S udies Managemen and Finance Economics, o Jou nal
0504-2644 (online): ISSN 0490,-2644 (p in ): ISSN
5202 Oc obe 10 Issue 80 Volume
8.317 Fac o : Impac ,02-i10-10.47191/je ms/ 8 DOI: A icle
9565 -6582 No: Page
JEFMS, Volume 08 Issue 10 Oc obe 2025 www.ije m.co.in Page 6582
B idging Technology and Readiness: AI, IoT, and he E ec i eness o Disas e
P e en ion in Clima e-Vulne able Regions
Hoang Minh Quan1, Phung Quang Thang2, Vo Minh Vinh3
*1Facul y o Business Managemen , Uni e si y o G eenwich, London
2Le Hong Phong High School o he Gi ed, Ho Chi Minh, VIETNAM
3Facul y o Managemen and Economics, Uni e si y o Tomas Ba a, Czech Republic
ABSTRACT: In he con ex o in ensi ying clima e change, his s udy explo es how a i icial in elligence (AI)–d i en lood p edic ion
accu acy and In e ne o Things (IoT)–based en i onmen al moni o ing co e age con ibu e o disas e p e en ion e ec i eness,
wi h communi y eadiness o echnology adop ion as a mode a ing ac o . G ounded in Socio-Technical Sys ems (STS) Theo y and
he Technology Accep ance Model (TAM)/Uni ied Theo y o Accep ance and Use o Technology (UTAUT), he s udy adop s a
quan i a i e design wi h 385 alid esponses om Vie nam, Singapo e, and Malaysia, ep esen ing enginee s, ICT manage s,
disas e o icials, and communi y s akeholde s. A s uc u ed 5-poin Like scale ques ionnai e was de eloped, o e alua e
pe cep ions o p edic i e accu acy, moni o ing co e age, disas e p e en ion e ec i eness, and communi y eadiness. Da a
analysis was conduc ed using SPSS, including eliabili y assessmen (C onbach’s alpha), explo a o y ac o analysis (EFA), linea
eg ession, and mode a ion analysis wi h he PROCESS Mac o o es he hypo hesized ela ionships. Findings con i m ha AI
p edic i e accu acy enhances p e en ion no me ely h ough nume ical p ecision bu by p o iding imely, ac ionable wa nings.
Likewise, IoT moni o ing imp o es si ua ional awa eness, ye i s alue depends on s a egic deploymen , in e ope abili y, and
usabili y. Impo an ly, communi y eadiness encompassing us , li e acy, a o dabili y, and willingness o adop eme ges as he
decisi e ac o ha enables echnological in as uc u es o ansla e in o p o ec i e ac ion. The s udy ad ances heo y by
in eg a ing socio- echnical and adop ion pe spec i es and o e s p ac ical insigh s, u ging policymake s o in es no only in
in as uc u es bu also in eadiness-building, pa icipa o y engagemen , and us -enhancing ini ia i es.
KEYWORDS: Disas e p e en ion e ec i eness; AI-d i en lood p edic ion; IoT en i onmen al moni o ing; Communi y eadiness;
Socio-Technical Sys ems Theo y; TAM; UTAUT; Clima e change esilience.
I. INTRODUCTION
In he con ex o in ensi ying clima e change and he inc easing equency o ex eme wea he e en s, disas e
p e en ion has become a c i ical ocus o bo h policymake s and esea che s (Wen e al., 2023). Technological inno a ion,
pa icula ly he in eg a ion o a i icial in elligence (AI) and he In e ne o Things (IoT), o e s ans o ma i e po en ial o
enhancing disas e p edic ion, moni o ing, and p epa edness (Ali e al., 2022; Na ayana e al., 2024). AI-d i en lood p edic ion
sys ems p o ide ad anced o ecas ing accu acy, while IoT-enabled en i onmen al moni o ing ne wo ks expand eal- ime
si ua ional awa eness ac oss haza d-p one egions. Toge he , hese echnologies p omise o shi disas e managemen om
eac i e esponse o p oac i e p e en ion, he eby educing isks and ulne abili ies.
Despi e hese ad ancemen s, disas e p e en ion e ec i eness canno be gua an eed by echnological inno a ion alone.
Resea ch highligh s ha echnological p ecision and moni o ing co e age a e o en unde mined when communi ies lack he
eadiness, us , and ins i u ional suppo o ac upon he gene a ed da a (Sinha e al., 2019; B a e al., 2022). Socio-Technical
Sys ems (STS) Theo y emphasizes he need o alignmen be ween echnical sys ems and social ac o s, while he Technology
Accep ance Model (TAM) and Uni ied Theo y o Accep ance and Use o Technology (UTAUT) explain he mic o-le el mechanisms
ha d i e adop ion and beha io al change (Venka esh e al., 2003; Tseng and S ojadino ić, 2024). Howe e , exis ing schola ship
equen ly e alua es ei he echnical pe o mance o social adop ion in isola ion, o e looking how communi y eadiness
mode a es he ansla ion o AI and IoT inno a ions in o e ec i e disas e p e en ion ou comes. This c ea es a signi ican esea ch
gap in unde s anding he in eg a i e pa hways h ough which socio- echnical sys ems shape esilience.
B idging Technology and Readiness: AI, IoT, and he E ec i eness o Disas e P e en ion in Clima e-Vulne able
Regions
JEFMS, Volume 08 Issue 10 Oc obe 2025 www.ije m.co.in Page 6583
This s udy add esses his gap by examining how AI-d i en lood p edic ion accu acy and IoT-based en i onmen al
moni o ing co e age in luence disas e p e en ion e ec i eness, wi h a pa icula ocus on he mode a ing ole o communi y
eadiness o echnology adop ion. By si ua ing hese a iables wi hin an in eg a ed STS–TAM/UTAUT amewo k, he esea ch
con ibu es o a holis ic unde s anding o disas e echnology adop ion and i s ole in ad ancing esilience agains clima e-induced
haza ds. To guide his in es iga ion, he s udy add esses he ollowing esea ch ques ions:
1. To wha ex en does AI-d i en lood p edic ion accu acy enhance disas e p e en ion e ec i eness?
2. How does IoT-based en i onmen al moni o ing co e age con ibu e o disas e p e en ion ou comes?
3. In wha ways does communi y eadiness mode a e he ela ionship be ween AI-d i en lood p edic ion accu acy and
disas e p e en ion e ec i eness?
II. LITERATURE REVIEW
2.1. Disas e P e en ion E ec i eness
Disas e p e en ion e ec i eness e e s o he ex en o which s a egies, echnologies, and policies success ully educe
o elimina e he isks and impac s o na u al haza ds be o e hey occu . Unlike disas e esponse, which ocuses on pos -e en
eco e y, p e en ion emphasizes p oac i e measu es ha minimize exposu e and ulne abili y h ough an icipa o y ac ion (Ma h
e al., 2015; Gao e al., 2025; Keim, 2018). I is concep ually dis inc om mi iga ion, which educes he se e i y o consequences
once a haza d has al eady un olded. P e en ion aims o a oid o neu alize disas e impac s al oge he , hough his dis inc ion is
o en deba ed in disas e s udies (Ma h e al., 2015). Schola s no e ha while p e en ion is aspi a ional, mi iga ion is o en
ega ded as mo e p ac ically achie able. Ne e heless, p e en ion e ec i eness emains a i al e alua i e lens in an e a whe e
clima e change in ensi ies haza d equency and complexi y, equi ing s a egies ha mo e beyond eac i e managemen
(Gailma d and Pa y, 2018; Wen e al., 2023).
The e m “e ec i eness” i sel is con es ed, as i s meaning a ies ac oss ins i u ional and communi y pe spec i es. Fo
go e nmen s, e ec i eness is equen ly measu ed in e ms o economic sa ings, in as uc u e p o ec ion, and con inui y o
c i ical se ices. By con as , communi ies o en judge p e en ion ou comes by he p o ec ion o li es, li elihoods, and cul u al
asse s, as well as educed displacemen du ing haza d e en s (Mechle , 2016). These di e gen pe spec i es expand e ec i eness
me ics beyond immedia e indica o s such as educed mo ali y o minimized p ope y loss, o include b oade measu es such as
imp o ed ea ly wa ning lead imes, enhanced adap i e capaci y, and s eng hened us in disas e go e nance sys ems. Such
mul iplici y unde sco es he need o cla i y e alua i e c i e ia, as inconsis en de ini ions limi he compa abili y o esea ch
indings and hinde alignmen ac oss policy amewo ks.
A c i ical gap in exis ing esea ch lies in he limi ed examina ion o end- o-end socio- echnical pa hways ha shape
p e en ion ou comes. Much o he cu en schola ship e alua es isola ed componen s o example, he p edic i e accu acy o
a i icial in elligence (AI) models o he eliabili y o IoT-enabled senso s wi hou conside ing how hese echnologies in e ac
wi hin he b oade disas e p e en ion chain (Yu and He, 2022). T ue p e en ion e ec i eness equi es success ul in eg a ion
ac oss mul iple s ages: haza d de ec ion, p edic ion and modeling, ale dissemina ion, communi y in e p e a ion, and p o ec i e
esponse (Haque e al., 2024). B eakdowns a any link whe he due o echnical ailu es, inadequa e communica ion, o social
non-adop ion unde mine o e all e ec i eness. This highligh s he need o in eg a i e app oaches ha mo e beyond na ow
echnical assessmen s o examine how socio- echnical sys ems as a whole ansla e inno a ions in o angible esilience gains.
Wi hou his ull-pipeline pe spec i e, i emains unclea whe he ad ancemen s in AI and IoT meaning ully enhance disas e
p e en ion o me ely s eng hen isola ed echnical capabili ies.
2.2. Theo e ical amewo k
The p esen s udy is g ounded in an in eg a i e heo e ical amewo k ha combines Socio-Technical Sys ems (STS)
Theo y and he Technology Accep ance Model (TAM)/Uni ied Theo y o Accep ance and Use o Technology (UTAUT). Toge he ,
hese heo ies cap u e bo h he echnical dimension (AI-d i en p edic ion and IoT-based moni o ing) and he social dimension
(communi y eadiness), he eby enabling a holis ic unde s anding o how disas e p e en ion e ec i eness eme ges.
2.2.1. Socio-Technical Sys ems (STS) Theo y
Socio-Technical Sys ems (STS) Theo y posi s ha o ganiza ional pe o mance esul s om he join op imiza ion o
echnical subsys ems ( ools, echnologies, and p ocesses) and social subsys ems (people, communi ies, and ins i u ions).
Acco ding o T is and Bam o h (1951), sys ems only achie e hei ull po en ial when bo h subsys ems a e aligned a he han
op imized in isola ion. In disas e managemen , his pe spec i e highligh s ha he echnologies such as AI-based p edic ion
B idging Technology and Readiness: AI, IoT, and he E ec i eness o Disas e P e en ion in Clima e-Vulne able
Regions
JEFMS, Volume 08 Issue 10 Oc obe 2025 www.ije m.co.in Page 6584
models and IoT moni o ing in as uc u es will no au oma ically deli e e ec i eness wi hou pa allel social p epa edness (Tseng
and S ojadino ić, 2024; Walke e al, 2007).
Disas e p e en ion e ec i eness can be concep ualized as an eme gen sys em-le el ou come ha a ises when echnical
solu ions (p edic ion accu acy, moni o ing co e age) a e in eg a ed wi h social accep ance and eadiness. F om an STS
pe spec i e, e ec i eness is no educible o echnical capaci y alone bu depends on whe he communi ies unde s and and
espond o echnological ou pu s (Chil e s, Palle and Ha g ea es, 2018). STS has been alida ed ac oss domains whe e
echnology–human in e ac ions a e c i ical. Fo ins ance, Kleine e al. (2011) demons a ed i s explana o y powe in complex
sa e y sys ems, a guing ha echnical imp o emen s wi hou human adap a ion a ely lead o be e ou comes. Simila ly, Tseng
and S ojadino ić (2024) showed ha socio- echnical in eg a ion inc eases sys em esilience du ing c ises, amewo ks like CI-STR
demons a e ha esilience is shaped by he dynamic in e play be ween social and echnical ac o s, wi h human capabili ies
se ing as he in e ace ha links indi iduals’ social cha ac e is ics o communi y esou ces and echnical in as uc u e. This
in eg a ion c ea es eedback loops whe e changes in one domain in luence he o he , enabling communi ies o adap and eco e
mo e e ec i ely a e disas e s. These indings ein o ce ha disas e p e en ion echnologies a e e ec i e only when embedded
in socially a uned sys ems.
The ele ance o STS becomes mo e e iden when conside ing he in e play among all a iables in his s udy. Wi hin he
echnical subsys em, AI-d i en p edic i e accu acy p o ides a c ucial in o ma ional esou ce, and IoT-enabled en i onmen al
moni o ing expands he scope o da a collec ion ac oss geog aphies. AI models, such as a i icial neu al ne wo ks and hyb id
a chi ec u es, ha e demons a ed high p edic i e accu acy in applica ions like wa e quali y assessmen , pollu ion de ec ion, and
e en ea hquake p edic ion, wi h some models achie ing o e 95% accu acy and signi ican educ ions in e o a es compa ed o
adi ional me hods (Zhang e al., 2021). Howe e , STS eminds us ha disas e p e en ion echnologies each hei ull po en ial
only when embedded in socially a uned sys ems ha os e unde s anding, p epa edness, and coo dina ed ac ion (Velazquez e
al., 2020; Ma chezini e al., 2018). Fo ins ance, a lood p edic ion model ha achie es high accu acy may s ill be dis ega ded i
local communi ies lack he li e acy o in e p e isk ca ego ies, o i decision-make s ail o ins i u ionalize he model in o
eme gency p o ocols (Munawa , Hammad and Walle , 2022; Mosa i, Öz ü k and Chau, 2018). Simila ly, b oad moni o ing co e age
may only gene a e esilience when local esponde s ha e clea p ocesses o in eg a ing senso ou pu s in o coo dina ed ac ion.
In his way, he mode a o a iable communi y eadiness o echnology adop ion becomes no simply an ex e nal ac o bu an
essen ial componen o he socio- echnical balance. I cap u es whe he social ac o s possess he us , in e p e i e capaci y, and
o ganiza ional ou ines necessa y o con e aw echnical ou pu s in o collec i e p e en i e ac ion. Wi hou his synch oniza ion,
he sys em isks becoming echnologically sophis ica ed ye socially ine , a condi ion STS explici ly cau ions agains .
Al hough STS o e s a obus lens o unde s and he join op imiza ion o social and echnical subsys ems, i emains
limi ed in i s explana o y p ecision when applied o disas e echnology adop ion. One key issue is i s le el o abs ac ion: STS
emphasizes “alignmen ” be ween subsys ems bu p o ides li le guidance on how his alignmen is achie ed in p ac ice. Fo
ins ance, while he heo y would p edic ha communi y eadiness mus complemen AI-d i en p edic ion and IoT moni o ing, i
does no speci y he cogni i e o beha io al mechanisms h ough which indi iduals decide o us o ejec such sys ems.
Fu he mo e, STS ends o assume a ela i ely symme ical ela ionship be ween social and echnical domains, ye in eal-wo ld
disas e con ex s, powe imbalances, policy cons ain s, and socio-economic inequali ies o en p i ilege one subsys em o e he
o he . This means ha e en when echnical sys ems a e well designed, ma ginalized communi ies may lack he ins i u ional
capaci y o poli ical oice o align e ec i ely, lea ing he STS amewo k insu icien o cap u e s uc u al inequi ies. Thus, while
STS highligh s he impo ance o in eg a ion, i equi es supplemen a ion by adop ion models ha accoun o he mic o-le el
decision p ocesses and socio-poli ical cons ain s shaping eal ou comes.
2.2.2. Technology Accep ance Model (TAM) / Uni ied Theo y o Accep ance and Use o Technology (UTAUT)
The Technology Accep ance Model (TAM) de eloped by Da is (1989) and he Uni ied Theo y o Accep ance and Use o
Technology (UTAUT) de eloped by Venka esh e al. (2003) explain why indi iduals and g oups adop o esis new echnologies.
These models emphasize pe cep ions o use ulness, ease o use, acili a ing condi ions, and us as key de e minan s o
echnology accep ance. UTAUT ex ends TAM by inco po a ing social in luence and beha io al in en ion, which a e pa icula ly
ele an in communi y-le el echnology adop ion.
Whe eas STS p o ides a mac o-le el iew o sys em in eg a ion, TAM and UTAUT illumina e he mic o-mechanisms ha
link he echnical and social elemen s ac oss a iables. The accu acy o AI-d i en p edic ions di ec ly shapes pe cep ions o
use ulness: he mo e eliable he o ecas s, he mo e likely communi ies a e o ega d hem as wo h adop ing (Kelly, Kaye and
O iedo-T espalacios, 2022). A he same ime, IoT-based moni o ing in luences pe cep ions o acili a ing condi ions by ensu ing
ha ale s a e localized, equen , and easily accessible, which in u n educes pe cei ed ba ie s o adop ion. C ucially, hese
B idging Technology and Readiness: AI, IoT, and he E ec i eness o Disas e P e en ion in Clima e-Vulne able
Regions
JEFMS, Volume 08 Issue 10 Oc obe 2025 www.ije m.co.in Page 6585
pe cep ions eed in o he mode a o a iable communi y eadiness since eadiness i sel is an agg ega e ou come o pe cei ed
ease o use, us , and social in luence (Sneesl e al., 2022). I communi ies belie e ha p edic ion ou pu s a e bo h accu a e and
ac ionable, and i moni o ing sys ems a e seen as compa ible wi h exis ing p ac ices, willingness o adop ises sha ply. This
psychological accep ance is wha ul ima ely con e s echnical in as uc u es in o measu able disas e p e en ion e ec i eness.
Thus, TAM/UTAUT cla i y why he same echnological deploymen yields une en esul s ac oss di e en egions: e ec i eness
depends less on he absolu e le el o accu acy o co e age, and mo e on how use s in e p e , us , and embed hese ea u es in o
collec i e ou ines.
F om he pe spec i e o TAM and UTAUT, he explana o y powe o he amewo k goes beyond communi y eadiness o
encompass how echnical ea u es shape pe cep ions and beha io al esponses. The pe cei ed use ulness o accu a e p edic ions
o he accessibili y o widesp ead moni o ing co e age di ec ly a ec s whe he indi iduals and communi ies accep o dismiss
hese ools. Impo an ly, disas e p e en ion e ec i eness eme ges no me ely om he a ailabili y o echnological in as uc u es
bu om hei accep ance, in eg a ion, and habi ual use (Sinha e al., 2019). Thus, adop ion models highligh he psychological
and social mechanisms ha ansla e echnical p ecision in o us ed sys ems, cla i ying why simila echnologies succeed in some
con ex s while ailing o deli e in o he s.
In con as , TAM and UTAUT p o ide sha pe ools o explaining adop ion a he indi idual and communi y le el, bu hey
isk o e simpli ica ion by educing complex socio- echnical p ocesses o pe cep ions such as use ulness, ease o use, and us .
These models a e highly e ec i e in p edic ing ini ial adop ion beha io , ye disas e managemen equi es mo e han adop ion;
i equi es sus ained, collec i e, and con ex -sensi i e use unde condi ions o unce ain y and s ess. TAM/UTAUT a e less
equipped o explain how s uc u al ac o s such as go e nmen egula ion, cul u al no ms o isk pe cep ion, o unequal access o
in as uc u e shape eadiness and e ec i eness beyond indi idual a i udes (Blu e al., 2022). Mo eo e , hese models assume
a ional decision-making guided by pe cep ions, bu disas e si ua ions o en in ol e emo ional esponses, misin o ma ion, and
collec i e dynamics ha all ou side he a ional-ac o pa adigm (Lee, Ramasamy, and Subba ao, 2025). As such, TAM/UTAUT may
o e s a e he explana o y powe o psychological a iables while unde es ima ing he sys emic and ins i u ional dimensions o
adop ion (Blu e al., 2024). This limi a ion ein o ces he need o in eg a e hem wi h a sys ems-le el heo y such as STS, ensu ing
ha adop ion is si ua ed no jus in indi idual cogni ion bu in b oade socio- echnical and poli ical con ex s.
2.3. De e minan s o Disas e P e en ion E ec i eness
2.3.1. AI-D i en Flood P edic ion Accu acy
AI-d i en lood p edic ion accu acy e e s o he abili y o a i icial in elligence models o o ecas lood e en s in e ms
o iming, loca ion, and se e i y wi h measu able eliabili y. Using machine lea ning and deep lea ning app oaches, hese models
in eg a e hyd ological, me eo ological, and geospa ial da a o de ec pa e ns ha adi ional physical models o en canno
cap u e (Ali e al., 2022). Techniques such as neu al ne wo ks, andom o es s, and hyb id ensemble me hods a e inc easingly
employed o ep esen nonlinea en i onmen al in e ac ions ac oss mul iple scales. Thei p edic i e pe o mance is ypically
assessed using s a is ical me ics such as Roo Mean Squa e E o (RMSE), p ecision, ecall, and A ea Unde he Cu e (AUC), which
indica e how closely o ecas s align wi h obse ed ou comes (Islam e al., 2020). These indica o s es ablish he echnical alidi y
o AI sys ems, p o iding an impo an baseline o assessing model quali y. Ye , ocusing solely on hese benchma k isks educing
he meaning o “accu acy” o a na ow se o s a is ical ou comes di o ced om p ac ical disas e p e en ion needs.
While AI models o en demons a e s ong nume ical pe o mance, he no ion o “accu acy” emains con es ed because i ca ies
di e en implica ions o esea che s, policymake s, and communi ies (Fang e al., 2020). F om a echnical pe spec i e, accu acy
is de ined by minimizing e o ma gins and maximizing s a is ical i . Howe e , o end use s, ope a ional accu acy he abili y o
o ecas s o p o ide su icien lead ime o p o ec i e ac ions can be a mo e aluable han pe ec nume ical p edic ions
deli e ed oo la e. A o ecas ha p edic s loodwa e le els wi h minimal RMSE bu issues wa nings hou s a e communi ies
ha e al eady been a ec ed o e s li le p e en i e u ili y (Sande s e al., 2022). Con e sely, a model wi h less s a is ical p ecision
ha p o ides ea ly, ac ionable wa nings can sa e mo e li es and p ope y. This ension e eals a gap in exis ing esea ch, which
o e whelmingly p i ileges echnical accu acy while neglec ing how p edic ions unc ion wi hin eal-wo ld decision-making
con ex s (Nea ing e al., 2024). Accu acy should he e o e be e amed as a mul idimensional cons uc ha b idges echnical
p ecision wi h ope a ional usabili y.
Unde s anding AI-d i en lood p edic ion accu acy in bo h echnical and ope a ional e ms is essen ial o ad ancing
disas e p e en ion e ec i eness. Accu a e p edic ions, when aligned wi h lead ime equi emen s, enable imely e acua ion
planning, e icien alloca ion o eme gency esou ces, and he sa egua ding o c i ical in as uc u e (Nea ing e al., 2024; Adika i
e al., 2021). Howe e , wi hou mechanisms ha ansla e echnical accu acy in o ac ionable communi y ale s and ins i u ional
B idging Technology and Readiness: AI, IoT, and he E ec i eness o Disas e P e en ion in Clima e-Vulne able
Regions
JEFMS, Volume 08 Issue 10 Oc obe 2025 www.ije m.co.in Page 6586
esponses, he p ac ical con ibu ion o AI sys ems emains limi ed. This unde sco es he need o si ua e AI lood p edic ion wi hin
b oade socio- echnical pa hways, whe e model ou pu s a e linked o communica ion sys ems, go e nance eadiness, and
communi y adop ion. Fu u e esea ch mus he e o e mo e beyond isola ed s a is ical alida ion o e alua e whe he AI
p edic ion accu acy ansla es in o e ec i e p e en i e ou comes. Only by in eg a ing accu acy in o he ull chain o disas e isk
educ ion can echnological p og ess meaning ully educe ulne abili y and enhance esilience.
Building on heo e ical insigh s and empi ical e idence, he s udy p oposes he i s hypo hesis ou lined below:
H1: AI-D i en lood p edic ion accu acy posi i ely impac s disas e p e en ion e ec i eness.
2.3.2. IoT-Based En i onmen al Moni o ing Co e age
IoT-based en i onmen al moni o ing co e age e e s o he ex en and densi y o in e connec ed senso ne wo ks ha
con inuously collec , ansmi , and analyze en i onmen al da a o disas e p e en ion pu poses. These sys ems le e age he
In e ne o Things (IoT) o in eg a e di e se de ices such as ain all gauges, i e -le el senso s, soil mois u e p obes, and ai quali y
moni o s in o uni ied pla o ms capable o gene a ing eal- ime in o ma ion lows (Na ayana e al., 2024). By expanding spa ial
co e age ac oss haza d-p one a eas and b oadening he ange o moni o ed a iables, IoT sys ems s eng hen ea ly wa ning
capaci y and si ua ional awa eness. Con en ional measu es o co e age ypically ocus on senso densi y and geog aphic
dis ibu ion, e lec ing he echnical in as uc u e equi ed o eliable moni o ing. These quan i a i e indica o s es ablish a
baseline o e alua ing sys em each and unc ionali y, bu hey cap u e only one dimension o wha makes IoT moni o ing
e ec i e o disas e p e en ion (Heimann e al., 2021; Oka o , Algho ani and Delaney, 2020).
Al hough senso deploymen densi y is o en used as he p ima y measu e o IoT co e age, he concep emains
con es ed, since e ec i e co e age in ol es mo e han he aw numbe o de ices (Fa hy, Ba naghi and Ta azolli, 2018). F om
echnical app oaches o enhance co e age include le e aging geo-con ex in o ma ion o con ol da a dis ibu ion, ensu ing ha
only ele an and loca ion-speci ic da a a e collec ed and dissemina ed, which inc eases sys em e iciency and ele ance
(Hasenbu g and Be mbach, 2020). Fo decision-make s, “co e age” is meaning ul only when in o ma ion eaches he igh
ins i u ions in usable o ma s and in ime o in o m p o ec i e ac ion. A senso - ich ne wo k ha gene a es massi e da ase s bu
ails o deli e imely, in e p e able in o ma ion o au ho i ies may be echnically expansi e bu ope a ionally ine ec i e (Shah e
al., 2019). Con e sely, a smalle , s a egically placed ne wo k wi h high eliabili y and eal- ime communica ion may p o ide
g ea e disas e p e en ion bene i s. This di e gence e eals a gap in he li e a u e, which o en p i ileges quan i a i e me ics o
co e age while o e looking how hese ne wo ks unc ion wi hin socio- echnical sys ems (Yabe e al., 2022).
Unde s anding IoT-based moni o ing co e age as bo h a quan i a i e and quali a i e cons uc has c i ical implica ions
o disas e p e en ion e ec i eness. Comp ehensi e and eliable co e age enhances ea ly wa ning sys ems, imp o es haza d
de ec ion, and p o ides decision-make s wi h he si ua ional awa eness needed o imely in e en ions such as e acua ions and
esou ce alloca ion (Ray, Mukhe jee and Shu, 2017). Howe e , wi hou add essing issues o ep esen a i eness, accessibili y, and
exposu e-weigh ed placemen , expanded senso ne wo ks isk p oducing la ge olumes o da a wi hou ansla ing in o ac ionable
insigh s. This unde sco es he impo ance o ede ining co e age in e ms o e ec i e moni o ing, whe e sys em eliabili y,
in e ope abili y, and da a usabili y a e gi en equal weigh alongside senso densi y (Ejaz e al., 2019; Na ayana e al., 2024). Fu u e
esea ch mus he e o e mo e beyond coun ing de ices owa d in eg a i e assessmen s ha e alua e how IoT co e age suppo s
eal-wo ld disas e p e en ion ou comes. By embedding IoT moni o ing wi hin b oade socio- echnical amewo ks, schola s and
p ac i ione s can be e de e mine whe he inc eased co e age leads o angible esilience gains.
G ounded in bo h heo y and empi ical indings, he s udy ad ances he ollowing second hypo hesis:
H2: IoT-based en i onmen al moni o ing co e age posi i ely impac s disas e p e en ion e ec i eness.
2.3.3. Communi y Readiness o Technology Adop ion
Communi y eadiness o echnology adop ion e e s o he willingness, capaci y, and p epa edness o local communi ies
o accep , in eg a e, and ac upon echnological ools such as IoT-based moni o ing sys ems and AI-d i en p edic ion models.
D awing on amewo ks such as he Technology Accep ance Model (TAM) and he Di usion o Inno a ions heo y (Basa i -Ozel e
al., 2023), eadiness depends on ac o s like pe cei ed use ulness, ease o use, us in echnology, and ins i u ional suppo . In
he disas e p e en ion con ex , eadiness ex ends beyond simple access o digi al in as uc u e o include echnological li e acy,
a o dabili y, and social accep ance o au oma ed decision-suppo sys ems (B a e al., 2022). A high le el o communi y eadiness
ensu es ha wa nings gene a ed by IoT sys ems a e no only ecei ed bu also unde s ood and ac ed upon, he eby ansla ing
echnical co e age in o p o ec i e ac ion.
The concep o “ eadiness” emains con es ed in bo h academic and p ac ical deba es, as i does no ca y a single ag eed-
upon de ini ion. One pe spec i e in e p e s eadiness in na ow, echnical e ms, ocusing on in as uc u al capaci y such as
in e ne connec i i y, de ice a ailabili y, and sys em in e ope abili y, which a e o en ea ed as su icien indica o s o whe he a
B idging Technology and Readiness: AI, IoT, and he E ec i eness o Disas e P e en ion in Clima e-Vulne able
Regions
JEFMS, Volume 08 Issue 10 Oc obe 2025 www.ije m.co.in Page 6587
communi y can adop echnological solu ions (Blu and Wang, 2019). F om his angle, eadiness is equa ed wi h he physical
p esence o digi al ools. In con as , disas e managemen esea ch inc easingly emphasizes ha eadiness ex ends beyond
in as uc u e o encompass socio-cul u al and psychological dimensions, including us in ins i u ions, pe cei ed legi imacy o
echnology, and willingness o ac upon da a-d i en o ecas s (Blu and Wang, 2019; Son and Han, 2011). Some s udies sugges
ha e en communi ies wi h limi ed digi al in as uc u e can s ill bene i om disas e echnologies i hey demons a e high le els
o us , collec i e p epa edness, and social accep ance o echnological inpu s (Peng e al., 2020; Rayamajhee and Boha a, 2020).
The coexis ence o hese wo pe spec i es illus a es ha eadiness is no educible o a single dimension. Ins ead, i is a mul i-
laye ed cons uc ha combines bo h ma e ial access and he beha io al and social condi ions necessa y o ope a ionalize
echnological ools (Blu and Wang, 2019). Cla i ying his dual na u e is essen ial, as i shapes how communi y eadiness is
posi ioned wi hin disas e p e en ion esea ch whe he as a ma e o in as uc u e p o ision o as a b oade ques ion o social
willingness and ins i u ional us .
Communi y eadiness c i ically mode a es he ela ionship be ween IoT-based en i onmen al moni o ing co e age and
disas e p e en ion e ec i eness. While b oade senso ne wo ks expand he po en ial o si ua ional awa eness, he ac ual
p e en i e impac depends on whe he communi ies a e p epa ed and willing o ac on he da a (Ryan e al., 2020). Low eadiness
can weaken o nulli y he bene i s o IoT sys ems: dense ne wo ks may gene a e accu a e, eal- ime da a, bu wi hou us , digi al
li e acy, o economic capaci y, wa nings may go unheeded. Con e sely, high eadiness ampli ies he e ec i eness o IoT co e age
by ensu ing apid in e p e a ion, collec i e mobiliza ion, and p o ec i e beha io (Sinha e al., 2019). Despi e he ele ance o his
dynamic, limi ed empi ical esea ch has examined eadiness h esholds in disas e con ex s, lea ing unclea whe he ac o s like
us , a o dabili y, and cul u al accep ance sys ema ically enhance o cons ain IoT’s p e en i e alue. Add essing his gap is
cen al o unde s anding how echnological in as uc u e in e ac s wi h human and social sys ems, and highligh s why communi y
eadiness should be posi ioned as a mode a ing a iable in disas e echnology adop ion esea ch.
D awing om he heo e ical and empi ical ounda ions, he s udy de elops he hi d hypo hesis p esen ed below:
H3: Communi y eadiness o echnology adop ion posi i ely mode a es he impac o IoT-based en i onmen al
moni o ing co e age on disas e p e en ion e ec i eness.
Buil on es ablished heo e ical ounda ions, his s udy enhances i s academic alue by p esen ing he ollowing
concep ual amewo k:
Figu e 1. The Pape ’s Concep ual F amewo k. Sou ce: (The au ho s, 2025)
III. METHODOLOGY
This s udy adop s a quan i a i e esea ch design o in es iga e he ela ionship be ween AI-d i en lood p edic ion
accu acy, IoT-based en i onmen al moni o ing co e age, and disas e p e en ion e ec i eness, wi h communi y eadiness o
echnology adop ion as a mode a ing ac o . A pu posi e s a i ied sampling s a egy was employed o ensu e ep esen a ion
om di e se p o essional, ins i u ional, and communi y g oups di ec ly in ol ed in he de elopmen , deploymen , and use o
disas e p e en ion echnologies.
Pa icipan s we e d awn om h ee coun ies Vie nam, Singapo e, and Malaysia selec ed o hei high exposu e o
clima e-induced haza ds, apid adop ion o AI and IoT inno a ions, and di e ing le els o communi y eadiness in disas e
go e nance. The sample was s a i ied in o ou s akeholde g oups: (1) sys em enginee s and da a scien is s engaged in
de eloping AI-based disas e p edic ion algo i hms; (2) ICT and in as uc u e manage s o e seeing IoT-enabled moni o ing
B idging Technology and Readiness: AI, IoT, and he E ec i eness o Disas e P e en ion in Clima e-Vulne able
Regions
JEFMS, Volume 08 Issue 10 Oc obe 2025 www.ije m.co.in Page 6588
sys ems; (3) disas e managemen p ac i ione s, eme gency esponse o icials, and policy make s esponsible o in eg a ing
echnological ou pu s in o p e en i e ac ion; and (4) communi y ep esen a i es and academic esea che s ocusing on disas e
esilience, echnology adop ion, and socio- echnical go e nance. Eligibili y c i e ia equi ed pa icipan s o ha e a leas one yea
o di ec expe ience in disas e isk managemen , AI/IoT deploymen , o communi y-based esilience p og ams, as well as
amilia i y wi h p edic i e modeling, en i onmen al moni o ing, o disas e p e en ion s a egies. This ensu ed ha esponses
e lec ed bo h echnical expe ise and p ac ical engagemen wi h disas e echnologies.
Da a was collec ed h ough an online s uc u ed ques ionnai e using a 5-poin Like scale (1 = “S ongly disag ee” o 5 =
“S ongly ag ee”). The ins umen was designed o measu e ou cons uc s: (1) AI-d i en p edic ion accu acy, (2) IoT-based
moni o ing co e age, (3) pe cei ed disas e p e en ion e ec i eness, and (4) communi y eadiness o echnology adop ion. The
su ey was dis ibu ed ia p o essional ne wo ks, egional ICT and disas e managemen associa ions, esea ch ins i u es, and
communi y-based o ganiza ions. In Vie nam, dissemina ion was suppo ed by collabo a ions wi h he Na ional S ee ing
Commi ee o Na u al Disas e P e en ion and uni e si y labs specializing in AI and disas e esea ch. In Singapo e and Malaysia,
dis ibu ion was acili a ed h ough egional disas e managemen councils, ICT indus y associa ions, and LinkedIn p o essional
communi ies ocusing on sma ci ies and disas e echnologies. A o al o 600 esponses we e ecei ed, o which 385 alid cases
we e e ained a e sc eening o comple eness, ele ance, and ole-speci ic expe ise. The inal sample achie ed balanced
ep esen a ion ac oss s akeholde ca ego ies and coun ies, ensu ing a obus empi ical basis o examining how AI-d i en lood
p edic ion, IoT moni o ing, and communi y eadiness join ly shape disas e p e en ion e ec i eness in clima e- ulne able egions.
IV. RESULTS
4.1. Reliabili y analysis
Table 1: Reliabili y analysis o he dependen a iable. Sou ce: (The au ho s, 2025)
Reliabili y S a is ics
C onbach's
Alpha
N o
I ems
.743
4
I em-To al S a is ics
Scale
Mean i I em
Dele ed
Scale
Va iance i I em
Dele ed
Co ec ed
I em-To al
Co ela ion
C onbach's
Alpha i I em
Dele ed
DPE1
6.315
8.211
.672
.683
DPE2
6.023
8.100
.610
.704
DPE3
6.767
7.810
.678
.689
DPE4
7.098
8.096
.664
.675
Whe e DPE1 o DPE4 a e coded o su ey ques ions 1 o 4 o Disas e P e en ion E ec i eness espec i ely.
In Table 1, he dependen obse ed a iables all p oduced co ec ed i em– o al co ela ion alues abo e he 0.3 h eshold.
The o e all C onbach’s alpha o 0.743 su passed he con en ional eliabili y s anda d o 0.7 and emained highe han he alues
ha would ha e esul ed i any i em had been emo ed. Mo eo e , each obse ed a iable eco ded an alpha coe icien g ea e
han i s co esponding adjus ed i em– o al co ela ion, e en unde hypo he ical i em dele ions. Consequen ly, all ou i ems we e
e ained o subsequen analysis. Compa able eliabili y ou comes we e also obse ed ac oss he emaining cons uc s.
4.2. Explo a o y ac o analysis (EFA)
Table 2: Ro a ed Componen Ma ix. Sou ce: (The au ho s, 2025)
Ro a ed Componen Ma ixa
Componen wi h loading ac o s
1
2
3
4
DPE1 .816
DPE2 .842
DPE3 .850
FPA1 .676
FPA2 .629
FPA3 .683
EMC1 .752
EMC2 .754
EMC3 .691
CRTA1 .669
CRTA2 .685
CRTA3 .658
B idging Technology and Readiness: AI, IoT, and he E ec i eness o Disas e P e en ion in Clima e-Vulne able
Regions
JEFMS, Volume 08 Issue 10 Oc obe 2025 www.ije m.co.in Page 6589
DPE4 .841
FPA4 .599
EMC4 .700
CRTA4 .705
Ex ac ion Me hod: P incipal Componen Analysis.
Ro a ion Me hod: Va imax wi h Kaise No maliza ion.
a. Ro a ion con e ged in 7 i e a ions.
Whe e he su ey i ems FPA1–FPA4, EMCI1–EMC4, and CRTA1–CRTA4 we e de eloped o code he wo independen
a iables and he mode a o a iable, wi h ou i ems alloca ed o each cons uc .
Table 2 indica es ha he o a ed componen ma ix success ully clus e ed he 16 obse ed a iables in o ou clea
ac o s ep esen ing he dependen a iable, he wo independen a iables, and he mode a o . Each obse ed a iable achie ed
a ac o loading g ea e han 0.5, and all i ems we e e ained in he analysis.
4.3. Mul iple linea eg ession model
Table 3: Coe icien sa. Sou ce: (The au ho s, 2025)
Model
Uns anda dized
Coe icien s
S anda dized
Coe icien s
Sig.
B
S d. E o
Be a
1
(Cons an )
7.030
.862
4.333
.000
FPA
.563
.745
.551
3.722
.000
EMC
.356
.892
.347
3.568
.000
a. Dependen Va iable: DPE
Whe e DPE: mean o DPE1 o DPE4; FPA: mean o FPA1–FPA4;
EMC: mean o EMCI1–EMC4
As shown in Table 3, he - es p oduced signi icance (Sig.) alues o .000, which a e lowe han he con en ional alpha
h eshold o 0.05. This indica es ha he independen a iables exe a s a is ically signi ican impac on he dependen a iable.
Consequen ly, bo h hypo heses a e suppo ed by he empi ical e idence.
4.4. Mode a o analysis
Table 4: Resul s analysis o “Communi y Readiness o Technology Adop ion”. Sou ce: (The au ho s, 2025)
Model : 1
Y : DPE
X : FPA
W : CRTA
Sample Size: 385
**************************************************************************
OUTCOME VARIABLE:
DPE
Model Summa y
R
R-sq
MSE
F
dl1
dl2
p
.693
.480
.536
5.932
3.000
381.000
.000
Model
coe
se
p
LLCI
ULCI
cons an
5.600
.775
60.006
.000
7.455
7.356
FPA
.516
.830
4.444
.000
.588
.576
CRTA
.523
.708
4.932
.000
.600
.582
In _1
.392
.438
4.788
.000
.585
.563
Whe e CRTA: mean o CRTA1 o CRTA4
B idging Technology and Readiness: AI, IoT, and he E ec i eness o Disas e P e en ion in Clima e-Vulne able
Regions
JEFMS, Volume 08 Issue 10 Oc obe 2025 www.ije m.co.in Page 6590
As shown in Table 4, he in e ac ion e m (In _1) p oduced a p- alue o 0.000, which is signi ican ly below he 0.05
benchma k. This con i ms a s a is ically signi ican in e ac ion be ween communi y eadiness o echnology adop ion and AI-
d i en lood p edic ion accu acy in in luencing Disas e P e en ion E ec i eness. The in e ac ion coe icien o .392 indica es ha
highe le els o communi y eadiness o echnology adop ion in ensi y he posi i e e ec o AI-d i en lood p edic ion accu acy
on Disas e P e en ion E ec i eness. The e o e, hypo hesis H3 is empi ically suppo ed.
V. DISCUSSION
5.1. Summa y esul s
The esul s om he linea eg ession analysis e eal ha AI-d i en lood p edic ion accu acy has he s onges in luence
on disas e p e en ion e ec i eness, wi h a coe icien o 0.551. In compa ison, IoT-based en i onmen al moni o ing co e age
also con ibu es posi i ely, hough o a lesse ex en , as e lec ed by i s coe icien o 0.347. Addi ionally, he le el o communi y
eadiness o echnology adop ion se es as a mode a ing ac o , exe ing a posi i e in luence on he ela ionship be ween
communi y eadiness o echnology adop ion and disas e p e en ion e ec i eness, wi h a mode a ion coe icien o 0.392. This
analy ical app oach was employed o comp ehensi ely add ess he esea ch ques ions and elucida e he in e connec ions among
he p incipal a iables.
5.2. Theo e ical implica ion
The indings s ongly a i m ha AI-d i en lood p edic ion accu acy enhances disas e p e en ion e ec i eness, bu his
suppo is nuanced. On one hand, esul s concu wi h Ali e al. (2022) and Islam e al. (2020), who a gued ha ad anced models
signi ican ly imp o e o ecas ing eliabili y. Ou e idence also ein o ces Nea ing e al. (2024), emphasizing ha p edic i e sys ems
p o ide ac ionable lead imes essen ial o imely e acua ion. Howe e , his s udy depa s om Fang e al. (2020) and Sande s e
al. (2022), who equa ed e ec i eness wi h s a is ical p ecision alone. The esul s ins ead highligh ope a ional usabili y, aligning
wi h Adika i e al. (2021), who s essed ha impe ec bu imely wa nings can sa e mo e li es han echnically lawless ye delayed
o ecas s. This esea ch he e o e con es s educ ionis no ions o “accu acy” and suppo s a mul idimensional de ini ion ha
b idges echnical p ecision wi h social usabili y. Thus, while b oadly consis en wi h op imism abou AI, he indings c i ically
oppose claims ha nume ical accu acy alone gua an ees p e en i e alue.
The e idence pa ially concu s wi h schola ship ha equa es wide IoT co e age wi h g ea e disas e p e en ion
(Na ayana e al., 2024; Heimann e al., 2015). Consis en wi h Ray, Mukhe jee and Shu (2017), b oade ne wo ks do imp o e
si ua ional awa eness. Ye , he indings c i ically challenge he dominan assump ion, oiced by Fa hy, Ba naghi and Ta azolli
(2018), ha senso densi y i sel cons i u es e ec i eness. Ins ead, his s udy co obo a es Shah e al. (2019) and Yabe e al.
(2022), showing ha in o ma ion quali y, ep esen a i eness, and in e p e abili y ma e mo e han shee quan i y. The esul s
also echo Ejaz e al. (2019), sugges ing ha e en small, s a egically posi ioned ne wo ks can ou pe o m dense bu poo ly
in eg a ed sys ems. By doing so, he s udy unse les echno-de e minis ic claims ha co e age expansion au oma ically ansla es
o esilience, showing ins ead ha socio- echnical alignmen and usabili y go e n p e en i e ou comes. Thus, while ag eeing ha
IoT co e age is in luen ial, he pape dispu es simplis ic me ics and a gues o a e aming o co e age as e ec i e moni o ing
a he han me e de ice p oli e a ion.
The indings s ongly suppo he mode a ing ole o communi y eadiness, aligning wi h Sinha e al. (2019) and Ryan e
al. (2020), who highligh ed ha echnological in as uc u e alone canno yield e ec i eness wi hou social willingness o ac . This
esea ch concu s wi h B a e al. (2022), emphasizing ha us and in e p e i e capaci y ampli y IoT bene i s. Howe e , he esul s
di e ge om Blu and Wang (2019), who na owly equa ed eadiness wi h in as uc u e p o ision. E idence ins ead con i ms he
b oade pe spec i e o Peng e al. (2020) and Rayamajhee and Boha a (2020), showing ha e en communi ies wi h limi ed
in as uc u e can exhibi high eadiness i hey possess us and collec i e ac ion capaci y. Impo an ly, he s udy con es s
echno-cen ic a gumen s ha assume IoT sys ems succeed ega dless o social adop ion. Ins ead, indings demons a e ha
eadiness undamen ally shapes whe he moni o ing ou pu s ansla e in o p e en i e ac ion. Thus, he esea ch ad ances an
a gumen a i e s ance ha posi ions eadiness as he decisi e socio- echnical hinge upon which IoT’s p e en i e alue depends.
5.3. P ac ical implica ion
The indings unde line ha disas e go e nance bodies mus mo e beyond na ow eliance on s a is ical benchma ks o
AI-d i en lood p edic ion and ins ead p io i ize ope a ional usabili y. While Ali e al. (2022) and Islam e al. (2020) highligh
accu acy as a echnical achie emen , his s udy shows ha i s eal alue lies in ea ly ac ionable ale s ha sa e li es. Policymake s
should hus ins i u ionalize AI ou pu s in o wa ning p o ocols, ensu ing ha o ecas esul s each communi ies in accessible
o ma s and lead imes (Adika i e al., 2021). This equi es in es men s no only in model sophis ica ion bu also in communica ion
