ARTICLE IN PRESS

Building and Environment 42 (2007) 3261–3276

www.elsevier.com/locate/buildenv

Improving energy performance of school buildings while ensuring

indoor air quality ventilation
Rachel Beckera,, Itamar Goldbergera, Monica Paciukb
a
Department of Structural Engineering and Construction Management, Faculty of Civil and Environmental Engineering,
Technion—Israel Institute of Technology, Haifa 32000, Israel
b
National Building Research Institute, Technion, Haifa, 32000, Israel
Received 6 June 2006; received in revised form 20 July 2006; accepted 28 August 2006

Abstract

Energy conscious design of school buildings, as well as deemed-to-satisfy provisions in a Performance Based Energy Code, should
address the problem known as the energy efficiency—thermal comfort—indoor air quality dilemma (EE-TC-IAQ Dilemma). In warm
and moderate climates, the large internal heat sources usually found in school buildings prevent achieving thermal comfort without
active cooling in summer, but are not sufficient to eliminate the need for heating in winter. Commonly used air-conditioners do not
improve air quality, while natural ventilation induces uncontrolled energy losses. In this study, a step by step process was used for the
development of deemed-to-satisfy design solutions, which cope with the EE-TC-IAQ Dilemma, for a performance based code.
A distinction is made between improving building design variables and improving ventilation schemes. Results indicate that
implementation of improved ventilation schemes in an otherwise well designed energy-conscious building result in savings of 28–30%
and 17–18% for northern and southern classroom orientations, respectively.
r 2006 Elsevier Ltd. All rights reserved.

Keywords: School buildings; Indoor air quality; Energy performance; Thermal comfort; Ventilation; Thermal insulation; Shading

1. Introduction The identification of preferred solutions for school

The methodology and results presented in this paper
were developed within the framework of an extended
Israeli research program, which aims at the establishment
of a modern Building Energy Code that addresses all
building occupancies.
The Code will enable two design options: a performance-
based option that requires a comparative assessment of
computationally-estimated energy expenditure against a
calculated energy budget, and a prescriptive option. For
the latter the Code will provide a set of solutions that are
considered deemed-to-satisfy, and are based on a systema-
tic investigation and identification of preferred sets of
solutions. The reference energy budget will be based on
such a preferred solution set.
Corresponding author. Fax: +972 4 8295697.
E-mail address: becker@tx.technion.ac.il (R. Becker).
buildings must address the inherent conflict that stems
from the wish to save energy while having to provide
adequate indoor air quality in addition to thermal comfort.
School buildings include nowadays various functional
spaces. However, classrooms are still the most common
functional space. They occupy the largest area of every
school building, and host the largest part of daily activities
and occupants. One of the most dominant features of a
classroom is its high occupancy density, which results in
very large values of the internal heat sources (approxi-
mately 5 kW), as well as of the internal emissions of body
odors, water vapor and CO2, causing an increasing concern
with regard to the indoor air quality youngsters are
exposed to along the major period of their growing up
years [1,2].
Extensive natural or controlled ventilation, intended to
remove internally generated contaminants, without active
heating or cooling, is only seldom sufficient for the
provision of required thermal comfort conditions.

0360-1323/$ - see front matter r 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.buildenv.2006.08.016
 ARTICLE IN PRESS
3262 R. Becker et al. / Building and Environment 42 (2007) 3261–3276
In summer, whenever external temperatures are lower
than the required indoor temperature, ventilation can
remove the excessive heat load produced by both incident
solar radiation and internal sources, thus allowing the
achievement of comfortable indoor temperatures. How-
ever, whenever external temperatures exceed the required
comfort level temperature, active mechanical cooling must
be provided.
Heating classrooms in winter is significantly assisted by
the large internal heat sources, which can replace a large
part of the heating energy demand. However, when
external temperatures are smaller than the required indoor
temperature, ventilation removes most of the internal heat
load as well as the heat gains from solar radiation. Natural
non-controlled ventilation leads then to excessive energy
losses, as well as to chilling draughts and loss of thermal
comfort. Controlled ventilation in winter is thus essential.
Consequently, simultaneous catering for thermal com-
fort, indoor air quality and energy conservation in schools
is a design dilemma with apparently no obvious solution.
In the sequel, we denote this dilemma by: the EE-TC-
IAQ Dilemma (energy efficiency—thermal comfort—in-
door air quality dilemma). Efficient energy design of the
classroom wing of school buildings should thus be
primarily concerned with providing optimal solutions able
to cope with the above dilemma.
Within the framework of a performance-based building
code, the aim is to propose a range of engineering valid
‘‘preferred’’ solutions that establish a basis for creative and
routine design, but not to prescribe an optimal unique
solution, which may be too restrictive and become an
unreasonable barrier to creative design.
This paper presents the methodology adopted for
deriving such ‘‘preferred’’ solutions that cope with the
EE-TC-IAQ Dilemma in school buildings, as well as some
results that may be of general interest though they were
derived for a typical local Mediterranean climate (see
Appendix for some typical data). Locally significant
observations or conclusions that are not relevant elsewhere
have been intentionally omitted.
2. Energy performance, thermal comfort and IAQ in schools
Most of the literature concerned with energy perfor-
mance of school buildings is devoted to savings via specific
features such as utilization of solar energy [3–6], construc-
tion features, such as thermal insulation, thermal mass, and
shading [5,7–9], HVAC performance [10–13], and geother-
mal pumps [14–17]. However, basic assumptions regarding
thermal comfort, indoor air quality, occupancy and
acclimatization schedules, internal loads, and architectural
features of the school building are not identical in the
various publications, and are usually based on local
preferences. Consequently, even when similar climatic
conditions prevail, conclusions cannot be regarded as
sufficiently general. Moreover, the topic of energy perfor-
mance of schools located in the Mediterranean region’s
climatic conditions and culture has not been explored
at all.
The basic assumptions relevant to this paper’s metho-
dology, analysis and results are presented in Sections
2.2–2.4 below, while Section 2.1 briefly summarizes the
literature concerned with the EE-TC-IAQ Dilemma.
2.1. Coping with the EE-TC-IAQ Dilemma in schools—
literature overview
Despite the obvious need to cope with the EE-TC-IAQ
Dilemma in school buildings, and the wealth of literature
emphasizing the need to improve IAQ on one hand
[1,2,18,19], and on the other one addressing the impact
on acclimatization energy imposed by direct IAQ ventila-
tion [4,18,20,21], only a few publications were found that
present integrated solutions accompanied by thermal and
energy analysis. Of these, only a few were concerned with
the main classroom wings, addressing the inherent large
internal loads as well as the more stringent IAQ require-
ments that exist there.
Studies since 1997 include Davanagere and colleagues
[22], who studied the effect of the new ASHRAE-62 IAQ
requirements for school buildings on life cycle costs when
using different HVAC systems, but did not address the
architectural or construction features of the building;
Dorer and Weber [23], who pointed to the significance of
an integrated evaluation of the energy and IAQ response of
a multi-story school building, but addressed only two
specific features: natural night ventilation enabled by two
modes of window opening, and shaft ventilation via a
double glazed fac- ade; Kavanaugh and Xie [16], who
demonstrated the significance of addressing the fan energy
required for ventilation, either IAQ ventilation or heat
recovery ventilation, in the total energy analysis; Eriksson
and Whalstrom [24], who analyzed the performance of a
hybrid ventilation system based on a solar chimney
implemented in a Swedish school, utilizing multi-zone air
transfer to model effects of wind conditions as well as door
opening strategies; Becker and Paciuk [25], who showed the
improved effect on total energy loads of various IAQ
ventilation and night ventilation schemes, enabled by
utilizing the buffering effect of an existing central atrium,
but did not address their impact on total electricity
consumption.
Additional articles are devoted to some specific solu-
tions or features of the EE-TC-IAQ Dilemma in other
functional spaces of school buildings. Recent publications
address atria [26], staff rooms and auditoria [19], and sport
halls [27].
The literature survey revealed that research in the area
has been mostly concerned with improving the implemen-
tation and control of natural ventilation, or replacing
some specific features of the HVAC system. No publica-
tions were found that address enhanced energy conserva-
tion by utilizing some of the regular, though specific,
architectural features of classroom wings (such as the
 ARTICLE IN PRESS
R. Becker et al. / Building and Environment 42 (2007) 3261–3276
layout, which always includes corridors, lobbies or atria
that are vacant during classroom hours), and the special
opportunity they provide for rational design and manage-
ment of mechanically controlled day and night ventilation
schemes that are concerned with solving the EE-TC-IAQ
Dilemma.
2.2. Typical architectural features
Most Israeli schools consist of two to three story-high
buildings. Fig. 1 depicts some schematic configurations of
the classroom wings in these buildings, as derived by
analyzing various plans of existing schools. In the absence
of restrictions on orientation, classroom windows may face
any direction.
It is noticed that classrooms are usually clustered to the
sides of a corridor or around a hall, with similar clusters
stacked repeatedly throughout the building height. In a one
sided corridor configuration, the opposite corridor wall
includes very large openings, which provide sufficient
daylight. When operable windows are installed in the
openings, they will usually consist of sliding wings, which
are closed in winter and open in summer. A two sided
corridor or an internal hall does not have sufficient access
to natural delighting, and these spaces are served by
electrical lighting. Despite the lack of natural ventilation in
such corridors and lobbies, it is not common to equip them
with mechanical ventilation.
Outside, shaded horizontally Atrium Corridor/Hall
Interior non conditioned space Interior conditioned space
Fig. 1. Schematic presentation of some classroom wing configurations in
Israeli school buildings.
7.2m
6.8m
Fig. 2. Schematic presentation of classroom layout.
3263
The classroom’s schematic layout is given in Fig. 2.
Dimensions are typical of Israeli classrooms, which are
designed for an occupancy of up to 40 pupils, based on
Design Guidelines provided by the Ministry of Education.
In most cases the entrance door is located on the wall
opposite to the windows, with only one wall including
windows. The blackboard is located on the transverse wall,
so that when pupils are facing the board windows are on
their left, allowing left-sided day lighting on their desks.
2.3. Thermal comfort
The American ASHRAE Comfort Standard 55 [28], and
the International ISO Standard 7730 [29] are considered as
standard tools for the evaluation of thermal comfort levels by
means of the thermal predicted mean vote (PMV) and the
predicted percentage of thermally satisfied (PPS) occupants
in the vicinity of neutral thermal sensation, for different
combinations of hygro-thermal conditions and personal
factors. These Standards are essentially identical, as both
have adopted Fanger’s formulation and accompanying
values of the equations’ parameters [30]. However, research
conducted in various countries has indicated that significant
discrepancies exist between the predicted values and actual
responses in occupied buildings. Statistically valid studies of
thermal comfort in school buildings are scarce, and published
data includes only those conducted in Japan [31,32], Brazil
[33], Hawaii [34], and Singapore [19]. The discrepancies
between the actual responses and the predicted values were
different in the various studies, with no uniform direction on
the comfort scale. This indicates that local aspects affect the
deviation from Fanger’s formula, and that assessment of
thermal comfort conditions cannot fully rely on the results
derived by means of the ASHRAE and ISO Standards.
The literature search indicated that data on thermal
response and thermal comfort in local Israeli, or Medi-
terranean schools is not available. No specific field study
was conducted within the framework of the current
research program. However, studies performed previously
by the authors in Israeli offices and residential buildings
[35–37] indicated that preferred conditions are usually
somewhat cooler, in summer as well as in winter, than
those predicted by the standard tools.
Assuming that the preferences in schools are similar to
those found in other buildings, the set points for the
present thermal analysis were established as follows:
Assuming a 50% RH and a classroom-adequate low air
velocity of 0.1 m/s, PMV was established for all possible
combinations of the variables given in Table 1. The range
of metabolic rates, M, addresses the need to satisfy the
comfort requirements of seated pupils as well as of the
standing teacher. The ranges of clothing’s thermal resis-
tance, Icl, stem from the difference in pupils’ clothing or
uniform and that of teachers.
A plot of air temperature, Ta, versus PMV is shown in
Fig. 3. Combinations leading to PMVo0.5 and

3264 R. Becker et al. / Building and Environment 42 (2007) 3261–3276
PPSo90% are denoted by ‘‘too cold’’. Combinations
leading to PMV4+0.5 and PPSo90% are denoted by
‘‘too hot’’. Combinations with PMV within the range 0.5
to +0.5 and PPSX90% are denoted by ‘‘acceptable’’.
Circular ticks denote combinations with regular clothing
and x’s denote those for which clothing restrictions are
applied (e.g., long sleeves only).
It was assumed that in schools, similarly to preferences
recorded in residential and office buildings, the cooler side
temperatures of the ‘‘acceptable’’ range would be preferred
and considered as adequate for providing thermal comfort.
Consequently, Ta ¼ 20.5 1C, was established as the set-
point for winter heating, and Ta ¼ 24.5 1C for summer
cooling. Confirming the statistical validity of these values
was beyond the scope of the current research program.
2.4. Indoor air quality
Requirements for indoor air quality are determined in
Northern America and some other countries by means of
the American Standard ASHRAE 62 [38], whereas in most
European, as well as in some other countries, the prevailing
background document is the European pre-standard prEN
13779 [39]. These two documents are not identical, neither
in concept nor in the established norms. The American
standard stipulates a minimum value of 8 l/s/person for the
required flux of ‘‘fresh’’ air in classrooms based on an
indoor maximum excess CO2 concentration of 700 ppm.
The European document defines three performance levels,
IDA1, IDA2, and IDA3, based on CO2 excess concentra-
tions of 800, 1000, and 1500 ppm, respectively. Obviously,
even the most stringent European level IDA1 is more
Table 1
Combinations of variables used for thermal comfort analysis
Item (1) Summer (2)
2
M (W/m ) 58–70
Icl (m2K/W) 0.07–0.11
Ta (1C) 23–26
MRT (1C) Ta+1
M, metabolic rate (W/m2); Icl, clothing insulation index (m21K/W); Ta, air
temperature (1C); MRT, mean radiant temperature (1C).
winter
22.5
Acceptable
22.0
Temperature (C)
21.5
21.0
Too Cold
20.5
20.0
19.5
-1.0 -0.5 0.0
PMV
Fig. 3. Ta vs. PMV in winter and summer, for the various combinations of M, Icl, and MRT.
ARTICLE IN PRESS
lenient than the unique American level. For classrooms,
where there is no smoking, prEN 13779 enables a
ventilation rate of 4 l/s/person for the IDA3 level, which
is half the ventilation rate established by ASHRAE. A local
Israeli regulation is under preparation, which intends to
compromise between the two mentioned documents. Using
this foreseen regulatory provision, a ‘‘fresh’’ air require-
ment of approximately 5 l/s/person ensues, leading to a
required ventilation rate of 720 m3/h in the given 40-pupil
classroom (i.e., 5 air changes p/h).
3. Improving energy performance of schools while coping
with the EE-TC-IAQ Dilemma
3.1. Methodology
The following methodology is suggested for identifying
the range of solutions for construction of schools in warm
climates that may be considered as energy efficient while
providing thermal comfort and indoor air quality, and thus
accepted as deemed-to-satisfy solutions for school build-
ings in the forthcoming Energy Code:
A distinction is made between the modification of
building design variables (such as orientation, size of
windows, thermal insulation, internal mass, color and hue
of facades, etc.) to bring about energy savings while
providing the required amount of fresh air by means of
direct ventilation (presented in Section 3.2 below), and the
modification of ventilation schemes with the intention of
further diminishing energy losses (presented in Section 3.3).
Taking into account the uncontrolled energy losses
associated with natural ventilation, only actively controlled
ventilation schemes are considered.
The effects of building design variables are studied on a
basic representative three-story module of the classroom
Winter (3) wing, comprised of construction elements with the stan-
dard thermal properties. This module is designated as a
58–70 ‘‘basic building’’. The analysis yields a so called ‘‘preferred
0.16–0.24 building’’, which achieves improved energy performance.
20–22
Ta1
The modified ventilation schemes are then applied to the
module representing the ‘‘preferred building’’, and enable
the selection of ‘‘preferred ventilation schemes’’. Perturba-
tions of some building features are then used to re-examine
Summer
26.5
Too Hot 26.0 Acceptable Too Hot
Temperature (C)
25.5
25.0
24.5 Too Cold
24.0
23.5
23.0
22.5
0.5 1.0 -1.0 -0.5 0.0 0.5 1.0
PMV
 ARTICLE IN PRESS
R. Becker et al. / Building and Environment 42 (2007) 3261–3276
previous conclusions regarding the selection of the ‘‘pre-
ferred building’’.
Due to different efficiencies of heating, cooling, lighting
and ventilation devices, energy demands cannot be
summed up and must first be converted to electrical energy
or fuel equivalents. As local devices use only electrical
power, the target function for energy conservation was
chosen as the total annual electrical energy demand per
1 m2 of classroom floor area.
Thermal and energy performance analysis was per-
formed by means of the public-domain program Energy-
Plus [40], which enables the prediction of detailed energy
demand as well as of electricity required for the various
electro-mechanical devices and systems.
Some common assumptions, which stem from non-
energy-related functional requirements of schools, are
made throughout the investigation. In the present study
their values are as follows:
 Occupancy periods: all classrooms are fully occupied
during classroom hours the entire school day, which
lasts from 8:00 to 17:00 Sundays through Thursdays and
from 8:00 to 13:00 on Fridays (loads were adjusted for
the intermission periods between classroom hours),
while no activity takes place on Saturdays. The school
year extends from September 1 until July 14, except for
official holidays.
 Indoor climate control: heating and cooling in every
classroom are provided by means of an electrically-
operated split air-conditioner controlled by an air-
temperature thermostat. Heating is turned on at 7:00,
while cooling only at 8:00. The heating period lasts from
October 1 to March 31, and the cooling period from
April 1 to July 14.
 Ventilation provision: scheduled controlled ventilation
(IAQ ventilation as well as summer night ventilation) is
achieved by means of electrical fans providing the
designated air flow rates in or out of a given space
(between five air changes/h for IAQ ventilation to a
varying night ventilation of 0–30 air changes/h). In
addition, a constant infiltration background rate of
0.5 h1 prevails at all times.
 Lighting control: electrical lighting, controlled by
dimmers, provides backup lighting to ensure at least
300 Lux on all students’ desks during occupancy. The
total maximum heating load, using efficient lighting
devices, amounts to 750 W/class.
 Internal heating loads: internal loads exist during
occupancy, and stem from lighting (maximum 750 W),
people (4000 W), and computers (200 W).
 School location: the school is located in the Mediterra-
nean region, which is simulated by Jerusalem’s Typical
Meteorological Year. A stand-alone building is assumed,
so that no shading is provided by other buildings.
 Construction features: internal floors are heavy-weight
(pre-stressed concrete slabs with 14 cm effective
thickness) with dark flooring (a ¼ 0.7). Internal wall
3265
and ceiling surfaces are light colored (a ¼ 0.4). The
construction features of the external and internal walls
have been part of the investigated design factors, as
explained in Section 3.2.
3.2. Effects of building design factors
The prevailing assumption in this part of the analysis is
that air is drawn into every classroom directly from the
outside by means of a fan designed to provide the
designated air flow rate of 720 m3/h during occupancy.
No active ventilation is provided beyond occupancy hours.
Improving energy efficiency of the building is the primary
concern at this stage, leading to the selection of a
‘‘preferred building’’ that yields the most significant energy
savings. The module representing this building will then
serve as the base-reference for investigating the potential of
energy improvements achieved by utilizing modified
ventilations schemes.
3.2.1. Investigated factors
Energy performance analysis of the module with direct
ventilation addresses the factors presented in column 1 of
Table 2. Column 3 presents the set of regularly used values
of these factors. The effect of each factor was studied
separately, with this factor varied in its entire feasible
range, as presented in column 2 for locally prevalent
values, while the other factors are kept at their regular
values (unless indicated otherwise in the text).
3.2.2. Results and discussion
Detailed results of a thermal and energy analysis always
depend on the climatic conditions and basic assumptions,
and are thus of a local nature. This section presents and
discusses only the main observations that may be of general
interest for the Mediterranean climatic zone and other
similarly warm and moderate climates. All energy con-
servation achievements are compared to the performance
of the ‘‘Basic Building’’ module with construction features
according to column 3 in Table 2.
Results indicated that the following four factors had a
negligible effect on total annual heating and cooling
demands: external wall structure, roof structure, internal
partition structure, and facade color and hue. It was thus
concluded that for these factors the regularly used values
can be considered in all further analyses. Moreover, this
implies that for the sake of a local Energy Code, no
restrictions should be imposed on these factors in the
deemed-to-satisfy solutions.
Analysis of results for bottom floor structure revealed
that this factor has very little impact for classrooms with
eastern and western window orientations. For northern
and southern orientations insulating the floor is beneficial
in the heating season without excessively increasing the
cooling demand, thus enabling a total decrease of up to 3%
 ARTICLE IN PRESS
3266 R. Becker et al. / Building and Environment 42 (2007) 3261–3276

Table 2
Building design factors investigated in order to detect the ‘‘preferred building’’ when direct IAQ ventilation prevails

Design factor (1) Range in investigation (2) Regular value (3) Recommended value (4)

Orientation 0–3601 0–3601 North (01), or South (1801)

External wall structure Internal mass (InM), equally Internal mass (InM) No restrictions
distributed mass (EDM), or
external mass (ExM)
Wall thermal-resistance 0.05–2.55 0.55–1.0 (0.85 for ‘‘basic rX0.85
(m2K/W) building’’)
Bottom floor structure On ground, or thermally On ground When economically justified—
insulated thermally insulated, otherwise—
on ground
Roof structure Regular, or inverted Roof Regular roof No restrictions
Roof thermal-resistance 0.05–2.55 0.55–2.05 (1.65 for ‘‘basic rX2.00
(m2K/W) building’’)
Internal partition structure Lightweight, or heavyweight Heavyweight No restrictions
Fac- ade color and hue Light, gray, or dark Light No restrictions
Window area 8–19.4% of floor area 14.7% of floor area Southern: X10% Northern:
X13%
Window glazing type 1. Single, SC ¼ 1 , Tvis ¼ 0.96 2. Single, SC ¼ 1 Tvis ¼ 0.96 Double Glazing. Southern:
Double with: SC: 0.96–0.15 Tvis: 0.554SC40.5 0.55XTvis40.5
0.84–0.05 Northern: SCX0.6 for tinted
glazing (no restriction for Low_E
glazing) TvisX0.6

in the total annual electricity demand. The recom-
mendation that follows is to opt for an insulated floor
whenever it is economically justified by the construction
technology.
The remaining five factors (window orientation, window
area, glazing type, external wall thermal resistance, and
roof thermal resistance) had much larger effects on annual
acclimatization energy and electricity demands, as de-
scribed below.
Fig. 4 shows the effect of classroom window orientation
on annual electrical energy demands for the internal mass
wall (InM) with three levels of wall insulation: 12 cm
concrete wall without any insulation (dI ¼ 0 cm), and same
wall with external insulation of 3 cm expanded polystyrene
(dI ¼ 3 cm), as well as with 12 cm (dI ¼ 12 cm). All walls
include a 4 cm external stone cladding.
Similar trends can be observed for the poorly insulated,
as well as for the highly insulated cases. Due to solar
radiation, heating demand is lowest for the southern
orientation and largest for the northern one, while cooling
demand is lowest for the northern and southern orienta-
tions and largest for the eastern and western ones.
Electricity demand for lighting is smallest for the south–
western orientation and largest for the eastern one. The
difference between the eastern and western orientation can
probably be attributed to the generally larger level of
cloudiness in the morning hours. Consequently, values of
total electrical energy demand expected in the eastern
and western orientations are more than 20% larger than
those expected in the northern and southern orienta-
tions. Furthermore, addressing the need to prevent direct
penetration of sun radiation in summer, it is concluded that
eastern and western orientations of classroom windows are
not desirable, and should be avoided (eastern and western
orientation have thus been excluded from the current
study).
It is also observed that heating demand is largest, as
expected, for the non-insulated walls, whereas cooling
demand is smallest for these walls. This trend is observed
also in Fig. 5, which shows the effect of wall thermal
resistance (surface to surface), r, for the InM wall, on
annual electrical energy demand for classrooms with
northern and southern windows. As for every other
building type, heating demand decreases with an increase
in thermal insulation. However, thermal insulation pre-
vents the withdrawal of heat generated the large internal
heat loads, increasing the cooling demand as the walls’
thermal resistance increases. Nonetheless, even for the
southern orientation, where cooling is more dominant than
for the northern orientation, the graph of total annual
demand retains a monotonous decreasing concave trend
along the entire feasible range of thermal resistance
values. Similar effects and trends were obtained for roof
insulation. For a specific building, economic considerations
must thus be taken into consideration in order to establish
the optimal insulation thickness for different insula-
tion materials and construction types. However, the
minimal deemed-to-satisfy thermal resistance for walls
and roofs to be stipulated by the Code cannot depend on
the cost of different solutions, but should rather be
determined according to an energy criterion. This criterion
can be chosen so that at the prescribed value, a major
decrease (say, 90%) of the feasible total electricity
reduction range is achieved. For the given climate, the
value of rX0.85 m2 K/W was derived for external walls and
rX2.0 m2 K/W for the roof.
 ARTICLE IN PRESS
R. Becker et al. / Building and Environment 42 (2007) 3261–3276 3267

40
Total - dI=0 cm
Total - dI=3 cm

Annual Electricity (kWh/m2)

30 Total - dI=12 cm
Cooling - dI=0 cm

20 Cooling - dI=3 cm
Cooling - dI=12 cm
Heating - dI=0 cm
10
Heating - dI=3 cm
Heating - dI=12 cm
- Lighting
- 45 90 135 180 225 270 315 360
Window Orientation (Degrees Clockwise) Ventilation

Fig. 4. Effects of classroom window orientation (degrees clockwise from north) on annual electrical energy demands for three levels of thermal insulation
thickness dI ¼ 0, 3, and 12 cm.

40
Total -
South
Annual Electricity (kWh/m2)

30 Cooling -
South

Heating -
20
South

Total -
North
10
Cooling -
North

Heating -
-
North
- 0.5 1.0 1.5 2.0 2.5 3.0
r (m2K/W)

Fig. 5. Effect of wall thermal resistance, r, on annual electrical energy demand for classrooms with northern and southern windows.

Fig. 6 shows the effect of window relative area (defined Two groups are observed according to their glazing:
by the ratio of glazed window area to floor area) on annual LE&C—that includes double glazed windows with low-E
electrical energy demands in the northern and southern and clear glass (Tvis ¼ 0.7 to 0.84, SC ¼ 0.51 to 0.96) as
orientations. Increased window area improves energy well as the single glazed clear glass window (Tvis ¼ 0.96,
performance in both orientations. However, even with SC ¼ 1.0), and T—that includes windows with tinted glass
the largest windows, demand in the northern orientation (Tvis ¼ 0.05–0.62, SC ¼ 0.15–0.72). Properties of readily
remains larger than the largest demand in the southern available commercial double-glazed windows have been
orientation, when windows are smallest. Increasing the used in the investigation. Consequently, the U-value was
window area is much more effective in the northern not identical for all the double glazed windows, and ranged
orientation, but the graphs reveal that the marginal within 1.78–3.23 W m2 K1. Moreover, even within the
improvement achieved beyond a northern glazed ratio of same group, SC values do not vary monotonously with
13% (window size at least 4.0  1.6 m) or beyond a Tvis, thus, when the effect of one factor is considered, the
southern glazed ratio of 10% (window size at least other is not constant, neither changing exactly in the same
5.0  1.0 m) is negligible. Deemed-to-satisfy window areas direction. In addition, for similar SC values, low-E glazing
are thus X13% and X10% of floor area in the northern has much larger Tvis values than tinted glazing. These
and southern orientations, respectively. characteristics of glazing properties explain the distinct
Figs. 7 and 8 show the effects of window glazing type (as difference in the overall energy-related behavior of the two
defined by its light transmission coefficient, Tvis, and glazing groups and the observed ‘‘jumpiness’’ of the
shading coefficient, SC) on annual electrical energy graphs. For LE&C glazing, the regular double-glazed
demands for northern and southern orientations. window with clear glass (Tvis ¼ 0.84, SC ¼ 0.96) yields the
 ARTICLE IN PRESS
3268 R. Becker et al. / Building and Environment 42 (2007) 3261–3276

Southern Orientation Northern Orientation

40

Annual Electricity (kWh/m2)

35

30

25

20
0.08 0.09 0.10 0.11 0.12 0.13 0.14 0.15 0.16 0.17 0.18 0.19 0.20
Aw/Af (-)

Fig. 6. Effect of window relative area, Aw/Af, on annual electrical energy demand for classrooms with northern and southern window orientations.

60
Total - South (LE&C)
Cooling - South (LE&C)
50 Heating - South (LE&C)
Annual Electricity (kWh/m2)

Lighting - South (LE&C)

Total - North (LE&C)
40
Cooling - North (LE&C)
Heating - North (LE&C)
30 Lighting - North (LE&C)
Total - South (T)
Cooling - South (T)
20
Heating - South (T)
Lighting - South (T)
10 Total - North (T)
Cooling - North (T)
Heating - North (T)
0
Lighting - North (T)
0 0.2 0.4 0.6 0.8 1
Tvis

Fig. 7. Effect of window glazing light transmission coefficient, Tvis, on annual electrical energy demand for classrooms with northern and southern
window orientations.

lowest annual energy demand, but for the variety of should be addressed as well. Two indicators have thus been
investigated glazing types the ranges of Tvis and SC hardly defined:
affect the expected total electrical energy demands. Discomfort glare severity indicator (DGSI), indicating
Consequently, from an energy viewpoint, all the glazing the annual cumulative severity of excessive discomfort
types in this group are equivalent, and no preference could glare, GI. It is given by
be established. For T glazing, the expected total electrical X
energy demand generally increases with decreasing Tvis as DGSI ¼ ðGI  20Þ, (1)
well as with decreasing SC, mainly due to the effects of tG

these factors on electricity demand for lighting. For the

where tG denotes the hours for which GI420.
southern orientation, as long as Tvis40.5 and SC40.5,
Summer thermal discomfort severity indicator (STDSI),
the increased demand is moderate in comparison to the
indicating the cumulative severity of excessive mean
expected electrical energy demand with LE&C glazing,
radiant temperature in summer. It is given by:
whereas for the northern orientation only TvisX0.6 and
SCX0.6 yield a comparatively moderate increase. X
STDSI ¼ ðMRT  T a Þ, (2)
In the pursuit of recommended deemed-to-satisfy solu- tMRT
tions the effect of light transmission on glare and the
effect of penetrating solar radiation on summer thermal where MRT is the mean radiant temperature obtained
comfort in classes with southern orientation windows by EnergyPlus at the room center, Ta is the room air
 ARTICLE IN PRESS
R. Becker et al. / Building and Environment 42 (2007) 3261–3276 3269

60
Total - South (LE&C)
Cooling - South (LE&C)
50 Heating - South (LE&C)

Annual Electricity (kWh/m2)

Lighting - South (LE&C)
Total - North (LE&C)
40
Cooling - North (LE&C)
Heating - North (LE&C)
30 Lighting - North (LE&C)
Total - South (T)
Cooling - South (T)
20
Heating - South (T)
Lighting - South (T)
10 Total - North (T)
Cooling - North (T)
Heating - North (T)
0
Lighting - North (T)
0 0.2 0.4 0.6 0.8 1
SC

Fig. 8. Effect of Window Glazing Shading Coefficient, SC, on annual electrical energy demand for classrooms with northern and southern window
orientations.

South North
22,000
20,000
18,000
16,000
14,000
12,000
DGSI

10,000
8,000
6,000
4,000
2,000
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Tvis

Fig. 9. Effect of window glazing light transmission coefficient, Tvis, on the excessive glare index, DGSI, for classrooms with northern and southern
window orientations.

temperature, and tMRT denotes the hours for which for southern classroom windows—0.55XTvis40.5 and
MRT4Ta during the cooling season. 0.554SC40.5.
The effects of Tvis and SC on DGSI and STDSI, The entire set of recommendations derived for the
respectively, are illustrated in Figs. 9 and 10. deemed-to-satisfy Energy Code solutions are summarized
As expected, DGSI and STDSI generally decrease with in column 4 of Table 2. The implementation of these
decreasing Tvis and SC, but their values in the southern recommendations yields the so called ‘‘preferred building’’.
orientation are substantially larger than those obtained in The ‘‘preferred building’’ does not have windows in
the northern orientation. Accepting the performance level western or eastern orientations, which may cause energy
of northern classes with regular clear glass double-glazed demands by some 7% to 21% larger than those predicted
windows (Tvis ¼ 0.84, SC ¼ 0.96) as a threshold level, for the northern and southern orientations, as well as
glazing with 0.55XTvis and 0.554SC should be preferred excessive glare and direct solar radiation on pupils. The
in the southern orientation. representative modules of the ‘‘preferred building’’ are
Combining these recommendations with those derived denoted by IAQ_5_N and IAQ_5_S for the northern and
when observing Fig. 7, the following deemed-to-satisfy southern orientations, respectively. Minor electrical energy
ranges are suggested: for northern classroom windows— savings of 2% are obtained in northern classrooms in
TvisX0.6 and, when tinted glazing is used, SCX0.6; comparison to the electrical energy demand of the
 ARTICLE IN PRESS
3270 R. Becker et al. / Building and Environment 42 (2007) 3261–3276

South North
1,400
1,300
1,200
1,100
STDSI 1,000
900
800
700
600
500
400
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
SC

Fig. 10. Effect of window glazing shading coefficient, SC, on the excessive mean radiant temperature index, STDSI, for classrooms with northern and
southern window orientations.

18

16
Annual Electricity(kWh/m2)

14
Cooling +
Ventilation_S
12
Cooling_S
10
Ventilation_S
8 Cooling +
Ventilation_N
6
Cooling_N
4 Ventilation_N
2

-
- 5 10 15 20 25 30
N (h-1)

Fig. 11. Effect of night ventilation rate, N, on annual electrical energy demand for classrooms with northern and southern window orientations.

commonly constructed ‘‘Basic Building’’, but a slight
increase of 2.5% is expected for the southern classrooms
due to the preferred use of glazing with lower Tvis and SC
in order to prevent glare and excessive mean radiant
temperatures.
The ‘‘preferred building’’ modules will be used in the
next steps for the investigation of modified ventilation
strategies and for the evaluation of their effect on
additional energy savings. Results will be compared to
those obtained for IAQ_5_N and IAQ_5_S.
3.3. Effects of ventilation schemes
3.3.1. Night ventilation
The existence of ventilation fans for the provision of
IAQ ventilation enables the introduction of summer night
ventilation without adding new equipment. The analysis in
this section is based on the assumption that during the
cooling season the fan is operated automatically and
exterior air is introduced into the classrooms whenever
TaToX0.5 1C. This night ventilation scheme is denoted
by the symbol DN.
Fig. 11 shows the effect of night ventilation rate, N, on
total annual electrical energy demands.
As expected, fan electricity demand increases with
increasing N, while active cooling electricity demand
decreases. A very flat region in the total annual electricity
demand is evident for air change rates greater than 5 h1,
with an optimum at NE10 h1.
The annual electricity saving enabled by the implemen-
tation of DN night ventilation is 11.7% and 13% in
northern and southern orientations, respectively.
Based on these results, it was deduced that deemed-
to-satisfy solutions should recommend the provision of
night ventilation during the cooling season at a rate of
N45 h1.
In order to check whether previously recommended
values for the building factors, which were derived in
 ARTICLE IN PRESS
R. Becker et al. / Building and Environment 42 (2007) 3261–3276 3271

Section 3.2, need to be altered after the addition of night
ventilation, a sensitivity analysis was conducted at the
N ¼ 10 h1 level. Perturbations around the preferred value
were investigated separately for every factor. None of the
previous recommendations had to be changed.
3.3.2. Effects of improved IAQ and night ventilation schemes
The basic assumption in the next step is that the specific
geometrical configuration of a school building and its
occupation pattern provide an additional feature for
improving energy saving. The main contribution stems
from the large public spaces in front of classrooms
(corridor, hall, or atrium), which are only partially
occupied during intermissions between lectures and are
completely vacated during class hours. When open to
external air, these spaces respond passively to the
surrounding conditions and tend to develop air tempera-
tures beyond the comfort limits during many hours. The
main hypothesis is that with adequate ventilation schemes,
these spaces can be utilized as a buffer zone, which is
warmer than the external air in winter and cooler in
summer, thus enabling pre heating of incoming IAQ
ventilation air in winter, and increasing heat losses through
the door-sided wall of the classrooms in summer. The
improved thermal comfort level in the buffer zones is in
itself a minor additional-value byproduct that is not
discussed quantitatively in this paper.
IAQ ventilation schemes that were considered, and their
symbols, include:
ID exterior air drawn directly into
classrooms and withdrawn directly to
the outside.
IDC exterior air drawn directly into
classrooms and withdrawn to the outside
through a window-closed corridor.
IVC exterior air drawn into classrooms
through the window-closed corridor and
withdrawn directly to the outside.
ISDCWVC a combination of IDC in summer and
IVC in winter.
Night ventilation schemes that were considered, and their
symbols, include:
ND exterior air drawn directly into classrooms and
withdrawn directly to the outside.
NDC exterior air drawn directly into classrooms and
withdrawn to the outside through a window-closed
corridor.
NVC exterior air drawn into classrooms through the
window-closed corridor and withdrawn directly to
the outside.
The ventilation combinations that have been analyzed in
order to study the effect of ventilation schemes are listed in
Table 3.
In order to portray the adverse effect of IAQ ventilation
on energy demand the analysis includes the cases with no
active ventilation, IAQ_0_N and IAQ_0_S, which cannot
ensure an adequate indoor air quality, but are otherwise
identical to the ‘‘preferred building’’ modules IAQ_5_N
and IAQ_5_S.
Fig. 12 shows the resulting annual electrical energy
demands for the investigated combinations.
It can be noticed that without the provision of IAQ
ventilation, heating demand is almost nil for the southern
as well as for the northern orientation due to the large
internal heat sources. Provision of the required minimum
IAQ ventilation rate of 5 h1 increases heating demand
significantly (by more than 700%), although somewhat
reducing the cooling demand (by less than 11%). At this
point, additional runs, which are not presented here, were
performed in order to obtain the optimal summer daytime
ventilation rate. It was found to be smaller than the
required 5 h1. Consequently, this optimum has no
practical significance and the IAQ air change rate of
5 h1 is kept throughout the year. Results obtained for the
improved ventilation schemes, as analyzed below, are thus
compared to those obtained for cases IAQ_5_N and
IAQ_5_S.
Changing IAQ ventilation from ID to IVC (schemes
DIR_NC0_N and DIR_NC0_S compared to IAQ_5_N

Table 3
Combinations of analyzed ventilation schemes

Option index Night ventilation IAQ ventilation

1
South (1) North (2) Scheme (4) Rate (h ) (5) Scheme (6) Rate (h1) (7)

IAQ_0_S IAQ_0_N — 0 — 0
IAQ_5_S IAQ_5_N — 0 ID 5
IAQ_5_NC10_S IAQ_5_NC10_N ND 10 ID 5
DIR1_NC0_S DIR1_NC0_N — 0 IVC 5
DIR2_NC0_S DIR2_NC0_N — 0 IDC 5
MIX1_NC0_S MIX1_NC0_N — 0 ISDCWVC 5
MIX2_NC10_S MIX2_NC10_N ND 10 ISDCWVC 5
MIX3_NC10_S MIX3_NC10_N NVC 10 ISDCWVC 5
MIX4_NC10_S MIX4_NC10_N NDC 10 ISDCWVC 5
 ARTICLE IN PRESS
3272 R. Becker et al. / Building and Environment 42 (2007) 3261–3276

Lighting Ventilation Heating Cooling

30

25

Annual Electricity (kWh/m2) 20

15

10

0
MIX2_NC10_N
MIX3_NC10_N
MIX4_NC10_N

IAQ_5_NC10_S
IAQ_0_N
IAQ_5_N
IAQ_5_NC10_N
DIR1_NC0_N
DIR2_NC0_N
MIX1_NC0_N

PREF_VENT_N
IAQ_0_S
IAQ_5_S

DIR1_NC0_S
DIR2_NC0_S
MIX1_NC0_S
MIX2_NC10_S
MIX3_NC10_S
MIX4_NC10_S
PREF_VENT_S
Fig. 12. Effects of various ventilation schemes on annual electrical energy demand for classrooms with northern and southern window orientations.

and IAQ_5_S) proves to be very efficient for reducing the Adding night ventilation to the daytime ISDCWVC IAQ
heating energy demand in both orientations. Exterior air ventilation further reduces cooling energy. Results in
warms up while passing through the closed corridor, Fig. 12 indicate that for the northern classroom, ND and
which, although not actively heated is at an elevated NDC night ventilation are much more effective than NVC,
temperature compared to the external ambient conditions. with NDC (scheme MIX4_NC10_N) yielding a total
In both orientations the corridor warms up due to heat annual reduction of 28% (in comparison to IAQ_5_N).
gains transferred through the separation wall with the For the southern classroom, the three night ventilation
adjacent classroom. A south facing corridor gains addi- schemes yield similar improvements, with a total annual
tional substantial heat from solar radiation and serves as a reduction of nearly 17% (in comparison to IAQ_5_S).
‘‘sun space’’. Consequently, the pre-warmed fresh air These total reductions are larger than the 11.7% and 13%
enables a reduction of some 23% in heating electricity for reductions derived by night time ventilation alone (see
the southern classrooms and 68% for the northern ones. Section 3.3.1).
However, due to the above detailed mechanisms, this Combining the most efficient schemes for winter and
ventilation scheme, when applied in summer, may increase summer improvements yielded the PREF_VENT_N and
the cooling electricity demand for the northern classroom PREF_VENT_S possible solutions for the northern and
by some 29%. For the southern classrooms the increment southern orientations, respectively (see Fig. 12). These
is not that large (only 13%) since the north facing include: In winter—for both orientations a closed corridor
corridor experiences much smaller solar gains during with IAQ ventilation according to the IVC scheme
summer. (reducing heating electricity by 68% for the northern
Changing IAQ ventilation from ID to IDC (schemes classroom and 23% for the southern one). In summer—for
DIR2_NC0_N and DIR2_NC0_S) reduces the buffer both orientations an open corridor with IAQ as well as
space temperatures in summer, and consequently summer night ventilation according to the IDC scheme (reducing
cooling electricity increments are now only 15% and 7.5% cooling electricity by 23% for the northern classroom and
in northern and southern classroom window orientations, 28% for the southern one). The total electricity reductions
respectively. obtained by these schemes (in comparison to IAQ_5_N
The combination of IVC in winter and IDC in summer and IAQ_5_S) are 30% for the northern classrooms and
(ISDCWVC, as implemented in schemes MIX1_NC0_N 18% for the southern ones.
and MIX1_NC0_S) can reduce heating electricity demand Consequently, it was deduced that improved ventilation
by 68% and 23% for northern classrooms and southern schemes are largely effective for northern classrooms,
classrooms, respectively, while increasing cooling demands whereas for southern classrooms they yield only minor
by only 15% and 7.5%, respectively. Consequently, it may energy reductions.
reduce total electrical energy demands by 11% for northern Recommended deemed to satisfy solutions thus include:
classrooms and only 1% for southern ones. closing corridor windows in winter and opening them in
 ARTICLE IN PRESS
R. Becker et al. / Building and Environment 42 (2007) 3261–3276 3273

35 InM-Basic-S
ExM-Basic-S
Basic-S Basic-N

Annual Electricity (kWh/m2)

EDM-Basic-S
30
InM-Basic-N
ExM-Basic-N

25 EDM-Basic-N
InM-PrefVent-S
PrefVent-S
ExM-PrefVent-S
20 EDM-PrefVent-S
PrefVent-S InM-PrefVent-N
ExM-PrefVent-N
15
0 0.5 1 1.5 2 2.5 3 EDM-PrefVent-N

Thermal Resistance (m2K/W)

Fig. 13. Sensitivity analysis of total electricity demand in southern and northern classrooms to external wall structure (InM, ExM and EDM) and thermal
resistance for the ‘‘basic building’’ and the one with ‘‘preferred ventilation schemes’’.

summer for both window orientations. For northern 4. Conclusions

classrooms—IAQ ventilation according to the IVC scheme
in winter and the ID or IDC scheme in summer, and The large ventilation rates required in crowded spaces in
summer night ventilation according to the ND or NDC order to provide adequate indoor air quality may cause
scheme. significant energy losses in school buildings. Seeking to
For southern classrooms—any combination of IAQ establish deemed-to-satisfy solutions as the base-line for a
ventilation and summer night ventilation. performance-based Energy Code for schools, two tasks
It may also be noted that the recommended ventilation have been accomplished: (i) establishing the preferred set of
schemes reduce electricity demands below the levels of the building design variables for schools that are energy
non-ventilated cases IAQ_0_N and IAQ_0_S, thus redu- efficient while providing acceptable levels of indoor air
cing the severity of the EE-TC-IAQ Dilemma. quality, as well as thermal and visual comfort, and (ii)
establishing a set of preferred ventilation schemes which
enhance the energy efficiency without reducing other
3.3.3. Integration of results performance levels. The first task is a classical energy
The preferred building design variables deduced in efficiency problem, leading in the present study to the
Section 3.2 were used in Sections 3.3.1 and 3.3.2 in order provision of the ‘‘preferred building’’ which served as a
to derive the preferred ventilation schemes. In order reference for the second task. It may enable a minor
to check whether the two sets of decisions can be electrical energy saving of 2% in northern classrooms in
simply integrated, sensitivity analyses were conducted. comparison to the electrical energy demand of the
In each analysis one building design variable was commonly constructed ‘‘basic building’’, but may cause a
varied while the ventilation schemes and all other design slight increase of 2.5% for southern classrooms due to the
variables were kept at their preferred solutions and values. preferred use of glazing with lower lighting and solar
Fig. 13 shows the analysis results for total electricity transmittance in order to prevent glare and excessive mean
demand in southern and northern orientations (indicated radiant temperatures.
by S and N, respectively) versus the external wall’s The existence of large, usually non-occupied, public
structure (location of mass indicated by InM, ExM or spaces in school buildings enables their utilization as buffer
EDM) and thermal resistance. Except for a shift in the zones for pre-warming the fresh ventilation air in winter,
overall range of total electricity demand, the trends of and for increasing removal of heat stored in classrooms’
the lines for the buildings with a direct IAQ ventilation floors and walls in summer. Analysis of various combina-
scheme and no night ventilation (indicated by Basic) tions of ventilation schemes with identical ventilation rates
are similar to those obtained for the buildings with during class hours, revealed that additional electricity
preferred ventilation schemes (indicated by PrefVent). savings can be achieved, allowing overcoming the de-
Similar results were obtained for other variables, indicating creased energy performance induced by indoor air quality
that the recommendations derived in Section 3.2 are (IAQ) ventilation. The preferred ventilation mode for the
relevant to the buildings with preferred ventilation schemes analyzed weather conditions (which represent the moderate
as well. Mediterranean climate) includes improved schemes for
 ARTICLE IN PRESS
3274 R. Becker et al. / Building and Environment 42 (2007) 3261–3276

year round IAQ ventilation combined with night ventila-
tion during the summer cooling period. The main outcomes
include:
(1) Corridor windows should be closed in winter and open
in summer.
(2) The preferred IAQ ventilation schemes in northern as
well as in southern classrooms are composed of: In
winter—drawing fresh outside air into the classrooms
through closed corridors, halls or atrium. In summer—
drawing exterior air directly into the classrooms and
exhausting it via open corridors, halls or atrium.
(3) The preferred summer night ventilation in northern
classrooms is composed of drawing exterior air directly
into the classroom and exhausting it directly to the
outside or via open corridors, halls or atrium. For
southern classrooms all night ventilation schemes have
very similar effects.
(4) When compared to the energy performance of the
‘‘preferred building’’, the total electrical energy saving
achieved by the application of the ‘‘preferred ventilation
schemes’’ may be in the order of 28% to 30% for north
facing classrooms and 17% to 18% for south facing ones.
(5) Sensitivity analyses showed that recommendations for
the building design variables need not to be altered
when the ‘‘Preferred Ventilation Schemes’’ are applied.
Acknowledgments
The paper is based on a research project supported by
the Israeli Ministry of National Infrastructures and the
Rachel Shalon Foundation.

Dry Bulb Tempe rature {C} Dew Point Temperature {C}

35
January

30

25
Temperature (˚C)

20

15

10

-5
0 24 48 72 96 120 144 168 192 216 240 264 288 312 336 360 384 408 432 456 480 504 528 552 576 600 624 648 672 696 720 744

Time (h)

35
June

30

25
Temperature (˚C)

20

15

10

-5
0 24 48 72 96 120 144 168 192 216 240 264 288 312 336 360 384 408 432 456 480 504 528 552 576 600 624 648 672 696 720

Time (h)

Fig. A.1. Ambient air temperature and dew-point temperature variations in January (winter) and June (summer) in Jerusalem.
 ARTICLE IN PRESS
R. Becker et al. / Building and Environment 42 (2007) 3261–3276
Table A.1
Characteristic data (external air temperature and global horizontal radiation) for the Jerusalem climate during the heating (November–March) and
cooling (September, October and April–July) periods
Heating period
2
Ext. air temperature (1C) Radiation (W/m )
Ave. (1) Std. dev. (2) Max. (3) Min. (4) Max. (5)
10.10 4.04 28.0 0.8 957
Appendix A. Typical climatic data
This appendix includes some characteristic values for the
typical meteorological year of Jerusalem, which was used in
this paper as representing the weather in the Mediterranean
region Fig A.1.
Periodic averages, maxima and minima of dry-bulb air
temperature, global solar radiation on a horizontal surface,
and cloudiness are given in Table A.1.
References
[1] Besler GJ, Besler M. Towards healthy microclimate of closed spaces
and habitats. Environmental Protection Engineering 2000;26(3):
23–38.
[2] Ellis R. Commissioning for indoor air quality. Engineering Systems
2003;20(4):30.
[3] Davies MG. Passive solar heated school in Wallasey. IV: an
observational study of the thermal response of a passive school
building. International Journal of Energy Research 1986;10(4):
305–32.
[4] Davies MG, Davies DM. Passive solar heated school in
Wallasey. VII: window opening behaviour and the microclimate of
a classroom. International Journal of Energy Research 1987;
11(3):315–26.
[5] Norton DI. Elementary school utilizes solar design techniques and
groundwater cooling system for energy efficiency. ASHRAE Journal
1995;37(3):50–2.
[6] Ubertini S, Desideri U. Design of a solar collector for year round
climatization. Renewable Energy 2003;28(4):623–45.
[7] Penman JM, Murdoch N. Thermal response of temporary school
buildings: I. Heating. Building Services Engineering Research and
Technology 1989;10(2):61–8.
[8] Ghrab MN, Bouden C, Franchisseur R. Overheating caused by solar
elements in Tunis. Effectiveness of some ways to prevent it.
Renewable Energy 1993;3(6–7):801–11.
[9] Zurigat YH, Al-Hinai H, Jubran BA, Al-Masoudi YS. Energy
efficient building strategies for school buildings in Oman. Interna-
tional Journal of Energy Research 2003;27(3):241–53.
[10] Stotz RB. An energy efficient HVAC system at a high school.
ASHRAE Transactions 1992;98(2):593–8.
[11] Butala V, Novak P. Energy consumption and potential energy
savings in old school buildings. Energy and Buildings 1999;29(3):
241–6.
[12] Henderson Jr HI, Walburger AC, Sand JR, Mescher K. The impact
of dessicant dehumidification on classroom humidity levels. ASH-
RAE Transactions 2002;108(2):546–54.
[13] Rosenbaum M. A green building on campus. ASHRAE Journal
2002;44(1):41–4.
[14] Rafferty K. Groundwater heat pump systems: experience at two high
schools. ASHRAE Transactions 1996;102(1):922–8.
3275
Cooling period
Ext. air temperature (1C) Radiation (W/m2)
Ave. (6) Std. dev. (7) Max. (8) Min. (9) Max. (10)
19.75 5.25 34.2 2.5 1054
[15] Shonder JA, Martin MA, Sharp TR, Durfee DJ, Hughes
PJ. Benchmark for performance: geothermal application in
Lincoln Public Schools. ASHRAE Transactions 1999;105(Part 2):
1199–207.
[16] Kavanaugh SP, Xie L. Energy use of ventilation air conditioning
options for ground-source heat pump systems. ASHRAE Transac-
tions 2000;106:543–50.
[17] Scott A. Geothermal heat pumps the answer for college. HPAC—
Heating, Piping, Air Conditioning Engineering 2002;74(7):65.
[18] Downing CC, Bayer CW. Classroom indoor air quality vs.
ventilation rates. ASHRAE Transactions 1993;99(2):1099–103.
[19] Cheong KWD, Lau HYT. Development and application of an indoor
air quality audit to an air-conditioned tertiary institutional building
in the tropics. Building and Environment 2003;38(4):605–16.
[20] Alexander DK, Vaughan ND, Jenkins HG, O’Sullivan PE. Energy
performance assessment of Looe junior and infant school. Interna-
tional Journal of Ambient Energy 1990;11(1):17–30.
[21] Mysen M, Schild PG, Hellstrand V, Thunshelle K. Evaluation of
simplified ventilation system with direct air supply throught he
fac- ade in a school in a cold climate. Energy and Buildings
2005;37(2):157–66.
[22] Davanagere BS, Shirley DB, Rengarajan K, Colacino F. Mitigating
the impacts of ASHRAE Standard 62-1989 on Florida schools.
ASHRAE Transactions 1997;103(1):241–58.
[23] Dorer V, Weber A. Air, contaminant and heat transport models:
Integration and application. Energy and Building 1999;30(1):
97–104.
[24] Eriksson J, Wahlstrom A. Use of multizone exchange simulation to
evaluate a hybrid ventilation system. ASHRAE Transactions
2002;108(2):811–7.
[25] Becker R, Paciuk M. Upgrading energy efficiency, comfort and air
quality in an Israeli school. Proceedings of the 2nd international
conference on building physics, Leuven, Belgium, September. 2003,
p. 929–35.
[26] Harris DJ. Performance of atria in two school buildings. Interna-
tional Journal of Ambient Energy 1997;18(4):171–6.
[27] Adel El-Kadi AEWM, Fanny MA. Architectural design and thermal
performance of school sports halls. Applied Energy 2003;76(1–3):
289–303.
[28] ASHRAE. Comfort Standard 55. Thermal Requirements for Human
Occupancy. Atlanta: American Society of Heating, Refrigerating and
Air-Conditioning Engineers; 1992.
[29] ISO. International Standard 7730. Moderate Thermal Environ-
ments—determination of the PMV and PPD indices and specification
of the conditions for thermal comfort. Geneva: ISO—International
Standards Association; 1994.
[30] Fanger PO. Thermal comfort: analysis and applications in environ-
mental engineering. Copenhagen: Danish Technical Press; 1970.
[31] Kwok AG, Chun C. Thermal comfort in Japanese schools. Solar
Energy 2003;74(3):245–52.
[32] Yoshida JA, Nomura M, Mikami K, Hachisu H. Thermal comfort of
severely disabled children in nursery schools in Japan. Proceedings of
the XIVth triennial congress of the international ergonomics
association, Ergonomics for the New Millenium. San Diego, CA,
USA. 2000. pp. 712–5.
 ARTICLE IN PRESS
3276 R. Becker et al. / Building and Environment 42 (2007) 3261–3276
[33] Xavier AADP, Lamberts R. Indices of thermal comfort developed
from field survey in Brazil. ASHRAE Transactions 2000;106:
45–58.
[34] Kwok AG. Thermal comfort in tropical classrooms. ASHRAE
Transactions 1998;104(1B):1031–47.
[35] Paciuk M. The role of personal control of the environment in thermal
comfort and satisfaction at the workplace. In: Proceedings of the 21th
annual conference of the environmental design research association.
Coming of Age, Champaign-Urbana, April 6–9, 1990. p. 303–12.
[36] Paciuk M. An expanded model of thermal comfort in the office
environment. In: Proceedings of the 6th international conference on
indoor air quality and climate, indoor air vol. 93, Helsinki, July 4–8,
1993. p. 49–54.
[37] Becker R, Paciuk M. Thermal comfort in residential buildings—
failure to predict by standard model. Building and Environment
Journal. 2006, Submitted for publication.
[38] ANSI/ASHRAE 62-1999. Ventilation for acceptable indoor air
quality. American Society of Heating, Refrigerating and Air-
Conditioning Engineers.
[39] CEN, prEn13779:1999 E. Ventilation for buildings—performance
requirements for ventilation and air-conditioning systems. CEN,
European committee for standardization, Brussels.
[40] US DOE. Energy efficiency and renewable energy, building
technologies program and software tools. Website: /http://www.
eere.energy.gov/buildings/energy_tools/energyplus/S. USA Depart-
ment of Energy.