ScienceDi ec
A ailable online a www.sciencedi ec .com
T anspo a ion Resea ch P ocedia 82 (2025) 530–546
2352-1465 © 2024 The Au ho s. Published by ELSEVIER B.V.
This is an open access a icle unde he CC BY-NC-ND license (h ps://c ea i ecommons.o g/licenses/by-nc-nd/4.0)
Pee - e iew unde esponsibili y o he scien i ic commi ee o he 16 h Wo ld Con e ence on T anspo Resea ch
10.1016/j. p o.2024.12.058
Keywo ds: T a el beha iou ; agmen a ion analysis; gende ;li e cou se, ac i i y pa icipa ion
1. In oduc ion
The spa ial and empo al dis ibu ion o housing and ac i i ies in he u ban en i onmen de e mine how people
mo e o each places in a imely and secu e manne . Howe e , o he issues de ine how we ge access di e en se ices,
* Co esponding au ho .
E-mail add ess: lucia@mejia-do an es.com
16 h Wo ld Con e ence on T anspo Resea ch (WCTR 2023)
Analysing gende equali y in Ba celona h ough
(spa io empo al) segmen a ion
Lídia Mon e oa, Lucía Mejía-Do an esb*, and Jaume Ba celóa
aDepa men o S a is ics and Ope a ions Resea ch Campus No d, Uni e si a Poli ècnica de Ca alunya (UPC) Ba celona Spain 08034
bConsul an . Ka ls uhe, Ge many, 76185
Abs ac
Ac i i y pa icipa ion is in luenced by many ac o s, such as he ones ela ed o he buil en i onmen , bu also o indi idual
a ibu es. He ein we explo e sequences o daily ac i i ies and a el employing echniques om he sequencing o e en s in he
li e cou se o indi iduals. S udying sequences o daily episodes (each ac i i y and each ip) allows us o s udy he en i e ajec o y
o a pe son’s ac i i y du ing a day while a he same ime conside ing he numbe o ac i i ies, o de o ac i i ies in a day, and hei
du a ions join ly. We applied his me hod o a sample o esiden s in he Me opoli an A ea o Ba celona (RMB) in 2018, 2019,
and 2020 T a el Su eys. We ha e ocused on agmen a ion analysis in ac i i y pa icipa ion, especially conce ning gende , age,
ac i i y, and anspo a ion mode. As expec ed, he su ey om 2020 dese es a pa icula app oach since ac i i y pa e ns a y
compa ed o su eys be o e he COVID-19 sp ead ou b eak. In his espec , ac i e anspo shows o be pa icula ly impo an ha
yea . O e all, he esul s show ha ac i i y pa icipa ion canno be disen angled om gende , indi idual li e-cou se, and he buil
en i onmen .
© 2024 The Au ho s. Published by ELSEVIER B.V.
This is an open access a icle unde he CC BY-NC-ND license (h ps://c ea i ecommons.o g/licenses/by-nc-nd/4.0)
Pee - e iew unde esponsibili y o he scien i ic commi ee o he 16 h Wo ld Con e ence on T anspo Resea ch
Lídia Mon e o e al. / T anspo a ion Resea ch P ocedia 82 (2025) 530–546 531
o example, he anspo sys ems a ailable in he a ea, along wi h he opog aphic and wea he cha ac e is ics o he
si e. Fu he mo e, he in insic a ibu es o he pe son who needs o access di e en places, such as he socioeconomic
le el, p o ession, gende , disabili ies, amily s a us, eligion, e hnici y, among o he s, play a key ole in shaping
mobili y choices.
In his ega d, di e en s udies obse e ha li es yles di e no ably in e ms o gende , which in luences mobili y
pa e ns (EIGE, 2019; So o Villag án and Mejía Do an es, 2022). S udies show ha wo king schedules depend on
gende (Hyde e al., 2020). This is ansla ed in o women ha ing mo e complica ed commu e pa e ns han men, based
on sho e , chained ips (C esswell and U eng, 2008; EIGE, 2022),and p e e ing ce ain anspo modes (Cubells e
al., 2020; Hanson, 2010).
E en i a e he COVID-19 pandemic ou b eak, we ha e seen a ema kable inc ease in ICTs in daily ac i i ies, as
many s a is ics show in Eu ope (Eu os a , 2020)and in ou speci ic case s udy (Ins i u o Nacional de Es adís ica, 2021).
Howe e , as Fana e al. (2020) obse e, gene ally people ha wo k om home (W H) end o be high-skilled wo ke s
wi h be e wages and con ac condi ions.
No wi hs anding his inc ease in ICT in daily ac i i ies, many ac i i ies ha need o be conduc ed physically, which
canno be ans e ed o i ual se ices: isi ing schools, doc o s, and esco ing ac i i ies. Fu he mo e, o he ac i i ies
a e p e e ably ca ied ou pe sonally, o example hose ha make us eel pa o a communi y o socie y, ha
con ibu e o people’s well-being. In all hese cases, ci izens depend on he anspo al e na i es a ailable.
In his s udy, we ca y ou a sequence analysis o measu e he agmen a ion in ac i i y pa icipa ion in a T anspo
O ien ed De elopmen (TOD) a ea o discuss he simila i ies o di e ences be ween esul s. This pape aims o gain a
be e unde s anding o he spa io empo al pa e ns and a el beha io in Ba celona, analyze gende ed mobili y
pa e ns, and discuss which cha ac e is ics play a signi ican ole in shaping mobili y. Ou app oach is based on
sequence analysis, making use o h ee ecen and consecu i e mobili y su eys in he Me opoli an Region o
Ba celona (RMB).
This pape is di ided in o 7 sec ions and o ganized as ollows. A e his in oduc ion, we p esen he li e a u e
e iew. A e wa d, we p esen he case s udy, da ase s, and me hodological app oach. Sec ion 5 p esen s he esul s.
Finally, in he las sec ion we p esen he discussion,main conclusions and implica ions o his s udy.
2. Li e a u e e iew
A sequence is a lis o e en s and ac ions pe o med in an o de ed manne . As Abbo and Fo es (Abbo and
Fo es , 1986)desc ibe, a sequence da ase may desc ibe wo di e en pa e ns. One may desc ibe ac i i ies ha occu
only once and he o he may desc ibe ac i i ies ha occu se e al imes in a ce ain sequence. Whe eas he i s idea
may be sol ed wi h pe mu a ion, he second analyzes ecu en e en s.
Many au ho s in he sociology ield ha e s udied li e cou se e en s such as ma iage, childbea ing, and employmen
h ough ac i i y sequence analysis (Elzinga and Lie b oe , 2007a; Giele and Elde , 1998). This analysis has been
ecen ly applied o mobili y beha iou (Ahmed e al., 2021; Shi e al., 2022). In his espec , Bha and Pnja i (Bha
and Pinja i, 2007)obse e ha sequences o ac i i ies and he daily ansi ioning om one ac i i y o ano he as well
as he amoun o ime spen in each ac i i y ep esen an impo an di ec ion o a el beha io analysis.
The analysis o ac i i ies’ sequences and a el is c i ical in o mula ing econome ic models embedded in ac i i y-
based daily simula ions o household ac i i y- a el pa e ns o la ge-scale a el demand analysis as many au ho s
obse e (Bha e al., 2013; Bu chell e al., 2020; Pale i e al., 2017; Rasouli and Timme mans, 2014).
Au ho s, such as S ude and Ri scha d (S ude and Ri scha d, 2016), iden i y he ollowing impo an in e wined
cha ac e is ics:
•Expe ienced s a es: he dis inc al e na i es p esen in he sequence
•Dis ibu ion: he o al ime o s a e dis ibu ion wi hin a sequence
•Timing: he momen in which each s a e appea s
•Du a ion: he pe iod o ime in he di e en successi e s a es
•Sequencing: he o de in which he dis inc successi e s a es ake place
532 Lídia Mon e o e al. / T anspo a ion Resea ch P ocedia 82 (2025) 530–546
On he o he hand, he agmen a ion concep has been many imes add essed (Alexande e al., 2010; Couclelis,
2006; Hube s e al., 2008; E. McB ide e al., 2020). They de ine he agmen a ion o a el and ac i i ies as he
sequence o many sho ips ha ake place du ing he daily schedule o a pe son. Whe eas empo al agmen a ion
is ela ed o he di e en imes ha ac i i ies a e ca ied ou , spa ial agmen a ion is ela ed o he loca ions whe e
ac i i ies a e pe o med. Toge he wi h o he ac i i ies and mo emen s ha occu in a la ge ame ime, hey build up
a s ing o ac i i ies wi h di e en du a ions and pu poses. This s ing may ha e di e en complexi ies depending on
di e en ex insic and in insic cha ac e is ics o he indi idual and he u ban en i onmen which shape mobili y
beha iou .
Fu he mo e, despi e no explaining how and why indi iduals engage in ac i i y- a el agmen a ion, he
classi ica ion o ac i i y- a el agmen a ion in o clus e s makes i possible o unde s and di e en g oups. Fo
ins ance, i allows us, o es ablish he ela ionship wi h he socioeconomic cha ac e is ics o he segmen s. Some
au ho s, such as McB ide e al. (2020), ha e poin ed ou he need o unde s and gende oles ha i is conside ed o be
ela ed o ime alloca ion o ac i i ies and hus, ac i i y- a el.
Addi ionally, some esea che s obse e gende -di e en ia ed pa e ns in segmen a ion analyses. Fo example,
Leszczyc and Timme mans (2002) analyze he Du ch dia y and concluded ha gende and age a e impo an
de e minan s o mo ing om one ac i i y ype o ano he . Bu chell e al., (2020) analyze he gende di e ences in he
segmen a ion o wo kplace pa e ns using he 2015 Eu opean Wo king Condi ions Su ey (EWCS, 2015), which
p esen s in o ma ion be o e he pandemic. The au ho s obse e ha he e a e clea di e ences when analyzing gende ,
o example, women we e mo e likely han men o wo k a he employe ’s o ices. On he o he hand, on Beh en e
al. (2020) ca ied ou an image-based clus e ing analysis o he indi iduals’ pa e n segmen a ion using he Ge man
Mobili y Panel o ac i i y (BMDV, 2020). The au ho s iden i y wo clus e s wi h child en in he household, one is
p edominan ly cha ac e ised by women and pa - ime wo ke s.
In he nex sec ions, we apply his heo e ic app oach o analyze he gende ed mobili y pa e ns in he Me opoli an
Region o Ba celona. To ou knowledge, his is he i s a emp o apply his echnique in he Spanish con ex ,
u he mo e by making use o a longi udinal da ase .
3. Case S udy
We apply his heo y in he Me opoli an Region o Ba celona (RMB). I is di ided in o 36 municipali ies in he
Me opoli an A ea o Ba celona (AMB), which accoun s o mo e han 3.2 Million inhabi an s. I has a well-sca e ed
public anspo a ion ne wo k wi h mo e han 200 bus lines, 10 me o lines, 15 ailways lines, and wo amway lines.
Mo e han 9 million ips a e made e e y day. The es o he RMB a ea consis s o 164 municipali ies and 1,848,514
inhabi an s. A de ailed a el demand modelling is egula ly conduc ed in he AMB suba ea .A case o special in e es
is he P ima y C own o he Me opoli an a ea ha includes he 18 mos popula ed municipali ies. Mo e in o ma ion
on he Me opoli an Region o Ba celona may be ound in (Mejía-Do an es e al., 2021).
4. Me hodology and Da a Desc ip ion
The en i e daily sequences o ac i i ies and a el pa e ns a e quan i ied by ou indica o s, de ined in McB ide e
al. (2019, 2020).
•No malized En opy, which explo es he a ie y in daily schedules
•Tu bulence, which shows he complexi y in daily schedules, and
•Complexi y, a no malized [0,1] sco e based on en opy, and di e en sequences o he indi idual’s schedule
•T a el Time Ra io (TTR): he o al a el ime in a day di ided by he sum o he o al ime ou side he home plus
he o al a el ime in a day
As explained by McB ide e al. (2020), hese summa y indica o s a e co ela ed o each o he . They quan i y daily
ac i i y- a el pa e ns o each indi idual in a nume ical way.
Lídia Mon e o e al. / T anspo a ion Resea ch P ocedia 82 (2025) 530–546 533
The s a is ical analysis is ca ied ou by he T aMineR package in R (Alexis Gabadinho e al., 2011). I has been
widely used o he analysis o biog aphical longi udinal da a in social sciences, bu o he app oaches ha e been also
desc ibed.
4.1. Me hodology
A sequence is de ined as a se ies o ime poin s a which a subjec can mo e om one disc e e ‘‘s a e’’ o ano he .
People wi h many s a es in hei daily schedule ha e agmen ed schedules. In his esea ch we used sequence analysis
o s a is ically analyze he agmen a ion o esponden s’ days using a minu e-by-minu e ime se ies, in which e e y
minu e o he day con ains a speci ic s a e o each pe son in he s udy. These “s a es” a e he places which indi iduals
isi du ing hei dia y day. Ac i i ies ini ially conside ed a e: home (H); wo k (W); casual (C) o no equen ly
isi ed places; o he (O) o equen ly isi ed places ha a e no he wo king place; and a el (T).
Among he echniques in he a el beha io ield ha can be used o measu e he du a ion o ac i i ies and
ansi ion a es om one ac i i y o he nex , we make use o en opy, u bulence and complexi y. En opy is he
p opo ion o o al ime spen in each s a e and he numbe o s a e ansi ions is no aken in o accoun (A Gabadinho
e al., 2011):
ℎ(𝑥𝑥)=ℎ(𝜋𝜋1…𝜋𝜋𝑆𝑆)=−∑𝜋𝜋𝑖𝑖𝑙𝑙𝑙𝑙𝑙𝑙(𝜋𝜋𝑖𝑖)
𝑆𝑆
𝑖𝑖=1
(1)
Whe e 𝜋𝜋𝑖𝑖is he p opo ion o occu ences o he i h s a e in he conside ed sequence. S is he numbe o po en ial
s a es, and 𝒙𝒙is he sequence de ined om minu e o minu e day ac i i ies.
The p opo ion o minu es alloca ed o each s a e du ing a day de ines he en opy indica o . Fo his measu e he
numbe changes o s a e is i ele an . I a pe son has no s a e change du ing he en i e day, his/he en opy would be
0. In con as , ha ing se e al s a es makes en opy o inc ease. The ange o possible alues depends on he numbe
o s a es and he maximum is alloca ed a sequences showing equal amoun o ime in each s a e. In ou case maximum
en opy is 1.61. A no malized en opy sco e is o en used consis ing on di iding en opy in o maximum en opy, hus
a 0 o 1 ange is ob ained.
The sequence u bulence is a measu e p oposed by Elzinga and Lie b oe (2007b) o measu ing schedule
complexi y in daily ac i ies. I is based on sequence pe manence and uses he numbe o dis inc subsequences ha
can be ex ac ed om he dis inc s a e sequence and he a iance o consecu i e ime poin s spen in a dis inc s a e.
Fo a sequence 𝒙𝒙, he o mula o 𝑇𝑇(𝒙𝒙) u bulence, as s a ed by McB ide e al. (2019) as ollows:
𝑇𝑇(𝑥𝑥)=𝑙𝑙𝑙𝑙𝑙𝑙2(𝜙𝜙(𝑥𝑥)𝑠𝑠𝑚𝑚𝑚𝑚𝑚𝑚
2+1
𝑠𝑠2+1 )
(2)
Whe e 𝜙𝜙(𝒙𝒙)is he o dis inc subsequences ha can be ex ac ed om he dis inc s a e sequence accoun ing on
ime p ecedence.
𝑠𝑠2is he a iance o he s a e du a ion
𝑠𝑠𝑚𝑚𝑚𝑚𝑚𝑚
2is he maximum a iance o be gi en based on he du a ion o he sequence and i is compu ed as
𝑠𝑠𝑚𝑚𝑚𝑚𝑚𝑚
2=(𝑛𝑛−1)(1−𝑡𝑡)2, whe e n-1 is he numbe o ansi ions in he sequence and 𝑡𝑡is he sequence
du a ion di ided by he numbe o dis inc s a es in he sequence.
𝐶𝐶(𝒙𝒙)=√(𝑛𝑛𝑡𝑡(𝒙𝒙)
(𝑙𝑙(𝒙𝒙)−1)ℎ(𝒙𝒙)
ℎ𝑚𝑚𝑚𝑚𝑚𝑚)
(3)
534 Lídia Mon e o e al. / T anspo a ion Resea ch P ocedia 82 (2025) 530–546
Complexi y C(x) includes
𝑛𝑛𝑛𝑛(𝒙𝒙)
as he numbe o dis inc ansi ions wi hin a sequence,
𝑙𝑙(𝒙𝒙)
is he leng h o he
sequence,
ℎ(𝒙𝒙)
is he en opy indica o and
ℎ
𝑚𝑚𝑚𝑚𝑚𝑚is he maximum en opy in he sample. This indica o will ha e a
alue be ween 0 and 1, wi h ze o co esponding o en opy ze o and no ansi ions (e.g., s aying a a single place o
he en i e day o obse a ion).
The a el ime a io (TTR), de ined by McB ide e al. (2020) as a compac indica o ha ep esen s he ade-o s
o people be ween a el and ac i i y ime. In his pape , TTR is de ined as ‘ he o al a el ime in a day di ided by
he sum o he o al ime a home plus he o al a el ime in a day’. Thus, TTR anges om 0.5 (no ip-make s) o
1.0 (whole day ou o home).
To achie e a be e unde s anding o he beha iou o he u bulence and i s dependency on indi idual ea u es like
gende , age g oup, ac i i y, handicapped s a us,W H possibili y, day o he week and mac ozone in he a ea o
Ba celona, along wi h di e en desc ip i e a iables including he pe cen age o households o e 100m2, numbe o
bus s ops, numbe o me o, am and ain s ops, a linea model o hei loga i hms has been applied o he 2018-2020
subsample o ip make s.
La e we compa e all sequences wi h each o he o add ess sequence dissimila i y and compu e pai wise mean
agmen a ion indica o s by yea using Tukey’s Hones ly Signi ican Di e ence (HSD) Tes (Tukey e al., 1984).
Finally, we use a clus e ing echnique o g oup sequences o ac i i ies wi h simila dissimila i y sco es. The inal
numbe o clus e s is op imized o ep esen he da a using a c i e ion o wi hin-g oup simila i y and ac oss g oups
dissimila i y.
4.2. Da a desc ip ion
The me hodology desc ibed in Sec ion 4.1 has been applied o he da a om EMEF (Weekday Mobili y Su ey, in
Ca alan) su eys o 2018, 2019, and 2020 o de ine he sequences o isi ed places by a pe son du ing a day join ly
wi h he du a ion o ac i i ies a each place and he a el imes o each hese places. They a e adi ional mobili y
su eys ha analyze he mobili y o esiden s in he Me opoli an Region o Ba celona (RMB) o indi iduals aged
16 and o e . The spa ial g anula i y is a he municipali y le el, bu as Ba celona is di ided in o en dis ic s, i leads
o a o al o 296 mac o-zones, whe e only 45 o hem a e in he AMB a ea. The EMEF 2020 su ey was launched
du ing he Fall o 2020 when some mobili y es ic ions and he p e alence o some online ac i i ies we e s ill p esen
due o he COVID-19 si ua ion. The EMEF2020 dese es a pa icula analysis since ac i i y pa e ns a e expec ed o
a y compa ed o he yea s be o e he COVID-19 ou b eak.
The da a collec ed o each jou ney e e s o ips o he day be o e: o igin and des ina ion (mac o-zones), pu pose,
mode (a e y de ailed lis o possibili ies), a el s a ime and du a ion (min), ehicle use, pa king use, e c. The
sample uni s a e indi idual esiden s, no households. The sample size o each yea a e emo ing he ca ego y
“p o essional d i e s” a e: 9,930; 9,934; and 10,024, espec i ely o 2018 o 2020 in he AMB a ea. The o al ips
a e 36,368; 37,463; and, 30,591, espec i ely o 2018 o 2020 in he RMB a ea. A e il e ing (p o essional a elle s,
esiden ial a ea missings, e c.), he inal numbe o esiden s included in he o al sample was ound o be 26,860.
We also make use o demog aphic and land use da a.This in o ma ion is de ined wi h he same spa ial g anula i y
de ined by he EMEF su eys. I consis s on popula ion segmen ed by gende (only men and women), age g oup (5
g oups), educa ion le el, educa ional places, se ices, land use, esiden ial mo phology, a e age pe capi a en , and
numbe o s ops in he public ne wo k.
5. Resul s
A i s analysis was pe o med by simpli ying he kind o ac i i ies ha people ca y ou , o places ha people isi .
The ac i i ies ini ially conside ed a e home (H); wo k (W); casual (C) o no equen ly isi ed places; o he (O) o
equen ly isi ed places ha a e no he wo king place; and a el (T). The nex Fig. 1shows as an example some
sequences o ac i i ies and du a ions o he 3 i s elemen s o ou sample and some de ined sco es.
The i s wo pa e ns belong o male pe sons coming om a place ou o Ba celona, who li e in he i s dis ic o
Ba celona ci y. They ha e nei he a ca , no a mo o bike.The hi d pa e n belongs o a s uden who no mally uses
Lídia Mon e o e al. / T anspo a ion Resea ch P ocedia 82 (2025) 530–546 535
public anspo o walks. The ac i i y se is de ined as he ac i i y alphabe and, he ein, o illus a ion pu poses, se
up as casual (C), home (H), o he (O), a el (T), and wo k (W).
Fig. 1. G aphical ep esen a ion o daily- a el pa e n o 3 uni s o he wo king sample.
Fo ilus a ion pu poses, he ac i i y dis ibu ion ac oss he day in a subse o he wo king sample is shown in Fig.
2. An in e p e a ion o his igu e indica es a 13:00h, 3% o he sample uni s show a ecu en ac i i y, 32% s ays a
home, 14% a e in ol ed in o he ac i i y, 9% is a eling, and he es , 58%, is wo king.
Fig. 2. Ac i i y Dis ibu ion by ime.
Fig. 3shows no malized en opy, u bulence, and complexi y dis ibu ions use ul o he gene a ion o syn he ic
536 Lídia Mon e o e al. / T anspo a ion Resea ch P ocedia 82 (2025) 530–546
popula ions (anonymized popula ion ha is ep esen a i e o eal popula ion numbe s), whe e some ques ions a ise.
Fo example, why has he no malized en opy a maximum abou 0.3-0.4 co esponding o daily pa e n (H-T-W-T-H-
T-O-T-H 450-10-740-10-110-20-280-40), and he 0 alue co esponds o non- ip-make s, he same high ba a ze o
is seen o u bulence and complexi y his og ams.
Fig. 3No malized en opy, Tu bulence and Complexi y dis ibu ions. EMEF 2018 o 2020 da a.
To achie e a be e unde s anding o he beha iou o he u bulence and i s dependency on he indi idual ea u es,
a linea model o hei loga i hms has been applied o he 2018-2020 EMEF subsample o ip make s. These indi idual
cha ac e is ics include gende , age g oup, ac i i y, disabili y s a us, W H possibili y (wi h ega ds o yea 2020), day
o he week, and mac o-zone along wi h desc ip i e a iables such as pe cen age o households o e 100 m2, numbe
o bus s ops, numbe o me os, am, and ain s a ions. The ma ginal e ec o he a iables ha ing signi ican ne -
e ec s is shown in Fig. 4The esul s show ha he agmen a ion o emale ou -o -home ac i i ies is signi ican ly
g ea e han hose o men in he 30-44 age g oup. A ema kable e ec o yea 2020 on he loga i hm o he u bulence
is seen once con olled by he o he conside ed ac o s.
Fig. 4. Tu bulence ans o ma ion ma ginal e ec o 2018-2020 EMEF (only ip make s).
Lídia Mon e o e al. / T anspo a ion Resea ch P ocedia 82 (2025) 530–546 537
Fu he mo e, Fig. 5shows he loga i hm o he u bulence acco ding o gende , age g oup, and yea , once educa ion,
bi hplace, disabili y s a us, and W H possibili ies a e accoun ed o in he model. Women in he 30-44 age g oup
show he highes u bulence in any yea ; while men u bulences seem s able be ween age 16 o 64 g oups in 2018 and
2019, he younges g oups, ei he men’s o women a e a ec ed by a se e e dec ease in agmen a ion in 2020. In
gene al e ms, he 2020 u bulences a e educed a e he COVID-19 b eakou .
Fig. 5. Ma ginal e ec o he u bulence ans o ma ion wi h ega ds o yea , gende , and age-g oup in e ac ion in he o al 2018-2020 EMEF
da ase (only ip make s) a e con olling by educa ion, o igin coun y, handicapped and W H e ec s.
538 Lídia Mon e o e al. / T anspo a ion Resea ch P ocedia 82 (2025) 530–546
Fig. 6. No malized En opy ma ginal e ec s (uppe igu e) and T a el Time Ra e (TTR, lowe igu e) o gende and age-g oup and yea
in e ac ion in 2018-2020 EMEF subsample o ip-make s a e con olling by yea , educa ion, ac i i y, coun y o o igin, disabili y, and wo king-
om-home e ec s.
Gi en he na u e o he esul s and he no jus i ica ion o any no mali y hypo hesis, u he analysis o ge insigh s
on hei beha iou we e conduc ed eso ing o nonpa ame ic s a is ical me hods, as o example K uskal-Wallis es
o homogenei y in means ac oss g oups. The Tukey HSD es has been employed o add ess pai wise compa isons
be ween means in g oups.
Conside ing he (no malized) en opy indica o , he g oss e ec o gende is signi ican acco ding o K uskal-
Wallis non-pa ame ic es o means depending on yea a 95% con idence le el. The male en opy is g ea e han
he emale en opy being he highes alue a he younges age g oup (see Fig. 6).
5.1. Sequence analysis wi h a new alphabe o ac i i ies
A ede ined se o ac i i ies based on he ypes o places indi iduals isi du ing hei dia y day we e used o
u he analyses: Home (H), Wo k (W), School (only o s uden s, S), Casual ( o no equen ly isi ed places, C),
and O he s ( equen ly isi ed places ha a e no he wo king place, O). Fu he mo e, he alphabe should and mus
be ex ended and e ined since ac i i ies like Esco ing/Accompanying (A) can be conside ed, as well as he
di e en ia ion o ac i i ies by a el mode since hei da a a e a ailable. T a el ac i i ies a e di ided in o TW o
ac i e modes (walking, cycling), TC o p i a e anspo (ca , an, mo o cycle), TP o public anspo (bus, me o,
am, and ain), and TM o o he anspo modes, such as an and uck. An alphabe o 10 ac i i ies is conside ed.
Ei he ip make s o non- ip make s a e included and a sequence o 1440 minu es (o mo e) wi h each minu e
classi ied as a ca ego y o he alphabe . Fig. 7shows he e ined daily ac i i y pa e n dis ibu ion ac oss yea s. The
2020 s a e ac i i y pa e n dis ibu ion shows an o e ep esen a ion o whole day home ac i i ies.
The pe cen ages o ac i i y dis ibu ion ac oss he day o men and women and by yea a e shown in Table 1. Wo k
ac i i ies ou o home a e g ea e o men han women, school ac i i y is simila in bo h cases. On he o he hand,
home ac i i y in 2020 (73.5%) inc eased by almos 13 poin s compa ed o p e ious yea s, while school and wo k
ac i i ies ou o home a e clea ly educed. The pe cen age o p i a e anspo use is se e ely a ec ed by esiden ial
a ea (analysed he e as c owns), showing ha in non-cen al c owns he e is an inc ease in p i a e anspo ac i i y
(3.4%) compa ed o Ba celona ci y (1.5%); consequen ly, public anspo in he cen al c own as Ba celona ci y
shows 2.8% incidence, while ex e nal c owns lie be ween 1.2-1.4% incidence. The spa ial e ec s mus be add essed
in ac i i y analysis (Table 1).
Lídia Mon e o e al. / T anspo a ion Resea ch P ocedia 82 (2025) 530–546 545
In his pape an ac i i y analysis ac oss days in he RMB a ea has been add essed. Focusing on agmen a ion
indica o s and he po en ial connec ion o modal p e e ences in one o he g ea es Eu opean ci ies such as Ba celona
is no ably ele an .
Ou esea ch analysis is ocused in he Me opoli an A ea o Ba celona, whe e he spa ial and empo al beha iou
o anspo demand is a he di e en om o he case s udies using he same me hodology in e ms o he unde lying
socioeconomic eali y, he anspo -o ien ed de elopmen (TOD) which can be seen h ough he ich public anspo
ne wo k, he u ban s uc u e, and di e en ac i i y al e na i es and se ices. A sequence analysis is used in his pape
o measu e agmen a ion in ac i i y pa icipa ion and a el.
S udying sequences o daily ac i i ies (each ac i i y a a place and each ip) includes he en i e ajec o y o a
pe son’s ac i i y du ing a day while join ly conside ing he numbe o ac i i ies and ips, hei o de ing, and hei
du a ions. The complexi y o he da a esul ing om agmen a ion in ha case led us o eso i o make use o an
OD dimensionali y educ ion and clus e ing echniques o conduc he analysis. The clus e ing analysis shows clea
di e ences be ween clus e s o e ep esen ed by women and by men.
Fu he mo e, as i has al eady been highligh ed in Table 1, ac i i y dis ibu ions by yea , gende and geog aphic
esiden ial a ea show signi ica i e di e ences, as o ins ance he wo k ac i i y p opo ion in men is highe han in
women. Fu he mo e, he pe cen age o p i a e anspo use is no ably in luenced by esiden ial a ea, showing ha in
non-cen al c owns he e is an inc ease o p i a e anspo ac i i y compa ed o Ba celona ci y. On he o he side,
public anspo in he cen al c own as Ba celona ci y shows a highe incidence, compa ed o he ex e nal c owns.
This s udy makes e iden , ha i is impossible o disen angle he ole o sus ainable anspo wi h he buil
en i onmen and dwelling loca ion.
This s udy has also e ealed some beha io s ha o be p ope ly unde s ood equi e a deepe analysis, and likely
mo e de ailed da a no a ailable in his case. An example o his would be ha o he bimodali y o he u bulence.
This esea ch lea es many doo s open o u u e analysis. Fo example, a deepe analysis o all clus e s.
Addi ionally, ega ding he in luence o he u ban en i onmen , which clea ly dese es a mo e de ailed analysis a
ine spa ial scale.
Acknowledgemen s
This esea ch was unded by PID2020-112967GB-C31 Spanish R+D P og ams and by Sec e a ia d’Uni e si a s-i-
Rece ca-Gene ali a de Ca alunya-2017-SGR-1749. The da ase s we e kindly sha ed by he Au o i a del T anspo
Me opoli à (ATM). Thei con ibu ion o ou esea ch is g a e ully acknowledged.
Re e ences
Abbo , A., Fo es , J., 1986. Op imal Ma ching Me hods o His o ical Sequences. Jou nal o In e disciplina y His o y 16, 471.
h ps://doi.o g/10.2307/204500
Ahmed, U., Mo eno, A.T., Moeckel, R., 2021. Mic oscopic ac i i y sequence gene a ion: a mul iple co espondence analysis o explain a el
beha io based on socio-demog aphic pe son a ibu es. T anspo a ion 48, 1481–1502. h ps://doi.o g/10.1007/S11116-020-10103-
1/FIGURES/6
Alexande , B., E ema, D., Dijs , M., 2010. F agmen a ion o wo k ac i i y as a mul i-dimensional cons uc and i s associa ion wi h ICT,
employmen and sociodemog aphic cha ac e is ics. Jou nal o T anspo Geog aphy 18, 55–64.
h ps://doi.o g/10.1016/j.j angeo.2009.05.010
Bha , Chand a R, Kons adinos, •, Goulias, G., Pendyala, Ram M, Pale i, Rajesh, Sidha han, Raghup asad, Schmi , Lau a, Hu, Hsi-Hwa, Bha , C
R, Sidha han, Á.R., Sidha han, R, Goulias, K.G., Pendyala, R M, Pale i, R, B incke ho , P., Plaza, P., Schmi , L, Hu, H.-H, 2013.
A household-le el ac i i y pa e n gene a ion model wi h an applica ion o Sou he n Cali o nia. T anspo a ion 2013 40:5 40, 1063–
1086. h ps://doi.o g/10.1007/S11116-013-9452-Y
Bha , C.R., Pinja i, A.R., 2007. Du a ion Modeling, in: Eme ald G oup Publishing (Ed.), Handbook o T anspo Modelling. pp. 105–131.
h ps://doi.o g/10.1108/9780857245670-006
BMDV, 2020. Mobili y in Ge many [WWW Documen ]. Mobili y in Ge many. URL h ps://www.bm i.de/EN/Se ices/S a is ics/Mobili y -in-
Ge many/mobili y-in-ge many.h ml (accessed 7.15.22).
Bu chell, B., Reuschke, D., Zhang, M., 2020. Spa ial and empo al segmen ing o u ban wo kplaces: The gende ing o mul i-loca ional wo king:
h ps://doi.o g/10.1177/0042098020903248 58, 2207–2232. h ps://doi.o g/10.1177/0042098020903248
Couclelis, H., 2006. Pizza o e he In e ne : e-comme ce, he agmen a ion o ac i i y and he y anny o he egion. En ep eneu ship &
Regional De elopmen 16, 41–54. h ps://doi.o g/10.1080/0898562042000205027
C esswell, T., U eng, T.P., 2008. Gende ed mobili ies: owa ds an holis ic unde s anding. Gende ed mobili ies 1–12.
Cubells, J., Ma que , O., Mi alles-Guasch, C., 2020. Gende and Age Di e ences in Me opoli an Ca Use. Recen Gende Gap T ends in P i a e
546 Lídia Mon e o e al. / T anspo a ion Resea ch P ocedia 82 (2025) 530–546
T anspo . Sus ainabili y 12, 7286. h ps://doi.o g/10.3390/su12187286
EIGE, 2022. Employmen [WWW Documen ]. Eu opean Ins i u e o Gende Equali y. URL h ps://eige.eu opa.eu/gende -mains eaming/policy-
a eas/employmen (accessed 2.15.22).
EIGE, 2019. Gende S a is ics | Gende S a is ics Da abase [WWW Documen ]. Eu opean Ins i u e o Gende Equali y. URL
h ps://eige.eu opa.eu/gende -s a is ics/dgs/da a- alks/wha -lies-behind-gende -pay-gap (accessed 2.23.22).
Elzinga, C.H., Lie b oe , A.C., 2007a. De-s anda diza ion o Family-Li e T ajec o ies o Young Adul s: A C oss-Na ional Compa ison Using
Sequence Analysis: Dé-s anda disa ion des ajec oi es de ie amiliale des jeunes adul es: compa aison en e pays pa analyse
séquen ielle. Eu J Popula ion 23, 225–250. h ps://doi.o g/10.1007/s10680-007-9133-7
Elzinga, C.H., Lie b oe , A.C., 2007b. De-s anda diza ion o amily-li e ajec o ies o young adul s: A c oss-na ional compa ison using sequence
analysis. Eu opean Jou nal o Popula ion 23, 225–250. h ps://doi.o g/10.1007/S10680-007-9133-7/FIGURES/2
Eu os a , 2020. How usual is i o wo k om home? [WWW Documen ]. Eu os a Eu opean Commission. URL
h ps://ec.eu opa.eu/eu os a /web/p oduc s-eu os a -news/-/ddn-20200206-1 (accessed 7.1.22).
EWCS, 2015. Six h Eu opean Wo king Condi ions Su ey: 2015 [WWW Documen ]. Eu o ound. URL
h ps://www.eu o ound.eu opa.eu/su eys/eu opean-wo king-condi ions-su eys/six h-eu opean-wo king-condi ions-su ey-2015
(accessed 7.15.22).
Fana, M., Tolan, S., To ejón, S., U zi B anca i, C., & Fe nández-Macías, E., 2020. The COVID con inemen measu es and EU labou ma ke s.
Publica ions O ice Eu opean Commission. Join Resea ch Cen e., LU.
Gabadinho, Alexis, Ri scha d, G., S ude , M., Mulle , N.S., 2011. Mining sequence da a in R wi h he T aMineR package: A use ’s guide (Use ’s
guide). Uni e si y o Gene a, Swi ze land.
Gabadinho, A, Ri scha d, G., S ude , M., Mülle , N.S., 2011. Mining sequence da a in R wi h he T aMineR package: A use s gu ide.
Giele, J.Z., Elde , G.H., 1998. Li e Cou se Resea ch: De elopmen o a Field, in: Me hods o Li e Cou se Resea ch: Quali a i e and Quan i a i e
App oaches. SAGE Publica ions, Inc., 2455 Telle Road, Thousand Oaks Cali o nia 91320 Uni ed S a es, pp. 5–27.
h ps://doi.o g/10.4135/9781483348919.n1
Hanson, S., 2010. Gende and mobili y: new app oaches o in o ming sus ainabili y. Gende , Place & Cul u e 17, 5–23.
h ps://doi.o g/10.1080/09663690903498225
Hube s, C., Schwanen, T., Dijs , M., 2008. ICT and Tempo al F agmen a ion o Ac i i ies: An Analy ical F amewo k and Ini ial Empi ical
Findings. Tijdsch i oo economische en sociale geog a ie 99, 528–546. h ps://doi.o g/10.1111/j.1467-9663.2008.00490.x
Hyde, E., G eene, M.E., Da ms ad , G.L., 2020. Time po e y: Obs acle o women’s human igh s, heal h and sus ainable de elopmen . Jou nal
o Global Heal h 10, 020313. h ps://doi.o g/10.7189/jogh.10.020313
Ins i u o Nacional de Es adís ica, 2021. Su ey on Equipmen and Use o In o ma ion and Communica ion Technologies in Households 2021
(Encues a sob e equipamien o y uso de ecnologías de in o mación y comunicación en los hoga es 2021) [WWW Documen ]. URL
h ps://ine.es/dyn 3/inebase/index.h m?pad e=8320&capsel=8339 (accessed 7.1.22).
Leszczyc, P.T.L.P., Timme mans, H., 2002. Uncondi ional and condi ional compe ing isk models o ac i i y du a ion and ac i i y sequencing
decisions: An empi ical compa ison. Jou nal o Geog aphical Sys ems 4, 157–170. h ps://doi.o g/10.1007/S101090200083
McB ide, E., Da is, A., Goulias, K., 2020. Explo a ion o S a ewide F agmen a ion o Ac i i y and T a el and a Taxonomy o Daily Time Use
Pa e ns using Sequence Analysis in Cali o nia. T anspo a ion Resea ch Reco d 2674, 38–51.
h ps://doi.o g/10.1177/0361198120946011
McB ide, E., Da is, A., Goulias, K., 2019. F agmen a ion in Daily Schedule o Ac i i ies using Ac i i y Sequences: T anspo a ion Resea c
Reco d 2673, 844–854. h ps://doi.o g/10.1177/0361198119837501
McB ide, E.C., Da is, A.W., Goulias, K.G., 2020. Explo a ion o S a ewide F agmen a ion o Ac i i y and T a el and a Taxonomy o Daily
Time Use Pa e ns using Sequence Analysis in Cali o nia. T anspo a ion Resea ch Reco d 2674, 38–51.
h ps://doi.o g/10.1177/0361198120946011
McB ide, E.C., Da is, A.W., Goulias, K.G., 2019. F agmen a ion in Daily Schedule o Ac i i ies using Ac i i y Sequences. T an spo a ion
Resea ch Reco d 2673, 844–854. h ps://doi.o g/10.1177/0361198119837501
Mejía-Do an es, L., Mon e o, L., Ba celó, J., 2021. Mobili y T ends be o e and a e he Pandemic Ou b eak: Analyzing he Me opoli an A ea o
Ba celona h ough he Lens o Equali y and Sus ainabili y. Sus ainabili y 13, 7908. h ps://doi.o g/10.3390/su13147908
Pale i, R., Vo sha, P., Vyas, G., Ande son, R., Giaimo, G., 2017. Ac i i y sequencing, loca ion, and o ma ion o indi idual non-manda o y
ou s: applica ion o he ac i i y-based models o Columbus, Cincinna i, and Cle eland, OH. T anspo a ion 44, 615–640.
h ps://doi.o g/10.1007/S11116-015-9671-5/FIGURES/2
Rasouli, S., Timme mans, H., 2014. Ac i i y-based models o a el demand: p omises, p og ess and p ospec s. In e na ional Jou nal o U ban
Sciences 18, 31–60. h ps://doi.o g/10.1080/12265934.2013.835118
Shi, H., Su, R., Xiao, J., Goulias, K.G., 2022. Spa io empo al analysis o ac i i y- a el agmen a ion based on spa ial clus e ing and sequence
analysis. Jou nal o T anspo Geog aphy 102, 103382. h ps://doi.o g/10.1016/J.JTRANGEO.2022.103382
So o Villag án, P., Mejía Do an es, L., 2022. U ban P ojec s and Public Space in La in Ame ica: Walking Towa ds a Pa adigm o Gende
Equali y (Chap e 23), in: The Rou ledge Handbook o U ban S udies In La in Ame ica And The Ca ibbean. Rou ledge.
S ude , M., Ri scha d, G., 2016. Wha ma e s in di e ences be ween li e ajec o ies: a compa a i e e iew o sequence diss imila i y measu es
on JSTOR. Jou nal o he Royal S a is ical Socie y A 179, 481–511.
Tukey, J.W. (John W., Tukey, J.W. (John W., B illinge , D.R., Cox, D.R. (Da id R., B aun, H.I., 1984. The collec ed wo ks o John W. Tukey.
Wadswo h Ad anced Books & So wa e, Belmon Cali .
on Beh en, S., Hilge , T., Ki chne , S., Chlond, B., Vo isch, P., 2020. Image-based ac i i y pa e n segmen a ion using longi udinal da a o he
Ge man Mobili y Panel. T anspo a ion Resea ch In e disciplina y Pe spec i es 8, 100264. h ps://doi.o g/10.1016/j. ip.2020.100264
