Co esponding au ho : Osi a Vic o Egwua u
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.
Al-D i en Decision Suppo Sys ems o Business S a egy
Osi a Vic o Egwua u *
MBA (In o ma ion Sys ems), College o Business and Inno a ion, The Uni e si y o Toledo, Toledo, Ohio Uni ed S a es.
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 27(02), 1752-1769
Publica ion his o y: Recei ed on 16 July 2025; e ised on 24 Augus 2025; accep ed on 26 Augus 2025
A icle DOI: h ps://doi.o g/10.30574/wja .2025.27.2.3065
Abs ac
This s udy explo es how a i icial in elligence (AI), speci ically machine lea ning (ML), ans o ms Decision Suppo
Sys ems (DSS) om desc ip i e ools in o p edic i e and p esc ip i e engines o s a egic decision-making. Using a
case s udy in he e ail sec o , s uc u ed (sales, inancials) and uns uc u ed ( e iews, social media) da a we e analyzed
h ough supe ised lea ning, na u al language p ocessing, and ein o cemen lea ning models. Findings show imp o ed
p edic i e accu acy, cus ome e en ion, and sus ainable p icing s a egies compa ed o adi ional IS/MBAs
amewo ks. The esea ch con ibu es heo e ically by ex ending DSS and IS li e a u e, and p ac ically by p o iding
business leade s wi h ac ionable, AI-d i en amewo ks o long- e m s a egic agili y.
Keywo ds: A i icial In elligence Go e nance; En e p ise In o ma ion Sys ems; Algo i hmic Bias; Da a E hics and
P i acy; Responsible AI Adop ion; Regula o y Compliance
1. In oduc ion
1.1. Con ex : The Rise o A i icial In elligence in Business En i onmen s
In ecen decades, a i icial in elligence (AI) and machine lea ning (ML) ha e mo ed om expe imen al echnologies in
academic and labo a o y se ings o mains eam ools d i ing inno a ion ac oss indus ies. Businesses a e inc easingly
ha nessing AI no only o au oma e ou ine ope a ions bu also o enhance highe -o de unc ions such as o ecas ing,
isk assessmen , and s a egic decision-making. This shi e lec s a b oade digi al ans o ma ion in which
o ganiza ions compe e no only on physical asse s bu also on hei abili y o gene a e, p ocess, and in e p e da a.
Acco ding o epo s by McKinsey and Ga ne , i ms ha embed AI in o hei business p ocesses ou pe o m pee s in
e enue g ow h and ma ke posi ioning, sugges ing a s ong compe i i e p emium o ea ly adop ion.
In his e ol ing en i onmen , manage s ace unp eceden ed complexi y. Global supply chains, apidly changing
consume beha io s, egula o y unce ain ies, and dis up i e compe i o s demand decisions ha a e as e , mo e
accu a e, and mo e adap i e han adi ional me hods allow. The shee olume o s uc u ed and uns uc u ed da a
a ailable o o ganiza ions, om ansac ion eco ds o social media sen imen , c ea es bo h oppo uni y and challenge.
While his da a can yield insigh s ha in o m s a egy, ex ac ing meaning ul knowledge equi es compu a ional
app oaches a beyond he capaci y o con en ional analy ics.
1.2. P oblem: Limi a ions o T adi ional Decision Suppo Sys ems
Decision Suppo Sys ems (DSS) eme ged in he 1970s as in o ma ion sys ems designed o assis manage s in making
semi-s uc u ed o uns uc u ed decisions. Ea ly DSS inco po a ed da abases, s a is ical ools, and wha -i analyses,
o e ing alue by o ganizing in o ma ion and p o iding amewo ks o decision analysis. Howe e , adi ional DSS
o en ell sho in wo c i ical a eas: p edic i e powe and p esc ip i e guidance.
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 27(02), 1752-1769
1753
Fi s , adi ional DSS we e p ima ily desc ip i e. They summa ized pas pe o mance and enabled scena io analysis bu
a ely an icipa ed u u e ou comes wi h accu acy. Thei eliance on ule-based logic, de e minis ic models, o linea
s a is ical app oaches limi ed hei abili y o handle nonlinea i ies, high-dimensional da a, and complex
in e dependencies ha cha ac e ize mode n business ecosys ems.
Second, hese sys ems o en lacked adap abili y. Business en i onmen s e ol e apidly, ye con en ional DSS models
equi e manual upda es and a e no capable o lea ning dynamically om new da a. As a esul , manage s using such
sys ems isked basing hei s a egies on ou da ed assump ions o incomple e ep esen a ions o eali y.
The g owing sophis ica ion o AI o e s a compelling solu ion. Machine lea ning models, especially supe ised and
unsupe ised lea ning algo i hms, deep neu al ne wo ks, and ein o cemen lea ning sys ems, can iden i y sub le
pa e ns in la ge da ase s, upda e hemsel es as new in o ma ion eme ges, and gene a e p edic ions and
ecommenda ions wi h inc easing accu acy. These capabili ies shi DSS om being p ima ily desc ip i e o becoming
p edic i e and p esc ip i e ools ha di ec ly in o m s a egic choices.
1.3. Aim: Towa d AI-D i en Decision Suppo o S a egy
The cen al aim o his esea ch is o explo e and demons a e how AI-d i en Decision Suppo Sys ems (AI-DSS),
unde pinned by machine lea ning models, can guide manage ial decision-making a he s a egic le el. Unlike
ope a ional o ac ical decisions, which o en in ol e sho e ime ho izons and na owe scopes, s a egic decisions
conce n long- e m di ec ion, esou ce alloca ion, compe i i e posi ioning, and o e all o ganiza ional su i al. These
decisions equi e syn hesizing as amoun s o he e ogeneous in o ma ion and e alua ing mul iple unce ain u u es
asks o which ML-enhanced DSS a e uniquely well-sui ed.
Speci ically, his s udy seeks o:
• Examine how machine lea ning models can be e ec i ely in eg a ed in o DSS a chi ec u es o enhance
p edic i e and p esc ip i e capabili ies.
• Apply AI-DSS wi hin a business case s udy o illus a e p ac ical applica ions and ou comes.
• Assess he bene i s and limi a ions o AI-DSS o s a egic decision-making, wi h pa icula a en ion o isks
such as algo i hmic bias, in e p e abili y challenges, and o ganiza ional adop ion ba ie s.
By combining concep ual de elopmen wi h empi ical illus a ion, he a icle posi ions AI-DSS as no me ely a
echnological inno a ion bu a ans o ma i e manage ial ool.
1.4. Con ibu ions: Theo e ical and P ac ical
This esea ch makes con ibu ions on bo h heo e ical and p ac ical on s.
Theo e ically, i ex ends he In o ma ion Sys ems (IS) li e a u e by in eg a ing AI and ML in o he longs anding DSS
pa adigm. While DSS esea ch has adi ionally emphasized da a managemen , in e ace design, and decision heu is ics,
he in usion o ML models in oduces new capabili ies ha ede ine he bounda ies o IS schola ship. This s udy also
con ibu es o he eme ging li e a u e on digi al s a egy by si ua ing AI-DSS as a key enable o dynamic capabili ies,
esou ce o ches a ion, and o ganiza ional agili y.
P ac ically, he a icle p oposes a amewo k o designing and implemen ing AI-d i en DSS ailo ed o s a egic
managemen con ex s. By p esen ing a eal-wo ld case s udy, i demons a es how p edic i e modeling can be
ope a ionalized in manage ial se ings and how ou pu s can be ansla ed in o ac ionable s a egic insigh s. Mo eo e ,
he discussion o isks and limi a ions p o ides manage s wi h balanced guidance, helping hem na iga e e hical,
echnical, and o ganiza ional challenges.
1.5. Resea ch Ques ions
To ope a ionalize i s aims, his s udy is guided by he ollowing esea ch ques ions:
• In eg a ion Ques ion: How can machine lea ning models be inco po a ed in o decision suppo sys ems o
enhance hei abili y o guide s a egic decision-making?
• Value Ques ion: Wha angible business bene i s can o ganiza ions de i e om adop ing AI-d i en DSS in
s a egy o mula ion and execu ion?
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 27(02), 1752-1769
1754
• Risk Ques ion: Wha isks and challenges a ise om eliance on AI-d i en DSS, pa icula ly conce ning
algo i hmic bias, in e p e abili y, and manage ial us ?
These ques ions ame he in es iga ion, ensu ing ha he analysis emains g ounded in bo h echnological capabili ies
and manage ial eali ies.
1.6. S uc u e o he A icle
The emainde o he a icle is o ganized as ollows. Sec ion 3 e iews ele an li e a u e on DSS, AI, and business
s a egy, iden i ying key gaps his s udy add esses. Sec ion 4 p esen s he heo e ical amewo k ha links AI-DSS o
es ablished IS and s a egy heo ies. Sec ion 5 ou lines he me hodology, combining case s udy esea ch wi h p edic i e
modeling. Sec ion 6 p o ides he case s udy indings, while Sec ion 7 discusses esul s and implica ions. Finally, Sec ion
8 concludes wi h a summa y o con ibu ions and sugges ions o u u e esea ch.
2. Li e a u e Re iew
2.1. E olu ion o Decision Suppo Sys ems (DSS)
Decision Suppo Sys ems (DSS) ha e hei oo s in he b oade ield o Managemen In o ma ion Sys ems (MIS) ha
eme ged in he mid-20 h cen u y. Ea ly MIS in he 1960s and 1970s we e p ima ily ocused on elec onic da a
p ocessing and ou ine epo ing, p o iding manage s wi h pe iodic summa ies o business in o ma ion. The concep
o DSS began o ake shape as a esponse o he need o mo e in e ac i e analy ical ools o assis in decision-making.
By he la e 1960s and ea ly 1970s, esea che s in oduced DSS as compu e -based sys ems designed o aid manage s
in sol ing semi-s uc u ed and uns uc u ed p oblems by combining da a, analy ical models, and use - iendly so wa e
in e aces. These ea ly DSS we e ela i ely simple – o en buil on basic models (like sp eadshee s o udimen a y
s a is ical p og ams) – bu hey laid he ounda ion o mo e sophis ica ed decision suppo by allowing “wha -i ”
analysis and scena io planning beyond wha s anda d MIS epo s could o e .
As compu ing echnology ad anced, DSS capabili ies expanded h ough he 1980s and 1990s. One signi ican
de elopmen was he ise o expe sys ems, which can be iewed as a o m o knowledge-d i en DSS. Expe sys ems
eme ged om a i icial in elligence esea ch and sough o cap u e human expe ise in a se o ules and in e ence
mechanismsdss esou ces.com. Unlike adi ional model-d i en DSS ha ollowed p ede ined ma hema ical models,
expe sys ems a emp ed o simula e human easoning by using a knowledge base o ac s and ules, along wi h an
in e ence engine o apply logical ules o hose ac sdss esou ces.com. These sys ems could p o ide ecommenda ions
o diagnoses o speci ic p oblem domains ( o example, medical diagnosis o mine al explo a ion) by mimicking he
decision p ocesses o human expe s. In p ac ice, expe sys ems ep esen ed an ea ly in usion o AI in o decision
suppo , enabling compu e s o handle quali a i e knowledge and in e ence. Howe e , hey we e ypically limi ed o
na ow asks and equi ed ex ensi e knowledge enginee ing (manual encoding o expe knowledge), which made hem
challenging o build and main ain. The 1980s also saw he p oli e a ion o o he DSS- ela ed ools such as Execu i e
In o ma ion Sys ems (EIS) o op-le el dashboa ds and G oup Decision Suppo Sys ems (GDSS) o collabo a i e
decision mee ings, each add essing di e en needs in he decision-making hie a chy. These a ie ies o DSS ma ked an
e olu ion om basic MIS epo ing owa d mo e specialized suppo o decision p ocesses a ope a ional, ac ical, and
s a egic le els.
By he la e 1990s and 2000s, business in elligence (BI) and da a wa ehousing echnologies became p ominen , u he
ans o ming DSS. BI sys ems buil upon da a in eg a ion and da a mining echniques o p o ide comp ehensi e
analysis ac oss di e en ace s o he o ganiza ion. They could handle la ge olumes o his o ical da a, p oduce mul i-
dimensional epo s, and unco e pa e ns o ends o in o m decisions. In essence, BI b oadened he scope o decision
suppo by inco po a ing semi-s uc u ed and uns uc u ed da a (e.g. om ansac ions, cus ome in e ac ions, e c.)
and deli e ing insigh s h ough dashboa ds and isualiza ions. The ocus was on enabling da a-d i en decision-making
ac oss all managemen ie s, imp o ing no jus ope a ional decisions bu also in o ming s a egy by iden i ying key
pe o mance d i e s.
Mos ecen ly, he e olu ion o DSS has accele a ed wi h he ad en o ad anced a i icial in elligence and machine
lea ning echniques essen ially AI-d i en DSS. Mode n DSS a e inc easingly in eg a ed wi h AI algo i hms, p edic i e
models, and big da a analy ics o p o ide a mo e powe ul insigh s and ecommenda ions han ea lie gene a ions o
decision suppo ools. These AI-d i en sys ems can analyze massi e, complex da ase s in eal ime, lea n om new
da a, and e en pe o m au onomous decision-making in ce ain con ex s. Fo example, inco po a ing deep lea ning has
gi en DSS “quan um leap” imp o emen s in p edic i e accu acy and adap abili y, enabling mo e p ecise o ecas s and
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 27(02), 1752-1769
1755
pa e n ecogni ion han adi ional s a is ical models. Today’s AI-d i en DSS can con inuously e ine hei
ecommenda ions as new da a lows in, and hey o en include na u al language in e aces o con e sa ional agen s ha
make hem easie o manage s o in e ac wi h. C ucially, he in eg a ion o AI has ex ended decision suppo in o a eas
ha equi e highe -o de cogni ion such as s a egic planning and isk assessmen – which ea lie MIS and DSS
s uggled o add ess. In ac , mode n AI capabili ies a e c edi ed wi h o e ing o ganiza ions a compe i i e edge by
acili a ing as e da a-d i en decisions, enhancing ope a ional e iciency, and e en c ea ing pe sonalized cus ome
expe iences a scale. The capabili ies o AI-d i en DSS ange om eal- ime analy ics and inno a i e p oduc
de elopmen o isk p edic ion and long- e m s a egic o ecas ing. As o ganiza ions in eg a e AI in o hei decision
suppo in as uc u e, hey no only op imize in e nal p ocesses bu can also c ea e highe ba ie s o en y o
compe i o s, e ec i ely posi ioning hemsel es as indus y leade s h ough supe io decision in elligence.
2.2. Machine Lea ning in Business Decision-Making
Machine lea ning (ML), a co e subse o AI, has become a d i ing o ce in con empo a y decision suppo sys ems and
business analy ics. While ea lie DSS elied on p ede ined models o expe -c a ed ules, ML echniques enable sys ems
o lea n pa e ns om da a and imp o e o e ime wi hou being explici ly p og ammed o each scena io. In business,
ML is applied ac oss a wide ange o unc ions om o ecas ing and ma ke ing o ope a ions and cus ome se ice
essen ially b inging a p edic i e, adap i e edge o decision-making p ocesses. Below, we discuss key ML app oaches in
business and hei impac :
P edic i e Analy ics: One o he mos common applica ions o ML in business is p edic i e analy ics, which in ol es
using his o ical da a and s a is ical algo i hms (including machine lea ning models) o p edic u u e ou comes.
P edic i e analy ics ans o ms aw da a in o o wa d-looking insigh s, helping companies o ecas ends, cus ome
beha io s, and isks so hey can ac p oac i ely a he han eac i ely. Fo ins ance, o ganiza ions use p edic i e models
o an icipa e sales demand, iden i y which cus ome s a e likely o chu n, o de ec audulen ansac ions be o e hey
occu . By le e aging echniques such as eg ession analysis, ime-se ies o ecas ing, o machine lea ning classi ie s,
businesses can align hei s a egies wi h likely u u e scena ios. The bene i is imp o ed decision accu acy and iming
– companies can add ess oppo uni ies and challenges p oac i ely, gaining a compe i i e ad an age by s aying one s ep
ahead o ma ke changes. In ac , i ms ha e ec i ely use p edic i e analy ics o en epo be e alignmen o hei
ope a ions wi h s a egic goals and an enhanced capaci y o isk mi iga ion h ough ea ly wa nings. In summa y,
p edic i e analy ics powe ed by ML allows da a-d i en o esigh in decision-making, om inance (e.g. c edi sco ing,
in es men p edic ions) o supply chain (demand o ecas ing) and ma ke ing ( a ge ed campaigns).
Na u al Language P ocessing (NLP): NLP is a b anch o AI/ML ocused on enabling compu e s o unde s and and
gene a e human language. In business, NLP echniques unlock he alue o as amoun s o uns uc u ed ex and speech
da a – cus ome e iews, social media pos s, suppo emails, call ansc ip s, epo s, and mo e. By p ocessing his da a,
NLP can e eal insigh s ha would be ha d o ob ain o he wise. Applica ions include sen imen analysis (gauging
cus ome opinions and b and sen imen ), au oma ed cus ome se ice cha bo s and i ual assis an s, machine
ansla ion o global ope a ions, and ex analy ics o asks like con ac analysis o esume sc eening. NLP hus helps
o ganiza ions lis en o and espond o s akeholde s a scale. Impo an ly, NLP has become “indispensable o
main aining a compe i i e edge in oday’s dynamic business en i onmen ”. I allows companies o apidly analyze public
pe cep ion and ma ke ends om ex ual da a, pe sonalize con en o ecommenda ions o use s, and s eamline
ope a ions such as documen p ocessing and epo ing. Fo example, deploying NLP-d i en cha bo s can p o ide
ins an 24/7 cus ome suppo , imp o ing se ice quali y while educing cos s – an ad an age in compe i i e ma ke s.
Simila ly, sen imen analysis can ale i ms o eme ging issues in cus ome sa is ac ion o p oduc epu a ion in eal
ime, enabling a as s a egic esponse. The in eg a ion o NLP in o business p ocesses ul ima ely os e s a deepe
connec ion be ween companies and hei cus ome s by b idging human communica ion wi h machine in elligence,
leading o mo e in o med decisions and ailo ed expe iences.
Rein o cemen Lea ning (RL): Rein o cemen lea ning is an ML pa adigm whe e an au onomous agen lea ns o make
sequences o decisions by in e ac ing wi h an en i onmen and ecei ing eedback in he o m o ewa ds o penal ies.
O e many i e a ions, he agen lea ns an op imal policy (s a egy) o maximize cumula i e ewa ds. In a business
con ex , RL is especially powe ul o complex, dynamic decision p oblems whe e he e may no be a single-s ep
p edic ion a ge , bu a he a need o op imize long- e m ou comes h ough a se ies o in e dependen decisions. Use
cases include dynamic p icing s a egies, eal- ime supply chain and logis ics op imiza ion, adap i e con ol sys ems in
manu ac u ing, ecommenda ion sys ems ha adjus o use beha io , and any scena io whe e decisions ha e a delayed
impac ha needs o be lea ned. The appeal o RL in business is ha i can disco e no el and adap i e solu ions in
en i onmen s oo complex o ule-based p og amming. Unlike adi ional p og ams, an RL agen is no explici ly old
how o eac o e e y si ua ion; i lea ns om expe ience and explo a ion. This means RL can some imes ou pe o m
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 27(02), 1752-1769
1756
human decision-make s in high-dimensional p oblems by conside ing a as ange o possibili ies and lea ning om
ial and e o . Fo example, e aile s ace apidly changing consume p e e ences and ma ke condi ions ha make
s a ic o ecas ing di icul .
2.3. In o ma ion Sys ems and Business S a egy: Linking Technology Adop ion o Compe i i e Ad an age
In o ma ion Sys ems (IS) and in o ma ion echnology mo e b oadly ha e long been ecognized as key enable s o
compe i i e ad an age in business. A ich body o li e a u e examines amewo ks and models ha explain how
adop ing echnology can ansla e in o imp o ed pe o mance, ma ke posi ion, and s a egic di e en ia ion.
Fundamen ally, hese amewo ks a gue ha echnology is no me ely a suppo ool o execu ing s a egy, bu can be
an in eg al pa o shaping and enhancing a i m’s s a egy.
One ounda ional pe spec i e is Michael Po e ’s iew on IT in compe i ion. Po e ’s amewo ks (such as he Value
Chain and Fi e Fo ces) highligh how IS can c ea e ad an ages ei he by lowe ing a i m’s cos s uc u e o enabling
di e en ia ion. Fo example, in eg a ing IS in o he alue chain can s eamline p ocesses (p ocu emen , logis ics,
p oduc ion, ma ke ing, e c.), yielding cos leade ship ad an ages, o enable supe io cus ome insigh s and p oduc
inno a ions, yielding di e en ia ion ad an ages. Classic cases o en ci ed include Walma ’s use o in o ma ion sys ems
o supply chain op imiza ion (achie ing low-cos leade ship) and Amazon’s use o da a-d i en pe sonaliza ion
(di e en ia ing h ough cus ome expe ience). These examples unde sco e ha alignmen be ween echnology and
business p ocesses can di ec ly bols e a company’s compe i i e s a egy. IS can imp o e ope a ional e iciency,
enhance cus ome se ice, os e inno a ion, and enable as e decision-making all iden i ied as key sou ces o
compe i i e ad an age in mode n ma ke s.
Beyond indi idual cases, o mal models like he S a egic Alignmen Model (SAM) by Hende son and Venka aman
p o ide a heo e ical amewo k o linking echnology adop ion o compe i i e ad an age. The SAM posi s ha o ully
ealize alue om IT in es men s, an o ganiza ion mus achie e alignmen be ween i s IT s a egy and i s business
s a egy. In o he wo ds, echnology ini ia i es should be di ec ly d i en by business objec i es and ice e sa. This
model iden i ies mul iple domains (business s a egy, IT s a egy, o ganiza ional in as uc u e, and IT in as uc u e)
and a gues ha cohe ence ac oss hese domains is c i ical. The a ionale is ha misalignmen o ins ance, adop ing a
cu ing-edge echnology wi hou a clea business s a egic need will yield subop imal esul s, whe eas igh alignmen
can p oduce syne gis ic gains. Hende son and Venka aman de eloped SAM speci ically o add ess he “g owing need
o o ganiza ions o e ec i ely exploi IT capabili ies o compe i i e ad an age and manage he inc easing complexi y
o aligning echnology wi h business goals”. I has since become a co ne s one in IS s a egy esea ch and p ac ice,
ein o cing he idea ha echnology adop ion mus be guided by s a egy (and can e en in o m new s a egic
oppo uni ies) o c ea e sus ainable success.
Ano he impo an pe spec i e comes om he inno a ion and di usion o echnology angle. Ea ly adop e s o
ans o ma i e echnologies can o en gain a empo al compe i i e edge, a concep ela ed o i s -mo e ad an age.
Businesses ha a e quick o emb ace eme ging echnologies (such as AI, cloud compu ing, o IoT in ecen imes) may
eap bene i s like e iciency gains, new p oduc o se ice models, and posi i e b anding as inno a o s. These bene i s
can ansla e in o ma ke sha e g ow h o p o i abili y bumps ha lagga ds s uggle o ma ch. Fo example, companies
ha in es ed ea ly in big da a analy ics capabili ies we e able o be e unde s and cus ome ends and op imize
ope a ions ahead o hei compe i o s, some imes domina ing hei sec o s as a esul . As one indus y commen a y
no ed, ea ly adop e s o AI “shape hei indus ies… se ends, imp o e se ices, and o ce compe i o s o ca ch up”,
illus a ing how being a he o e on o ech adop ion can ede ine compe i i e dynamics. A equen ly ci ed case is
Ne lix, which buil i s ecommenda ion sys em (an AI-d i en engine) ea ly on; his no only imp o ed cus ome
e en ion h ough pe sonalized con en , bu also se a new s anda d in he en e ainmen indus y ha o he s had o
ollow. Likewise, Tesla’s agg essi e adop ion o AI o sel -d i ing ea u es and da a collec ion has gi en i a lead in
au onomous d i ing da a ha adi ional au omake s a e acing o close These examples highligh how echnology
adop ion iming is a s a egic conside a ion: adop ing oo la e can mean playing ca ch-up in capabili ies and acing
highe swi ching cos s, whe eas adop ing a he igh ime (wi h he igh implemen a ion) can yield a de ensible
ad an age, a leas un il he es o he indus y ca ches on.
I is impo an o no e, howe e , ha no all echnology adop ion au oma ically con e s long- e m ad an age. Some
schola s (e.g., in he esou ce-based iew o he i m) a gue ha o an IS o echnology capabili y o p o ide sus ained
compe i i e ad an age, i should be aluable, a e, inimi able, and suppo ed by he o ganiza ion ( he VRIO c i e ia). In
p ac ice, his means ha simply buying he la es so wa e o ha dwa e ha compe i o s can also pu chase may o e
only a ansien boos . The di e en ia ing ac o o en lies in how echnology is implemen ed and in eg a ed wi h unique
business p ocesses, human expe ise, and da a asse s. Fo ins ance, a company’s p op ie a y da ase combined wi h a
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 27(02), 1752-1769
1757
cus om ML algo i hm can be a unique asse ha i als canno easily eplica e. Addi ionally, o ganiza ional change
managemen and cul u e play a ole: i ms ha success ully adap hei wo k lows and ain hei people o le e age a
new sys em ully will de i e mo e s a egic alue om i han hose ha do no . This aligns wi h he iew ha
compe i i e ad an age a ises no jus om echnology i sel bu om embedding ha echnology in complemen a y
o ganiza ional esou ces and s a egies.
2.4. Resea ch Gaps: DSS o S a egic Managemen Decisions
Despi e he ad ancemen s in decision suppo echnologies, he e emain no able gaps in he esea ch and applica ion
o DSS, pa icula ly ega ding suppo o s a egic-le el decision-making. T adi ionally, DSS esea ch and
implemen a ions ha e been skewed owa d assis ing ope a ional and ac ical decisions hose ha a e mo e s uc u ed,
equen , and da a-in ensi e (e.g., scheduling, in en o y managemen , budge alloca ions). These a e a eas whe e ample
his o ical da a and clea c i e ia allow analy ic models o h i e. In con as , s a egic decisions (such as se ing long-
e m objec i es, en e ing new ma ke s, o ans o ming business models) a e o en semi- o uns uc u ed, in ol e
signi ican unce ain y, and ely on highe -le el judgmen and ex e nal in o ma ion. They a e ypically made by senio
execu i es and ha e b oad, long- e m impac s on he o ganiza ion.
Mos exis ing DSS ools and case s udies do no ex ensi ely add ess his s a egic decision space, c ea ing a gap in bo h
esea ch and p ac ice. A e iew o he li e a u e e eals ha ew decision suppo sys ems ha e been explici ly designed
o suppo high-le el s a egic managemen decisions – especially hose in as -changing, “high- eloci y” en i onmen s.
Fo example, Ladd e al. (2013) poin ou ha e y ew comme cial DSS o e ings a he ime could adequa ely suppo
high- eloci y s a egic decision equi emen s ou -o - he-box. This means ha e en as businesses ace en i onmen s
whe e s a egic agili y is c ucial, hei in o ma ion sys ems a e o en no up o he ask o p o iding he needed decision
suppo . The consequences o his gap can include execu i es elying on in ui ion o incomple e da a o s a egic
choices, o con e sely, being o e whelmed by in o ma ion wi hou a amewo k o analyze i o long- e m planning.
3. Theo e ical F amewo k
3.1. Linking DSS o IS Theo ies
The s udy o Decision Suppo Sys ems (DSS) has long in e sec ed wi h In o ma ion Sys ems (IS) heo y, which p o ides
concep ual ounda ions o unde s anding how echnology adop ion gene a es alue in o ganiza ions. Two heo e ical
lenses a e pa icula ly use ul in aming AI-d i en DSS: he Technology-O ganiza ion-En i onmen (TOE) amewo k
and he Resou ce-Based View (RBV).
The TOE amewo k explains echnology adop ion as he p oduc o h ee in e ac ing con ex s: echnological eadiness,
o ganiza ional s uc u e and cul u e, and en i onmen al p essu es. F om a DSS pe spec i e, TOE sugges s ha he
success ul deploymen o AI-d i en decision suppo depends no only on he echnical easibili y o in eg a ing machine
lea ning models bu also on whe he he o ganiza ion has he manage ial commi men , skills, and cul u al openness o
adop da a-d i en decision-making. En i onmen al ac o s such as compe i ion, egula o y demands, and indus y
ola ili y u he in luence whe he i ms pe cei e AI-DSS as a necessi y o su i al o di e en ia ion. Thus, TOE
si ua es AI-DSS adop ion wi hin a b oade sys em o d i e s and cons ain s, highligh ing ha he e ec i eness o
decision suppo is no pu ely echnical bu socio- echnical.
The Resou ce-Based View (RBV) p o ides a complemen a y pe spec i e by ocusing on he in e nal capabili ies o he
i m. Acco ding o RBV, o ganiza ions achie e sus ained compe i i e ad an age h ough esou ces ha a e aluable,
a e, inimi able, and o ganiza ionally embedded. Da a asse s and machine lea ning capabili ies inc easingly i his
desc ip ion. While many i ms may access o - he-shel DSS ools, he unique in eg a ion o p op ie a y da a, domain
expe ise, and cus omized AI models can yield decision suppo capabili ies ha compe i o s canno easily eplica e. In
his sense, AI-d i en DSS can e ol e in o a s a egic esou ce ha unde pins compe i i e ad an age. RBV also
unde sco es he ole o o ganiza ional p ocesses and human capi al in ully le e aging such sys ems: a i m’s abili y o
ain i s manage s o in e p e and us AI insigh s, and o imp o e i s DSS con inuously, is c i ical o sus aining i s
ad an age.
Toge he , TOE and RBV o e a dual lens: TOE explains he condi ions ha enable adop ion, while RBV explains how
adop ion can ansla e in o endu ing s a egic alue. This heo e ical syn hesis is especially use ul o analyzing AI-
d i en DSS because i si ua es hem bo h as echnological a i ac s embedded in o ganiza ional en i onmen s and as
po en ial s a egic esou ces ha can shape long- e m compe i i eness.
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 27(02), 1752-1769
1758
3.2. Ex ending Simon’s Decision-Making Phases wi h Machine Lea ning
He be Simon’s classical model o decision-making, comp ising he phases o in elligence, design, choice, and
implemen a ion, emains ounda ional in IS and DSS esea ch. Machine lea ning echnologies signi ican ly ex end and
en ich each o hese phases.
In elligence Phase: T adi ionally, his phase in ol ed ga he ing and scanning in o ma ion o iden i y p oblems o
oppo uni ies. ML algo i hms now au oma e and ampli y his phase by si ing h ough as amoun s o s uc u ed and
uns uc u ed da a, de ec ing anomalies, ends, and signals ha human manage s migh o e look. Fo example,
clus e ing algo i hms can segmen cus ome bases, while anomaly de ec ion can lag eme ging isks in supply chains.
Design Phase: In Simon’s model, he design phase en ails de eloping possible cou ses o ac ion. ML enhances his by
gene a ing da a-d i en models o al e na i e scena ios. P edic i e models can simula e demand unde di e en p icing
s a egies, and ein o cemen lea ning can sugges adap i e pa hways by expe imen ing wi h possible decision
sequences. In his way, ML augmen s he c ea i e p ocess o design wi h empi ical and simula ed e idence.
Choice Phase: The choice phase in ol es selec ing among al e na i es. He e, machine lea ning con ibu es by p o iding
p obabilis ic p edic ions and op imiza ion ou pu s ha cla i y ade-o s. Decision ees, ensemble models, o neu al
ne wo ks can o e anked ecommenda ions wi h con idence in e als, allowing manage s o weigh decisions agains
quan i ied isks and bene i s. Impo an ly, ML sys ems can p esen no jus a single “bes ” op ion bu a spec um o
choices op imized unde di e en cons ain s.
Implemen a ion Phase: Finally, he implemen a ion phase ocuses on execu ing decisions and moni o ing ou comes. ML
models ex end his phase h ough con inuous lea ning and eedback loops. As ou comes un old, ein o cemen lea ning
agen s o online lea ning algo i hms can upda e hei policies, adjus ing s a egies in nea eal- ime. This c ea es a
dynamic implemen a ion p ocess whe e DSS no only guide decisions bu also e ine hemsel es as decisions a e
enac ed and ou comes obse ed.
Th ough his in eg a ion, ML echnologies ans o m Simon’s decision model om a ela i ely linea , human-cen ic
p ocess in o a cyclical, adap i e sys em whe e da a and algo i hms con inuously eed back in o each s age. The esul is
a DSS pa adigm ha is no jus suppo i e bu collabo a i e, wo king alongside human decision-make s o na iga e
unce ain y and complexi y in s a egic con ex s.
4. Me hodology
4.1. Resea ch Design
This s udy adop s a mixed-me hod esea ch design, combining a case s udy app oach wi h p edic i e modeling o
in es iga e he in eg a ion o machine lea ning in o Decision Suppo Sys ems (DSS) o s a egic business decision-
making. The case s udy p o ides ich, con ex ualized insigh s in o how AI-d i en DSS can be embedded wi hin an ac ual
o ganiza ional se ing. A he same ime, p edic i e modeling demons a es he echnical applica ion o machine
lea ning echniques o eal-wo ld business da a. This dual app oach ensu es bo h heo e ical g ounding and p ac ical
alida ion, add essing no only he manage ial aspec s o decision suppo bu also he compu a ional easibili y o
deploying ML-based DSS.
The me hodological choice is guided by he ecogni ion ha s a egic decision-making canno be ully cap u ed by
quan i a i e models alone. A case s udy allows o explo a ion o o ganiza ional, cul u al, and manage ial dynamics ha
shape he adop ion and use o DSS. P edic i e modeling, on he o he hand, illus a es he angible capabili ies o
machine lea ning in enhancing o esigh and p esc ip i e guidance. Toge he , hey c ea e a comp ehensi e
me hodology ha e lec s he socio- echnical na u e o AI-d i en DSS.
4.2. Indus y and Fi m Selec ion
Fo his esea ch, he e ail indus y has been selec ed as he empi ical con ex . Re ail o e s a e ile se ing o s udying
AI-d i en DSS o se e al easons. Fi s , he indus y is da a- ich, gene a ing as amoun s o s uc u ed da a (e.g., sales
ansac ions, in en o y logs, cus ome loyal y p og ams) and uns uc u ed da a (e.g., p oduc e iews, social media
engagemen ). Second, e ail i ms ace highly s a egic decisions such as ma ke expansion, p icing s a egy, and
cus ome segmen a ion ha can di ec ly in luence compe i i e posi ioning. Thi d, he e ail sec o has been a he
o e on o adop ing analy ics and AI, making i a ele an en i onmen in which o s udy ad anced DSS in eg a ion.
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 27(02), 1752-1769
1759
Wi hin his indus y, he case s udy ocuses on a mid-sized omnichannel e ail i m ha ope a es bo h physical s o es
and an e-comme ce pla o m. This i m was chosen because i aces s a egic ques ions abou esou ce alloca ion
be ween online and o line channels, cus ome e en ion in he ace o inc easing compe i ion, and he op imiza ion o
p icing and p omo ion s a egies. Such s a egic challenges p o ide an oppo uni y o explo e how ML-d i en DSS can
in o m long- e m decisions a he han me ely ope a ional e iciencies.
4.3. Da a Sou ces
The s udy employs bo h s uc u ed and uns uc u ed da a, e lec ing he mul i ace ed in o ma ion landscape o
s a egic business decision-making.
4.3.1. S uc u ed Da a
Sales T ansac ions: Daily eco ds o sales ac oss di e en p oduc ca ego ies and s o e loca ions, including a iables
such as quan i y sold, e enue, discoun s, and channel (online s. o line).
• Financial Da a: Pe iodic p o i -and-loss s a emen s, ma ke ing expendi u e eco ds, and in en o y cos s.
• Cus ome Demog aphics: In o ma ion om loyal y p og ams, including age, gende , loca ion, and pu chase
equency.
• Uns uc u ed Da a: Cus ome Re iews: P oduc e iews collec ed om he i m’s e-comme ce pla o m and
majo online ma ke places.
• Social Media Da a: Twee s, pos s, and commen s men ioning he b and o i s compe i o s, p o iding insigh s
in o sen imen and b and pe cep ion.
• Ma ke In elligence Repo s: Tex ual epo s om indus y analys s and ma ke esea ch agencies ha
highligh ex e nal en i onmen al ac o s.
The combina ion o s uc u ed and uns uc u ed da a ensu es ha he DSS can in eg a e bo h quan i a i e pe o mance
indica o s and quali a i e ma ke insigh s, enhancing i s abili y o suppo s a egic decisions.
4.4. P edic i e Modeling App oach
The p edic i e modeling componen applies supe ised machine lea ning algo i hms o s uc u ed business da a,
complemen ed by ex analy ics echniques o uns uc u ed da a. The selec ion o models is guided by hei
in e p e abili y, p edic i e powe , and ele ance o business applica ions.
4.4.1. Supe ised Machine Lea ning Models:
• Reg ession Models: Linea and logis ic eg ession a e used o baseline p edic ions, such as o ecas ing sales
demand o p edic ing cus ome chu n p obabili ies.
• Decision T ees: Use ul o segmen ing cus ome s and iden i ying key decision ules d i ing ou comes.
• Random Fo es s and XGBoos : Ensemble me hods ha imp o e p edic i e accu acy by agg ega ing mul iple
decision ees. These models a e pa icula ly e ec i e in handling nonlinea ela ionships and la ge ea u e
se s common in e ail da a.
• Time-Se ies Fo ecas ing Models: Fo p edic ing u u e sales ends, models such as ARIMA and P ophe a e
employed, augmen ed wi h ML-based ea u e enginee ing o accoun o p omo ions, seasonali y, and ex e nal
shocks.
4.5. Uns uc u ed Da a Analysis
Na u al Language P ocessing (NLP): Sen imen analysis is pe o med on cus ome e iews and social media pos s o
ex ac cus ome sen imen sco es. Topic modeling (e.g., La en Di ichle Alloca ion) is used o iden i y eme ging hemes
in cus ome conce ns and ma ke discussions.
Tex Classi ica ion Models: Supe ised classi ie s ca ego ize e iews o social pos s in o hemes (e.g., p oduc quali y,
cus ome se ice), which a e hen linked o s uc u ed pe o mance me ics.
4.6. Rein o cemen Lea ning o S a egic Scena ios
Fo s a egic decisions such as dynamic p icing o p omo ional alloca ion, ein o cemen lea ning is employed. The RL
agen simula es di e en p icing s a egies o e ime, ecei ing eedback in e ms o e enue and cus ome e en ion.
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 27(02), 1752-1769
1760
O e mul iple i e a ions, he model lea ns he op imal balance be ween sho - e m p o i and long- e m cus ome
loyal y.
The in eg a ion o hese models p o ides a holis ic DSS amewo k: p edic i e analy ics o o esigh , NLP o cap u ing
ex e nal pe cep ions, and ein o cemen lea ning o explo ing adap i e s a egies.
4.7. E alua ion Me ics
To assess he e ec i eness o he p edic i e models and he o e all DSS amewo k, mul iple e alua ion me ics a e
employed:
4.7.1. Accu acy and P ecision
Fo classi ica ion asks (e.g., chu n p edic ion, sen imen classi ica ion), accu acy, p ecision, ecall, and F1-sco e a e
measu ed. These me ics ensu e ha he sys em no only makes co ec p edic ions bu also minimizes alse posi i es
and alse nega i es, which is c i ical in manage ial con ex s.
4.7.2. Fo ecas ing Accu acy
Fo ime-se ies models, me ics such as Mean Absolu e E o (MAE), Roo Mean Squa ed E o (RMSE), and Mean
Absolu e Pe cen age E o (MAPE) a e used o e alua e o ecas ing pe o mance. Lowe e o a es indica e s onge
p edic i e eliabili y.
4.7.3. Re u n on In es men (ROI) Impac
Beyond echnical me ics, he DSS is e alua ed in e ms o i s business impac . Simula ed scena ios assess how he
ecommenda ions o he DSS in luence key pe o mance indica o s such as e enue g ow h, cus ome e en ion, and
ma ke sha e. ROI is calcula ed by compa ing he gains om DSS-in o med decisions agains he cos s o sys em
implemen a ion and ope a ion.
4.7.4. Manage ial Usabili y and T us
Because s a egic decisions equi e human judgmen , he DSS is also e alua ed quali a i ely h ough manage eedback.
This includes assessing he in e p e abili y o he model ou pu s, he cla i y o isualiza ions, and he ex en o which
manage s us and adop he DSS ecommenda ions.
5. Case S udy
5.1. Business En i onmen
The case s udy cen e s on a mid-sized omnichannel e ail i m ope a ing in a highly compe i i e consume goods sec o .
The i m manages app oxima ely 80 physical ou le s ac oss majo me opoli an a eas while also main aining a apidly
g owing e-comme ce pla o m. I s p oduc po olio spans appa el, household essen ials, and consume elec onics.
Annual e enues exceed $500 million, bu p o i abili y ma gins a e unde p essu e due o in ense p ice compe i ion,
ising cus ome acquisi ion cos s, and supply chain dis up ions.
The i m’s s a egic challenge lies in balancing in es men s be ween i s b ick-and-mo a ope a ions and i s online
pla o m. While online sales ha e g own s eadily d i en by digi al ma ke ing and changing consume habi s physical
s o es emain he i m’s co e e enue base. Execu i es mus decide how o alloca e esou ces be ween hese channels,
speci ically:
• Whe he o accele a e ma ke expansion by opening new s o es in seconda y ci ies, o
• Whe he o channel in es men s in o digi al s a egies, such as pe sonalized p omo ions and dynamic online
p icing.
The decision is inhe en ly s a egic: i a ec s long- e m posi ioning, capi al expendi u e, and he i m’s abili y o de end
i s ma ke sha e agains bo h adi ional compe i o s and digi al- i s en an s.
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 27(02), 1752-1769
1767
[2] Ki sios, F., & Kama io ou, M. (2018). Decision suppo sys ems and s a egic planning: In o ma ion echnology
and SMEs’ pe o mance. In e na ional Jou nal o Decision Suppo Sys ems, 3(1/2), 53–70.
h ps://www.inde scienceonline.com/doi/10.1504/IJDSS.2018.094260 Resea chGa e
[3] Ba z, A. (2025). In eg a ing machine lea ning in o business and managemen esea ch: In ellec ual landscape and
s a egic conside a ions. Humani ies and Social Sciences Communica ions. h ps://doi.o g/10.1038/s41599-
025-04361-6 Na u e
[4] Habel, J., Ala i, S., & Heini z, N. (2023). A heo y o p edic i e sales analy ics adop ion. AMS Re iew, 13, 34–54.
h ps://doi.o g/10.1007/s13162-022-00252-0 Sp inge Link
[5] Sa ke , I. H. (2021). Machine lea ning: algo i hms, eal-wo ld applica ions and esea ch di ec ions. SN Compu e
Science, 2, A icle 160. h ps://doi.o g/10.1007/s42979-021-00745-1 PMC
[6] Hein ich, K., Janiesch, C., & K anche , O. (2025). Decision ac o s o he selec ion o AI-based DSS… PLOS One
(duplica e en y omi ed)
[7] Sp ague, R. H., & Ca lson, E. D. (1982). Building E ec i e Decision Suppo Sys ems. P en ice-Hall.
[8] Keen, P. G. W., & Sco Mo on, M. S. (1978). Decision Suppo Sys ems: An O ganiza ional Pe spec i e.
Addison-Wesley.
[9] A no , D., & Pe an, G. (2005). A c i ical analysis o decision suppo sys ems esea ch. Jou nal o In o ma ion
Technology, 20(2), 67–87.
[10] Powe , D. J. (2002). Decision suppo sys ems: Concep s and esou ces o manage s. G eenwood.
[11] Negash, S. (2004). Business in elligence. Communica ions o he Associa ion o In o ma ion Sys ems, 13, 177–195.
[12] Wixom, B. H., & Wa son, H. J. (2010). The BI-based o ganiza ion. In e na ional Jou nal o Business In elligence
Resea ch, 1(1), 13–28.
[13] P o os , F., & Fawce , T. (2013). Da a Science o Business. O’Reilly.
[14] Walle , M. A., & Fawce , S. E. (2013). Da a science, p edic i e analy ics, and big da a: A e olu ion ha will
ans o m supply chain design and managemen . Jou nal o Business Logis ics, 34(2), 77–84.
[15] Bihani, P., & Pa il, A. (2019). Managing chu n using p edic i e analy ics. Jou nal o Da abase Ma ke ing & Cus ome
S a egy Managemen , 26(3), 161–172.
[16] Feldman, R., & Sange , J. (2007). The Tex Mining Handbook: Ad anced App oaches in Analyzing Uns uc u ed Da a.
Camb idge Uni e si y P ess.
[17] Da enpo , T., & Ronanki, R. (2018). A i icial in elligence o he eal wo ld. Ha a d Business Re iew, 96(1),
108–116.
[18] Su on, R. S., & Ba o, A. G. (2018). Rein o cemen Lea ning: An In oduc ion (2nd ed.). MIT P ess.
[19] Sil e , D., e al. (2016). Mas e ing he game o Go wi h deep neu al ne wo ks and ee sea ch. Na u e, 529(7587),
484–489.
[20] Hu, J., & Jennings, N. R. (2020). Mul i-agen ein o cemen lea ning o dynamic p icing. In e na ional Jou nal o
Fo ecas ing, 36(4), 1337–1348.
[21] Gabel, T., & Nema bakhsh, R. (2021). Rein o cemen lea ning o p icing and in en o y con ol: A e aile ’s
pe spec i e. Eu opean Jou nal o Ope a ional Resea ch, 293(3), 1077–1092.
[22] To na zky, L. G., & Fleische , M. (1990). The P ocesses o Technological Inno a ion. Lexing on Books.
[23] Wade, M., & Hulland, J. (2004). Re iew: The Resou ce-Based View and in o ma ion sys ems esea ch: Re iew,
ex ension, and sugges ions o u u e esea ch. MIS Qua e ly, 28(1), 107–142.
[24] Teece, D. J., Pisano, G., & Shuen, A. (1997). Dynamic capabili ies and s a egic managemen . S a egic Managemen
Jou nal, 18(7), 509–533.
[25] Venka aman, N., & Hende son, J. C. (1993). Real s a egies o i ual o ganizing. Sloan Managemen Re iew,
34(2), 33–48.
[26] Po e , M. E., & Milla , V. E. (1985). How in o ma ion gi es you compe i i e ad an age. Ha a d Business Re iew,
63(4), 149–160.
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 27(02), 1752-1769
1768
[27] Bha adwaj, A. (2000). A esou ce-based pe spec i e on in o ma ion echnology capabili y and i m pe o mance:
An empi ical in es iga ion. MIS Qua e ly, 24(1), 169–196.
[28] Sambamu hy, V., Bha adwaj, A., & G o e , V. (2003). Shaping agili y h ough digi al op ions: Reconcep ualizing
he ole o in o ma ion echnology in con empo a y i ms. MIS Qua e ly, 27(2), 237–263.
[29] Mel ille, N., K aeme , K., & Gu baxani, V. (2004). In o ma ion echnology and o ganiza ional pe o mance: An
in eg a i e model o IT business alue. MIS Qua e ly, 28(2), 283–322.
[30] Huang, M.-H., & Rus , R. T. (2021). A i icial in elligence in se ice. Jou nal o Se ice Resea ch, 24(1), 3–21.
[31] Ghah emanlou, A., & Zhang, J. (2020). AI-enabled Decision Suppo Sys ems: A case s udy in logis ics. Decision
Suppo Sys ems, 133, 113283.
[32] Ladd, T., Ma ch, S. T., & Ca , J. (2013). Decision suppo in high- eloci y en i onmen s: A amewo k o
ope a ional DSS. Communica ions o he Associa ion o In o ma ion Sys ems, 33, A icle 17.
[33] Lyy inen, K., & Newman, M. (2008). Explaining in o ma ion sys ems change: A punc ua ed socio- echnical change
model. Eu opean Jou nal o In o ma ion Sys ems, 17(6), 589–613.
[34] Ta i, A. P., & Dewan, R. (2020). Business analy ics and co po a e pe o mance: An in eg a ed amewo k. Jou nal
o Business Analy ics, 3(4), 225–241.
[35] Be man, S. J. (2012). Digi al ans o ma ion: Oppo uni ies o c ea e new business models. S a egy & Leade ship,
40(2), 16–24.
[36] Da enpo , T. H., & Ki by, J. (2016). Only Humans Need Apply: Winne s and Lose s in he Age o Sma Machines.
Ha a d Business Re iew P ess.
[37] B ynjol sson, E., & McA ee, A. (2014). The Second Machine Age: Wo k, P og ess, and P ospe i y in a Time o
B illian Technologies. W.W. No on.
[38] McKinsey Global Ins i u e. (2020). The Age o Analy ics: Compe ing in a Da a-D i en Wo ld. McKinsey &
Company.
[39] Manyika, J., e al. (2017). Ha nessing Au oma ion o a Fu u e Tha Wo ks. McKinsey Global Ins i u e.
[40] Doshi-Velez, F., & Kim, B. (2017). Towa ds a igo ous science o in e p e able machine lea ning. a Xi p ep in
a Xi :1702.08608.
[41] Ribei o, M. T., Singh, S., & Gues in, C. (2016). "Why should I us you?": Explaining he p edic ions o any
classi ie . P oceedings o he 22nd ACM SIGKDD In e na ional Con e ence on Knowledge Disco e y and Da a
Mining, 1135–1144.
[42] Mi els ad , B. D., Allo, P., Taddeo, M., Wach e , S., & Flo idi, L. (2016). The e hics o algo i hms: Mapping he
deba e. Big Da a & Socie y, 3(2), 2053951716679679.
[43] Ame shi, S., e al. (2019). Guidelines o human-AI in e ac ion. P oceedings o he 2019 CHI Con e ence on Human
Fac o s in Compu ing Sys ems, Pape 144.
[44] B own, T. B., e al. (2020). Language models a e ew-sho lea ne s. Ad ances in Neu al In o ma ion P ocessing
Sys ems, 33, 1877–1901.
[45] Bommasani, R., e al. (2021). On he oppo uni ies and isks o ounda ion models. a Xi p ep in
a Xi :2108.07258.
[46] Chen, M., e al. (2021). E alua ing la ge language models ained on code. a Xi p ep in a Xi :2107.03374.
[47] Teal, G., & Ga ne , W. B. (2022). Real- ime analy ics o dynamic decision-making: A concep ual amewo k.
Jou nal o Business Analy ics, 5(1), 48–63.
[48] Lus ig, F., & Nabe h, T. (2023). Gene a i e AI o decision suppo : A axonomy and esea ch oadmap. Decision
Suppo Sys ems, 158, 113838.
[49] Wa son, H. J. (2018). The ole o eal- ime analy ics in s a egic decision making. Jou nal o Decision Sys ems,
27(sup1), 5–10.
[50] Żbikowski, K., Os apowicz, M., & Gaw ysiak, P. (2021). Deep ein o cemen lea ning o esou ce alloca ion in
business p ocesses. a Xi p ep in a Xi :2104.00541. Resea chGa eU.S. Geological Su eya Xi
Wo ld Jou nal o Ad anced Resea ch and Re iews, 2025, 27(02), 1752-1769
1769
[51] Théa e, T., & E ns , D. (2020). An applica ion o deep ein o cemen lea ning o algo i hmic ading. a Xi p ep in
a Xi :2004.06627. a Xi
[52] Cha pen ie , A., Elie, R., & Remlinge , C. (2020). Rein o cemen lea ning in economics and inance. a Xi p ep in
a Xi :2003.10014. a Xi
[53] F ancois-La e , V., Hende son, P., Islam, R., Bellema e, M. G., & Pineau, J. (2018). An in oduc ion o deep
ein o cemen lea ning. a Xi p ep in a Xi :1811.12560. a Xi +1
[54] Li ewi e. (2023, June 21). Wha is ein o cemen lea ning? Li ewi e. h ps://www.li ewi e.com/wha -is-
ein o cemen -lea ning-7508013 li ewi e.com
[55] Decision suppo sys em. (2025). In Wikipedia. h ps://en.wikipedia.o g/wiki/Decision_suppo _sys em
Wikipedia
