Decision-making o eno a ing he Medi e anean social housing: A
p ac ical app oach h ough an in e ac i e open access ool
C.M. Calama-Gonz´
alez
a,*
, R. Escand´
on
b
, R. Su´
a ez
b
, F. Ascione
c
a
Depa amen o de Cons ucciones A qui ec ´
onicas y su Con ol, Escuela T´
ecnica Supe io de Edi icaci´
on, Uni e sidad Poli ´
ecnica de Mad id, A da. Juan de He e a 4,
28040 Mad id, Spain
b
Ins i u o Uni e si a io de A qui ec u a y Ciencias de la Cons ucci´
on, Escuela T´
ecnica Supe io de A qui ec u a, Uni e sidad de Se illa, A . de Reina Me cedes 2, 41012
Se ille, Spain
c
Depa men o Indus ial Enginee ing, Uni e si `
a degli S udi di Napoli Fede ico II, Piazzale Tecchio 80, 80125 Napoli, I aly
ARTICLE INFO
Keywo ds:
Pa ame ic building s ock modelling
Mul i-objec i e op imiza ion
In es men cos s
Adap i e he mal com o
Clus e ing e o i s a egies
Deep s pa ial eno a ion
ABSTRACT
To achie e 2050 Clima e Neu ali y, building s ock equi es a mul idimensional eno a ion p ocess. This is
pa icula ly u gen in mos ulne able households, wi h highe exposu e o clima e change, whe e his p ocedu e
should ocus on cos -con olled passi e measu es. Gi en he complexi y o iden i ying op imal s a egies, i is
impe a i e o imp o e he e o i ing p ocess o he social housing s ock o enhance i s ene gy pe o mance
gua an eeing heal h and com o . Fo his, an in e ac i e ool was de eloped ocused on he case o sou he n
Spain. Able o p o ide op imized combina ions o ene gy e o i s a egies, using NSGA-II gene ic algo i hms
and se ing wo op imiza ion objec i es: minimizing he mal discom o and economic cos s. The eely acces-
sible ool was designed wi h p ac ical and didac ic app oach o acili a e decision-making. The esul s ob ained
sugges he easibili y o implemen ing phase ac ions ins ead o a single la ge-scale in e en ion and show he
ool’s abili y o quan i y he pe cen age o he mal com o imp o emen achie es a each phase.
1. In oduc ion
The phenomenon o clima e change, and he new eali y i ep e-
sen s, is ans o ming ou way o li e and p ecipi a ing a me amo phosis
a ene gy, poli ical, and social le els [1]. In his con ex , a a ie y o
p ocesses and ene gy policies a e pu o wa d wi h he objec i e o
acili a ing he ecological ansi ion and ul ima ely achie ing clima e
neu ali y by 2050. The Eu opean G een Deal [2] unde sco es he
impe a i e o long- e m esiden ial eno a ion as a means o enhancing
ene gy e iciency in he building sec o . In he wake o he global
pandemic, he Eu opean Commission ad anced a comp ehensi e und-
ing plan o d i e eno a ion ini ia i es. Consequen ly, he s a egy,
en i led “A Reno a ion Wa e o Eu ope – G eening ou buildings,
c ea ing jobs, imp o ing li es” [3] was de ised wi h he objec i e o
doubling he eno a ion a e by 2030 while also add essing he issue o
ene gy po e y.
The no el ou h e sion o he Ene gy Pe o mance o Buildings
Di ec i e (EPBD) (EU) 2024/1275 [4] p o ides an oppo uni y o
add ess he le el and pace o eno a ion ac ions. This di ec i e en-
compasses he implemen a ion o measu es such as he Building
Reno a ion Passpo (BRP), which supplies a pe sonalized oadmap o
indi idual buildings, enabling emission educ ion a ge s o be me .
Fu he mo e, he inco po a ion o Building Digi al Logbooks (DBLs) is
expec ed o add ess ene gy po e y and acili a e decision-making p o-
cesses o e o i . Despi e he absence o a uni ied me hodological
app oach o he de ini ion o long- e m eno a ion s a egies, indi id-
ual membe s a es a e equi ed o adap his app oach in line wi h hei
speci ic condi ions. Ne e heless, he app oach o in oducing he
mul idimensional concep o deep eno a ion (DR) [5] has been widely
adop ed o achie e clima e neu ali y.
The de ini ions and implica ions o deep eno a ion (DR) a e b oad,
gi en he comp ehensi e eno a ion equi ed o his building s ock.
The p ocess calls o he implemen a ion o a ange o simul aneous
passi e and ac i e measu es, along wi h no able enhancemen s in pe -
o mance and educ ions in ene gy consump ion when compa ed o he
p e- eno a ion s a e [6]. Di e en membe s a es ha e been inco po-
a ing he concep o DR in o long- e m eno a ion s a egies (LTRS)
using a ious c i e ia, and wi h a p ima y ocus on buildings. This is
p ima ily made possible by a educ ion in p ima y ene gy consump ion,
exp essed as a pe cen age o ene gy sa ings. Howe e , ene gy e iciency
* Co esponding au ho .
E-mail add ess: [email p o ec ed] (C.M. Calama-Gonz´
alez).
Con en s lis s a ailable a ScienceDi ec
Ene gy & Buildings
jou nal homepage: www.else ie .com/loca e/enb
h ps://doi.o g/10.1016/j.enbuild.2025.115629
Recei ed 18 Janua y 2025; Recei ed in e ised o m 4 Ma ch 2025; Accep ed 16 Ma ch 2025
Ene gy & Buildings 336 (2025) 115629
A ailable online 21 Ma ch 2025
0378-7788/© 2025 The Au ho s. Published by Else ie B.V. This is an open access a icle unde he CC BY license (
h p://c ea i ecommons.o g/licenses/by/4.0/ ).
equi emen s implemen ed mus go hand-in-hand wi h imp o emen s in
he quali y o li e o use s and hei capaci y o adap o clima e condi-
ions, which is pa icula ly c ucial agains he cu en backg ound o
inc easingly equen hea wa es [7].
Gi en he po en ial en i onmen al, social, and economic bene i s
esul ing om he deep eno a ion p ocess, se e al di e en app oaches
can be iden i ied, namely (wi h some explo a o y s udies).
- E alua ion o a echno-economic me hod o deep eno a ion wi h
mul iple in e en ion scena ios. Aglia di e al. [8] applied his
app oach o a eal case s udy o social housing in Reggio Emilia
(I aly).
- Analysis o he deep eno a ion ma ke and explo a ion o he policy,
economic, social and echnical (PEST) ac o s in luencing he ma -
ke . Mainali e al. [9] p oposed ma ke ing s a egies o de ached
houses in Sweden and Denma k.
- Applica ion o a package o measu es ac oss all buildings, spa ially
di e en ia ing cos -e ec i eness. ¨
Os e b ing e al. [10] concluded
ha he cos -e ec i eness o deep eno a ion should be assessed on a
building-by-building basis a he han by a ea o neighbou hood.
Gi en he inhe en complexi y o his issue, he issue o e o i ing a
building mus be iewed as mul idimensional, wi h a solu ion ha is
nei he singula no app oxima e. Fu he mo e, he p ocess engages
nume ous s akeholde s, including use s, echnicians, manage s, and
builde s, each wi h a ying app oaches and p io i ies. The eno a ion o
he esiden ial building s ock in esponse o clima e change can be
conside ed a ’wicked p oblem’, one which Ri el and Webbe [11]
applied o social policy planning, desc ibing a complex p oblem ha
a ies o e ime and canno be immedia ely e i ied o easily e e sed.
I also di e s in indi idual buildings, equi es signi ican in es men ,
and does no comple ely sol e he p oblem, al hough i does imp o e
pe o mance.
The p elimina y measu es conside ed should ocus on he building
en elope in o de o educe he he mal load. Such measu es can be
implemen ed in a phased manne , ha e a long ope a ional li espan, and
acili a e ene gy sa ings o e ex ended pe iods. This sugges s ha , in he
con ex o deep eno a ion, a posi i e cumula i e e ec can be expec ed.
Howe e , as decision-making wi h combined in e en ion s a egies is a
complex ask, i becomes di icul o iden i y he mos echnically e i-
cien and cos -e ec i e in e en ion solu ions among a di e se se o
al e na i es. A p elimina y comp ehensi e ene gy assessmen is hus
equi ed o asce ain po en ial ene gy enhancemen s, along wi h an
app op ia e economic e alua ion o con as in es men cos s and he
imp o emen p edic ed.
The inco po a ion o digi al echnologies is one o he essen ial
mechanisms which can be implemen ed o adequa ely add ess his
p ocess. The use o sophis ica ed me hodologies o o ecas , examine
and mi iga e he impac s o clima e change on cons uc ions, coupled
wi h he deploymen o a i icial in elligence (AI) echnologies, can be
e icien in add essing his challenge, b inging abou enhanced, op i-
mized and supe io ou comes.
The e alua ion o imp o emen po en ial is ypically conduc ed
using Building Ene gy Modelling (BEM), a he base o Building Ene gy
Simula ion (BES). This echnique, which allows he cons uc ion o
simula ion models o ep esen a i e buildings o a che ypes based on
adi ional ma hema ical models, o e s accu a e p edic ions o ene gy
beha iou a building le el. Howe e , his me hodology has a clea
limi a ion in ob aining esul s a building s ock le el, as his la ge scale
would equi e signi ican compu a ional esou ces o simula e all po-
en ial cases [12]. To add ess hese limi a ions, new eme ging digi al
ools [13] such as a i icial in elligence echniques ha e been de eloped
o applica ion o imp o ing ene gy e iciency and com o condi ions
[14]. These echniques enable he implemen a ion o lea ning algo-
i hms and compu e codes capable o in e ela ing and coupling wi h
he ene gy simula ion models hemsel es. This way i is possible o
c ea e pa ame e ized Building S ock Models (BSMs) based on exis ing
building a che ypes, capable o ep esen ing a la ge building collec i e,
using ewe compu a ional esou ces and sho e calcula ion imes, and
achie ing a la ge building s ock scale [15].
Based on a da a-d i en app oach, Machine Lea ning algo i hms
p o ide signi ican ad an ages in analy ical p edic ion p ocesses, d as-
ically educing calcula ion imes and esul ing in subs an ial p og ess
o building managemen , accu a e es ima ions and decision-making o
he buil en i onmen [16]. Among hese, he use o e olu iona y
echniques o mul i-objec i e nume ical op imiza ion [17] such as
NSGA-II gene ic algo i hms [18], au oma es simula ion and ene gy
e alua ion p ocesses. This app oach equally bene i s selec i e p ocesses,
and he assessmen o op imal e o i solu ions based on selec ed a i-
ables. I cons i u es a igo ous me hodological amewo k o e ing a
b oad pano ama o combina ions o possibili ies and s a egies, enabling
use s o choose he op ions bes sui ed o hei speci ic cons ain s [19],
and p oposing e ec i e ene gy-e icien and ene gy-sa ing ac ions [20].
Thus, he use o au oma ed ma hema ical building pe o mance op i-
miza ion (BPO) pai ed wi h building pe o mance simula ion (BPS) is a
means o e alua ing many di e en design op ions and ob ain he
op imal o nea op imal while achie ing ixed objec i es. Despi e he
exis ing limi a ions, including model unce ain y, compu a ion ime o
di icul y o use, as ou lined by A ia e al. [21] in hei e iew o hei
applica ion in ne ze o ene gy buildings, he ad an ages s ill ou weigh
hese limi a ions. Conside ing his me hodology, he e is a high numbe
o s udies ha op imized he mal en elope condi ions [22], hei com-
bined e ec wi h ene gy supply sys ems [23], building sys em ope a ion
schedules [24] and sola p o ec ion sys ems [25], among o he s, wi h
he main op imiza ion objec i es o minimizing p ima y ene gy con-
sump ion o CO
2
emissions [26], ene gy- ela ed global cos s [27] o
isual and he mal com o [28].
In his con ex , i is also o he u mos impo ance o highligh he
de ec ed esea ch gap on exis ing open access and ee ools ha p o ide
eal op imized e o i solu ions applied o he housing building a he
le el s ock. Se e al wo ks p o ide in o ma ion on cu en building
he mal and ene gy pe o mance, o bo h speci ic case s udies a he
single building le el o speci ic neighbou hoods, no mally h ough
g aphical in o ma ion, mapping echniques o GIS pla o ms, ha usu-
ally classi y buildings acco ding o hei ene gy demand o consump ion
alues, which a e commonly ob ained om Ene gy Pe o mance Ce -
i ica es o s a ic calcula ions [29,30]. Al hough se e al pape s p esen
ene gy e o i s a egies applied o he exis ing housing s ock using
dynamic modelling, hese wo ks no mally ocus on a single building
le el case s udy. In ac , as can be obse ed in he e iew o 153 pape s
conduc ed by Hashempou e al. [31], only 14.3 % o he s udies on
building pe o mance p esen e o i s a egies and, 85.7 % o hem
conside he single building le el. Fo ins ance, he wo k conduc ed by
Ascione e al. [32] is signi ican ly ele an , since op imized e o i
s a egies o he I alian housing building s ock conside ing cos sa ings,
ca bon dioxide emissions and p ima y ene gy consump ion a e de e -
mined. Ne e heless, hese au ho s do no p o ide any ac ual e o i
ool which con ain he ob ained esul s, simply p esen ing hei indings
and conclusions h ough s a ic igu es and w i en desc ip ions.
Th oughou he eno a ion p ocess, he e is one key agen , he
homeowne o occupan , especially in cases o ene gy and socioeco-
nomic ulne abili y [33]. Gene ally, a majo ba ie a ises due o he
limi ed aining o homeowne s and p ope y manage s, coupled wi h a
lack o eliable in o ma ion [34]. Mo eo e , ew s udies connec ene gy
e iciency policies and objec i es wi h homeowne s’ pe spec i es and
hei abili y o eno a e hei homes [35]. The necessa y eno a ion
p ocess should be unde s ood as pa o long- e m housing s ock man-
agemen , so ha he elemen o ime should also be inco po a ed [36]. In
his con ex and li e a u e amewo k, se e al esea ch ques ions a ise:
Wha a e he oppo uni ies o homeowne s o unde ake deep ene gy
eno a ions? Can di e en e ec i e eno a ion s a egies be imple-
men ed in homes ha align wi h he economic capabili ies o hei
C.M. Calama-Gonz´
alez e al.
Ene gy & Buildings 336 (2025) 115629
2
owne s?
The no el objec i e o he p esen esea ch p ojec is o de elop an
open-access and in e ac i e ool o assessing ene gy e o i s a egies,
helping o ob ain aluable in o ma ion o add essing he u gen ly
needed eno a ion o exis ing social housing buildings in sou he n Spain
(Medi e anean clima e). The main aim o his ool is o p o ide a
comp ehensi e ange o di e se, op imized combina ions o ene gy
e o i solu ions, in combina ion wi h a basic analy ical decision-
making p ocess ha conside s a ange o c i e ia (including he
imp o emen o use s’ he mal com o , he minimiza ion o economic
aspec s and he conside a ion o he le el o building in e en ion),
a he han ad oca ing o a single op imal solu ion. This open-access
and in e ac i e ool may p o e bene icial o public s akeholde s and
use s in ol ed in he decision-making p ocesses ela ed o building
e o i ing. This esea ch s ands ou om exis ing s udies in he
ollowing ways:
- Al hough he implemen a ion o pa ame ic modelling and a i icial
in elligence echniques on a mul i-objec i e analysis amewo k may
be s anda d in nume ic op imiza ion app oaches, he no el y lays in
hei applica ion o p edic p ecise he mal pe o mance esul s o an
speci ic building s ock which is in u gen need o ene gy eno a ion
gi en i s high social, economic and clima e ulne abili y: he social
housing building in sou he n Spain. Besides, hese me hodologies a e
applied a he building s ock le el a he han a single building le el,
wi h educed compu a ional cos s and high p edic i e capaci y. To
do so, alida ed pa ame e ized s ock building models a e c ea ed
using cha ac e iza ion building in o ma ion con ained in a la ge
building da abase (a ound 39,500 dwellings) which has been p e-
iously s a is ically analysed.
- This app oach allows o ob ain ene gy simula ion esul s o he s ock
le el based on dynamic calcula ions ha o e he possibili y o
simul aneously and concisely assess di e en pa ame e s (such as,
he mal com o and s ess, in e en ion cos s, e c.), and in eg a e
ime-based scena io e alua ions In ac , o mo e accu a ely ep esen
he eal he mal pe o mance o his ulne able social s ock, adap i e
he mal com o has been conside ed as an op imiza ion objec i e, in
con as o he mos commonly used app oach based on ene gy-
ela ed aspec s. Speci ically, a p oposal o op imized and he mally
e icien e o i s a egy packages o exis ing buildings h ough a
p ocess o di e gen and hinking analysis in ol ing mul iple op ions
and al e na i es is p esen ed h ough he de elopmen o an open-
access ool. This ool helps egional decision-making, allowing
use s, echnicians o public en i ies o choose he mos sui able op-
ions o hei speci ic needs., since i also includes empo al aspec s
ha conside s aging eno a ion solu ions.
2. Me hodology
The me hodology ollowed in his esea ch is explained in he wo k
s ages shown in Fig. 1 and includes he combina ion o di e en
me hods: s a is ical echniques o he assessmen o ex ensi e da abases
and alida ion o building ene gy models, collec ion o on-si e mea-
su emen s o eal case s udies, cons uc ion o dynamic simula ion
models h ough ene gy simula ion ools and in elligen compu a ion
h ough mul i-objec i e analysis and au oma ic nume ical op imiza ion
wi h gene ic algo i hms.
2.1. Desc ip ion o he me hodology s ages
2.1.1. S age 1. Building cha ac e iza ion o he exis ing social housing s ock
Building cha ac e iza ion da a ha e been ob ained om a public
da abase p o ided by he Andalusian Housing and Rehabili a ion
Agency (AVRA in Spanish), whose con en has been enhanced by
inco po a ing new s udy a iables. Following he s a is ical analyses
ca ied ou on his da abase, con aining in o ma ion on app oxima ely
39,500 public social housing dwellings in sou he n Spain buil be ween
1950–2010, he mos ep esen a i e a iabili y anges o he ypolog-
ical, mo phological, and cons uc ion cha ac e is ic a iables o he
exis ing building s ock ha e been de ined. In he p esen esea ch, he
s udies ha e ocused on he p edominan building ypologies (H-block
and linea block, which oge he ep esen mo e han 82 % o he
buildings included in he a o emen ioned da abase) [37,38] loca ed in
he clima ic zone wi h he la ges ex ension and ep esen a ion wi hin
he Andalusian Medi e anean a ea (including ci ies such as Se ille,
C´
o doba, o Huel a). Acco ding o he K¨
oppen-Geige classi ica ion his
a ea is classi ied as Medi e anean clima e (Csa) [39].
2.1.2. S age 2. Cons uc ion o alida ed building ene gy models o
ep esen a i e building a che ypes
A his s age, ep esen a i e building a che ypes o he wo p e-
dominan building ypologies included in he public da abase (H-block
and linea block) a e selec ed and moni o ed o e an ex ended leng h o
ime, conside ing di e en seasonal pe iods (summe , win e , and mid-
season). Subsequen ly, dynamic simula ion models a single-building
le el a e cons uc ed based on he geome ic, physical and cons uc-
i e da a o he case s udies, using he Ene gyPlus e sion 9.2 ene gy
simula ion so wa e and linking clima ic da a h ough “.epw” clima e
iles. These models unde go a p io alida ion p ocess and a e calib a ed
using Bayesian s a is ical echniques and he compa ison o moni o ed
in-si u da a wi h dynamic simula ion so wa e p edic ions. The esul s
a e hen alida ed based on he unce ain y coe icien s es ablished in
he ASHRAE Guidelines [40]. P e iously published wo ks ha e de ailed
his calib a ion and alida ion p ocess o bo h he building a che ype o
he H-block [41] and linea block models [42].
2.1.3. S age 3. Cons uc ion o pa ame e ized and alida ed building s ock
ene gy models
A pa ame e iza ion p ocess o he alida ed single-building le el
models is la e ca ied ou , aiming o cons uc pa ame e ized and
alida ed models a building s ock le el, applying a s a is ical bo om-up
app oach o analyse buildings a a neighbou hood o egional scale. This
s ep allows he cons uc ion o ep esen a i e a che ypes o he exis ing
esiden ial s ock based on p edominan building ypologies (linea and
Fig. 1. Wo k s ages included in he me hodology ollowed.
C.M. Calama-Gonz´
alez e al.
Ene gy & Buildings 336 (2025) 115629
3
H-block buildings, as ep esen ed in Fig. 2), and h ough au oma ic
inpu se ing a ia ion o he pa ame e ized a iables, simul aneous
simula ions can be ca ied ou o housands o ep esen a i e buildings,
a he han jus a single-building le el. Fo his pu pose, jEPlus e sion
2.1 ene gy simula ion so wa e and bo h EP and Py hon p og amming
languages ha e been used, ca ying ou he pa ame e iza ion o di e en
a iables:
1) Geome ic a iables. Fi s ly, he h ee dimensions o space (X, Y, Z)
o he building a che ype model a e pa ame e ized. Two key a i-
ables a e de ined: he a e age buil a ea o he dwelling (m
2
), and he
loo - o-ceiling heigh (m), bo h pa ame e s which depend on he
geome ic X, Y and Z a iables. Addi ionally, he building o ien a-
ion (◦wi h espec o No h) and he window- o-wall a io o he
acades (%) a e pa ame e ized, as a e he numbe o loo s o he
building and he ype o u ban ypology. This las a iable allows he
analysis o di e en cases o building block posi ioning wi hin he
u ban con ex : a single building wi h no shading e ec s om he
su oundings (Isola ed), a ached case wi h wo pa y walls
(Te aced), o a ached case wi h one pa y wall (Co ne ). In o de o
conside he numbe o loo s and u ban ypologies, i is necessa y o
cons uc as many “.id ” iles in Ene gyPlus as de ined op ions, which
a e la e impo ed in o he simula ion so wa e as a pa ame e ized
ile whe e ma hema ical calcula ions can be conduc ed (“.im ”). All
his allows he ep esen a ion o di e en building geome ies.
2) Physical and cons uc i e a iables ela ed o he en elope. This
includes pa ame e ized a iables o de ine he he mal conduc i i y
p ope ies (W/m⋅K), speci ic hea (J/kg⋅K), densi y (kg/m
3
), and
hickness (m) o he oo , acade, and in e media e loo s, along wi h
he sola abso p ance p ope ies o he oo and acade. I is hus
possible o simula e nume ous cons uc ion en elope solu ions wi h
di e en U- alues (W/m
2⋅
K). Addi ionally, pa i ion wall hickness
(m), window glazing and ame ypes ( o he e alua ion o windows
wi h di e en U- alues), and in il a ion a e (ACH) ha e been
pa ame e ized in he models o conside di e en scena ios.
3) Ope a ional a iables. In his case, people densi y (people/m
2
) has
been pa ame e ized o simula e di e en occupancy loads acco ding
o he ele an Spanish egula ions, as well as he opening o he mos
commonly used ex e io sola p o ec ion sys em in sou he n Spain
( olle blinds). Simila ly, he models ha e been pa ame e ized wi h
he a iable o nigh - ime na u al en ila ion a e (ACH) applied
du ing summe pe iods in esiden ial buildings in sou he n Spain
(Medi e anean clima e) o simula e di e en scena ios.
In o he wo ds, all de ining a iables o he building a che ypes ha e
been pa ame e ized. Thus, he a iabili y anges which esul om he
s a is ical e alua ion o he ex ensi e building da abase con aining in-
o ma ion o he social housing s ock in sou he n Spain (p esen ed in
subsec ion 2.1.1 and la e 2.2), ha e been assigned o he pa ame e ized
a iables o he a che ypes simula ion models, as possible alue anges.
As a esul , he cu en pe o mance o he exis ing public social housing
o sou he n Spain (named as “base case”, wi h no e o i s a egies
implemen ed) a he s ock le el (in con as o he single-building le el)
can be assessed, p o iding gene al esul s a egional scale, h ough
dynamic simula ions o ep esen a i e case s udy buildings, and also
epo ing aluable in o ma ion o he la e decision-making e o i
p ocess.
In ega d o simula ion assump ions, shadows cas ed by neighbou -
ing buildings a e no conside ed in he modelling p ocess gi en ha
hese aspec s canno be gene ally pa ame e ized. Howe e , shading
cas ed by he u ban g ouping o he building ypologies hemsel es,
ha e been aken in o accoun in he case o e ace and co ne building
ypologies. In ela ion o he mal b idges, adjus men s o he conduc ion
calcula ions o linea and poin he mal b idges ha e been conside ed
du ing he calib a ion and alida ion p ocess o he pa ame e ized
building s ock model.
2.1.4. S age 4. De ini ion o passi e and low-cos ene gy e o i solu ions
applicable o he social housing s ock
The p oposal o possible e o i solu ions o imp o e he ene gy
pe o mance o he exis ing social housing s ock has been o mula ed
conside ing di e en c i e ia, which akes in o accoun expe knowl-
edge and cons uc ion p ac ices in he egion. Fi s ly, use s’ social
ulne abili y based on socio-economic aspec s was aken in o accoun ,
as well as he inhe en need o p opose cos -app op ia e e o i solu-
ions o he use s o hese dwellings, in keeping wi h he esul s e-
po ed in [33]. In addi ion, e o i measu es p oposed a ge he mos
in luen ial a iables on he mal building pe o mance, p e iously ana-
lysed hough sensi i i y analysis conduc ed in o he wo ks [43]. In his
case, whe e he use o HVAC sys ems in social dwellings is limi ed due o
he lowe spending powe o he use s o hei ins alla ion and ope a-
ion, he mal com o has been conside ed a ele an ac o in he pe -
o mance balance o social esiden ial buildings. Fo his eason, he
e o i measu es p oposed ocus p ima ily on passi e imp o emen o
he he mal en elope o he buildings, he enhancemen o ope a ional
measu es ela ing o ai -exchange a iables, and window-con igu a ion
aspec s. These measu es also add ess he possible applica ion o low-cos
ac i e solu ions, mos ly linked o mechanical en ila ion sys ems. All
hese conside a ions a e in line wi h he cu en Eu opean and na ional
ene gy s a egies aimed a achie ing social well-being h ough he
deca boniza ion and e o i o exis ing building s ock. In iew o all he
abo e, a se o ene gy e o i solu ions is p oposed in sec ion 3.2.
To e ec i ely inco po a e he e o i s a egies p oposed in he en-
e gy simula ion model a he building s ock le el, i is necessa y o
pe o m a second pa ame e iza ion p ocess, which may allow o simu-
la e he mal pe o mance a e ene gy e o i ing he building s ock.
Fig. 2. Sample o linea and H-block buildings a che ypes.
C.M. Calama-Gonz´
alez e al.
Ene gy & Buildings 336 (2025) 115629
4
This en ails: A) inco po a ing new pa ame ized a iables ( o example,
de ining he possible implemen a ion o a mechanical en ila ion sys em
o he implemen a ion o cons uc i e e o i solu ions o he building
en elope o e he exis ing building (base case); o B) modi ying he
al eady pa ame e ized a iables ( o ins ance, adding new schedules o
he use o na u al en ila ion o di e en scena ios o sola p o ec ion
sys ems). Thus, his imp o emen makes i possible o assess he pe -
o mance o he exis ing building social housing s ock a egional scale
h ough p edominan building simula ion case s udies, conside ing
possible ene gy e o i s a egies.
2.1.5. S age 5. Mul i-objec i e analysis o nume ical op imiza ion o
e o i solu ions
The objec i e a iables de ined in his esea ch aim o minimize he
he mal discom o o he social use s in hei home, aking in o accoun
hei economic and he mal adap a ion capabili ies. The e o e, he
objec i e a iables co espond o he annual pe cen age o o e hea ing
hou s (%), he annual pe cen age o unde cooling hou s (%), and he
ini ial in es men cos s in ela ion o he e o i solu ions (
€
/m
2
, whe e
m
2
e e s o he buil a ea pe housing uni ). All h ee pa ame e s a e
equally weigh ed in he op imiza ion p oblem.
Pe cen ages o o e hea ing and unde cooling hou s a e calcula ed
based on he adap i e he mal com o model de ined in EN 16798-
1:2019 [44] and a e de e mined in ela ion o he pe cen age o hou s
exceeding he uppe limi o he adap i e com o band and ha alling
below he lowe limi o he band, espec i ely. The adap i e he mal
com o band is ob ained by de ining a p edic ed pe cen age o dissa -
is ied (PPD) lowe han 10 %, co esponding o an in e al o +3◦C and
−4◦C (uppe and lowe limi s, espec i ely) on he adap i e com o
empe a u e (T
c
). This empe a u e is de i ed om he unning mean
d y bulb ou doo empe a u e (T
e
) which in u n depends on he daily
mean d y bulb ou doo empe a u e o he p e ious 1 o 7 days (T
e1
o
T
e7
) (Eqs. (1) and (2)). Fu he in o ma ion on his me hodology can be
ound in he e e enced s anda d.
Tc=0.33 ×Te+18.8 (1)
Te= (Te1 +0.8⋅Te2 +0.6⋅Te3 +0.5⋅Te4 +0.4⋅Te5 +0.3⋅Te6
+0.2⋅Te7)/3.8 (2)
Rega ding in es men cos s, each e o i solu ion has been associ-
a ed wi h a speci ic Ini ial In es men Cos (
€
/m
2
o
€
/uni ) o i s
e ec i e implemen a ion in he building. These cos s include aspec s
such as p ices linked o he disman ling o laye s o elemen s o he
o iginal s a e, whene e equi ed, o he cos o sca olding o auxilia y
elemen s, aiming o p o ide he mos ealis ic cos assessmen s possible.
I should also be no ed ha hese cos s we e ob ained om he Spanish
CYPE P ice Cons uc ion Gene a o [45] and/o eal execu ion p ojec s.
Howe e , hey should only be used o he compa ison o solu ions, as
hey a e subjec o cha ac e is ic changes in he cons uc ion sec o and
may a y wi h ime.
In he op imiza ion p ocess, he NSGA-II gene ic machine lea ning
algo i hm was used in he de ini ion o he mul i-objec i e p oblem
app oach, using he La in Hype cube Sampling me hod o andomly
and s a is ically de ining he case s udies o be analysed. The maximum
numbe o i e a ions was se a 100, wi h a c osso e a e o 100 %
(pa ame e which ep esen s he equency wi h which new solu ions
a e de ined by me ging ea u es o p e ious ones). The mu a ion a e,
which e e s o he equency o andom changes in new solu ions, was
conside ed o be 20 %. Finally, he ou namen selec ion size was se a
2, ha is o say, om wo andom solu ions in he popula ion, he al-
go i hm only kep he bes solu ion i i was conside ed i . These alues
chosen we e based on he compu a ional esou ces a ailable and
p e ious expe iences conduc ing mul i-objec i e calcula ions [46]. In
he op imiza ion p oblem, a se o domina ed solu ions a e ob ained,
which ep esen he “Pa e o F on ” o se o minimum alues, since all
he h ee op imiza ion objec i es a e minimized. This means ha a se o
op imal s a egies may be ob ained, o e ing use s a basic p ocess o
p ima y analysis and decision-making adap ed o hei speci ic needs.
Fo his pu pose, he jEPlus +EA e sion 2.1 simula ion so wa e, as
well as he EP and Py hon p og amming languages a e used o he
o mula ion o he op imiza ion p oblem.
2.1.6. S age 6. C ea ion o an in e ac i e open-access ool o isualizing
op imized e o i s a egies
To p omo e he dissemina ion o esul s while ensu ing a signi ican
social impac , a ee ool ha con ains he scien i ic indings epo ed in
his esea ch in e ms o op imized e o i s a egies applied o he
exis ing social housing s ock in sou he n Spain has been gene a ed using
he open-access package HiPlo (High-dimensional in e ac i e plo ing),
an in e ac i e isualiza ion ool o high-dimensional da a h ough
pa allel plo s [47] c ea ed wi h Py hon p og amming language. The
in e ac i e pa allel axes plo has been expo ed o “.h ml” o ma ,
enabling any use o isualize and manipula e he esul s ia a web
b owse . Mo eo e , use s ha e he capabili y o expo he da a o a “.
cs ” ile o u he da a p ocessing. Thus, he ool becomes a ully open
and ope a ional esou ce o he use .
2.2. Case s udy desc ip ion
2.2.1. S a is ical cha ac e iza ion o he social housing building in sou he n
Spain
The clima ic se e i y o he loca ions e alua ed can be de ined by
Cooling Deg ee-Days (CDD) and Hea ing Deg ee-Days (HDD). A base
empe a u e o 20 ◦C, he alue used by he cu en Spanish CTE s an-
da d [48] o de ine clima ic a eas, has been de ined o he calcula ion o
CDD and HDD. CDD alues o he analysed clima e zone ange om
1190 o 1490, while HDD alues ange om 700 o 970.
Table 1 summa izes he a iabili y anges de ec ed o he cha ac-
e iza ion o he mos ep esen a i e building ypologies (H-block and
linea block). Taking in o conside a ion hese anges, i should be no ed
ha 64 % o he H-shaped buildings ound in hese e i o ies a e 3 o 5
loo s high. Fu he mo e, 89 % o hem ha e a window- o-wall a io o
be ween 10–30 %, and in 91 % o cases, he dwellings ha e an a e age
buil a ea o a ound 70 o 115 m
2
. Rega ding he linea blocks, 88 % o
he buildings o his ypology a e be ween 3 and 4 loo s high, wi h 78 %
ha ing a window- o-wall a io o be ween 10 and 20 %. Simila ly, a
simula ion o linea block cases wi h a e age dwelling buil a eas o
be ween 70 and 115 m
2
would include 81 % o he buildings in his
clima ic egion. These da a u he suppo he ep esen a i eness o he
a iabili y anges ob ained.
2.2.2. Ene gy e o i solu ions p oposed o he social housing s ock
Based on he c i e ia desc ibed in sec ion 2.4, he ollowing ene gy
e o i solu ions ha e been conside ed (Table 2):
1) Fo he oo e o i , a o al o 15 solu ions we e p oposed, in addi-
ion o he non- e o i ed case. Bo h he imp o emen associa ed
wi h he applica ion o a low emissi i y ex e nal pain , wi h a sola
abso p ance below 0.3 (P), and solu ions in ol ing he addi ion o
he mal insula ion o he o iginal oo we e conside ed. In his las
case, he mal insula ion was added in e nally (In) o ex e nally (Ou )
o he base case oo , in di e en hicknesses (0.06, 0.08, 0.09, 0.10,
0.12, 0.14 m) depending on he ype o insula ion used (MW o
mine al wool wi h a he mal conduc i i y o 0.045 W/m⋅K and XPS
C.M. Calama-Gonz´
alez e al.
Ene gy & Buildings 336 (2025) 115629
5
o ex uded polys y ene wi h a he mal conduc i i y o 0.034 W/
m⋅K). Addi ionally, he possibili y o inco po a ing a g een oo so-
lu ion (G een) was also conside ed. The 15 solu ions a e de ined in
ela ion o he combina ion o all hese pa ame e s, p esen ing he
mos commonly used e o i solu ions in he ma ke .
2) Fo he acade, 15 possible solu ions we e analysed, in addi ion o he
non- e o i ed case. As seen on he oo s, he possible applica ion o
a low emissi i y ex e nal pain wi h a sola abso p ance lowe han
0.3 (P) was also conside ed o he acade, as was he inco po a ion
o he mal insula ion on he base case solu ion. Fo he acade,
insula ion hicknesses conside ed we e 0.04, 0.05, 0.06, 0.075, 0.08,
0.10, and 0.12 m, depending on he ype o insula ion used and i s
posi ion on he açade. As seen in he able men ioned, he p oposed
insula ions co espond o MW (mine al wool wi h he mal conduc-
i i y o 0.037 W/m⋅K), PUR (polyu e hane wi h 0.028 W/m⋅K), EPS
(expanded polys y ene wi h 0.037 W/m⋅K), and RW (mine al wool
wi h 0.035 W/m⋅K). Rega ding he posi ion o he insula ion, in e io
insula ion (In), ex e io insula ion (Ou ), and a scena io o injec ing
he mal insula ion in o he p e-exis ing ai ca i y (Ca) we e
con empla ed. A en ila ed acade solu ion (FV) adap ed o he
Medi e anean clima e was also p oposed, conside ing an opaque
ce amic cladding.
3) 12 con igu a ions we e analysed o windows, in addi ion o he non-
e o i ed case (single glazing wi h aluminium ame being he mos
common solu ion based on he in o ma ion eco ded in he public
da abase). These solu ions a ose om he combina ion o di e en
double glazing wi h di e en ame ypes (PVC o aluminium wi h
he mal b idge b eak). Rega ding double glazing, he possibili y o
including low emissi i y glasses is con empla ed, bo h on he ex e-
io and in e io aces, wi h hicknesses o 4 and 6 mm, and wi h 8,
10, o 12 mm ai gaps. T iple glazing windows and ca i y gases wi h
be e pe o mance (a gon, xenon, k yp on) we e no conside ed due
o hei high economic cos o social housing.
Likewise, e o i measu es linked o ope a ional and usage pa am-
e e s o he dwellings a e p oposed (Table 3), speci ically associa ed
wi h:
1) Na u al en ila ion. Op imizing he use schedule o he na u al
en ila ion sys ems in he dwellings (windows) acco ding o he
seasonal pe iod (summe and win e ) was conside ed as a e o i
s a egy. As a esul , ou di e en schedules o na u al en ila ion
ha e been included.
2) Sola p o ec ion sys ems. Fou possible scena ios we e aken in o
accoun : absence o sola p o ec ion sys ems (non- e o i ed case);
implemen a ion o ex e nal mobile sola p o ec ion sys ems based on
PVC olle blinds, ei he wi h an a e age annual ape u e o 50 % o
he ime open; o wi h an op imized opening schedule depending on
he season (summe and win e ); and, inally, he inco po a ion o a
ixed ex e nal sola p o ec ion sys em wi h sla s.
3) Mechanical en ila ion sys em. The conside a ion o mechanical
en ila ion as a low-cos HVAC sys em was included, whose ope a-
ion is es ablished based on he equi emen s included in he appli-
cable Spanish egula ions [50], which es ablish a con inuous use
schedule. The e o e, o assess i s implemen a ion, wo possible sce-
na ios a e conside ed (ON, wi h mechanical en ila ion sys em) and
OFF (wi hou he mechanical sys em).
3. Analysis and esul s
3.1. Op imiza ion o e o i solu ions
The esul s analysed in his sec ion ha e been ob ained om
nume ous dynamic simula ions comple ed. In he case o he H-block
building, a o al o 115,538 simula ions we e pe o med, yielding
15,740 sub-op imal solu ions (13.6 %). Fo he linea block, 7,440 sub-
op imal solu ions (20 %) we e ob ained ou o 37,031 simula ions
conduc ed in o al. I is impo an o no e ha he execu ion o such a
high numbe o simula ions is due o se e al ac o s:
1) The exis ing di e ences in he a iabili y anges o he building
a iables which de ine he a che ypes o he H-blocks and linea
blocks.
2) The a ying le els o geome ic complexi y and de ini ion o he
pa ame ic s ock simula ion models associa ed wi h each building
ypology.
3) The need o p o ide a ool which allows esul s o be ob ained based
on di e en case s udies wi h speci ic ini ial condi ions. Tha is,
ce ain de ined inpu s which enable he use o selec he case s udy
ha mos closely esembles hei s a ing condi ions (o ien a ion,
numbe o building loo s, u ban ypology, a e age buil a ea o he
dwelling, and pe cen age o glazing su ace), la e analysing he
ou pu a iables ( e o i solu ion packages, he mal com o
Table 1
Va iabili y anges o he exis ing social housing s ock o he p edominan
ypologies.
Va iables H-
block
Linea
block
Geome y U ban ypology Isola ed, Te aced,
Co ne *
O ien a ion (◦) N-S, E-W
Floo a ea (m
2
) 70–115
Floo heigh (m) 2.50–3.00
Window- o-wall a io (%) 10–30 10–20
Numbe o s o eys 3–5 3–4
Building
en elope
Roo sola abso p ance 0.3–0.9
Roo U- alue (W/m
2
⋅K) 1.2–2.4
Roo hickness (m) 0.25–0.45
Roo he mal conduc i i y (W/
m⋅K)
0.3–0.6
Roo densi y (kg/m
3
) 1000–1800
Roo speci ic hea (J/kg⋅K) 500–1500
Floo U- alue (W/m
2
⋅K) 3.0–7.00
Floo hickness (m) 0.15–0.30
Floo he mal conduc i i y (W/
m⋅K)
0.7–1.7
Floo densi y (kg/m
3
) 1200–1800
Floo speci ic hea (J/kg⋅K) 500–1500
Facade sola abso p ance 0.3–0.9
Facade U- alue (W/m
2
⋅K) 1.2–2.5
Facade hickness (m) 0.10–0.30
Facade conduc i i y (W/m⋅K) 0.2–0.4
Facade densi y (kg/m
3
) 1000–3000
Facade speci ic hea (J/kg⋅K) 500–1500
Pa i ion hickness (m) 0.07–0.12
Type o window glass Single
Type o window ame Aluminium
Window U- alue (W/m
2
⋅K) 5.50–5.70
In il a ion a e (ACH) 0.30–1.00
Ope a ion People densi y (people/m
2
) 0.01–0.15
Na u al en ila ion a e (ACH) 0–4
Summe nigh - ime na u al
en ila ion
22:00–8:00
Blinds ape u e No blind, 50 %, o ally
open
*
Explained in de ail in subsec ions 2.3 and 2.6.
C.M. Calama-Gonz´
alez e al.
Ene gy & Buildings 336 (2025) 115629
6
Table 2
Passi e solu ions conside ed o e o i ing he social housing s ock.
Elemen Label TI
posi ion
TI
ype
TI hickness
(m)
Pa emen G een
co e age
Rende ing Ou . Glass
(mm)
Gap
(mm)
In. glass
(mm)
F ame U- alue (W/
m
2
⋅K)
App oxima e in es men
cos (
€
/m
2
)
Roo Un e o i ed
(non- e o i ed)
− − − X− − − − − − 1.2–2.4 −
P_Roo − − − X−P− − − − 1.2–2.4 29.12
InMW_0.08 In MW 0.08 X − − − − − − 0.26–0.37 16.63
InMW_0.09 0.09 X − − − − − − 0.24–0.33 18.71
InMW_0.10 0.10 X − − − − − − 0.23–0.30 19.97
InMW_0.12 0.12 X − − − − − − 0.20–0.26 23.28
P_InMW_0.08 0.08 X −P− − − − 0.26–0.37 45.75
Ou XPS_0.06 Ex XPS 0.06 X − − − − − − 0.29–0.43 34.59
Ou XPS_0.08 0.08 X − − − − − − 0.25–0.34 37.16
Ou XPS_0.10 0.10 X − − − − − − 0.22–0.28 46.30
Ou XPS_0.14 0.14 X − − − − − − 0.17–0.21 53.62
P_Ou XPS_0.08 0.08 X −P− − − − 0.25–0.34 66.28
G een_Ou XPS_0.06 0.06 −X− − − − − 0.27–0.39 129.17
G een_Ou XPS_0.08 0.08 −X− − − − − 0.23–0.32 133.36
G een_Ou XPS_0.10 0.10 −X− − − − − 0.21–0.27 138.8
G een_Ou XPS_0.14 0.14 −X− − − − − 0.17–0.20 146.12
Wall Un e o i ed
(non- e o i ed)
− − − − − X− − − − 1.2–2.5 −
P_Wall − − − − − P− − − − 1.2–2.5 23.96
InMW_0.05 In MW 0.05 − − X− − − − 0.37–0.56 60.31
InMW_0.06 0.06 − − X− − − − 0.33–0.48 61.02
InMW_0.075 0.075 − − X− − − − 0.29–0.41 64.22
P_InMW_0.06 0.06 − − P− − − − 0.33–0.48 84.98
CaPUR_0.04 Ca PUR 0.04 − − X− − − − 0.36–0.54 36.61
CaPUR_0.05 0.05 − − X− − − − 0.32–0.45 38.54
P_CaPUR_0.05 0.05 − − P− − − − 0.32–0.45 62.51
Ou EPS_0.05 Ex EPS 0.05 − − X− − − − 0.37–0.56 79.47
Ou EPS_0.06 0.06 − − X− − − − 0.33–0.49 80.46
Ou EPS_0.08 0.08 − − X− − − − 0.28–0.39 82.45
Ou EPS_0.10 0.10 − − X− − − − 0.25–0.32 84.44
Ou EPS_0.12 0.12 − − X− − − − 0.22–0.27 86.40
P_Ou EPS_0.06 0.06 − − P− − − − 0.33–0.49 104.43
VF_Ou RW_0.06 RW 0.06 − − − − − − − 0.38–0.56 133.00
Window Un e o i ed
(non- e o i ed)
− − − − − − 4− − Al 5.5–5.7 F om 616.91 o 702.07
€
/uni
4LE-Ai 8-6 − − − − − − 4LE Ai 8 6 PVC o
Al
TBB
2.0–2.3
4LE-Ai 10-6 − − − − − − 4LE Ai 10 6 1.8–2.0
4LE-Ai 12-6 − − − − − − 4LE Ai 12 6 1.7–1.9
4-Ai 8-6LE − − − − − − 4 Ai 8 6LE 2.0–2.3
4-Ai 10-6LE − − − − − − 4 Ai 10 6LE 1.8–2.0
4-Ai 12-6LE − − − − − − 4 Ai 12 6LE 1.7–1.9
TI: he mal insula ion. Ou : ex e nal. In: in e nal. P: low emissi i y ex e nal pain . MW: mine al wool (0.045 W/m⋅K in oo , 0.037 W/m⋅K in wall). XPS: ex uded polys y ene (0.034 W/m⋅K). PUR: polyu e hane (0.028 W/
m⋅K). EPS: expanded polys y ene (0.037 W/m⋅K). RW: mine al wool (0.035 W/m⋅K). Ca: ai ca i y. VF: ce amic en ila ed acade. LE: low emissi i y. Al: aluminium. TBB: he mal b idge b eak. All oo and wall solu ions
main ain he exis ing base solu ion. Roo solu ions o e he possibili y o modi ying he ex e nal co e age (pa emen o g een). In es men cos s include disman ling old elemen s in he solu ion when needed.
C.M. Calama-Gonz´
alez e al.
Ene gy & Buildings 336 (2025) 115629
7
condi ions and ini ial in es men cos s o he in e en ions) in o de
o selec he mos sui able op ion.
The sca e plo s in Fig. 3 show he se o non-domina ed o op imal
solu ions (Pa e o F on ) ob ained in he mul i-objec i e analysis o he
H-block and linea block, espec i ely. The annual pe cen age o o e -
hea ing hou s on he X-axis and unde cooling hou s on he Y-axis a e
indica ed. The solu ions ha e been classi ied by in e en ion cos le el:
low cos (up o 50
€
/m
2
, in g een), medium cos (be ween 50 and 100
€
/m
2
, in blue), and high cos (o e 100
€
/m
2
, in ed). Each poin on he
g aph co esponds o a package o se o e o i solu ions ha a ec he
he mal en elope and bo h ope a ional and sola p o ec ion sys ems.
Gene ally, i can be obse ed ha he e is a wide ange o op imized
e o i solu ions, especially low-cos ones, which lead o a signi ican
imp o emen in he he mal pe o mance o exis ing homes. I can also
be deduced ha solu ions o e ing a mo e comp ehensi e and comple e
ene gy e o i (high-cos s a egies) do no always esul in a subs an ial
imp o emen o he bes he mal pe o mance om an op imiza ion
pe spec i e. Ins ead, hey p esen mo e bounded esul s, wi h pe cen -
ages o o e hea ing hou s anging om app oxima ely 30 o 67 % and
unde cooling hou s om 20 o 50 %. This con as s wi h he g ea e
dispe sion o he es o he lowe -cos solu ions, o example, wi h
anges o 33 o 75 % o o e hea ing hou s and 20 o 75 % o unde -
cooling hou s o he low-cos s a egies. Mo eo e , in solu ions up o 50
€
/m
2
, a clea e linea end is obse ed, especially in he case o he
linea block. Ano he in e es ing aspec is ha , he highe he economic
cos o he e o i s a egies, he g ea e he educ ion in he pe cen age
o unde cooling hou s, which is mo e signi ican han ha o o e -
hea ing hou s. This is a c ucial aspec : since he analysis has been
de eloped on an annual basis, inco po a ing bo h summe and win e ,
achie ing a balance be ween solu ions o ensu e he mal imp o emen
h oughou he whole yea is a complex ask.
3.2. In e ac i e open-access ool
The in e ac i e ool de eloped o isualize he op imized combina-
ions o ene gy e o i s a egies o he exis ing social housing s ock in
sou he n Spain, and he co esponding use manual, a e eely a ailable
as Mendeley Resea ch Da a: h ps://doi.o g/10.17632/c7 cc3y yj.1 (H-
block buildings) and h ps://doi.o g/10.17632/9kz9dchhj3.1 (linea -
block buildings). This ool is also egis e ed in he in ellec ual p ope y
egis ies 04/2024/3551 and 04/2024/3552.
Fig. 4, which se es o illus a e he ool’s capabili ies, will subse-
quen ly be employed o explain i s usage and unc ionali y in his pape .
In de ining he pa allel axes plo and in o de o ensu e he mos use -
iendly manipula ion possible, a se o inpu pa ame e s o a iables
was de ined o allow he use o speci y he ini ial condi ions o hei
pa icula case s udy. Consequen ly, he ool p o ides esul s on how he
selec ion o di e en ene gy e o i packages leads o speci ic he mal
com o ou comes in social dwellings in sou he n Spain, based on p e-
de e mined in e en ion cos s.
Thus, he use is also able o selec he desi ed le el o in e en ion in
he building, any possible economic cons ain s, o he desi ed
imp o emen in com o condi ions o be achie ed. Thus, he ollowing
a iables a e de ined (Fig. 4a):
1) Inpu da a. “O ien a ion” is he i s a iable conside ed, analysing
he p edominan No h-Sou h (“N-S”) and Eas -Wes (“E-W”) o ien-
a ions. Following his, he numbe o “Building loo s” (3, 4, o 5
loo s) is es ablished. Subsequen ly, he selec ion o he u ban “Ty-
pology” ca ego y allows he use o choose be ween an “Isola ed”
building (wi hou shading e ec s om he su oundings), a
“Te aced” building (an a ached case wi h wo pa y walls), o
“Co ne ,” which is an a ached case wi h one pa y wall. In he las
case, and depending on he selec ed o ien a ion, o buildings wi h
no h- o eas - acing pa y walls, he use should selec “Co ne
(uppe , igh )”, while i he pa y wall aces sou h o wes , he
“Co ne (lowe , le )” op ion should be selec ed. Addi ionally, he
use can app oxima e he a e age buil a ea o hei home acco ding
o he anges included in he g aph on he “A ea” axis ( om 70 o
115 m
2
, each 5 m
2
). The las inpu co esponds o he pe cen age o
glazing su ace (“WWR”) o he home ( om 10 o 20 % o 30 % e e y
5 %, depending on he p edominan building ypology analysed). All
hese pa ame e s and hei possible op ions ha e been de ined based
on he p edominan a iabili y anges ob ained in he building
cha ac e iza ion in s age 1.
2) Ou pu da a. This co esponds o he a iables ha he use can
analyse. Fi s ly, he ou pu s ela ed o he e o i solu ion packages
a e p o ided. These include he ype o glazing and ame (“Glass”
and “F ame” in he in e ac i e igu e), as well as he cons uc i e
solu ion o he opaque en elope (“Wall” and “Roo ”). Addi ionally,
a iables e e ing o he ype o sola p o ec ion sys em (“Sola
p o ec ion”), he schedule o he na u al en ila ion sys em (“Na
Ven ”), and he use o a mechanical en ila ion sys em (“Mech Ven ”)
a e included. All he a o emen ioned a iables include a non-
e o i ed case. The desc ip ion o he symbols used in he in e ac-
i e igu e o de ine each possible op ion can be ound in he “Label”
column in Tables 2 and 3 o his pape . Finally, also as ou pu da a,
he g aph displays he esul s o he mal com o condi ions, based
on he pe cen age o annual “O e hea ing hou s” and “Unde cooling
Table 3
Ope a ional solu ions conside ed o e o i ing he social housing s ock.
Solu ion Label Summe Win e App oxima e in es men cos (
€
/ uni )
Na u al en ila ion O −−−
Sum&Win_8-9 h 8:00–9:00 8:00–9:00 −
Sum_8-9 h, Win_14-15 h 8:00–9:00 14:00–15:00 −
Sum_22-8 h, Win_8-9 h 22:00–8:00 8:00–9:00 −
Sum_22-8 h, Win_14-15 h 22:00–8:00 14:00–15:00 −
Rolle blinds No p o ec ion No sola p o ec ion sys em −
Blinds 50 % 50 % opened 137.53
Blinds op imized 0*% om 8:00–16:00
50 % om 16:00–21:00
100 % 21:00–7:00
100 % om 9:00–19:00
0*% om 19:00–9:00
137.53
Ex e nal p o ec ion Ex e nal sola p o ec ion (sla s) Ex e nal sola p o ec ion (sla s) 394.06
Mechanical en ila ion On con inuous ON 650.00
OFF OFF −
*
0 % blind ape u e le el means o ally closed and 100% e e s o o ally open.
C.M. Calama-Gonz´
alez e al.
Ene gy & Buildings 336 (2025) 115629
8
Fig. 3. Dispe sion plo s o he annual % o o e hea ing and unde cooling hou s o he op imized esul s, classi ied acco ding o economic in es men le el: a) up o
50
€
/m
2
o he H-block and b) up o 50
€
/m
2
o he linea block; c) be ween 50 and 100
€
/m
2
o he H-block and d) be ween 50 and 100
€
/m
2
o he linea block;
e) o e 100
€
/m
2
o he H-block and ) o e 100
€
/m
2
o he linea block.
C.M. Calama-Gonz´
alez e al.
Ene gy & Buildings 336 (2025) 115629
9
[25] A. Allouhi, M.B. Amine, C. Reisch, Mul i-objec i e op imiza ion o sola ene gy
sys ems o elec ici y and ho wa e gene a ion in collec i e esiden ial buildings
conside ing he powe - o-hea concep , Appl. The m. Eng. 230 (2023) 120658,
h ps://doi.o g/10.1016/j.appl he maleng.2023.120658.
[26] G. Manc`
o, U. Tesio, E. Guelpa, V. Ve da, A e iew on mul i ene gy sys ems
modelling and op imiza ion, Appl. The m. Eng. 121871 (2023), h ps://doi.o g/
10.1016/j.appl he maleng.2023.121871.
[27] F. Ascione, N. Bianco, G.M. Mau o, D.F. Napoli ano, Building en elope design:
mul i-objec i e op imiza ion o minimize ene gy consump ion, global cos and
he mal discom o . Applica ion o di e en I alian clima ic zones, Ene gy 174
(2019) 359–374, h ps://doi.o g/10.1016/j.ene gy.2019.02.182.
[28] M. Baghoolizadeh, M. Ros amzadeh-Renani, R. Ros amzadeh-Renani, D. Togh aie,
Mul i-objec i e op imiza ion o Vene ian blinds in o ice buildings o educe
elec ici y consump ion and imp o e isual and he mal com o by NSGA-II,
Ene g. Buildings 278 (2023) 112639, h ps://doi.o g/10.1016/j.
enbuild.2022.112639.
[29] C. Bel ´
an-Velamaz´
an, M. Monz´
on-Cha a ías, B. L´
opez-Mesa, A new app oach o
na ional-scale Building Ene gy Models based on Ene gy Pe o mance Ce i ica es in
Eu opean coun ies: he case o Spain, Heliyon 10 (3) (2024), h ps://doi.o g/
10.1016/j.heliyon.2024.e25473.
[30] C. Villanue a-Díaz, M. ´
Al a ez-Sanz, ´
A. Campos-Celado , J. Te ´
es-Zubiaga, The
open da a po en ial o he geospa ial cha ac e isa ion o building s ock on an
u ban scale: me hodology and implemen a ion in a case s udy, Sus ainabili y 16 (2)
(2024) 652, h ps://doi.o g/10.3390/su16020652.
[31] N. Hashempou , R. Tahe khani, M. Mahdikhani, Ene gy pe o mance op imiza ion
o exis ing buildings: a li e a u e e iew, Sus ain. Ci ies Soc. 54 (2020), h ps://doi.
o g/10.1016/j.scs.2019.101967.
[32] F. Ascione, N. Bianco, C. De S asio, G.M. Mau o, G.P. Vanoli, Add essing la ge-scale
ene gy e o i o a building s ock ia ep esen a i e building samples: public and
p i a e pe spec i es, Sus ainabili y 9 (6) (2017) 940, h ps://doi.o g/10.3390/
su9060940.
[33] C.M. Calama-Gonz´
alez, R. Escand´
on, R. Su´
a ez, A. Alonso, ´
A.L. Le´
on-Rod íguez,
Household ene gy ulne abili y e alua ion in sou he n Spain h ough pa ame ic
ene gy simula ion models and socio-economic da a, Sus ain. Ci ies Soc. 103 (2024)
105276, h ps://doi.o g/10.1016/j.scs.2024.105276.
[34] P. Tuominen, K. Klobu , A. Tolman, A. Adjei, M. de Bes -Waldhobe , Ene gy
sa ings po en ial in buildings and o e coming ma ke ba ie s in membe s a es o
he Eu opean Union, Ene g. Buildings 51 (2012) 48–55, h ps://doi.o g/10.1016/j.
enbuild.2012.04.015.
[35] R. Chen, R. Fan, Q. Yao, R. Qian, E olu iona y dynamics o homeowne s’ ene gy-
e iciency e o i decision-making in complex ne wo k, J. En i on. Manage. 326
(2023) 116849, h ps://doi.o g/10.1016/j.jen man.2022.116849.
[36] T. Fawce , Explo ing he ime dimension o low ca bon e o i : owne -occupied
housing, Build. Res. In . 42 (4) (2014) 477–488, h ps://doi.o g/10.1080/
09613218.2013.804769.
[37] C.M. Calama-Gonz´
alez, R. Su´
a ez, A.L. Le´
on-Rod íguez, 2020. Building
cha ac e isa ion and assessmen me hodology o social housing s ock in he
wa me Medi e anean clima e: he case o sou he n Spain, In: IOP Con e ence
Se ies: Ea h and En i onmen al Science, Vol. 410, No. 1, p. 012049, IOP
Publishing, doi: 10.1088/1755-1315/410/1/012049.
[38] C.M. Calama-Gonz´
alez, R. Escand´
on, A. Alonso, A.L. Le´
on-Rod íguez, R. Su´
a ez,
Building assessmen and s a is ical cha ac e isa ion o he medi e anean social
housing s ock in Sou he n Spain, in IOP Con e ence Se ies: Ea h and
En i onmen al Science, Vol. 1050, No. 1, p. 012020, IOP Publishing, 2022, July.
doi: 10.1088/1755-1315/1050/1/012020.
[39] M. Ko ek, J. G iese , C. Beck, B. Rudol , F. Rubel, Wo ld map o he K¨
oppen-Geige
clima e classi ica ion upda ed, Me eo ol. Z. 15 (3) (2006) 259–263, h ps://doi.
o g/10.1127/0941-2948/2006/0130.
[40] ASHRAE, Ame ican, Socie y o Hea ing, Re ige a ing and Ai Condi ioning
Enginee s. ASHRAE Guideline 14-2014: Measu emen o Ene gy, Demand and
Wa e Sa ings, ASHRAE, 2014.
[41] C.M. Calama-Gonz´
alez, R. Su´
a ez, ´
A.L. Le´
on-Rod íguez, The mal com o
p edic ion o he exis ing housing s ock in sou he n Spain h ough calib a ed and
alida ed pa ame e ized simula ion models, Ene g. Buildings 254 (2022) 1–14,
h ps://doi.o g/10.1016/j.enbuild.2021.111562.
[42] R. Escand´
on, F. Ascione, N. Bianco, G.M. Mau o, R. Su´
a ez, J.J. Send a, The mal
com o p edic ion in a building ca ego y: A i icial Neu al Ne wo k gene a ion
om calib a ed models o a social housing s ock in sou he n Eu ope, Appl. The m.
Eng. 150 (2019) 492–505, h ps://doi.o g/10.1016/j.
appl he maleng.2019.01.013.
[43] C.M. Calama-Gonz´
alez, R. Su´
a ez, R. Escand´
on, Building pa ame e in luence on
o e hea ing and unde cooling isks in he Medi e anean social housing s ock o
sou he n Spain, in: E3S Web o Con e ences, Vol. 545, p. 04001, EDP Sciences,
2024, doi: 10.1051/e3scon /202454504001.
[44] Eu opean s anda ds commi ee, EN 16798–1:2019 Ene gy Pe o mance o
Buildings – Ven ila ion o buildings - Pa 1: Indoo En i onmen al Inpu
Pa ame e s o Design and Assessmen o Ene gy Pe o mance o Buildings
Add essing Indoo Ai Quali y 2019 The mal En i onmen Ligh ing and Acous ics -
Module M1 6.
[45] C.Y.P.E. Ingenie os, Gene ado de p ecios de la cons ucci´
on. Espa˜
na, h p://www.
gene ado dep ecios.in o/#gsc. ab=0(Accessed 17 May 2024).
[46] C.M. Calama-Gonz´
alez, P. Symonds, ´
A.L. Le´
on-Rod íguez, R. Su´
a ez, Op imal
e o i solu ions conside ing he mal com o and in e en ion cos s o he
Medi e anean social housing s ock, Ene g. Buildings 259 (2022) 1–13, h ps://doi.
o g/10.1016/j.enbuild.2022.111915.
[47] HiPlo , High Dimensional In e ac i e Plo ing Ci cle Cl. A ailable online: h
ps://pypi.o g/p ojec /hiplo / (accessed 17 May 2024).
[48] CTE, Spanish Technical Building Code, Basic documen : ene gy sa ing, 2019.
Spanish Go e nmen . A ailable online: h ps://www.codigo ecnico.o g (accessed
17 May 2024).
[49] L. De Boeck, S. Ve beke, A. Audenae , L. De Mesmaeke , Imp o ing he ene gy
pe o mance o esiden ial buildings: a li e a u e e iew, Renew. Sus ain. Ene gy
Re . 52 (2015) 960–975, h ps://doi.o g/10.1016/j. se .2015.07.037.
[50] G. Semp ini, R. Gulli, A. Fe an e, Deep egene a ion s shallow eno a ion o
achie e nea ly Ze o Ene gy in exis ing buildings: ene gy sa ing and economic
impac o design solu ions in he housing s ock o Bologna, Ene g. Buildings 156
(2017) 327–342, h ps://doi.o g/10.1016/j.enbuild.2017.09.044.
[51] S. F i z, M. Pehn , P. Mellwig, J. Vol , Planned s aged deep eno a ions as he main
d i e o a deca bonized Eu opean building s ock, in: ECEEE Summe S udy
P oceedings, Vol. 7, 2019.
[52] P. Femenías, K. Mj¨
o nell, L. Thu ande , Re hinking deep eno a ion: he
pe spec i e o en al housing in Sweden, J. Clean. P od. 195 (2018) 1457–1467,
h ps://doi.o g/10.1016/j.jclep o.2017.12.282.
C.M. Calama-Gonz´
alez e al.
Ene gy & Buildings 336 (2025) 115629
16
