Op imising Machine y U ilisa ion by Applying
A i icial In elligence
Miguel Ángel Ma eo-Casali(B), Juan Pablo Fiesco, Bea iz And es, and Raul Pole
Resea ch Cen e on P oduc ion Managemen and Enginee ing (CIGIP), Uni e si a Poli ècnica
de València (UPV), Camino de Ve a s/n, 46022 Valencia, Spain
{mma eo,j iesco,band es, pole }@cigip.up .es
Abs ac . The a icle discusses he impo ance o sma p oduc ion du ing he
p og ess o Indus y 4.0 and he challenges ha Big Da a analy ics and a i i-
cial in elligence (AI) ools ace. Using AI ools, such as p edic i e main enance,
p oduc ion op imisa ion and quali y con ol sys ems, can imp o e p oduc ion e i-
ciency, quali y and sa e y. This a icle also highligh s he goals o AI echnologies,
such as educing p oduc ion down imes, op imising p oduc ion, imp o ing p od-
uc quali y and sa e y, and inc easing au oma ion o achie e he ze o-de ec phi-
losophy. I concludes ha applying AI solu ions can help o educe de ec s, was e
and e o s in p oduc ion p ocesses, which will esul in inc easing he e iciency
and quali y o p oduc ion p ocesses.
Keywo ds: A i icial in elligence ·Use li e cycle ·Sma ac o ies ·Ze o De ec s
1 In oduc ion
In ecen yea s, ocusing on sma p oduc ion as a key issue o ad ancing Indus y 4.0
has g own (Lin e al., 2019). Wi h he deploymen o hund eds o e en housands o
senso s and sma de ices, sma ac o ies a e now able o enhance p oduc quali y using
a ious digi al echnologies. This has led o he apid g ow h o he In e ne o Things
(IoT) and he eme gence o IoT-based sma ac o ies, which b ing new challenges o Big
Da a analy ics and he implemen a ion o machine-lea ning (ML) echniques (Yu e al.,
2022). Sma indus ies ely on no only supe ised lea ning o op imise p oduc ion and
o ensu e maximum quali y (Sha iq e al., 2023), bu also on he ac ha e o s always
a ec p oduc ion, bu emphasise he apid de ec ion and educ ion o aul s and de ec s
in p oduc ion. In addi ion, no ou -o -speci ica ion esul s a e passed on o he nex s ep
o o cus ome s (Se ano-Ruiz e al., 2021).
To o e come his hu dle, i is essen ial o analyse he obs acles ha small- and
medium-sizeden e p ises(SMEs)encoun e whenadop ingin o ma ion echnologyin o
Indus y 4.0 (Moeu e al., 2020). This app oach can accele a e he widesp ead adop ion
o ad anced manu ac u ing p ac ices among SMEs, while la ge co po a ions may lag
in inno a ing. Addi ionally, expe s sugges ha a key ac o o success would be o
simpli y Indus y 4.0 ools by making hem mo e accessible o SMEs. This app oach
would no only educe he impac o lack o expe ise, bu would also p omo e hese
© The Au ho (s), unde exclusi e license o Sp inge Na u e Swi ze land AG 2024
J. Bau is a-Valhondo e al. (Eds.): CIO 2023, LNDECT 206, pp. 444–449, 2024.
h ps://doi.o g/10.1007/978-3-031-57996-7_76
Op imising Machine y U ilisa ion 445
ools pene a ing SMEs by inc easing hei adap abili y and compe i i eness (Sande s
e al., 2016). The ollowing a icle will ca y ou a concep ual e iew based on he
iden i ica ion o AI uses o op imise machine u ilisa ion. In o ma ion is p esen ed on
how AI is ans o ming he way ac o ies manu ac u e p oduc s and manage hei p o-
duc ion p ocesses. I desc ibes di e en AI applica ions in indus y, such as p edic i e
main enance, p oduc ion op imisa ion, quali y con ol, imp o ed sa e y and inc eased
au oma ion. This a icle aims o concep ualise he AI ools de eloped in he AIDEAS
p ojec o co e he use phase o indus ial equipmen . The objec i e is o enhance he
e iciency and quali y p oduc ion p ocess du ing machine y u ilisa ion.
2 Me hods and Objec i es
The implemen a ion o AI is e olu ionising manu ac u ing and p ocess managemen
in ac o ies. In he machine u ilisa ion s age, a ious AI ools a e employed o imp o e
p oduc ione iciency,quali yandsa e y.Thesesolu ionsincludep edic i emain enance,
p oduc ion op imisa ion and quali y moni o ing. In he ealm o indus ial ope a ions,
se e al AI-powe ed ools ha e eme ged o enhance e iciency, quali y and sa e y. Le ’s
explo e h ee key examples:
•P edic i e main enance sys ems use AI algo i hms o analyse he da a ob ained om
he senso s ound on machines h oughou he p oduc ion p ocess. Thanks o hese
da a, ML models can be implemen ed o e icien ly p edic when a machine equi es
main enance.
•P oduc ion op imisa ion sys ems u ilise AI ools o analyse p oduc ion da a, p edic
op imal p oduc ion alues and adjus machine pa ame e s o maximise e iciency by
educing chances o e o .
•Quali y moni o ing sys ems employ came as o senso s o cap u e li e images o
eal- ime p oduc ion and analyse hem wi h AI algo i hms o de ec quali y de ec s
and ale ope a o s o co ec he p oblem.
AI solu ions, such as quali y moni o ing and machine ision sys ems, enable ope a-
o s o de ec any de ec s in p oduc ion in eal ime by allowing hem o quickly add ess
he p oblem and educe he numbe o de ec i e p oduc s. In addi ion, using ML algo-
i hms and p edic i e main enance sys ems can help o p e en ailu es and de ec s in
p oduc ion be o e hey occu . The Ze o De ec s philosophy (Cal in, 1983), and he
AI solu ions used in he use phase, aim o educe and elimina e de ec s in p oduc ion,
which can imp o e p oduc quali y and inc ease p oduc ion p o i abili y. By in eg a -
ing hese AI solu ions in o a ac o y’s daily ope a ions, companies can mo e owa ds
he Ze o De ec s goal and imp o e hei compe i i e ma ke posi ion (Naza enko e al.,
2021).The goals o AI echnologies can help o ul il hese goals in se e al ways, such
as:
•P o ide AI echnologies ha enable ini ial machine calib a ion by using algo i hms
and ML models o adjus and con igu e machine pa ame e s op imally om he s a
by ensu ing hei p ope and e icien ope a ion.
446 M. A. Ma eo-Casali e al.
•O e AI echnologies ha ensu e he quali y o indus ial p ocesses by cons an ly
assessing he s a us o machines and de ec ing anomalies by analysing he da a col-
lec ed by senso s. These echnologies make i possible o iden i y and co ec p ocess
de ia ions by op imising p oduc ion quali y and e iciency.
•De elop AI echnologies ha ensu e he quali y o manu ac u ed p oduc s by imple-
men ing AI-based quali y con ol sys ems using came as, senso s and algo i hms
o de ec and p e en p oduc de ec s o, hus, ensu e cus ome sa is ac ion and he
deli e y o high-quali y p oduc s.
•Es ablish app op ia e mechanisms o exchange machine u ilisa ion da a wi h o he
li e cycle s ages o indus ial equipmen . This in ol es in eg a ing sys ems and se ing
up s anda ds ha enable e icien secu e da a ans e by acili a ing collabo a ion and
da a analysis along he en i e alue chain.
•Con inuously manage he in eg a ion and alida ion o AI applica ions o use by
indus ial equipmen . Ha e a sys ema ic app oach o ensu e ha he AI applica ions
employedonequipmen a esui able, eliableandsecu e. Thisincludes egula es ing,
e alua ion and upg ades o op imise hei pe o mance and adap hem o changing
indus y equi emen s.
In eg a ing a i icial in elligence echnologies in o indus ial en i onmen s o e s a
numbe o bene i s andadd esses se e al challenges. Fi s ly, p edic i e main enance uses
algo i hms o analyse senso da aand p edic po en ial equipmen ailu es.This p oac i e
s a egy enables imely epai s by educing down imes and imp o ing ope a ional e i-
ciency. In addi ion, AI-based p oduc ion op imisa ion solu ions enable p oduc ion da a
o be analysed and o ind ways ha imp o e e iciency and educe was e. By making
da a-d i en decisions, companies can op imise p oduc ion p ocesses and make he bes
use o esou ces. Likewise, implemen ing AI-based quali y sys ems makes i possible
o de ec and educe de ec s, which imp o es p oduc quali y and educes was e. AI
algo i hms analyse da a in eal ime o ensu e ha p oduc s mee o exceed quali y s an-
da ds. By iden i ying haza ds in eal ime, AI echnology helps o p e en acciden s and
educe wo kplace inju ies. In eg a ing au oma ed sys ems like obo s, which can pe -
o m epe i i e and p ecise asks, inc eases au oma ion and educes he need o human
in e en ion in po en ially haza dous asks.
Implemen ing AI solu ions o ul il hese goals can help companies o achie e he
Ze o De ec s philosophy by educing de ec s in p oduc ion p ocesses. P edic i e main e-
nance using AI can help o p e en equipmen ailu es, which can lead o p oduc de ec s.
AI-enabled p oduc ion op imisa ion can iden i y and elimina e was e, which can lead o
p oduc de ec s. Implemen ing quali y sys ems based on AI use can help o de ec and
educe p oduc de ec s. Sa e y AI applica ions can help o de ec haza dous condi ions
o ope a o s ha can lead o inju ies and p oduc de ec s. Finally, in oducing obo s
o au oma ed sys ems using AI can help o pe o m asks mo e accu a ely and wi hou
e o s, which educes he isk o de ec s. The e o e, in oducing AI ools con ibu es o
imp o e he challenges p esen ed in he use phase o a machine in a ac o y in e iciency,
quali y and sa e y p oduc ion e ms.
Op imising Machine y U ilisa ion 447
3 Resul s
The Eu opean p ojec AI D i en Indus ial Equipmen P oduc Li e Cycle Boos ing
Agili y, Sus ainabili y and Resilience (AIDEAS, 2022) ocuses on de eloping AI ech-
nologies o suppo he en i e li e cycle o indus ial equipmen (design, manu ac u ing,
use, epai / euse/ ecycle) as a s a egic ins umen o imp o e he sus ainabili y, agili y
and esilience o Eu opean machine y manu ac u ing companies. This pape ocuses on
he use phase. A se o i e ools is in oduced o op imally suppo machine y u ilisa ion:
•The Machine Calib a o (MC) Toolki is an AI-d i en solu ion designed o help
manu ac u e s o quickly and e icien ly calib a e indus ial equipmen . The MC is
use ul when ins alling new equipmen in a ac o y o when e-calib a ion is needed.
By le e aging AI echniques, he MC can p o ide he mos well-sui ed calib a ion
pa ame e s o, hus educe he ime and cos associa ed wi h he calib a ion p ocess.
The MC can also imp o e he accu acy and p ecision o equipmen by ensu ing ha
i mee s he equi ed speci ica ions.
•The Condi ion E alua o (CE) Toolki is an ad anced solu ion ha can help man-
u ac u e s o de e mine he condi ion o a machine o i s componen s unde wo king
condi ions in a ac o y. The CE uses ad anced algo i hms o analyse he da a collec ed
om senso s and o he sou ces o de e mine he machine’s o e all condi ion. This
allows manu ac u e s o iden i y po en ial issues be o e hey become c i ical, which
educes down imes and main enance cos s. The CE can also help manu ac u e s o
op imise main enance schedules by ensu ing ha machines a e in op imal condi ion,
which educes he isk o unplanned down imes.
•The Anomaly De ec o (AD) Toolki is an AI-based solu ion ha can de ec anoma-
lies in machine componen s unde wo king condi ions in a ac o y. The AD uses ML
algo i hms o analyse he da a collec ed om senso s and o he sou ces, which allows
manu ac u e s o iden i y anomalies ha may indica e po en ial p oblems. By iden-
i ying anomalies ea ly, manu ac u e s can ake co ec i e ac ion be o e he p oblem
becomes c i ical, which educes he isk o unplanned down imes and main enance
cos s.
•The Adap i e Con olle (AC) Toolki is ano he AI-d i en solu ion ha can help
manu ac u e s o ain machine con olle s o accommoda e he machine’s condi ion
and equi emen s. The AC uses measu emen da a o ain models, which can hen
be employed o ain machine con olle s. By aining machine con olle s wi h hese
models, manu ac u e s can ensu e ha machines op imally ope a e and can adap o
changing condi ions in eal ime. This can help manu ac u e s o educe he isk o
unplanned down imes and imp o e o e all e iciency.
•The Quali y Assu ance (QA) Toolki comp ises AI-enabled ea u es o moni o -
ing manu ac u ed p oduc quali y. The QA can analyse da a om senso s and o he
sou ces, such as came as, o iden i y po en ial quali y issues. Manu ac u e s can hen
make any adjus men s be o e he p oduc is deli e ed o he cus ome . In his way,
manu ac u e s can inc ease cus ome sa is ac ion and lowe he isk o e u ns and
wa an y claims by ensu ing ha he quali y o i ems mee s ele an c i e ia.
448 M. A. Ma eo-Casali e al.
In sho , he manu ac u ing indus y is being e olu ionised by AI-based solu ions,
such as he p oposed MC, CE, AD, AC and QA oolki s. These solu ions use da a om
a ious sou ces, such as senso s o came as, and analyse hem o help manu ac u e s
o iden i y po en ial p oblems be o e hey become c i ical issues. As a esul , hese
ools educe down imes, main enance cos s and he isk o unplanned down imes. The
manu ac u e s ha adop hese AI-based solu ions a e mo e likely o keep in pace wi h
a apidly e ol ing and inc easingly demanding ma ke en i onmen , which will enable
hem o gene a e o main ain a compe i i e ad an age by imp o ing hei ope a ional
e iciency, educing cos s and imp o ing he quali y o hei p oduc s.
4 Conclusion
By way o conclusion, he in eg a ion o AI and Indus y 4.0 in o he manu ac u ing
indus y o e s nume ous bene i s, including imp o ed e iciency, educed was e and
enhanced sus ainabili y h oughou he li e cycle o indus ial equipmen . Employing
AI-based oolki s like he MC, CE, AD, AC and QA oolki s can u he op imise he
daily use o indus ial equipmen by allowing o nea eal- ime esponses o down imes.
The Ze o De ec s philosophy and he AI solu ions applied in he use phase aim o educe
and elimina e de ec s in p oduc ion, which can imp o e p oduc quali y and inc ease
p oduc ion p o i abili y. Thus AI echnologies will con inue o play a i al ole in he
use o indus y by enabling manu ac u e s o achie e e iciency, cos , e ec i eness and
cus ome sa is ac ion.
Acknowledgemen s. The esea ch ha led o hese indings ecei ed unding om wo sou ces.
The i s sou ceo undingwas om heHo izonEu opeF amewo kP og amme(HORIZON)wi h
G an Ag eemen No. 101057294 “AI D i en Indus ial Equipmen P oduc Li e Cycle Boos ing
Agili y, Sus ainabili y, and Resilience (AIDEAS)”. The second sou ce o unding was om he
Regional Depa men o Inno a ion, Uni e si ies, Science, and Digi al Socie y o he Gene ali a
Valenciana “P og ama In es igo” ( e . INVEST/2022/330), which he Eu opean Union suppo ed
- Nex Gene a ionEU unde he Plan de Recupe ación, T ans o mación y Resiliencia.
Re e ences
AIDEAS. AI D i en Indus ial Equipmen P oduc Li e Cycle Boos ing Agili y, Sus ainabili y
and Resilience. Eu opean Union’s Ho izon Eu ope esea ch and inno a ion p og amme unde
g an ag eemen No. 101057294 (2022)
Cal in, T.W.: Quali y con ol echniques o ze o de ec s. Tech. Pap. -Soc. Manu ac . Eng. C(3),
323–328 (1983). h ps://doi.o g/10.1016/0026-2714(84)90075-1
Lin, J.W., Liao, S.W., Leu, F.Y.: Senso da a comp ession using bounded e o piecewise linea
app oxima ion wi h esolu ion educ ion. Ene gies 12(13), 2523 (2019). h ps://doi.o g/10.
3390/en12132523
Moeu , A., Lamou i, S., Pelle in, R., Tamayo-Gi aldo, S., Tobon-Valencia, E., Ebu dy, R.: Iden i-
ica ion o c i ical success ac o s, isks and oppo uni ies o indus y 4.0 in SMEs. In . J. P od.
Res. 58(5), 1384–1400 (2020). h ps://doi.o g/10.1080/00207543.2019.1636323
Op imising Machine y U ilisa ion 449
Naza enko, A.A., Sa aipa, J., Cama inha-Ma os, L.M., G unewald, C., Do chain, M., Ja dim-
Goncal es, R.: Analysis o ele an s anda ds o indus ial sys ems o suppo ze o de ec s
manu ac u ing p ocess. J. Ind. In . In eg . 23(Sep embe 2020), 100214 (2021). h ps://doi.o g/
10.1016/j.jii.2021.100214
Sande s, A., Elangeswa an, C., Wul sbe g, J.: Indus y 4.0 implies lean manu ac u ing: esea ch
ac i i ies in indus y 4.0 unc ion as enable s o lean manu ac u ing. J. Ind. Eng. Manag.-JIEM
9(3), 811–833 (2016). h ps://doi.o g/10.3926/jiem.1940
Se ano-Ruiz, J.C., Mula, J., Pole , R.: Sma manu ac u ing scheduling: a li e a u e e iew. In:
Jou nal o Manu ac u ing Sys ems (Vol. 61, pp. 265–287). Else ie B.V (2021). h ps://doi.
o g/10.1016/j.jmsy.2021.09.011
Sha iq, M., Thak e, K., K ishna, K.R., Robe , N.J., Ku uppa h, A., Kuma , D.: Con inuous quali y
con ol e alua ion du ing manu ac u ing using supe ised lea ning algo i hm o Indus y 4.0.
In: In e na ional Jou nal o Ad anced Manu ac u ing Technology (2023). h ps://doi.o g/10.
1007/s00170-023-10847-x
Yu, W., Liu, Y., Dillon, T., Rahayu, W., Mos a a, F.: An in eg a ed amewo k o heal h s a e
moni o ing in a sma ac o y employing IoT and big da a echniques. IEEE In e ne Things J.
9(3), 2443–2454 (2022). h ps://doi.o g/10.1109/JIOT.2021.3096637
