Co esponding au ho : Ahlem HAMRI.
Copy igh © 2025 Au ho (s) e ain he copy igh o his a icle. This a icle is published unde he e ms o he C ea i e Commons A ibu ion Liscense 4.0.
The selec ion o he logis ics pla o m’s loca ion based on ELECTRE
Ahlem Ham i * and Ra aa M aihi
Depa men o economics. Highe School o Business o Tunis. Manouba 2010. Tunisia.
Wo ld Jou nal o Ad anced Resea ch and Re iews. 2025, 27(02), 2157-2169
Publica ion his o y: Recei ed on 27 Ap il 2025; e ised on 25 Augus 2025; accep ed on 29 Augus 2025
A icle DOI: h ps://doi.o g/10.30574/wja .2025.27.2.1703
Abs ac
This pape ocuses on inding he mos sui able loca ion o he logis ics pla o ms in Tunisia. S a egic global logis ics
pla o m’s loca ion decisions equi e se e al ac o s ha can be con lic ing in na u e. and may pose a di icul anking
p oblem. This pape uses ELECTRE III. Based on his mul ic i e ia decision making me hod. egions a e anked om he
bes o he wo s o he loca ion o logis ics pla o ms acco ding o a se o al e na i es and a amily o c i e ia and hei
e alua ion h ough decision make ’s p e e ences.
Keywo ds: Logis ics Pla o ms; Ranking P oblem; Mul ic i e ia Decision Making Me hod; ELECTRE III
1. In oduc ion
Logis ics in as uc u e can be de ined as a pi o al ool o low managemen . i allows s o age. so ing. consolida ion
and bu s o p oduc s lows [1]. Bu acco ding o he con ex , his wo d e e s o a speci ic equipmen (Wa ehouse
so ing cen e ) o an ac i i y a ea eg ouping se e al equipmen s [25]. In ac . i can be a logis ic si e which is a speci ic
logis ic ac i i y a ea ha may be named pa k o pla o m; o a logis ic hub which consolida es se e al logis ic si es in
he same pe ime e and inally. i is named logis ic a ea in a wide geog aphic scale. On a global scale. and in Eu ope [8.
25.14] no e a big pola iza ion o logis ic implan a ions which a e condensed in space. Howe e . his is a gene al
phenomenon o all logis ic in as uc u e ypes. Joignaux [10] speaks abou a double dimension o logis ic
in as uc u es: In as uc u al dimension and se ices dimension. In ac . he i s one e e s o ma e ials and physical
dimension which is ela ed o in es men s. Logis ic in as uc u es a e a suppo o de elopmen o p oduc ion and
se ices ac i i ies. Pe a and Zimme man [18] explain ha he aim o public au ho i ies in no a ac ing in es men s
bu o ensu e hei sus ainabili y.
The selec ion o op imal logis ic in as uc u e has become a decisi e ac o in supply chain pe o mance [2]. Indeed,
public au ho i ies a e in e es ed o hose logis ic implan a ions because hey ep esen an essen ial objec o
de elopmen and spa ial planning [1, 19, 24, 17] and should be ela ed o i m’s de elopmen s a egies [34].
2. Li e a u e e iew
The analysis’s esul s o Me enne-Schoumake [15] show ha logis ic sec o loca ion is based on bo h indus ial and
se ices loca ion logic. Indeed. hey ci e ou ac o s which close logis ic o he indus y: spaces buil and no buil .
na ow links wi h he anspo in as uc u e (Mos ly Road. ailway and ma i ime anspo ). and skilled labo
(no ably o manual asks)) and nuisances caused (noise. a ic. pollu ions…). The e is Two o he ac o s b ing logis ics
close o se ice: he ela ionship wi h cus ome s and he weigh o he eal es a e in loca ion choice. Howe e . hey
poin ou logis ic as a speci ic sec o which wo h a ull place in he economic ac i i ies loca ion. We can add he indings
o Hesse-Rod igue [8]. in one hand. bes accessibili y and possible nea by wi h cus ome s. in o he hand. low cos o
ins alla ion. Indus ial loca ion models appea ed in he 19 h cen u y. hey aimed o explain economic o ganiza ion in
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 27(02), 2157-2169
2158
space agains anspo a ion cos . Von Thünen model [31]. which co e s g ound en o ag icul u al c ops; al hough i
looks simple, we can use i in land use o a eas o economic ac i i y (indus ial a eas a e included). Also. Al ed Webe
model [32]is in e es ed in he op imal loca ion o p oduc ion ac i i ies. In ac . Webe [32] inds ha he anspo is
he base o he model because i plays a p ima y ole in he implan a ion choice. The bes loca ion minimizes supply
bu den o p ima y ma e ials. ene gy and inished p oduc expedi ion.
Lösch [13] buil simila models o explain he loca ion choice o comme cial and adminis a i e se ices. Theo e ical
models o Ch is alle and Lösch. a e known as « cen al loca ions’ heo y » The ansposi ion o his model o he Logis ic
pla o m loca ion mus include anspo a ion cos ups eam and downs eam. Ho elling models [13] a e models o
spa ial compe i ion which de ine equilib ium loca ion o many companies in compe i ion. when demand is spa ially
dispe sed. Also. company loca ion models and spa ial compe i ion models a e ools ha enligh en logis ic ac i i ies
loca ion choice.
Owen and Daskin [16]; Fa ahani and Hekma a [6] de ine he loca ion p oblem as a way o ind a co ec place o an
in as uc u e in a s udied a ea. and we should conside wo essen ial elemen s: he app op ia e cons ain s and
decision make ’s (DM’s) p e e ences. In addi ion. he loca ion p oblem can be applied in di e en ields and can conce n
di e en ca ego ies o si es/ acili ies [5].
Se e al me hods and app oaches ha e been used in o de o sol e his ques ion. including: ma hema ical model [7].
selec ion and anking o conside ed si es [11, 21]. Addi ionally. he loca ion p oblem has an impo an con ibu ion in
logis ics. whe e i e e s o anking o selec ing he mos desi able loca ion o logis ics si es. The ques ion o logis ics
loca ion is conside ed as a ypical p oblem wi h he de elopmen o logis ics in he wo ld and i e e s o choose he
app op ia e placemen o logis ics pla o ms which a e sepa a e en i ies wo king in a secu ed a ea. wi hin which all
logis ics ac i i ies ( anspo a ion and o wa ding. ma e ial handling. wa ehousing. in en o y managemen . c oss-
docking. in e modal ans-shipmen . physical dis ibu ion o goods) a e execu ed on a comme cial basis. In he mode n
concep s o 3PL and 4PL [24] logis ic pla o m plays a c ucial ole o logis ics se ice in eg a o ha p o ides cus ome s
wi h a comp ehensi e. high-quali y se ice and links all possible anspo a ion modes.
Finally. Sa y [25] and Joignaux [10] ag ee on he impo ance o logis ic ac i i ies loca ion which is a p imo dial ac o
in indus ial loca ion in 19 h cen u y and p obably an impo an elemen o e i o y o ganiza ion in he 21 h cen u y.
So. which ac o s de e mine loca ion logis ic ac i i ies?
2.1. Logis ic ac i i ies: loca ion ac o s
The implemen a ion o a logis ics si e is a complex decision whe e he pa ame e s o conside a e nume ous. In ac .
p o essionals ci e i e ac o s which de e mine Logis ic pla o m loca ions: ma ke . accessibili y and in as uc u es
a ailabili y. g ound cos and ole o public au ho i ies. These ac o s a e con i med by he s udies o majo in e na ional
consul an s as Buck [4] and labo agencies whe e Cushman and Wake ield s udies he “ op” loca ions o logis ic cen e s.
In 2004. he s udy includes 34 egions NUTS 2 and 61 egions in 2006. Howe e . hese s udy. is bases on a mul ic i e ia
analysis which includes six ields (cos . anspo sys em. accessibili y. buildings and e ains. labo and Know-how)
and 19 c i e ia). Me enne-Schoumake [15] quo ed i e c i e ia o anking he 15 i s egions NUTS2 o no hwes e n
Eu ope on he basis o logis ic pe o mances: p oximi y o ma ke . accessibili y o loca ion and a ailabili y
in as uc u e. A ailabili y and cos o land and building (wa ehouse exceed 10 000 m2 and may o m hubs o 60000 o
80000 m2). A ailabili y. cos and quali ica ions o he labo and inally he ole o public au ho i y.
Żak and Węgliński [33] ha e p oposed a se o 9 c i e ia in o de o selec he mos desi able a ea/ egion o placing
he logis ic pla o m on Poland which a e (Condi ion o anspo a ion in as uc u e – [km/100km2]. Economic
de elopmen . In es men cos – [$/m2]. Le el o anspo a ion and logis ics compe i i eness – [%]. In es men
a ac i eness – [km2]. T anspo a ion and logis ics a ac i eness – [mln km]. Social a ac i eness – [poin s].
En i onmen al- iendliness – [poin s]. and Sa e y and secu i y – [poin s] . Issam and Bouzidi [9] p opose 15 c i e ia in
o de o judge he app op ia eness o indus ial acili ies loca ion demons a ing he decisi e cha ac e o logis ics in
he indus ial posi ioning o « CASABLANCA-SETTAT »: he size. accessibili y. Cons uc abili y. land managemen .
egula ion and se i ude. en i onmen al impac . impac on landscape. u ban impac . si e secu i y possibili y. showcase
E ec . posi ioning. p oximi y o indus ial de elopmen . P oximi y o esiden ial and equipmen s. In e es om
in es o s. ma ke se ices. On he o he hand. Tomić [27] s udies he loca ion o logis ic cen e s in he Balkan Peninsula.
Indeed, he de ines six c i e ia acco ding o lows ypes (physical lows. economic lows. ins i u ional o owne ship lows.
goods lows. in o ma ion lows and o he lows). Then, Masson & Pe io [14] con i m ha es ablishing a si e is a complex
decision because i equi es nume ous se ings and hey se eigh loca ion ac o s: accessibili y and anspo cos .
access o ma ke . loca ion cos s. cos and a ailabili y o ac o s o p oduc ion. poli ic and egula o y ac o s. ac o s
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 27(02), 2157-2169
2159
ela ed o logis ic chain o ganiza ion. ex e nali ies and e i o ial asse s. In addi ion. Uysal and Ya uz [29] ca ied ou
an e alua ion o logis ics cen e loca ion in Tu key acco ding o en c i e ia: closeness o ai po and po . p oximi y o
esiden ial a eas. accessibili y o labo . en i onmen secu i y. accessibili y o highways and linking oads. subs uc u e.
a ic densi y. ai pollu ion. building si e.
Lee e al. [12] mixed quan i a i e and quali a i e c i e ia. ela ed o ma ke . labo . egional de elopmen policies. ax.
and legisla ion. Ren e al. [22] and Tu e al. [28] ake in o conside a ion he a ailabili y o suppo se ices (ene gy
p o ision and was e managemen ). Finally. S ale [26] s udies a loca ion o a sample o logis ic companies ( he
No hwes e n Eu opean u ban egions) in o de o ind loca ion ac o s and he concludes. in one hand. ha spa ial
dis ibu ion o logis ic ac i i ies s eng hens end o he es o economic ac i i y and in o he hand. i is a classic
economic geog aphy which is based on classic ac o s (p oximi y o he ma ke . op imal accessibili y condi ions and
easonable land p ices) which shows he need o well managing logis ic ac i i ies dis ibu ion and well choosing
loca ion ac o s.
3. Tunisia: E iciency o logis ic sec o
Wha e e he coun y. in he wo ld. decision make s. deal wi h egional inequali ies. Tunisia. su e s om his p oblem.
56% o popula ion and 92% o indus ial companies concen a ed on h ee big ci ies: Tunis. S ax. and Sousse. Those
ci ies ep esen ing 85% o GPD [20]. Howe e . hose egional inequali ies a e caused by economic poli ics. In ac . he e
was a egional imbalance in Tunisia (in equipmen s and in as uc u es) since he colonial pe iod and i becomes clea e
a e independence and his has an e ec on egional e i o ial a ac i eness. In ac . egions o he eas coas a ac ed
mo e in es men s han he o he s egions o he coun y. Thus. hey ha e mo e lows o se ices and goods which c ea e
as well as zones o in luence. Howe e . he o he egions we e o go en which leads o imbalance in weal h. Fo his
pu pose. a big weal h is concen a ed on he coas : whe e 68% o popula ion. 75% o added alue. 95% o indus ial
and ou is ic economy. On he o he hand. he o he egions inside he coun y (Jendouba. Sidi Bouzid. Siliana e
Kasse ine) ha e 33% o popula ion and a e poo because hey ha e he lowes le el o de elopmen , his is explained
by hei employmen and weal h le els: high unemploymen le el (up o 29% o kasse ine). low densi y o business
(0.2 agains 3.1 o Tunis). A high le el o po e y (27.5% o Sidi Bouzid agains 6.9% o Tunis) [3]. In ac . Tunisia
has 605 000 unemployed pe sons in he i s qua e o 2014. Unemploymen is a 15.2 % o he labo o ce (3 923 200
indi iduals). Joblessness ouches especially he sou h and wes o he coun y. I g ows o e he ime in 2011 (up o 43.3
% compa ed o 2010) by he dec ease o g ow h (-2.1 %). he decline in o eign in es men . The a e age po e y a e
is s ill ou imes highe inside he coun y. compa ed o weal hie coas al egions which a e iche [20].
In he o he hand expo a ions ep esen 45 % o he GPD. and 81% o expo a ions a e in ended o Eu opean wi h
5700 indus ial companies. Tunisia is he leading indus ial expo e in he sou he n sho e o he Medi e anean o he
Eu opean Union (Janua y 2008). Also. Tunisia has excep ional asse s in logis ic pe o mances: I ’s p oximi y o Eu ope.
i s s a egic posi ion be ween No h A ica. Eu ope and Asia.
4. P oblem pos ula e and me hodology
We can add o all hose asse s (ci ed in he las sec ion) ha Tunisia has a s a egic geog aphic posi ion and h ough
his. many ques ions a e asked abou he pe o mance o he Tunisian logis ics sec o (110 h wo ld class acco ding o
LPI anking 2018 he Wo ld Bank) [20]. So. poli ic decision go e ns economic logic and he e o e he loca ion o
economic ac i i ies. Hence he in e es o his wo k which i s o e all aim is o de ine a decision suppo ool o loca ion
o logis ic pla o ms in o de o ha e an e icien logis ic sys em. Howe e . he use o mul ic i e ia analysis helps in
compa ing ac ions and he e o e o gene a e one o many p oposals. In ac . h ough geog aphic in o ma ion.
mul ic i e ia analysis. we selec a ian s o sa is ying p e e ences o decision make s. C i e ia can be quan i a i e and
quali a i e. This ype o c i e ia is sui able o he e alua ion o u ban p ojec s. Thus. he choice o mul ic i e ia me hods
should ake in accoun p oblem na u e and decision make s choices
4.1. ELECTRE III me hod
ELECTRE III me hod belongs o a amily o ELECTRE me hods. p oposed by Roy [23] and i is based on he bina y
ou anking ela ion [ 23. 30]. In his me hod. he basic se o da a is composed o he ollowing elemen s: a ini e se o
a ian s A. a amily o c i e ia F. and he p e e en ial in o ma ion submi ed by he DM. The p e e en ial in o ma ion is
de ined in he o m o c i e ia weigh s - w and he indi e ence - q. p e e ence - p and e o - h esholds [23]. The
h esholds de ine he ollowing in e als o p e e ence be ween a ian s on each c i e ion: indi e ence (up o q). weak
p e e ence (be ween q and p). (s ong) p e e ence (be ween p and ) and incompa abili y (beyond ). Va ian s a and b
a e conside ed indi e en i he di e ence be ween hei e alua ions (a) and (b) on a speci ic c i e ion is so small
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 27(02), 2157-2169
2160
(smalle han q) ha he DM can’ make any dis inc ion be ween a ian s. Va ian a is weakly p e e ed agains a ian
b i he di e ence be ween hei e alua ions (a) and (b) on a speci ic c i e ion is no iceable o he DM (be ween q and
p) bu he/she is hesi an o p e e one o hem. Va ian a is s ongly p e e ed agains a ian b i he di e ence be ween
hei e alua ions (a) and (b) on a speci ic c i e ion is subs an ial o he DM (be ween p and ) and he/she is con inced
ha a is p e e ed o b. Va ian s a and b a e incompa able i he di e ence be ween hei e alua ions (a) and (b) on a
speci ic c i e ion is so la ge (la ge han ) ha he DM canno conside hem as compa able objec s.
The ou anking ela ion in he ELECTRE III me hod is buil on he basis o he conco dance and disco dance es s. In he
conco dance es a conco dance ma ix (Table 5). composed o he global conco dance indica o s C (a.b). is cons uc ed.
In he disco dance es . a disco dance index dj (a.b) o each c i e ion j is calcula ed. The ou anking ela ion indica es
he ex en o which” a ou anks b” o e all. This ela ion is exp essed by he deg ee o c edibili y d (a.b). which is
equi alen o he global conco dance indica o C (a.b) weakened by he disco dance indexes dj (a.b). The alues o d
(a.b) a e om he in e al [0.1]. C edibili y d (a.b) = 1 i and only i he asse ion
a S b (“a ou anks b”) is well ounded. D (a.b) = 0 i he e is no a gumen in a o o a S b (no a S b – “a does no ou anks
b”). The de ini ion o d (a.b) esul s in he cons uc ion o he c edibili y ma ix based on which he me hod es ablishes
wo p elimina y ankings (comple e p eo de s) using a classi ica ion algo i hm (dis illa ion p ocedu e).
Du ing his p ocedu e one can ob ain a descending and an ascending p eo de . In he descending dis illa ion he anking
p ocess s a s om he selec ion o he bes a ian . which is placed a he op o he anking. In he ascending
dis illa ion he a ian s a e anked in he in e se o de . The inal anking is gene a ed as an in e sec ion o he
abo e-men ioned comple e p eo de s. I can be p esen ed ei he in he o m o he anking ma ix (Table 4)
(o in he o m o he ou anking g aph. The ollowing si ua ions can be dis inguished he e: indi e ence (I).
p e e ence (P). lack o p e e ence (P~) and incompa abili y (R).
4.2. Rele an c i e ia
The decision o loca e logis ics pla o ms is a complex issue. because decision depends on se e al ac o s (in o ma ion.
he gene al con ex and he weigh o each c i e ion [12]. So. esul s should be based on s udied c i e ia and in o ma ion
ela ed o hose c i e ia can be quan i a i e and quali a i e. Hence he ole o mul ic i e ia analysis me hods which makes
hese c i e ia con lic ing. hese a e e alua ed by he weigh s assigned by he decision-make s (p e e ences) in o de o ind he
op imal si e loca ion o anking.
In his con ex . we conside ed a ian s (desc ibed below) which ha e been de ined as 24 go e no a es. dis ibu ed all
o e Tunisia and ep esen ing po en ial a eas o placing he logis ic pla o ms on hei e i o y. We p opose a se o 7
measu es. denomina ed by C1. C2…. C7 ha includes c i e ia ep esen ing di e en aspec s o e alua ion and in e es s
o di e en s akeholde s. All 7 c i e ia a e p esen ed below (yea 2024).
• C1 - Condi ion o anspo a ion in as uc u e – [km/100km2]. This maximized c i e ion is de ined as a densi y o
oad and ail in as uc u e in he conside ed egion. I measu es logis ics accessibili y o he egion and
anspo a ion e iciency o dis ibu ing goods. The c i e ion is exp essed as a a io o he o e all leng h o public
oads and s anda d gauge ailway lines and he o al a ea o he egion.
• C2 – socio Economic de elopmen – [%]. This maximized c i e ion is de ined as an annual alue o GDP pe capi a
in he analyzed egion. I measu es economic po en ial o a egion. which is a c i ical ac o in luencing on in es o s’
decisions conce ning he placemen o an in es men .
• C3 – Land p ice – [TND/m2]. I is de ined as an es ima ed uni pu chase cos o de eloping 1 sq. me e o he logis ic
pla o m in he conside ed egion.
• C4 - Le el o anspo a ion and logis ics compe i i eness – [%]. This minimized c i e ion is de ined as a pe cen age
sha e o logis ics se ice p o ide s ( anspo a ion. o wa ding and wa ehousing companies) ope a ing in he
egion in he o al numbe o such en i ies in he coun y. The c i e ion is minimized o guide he DM (in es o )
owa ds he selec ion o less compe i i e en i onmen s.
• C5 - In es men a ac i eness – [km2]. I is de ined as a o al a ea o all Special Economic Zones (SEZ) in he egion.
Thus. i helps o es ima e po en ial bene i s o he logis ic pla o m esul ing om special ules. including income
ax eleases. VAT educ ions. educed land use and employees’ social bene i s ees).
• C6 - T anspo a ion and logis ics a ac i eness – [mln onnes]. This maximized c i e ion measu es he indus y
a ac i eness o each egion. exp essed in e ms o he o al annual amoun o load being anspo ed in each egion
by all logis ics se ice p o ide s. C i e ion C6 is calcula ed as a sum o indus ial p oduc s o each egion.
• C7 -En i onmen al- iendliness – [m2/habi an ]. This maximized c i e ion de ines he en i onmen al iendliness
and he exis ing condi ion o en i onmen in each egion. I measu es le el o g een land in each egion
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 27(02), 2157-2169
2161
Table 1 E alua ion ma ix o he compa ed a ian s
C1
C2
C3
C4
C5
C6
C7
Di ec ion o p e e ences
Max
Max
Min
Min
Max
Max
Max
Uni
Km/100km2
%
D
%
Km2
Mille onnes
M2/habi an
A1: A iana
51.88
0.751
45
4.26
0
857.6
15.25
A2: Bize e
55.25
0.541
45
18.18
0.81
1072.0
16.25
A3: Beja
95.62
0.354
55
20.58
0
321.6
19.56
A4: Ben A ous
77.28
0.714
65
52.65
0
1608.0
17.39
A5: Gabes
34.56
0.354
50
25.86
0
2144.0
17.40
A6: Ga sa
48.51
0.388
50
50
0
1286.4
15.74
A7: Jendouba
88.7
0.231
20
20.83
0
536.0
22.22
A8: kai ouan
69.1
0.25
30
5.88
0
364.5
15.48
A9: kasse ine
65.75
0.234
45
10
0
214.4
16.39
A10: Kebili
7.3
0.445
50
5.76
0
857.6
16.23
A11: Le ke
70.75
0.284
25
13.33
0.1
128.6
16.65
A12: Mahdia
56.07
0.402
52
23.07
0
536.0
17.90
A13: Manouba
125.14
0.567
70
17.64
0
643.2
15.25
A14: Medenine
60.33
0.397
50
29.62
0
214.4
16.23
A15:Monas i
120.3
0.73
30
10.9
0
2358.4
15.28
A16: Nabeul
111.23
0.615
55
22.94
0
2144.0
23
A17: Seliana
76.2
0.262
35
10.52
0
750.4
19.20
A18: S ax
29.84
0.603
26
24.6
0
1715.2
17.52
A19: Sidi Bouzid
94.29
0.271
25
7.14
0
107.2
17.3
A20: Sousse
46.48
0.67
45
12
0
2144.0
15
A21: Ta aouine
15.11
0.301
30
28.57
1.6
364 .476
20.63
A22: Tozeu
18.72
0.495
25
25
0
428.8
17.20
A23: Tunis
114.3
0.762
50
21.39
0
1072.0
14.95
A24: Zaghouan
79.84
0.483
50
17.39
0
600.3
17.13
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 27(02), 2157-2169
2162
Figu e 1 Po en ial egions ( a ian s)
4.3. Modeling o DM’s and s akeholde s’ p e e ences
The model o he DM’s and s akeholde s’ p e e ences has been cons uc ed based on in e iews and su eys ca ied ou
wi h ep esen a i es o : in es o s. local esiden s. local companies and logis ics se ice p o ide s. Ini ially. p e e ences
o each indi idual. conce ning he impo ance o he c i e ia and he sensi i i y o each pe son wi h espec o he
changes o he c i e ia alues. ha e been de e mined. A e wa ds indi idual p e e ences and expec a ions ha e been
agg ega ed and ans e ed in o one. cohe en model o p e e ence. cha ac e is ic o he Elec e III/IV me hod. This
model. p esen ed in able 2. u ilizes weigh s o c i e ia – w and h esholds o indi e ence - q. p e e ence - p and e o –
. o di e en ia e a ian s be ween each o he . We don’ use he e o in ou case.
Table 2 Agg ega ed model o p e e ences o he DM and s akeholde s
C1
C2
C3
C4
C5
C6
C7
w
9
8
6
7
7.5
8.5
4
q
10
0.1
5
1
0.1
100
2
p
20
0.2
10
5
0.2
500
5
-
-
-
-
-
-
-
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 27(02), 2157-2169
2163
4.3.1. Compu a ional expe imen s
We use he ELECTRE III me hod o make compu a ional expe imen s. whose algo i hm has been p esen ed in sec ion
4.1. Compu a ional expe imen s ha e been ini ia ed by he cons uc ion o he e alua ion ma ix ha con ains da a
assessing all a ian s on all c i e ia (Table 1). A e wa ds he agg ega ed model o p e e ences has been de ined (Table
2). As a esul . he conco dance ma ix (Table 5) composed o he global conco dance indica o s. has been cons uc ed.
These indica o s de ine he deg ee o ul illing he asse a ion ha “a ou ank b”. based on he a gumen s suppo ing
his s a emen (a gumen s “ o ”). only. Thus. aking. o ins ance. he example o a ian s A3 and A5. A5 ou anks A3
wi h a global indica ion C (a. b) =0.82. A he same ime he in e se indica ion ha A3 ou anks A5 is ep esen ed by C
(b. a) =0.69. These igu es can be in e p e ed as an indica ion ha he e a e sligh ly mo e a gumen s suppo ing he
asse ion ha A5 ou anks A3. Simila easoning can be ca ied ou o all pai -wise compa ed a ian s.
Based on he c edibili y ma ix he me hod es ablished wo p elimina y ankings (comple e p eo de s) using
descending and ascending dis illa ions (Table 3). In he o me he anking is cons uc ed om he op down o he
bo om. while in he la e in he in e se sequence om he bo om up o he op. The esul an o hese wo dis illa ions
is he inal p eo de (a e age). In he inal anking he ollowing si ua ions – ela ions may occu : indi e ence – I –
a ian s a e indi e en and g aphically placed in he same box; p e e ence – P – a ian a is p e e ed o b; a is placed
abo e b on he g aph; non-p e e ence – P- - a ian a is no p e e ed o b. hus. a is placed below b on he g aph;
incompa abili y – R – a ian s a e no linked (in e connec ed) on he g aph. Using he same example o a ian s A3 and
A5 one can see ha hey a e incompa able. which is indica ed by le e R in he anking ma ix.
Table 3 Ascending and descending dis illa ion
Ascend.
Descend.
A e age
Si e
Ranking
A1
4.0
7.0
5.5
A4
1.0
A2
1.0
6.0
3.5
A16
1.0
A3
2.0
3.0
2.5
A23
1.5
A4
1.0
1.0
1.0
A13
2.0
A5
1.0
6.0
3.5
A3
2.5
A6
2.0
6.0
4.0
A2
3.5
A7
2.0
5.0
3.5
A5
3.5
A8
6.0
7.0
6.5
A7
3.5
A9
6.0
7.0
6.5
A15
3.5
A10
5.0
7.0
6.0
A18
3.5
A11
5.0
7.0
6.0
A20
3.5
A12
3.0
6.0
4.5
A6
4.0
A13
1.0
3.0
2.0
A14
4.0
A14
2.0
6.0
4.0
A21
4.0
A15
1.0
6.0
3.5
A24
4.0
A16
1.0
1.0
1.0
A12
4.5
A17
3.0
6.0
4.5
A17
4.5
A18
1.0
6.0
3.5
A22
5.0
A19
4.0
7.0
5.5
A1
5.5
A20
2.0
5.0
3.5
A19
5.5
A21
1.0
7.0
4.0
A10
6.0
A22
3.0
7.0
5.0
A11
6.0
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 27(02), 2157-2169
2164
A23
1.0
2.0
1.5
A8
6.5
A24
4.0
4.0
4.0
A9
6.5
Being a he op o he anking hey become he mos desi able solu ions: Va ian s A4 and A16 ep esen : Ben A ous
and Nabeul. espec i ely. They a e loca ed in he eas e n no h and Cen al-no he n pa o Tunisia. Based on he
esul s gene a ed in he anking ma ix (Table 4) one may conclude ha bo h a ian s a e equally good (22 indica ions
P-p e e ed). A4 is cha ac e ized by high le el o socio-economic de elopmen and A16 is cha ac e ized by high olume
o indus ial p oduc s and has he bes condi ion o anspo a ion in as uc u e condi ions. In o he hand. A8 and A9
ha e he lowes le el o socio-economic de elopmen and ha e no impo an indus ial ac i i ies.
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 27(02), 2157-2169
2165
Table 4 Ranking ma ix
A1
A2
A3
A4
A5
A6
A7
A8
A9
A10
A11
A12
A13
A14
A15
A16
A17
A18
A19
A20
A21
A22
A23
A24
A1
0
P-
P-
P-
P-
P-
P-
P+
P+
P+
P+
P-
P-
P-
P-
P-
P-
P-
I
P-
P-
P-
P-
P-
A2
P+
0
R
P-
I
P+
R
P+
P+
P+
P+
P+
P-
P+
I
P-
P+
I
P+
R
P+
P+
P-
R
A3
P+
R
0
P-
R
P+
P+
P+
P+
P+
P+
P+
P-
P+
R
P-
P+
R
P+
P+
R
P+
P-
P+
A4
P+
P+
P+
0
P+
P+
P+
P+
P+
P+
P+
P+
P+
P+
P+
I
P+
P+
P+
P+
P+
P+
P+
P+
A5
P+
I
R
P-
0
P+
R
P+
P+
P+
P+
P+
P-
P+
I
P-
P+
I
P+
R
P+
P+
P-
R
A6
P+
P-
P-
P-
P-
0
P-
P+
P+
P+
P+
P+
P-
I
P-
P-
P+
P-
P+
P-
R
P+
P-
R
A7
P+
R
P-
P-
R
P+
0
P+
P+
P+
P+
P+
P-
P+
R
P-
P+
R
P+
I
R
P+
P-
R
A8
P-
P-
P-
P-
P-
P-
P-
0
I
P-
P-
P-
P-
P-
P-
P-
P-
P-
P-
P-
P-
P-
P-
P-
A9
P-
P-
P-
P-
P-
P-
P-
I
0
P-
P-
P-
P-
P-
P-
P-
P-
P-
P-
P-
P-
P-
P-
P-
A10
P-
P-
P-
P-
P-
P-
P-
P+
P+
0
I
P-
P-
P-
P-
P-
P-
P-
P-
P-
P-
P-
P-
P-
A11
P-
P-
P-
P-
P-
P-
P-
P+
P+
I
0
P-
P-
P-
P-
P-
P-
P-
P-
P-
P-
P-
P-
P-
A12
P+
P-
P-
P-
P-
P-
P-
P+
P+
P+
P+
0
P-
P-
P-
P-
I
P-
P+
P-
R
P+
P-
R
A13
P+
P+
P+
P-
P+
P+
P+
P+
P+
P+
P+
P+
0
P+
P+
P-
P+
P+
P+
P+
P+
P+
P-
P+
A14
P+
P-
P-
P-
P-
I
P-
P+
P+
P+
P+
P+
P-
0
P-
P-
P+
P-
P+
P-
R
P+
P-
R
A15
P+
I
R
P-
I
P+
R
P+
P+
P+
P+
P+
P-
P+
0
P-
P+
I
P+
R
P+
P+
P-
R
A16
P+
P+
P+
I
P+
P+
P+
P+
P+
P+
P+
P+
P+
P+
P+
0
P+
P+
P+
P+
P+
P+
P+
P+
A17
P+
P-
P-
P-
P-
P-
P-
P+
P+
P+
P+
I
P-
P-
P-
P-
0
P-
P+
P-
R
P+
P-
R
A18
P+
I
R
P-
I
P+
R
P+
P+
P+
P+
P+
P-
P+
I
P-
P+
0
P+
R
P+
P+
P-
R
A19
I
P-
P-
P-
P-
P-
P-
P+
P+
P+
P+
P-
P-
P-
P-
P-
P-
P-
0
P-
P-
P-
P-
P-
A20
P+
R
P-
P-
R
P+
I
P+
P+
P+
P+
P+
P-
P+
R
P-
P+
R
P+
0
R
P+
P-
R
A21
P+
P-
R
P-
P-
R
R
P+
P+
P+
P+
R
P-
R
P-
P-
R
P-
P+
R
0
P+
P-
R
A22
P+
P-
P-
P-
P-
P-
P-
P+
P+
P+
P+
P-
P-
P-
P-
P-
P-
P-
P+
P-
P-
0
P-
R
A23
P+
P+
P+
P-
P+
P+
P+
P+
P+
P+
P+
P+
P+
P+
P+
P-
P+
P+
P+
P+
P+
P+
0
P+
A24
P+
R
P-
P-
R
R
R
P+
P+
P+
P+
R
P-
R
R
P-
R
R
P+
R
R
R
P-
0
