Energy and Buildings 150 (2017) 224238

Contents lists available at ScienceDirect

Energy and Buildings

journal homepage: www.elsevier.com/locate/enbuild

A comprehensive cost-optimal approach for energy retrot of existing

multi-family buildings: Application to apartment blocks in Turkey
Nese Ganic Saglam a, , A. Zerrin Ylmaz a , Cristina Becchio b , Stefano P. Corgnati b
a
Istanbul Technical University (ITU) Faculty of Architecture, Taskisla Campus, 34437, Istanbul, Turkey
b
DENERG, Politecnico di Torino, Corso Duca degli Abruzzi 24, 10129 Torino, Italy

a r t i c l e i n f o a b s t r a c t

Article history: This research presents a comprehensive cost-optimal approach for existing building retrots. The
Received 21 December 2016 approach incorporates energy efciency measures related to building envelope, building energy systems
Received in revised form 4 April 2017 and renewable energy, all at once. Moreover, the cost-optimal calculation scope is extended through
Accepted 10 June 2017
integration of occupant behaviour.
Available online 13 June 2017
The comprehensive approach is implemented for a reference building which represents high-rise apart-
ment blocks in Turkey; these building types accommodate 4500 families corresponding to 23.1% of
Keywords:
the national households. Three different climatic regions, that show signicant distinction from each
Energy efciency measures
Energy retrot
other, were selected for the sample application of the approach in order to refer different climates. Cost-
Cost-optimal level optimal calculations were performed for more than 1300 scenarios. Integration of occupant behaviour
Cost-effectiveness into this calculation procedure was demonstrated with sensitivity analyses; the present application of
Occupant behaviour the approach was focused on the use of window openings by occupants.
Multi-family buildings Results reveal that cost-effective energy saving potential is higher than 70% in high-rise apartments.
Climatic conditions Moreover, occupant behaviour related to window openings is able to raise this rate above 80%. The
ndings indicate the necessity of comprehensive approach and demonstrate that integration of occupant
behaviour into cost-optimal calculations is fundamental in order to reach signicant and stable results.
2017 Elsevier B.V. All rights reserved.

1. Introduction
Energy efciency in buildings is accepted as one of the key
strategies to achieve intended energy saving in the world since
energy consumed in building sector corresponds to one-third of
nal energy consumption [1]. Accordingly in European Union (EU),
building sector is responsible from the largest share as well and
leads to 40% of the nal energy consumption [2]. Therefore, building
energy efciency targets of EU were set through the recast of Energy
Performance of Buildings Directive 2010/31/EU (EPBD recast) [3].
Considering that realization of high energy performance in build-
ings is not only an energy aspect but also economic practice, EPBD
recast introduced cost-optimality and nearly-zero energy build-
ing (NZEB) concepts for building energy performance assessment.
According to EPBD recast, Member States are obliged to calculate
cost-optimal energy efciency levels for buildings by adapting the
methodology framework provided in EU Regulation No.244/2012
[4,5] and to compare the results with the existing requirements
Corresponding author.
E-mail addresses: neseganic@gmail.com, ganicn@itu.edu.tr (N.G. Saglam).
in force. Subsequently, after 2020, the cost optimal energy perfor-
mance level is expected to meet the NZEB level [6].
In Europe, approximately 40% of the building stock was con-
structed before 1960s [7] and these existing buildings represent the
highest energy saving potential [8]. Therefore, cost-optimality con-
cept of EPBD recast refers not only to new constructions but also
to existing building retrots. Since 75% of the existing buildings
in Europe are residential [9], considerable number of the studies
focused on the cost-optimal retrots of national residential build-
ing stock as presented below.
In Sweden, Brown et al. investigated residential building stock
and suggested to consider also national environmental ratings in
order to integrate indoor environmental quality assessment into
building energy retrot process [10]. Liu et al., on the other hand,
considered future targets and indicated that although there are
challenges, 50% saving potential exists in Swedish multifamily
buildings energy use by 2050 [11]. Bonakdar et al. presented a
method for cost optimal analysis of multifamily building retrots
which includes comparison between different nancial scenarios
and the national building codes of Sweden [12]. In Estonia, as the
result of their analyses on wooden apartment buildings, Arumgi
and Kalamees [13] declared that in order to ensure cost-effective

http://dx.doi.org/10.1016/j.enbuild.2017.06.026
0378-7788/ 2017 Elsevier B.V. All rights reserved.
 N.G. Saglam et al. / Energy and Buildings 150 (2017) 224238
energy savings, envelope retrots are required to be supported by
building service system retrots. Kuusk et al. [14] investigated the
brick apartments and pointed out the necessity of deep renova-
tion perspective and nancial support. Eventually these research
activities supported the Estonian energy roadmap development, as
presented by Kurnitski et al. in [15] and Pikas et al. in [16]. Similar
to the procedure followed in Estonia, also in Portugal the research
activities on residential building refurbishment resulted with a sup-
port to the policy-makers. Brando de Vasconcelos et al. started
from the reference building denition procedure [17] as the rst
stage and then focused on the whole cost-optimal approach for Por-
tuguese residential building stock [18] until the sensitivity analyses
which were concluded with policy implications [19]. The authors
reported that envelope measures are required to be combined as
packages in order to obtain more benets and discount rates are
one of the major factors affecting the cost-optimal calculations. In
Italy, Corrado et al. studied a new procedure for the optimization
of the cost optimal levels of an Italian residential reference build-
ing [20]. Fabbri et al. [21] identied strengths and weaknesses of
the cost-optimal approach through its application to the refurbish-
ment of a residential case study. Desogus et al. [22] investigated
existing masonry public dwellings in terms of cost-efciency with
different investment evaluation methods. Becchio et al. highlighted
the use of cost-optimal methodology as a decision-making tool for
supporting architects in the retrot energy design of a single family
house [23].
The abovementioned studies validate the signicance of cost-
effective retrot of existing residential buildings on achieving the
ambitious future energy efciency targets. These recent studies
also point out essentiality of examining retrot measure combina-
tions rather than considering only single measures and moreover
emphasise the importance of covering building service system
retrots in addition to envelope retrots. Nevertheless, as Pombo
et al. demonstrated in their review [24], among the research studies
on cost-effective retrot of existing residential buildings, passive
strategies are the most commonly addressed measures and there
are only limited number of studies combining retrot measures
related to passive strategies, building service systems and renew-
able energy use all at once. As a consequence of these ndings,
it is obvious that a comprehensive approach is required for the
cost-optimal analyses.
On the other hand, while the cost-optimal calculations for exist-
ing building retrots have still been progressing, new perspectives
were introduced. The recent studies in literature indicated the
necessity of addressing occupant behaviour within NZEB analyses
through cost-optimal approach. As Barthelmes et al. revealed in
their research [25], occupant behaviour affects the energy perfor-
mance of NZEBs more than 150% and accordingly it is needed to be
addressed in related studies. Moreover, Becchio et al., drew atten-
tion to the importance of regarding the effect of occupant behaviour
in cost-optimal analyses in order to actualize future targets based
on recently raised concepts such as post-carbon cities as explained
in [26]. Therefore, the existing cost-optimal approach is also needed
to be extended by regarding effect of occupant behaviour.
Considering the abovementioned gaps and necessities in cost-
optimal calculations, the aim of this study is to introduce a
comprehensive cost-optimal approach for evaluating existing
building retrots. The approach incorporates three main con-
stituents of a whole building retrot: envelope, building energy
systems and renewable energy use. Moreover, it extends the
scope of cost-optimal calculations by integrating effect of occupant
behaviour into the energy and economic assessment of building
retrots. In this way the approach regards occupant behaviour as
an integral part of cost-effective retrot of existing buildings to
guide actualization of recently introduced future targets.
225
In line with the aim, this paper demonstrates the proposed com-
prehensive approach through a sample application for Turkey. This
sample application of the approach may refer to the various build-
ing refurbishment strategies in Europe as long as Turkey involves
different climatic regions that show signicant distinction from
each other. Although the mild climate is observed in the greater
area of Turkey, Turkeys cold climatic region is able to refer North-
ern Europe since it is characterized by cold, strong and long winter
period where the air temperatures are mostly below zero. On the
contrary, hot humid climatic region, which appears in the Mediter-
ranean coast of Turkey with hot humid summers and warm wet
winters, is indeed able to refer to other Mediterranean countries.
Moreover, in Turkey, similar trends that characterize the Euro-
pean building stock are recognizable since buildings sector is the
second biggest energy consumer that is responsible for 30% of
total nal energy consumption [27] and correspondingly ensuring
energy efciency in buildings is among the future strategic tar-
gets of the country [28]. This strategic target is indeed related to
recent EU legislation on building energy performance that Turkey
follows within the EU harmonization process [29,30]. Since resi-
dential buildings represent 75% of the national building stock [31],
EPBD recast related research activities in Turkey also focused ini-
tially on the residential buildings [32,33]. Findings of these research
activities are in line with previously mentioned studies as well,
as Ashraan et al. indicated that it is necessary to develop stud-
ies which combine envelope retrots with mechanical system
improvements in order to achieve lower energy consumption levels
in building retrots [34].
Accordingly in this study, a Turkish residential reference build-
ing (RB) is examined for the sample application of the proposed
approach. The RB is a high-rise apartment building; the choice of
this building typology is due to the fact that in Turkey 23.1% of
households, which correspond to 4500 families, reside in dwellings
which have 6 or more oors [35].
In order to accomplish a complete view, mild, hot-humid and
cold climatic regions of Turkey were taken into consideration
based on their potentials to refer European climates. These cli-
matic regions are represented respectively by Istanbul, Antalya and
Erzurum cities.
In compliance with the cost-optimal methodology framework,
retrot scenarios, consist of energy efciency measures or packages
of measures, were investigated in terms of both primary energy
consumption and global cost with the aim of deriving cost-optimal
levels.
Energy consumption of the RB is calculated using the detailed
dynamic simulation tool EnergyPlus (version 8.2) [36]. Energy con-
sumption of different retrot scenarios are also evaluated using
this tool and converted in primary energy to make a comparative
assessment, that considers primary energy consumption for space
heating and cooling, domestic hot water production (DHW) and
lighting.
Economic calculations of this study are based on the global
cost concept of EN15459 standard [37]. Global cost calculations
were performed for each scenario considering 30 years calcula-
tion period. These calculations mainly consider investment cost,
replacement cost, energy cost, maintenance cost and residual value
at the end of the calculation period.
The cost-optimal levels of RB retrots were derived considering
a reference occupant behaviour at the beginning. Then, sensitivity
analyses on cost-optimal calculations were performed to analyse
the effect of occupant behaviour relevant to use of window open-
ings. This extended cost-optimal approach can be widen to all other
aspects of occupant behaviour as well. Finally, some observations
are deduced from the results of the analyses in order to give useful
226 N.G. Saglam et al. / Energy and Buildings 150 (2017) 224238

Fig. 1. Flow chart of the approach.

guidelines to policy-makers of Turkey but also to other countries whole approach is given with Fig. 1. This section presents detailed
with similar features. denition of this approach through a sample application while the
results and sensitivity analyses are presented in the following sec-
tion.
2. Methodology

The comprehensive approach applied in this research is based
on adaptation of the EU cost-optimal methodology framework to
national conditions. Accordingly the approach involves reference
building denition, identication of energy efciency measures,
constitution of packages of measures, energy performance calcu-
lations and global cost calculations. A owchart representing the
2.1. The reference building
The reference building (RB) is a residential building that was
dened in a national research project [32] and represents the high-
rise apartments constructed between 1985 and 1999. It has an
unconditioned basement and twelve superior oors with four ats
 N.G. Saglam et al. / Energy and Buildings 150 (2017) 224238 227

Fig. 2. The oor plan and geometry of the RB.

Table 1
Thermal properties of the building envelope.

Layers Thickness (m) (W/mK) U-value (W/m2 K)

External Concrete External Plaster 0.005 1.4 1.04

Wall Reinforced Concrete 0.2 2.1
Herapor Thermal Insulation 0.035 0.051
Internal Plaster 0.005 0.7
External Concrete Reinforced Concrete 0.12 2.1 1.09
Panel Wall Herapor Thermal Insulation 0.035 0.051
Internal Plaster 0.005 0.7
Basement Ceiling Oak Parquet 0.016 0.2 1.25
Asphalt 0.005 0.7
Concrete Screed 0.03 1.4
Reinforced Concrete 0.16 2.1
Herapor Thermal Insulation 0.025 0.09
Attic slab Rockwool 0.05 0.045 0.71
Reinforced Concrete 0.16 2.1
Ceiling Plaster 0.01 0.87

in each as shown in Fig. 2. Total net oor area of the RB is 5186 m2 ,
total facade area is 3823 m2 with 590 m2 glazing area.
The construction type of the RB is reinforced concrete tunnel
form system and the facades were completed using concrete panel
walls. Layers, thermal conductivities () and calculated overall heat
transfer coefcients (U-value) of the building envelope compo-
nents are shown with Table 1.
The windows have double glazing with two layers of 4 mm at
glass and 12 mm air gap between them while material of the frame
is polyvinyl chloride (PVC). Overall heat transfer coefcient of the
window glazing (Uwindow ) is 2.9 W/m2 K, visible transmittance (Tvis )
is 0.80 and solar heat gain coefcient (SHGC) is 0.75.
User prole and schedules related to occupancy, activities,
equipment use and lighting system are previously dened in
the national research project as a part of the reference building
denition considering national and international standards and
statistical data gathered from Turkish Statistical Institute (TURK-
STAT) database [38]. These are explained below.
Since average household size is 3.8 people in Turkey [35] and
54% of the households consist of a couple with children, it is
assumed as in each apartment at, a family consists of parents
and two children [39]. Moreover, the occupancy schedules were
dened based on the ofcial surveys on Turkish family structure
[40,41]. Activity level values are gathered from ASHRAE-55 stan-
dard according to the related activity [42]. The occupancy and
activity levels considered for every single at are presented in
Table 2. In order to take heat gain from home appliances into con-
sideration, the power and operating time of these appliances are
assumed as in Table 3 [32].
Lighting power density (LPD) of spaces were calculated using
DIALux evo software [43]. Boundaries for illuminance levels are
200 lx for kitchen, 300 lx for children bedroom and 100 lx for liv-
ing room, bedroom, corridor and bathroom. The required luminous
ux is met by compact uorescent lamps and the obtained light-
ing power densities are provided with Table 4 together with the
operating time of the lighting system [44,45]
228 N.G. Saglam et al. / Energy and Buildings 150 (2017) 224238

Table 2
Occupancy and Activity Level Schedules.

Hours Number of Person Activity Activity Level (W/m2 ) Name of the Space

WEEKDAYS 00:00 07:00 4 Sleeping 40 Bedrooms

07:00 07:30 4 Breakfast 60 Kitchen
07:30 12:30 1 Housework 115 All spaces
12:30 15:30 1 Rest 45 Living Room
15:30 16:30 1 Housework 115 All spaces
16:30 19:00 3 1 person: Housework 2 115 All spaces
people: Rest 45
19:00 20:00 4 1 person: Housework 3 115 Kitchen
person: Light work 60 Living Room
20:00 20:30 4 Dinner 60 Kitchen
20:30 23:00 4 Reclining, Light work 60 Living Room, Bedrooms
23:00 24:00 4 Sleeping 40 Bedrooms
WEEKEND 00:00 00:30 4 Reclining, Light work 60 Living Room, Bedrooms
00:30 08:30 4 Sleeping 40 Bedrooms
08:30 12:30 4 Reclining, Light work 60 Living Room, Bedrooms
12:30 15:30 0
15:30 18:30 2 Reclining, Light work 60 Living Room, Bedrooms
18:30 22:30 3 Reclining, Light work 60 Living Room, Bedrooms
22:30 24:00 4 Rest 45 Living Room, Bedrooms

Table 3 The energy efciency measures were applied on the RB individu-

Power and Operating Time of the Electrical Equipment.
ally and also as packages of measures. As displayed in Fig. 1, in order
Home Appliance Power (W) Operating Time to constitute packages, thermal insulation and glazing retrots and
Refrigerator 38 All day (24 h) their combinations were analysed initially. According to the energy
Oven 2600 4 h in a week performance and cost calculation results of this rst set of analy-
Dishwasher 1030 5 h in a week ses, a number of scenarios were selected to be used in following
Washing Machine 851 4 h in a week extensive packages. As an example, decrease in the primary energy
Tea Kettle 1650 Weekdays: 3 h in a day
consumption provided by thermal insulation retrots applied on
Weekend: 2 h in a day
Iron 2300 2 h for 2 days in a week the attic slab (IN-R) and basement ceiling (IN-F) are between 0.52%
Vacuum Cleaner 2000 2 h for 2 days in a week while the change in the global cost is less than 1%. Therefore these
TV 105 Weekdays: 5 h in a day measures were not included in the following packages in all three
Weekends: 4 h in a day
climates. On the other hand, among these building envelope retrot
Notebook 120 3 h in a day
Stove 1800 2.5 h in a day scenarios, glazing retrots (GL) represent the cost-optimal range
Cooker hood 290 1.5 h in a day for Antalya by providing up to 13% primary energy saving and
9% global cost saving; that these are widely addressed in the fol-
lowing packages for Antalya. Besides the cost-optimal results, the
The heating energy demand of the RB is met by a central hot most energy efcient scenarios were also considered in subsequent
water boiler using natural gas. The nominal thermal efciency of packages in order to analyse the possible global cost decreases by
this natural gas boiler is 80%. In each at, there are radiators for combining them with low cost measures. Together with these enve-
emitting the heat generated by the boiler through circulating hot lope scenarios, 500 scenarios for Istanbul, 472 scenarios for Antalya
water. The cooling energy demand is met by individual split air con- and 332 scenarios for Erzurum were analysed in total.
ditioners using electricity. SEER value (seasonal energy efciency
ratio) of the split air conditioners are equal to 5.8 kWh/kWh. DHW
system is also individual and an electric water heater with 80% ther- 2.3. Energy performance calculations
mal efciency that exists in each at in order to provide hot water
for the occupants. The heating and cooling systems were assumed The geometric model for the energy calculations was consti-
as being operated continuously with setpoints 20 C for heating and tuted using Legacy Open Studio Plug-in for SketchUp 8 [48,49]. In
26 C for cooling [32]. The air change rate is assumed as 0.5 h1 in this model, thermal zones were also identied considering each
accordance with National Building Energy Performance Calculation at as a thermal zone and the common circulation areas as different
Methodology of Turkey [46]. thermal zones at each oor. This model was exported to EnergyPlus
(version 8.2) tool [36] to prepare a complete energy model with all
2.2. Energy efciency measures other inputs and to perform energy calculations through conduc-
tion transfer function algorithm. EnergyPlus is a modular building
In order to analyse on the RB, 37 different energy efciency mea- energy analysis and thermal load simulation program, developed
sures, which refer to thermal insulation, glazing, shading devices, by U.S. Department of Energy. It has been chosen for the aim of this
heating, cooling and DHW systems, lamps and renewable energy study for being an open-source free software, well-known, good
use considering solar thermal panels and photovoltaics, were iden- enough in terms of capabilities and widely used all over the world
tied. The energy efciency measures and their abbreviated names for building and HVAC system design and dynamic simulation [50].
are presented in Table 5. For each analysed scenario, a detailed sub-hourly simulation of the
Since the maximum limits of U-values in national heat insula- building was conducted. In the calculations, IWEC (International
tion standard TS 8252013 [47] differ based on climatic regions, the Weather for Energy Calculations) weather le was used for Istan-
analysed thermal insulation levels applied on the building envelope bul while the weather les for Antalya and Erzurum were derived
differ according to the analysed city as well. The maximum allowed by integrating national weather data representing typical mete-
limits in TS 8252013 and the calculated U-values corresponding orological year with meteonorm les since there is no available
to the retrot measures are provided in Table 6. international weather data for these cities [51].
 N.G. Saglam et al. / Energy and Buildings 150 (2017) 224238 229

Table 4
Lighting Power Densities.

Room Area (m2 ) LPD (W/m2 ) Operating Time

Bedroom 12.5 9.6 2 h/day

Bedroom 10.0 8.0
Children Bedroom 13.8 17.4 Manually controlled depending on the
Children Bedroom 14.0 20.0 illuminance level in occupied hours
Living Room 28.0 5.7
Kitchen 9.0 10.7 4 h/day
Bathroom 5.4 7.4 2 h/day
Bathroom 4.8 8.3
WC 2.1 10.0
Corridor 4.5 8.9
Entrance 8.0 10.0

Table 5
Energy efciency measures.

Name Denition of the measure

THERMAL INSULATION IN1-W External wall insulation with extruded polystyrene (xps) panels to meet the maximum
allowed U-values in national heat insulation standard (TS825-2013) [47].
IN2-W External wall insulation with xps panels to meet 25% lower U-values than the maximum limits
of TS 8252013.
IN3-W External wall insulation with xps panels to meet 50% lower U-values than the maximum limits
of TS 8252013.
IN4-W External wall insulation with xps panels to meet 75% lower U-values than the maximum limits
of TS 8252013.
IN1-R Attic slab insulation with rock wool panels to meet the maximum allowed U-values in
TS825-2013.
IN2-R Attic slab insulation with rock wool panels to meet 25% lower U-values than the maximum
limits of TS 8252013.
IN3-R Attic slab insulation with rock wool panels to meet 50% lower U-values than the maximum
limits of TS 8252013.
IN4-R Application of rock wool panels on attic slab to meet 75% lower U values than the maximum
limits of TS 8252013.
IN1-F Basement ceiling insulation with xps panels to meet the maximum allowed U-values in TS
8252013.
IN2-F Basement ceiling insulation with xps panels to meet 25% lower U-values than the maximum
limits of TS 8252013.
IN3-F Basement ceiling insulation with xps panels to meet 50% lower U-values than the maximum
limits of TS 8252013.
IN4-F Basement ceiling insulation with xps panels to meet 75% lower U-values than the maximum
limits of TS 8252013.
IN1-E Whole envelope (external walls, attic slab, basement ceiling) insulation to meet maximum
allowed U-values in TS 8252013.
IN2-E Whole envelope insulation to meet 25% lower U values than the maximum limits of
TS 8252013.
IN3-E Whole envelope insulation to meet 50% lower U values than the maximum limits of
TS 8252013.
IN4-E Whole envelope insulation to meet 75% lower U values than the maximum limits of
TS 8252013.
GLAZING GL1 Window glass replacement: U-value = 1.8 W/m2 K, Tvis = 0.79, SHGC = 0.56
GL2 Window glass replacement: U-value = 1.6 W/m2 K, Tvis = 0.79, SHGC = 0.56
GL3 Window glass replacement: U-value = 1.6 W/m2 K, Tvis = 0.71, SHGC = 0.44
GL4 Window glass replacement: U-value = 1.3 W/m2 K, Tvis = 0.71, SHGC = 0.44
GL5 Window glass replacement: U-value = 1.1 W/m2 K, Tvis = 0.71, SHGC = 0.44
GL6 Window glass replacement: U-value = 0.9 W/m2 K, Tvis = 0.69, SHGC = 0.48
GL7 Window glass replacement: U-value = 0.9 W/m2 K, Tvis = 0.63, SHGC = 0.39
SOLAR CONTROL SHD1 Fixed aluminium shading device installation: 60 cm width overhangs on south facade, 60 cm
width overhang and ns on east and west facades.
SHD2 Installation of external semi-transparent textile blinds with manual control:
Solar transmittance: 35%, solar reectance: 60%, visible transmittance: 35%,
visible reectance: 65%
HEATING & COOLING & DHW SYSTEMS BOI Central boiler replacement: new condensing boiler with 95% nominal thermal efciency.
RF Replacement of the existing heating system with radiators to radiant oor system
CHW Change of the individual domestic hot water systems to the central hot water system supplied
by the central boiler.
AC Upgrade SEER value of split type air conditioners to 8.5 kWh/kWh by replacement.
VRV Installation of a central variable refrigerant volume (VRV) system in substitution for split air
conditioners. Gross rated cooling coefcient of performance (COP) is equal to 3.1.
LIGHTING SYSTEM LED Installation of LED lamps to provide same illuminance levels. LPD after retrot are as
following: Bedroom-1: 5.3 W/m2 , Bedroom-2: 4.4 W/m2 , Children bedroom-1: 10.4 W/m2 ,
Children bedroom-2: 11 W/m2 , Living room: 3.1 W/m2 , Kitchen: 6.9 W/m2 , Bathroom-1:
4.1W/m2 , Bathroom-2: 4.6W/m2 , WC: 7.6W/m2 , Corridor: 4.9W/m2 , Entrance: 5.5W/m2 .
RENEWABLE ENERGY SP Installation of 48 solar thermal panel at roof with 120m2 total gross area.
USE PV Installation of photovoltaic system at roof with 11 kW rated power
230 N.G. Saglam et al. / Energy and Buildings 150 (2017) 224238
Calculated energy consumptions for space heating and cooling,
lighting and DHW were converted to primary energy using national
primary energy conversion factors which are 1 for natural gas and
2.36 for electricity in Turkey.
2.4. Global cost calculations
Global cost calculation are based on the Net Present Value
Method (NPV) in accordance with EN15459 standard [37]. This
cost estimation method considers time value of the money by
converting all future incomes and outcomes to a present value
using a discount rate. Since 67% of residential units in Turkey are
owner-occupied [39], the calculation method followed in this study
concerns micro-economic (individual end-user) perspective and
global cost was calculated based on the following equation:
 
 
Cg () = CI + (Ca,i (j) Rd (i)) Vf, (j)
j i=1
where,  is calculation period, Cg () is global cost over the calcula-
tion period, CI is initial investment cost for the measure, Ca,i (j) is
annual cost during year i for measure j, Vf, (j) is residual value at
the end of calculation period and Rd (i) is discount rate. The formula
to calculate discount rate is as below:
 1
p
Rd (p) =
1 + RR
where, p is number of years and RR is real discount rate. Real dis-
count rate was calculated using Eq. (3).
R Ri
RR =
1 + Ri
In the formula R is the market interest rate and Ri is the ination
rate.
The xed expenses and the costs related to the building ele-
ments that does not affect the energy performance of the building
were not included in the cost calculations [4].
Investment costs of energy efciency measures are based on
market prices and presented with Table 7 both in Euros (D ) and
Turkish Liras (TL). Since beginning year of the calculations is 2015,
the exchange rate is assumed as the average of all months in 2015
and therefore D /TL is equal to 3.02 [52].
Annual cost items considered in this research are replacement
cost, maintenance cost and energy cost. Lifespan of the building
components are assumed as 50 years for thermal insulation of
walls, 40 years for thermal insulation of attic slab and basement
ceiling, 30 years for windows and shading devices. Lifespan of the
energy system components are gathered from Annex A of EN 15459
standard [37]. Annual maintenance and operation costs of these
systems are also calculated based on this annex. As an exception
only for the rst set of calculations, aimed at selection of appropri-
ate envelope retrot scenarios, costs related to energy systems are
not included since those are the same for all scenarios.
In order to calculate real discount rate, economic indicators were
investigated using statistics provided by Central Bank of the Repub-
lic of Turkey (CBRT) and TURKSTAT [52,38] and average of the last
5 years before 2015 were used in the calculations. In this con-
text, ination rate is assumed as 8.054% and market interest rate is
assumed as 14.3%. Using Eq. (3), real discount rate was calculated
as 5.78%.
Energy costs were calculated based on the calculated energy
consumptions and average energy prices for 2015. Tax included
prices are 0.1213 D /kWh for electricity, 0.0370 D /kWh for natu-
ral gas in Istanbul, 0.0368 D /kWh for natural gas in Antalya and
0.03457 D /kWh for natural gas in Erzurum [5356].
The global cost calculation period is 30 years and residual value
at the end of the calculation period is also considered.
3. Results
Following the abovementioned method, initially nal use and
primary energy consumptions of the RB were calculated. Fig. 3
presents the nal energy uses subdivided by different energy car-
riers and primary energy consumption results for the RB.
Final energy use for space heating is extremely high in Erzu-
rum while energy use for space cooling is the highest in Antalya.
Moreover, primary energy consumption of the RB is the highest in
Erzurum and the lowest in Istanbul.
Since it is not feasible and practical to analyse all combination
possibilities of measures, proper building envelope retrot scenar-
ios were selected among thermal insulation and glazing retrots. In
(1) accordance with the results obtained from rst set of calculations,
appropriate building envelope retrot scenarios were selected to be
combined with other measures. Primary energy consumption and
global cost results for the initially selected scenarios are presented
with Table 8. In the table, RB represents the existing situation of
the reference building and the measures constituting a package are
separated by a plus (+) sign.
After the rst set of calculations, the selected building envelope
measures and packages were combined with other measures refer-
(2) ring to heating, cooling, DHW and lighting systems, shading devices
and renewable energy systems considering the climatic context.
The results are displayed in cost-optimal graphs in order to enable
a comparison between different scenarios.
In the cost-optimal graphs, each point represents a different
(3) scenario. Horizontal axis of the cost-optimal graph expresses the
primary energy consumption in kWh/m2 y and the vertical axis
expresses the global cost in D /m2 considering the total net oor
area of the RB. In case of renewable energy system inclusion in the
scenarios, primary energy consumption refers to the net primary
energy consumption left after subtraction of produced primary
energy. In order to emphasise signicant scenarios, horizontal and
vertical dashed lines were used to highlight exact numerical result
of primary energy consumption and global cost. Fig. 4 presents the
cost-optimal graph for the retrot scenarios analysed for Istanbul.
In Istanbul, the retrot package combining GL7, BOI, CHW, LED
and PV retrots represents the cost-optimal level. This cost-optimal
package achieves 65.5 kWh/m2 y saving that corresponds 45% of
RB primary energy consumption and 30.9 D /m2 saving that corre-
sponds 27% of global cost of the RB. The simple payback period of
this cost-optimal package is 5.1 years.
As seen from the graph, it is possible to decrease energy con-
sumption of the RB until 39.8 kWh/m2 y in a cost-effective way
by applying the energy efciency package involving GL, BOI, VRV,
CHW, LED, RF, SP and PV retrots. This decrease corresponds to 73%
primary energy saving in Istanbul while the global cost is almost
the same with the RB. In comparison to this cost-effective package,
affording 19.9 D /m2 higher global cost for applying also thermal
insulation retrot brings 10 kWh/m2 y additional primary energy
saving. Beyond this primary energy consumption level around
29.8 kWh/m2 y, primary energy saving amount corresponding to
the unit global cost increase is lower. In detail, 21.4 D /m2 increase
in the global cost results only with 1.2 kWh/m2 y additional primary
energy saving since the effect of the additional thermal insulation
thickness is limited while the investment cost is respectively high.
All of the energy efciency packages that involve at least two
of the highest level of thermal insulation (IN4), shading devices
(SHD2) or installation of air conditioners (AC) resulted with a global
cost higher than 160 D /m2 since these measures have a very high
initial investment cost.
 N.G. Saglam et al. / Energy and Buildings 150 (2017) 224238 231

Table 6
U-values of building components corresponding to the retrot measures.

U-values (W/m2 K)

Istanbul Antalya Erzurum

TS 825 Limits Uwall 0.57 0.66 0.36

Uroof 0.38 0.43 0.21
Ufloor 0.57 0.66 0.36
Uwindow 1.80 1.80 1.80
Values corresponding IN1-W (wall) 0.56 0.60 0.34
to the retrots IN1-R (roof) 0.36 0.39 0.21
IN1-F (oor) 0.56 0.66 0.35
IN2-W 0.42 0.48 0.26
IN2-R 0.27 0.32 0.15
IN2-F 0.42 0.48 0.25
IN3-W 0.29 0.31 0.18
IN3-R 0.17 0.18 0.11
IN3-F 0.29 0.29 0.18
IN4-W 0.14 0.16 0.12
IN4-R 0.10 0.11 0.08
IN4-F 0.14 0.17 0.09

Table 7
Investment cost of the energy efciency measures.

Total Investment Cost (TL) Cost per unit oor area Total Investment Cost Cost per unit oor area
2 2
(TL/m ) (D /m ) (TL/m2 ) (D /m2 )

IN1W Istanbul: 281934.9 54.4 18.0 IN1E Istanbul: 312574.1 60.3 20.0
Antalya: 275852.9 53.2 17.6 Antalya: 305621.7 58.9 19.5
Erzurum: 335659.9 64.7 21.4 Erzurum: 368893.5 71.1 23.6
IN2W Istanbul: 306924.3 59.2 19.6 IN2E Istanbul: 339041.5 65.4 21.6
Antalya: 294099.1 56.7 18.8 Antalya: 325229.9 62.7 20.8
Erzurum: 390574.9 75.3 24.9 Erzurum: 431115.0 83.1 27.5
IN3W Istanbul: 371976.1 71.7 23.8 IN3E Istanbul: 406960.7 78.5 26.0
Antalya: 351878.8 67.9 22.5 Antalya: 386264.0 74.5 24.7
Erzurum: 646988.0 124.8 41.3 Erzurum: 691483.5 133.3 44.2
IN4W Istanbul: 729666.9 140.7 46.6 IN4E Istanbul: 778070.5 150.0 49.7
Antalya: 671405.3 129.5 42.9 Antalya: 716500.2 138.2 45.7
Erzurum: 855883.8 165.0 54.6 Erzurum: 908020.9 175.1 58.0
IN1R Istanbul: 7259.4 1.4 0.46 GL1 51985.3 10.0 3.3
Antalya: 6678.8 1.3 0.43 GL2 53377.7 10.3 3.4
Erzurum: 8982.2 1.7 0.57 GL3 59177.0 11.4 3.8
IN2R Istanbul: 7854.1 1.5 0.50 GL4 60569.4 11.7 3.9
Antalya: 7259.4 1.4 0.46 GL5 65213.1 12.6 4.2
Erzurum: 15113.5 2.9 0.96 GL6 69153.5 13.3 4.4
IN3R Istanbul: 10152.7 2.0 0.65 GL7 74723.1 14.4 4.8
Antalya: 9553.3 1.8 0.61 SHD1 114903.7 22.2 7.3
Erzurum: 17364.9 3.3 1.11 SHD2 517108.6 99.7 33.0
IN4R Istanbul: 19134.9 3.7 1.22 BOI Istanbul: 43348.8 8.4 2.8
Antalya: 17364.9 3.3 1.11 Antalya: 43348.8 8.4 2.8
Erzurum: 20305.4 3.9 1.30 Erzurum: 46013.8 8.9 2.9
IN1F Istanbul: 23379.8 4.5 1.49 RF 367067.4 70.8 23.4
Antalya: 23090.0 4.5 1.47 CHW 22157.7 4.3 1.4
Erzurum: 24251.4 4.7 1.55 AC Istanbul: 311788.2 60.1 19.9
IN2F Istanbul: 24263.1 4.7 1.55 Antalya: 323294.3 62.3 20.6
Antalya: 23871.4 4.6 1.52 Erzurum: 302035.2 58.2 19.3
Erzurum: 25426.6 4.9 1.62 VRV Istanbul: 386300.5 74.5 24.7
IN3F Istanbul: 24831.9 4.8 1.59 Antalya: 451824.3 87.1 28.8
Antalya: 24831.9 4.8 1.59 LED 84052.4 16.2 5.4
Erzurum: 27130.6 5.2 1.73 SP 107941.8 20.8 6.9
IN4F Istanbul: 29268.7 5.6 1.87 PV 53744.7 10.4
Antalya: 27730.0 5.3 1.77
Erzurum: 31831.7 6.1 2.03

The package combining IN1-E and GL1 retrots which cor-
responds to the U-values stated in the national heat insulation
standard is not a cost-effective solution in Istanbul although it pro-
vides 27.4 kWh/m2 y decrease in primary energy consumption of
the RB. However, cost-effective results are achievable when the
packages involving thermal insulation and glazing retrots are
combined with other measures.
The cost-optimal graph for Antalya is given with Fig. 5. Energy
efciency packages combining GL, CHW, LED and PV retrots
achieve the cost-optimal results for Antalya. The exact optimal
result is provided by using GL7 retrot within this package and
it corresponds to 96.4 kWh/m2 y primary energy consumption
and 97.8 D /m2 global cost. The cost-optimal package ensures
64.5 kWh/m2 y energy saving which is equal to 40% of RB primary
energy consumption and 35 D /m2 cost saving that is 26% of total
global cost of the RB. The payback period of the cost-optimal pack-
age is 4.3 years in Antalya.
It is possible to provide 43.8 kWh/m2 y further energy saving
by affording 4.6 D /m2 higher global cost than the cost-optimal
package in order to apply also VRV and SP measures in this Mediter-
232 N.G. Saglam et al. / Energy and Buildings 150 (2017) 224238

Fig. 3. Energy consumption of the RB.

Table 8
Primary energy consumption and global cost results for selected envelope retrot scenarios.

ISTANBUL

Scenario Primary Energy Consumption Global Cost Scenario Primary Energy Consumption Global Cost
(kWh/m2 y) (D /m2 ) (kWh/m2 y) (D /m2 )

RB 145.3 97.2 IN4-W + GL5 111.0 115.7

GL5 132.3 92.0 IN4-W + GL7 109.4 115.9
GL6 132.3 93.2 IN2-E + GL7 112.7 103.9
GL7 131.0 92.3 IN3-E + GL4 111.6 106.3
IN2-W + GL7 114.9 100.6 IN3-E + GL5 111.2 106.5
IN3-W + GL4 114.0 101.6 IN3-E + GL7 109.7 106.1
IN3-W + GL5 113.5 101.9 IN4-E + GL4 109.0 126.8
IN3-W + GL7 112.0 101.9 IN4-E + GL5 108.7 115.5
IN4-W + GL4 111.4 115.6 IN4-E + GL7 107.1 127,0

ANTALYA

Scenario Primary Energy Consumption Global Cost Scenario Primary Energy Consumption Global Cost
(kWh/m2 y) (D /m2 ) (kWh/m2 y) (D /m2 )

RB 160.9 114.9 IN3-W + GL5 128.7 118.0

GL3 142.0 104.7 IN3-W + GL7 126.6 117.2
GL4 141.5 104.5 IN4-W + GL7 125.3 135.3
GL5 141.4 104.8 IN2-E + GL7 126.8 116.0
GL6 143.5 106.9 IN3-E + GL3 126.0 117.9
GL7 139.8 104.3 IN3-E + GL4 126.1 118.1
IN2-W + GL7 128.8 115.0 IN3-E + GL5 126.4 118.7
IN3-W + GL3 128.5 117.4 IN3-E + GL7 124.3 118.0
IN3-W + GL4 128.4 117.5 IN4-E + GL7 123.2 136.6

ERZURUM

Scenario Primary Energy Consumption Global Cost Scenario Primary Energy Consumption Global Cost
(kWh/m2 y) (D /m2 ) (kWh/m2 y) (D /m2 )

RB 216.3 127.8 IN4-W + GL7 141.9 146.3

GL6 200.5 123.5 IN1-E + GL6 149.2 121.1
GL7 202.1 124.3 IN2-E + GL6 143.5 122.2
IN1-W + GL6 155.6 122.1 IN2-E + GL7 143.8 122.1
IN2-W + GL6 150.7 123.0 IN3-E + GL5 140.0 135.3
IN2-W + GL7 151.3 123.1 IN3-E + GL6 137.8 135.0
IN3-W + GL6 145.7 135.9 IN3-E + GL7 137.9 134.8
IN3-W + GL7 145.9 135.8 IN4-E + GL5 135.5 146.1
IN4-W + GL5 143.9 146.8 IN4-E + GL6 133.5 146.0
IN4-W + GL6 141.6 146.4 IN4-E + GL7 133.3 145.6

ranean climate. The package, that involves GL7, VRV, CHW, LED,
SP and PV retrot measures results with 52.6 kWh/m2 y primary
energy consumption and 102.4 D /m2 global cost. Investing also
on the thermal insulation together with this package is still cost-
effective until the third level of insulation (IN3). However, when the
heat insulation level applied with this package is increased further,
the result is not cost-effective. The package that includes IN3-E, GL7,
VRV, CHW, LED, SP and PV retrots results with 39.7 kWh/m2 y pri-
mary energy consumption and 118 D /m2 global cost. This result
shows that it is possible to achieve 75% primary energy saving in a
cost-effective way in Antalya. After this point, global cost shows a
rapid increase and only 3.5 kWh/m2 y decrease in energy consump-
tion is achieved with 50.3 D /m2 increase in global cost (IN4-E, GL7,
VRV, CHW, LED, SP, PV, SHD2).
 N.G. Saglam et al. / Energy and Buildings 150 (2017) 224238
Fig. 4. Cost-optimal graph of retrot scenarios analysed for Istanbul.
Similar to Istanbul, energy efciency packages which resulted
with a global cost higher than 170 D /m2 in Antalya involve at least
two of the IN4, AC and SHD2 measures.
IN1-E+GL1 retrot package is not cost-effective in Antalya
as well. Nevertheless, there are cost-effective building envelope
retrot scenarios such as IN2-W+GL7 package even though the ini-
tial investment cost is higher. Moreover, it is possible to decrease
the global cost more by combining the envelope retrot scenarios
with other measures for energy systems and renewable energy use.
The cost-optimal graph for Erzurum is presented with Fig. 6. As
seen from this graph, the energy efciency package, which consists
of GL6, BOI, CHW, LED and PV retrots, achieves the cost-optimal
result in Erzurum. This package results with 132.1 kWh/m2 y pri-
mary energy consumption that leads to 39% energy saving and
103.8 D /m2 saving which is equal to 28% of RB global cost. Payback
period of this cost-optimal package is 4.2 years in Erzurum.
In comparison with the cost-optimal package, 43.5 kWh/m2 y
further decrease in primary energy consumption is achievable only
with 1 D /m2 increase in the global cost by combining the cost-
optimal package with the thermal insulation retrot IN1-E. Instead,
only around 3 D /m2 rise in the global cost value enables to decrease
the primary energy consumption of a 28%, reaching 71.5 kWh/m2 y
(IN2-E, GL6, BOI, CHW, LED, SP, PV). Despite the global cost point
of view, when the investment costs are compared, the variation
between this scenario and the cost optimal scenario is equal to
102.7 D /m2 and consequently the payback period is the double of
the cost optimal solution. A further action shows that application
of RF measure together with this package decreases the primary
energy consumption of the RB until 53.3 kWh/m2 y and increases
the global cost up to 119.2 D /m2 .
The global cost level of the RB is associated with very different
energy performance levels up to 47.7 kWh/m2 y which is achieved
by applying IN4-E, GL7, BOI, CHW, LED, RF, SP, and PV retrot mea-
233
sures. This level corresponds to 78% primary energy saving with
168.6 kWh/m2 y decrease in energy consumption in Erzurum.
Contrary to Istanbul and Antalya, in Erzurum IN1-E+GL1 pack-
age which represents the national standard is cost-effective. This
package achieves 58 kWh/m2 y primary energy saving (27%) with
2% decrease in the global cost. Even though, the energy performance
level obtained through the national requirements corresponds to
158.3 kWh/m2 y primary energy consumption which is not poten-
tially reliable for future targets when it is compared with more
effective retrot scenarios.
3.1. Sensitivity analyses: the effect of occupant behaviour on
window openings
After the cost-optimal graphs were obtained, sensitivity anal-
yses were performed in order to examine effect of occupant
behaviour relevant to use of window openings. The scenarios on
the boundary of cost-optimal curve were selected for these analy-
ses. Besides the existing assumption of the RB description, which
refers to air change rate of 0.5 h1 , greater use of the window open-
ings in addition to this air change rate were also analysed. In the
analyses, it is considered that RB occupants control up to 3.6 m2
opening area and leave windows open while the outdoor tempera-
ture is between 21 C and 26 C. This alternate occupant behaviour
is regarded as a measure and abbreviated as OB.
Results of the sensitivity analyses considering the effect of occu-
pant behaviour for Istanbul are presented in Fig. 7. As shown in
this gure, proper use of window openings decreases the pri-
mary energy consumption of the cost-optimal package around
20.2 kWh/m2 y and decreases the global cost of this package around
14.7 D /m2 in Istanbul. Accordingly, the cost-optimal package
achieves 59.6 kWh/m2 y primary energy consumption level and
69.1 D /m2 global cost.
234 N.G. Saglam et al. / Energy and Buildings 150 (2017) 224238

Fig. 5. Cost-optimal graph of retrot scenarios analysed for Antalya.

Fig. 6. Cost-optimal graph of retrot scenarios analysed for Erzurum.

 N.G. Saglam et al. / Energy and Buildings 150 (2017) 224238
Fig. 7. Cost-optimal graph of the sensitivity analyses for RB retrots in Istanbul.
In Istanbul, the occupant behaviour affects results of all other
packages in a positive way as well. The package with IN3, GL7, BOI,
CHW, LED, SP and PV retrot measures results with 28.7 kWh/m2 y
primary energy consumption and 84.3 D /m2 global cost when OB is
applied in Istanbul. In this case, the package achieves 80% primary
energy saving in a cost-effective way. The energy efciency package
that combines IN4, GL, BOI, VRV, CHW, LED, RF, SP and PV retrots
achieves 17.0 kWh/m2 y primary energy consumption level when
the building occupants operates the windows as considered. This
package is able to achieve primary energy savings up to 88% in Istan-
bul and to save 128.3 kWh/m2 y primary energy with an appropriate
use of window openings, however it is still not cost-effective.
Results of the sensitivity analyses examining OB in Antalya are
displayed with Fig. 8. Conscious use of window openings changes
the cost-optimal result to 74.3 kWh/m2 y primary energy consump-
tion level and 81.8 D /m2 global cost level in Antalya. This scenario
saves 22.1 kWh/m2 y more primary energy and 16.0 D /m2 more
global cost in comparison to the cost-optimal scenario without
considering OB. Moreover, the primary energy consumption level
achieved by the package involving IN4-E, GL7, VRV, CHW, LED, SP
and PV retrots, is moved from 38.4 kWh/m2 y to 26.9 kWh/m2 y
while the global cost is also decreased around 8.4 D /m2 . This result
corresponds to 83% energy saving with 3% decrease in global cost
comparing to the RB without OB.
Fig. 9 shows results of the sensitivity analyses related to OB for
Erzurum. In Erzurum, energy savings related to OB is lower in com-
parison to Istanbul and Antalya since this city has a cold climate.
However, this occupant behaviour changes the cost-optimal sce-
nario in Erzurum since the global cost variation between the most
cost-effective packages is minor. When OB is considered, the pack-
age that combines IN1-E, GL6, BOI, CHW, LED, SP and PV measures
achieves 68.8 kWh/m2 -a primary energy consumption and 100.0
235
D /m2 global cost and appears as the cost-optimal solution for the
RB in Erzurum.
The package including IN4-E, GL7, BOI, CHW, LED, RF, SP, and PV
retrots, which was previously resulted with a global cost slightly
higher than RB, results in a cost-effective way when OB is con-
sidered. This package results with 39.1 kWh/m2 y primary energy
consumption and 138.7 D /m2 global cost; accordingly it achieves
82% primary energy saving.
4. Discussion
Results of this study reveal that cost-effective energy saving
potential for the existing high-rise apartment buildings is higher
than 70% in all analysed climatic regions. Moreover, occupant
behaviour related to conscious use of window openings is able to
raise this rate above 80%.
The obtained results verify the dependence of cost-optimal lev-
els to the climate. For the same RB, the cost-optimal scenarios range
between 79.8 and 132.1 kWh/m2 y (in terms of primary energy con-
sumption) depending on the climate while the reference occupant
behaviour is considered. This variation of primary energy consump-
tion corresponds to a 20 D /m2 alteration in global cost. On the
other hand, these variations refer to similar improvement rates
which are between 39% (Erzurum) and 45% (Istanbul) in terms of
primary energy consumption and between 26% (Antalya) and 28%
(Erzurum) in terms of global cost.
Results also display that, for all climatic zones, envelope retrots
have limited effect on the energy performance of the high-rise
apartment buildings. The highest primary energy saving ratio
achieved by the envelope retrots is only between 23% (Antalya)
and 38% (Erzurum) while cost-effectiveness is not assured in
every case. Moreover, combination of the envelope improvements
and other retrot actions related to building service systems and
236 N.G. Saglam et al. / Energy and Buildings 150 (2017) 224238

Fig. 8. Cost-optimal graph of the sensitivity analyses for RB retrots in Antalya.

Fig. 9. Cost-optimal graph of the sensitivity analyses for RB retrots in Erzurum.

 N.G. Saglam et al. / Energy and Buildings 150 (2017) 224238
renewable energy use is able to provide global cost decrease. It is
obvious that the comprehensive approach is not only needed to
achieve higher energy performance but also to ensure global cost
decrease.
On the other hand, the comparison between the results of build-
ing envelope retrot scenarios obtained in this research and the
previous study in [34] shows that a number of building envelope
retrots which are cost-effective for low-rise residential buildings
are not cost-effective for high-rise residential buildings while the
cost-optimal solutions are also different than each other. Therefore,
it can be clearly stated that besides the different building cate-
gories, different sub-categories also require particular attention in
cost-optimal analyses obliged by the EU Directive.
EPBD recast and related legislation specify that national mini-
mum energy performance requirements are needed to be revised
if these are less ambitious than the calculated cost-optimal levels
[3,6]. Calculation results for IN1-E+GL1 package, which represents
the thermal insulation level indicated in the national standard
TS825-2013 [47], show that, although this solution provides energy
efciency for the high-rise reference apartment building in all cli-
mates, it is not cost-effective for the RB in mild and hot-humid
climates. This package is slightly cost-effective in Erzurum which
represents cold climatic region. Nevertheless, the results display
that the national standard is less ambitious than the cost-optimal
levels in all three climates. In spite of these ndings, the national
standards are required to be likewise analysed for new construc-
tions since validity of these results is limited to existing building
renovations.
Another important result to consider in energy policy is the
presence of common measures that are involved in the cost-
optimal packages for all three climates. These measures, which are
GL (window glass replacement), LED (installation of LED lamps),
CHW (centralization of DHW production) and PV (installation of
photovoltaic system), are required to be promoted for building
renovation activities.
Although some measures may not be applicable for all buildings
that are represented by the analysed RB in practice, it is impor-
tant to consider them in cost-optimal calculations to consolidate
cost-effective energy efciency. For instance, even though CHW
(centralization of DHW production) or RF (transition to radiant oor
system) require certain conditions to implement, the results show
that these measures provide higher energy efciency in a cost-
effective way especially in cold climatic region. These results may
also refer to the measures to be involved in cost-optimal analyses
for new buildings.
An important consequence of the results is the effectiveness of
conscious occupant behaviour related to use of window openings
on the cost-optimal results. This alternate occupant behaviour (OB)
is able to slide the cost-optimal levels until the range between
59.6 kWh/m2 y (Istanbul) and 74.3 kWh/m2 y (Antalya) of primary
energy consumption. OB provides more than 20 kWh/m2 y addi-
tional primary energy saving and more than 14 D /m2 global cost
saving for the cost-optimal scenarios in mild and hot-humid cli-
mates. The effectiveness of OB in the cold climate is not as high as
in the other two climates. However, it is a signicant indicative on
the cost-optimal scenario in cold climate since the calculated global
costs of different energy efciency packages are not signicant.
This result shows that, OB is not only determinative on the cost-
optimal energy performance level but also on the identication of
the cost-optimal scenario.
5. Conclusion and future studies
This research proposes a comprehensive cost-optimal approach
for existing building retrots in order to maximise energy perfor-
237
mance level for multi-family buildings. The approach incorporates
building envelope improvements, energy system retrots and
renewable energy use all at once. Moreover, this comprehen-
sive approach was upgraded by integration of occupant behaviour
effect into the cost-optimal calculations. In particular, the proposed
approach was applied to a reference high-rise apartment block con-
sidering three different representative climates of Turkey which
accommodate a large population.
Application of the proposed approach demonstrates a signi-
cant cost-effective energy saving potential for the analysed RB. This
potential afrms a remarkable energy and cost saving when the
existing apartment stock is considered. In this respect, obtained
results are important for implementing Article 4 of Directive
2012/27/EU, which obliges to prepare future strategies including
cost-effective deep retrot approach for existing buildings [8].
In particular, the ndings of the research reveal that national
standard and regulations are required to be assessed in terms of
cost-optimality since these are not favourable for all cases as in
high-rise apartments in Turkey. The assessment is needed to be
done by extending the comprehensive approach presented in this
study to cover all building types. Subsequent to this assessment and
related modications, the further focusing point is to move towards
NZEB. This study also notify about the path towards NZEB by dis-
playing the scenarios that are resulted with more ambitious energy
performance level in comparison to the cost-optimal scenario.
The present study refers to the cost-effective retrot solutions
for multi-family buildings considering various climatic regions.
The sample application for the hot-humid climate (Antalya) rep-
resents a comprehensive Mediterranean cost-optimal approach
that requires particular attention to keep cooling energy consump-
tion under control. On the contrary, a comprehensive cost-optimal
approach for the cold climatic region is also displayed through
the analyses for Erzurum. Correspondingly, climate-specic retrot
strategies are demonstrated such as central cooling systems for
Mediterranean climate and radiant oor heating for the cold cli-
mate. Findings for these strategies guide also to the following
analyses on possible government support for cost-effective reno-
vation in different climates.
The research also signies that integration of occupant
behaviour into cost-optimal calculations is fundamental for further
studies since it is a determinative factor for the cost-optimal level.
In order to demonstrate the integration approach under sensitivity
analyses, this study addressed the occupant behaviour related to
the use of window openings. In accordance with the results, it can
be stated that the inuence of the occupant behaviour also varies
based on the climate. For instance, the highest variation for the cost-
optimal level is observed in the hot-humid climate when conscious
occupant behaviour is considered. The further studies following the
presented approach need to include also other occupant behaviour
congurations such as activities, equipment use, heating and cool-
ing system operations (temperatures, scheduling, etc.). In order to
have an absolute perspective, comfort analyses would be required
in these further research activities as well.
Besides revealing the essentiality of a comprehensive cost-
optimal approach for the existing building retrots, the study
indicates importance of integrating the occupant behaviour in the
approach. Considering the signicant effect of the occupant, build-
ing users trainings may also be considered as a tool within national
plans towards future targets.
Acknowledgement
This work was supported by The Scientic and Technological
Research Council of Turkey (TUBITAK) with a PhD researcher grant.
238 N.G. Saglam et al. / Energy and Buildings 150 (2017) 224238
References
[1] Transition to Sustainable Buildings, Strategies and Opportunitiesto 2050,
International Energy Agency, 2013 (Cedex).
[2] Communication From The Commission to the European Parliament, the
Council, the European Economic and Social Committee and The Committee of
the Regions, Energy Efciency Plan 2011, COM(2011) 109 nal, Brussels, 2011.
[3] EPBD recast, Directive 2010/31/EU of the European Parliament and of Council
of 19 May 2010 on the energy performance of buildings (recast), J. Eur. Union
L153/13-35 (2010).
[4] Commission Delegated Regulation (EU) No 244/2012 of 16 January 2012
supplementing Directive 2010/31/EU of the European Parliament and of the
Council on the energy performance of buildings by establishing a comparative
methodology framework for calculating cost-optimal levels of minimum
energy performance requirements for buildings and building elements, J. Eur.
Union L81/18-36 (2012).
[5] Guidelines Accompanying on Supplementing Directive 2010/31/EU, of the
European Parliament and of the Council on energy performance of buildings
(recast) by establishing a comparative methodology framework for
calculating cost-optimal levels of minimum energy performance
requirements for buildings and building element, J. Eur. Union C115/1-28
(2012).
[6] Commission Recommendation (EU) 2016/1318 of 29 July 2016 on guidelines
for the promotion of nearly zero-energy buildings and best practices to
ensure that, by 2020, all new buildings are nearly zero-energy buildings,
Brussels, 2016.
[7] BPIE (Building Performance Institute Europe) (2011). Europes building under
the microscope. Brussels.
[8] Directive 2012/27/EU of The European Parliament and of the Council of 25
October 2012 on energy efciency, amending Directives 2009/125/EC and
2010/30/EU and repealing Directives 2004/8/EC and 2006/32/EC, Ofcial
Journal of the European Union, 2012.
[9] Financing the energy renovation of buildings with Cohesion Policy funding,
Final Report, European Commission 2014. ISBN:978-92-79-35999-6.
[10] N.W.O. Brown, T. Malmqvist, W. Bai, M. Molinari, Sustainability assessment of
renovation packages for increased energy efciency for multi-family
buildings in Sweden, Build. Environ. 61 (2013) 140148.
[11] L. Liu, B. Moshfegh, J. Akander, M. Cehlin, Comprehensive investigation on
energy retrots in eleven multi-family buildings in Sweden, Energy Build. 84
(2014) 704715.
[12] F. Bonakdar, A. Dodoo, L. Gustavsson, Cost-optimum analysis of building
fabric renovation in a Swedish multi-story residential building, Energy Build.
84 (2014) 662673.
[13] E. Arumgi, T. Kalamees, Analysis of energy economic renovation for historic
wooden apartment buildings in cold climates, Appl. Energ. 115 (2014)
540548.
[14] K. Kuusk, T. Kalamees, M. Maivel, Cost-effectiveness of energy performance
improvements in Estonian brick apartment buildings, Energy Build. 77 (2014)
313322.
[15] J. Kurnitski, K. Kuusk, T. Tarka Aivar Uutar, E. Kalamees, Energy and
investment intensity of integrated renovation and 2030 cost-optimal savings,
Energy Build. 75 (2014) 5159.
[16] E. Pikas, J. Kurnitski, R. Liias, M. Thalfeldt, Quantication of economic benets
of renovation of apartment buildings as a basis for cost-optimal 2030 energy
efciency strategies, Energy Build. 86 (2015) 151160.
[17] EPBD cost-optimal methodology: application to the thermal rehabilitation of
the building envelope of a Portuguese residential reference building. Ana
Brando de Vasconcelos, Manuel Duarte Pinheiro, Armando Manso, Antnio
Cabaco, Energy Build. 111 (2016) 1225.
[18] A. Brando de Vasconcelos, M.D. Pinheiro, A. Manso, A. Cabaco, Portuguese
approach to dene reference buildings for cost-optimal methodologies, Appl.
Energ. 140 (2015) 316328.
[19] A. Brandode Vasconcelos, A. Cabaco, M.D. Pinheiro, A. Manso, The impact of
building orientation and discount rates on a Portuguese reference building
refurbishment decision, Energy Policy 91 (2016) 329340.
[20] V. Corrado, I. Ballarini, S. Paduos, Assessment of cost-optimal energy
performance requirements for the Italian residential building stock, Energy
Procedia 45 (2014) 443452.
[21] K. Fabbri, L. Tronchin, V. Tarabusi, Energy retrot and economic evaluation
priorities applied at an Italian case Study, Energy Procedia 45 (2014) 379384.
[22] G. Desogus, L. Di Pilla, S. Mura, G.L. Pisano, R. Ricciu, Economic efciency of
social housing thermal upgrade in Mediterranean climate, Energy Build. 57
(2013) 354360.
[23] C. Becchio, M.C. Bottero, S.P. Corgnati, C. Ghiglione, nZEB design: challenging
between energy and economic targets, Energy Procedia 78 (2015) 20702075.
[24] O. Pombo, B. Rivela, J. Neila, The challenge of sustainable building renovation:
assessment of current criteria and future outlook, J. Clean. Prod. 123 (2016)
88100.
[25] V.M. Barthelmes, C. Becchio, S.P. Corgnati, Occupant Behavior Lifestyles in a
residential nearly zero energy building: effect on energy use and thermal
comfort, Sci. Technol. Built Environ. 22 (2016) 960975.
[26] C. Becchio, S.P. Corgnati, C. Delmastro, V. Fabi, P. Lombardi, The role of
nearly-zero energy buildings in the transition towards Post-Carbon Cities,
Sust. Cities Soc. 27 (2016) 324337.
[27] Tapping the Potential for Energy Savings in Turkey, Sustainable Development
Department (ECSSD) Europe and Central Asia Region (ECA) Document of the
World Bank, 2010.
[28] Energy Efciency Strategy Paper 20122023, Republic of Turkey Ofcial
Gazette, 28215, Ankara, 2012.
[29] Binalarda Enerji Performans Ynetmeligi [Energy Performance of Buildings
Regulation], Ministry of Public Works and Settlements, Republic of Turkey
Ofcial Gazette, 27075, Ankara, 2008 (in Turkish).
[30] Enerji Verimliligi Kanunu [Energy Efciency Law], Law Number: 5627,
18.04.2007. Republic of Turkey Ofcial Gazette, 26510, Ankara, 2007 [in
Turkish].
[31] Building Census 2000, State Institute of Statistics Prime Ministry Republic of
Turkey, Ankara, 2001. ISBN 975 19 2819-2.
[32] A. Z. Yilmaz, N. Ganic Saglam, G., Gali, T., Ashraan, A., Akguc, Determination
of Turkish Reference Residential Buildings and National Method for Dening
Cost Optimum Energy Efciency Level of Buildings, Scientic and
Technological Research Council of Turkey (TUBITAK), 2015, Project
no:113M596.
[33] N. Ganic Saglam, A. Z. Ylmaz, G., Gali, T. Ashraan Bonab, A. Akgc, Reference
Building Establishment Procedure for the Residential Buildings in Turkey,
Heiselberg, P. K. (Ed.) (2016). CLIMA 2016 proceedings of the 12th REHVA
World Congress: volume 6. Aalborg: Aalborg University, Department of Civil
Engineering.
[34] T. Ashraan, A.Z. Yilmaz, S.P. Corgnati, N. Moazzen, Methodology to dene
cost-optimal level of architectural measures for energy efcient retrots of
existing detached residential buildings in Turkey, Energy Build. 120 (2016)
5877.
[35] Population and Housing Census, 2011, Turkish Statistical Institute, No: 15843,
2013.
[36] EnergyPlus Energy Simulation Software Web Site. Available online:
http://apps1. eere.energy.gov (data retrievedon November 2014).
[37] CEN. [European Committee for Standardization]. Energy performance of
buildings Economic evaluation procedure for energy systems in buildings.
Standard EN 15459:2007, Brussels : CEN; 2007.
[38] http://www.turkstat.gov.tr (accessed on November 2015).
[39] Income and Living Conditions Survey 2012, Turkish Statistical Institute
(TURKSTAT), Publication Number 4143, 2014.
[40] Family Structure Survey 2011, Turkish Republic Ministry of Family and Social
Policies, Ankara, 2011.
[41] Family Structure Survey 2013, Turkish Republic Ministry of Family and Social
Policies, Ankara, 2013.
[42] ASHRAE STANDARD 552010, Thermal Environmental Conditions for Human
Occupancy, ISSN 10412336, American Society of Heating, Refrigerating and
Air-Conditioning Engineers, Inc., 2010.
[43] https://www.dial.de (data retrieved on February 2016).
[44] . Smengen, A. Kknel Yener, Konutlarda Aydnlatma Enerjisi Performans
ve Grsel Konfor Kosullar [Lighting Energy Performance and Visual Comfort
Conditions in Residential Buildings], 7. Ulusal Aydnlatma Sempozyumu,
2013, Izmir, [in Turkish].
[45] IESNA/Illuminating Engineering SocietyThe Lighting Handbook Reference
and Application, in: D.L. DiLaura, K.W. Houser, R.G. Mistrick, G.R. Steffy, (Eds.),
10th edition, 2011.
[46] Bina Enerji Performans Hesaplama Yntemi, Bina Enerji Performans Istma
ve Sogutma icin Net Enerji Ihtiyacnn Hesaplanmas [Building Energy
Performance Calculation Method, Building Energy Performance Calculation
of net energy demand for heating and cooling], Republic of Turkey Ofcial
Gazette, 27778, Ankara, 2010.
[47] TSE. [Turkish Standards Institution]. Binalarda Is Yaltm Kurallar [Thermal
Insulation Requirements for Buildings], TS 825, ICS 91.120.10 Ankara, TSE,
2013 [in Turkish].
[48] www.sketchup.com (data retrieved on November 2014).
[49] www.openstudio.net (data retrieved on Novemner2014).
[50] D.B. Crawley, J.W. Hand, M. Kummert, B.T. Grifth, Contrasting the
capabilities of building energy performance simulation programs, Build.
Environ. 43 (2008) 661673.
[51] www.meteonorm.com (data retrieved on May 2014).
[52] http://www.tcmb.gov.tr (data retrieved on February 2016).
[53] http://www.tedas.gov.tr (data retrieved on February 2016).
[54] http://www.bedas.com.tr (data retrieved on February 2016).
[55] http://www.palen.com.tr (data retrieved on February 2016).
[56] http://www.enerya.com.tr (data retrieved on February 2016).