis of data from temperature and humidity monitored permanently for the whole year. With
these records indicators may be obtained such as hours of accumulated comfort, for hours of comfort by
work areas or working hours. In addition, this information may be contrasted with the values and acceptability
times from user surveys that occur intermittently over time. Comfort hours are grouped in three time sections:
AM-M-PM.
4.1.4 USI: User Satisfaction Index (+3 to -3)
This index measures the user satisfaction with the thermal environment in the last month. It ranges from +3
(very satisfied) to -3 (very unsatisfied) and 0 is indifferent (Fanger scale). Average values of apparent
temperature, acceptability, comfort, and work performance are obtained from the survey directly. The
average percentage of acceptability, weighted according to the number of people in the workspace, is used
to rate the performance of the building. The results have two applications: the particular evaluation of each
office considering the areas and types of workspaces that are evaluated; and the comparison of offices so
evaluated with standards and regulations, as well as with the offices samples analyzed in this study.
4.1.5 UPI: User Thermal Perception Index (+3 to -3)
This index measures the user perception with the thermal environment in the last month. It ranges from +3
(very satisfied) to -3 (very unsatisfied) and 0 is indifferent or comfortable.
4.2 Energy Consumption Indicators
In regard of energy consumption: Total Energy Use Index (EUI); Energy Costs Index (ICE); Electrical Load
Factor (ELF).
4.2.1 EUI: Total Energy Use Index (kWh/m2)
The purpose of this indicator is to establish a reference baseline of energy consumption based on electricity
bills in the studied offices and a pro rata share of the building costs. This indicator was analyzed on a
comparative basis to establish causal relationships of consumption relative to the total analyzed cases in an
annual and seasonal scale.
4.2.2 ECI: Energy Costs Index (CH$/m2)
Its purpose is to establish a reference baseline cost of energy based on electricity bills in the studied offices
and a pro rata share of the building costs. This indicator was analyzed on a comparative basis for
establishing causal relationships of costs relative to the total analyzed cases in an annual and seasonal
scale.

Figure 4: Box plot analysis for Comfort Hours Index (CHI) by season and by Cluster. Measurements for summer and spring,
2015. FONDECYT 1130815
The performance of the whole analysis group of buildings by season (Fig. 4) shows that thermal comfort is
better during spring than during summer since there are no 0% PPV / CHI events, which do occur during the
summer season. However, the central quartiles (25% to 75%) are better balanced and better located relative to
the proportion of comfort hours reached. On the other hand, the means are similar for both seasons, around
60%.
The comfort hours analysis (CHI) per cluster shows (Fig. 4) a notorious diversity in the results. Means do not
present a comparable pattern, nor do dispersions. Buildings in Cluster 2, which are small buildings with light
façades and prevailing in number in the sample, are the ones showing the most notorious dispersion. On the
other hand, buildings Cluster 1, which are the buildings with the biggest façade areas (excluding Cluster 3, with
only one case of study) have the least dispersion. The worst performance is that of Cluster 5, with a maximum
of 60% of comfort hours.
5. Results and discussion
The results obtained are presented in the following in two aspects: thermal comfort and energy consumption.
5.1 Thermal Comfort in spring and summer
Figure 5: Box plot analysis for PMV by season and by Cluster. Measurements for summer and spring, 2015.
C1#
C2#
C3#
C4#
C5#
0%# 20%# 40%# 60%# 80%# 100%#
##
Box#plots#
CHI#by#Cluster##
Mean#
Spring'
Summer'
0%' 20%' 40%' 60%' 80%' 100%'
!!
Box!plots!
CHI!by!Season!!
Mean'
C1#
C2#
C3#
C4#
C5#
0# 5# 10# 15# 20# 25# 30# 35# 40#
%"
Box"plots"
PPD"by"Cluster""
Mean#
Spring'
Summer'
0' 5' 10' 15' 20' 25' 30' 35' 40'
%"
Box"plots"
PPD"by"Season""
Mean'

The PMV analysis by season (Fig. 5) shows that the indoor environment conditions have a trend to predict
the perception of cold and that the data dispersion in summer season predict heat events that do not take
place during the spring. This indicates the operation of the AC systems during the summer is erratic. The
reference values for thermal comfort (PPD) are between -0,5 y + 0,5 (UNE ENE ISO 7730:2006), and thus
the season results are to an important degree unsatisfactory for both summer and spring seasons.
The PMV analysis by Cluster (Fig. 5) shows a trend to predict col events in all of them, without any trend to
predict heat events. Considering the reference values, only the buildings in Cluster 1 and Cluster 4 have
acceptable predictions, while all other Clusters have a dispersion above -0,5, considered to be the limit of the
acceptable.
The PPD analysis by season (Fig. 6) shows that the percentage of unsatisfied users is very different from
summer to spring. The acceptable reference values is 10% maximum (UNE ENE ISO 7730:2006), thus the
global prediction indicates that the whole buildings sample will have a uncomfortable perception by the
users.The PPD analysis by Cluster (Fig. 6) indicates no Cluster reaches acceptable results. However,
Clusters 1 and 4 tend to have a better user perception.
So far, the comfort parameters analyzed are objective, i.e. they are based on measurements and calculation
models based on measured data. A different nuance appears when analyzing the user perception,
considered subjective, as collected by means of polls.
Figure 7: Box plot analysis for the User Satisfaction Index (USI) by season and by Cluster. Polls for summer and spring,
2015.
Figure 6: Box plot analysis for PMV by season and by Cluster. Measurements for summer and spring, 2015.
C1#
C2#
C3#
C4#
C5#
(2# (1,5# (1# (0,5# 0# 0,5# 1# 1,5# 2#
!!
Box!plots!
USI!by!Cluster!!
Mean#
Spring'
Summer'
+2' +1,5' +1' +0,5' 0' 0,5' 1' 1,5' 2'
!!
Box!plots!
USI!by!Season!!
Mean'
C1#
C2#
C3#
C4#
C5#
0# 5# 10# 15# 20# 25# 30# 35# 40#
%"
Box"plots"
PPD"by"Cluster""
Mean#
Spring'
Summer'
0' 5' 10' 15' 20' 25' 30' 35' 40'
%"
Box"plots"
PPD"by"Season""
Mean'

The analysis of user satisfaction (USI) by season (Fig. 7) suggests that users are thermally satisfied with
their indoor work environment, as in the results of the polls the evaluation is positive. Only in the summer the
first quartile reaches negative values. The mean values are very similar for both seasons.
The USI analysis by Cluster (Fig. 7) disaggregates the previous and shows Clusters 1, 2 and 4 to have a
clear positive perception of the indoor work environment. For Cluster 3, where PPD presented a higher
dispersion, users declare the highest indifference to the indoor work environment, since the average
response tend to 0. Cluster 5 also presents a neutral perception, yet the CHI results suggested this was a
cluster with a poor performance.
Finally, when evaluating the user perception (UPI) by season (Fig. 8), we observe it tends to be comfortable,
since the responses are around 0. In summer the extreme quartiles present a higher dispersion, however in
both seasons the central quartiles remain consistently around 0. The results of the UPI analysis by Cluster
(Fig. 8) also indicate a good performance since users in all of them responded with values between +1 and -
1, considered acceptable.
Figure 8: Box plot analysis of the User Perception Index (UPI) by season and by Cluster. Polls for summer and spring,
2015.
In sum, based on the analyzed sample the results for thermal comfort present the following characteristics:
- In general, objective and subjective indexes do not show big seasonal differences, although in spring
the performance is better than that in summer
- The Comfort Hours Index (CHI) analysis, relating only temperature and Relative Humidity appears to
be inconsistent with the indexes including variables associated to the users in the calculation
models, such as clothing or activity.
- There is a divergence between the objective measurement values and the subjective perception by
the users, which can only be explained by a lack of critical appraisal and low expectations on behalf
of the users to have a comfortable work environment.
5.2 Energy Consumption
The analysis of the total energy consumption by Cluster, as expressed by the Energy Use Index (EUI) (Fig.
9) shows important differences in the energy consumption between spring and summer, with a very clear
trend towards a higher summer consumption. This leads to think it is a consequence of Air Conditioning
energy consumption, which in summer rises as a result of solar radiation. It can also be observed that during
the summer the dispersion of the second quartile is much more than those of the others; and that the third
and fourth quartile are more robust than the first and the second, the quartiles with the highest consumption,
and in a much higher proportion.
C1#
C2#
C3#
C4#
C5#
(2# (1,5# (1# (0,5# 0# 0,5# 1# 1,5# 2#
!!
Box!plots!
UPI!by!Cluster!!
Mean#
Spring'
Summer'
+2' +1,5' +1' +0,5' 0' 0,5' 1' 1,5' 2'
!!
Box!plots!
UPI!by!Season!!
Mean'

To go deeper into these results, we have elaborated a daily analysis of energy consumption for Air
Conditioning (Fig. 10), elaborated based on the measurements conducted directly on the electricity meters,
in working hours for spring and summer. The result shows that most of the clusters have a comparable level
of consumption, except for cluster 3, with only one case of study, different to the rest because of its size.
Figure 10: Daily HVAC Energy consumption by Cluster. Elaborated based on the measurements of consumption during
working hours for the work days in spring and summer. Measurements for 2015
An analysis base on the Energy Cost Index (ECI) (Fig. 11) by season was also conducted, reflecting also the
difference in consumption observed before; however, observing the same index by cluster (Figure 11),
notorious differences appear in what the buildings in the different clusters must pay for the energy they
consume. It is particularly notorious that Cluster 3, which is the one consuming the most energy for air
conditioning, is also the one that pays less for the energy consumed.
Figure 11: Energy Cost Index (ECI) by season and cluster. Elaborated based on the electricity bills for 2013 and 2014
Figure 9: Box plot analysis of Total Energy Use Index (EUI) by season and by Cluster. Elaborated based on the
electricity bills for years 2013 and 2014.
C1#
C2#
C3#
C4#
C5#
$#0# $#5.000# $#10.000# $#15.000# $#20.000# $#25.000#
$/m2#
Box$plots$
ECI$by$Cluster$$
Mean#
Spring'
Summer'
$'0' $'5.000' $'10.000' $'15.000' $'20.000' $'25.000'
$/m2'
Box$plots$
ECI$by$Season$$
Mean'
Spring'
Summer'
0' 0,05' 0,1' 0,15' 0,2' 0,25' 0,3'
kWh/m2'day'
Box$plots$
Daily$HVAC$Energy$Comsump8on$by$Season$$$
Mean'
C1#
C2#
C3#
C4#
C5#
0# 0,05# 0,1# 0,15# 0,2# 0,25# 0,3#
kWh/m2#day#
Box$plots$
Daily$HVAC$Energy$Comsump8on$by$Cluster$$
Mean#
C1#
C2#
C3#
C4#
C5#
0# 50# 100# 150# 200# 250# 300# 350#
kWh/m2#year#
Box$plots$
EUI$by$Cluster$$
Mean#
Spring'
Summer'
0' 50' 100' 150' 200' 250' 300' 350'
kWh/m2'year'
Box$plots$
EUI$by$Season$$
Mean'

In sum, the energy consumption in the studied cases can be characterized as follows:
- The seasons are relevant in the energy consumption. Climate conditions are thus a relevant factor in
the prediction of energy consumption.
- An anomaly is observed in the pricing, where the highest consumer has to pay the lowest prices for
the consumed energy. This is, the current pricing scheme plays a role as an incentive to energy
consumption.
6. Conclusions
To summarize the conclusions of this work, figure 12 compares the Energy Cost Index (ECI) with the
Comfort Hours Index (CHI), in order to correlate the energy consumption in the buildings and the efficiency
they reach at a the given air conditioning cost. It can be clearly read from the graph that the prices payed in
the different clusters have no relation at all with the efficiency of their performance. This suggests that it is
the energy market that regulates the consumption and not the efficiency in the use of energy.
Figure 12: ECI / CHI by Cluster. FONDECYT 1130815. Figure 13: PMV / UPI. FONDECYT 1130815
Figure 13 shows a comparison between the objective comfort index (PMV) and the subjective (UPI), where
no correlation can be observed between the two indexes. That is, the users perceive as thermally
comfortable certain situations that objectively are not. This is a divergence requiring further analysis;
however, a preliminary combination of both conclusions is that as a consequence of a consumption pattern
regulated by the energy price rather than by the efficiency of the buildings the indoor comfort conditions are
deficient and users tend to get used to these poor conditions.
7. Acknowledgments
This article has been produced with funding from the National Council of Scientific and Technological
Research of Chile (CONICYT) through FONDECYT Project 1130815: “Definición de la línea base de
consumo energético y funcionamiento del ambiente interior en edificios de oficina de Santiago”. It has also
had the support of the Center for Sustainable Urban Development (CEDEUS), CONICYT/FONDAP Project
15 110020.
8. References
[1] FONDECYT 1141240 Modelo para el diseno arquitectónico de sistemas de fachadas vidriadas
complejas energética y lumínicamente optimizadas. Conicyt 2014-2017
[2] Prieto Hoces, A., 2011. Interfaz ambiental en edificios de oficina: envolvente de espesor programático
variable como sistema de mediación ambiental pasivo.
0%#
20%#
40%#
60%#
80%#
100%#
$#0# $#5.000# $#10.000# $#15.000# $#20.000# $#25.000#
CHI$
ECI$
ECI/CHI$by$Cluster$
C1# C2# C3# C4# C5#
y"="0,521x"*"0,5494"
R²"="0,39595"
*2"
*1,5"
*1"
*0,5"
0"
0,5"
1"
1,5"
2"
*2" *1,5" *1" *0,5" 0" 0,5" 1" 1,5" 2"
PMV$
UPI$
Objec,ve$(PMV)$by$Subjec,ve$(UPI)$Comfort$
Percepcion$Analisys$

[3] FONDECYT 1100143 Base de Información para el diseño de fachadas transparentes eficientes en
Santiago, Conicyt, 2010-2012
[4] D.S. N°47, 1992. Ministerio de Vivienda y Urbanismo. Ordenanza General de Urbanismo y
Construcciones, O.G.U.C. (actualizada al 08 de Abril del 2014)
[5] MOP, 2012. Términos de referencia estandarizados de eficiencia energética y confort ambiental, para
licitaciones de diseño y obra de la Dirección de Arquitectura, según zonas geográficas del país y
según tipología de edificios. Con 10 Guías de Apoyo Técnico.
[6] D.S. N° 594, 1999. Ministerio de Salud. Aprueba Reglamento sobre condiciones sanitarias y
ambientales básicas en los lugares de trabajo. Actualizado DS N° 57, 2003.
[7] MOP – IC, 2008. Manual de Diseño Pasivo de Edificios Públicos. Santiago: Instituto de la
Construcción,
[8] European Union Directive 2010/31/EU of the European Parliament and of the Council of 19 May 2010
on the energy performance of buildings
[9] BPIE, 2011. Europe's Buildings under the Microscope: A country by country review of the energy
performance of buildings
[10] EN 12464: 2007 The Lighting of Workplaces
[11] EN 15251: 2007 Indoor environmental input parameters for design and assessment of energy
performance of buildings addressing indoor air quality, thermal environment, lighting and acoustics.
[12] ASHRAE, 2010. Performance Measurement Protocols for Commercial Buildings.
[13] ASHRAE, 2012. Performance Measurement Protocols for Commercial Buildings: Best Practice Guide.
---------------------------------------------------------------------------------------------------------------------------------
Copyright
By submitting the paper and PowerPoint presentation, the author understands that he/she is granting the
Economic Forum (‘Copyright Holder’) the copyright for said publication. The Copyright Holder shall have the
right to use this work, in whole or in part, in printed or digital form. The Copyright Holder shall have the right
to publish this work in conference proceedings or articles for advertising purposes. The author warrants that
the paper and PowerPoint presentation do not infringe the intellectual property rights of any third party.