S oibe , Ch is oph; Schönig, S e an
A icle — Published Ve sion
Le e aging he indus ial in e ne o hings o business
p ocess imp o emen : a me amodel and pa e ns
In o ma ion Sys ems and e-Business Managemen
P o ided in Coope a ion wi h:
Sp inge Na u e
Sugges ed Ci a ion: S oibe , Ch is oph; Schönig, S e an (2024) : Le e aging he indus ial in e ne o
hings o business p ocess imp o emen : a me amodel and pa e ns, In o ma ion Sys ems and e-
Business Managemen , ISSN 1617-9854, Sp inge , Be lin, Heidelbe g, Vol. 22, Iss. 2, pp. 285-313,
h ps://doi.o g/10.1007/s10257-024-00676-0
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/315079
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h p://c ea i ecommons.o g/licenses/by/4.0/
Vol.:(0123456789)
In o ma ion Sys ems and e-Business Managemen (2024) 22:285–313
h ps://doi.o g/10.1007/s10257-024-00676-0
1 3
ORIGINAL ARTICLE
Le e aging heindus ial in e ne o hings o business
p ocess imp o emen : ame amodel andpa e ns
Ch is ophS oibe 1 · S e anSchönig1
Recei ed: 27 May 2023 / Re ised: 28 Decembe 2023 / Accep ed: 13 Feb ua y 2024 /
Published online: 26 Ap il 2024
© The Au ho (s) 2024
Abs ac
Indus ial o ganiza ions o all kinds inc easingly ecognize he indus ial in e ne o
hing’s (IIoT) capabili ies o enable aluable business p ocess imp o emen (BPI).
Howe e , bo h heo e ically and p ac ically, he e is a lack o cla i y ega ding he
sys ema ic and success ul iden i ica ion, speci ica ion, and implemen a ion o co -
esponding applica ions. This a icle aims o b idge his esea ch gap by p esen ing
a comp ehensi e me amodel encompassing all ele an aspec s and elemen s o IIoT
applica ions wi h BPI p oposi ions. The me amodel is he ounda ion o de i ing
gene ic ye p ac ical pa e ns ha can assis o ganiza ions in e ec i ely execu ing
IIoT p ojec s. To e alua e he use ulness o he app oach, i e ini ial pa e ns we e
designed and applied by a ma ke -leading o ganiza ion. The me amodel and pa -
e ns con ibu e o he desc ip i e knowledge o he IIoT and acili a e sense-mak-
ing, heo y-led design, and p ac ical p ojec execu ion. To ensu e igo , he esea ch
endea o ollowed undamen al p inciples o he design science esea ch (DSR)
me hodology.
Keywo ds Indus ial in e ne o hings· Business p ocess imp o emen · Business
p ocess managemen · Me amodel· Pa e ns
1 In oduc ion
The impac o in e ne o hings (IoT) applica ions is pe asi e, in luencing a ious
aspec s o daily li e and in oducing dis up i e echnologies o p i a e households
and businesses ac oss di e en sec o s (Whi mo e e al. 2015; Sie e s e al. 2021).
Besides a ious sma home, sma g id, and sma ci y applica ions, especially
indus ial o ganiza ions can ema kably bene i om in eg a ing IoT echnologies
* Ch is oph S oibe
ch is oph.s oibe @u .de
1 P o esso ship o IoT-Based In o ma ion Sys ems, Uni e si y o Regensbu g, Uni e si ä ss aße
31, 93053Regensbu g, Ge many
286
C.S oibe , S.Schönig
1 3
in o hei business p ocesses. In his ega d, a pa adigm deno ed as he indus ial
IoT (IIoT) has e ol ed ha le e ages he IoT, albei anscending he concep o he
hing owa d indus ial applica ions. By ans o ming analog in o ma ion in o digi-
al da a ha can be p ocessed in eal- ime, he IIoT enables new business models,
e olu ionizes exis ing ones (Ng and Wakenshaw 2017; Anosike e al. 2021), and
enhances o ganiza ions’ compe i i e ad an age (Li e al. 2012). Gene a ing and u i-
lizing comp ehensi e p ocess da a and he in e connec edness o p ocess en i ies
o e oppo uni ies o imp o ing a ious business p ocesses and op imizing alue
c ea ion (Del Giudice 2016). Hence, in eg a ing IIoT echnology in o exis ing busi-
ness p ocesses can esul in aluable business p ocess imp o emen (BPI), which is
pa icula ly ele an o p ocess-o ien ed o ganiza ions (Janiesch e al. 2020). Fo
example, equipping in-s ock p oduc s wi h simple adio- equency iden i ica ion
(RFID) ags can enhance wa ehouse p ocess aceabili y and unlock nume ous pos-
sibili ies o imp o ing downs eam ope a ions (Fescioglu-Un e e al. 2015). As a
esul , he p essu e on o ganiza ions o in eg a e IIoT echnology is g owing s eadily,
so o ganiza ions ha do no adop IIoT may no be compe i i e soon (Liu 2017).
Howe e , a su ey o o e 500 business execu i es e ealed ha 90% o indus-
ial o ganiza ions emain in he p oo o concep o ea ly-s age planning phases
o IIoT p ojec s (Bosche e al. 2016). Many o ganiza ions encoun e se e e dis-
c epancies be ween he ini ial expec a ions o IIoT p ojec s and ac ual esul s
(Ska žauskienė and Kalinauskas 2015; Zhang e al. 2021). These indings high-
ligh signi ican ba ie s and hu dles ha impede success ul in eg a ion in o exis -
ing business p ocess landscapes. Cu en li e a u e has also d awn a en ion o
his issue and decla ed he in e ac ion o IIoT wi h business p ocesses as one o
he leas unde s ood phenomena in he business p ocess managemen (BPM) dis-
cipline (Be e ungen e al. 2020). O ganiza ions need a no el and e ec i e ech-
nology managemen o exploi he IIoT o imp o e business p ocess pe o mance
(Del Giudice 2016). This mo i a es e lec ing and upda ing s a egies and bes
p ac ices o edesigning business p ocesses (Be e ungen e al. 2020).
Fo an enhanced IIoT applica ion ma u i y in indus ial o ganiza ions, exis -
ing business p ocess p oblems mus be in es iga ed, and po en ial IIoT solu ions
iden i ied (Se hi and Sa angi 2017). In his ega d, he “Ac o Imp o emen ”,
i.e., how exis ing business p ocesses a e ans e ed o he imp o ed a ge s a e
by implemen ing IIoT applica ions, mus be de ined p ecisely. This enhances he
plannabili y and, hus, he chance o a success ully pe o med IIoT p ojec (Fo -
s e 2006). Acco dingly, o ganiza ions need suppo o h ee main challenges: (i)
he iden i ica ion o BPI po en ials, (ii) he speci ica ion o IIoT applica ions, and
(iii) he ac ual implemen a ion o hese applica ions. The cen al esea ch ques-
ion (RQ) o his a icle can be o mula ed acco dingly:
RQ1 How can indus ial o ganiza ions be suppo ed in he iden i ica ion, speci ica-
ion, and implemen a ion o IIoT-based BPI applica ions?
One auspicious app oach o add ess his RQ is he de elopmen o pa e ns.
Pa e ns a e gene ic and eusable a i ac s ha o e sui able solu ions o speci ic
287
1 3
Le e aging heindus ial in e ne o hings o business…
p oblems wi hin a gi en con ex (Alexande 1977). Conce ning IIoT-based BPI
applica ions, pa e ns se e as empla es o bluep in s ha can be applied ac oss
di e en indus ial o ganiza ions (Fo s e 2006). These pa e ns encompass all
ele an elemen s o he applica ions, including unde lying challenges, indus y
examples, pe o mance indica o s, and speci ic cha ac e is ics o he echnical
solu ions. To o mula e pa e ns e ec i ely, a well-de ined me amodel ha cap-
u es undamen al design p inciples is a p e equisi e. The me amodel ensu es he
pa e n desc ip ions’ comple eness, consis ency, and s uc u ed ep esen a ion
(Falk e al. 2013). Agains his backg ound, an addi ional RQ can be o mula ed:
RQ2 Which me amodel enables he illus a ion o IIoT-based BPI pa e ns?
The a icle a hand add esses bo h RQs by p oposing a me amodel con aining
all elemen s equi ed o undamen ally comp ehend he phenomenon o IIoT-based
BPI. To demons a e he p ac ical applica ion o he me amodel, an ini ial se o i e
IIoT-based BPI pa e ns was de i ed om a ho ough analysis o 34 eal-li e IIoT
applica ions and illus a ed using he me amodel. To e alua e he esea ch app oach,
he pa e ns we e applied by a ma ke -leading indus ial o ganiza ion.
The emainde o his a icle is s uc u ed as ollows. Sec ion2 p esen s he heo-
e ical ounda ions o IIoT and BPI and an o e iew o pa e ns and me amodels in
in o ma ion sys ems esea ch. Sec ion3 desc ibes he unde lying esea ch me hod-
ology, which has been applied o de elop and e alua e he me amodel and pa e ns.
Subsequen ly, he design and de elopmen o he me amodel is illus a ed in Sec .4,
including an ex ensi e Sys ema ic Li e a u e Re iew (SLR). Sec ion5 p esen s he
inal me amodel o IIoT-based BPI, including all aspec s and elemen s. In Sec .6,
a se o i e pa e ns is de i ed by an expe panel. Subsequen ly, in Sec .7, he
pa e ns a e applied in a eal-wo ld case s udy se ing as a summa i e e alua ion.
Finally, implica ions, limi a ions, and u u e esea ch oppo uni ies a e discussed in
Sec s.8 and 9.
2 Theo e ical ounda ions
2.1 Indus ial in e ne o hings mee s business p ocess imp o emen
The e a e many app oaches o de ining IoT, i s componen s, ea u es and capabili-
ies, and he hings hemsel es. Acco ding o he IEEE, he IoT is a ne wo k ha
connec s uniquely iden i iable hings o he in e ne . By exploi ing unique iden i i-
ca ion and sensing, in o ma ion abou he hing can be collec ed, and he s a e can
be changed om anywhe e, any ime, by any hing (Mine a e al. 2015). Hence,
he e m hing co esponds o he idea o c ea ing a ubiqui ous p esence o objec s
equipped wi h senso s, ac ua o s, o ags. On he o he hand, he e m in e ne e e s
o he abili y o hese hings o build a ne wo k o in e connec ed objec s based on
se e al speci ic ne wo k echnologies. These wo pe spec i es can be complemen ed
288
C.S oibe , S.Schönig
1 3
by a seman ic iew, which ep esen s he abili y o IoT o uniquely iden i y hings
and s o e, p ocess, and exchange da a (A zo i e al. 2010).
Wi h he inc easing p ominence o indus ial IoT applica ions, a mo e specialized
pa adigm called he IIoT has eme ged. Unlike he gene ic de ini ion o IoT, he IIoT
in ol es u ilizing speci ic IoT echnologies, such as pa icula ypes o sma objec s
wi hin indus ial cybe -physical sys ems, o pu sue objec i es speci ic o he indus-
ial domain. As a esul , he IIoT di e en ia es i sel om he IoT by he pu poses
o which he echnologies a e pu (Boyes e al. 2018). Cu en esea ch and imple-
men ed applica ions show ha IIoT echnology e eals many ex ensi e possibili ies
o imp o ing business p ocesses (S oibe and Schönig 2021).
Despi e he IIoT’s po en ial o enhance BPI and op imize o ganiza ional pe o -
mance, he e is a dea h o esea ch ocused on IIoT-based BPI. This esea ch gap
can be add essed by de eloping a me amodel ha acili a es he c ea ion o pa -
e ns and con ibu es o he desc ip i e knowledge o IIoT-based BPI. This app oach
has p o en success ul in nume ous esea ch disciplines and has gained accep ance
ac oss indus ies (Win e e al. 2009).
2.2 Me amodels andpa e ns
Pa e ns, o iginally in oduced by Alexande (1977) in he con ex o a chi ec-
u e, ha e ound applicabili y in a ious domains, including in o ma ion sys ems
esea ch. They ep esen ecu ing p oblems o challenges in he eal wo ld and
p o ide gene ic solu ions ha can be applied o simila p oblems in di e en con-
ex s. Acco ding o Fowle (1996), pa e ns a e ideas ha ha e p o en use ul in one
p ac ical applica ion and a e likely o be aluable in o he s. Following he de ini ion
o Gamma e al. (1994), pa e ns consis o ou essen ial elemen s. Fi s , a pa e n
mus ha e a name o iden i ica ion. Nex , he p oblem should be desc ibed, speci y-
ing he con ex in which he pa e n can be applied. The hi d elemen desc ibes he
p oblem solu ion, which should no be a conc e e solu ion bu a he an explana ion
o he in e ac ion o di e en mechanisms leading o a solu ion. Finally, he conse-
quences o applying he pa e n, including posi i e and nega i e e ec s, should be
ou lined.
Pa e ns ha e been ex ensi ely esea ched in di e en esea ch disciplines, signi i-
can ly con ibu ing o, e.g., so wa e enginee ing (Gamma e al. 1994). Fu he mo e,
pa e ns ha e p o en aluable in designing indi idual objec -o ien ed so wa e com-
ponen s and composing hem in o applica ions (Schmid e al. 2000). They b idge
he gap be ween high-le el in eg a ion plans and implemen a ion challenges, p o-
iding guidelines ha compensa e o decision-make s’ lack o expe ience (Hohpe
and Wool 2003), leading o imp o ed p ojec quali y, educed ime consump ion,
and cos sa ings. Pa e ns a e also applicable o p ocess- ela ed disciplines such as
Wo k low Managemen o p ocess-awa e in o ma ion sys ems (Webe e al. 2008).
Howe e , o he discipline o BPI, he c ea ion o speci ic pa e ns has ba ely
been add essed. Fo s e (2006) buil a amewo k and oolse o c ea ing and
s uc u ing BPI pa e ns while c ea ing he i s se o pa e ns. Ano he ele an
289
1 3
Le e aging heindus ial in e ne o hings o business…
con ibu ion by Falk e al. (2013) p oposes a me amodel ha acili a es he illus-
a ion o BPI pa e ns. These pa e ns cons i u e models de i ed om an o igin
me amodel.
Gene ally, a model can desc ibe objec s in he eal wo ld and abs ac cons uc s.
When he abs ac cons uc desc ibed is a model i sel , i is called a me amodel
(Gonzalez-Pe ez and Hende son-Selle s 2008). To s uc u e his me amodeling
app oach, he Me a-Objec Facili y (MOF) s anda d o he Objec Managemen
G oup (OMG) p o ides a conc e e me amodeling a chi ec u e illus a ed in di e -
en laye s (Objec Managemen G oup 2013). Laye M0 e e s o ins an ia ions,
eal objec s o ins ances. Laye M1 consis s o models, and hus mappings o eal
objec s. Fu he mo e, laye M2 e e s o me amodel, o example, w i en in he Uni-
ied Modeling Language (UML). A me amodel desc ibes he ypes o model build-
ing blocks a ailable, he ypes o ela ionships be ween he model building blocks,
he ules o linking be ween model building blocks by ela ionships, and he seman-
ics o he model building blocks and ela ionships. The ou h laye M3 p o ides a
me a-me amodel and cons i u es he language used by MOF o build me amodels.
2.3 Rela ed wo k
As desc ibed in subSec . 2.2, some wo ks ha speci ically add ess he combina-
ion o IIoT echnology and business p ocesses ha e been eme ging. Hence, BPM
esea ch inc easingly ecognizes he challenges and oppo uni ies o in eg a ing IIoT
and ice e sa. A ecen s uc u ed li e a u e e iew (SLR) o Vukšić e al. (2021)
highligh ed he opic’s ele ance by analyzing he empo al dis ibu ion o published
a icles. While esea ch has been ela i ely sca ce om 2010 o 2015, a no iceable
g ow h can be seen since 2016. Consequen ly, he opic is s ill in an ea ly scien i ic
esea ch s age bu can be conside ed a g owing end (Su i e al. 2018).
Some gene al amewo ks and e e ence a chi ec u es o IIoT also add ess busi-
ness p ocess pe spec i es. Fo example, he Re e ence A chi ec u al Model Indus y
4.0 (RAMI 4.0) is a h ee-dimensional model showing how o app oach Indus y
4.0, including IIoT, in a s uc u ed manne (DIN SPEC 91345, 2016). As one o he
main laye s o he model, business p ocesses a e add essed. Ano he example is he
Indus ial In e ne Re e ence A chi ec u e (IIRA), ha p o ides guidance o imple-
men ing IIoT echnology in o ganiza ions (Indus y IoT Conso ium 2022). Wi hin
he guidelines, a speci ic business iew is desc ibed ha highligh s he impo ance
o a business p ocess pe spec i e.
Focusing on IIoT as an enable o BPI, some a icles p o ide an o e iew o he
capabili ies o IIoT o edesigning and imp o ing business p ocesses. A publica-
ion by Halle e al. (2009) desc ibes majo applica ion a eas whe e IIoT can gene -
a e business alue. Also, Chui e al. (2010) speci ically highligh ed he impo ance
o IIoT o imp o ing business p ocesses while de ining six eme ging applica ions.
Yang e al. (2018) s a ed ha IIoT could be used o edesign p oduc ion p ocesses
and, hus, achie e highe e iciency. A s udy by Ses ino e al. (2020) shows how
IIoT can be used o BPI and illus a es se e al cen al opics in cu en li e a u e.
Mo eo e , A nold e al. (2016) p esen ed he impac o IIoT echnology on business
290
C.S oibe , S.Schönig
1 3
models and business p ocesses in di e en manu ac u ing indus ies. Mo e ecen
esea ch by D echsle e al. (2022) desc ibes how he eme gence o IIoT echnolo-
gies ollows a cascading g ow h pa e n.
Rega ding BPI pa e ns, especially no ewo hy is he con ibu ion o Falk e al.
(2013), who c ea ed an explici me amodel ha enables he c ea ion and o mula-
ion o BPI pa e ns and can be used as a empla e and basis o u he esea ch.
Fu he mo e, pa e ns ha e also been applied o se e al IoT o IIoT- ela ed opics.
As he IIoT consis s o di e en laye s, comp ising pe cei ing, ne wo king, o da a
p ocessing echnologies, many di e en pa e ns can be o mula ed ha suppo
sys em enginee s wi h in eg a ing whole applica ions in o business en i onmen s.
The design and a chi ec u e o IIoT sys ems can eminen ly bene i om pa e ns ha
assis in designing scalable and eplicable IIoT applica ions (Washizaki e al. 2020).
Ano he ocus wi hin his esea ch a ea is on da a exchange and ne wo k echnology
pa e ns along wi h mul iple connec ed de ices, machines, o p ocess en i ies (Rein-
u e al. 2016). Howe e , a comp ehensi e concep ualiza ion o IIoT-based BPI
ha explici ly desc ibes and speci ies he phenomenon has no been add essed ye .
3 Resea ch me hodology
Since he o mula ed RQs aim o c ea e no el a i ac s, he DSR pa adigm was aken
as a basis. DSR seeks o guide esea che s du ing he p o ision o usable a i ac s
such as cons uc s, models, me hods, o ins an ia ions (He ne e al. 2004). To
de elop he me amodel, he s uc u ed p ocedu e o Pe e s e al. (2007) was applied.
This p o en me hod is based on he me hodology o He ne e al. (2004) and p o-
ides de ailed phases o ca y ou DSR. In his espec , he conduc ed esea ch is
composed o he ollowing esea ch ac i i ies: (i) he design and de elopmen o a
me amodel, (ii) he de i a ion o an ini ial se o pa e ns, and (iii) a summa i e e al-
ua ion in he o m o a case s udy. Figu e1 gi es a b ie desc ip ion o he ac i i ies
and hei sec ion e e ences.
In he i s esea ch ac i i y, he me amodel o IIoT-based BPI applica ions was
designed, and a comp ehensi e o ma i e e alua ion was pe o med. In con as o
c ea ing an en i ely new me amodel om sc a ch, he imp o emen and e ision
o an exis ing and hema ically ela ed me amodel enabled he adop ion o p o en
concep s and ideas. The e o e, acco ding o Falk e al. (2013), he me amodel o
BPI pa e ns se ed as he base o de elopmen . I was gene ic enough o ep e-
sen all pa e ns o IIoT-based BPI since hese ep esen a subse o BPI pa e ns.
Howe e , i was no speci ic enough o app op ia ely illumina e he aspec s o he
IIoT domain due o i s complexi y and unique ea u es. Fo his eason, he baseline
me amodel equi ed adap ion conce ning IIoT. As in he o iginal me amodel, a class
diag am was used o modeling as i p o ided su icien seman ic exp essi eness o
me amodeling.
To adap he base me amodel, wo de elopmen i e a ions we e conduc ed.
Wi hin he i s design i e a ion, an explo a i e induc i e app oach was selec ed.
In his espec , an ex ensi e SLR in es iga ed li e a u e desc ibing IIoT applica-
ions wi h BPI e e ence. The SLR is u he de ailed in Sec . 4.2. Subsequen ly,
291
1 3
Le e aging heindus ial in e ne o hings o business…
he iden i ied li e a u e was analyzed applying G ounded heo y and i s me hods o
open and axial coding (Co bin and S auss 1990). This enabled he iden i ica ion o
indispensable aspec s o IIoT-based BPI, which could en ich he base me amodel.
Wi hin he au ho eam, he me hod o induc i e easoning (Hempel 1966) was
applied o c i ically discuss he indings and selec he mos app op ia e me amodel
adap ions. Wi hin he second i e a ion, addi ional expe knowledge was included
in he esea ch app oach. He eo , a Delphi s udy wi h nine indus y and academia
expe s was conduc ed o e ine he me amodel. In ou ounds, he expe s we e
asked o a e and e en ually adap he me amodel based on hei expe ise in he
esea ch a ea.
In he second esea ch ac i i y, he me amodel was demons a ed o h ee o gani-
za ions in he manu ac u ing, pha maceu ical, and au omo i e indus ies. Se en
p ac i ione s om di e en depa men s analyzed 34 IIoT applica ions in hei
di e en business a eas o de i e a i s se o pa e ns. The indus y expe s used
undamen al p inciples o induc i e easoning o analyze he applica ions, iden i y
gene ic aspec s, and c ea e i e pa e ns.
To e alua e he use ulness o he design a i ac s, he pa e ns we e in oduced o
an indus ial o ganiza ion in he hi d and inal esea ch ac i i y. In his espec , hey
se ed as he basis o undamen ally edesign and imp o e an impo an dis ibu ion
p ocess in he Scandina ian egion.
4 Me amodel de elopmen
4.1 The base me amodel
The de elopmen o he IIoT-based BPI me amodel builds upon p io wo k o Falk
e al. (2013), who designed a BPI me amodel, illus a ed in Fig.2. In hei app oach,
Ac i i y Desc ip ion
Me amodel de elopmen
Pa e n c ea ion
Fi s design i e a ion:
Adap ion o baseline me amodel
applyingSLR and G ounded heo y
Summa i e e alua ion
Second design i e a ion:
Re inemen o me amodel applying
Delphi s udy
De i a ion o i e pa e ns om34
indus y applica ions
Case s udy
Sec ion e e ence
4.2
4.3
6
7
Fig. 1 Pe o med esea ch ac i i ies
292
C.S oibe , S.Schönig
1 3
he BPI me amodel is ep esen ed as a class diag am acco ding o UML 2.0, whe e
each pa e n elemen is depic ed as a speci ic class. The p ope ies o hese classes
a e desc ibed using a ibu es, and he ela ionships be ween he classes a e ep e-
sen ed by undi ec ed bina y associa ions wi h speci ied mul iplici ies. These mul i-
plici ies de ine he ela ionships be ween indi idual objec classes. A he co e o he
me amodel is he BPI Pa e n class, which is ins an ia ed wi h a unique Name and
an Example. The name desc ibes he pa e n’s o e all pu pose and enables i s unique
iden i ica ion, while he example gi es in o ma ion on po en ial ins an ia ions. In
addi ion, he e is he class P oblem, de ined by he a ibu es Name, Desc ip ion,
and he ac ual Consequences o he p oblem o he p ocess. Fu he mo e, he Con-
ex class is di ec ly ela ed o he class BPI Pa e n. I is explained by a Name and
con ex -speci ic Cha ac e is ics and desc ibes he equi ed ci cums ances o he
pa e n o be applicable. Each pa e n exis s in exac ly one con ex , bu mul iple pa -
e ns can exis in he same con ex .
Each pa e n also con ains a Solu ion, desc ibed by a Name and he Measu es
equi ed o achie e he goal. The same solu ion can again be applied o mul iple
pa e ns, bu each pa e n has only one solu ion. Bound o he solu ion a e one o
mo e Mechanisms, each de ined by a Name and p ecise ac ion Ins uc ions. In addi-
ion, a solu ion can op ionally con ain one o mo e Building Blocks. These building
blocks a e p ede ined models ha can be implemen ed o sol e he p oblem wi hou
cus omiza ion. In addi ion, he pa e n is ela ed o an E ec , which is de ined by a
Name and he BPI dimensions o Cos , Time, Quali y, and Flexibili y (Dumas e al.
2018). Finally, each pa e n is ela ed o one o mo e Pe o mance Indica o s. These
a e de ined by a Name and Pe o mance Measu es ha can be used o ep esen he
imp o emen a e he pa e n has been implemen ed.
4.2 Fi s de elopmen i e a ion
To adap he base me amodel, a i s induc i e de elopmen i e a ion was conduc ed.
Induc i e app oaches in ol e p ocessing in o ma ion om subsys ems o o m a
pe cep ion o a op-le el sys em. This app oach is sui able o analyzing ini ially
Fig. 2 Base me amodel by Falk e al. (2013)
299
1 3
Le e aging heindus ial in e ne o hings o business…
senso s, ne wo king, and da a p ocessing echnologies. The i s a ibu e o he class
is Key Capabili ies. IIoT echnology encompasses no el and dis up i e capabili ies
ha se i apa om o he echnologies. To e ec i ely le e age hese capabili ies
o bene icial BPIs, hey mus be sys ema ically exploi ed. While he combina ion
o hese capabili ies is o en ele an in IIoT-based BPI, he e a e usually dis inc
key capabili ies ha a e pa icula ly ele an in each case. Such capabili ies include
uni e sal scalabili y, comp ehensi e pe cep ion, embedded in elligence, and in e -
ope abili y. U ilizing speci ic IIoT echnologies and exploi ing a se o capabili ies
can de ine he IIoT Ma u i y. In his con ex , ma u i y e e s o he complexi y o
an IIoT applica ion, i s le el o in eg a ion in o he p ocess, and he alue i gene -
a es. I anges om simple da a collec ion and analy ics o ully au oma ed asks. Tai
Angus Lai e al. (2018) ha e iden i ied a ious possibili ies o de ining IIoT ma u-
i y, such as si ua ional awa eness, decision-making suppo , in o ma ion exchange,
and au onomous sys ems.
Finally, he class P ocess Pe spec i e was added o he me amodel. I desc ibes
he pe spec i es and, he e o e, p ocess aspec s ha a e in luenced mos by he IIoT
applica ion. This class is pa icula ly use ul o illus a ing how he IIoT applica-
ion a ec s and edesigns he p ocess. Jablonski and Bussle (1996) ha e de ined six
p ocess pe spec i es ha can be applied in his con ex . The beha io al pe spec i e
includes elemen s such as he co ec p ocess wo k low o sequence, legal egula-
ions, and in e nal equi emen s. The o ganiza ional pe spec i e ocuses on he pe -
sonnel in ol ed in p ocess execu ion, including p ocess owne s, adminis a o s, and
use s. The unde lying sys em, such as he IT en i onmen , is also pa o his pe -
spec i e. The unc ional pe spec i e encompasses conc e e p ocess s eps, asks, and
e en s. In indus ial se ings, many p ocesses in ol e mul iple machines, ools, and
so wa e applica ions, which can be desc ibed om an ope a ional pe spec i e. The
da a pe spec i e in ol es all da a and documen s necessa y o p ocess execu ion.
Las ly, he loca ional pe spec i e desc ibes he speci ic loca ions o p ocess en i ies,
such as machines o wo ke s.
6 C ea ion o ini ial pa e ns
Once he me amodel was designed, i was used o illus a e an ini ial se o pa e ns.
Fo he pa e n de i a ion, se en p ac i ione s om h ee mul ina ional indus ial
o ganiza ions we e engaged. Thei oles included echnical p ojec manage s, IT
manage s, au oma ion expe s, and digi aliza ion manage s, and hey we e espon-
sible o o e seeing IIoT p ojec s wi hin hei espec i e o ganiza ions. Mo e de ails
abou hese expe s can be ound in Table1.
Du ing hei analysis, he p ac i ione s iden i ied 34 IIoT applica ions wi h BPI
p oposi ions ha we e assessed as sui able o u he examina ion. In a collabo a-
i e wo kshop conduc ed ia ideo con e ence in Feb ua y 2021, six dis inc pa -
e ns we e de i ed and illus a ed using he p o ided me amodel. These pa e ns a e
P ocess Guidance, De i a ion De ec ion, Au hen ica ion and Au ho iza ion, Task
Dis ibu ion, and Ac i i y Au oma ion. In he ollowing subsec ions, hey a e b ie ly
desc ibed.
300
C.S oibe , S.Schönig
1 3
In gene al, he pa e n de ini ion is a c ea i e p ocedu e ha is based p ima -
ily on unde lying da a. Fo example, se e al di e en IIoT use cases could be an
app op ia e basis o his pu pose. The ac ual de i a ion, o disco e ing, o pa -
e ns could be based on classi ica ion o coding echniques. The echniques o
open, axial, and selec i e coding a e pa icula ly sui able o his.
6.1 P ocess guidance
The i s pa e n P ocess Guidance (see Fig.7) p o ides a gene ic desc ip ion o
applica ions ha ocus on enhanced use guidance wi hin a business p ocess. By
cap u ing si ua ional and p ocess- ela ed da a, hese applica ions de e mine he
cu en s a e o he p ocess and iden i y he subsequen asks. The iden i ied asks
can hen be displayed o he p ocess pa icipan s, o example, h ough wea ables
o displays. This pa e n p ima ily a ec s he ope a ional and da a pe spec i es,
in ol ing inpu and ou pu da a o in luence ask pe o mance.
The sma de ices employed in his pa e n a e p ocess-awa e, as hey a e
esponsible o cap u ing p ocess- ela ed da a, p ocessing i , and p o iding i o
he p ocess pa icipan s based on he cu en p ocess s a e. An example men-
ioned in he pa e n is om an au omo i e o ganiza ion whe e employees ecei e
isual ins uc ions and indica ions ha guide hem h ough each p ocess ask. The
IIoT sys em, wi h i s in eg a ed senso s, comp ehensi e pe cep ion, and embed-
ded in elligence, enables acking o he ac ual p ocess low. This allows o he
p o ision o in o ma ion abou he cu en ask using di e en colo ed ligh ba s,
BPI Pa e n
P oblem
Con ex
Solu ion
Name: P ocessAmbigui ies.1
Desc ip ion: Fo new p ocess use s
o complex p ocesses, he co ec
ask pe o mances and p ocess
sequences a e di icul o
unde s and.
Consequences:Long p ocessing
imes, dissa is ied use s, e o a es
Name: Con ex .1
Cha ac e is ics: P ocess
sequences depend on p ocess and
si ua ional da a.
Name: Da a-based P ocess
Guidance.1
Measu es:In eg a ing senso s o
moni o si ua ional da a. Mapping
da a wi h p ocess sequences o
display guidance o asks.
Name: P ocessGuidance.1
Example: Du ing he assembly
o ca s, employees ge isual
ins uc ions and indica ions
ega ding hei p ocess asks.
The IIoT sys em ecognizes ask
pe o mances using senso s and
guides h ough asks acco ding
o he collec ed da a and he
unde lying p ocess model. The
sys ems guides du ing he
selec ion o pa s and indica es
he loca ion o moun ing.
IIoTTechnology
Name: Pe cep ion and P o ision.1
Sma De ices: P ocess-awa e
sys ems cap u ing p ocess and
si ua ional da a ia senso s.
Audio isual de ices displaying
p ocess in o ma ion, e.g., wea ables
o ligh ba s.
Da a P ocessing: P ocessing
si ua ional da a in edge-de ices and
mapping wi h expec ed p ocess-da a
om, e.g., BPMS. T igge ing
subsequen p ocess asks and
p o iding explana o y in o ma ion.
Mechanism
Name: Da a Collec ion and
P o ision.1
Ins uc ion: Collec ing p ocess-
ela ed and si ua ional da a o enable
p o ision o co ec p ocess
sequences and ask desc ip ions.
In e ac ion
Name: Use Guidance.1
Human In ol emen : P ocess use s
a e pe cei ing in o ma ion on p ocess
asks om he IIoT de ice. They need
o ollow he p o ided guidance.
Value P oposi ion
Name: In o ma ion Exchange.1
Key Capabili ies:Comp ehensi e
pe cep ion, embedded in elligence
IIoT Ma u i y: In o ma ion exchange
E ec
Name: Quali y.1
Pe o mance Measu es:
E o a e, epe i ion loops
Name: Ope a ional.1
Desc ip ion: Guided ask
execu ion
Name: E ec .1
Cos :(0) neu al
Time: (+) posi i e
Quali y: (+) posi i e
Flexibili y: (-) nega i e
Name: Da a.1
Desc ip ion: Cap u ing and
p o iding si ua ional da a
Pe o mance Indica o
P ocessPe spec i e
P ocessPe spec i e
Name: Time.1
Pe o mance Measu es:
P ocessing ime
Pe o mance Indica o
Fig. 7 P ocess guidance pa e n
301
1 3
Le e aging heindus ial in e ne o hings o business…
o ins ance. Implemen ing his pa e n has shown imp o emen s in p ocessing
ime and dec eased e o a es and epe i ion loops.
An exempla y li e a u e applica ion o his kind is he aining o new employ-
ees in a manu ac u ing o ganiza ion. Employees a e guided h ough p ocess asks
by acking he cu en p ocess da a and isualizing p ocess desc ip ions o subse-
quen asks. O he o ganiza ions ha e implemen ed applica ions o guide employ-
ees h ough p oduc ion o logis ic p ocesses by cap u ing en i onmen al and p ocess
da a, p ocessing i , ma ching i wi h p ocess models, and p o iding guidance o
asks (De V ies e al. 2015).
6.2 De ia ion de ec ion
The second pa e n De ia ion De ec ion is illus a ed using a gas bo le illing p o-
cess in he chemical indus y, as seen in Fig. 8. The p ima y challenge aced by
o ganiza ions is o de ec p ocess de ia ions du ing un ime, allowing hem o iden-
i y inco ec ask execu ions and app op ia ely adjus he subsequen p ocess lows.
De ia ions in he p ocess can esul in low p ocess quali y, p ocess deadlocks, o he
need o p ocess suppo . In he case o he gas bo le illing p ocess, i is c ucial o
ensu e ha a e illing he oxic gas bo les, hey a e placed in he co ec a eas as
speci ied by he p ocess desc ip ion. Inco ec ask execu ions in his con ex ca -
ies a high- isk po en ial. To add ess his, he pa e n p oposes implemen ing loca-
ion senso s ha collec ask execu ion da a and compa e i wi h he expec ed alues
de i ed om he p ocess desc ip ion. By analyzing his da a, de ia ions om he
expec ed beha io can be de ec ed.
BPI Pa e nE ec
P oblem
Con ex
Solu ion
Name: Quali y.2
Pe o mance Measu es:
E o ecogni ion, ewo k a e
Name: P ocess In anspa encies.1
Desc ip ion: The co ec p ocess
ask execu ion is ha d o e i y.
Inco ec execu ions ha e a nega i e
impac on subsequen asks and he
whole p ocess. The selec ion o
app op ia e coun e measu es
depends on he de ia ion
iden i ica ion.
Consequences:Low p ocess
quali y, p ocess deadlocks, sc ap
and ewo k
Name: Con ex .2
Cha ac e is ics: Poo ly o inco ec ly
execu ed asks mus be iden i ied.
App op ia e coun e measu es mus
be ini ia ed.
Name: P ocess Da a Reconcili aion.1
Measu es: In eg a ing senso s o
collec and analyze p ocess ask
da a. Mapping collec ed da a wi h
expec ed da a o ind de ia ions.
Ini ia ing coun e measu es, i
equi ed.
Name: De ia ion De ec ion.1
Example: A e he illing o oxic
gas cylinde s, he wo ke mus place
he cylinde s in he igh a ea o
subsequen asks. I he cylinde s
a e placed in he w ong a ea, his
in ol es high isk po en ial, e.g., o
placing in ood o chemical gas
a eas. Senso s collec loca ional
da a o he oxic cylinde s. This da a
is p ocessed, analyzed, and
mapped wi h expec ed da a o he
p ocess ask. I he pe cei ed da a
does no comply wi h he expec ed
loca ional da a, a de ia ion is
de ec ed. This in o ma ion can be
used o p e en a w ong placemen
o o ini ia e a eposi ioning o he
oxic cylinde s and a sc apping o
he ood o medical gases.
Name: Func ional.1
Desc ip ion: P ocess ask
moni o ing
Name: E ec .2
Cos :(0) neu al
Time: (0) neu al
Quali y: (+) posi i e
Flexibili y: (0) neu al
Name: Da a.2
Desc ip ion: Collec ion and
econcilia ion o da a
IIoT Technology
Name: Sensing.1
Sma De ices: Ac i i y-awa e
sys ems sensing and collec ing
da a.
Da a P ocessing: P ocessing
si ua ional da a in edge-de ices
o (cloud-) se e s. Compa ing
da a wi h de ined
alues/ h esholds.
Pe o mance Indica o
Mechanism
Name: Da a Collec ion and
Reconcilia ion.1
Ins uc ion: Sensing,
collec ing, and analyzing ask
da a. Mapping wi h expec ed
da a and ini ia ing
coun e measu es, i equi ed.
P ocessPe spec i e
P ocessPe spec i e
Value P oposi ion
Name: Si ua ional Awa eness.1
Key Capabili ies:
Comp ehensi e pe cep ion
IIoT Ma u i y: Si ua ional
awa eness
Fig. 8 De ia ion de ec ion pa e n
302
C.S oibe , S.Schönig
1 3
This de ec ion enables he ini ia ion o coun e measu es and imp o ed e o ec-
ogni ion a e, posi i ely impac ing he o e all p ocess quali y. The pa e n add esses
he unc ional and da a pe spec i es as he execu ion o he p ocess ask is moni-
o ed. The IIoT echnology includes ac i i y-awa e sma de ices ha p ocess si u-
a ional da a on edge de ices o (hyb id) cloud se e s. To iden i y de ia ions o any
kind, he key capabili y o comp ehensi e pe cep ion mus be exploi ed o enable
si ua ional awa eness o all p ocess de ails. The implemen ed pa e n imp o es p o-
cess quali y in e ms o inc eased e o ecogni ion a e and dec eased ewo k a e.
Simila indus ial applica ions can be ound o de ec machine ailu es whe e senso
da a is used o diagnos ics and de ec ion o de ia ions, e.g., a leakage de ec ion
(Ammi a o e al. 2019) o o he anomalies (Schneide e al. 2019).
6.3 Au hen ica ion andau ho iza ion
Many business p ocesses equi e au ho ized use s o gua an ee p ocess sa e y and
quali y. Thus, po en ial use s mus au hen ica e hei iden i y o check i hey a e
au ho ized o pe o m speci ic asks. The hi d pa e n Au hen ica ion and Au ho i-
za ion sol es his challenge. The exempla y p ocess is aken om a ca assem-
bly p ocess o a majo au omo i e o ganiza ion. Du ing he assembly p ocess, he
sa e y–c i ical an i-lock b aking sys em (ABS) module mus be co ec ly placed,
ixed, and connec ed o he ca . As his should only be pe o med by ained pe son-
nel, au ho iza ion is equi ed. The employee mus au hen ica e o access he espec-
i e ABS module s o age con aine . This con aine is only unlocked i he use has
au hen ica ed himsel wi h alid c eden ials and he employee’s loca ion is di ec ly
a he espec i e ca . This can be done using sma wa ches o o he pe sonal IIoT
de ices (Fig.9).
The pa e n mainly in luences he o ganiza ional and da a p ocess pe spec i es, as
use and p ocess da a a e collec ed and used o au hen ica ing he p ocess pa ici-
pan . This has a highly posi i e impac on p ocess quali y as i imp o es he o e all
p ocess sa e y. Simila IIoT applica ions can be iden i ied in se e al indus y appli-
ca ions aiming a checking cus ome au ho iza ion and au hen ica ing employees in
he p oduc ion a ea o he logis ics sec o . Fo ins ance, he ma e ial handling o
special p oduc s is only allowed o ained pe sonnel, whe e o e an au hen ica ion is
necessa y (De Souza e al. 2020).
6.4 Task dis ibu ion
The ou h pa e n Task Dis ibu ion o igina es om he p oblem o au oma ed and
e icien alloca ion o p ocess asks o a mul i ude o p ocess use s wi h speci ic
knowledge and skills. In he co uga ion indus y, he di e en p oduc ion p ocess
asks mus be dis ibu ed o wo ke s acco ding o hei si ua ional and pe sonal cha -
ac e is ics. This is pe o med by cap u ing da a on hei cu en loca ion, a igue,
skills, and compe encies and ma ching i wi h p ocess ask equi emen s (Fig.10).
The pa ame e s can be mapped wi h he ask da a by p ocessing da a om
ac i i y-awa e sys ems on edge o cloud de ices. The ask can be displayed ia
303
1 3
Le e aging heindus ial in e ne o hings o business…
BPI Pa e n
E ec
P oblem
Con ex
Solu ion
Name: Quali y.3
Pe o mance Measu es:
P ocesssa e y
Name: Lack o Use Au hen ica ion.1
Desc ip ion: Speci ic p ocess asks
equi e au ho ized use s o
gua an ee co ec and sa e
pe o mances. Use and si ua ional
da a is no a ailable o use
au hen ica ion.
Consequences:Sa e y isks,
malp ac ices, he cases, poo
p ocessquali y
Name: Con ex .3
Cha ac e is ics: Be o e s a ing a
ask, use s mus au hen ica e o
check o au ho iza ion. This is done
using pe sonal and si ua ional da a.
Name: Use Da a Ve i ica ion.1
Measu es:Implemen ing s eps o
au ho iza ion. Use s mus
au hen ica e hemsel esbased on
pe sonal and si ua ional da a.
Name: Au hen ica ion and
Au ho iza ion.1
Example: Du ing ca assembly,
sa e y-c i ical ABS modules ha e o
be ins alled a a la e s age in he
p ocess. This may only be ca ied
ou by au ho ized employees. Be o e
each indi idual ABS module
ins alla ion, an au ho ized employee
mus au hen ica e o ge access o
he ABS module and s a he
assembly ask. In addi ion, he
employee’s loca ion is acked using
his espec i e sma wa ch. This
ensu es ha he employee is di ec ly
on si e a he co ec loca ion and
ca .
Name: O ganiza ional.1
Desc ip ion: P ocess ask
moni o ing
Name: E ec .3
Cos :(0) neu al
Time: (0) neu al
Quali y: (+) posi i e
Flexibili y: (0) neu al
Name: Da a.3
Desc ip ion: Use da a
collec ion, da a econcilia ion
IIoT Technology
Name: Da a Collec ion and
Ve i ica ion.1
Sma De ices: Policy-awa e
de ices ha lock o unlock he
p ocessacco ding o use da a.
So wa e applica ions o
connec edphysical de ices o
use da acap u ing.
Da a P ocessing: P ocessing
and e i ica iono use da ain
cloud da abase. T igge ing
subsequen p ocesss eps
acco dingly.
Pe o mance Indica o
Mechanism
Name: Au ho izeP ocess
Pa icipan s.1
Ins uc ion: Collec inguse
da a, de e mining au ho iza ion
le el, and ini ia ing p ocesss eps
acco dingly.
P ocessPe spec i e
P ocessPe spec i e
Value P oposi ion
Name: Decision-making Suppo .1
Key Capabili ies:Embedded
in elligence
IIoTMa u i y: Decision-making
suppo
In e ac ion
Name: Da a En y.1
Human In ol emen : P ocess
use sa e ac i ely o passi ely
en e ingda a ia senso so
applica ions.
Fig. 9 Au hen ica ion and au ho iza ion pa e n
BPI Pa e n
E ec
P oblem
Con ex
Solu ion
Name: Quali y.4
Pe o mance Measu es:
Task pe o mance
Name: Responsibili yAmbigui ies.1
Desc ip ion: Responsibili ies o
asksa e no clea . Dis ibu ion is
pe o medmanually and/o
ine icien ly.
Consequences:High ime
consump ion, Poo p ocessquali y
Name:Con ex .4
Cha ac e is ics: Whileha ing
se e alwo ke s andcomplex p ocess
ac i i ies,e icien alloca ion o asks
is challenging.
Name: Op imizing Dis ibu ion.1
Measu es: Cap u ingsi ua ional
and/o pe sonal da a o dis ibu e
p ocess asks (o dina yand
excep ional) in an e icien manne .
Displaying asks o espec i e
pe sonnel ia audio isualde ices.
Name: Task Dis ibu ion.1
Example: Du ing hep oduc ion
p ocesso co uga o s,wo ke s ge
no i ica ionso necessa y asks ha
need o be pe o med. These asks
a edis ibu ed anddisplayed o he
app op ia ewo ke o g oup o
wo ke s. The dis ibu ion is basedon
si ua ional and/o pe sonal da a
cap u ed by senso s, e.g., wo ke
loca ion, a igue,o skill se .
Displaying he askin o ma ion is
done ia wea ables, e.g.,
sma wa ches o augmen ed eali y
glasses.
Name: O ganiza ional.2
Desc ip ion: Selec iono
p ocessuse s
Name: E ec .4
Cos :(0) neu al
Time: (+) posi i e
Quali y: (0) neu al
Flexibili y: (0) neu al
Name: Da a.4
Desc ip ion: Cap u ing
si ua ional o pe sonal da a,
p o idingp ocess da a
IIoTTechnology
Name: Sensingand Op imizing.1
Sma De ices: Simple ac i i y-
awa e de iceswi hsenso s
cap u ingsi ua ional and/o
pe sonal da a.Audio isualdisplay
de ices, e.g. wea ables.
Da a P ocessing: P ocessing
da a on edge de iceso cloud.
Mapping wi h ask equi emen
da a.Op imizing askalloca ion
andp o iding da a.
Pe o mance Indica o
Mechanism
Name: E icien Task Alloca ion.1
Ins uc ion: Cap u ingsi ua ional
and/o pe sonal da a and
ma ching ask ela ed
equi emen s.Dis ibu ing asks
anddisplay in o ma ion ouse s.
P ocessPe spec i e
P ocessPe spec i e
Value P oposi ion
Name: Decision-making Suppo .1
Key Capabili ies:Embedded
in elligence
IIoT Ma u i y: Decision-making
suppo
In e ac ion
Name: Da a En y.1
Human In ol emen : P ocess
use sa e ac i ely o passi ely
en e ing da a ia senso so
applica ions.
In e ac ion
Name: Da a Visualiza ion.1
Human In ol emen : P ocess
use sa e ecei ingp ocess da a
ia audio isualin e aces, e.g.,
wea ableso applica ions.
Fig. 10 Task dis ibu ion pa e n
304
C.S oibe , S.Schönig
1 3
wea ables, e.g., sma wa ches o o he audio isual de ices, when dis ibu ed o a
speci ic wo ke . This imp o es he p ocess quali y, as he mos app op ia e pe son
pe o ms he asks. The pa e n can also be iden i ied along whole supply chains o
achie e an IIoT-enabled ask alloca ion op imiza ion. By pe o ming in o ma ion-
d i en dynamic op imiza ions based on IIoT da a, he dis ibu ion o logis ics asks
along he supply chain en i ies can be signi ican ly imp o ed (Liu e al. 2018).
6.5 Ac i i y au oma ion
Replacing manual p ocess asks wi h au oma ed asks con ains signi ican bene i s
o o ganiza ions. As manual p ocess s eps equi e ained pe sonnel, i is associa ed
wi h high labo cos s and high wo king ime consump ion. The i h pa e n Ac i i y
Au oma ion add esses his p oblem. The example desc ibed in Fig.11 is aken om
he manu ac u ing indus y. He e, o kli d i e s needed o manually scan palle
ba codes and s o age loca ion ba codes o enable e ec i e ma e ial acking. Using
loca ion senso s and p o iding his in o ma ion o o e a ching con ol sys ems can
enable au oma ed acking and acing. Senso s cap u e loca ion da a ha is p o-
cessed on edge de ices o cloud se ices. Based on he da a inpu , mechanical o
so wa e-based eac ions can simula e manual ac i i ies. These sys ems o m a he
complex p ocess-awa e IIoT de ices ha can educe labo cos s, wo king ime, and
o e all p ocessing ime. As he IIoT echnology edesigns he ac ual p ocess low, i
mainly in luences he unc ional p ocess pe spec i e.
The pa e n has a posi i e impac on labo cos s as well as wo king ime con-
sump ion. Imp o ing business p ocesses by au oma ing ac i i ies is one o he li -
e a u e’s mos ele an and equen pa e ns. Se e al use cases ha e been iden i ied
du ing he li e a u e e iew desc ibing au onomous sys ems con aining high-com-
plexi y IIoT sys ems. Li e al. (2017), e.g., desc ibe a ully au onomous sys em in
BPI Pa e n
E ec
P oblem
Con ex
Solu ion Name: Cos .1
Pe o mance Measu es:
Labo cos
Name: Manual Ac i i ies.1
Desc ip ion: Manual p ocess asks
equi e ained pe sonnel.
Consequences:High labo cos s,
high wo king ime consump ion
Name: Con ex .6
Cha ac e is ics: Many manual
p ocesss eps a eine icien and
equi eexpensi e human esou ces.
Name: Au oma ing P ocessS eps.1
Measu es:Re o i ing legacy
machines wi h senso sand ac ua o s
o ins alling newequipmen o
cap u eda aand eac acco dingly.
Redesign o p ocessbyau oma ing
exis ing ac i i ies.
Name: Ac i i y Au oma ion.1
Example: A s o agea eas,palle s
mus be loca edonspeci ics o age
posi ions. Fo kli d i e sa e placing
hepalle s andscan he palle
ba codeaswellas he s o age
ba code. This enablesaclea
acking o ma e ial. Usingloca ion
senso s, his scanning canbe
eplaced by acking heexac
posi ion o palle s in heou side
ya d. The Manu ac u ing Execu ion
Sys em can ack and ace all
ma e ial lowsau oma ically wi hou
manualac ions.
Name: Func ional.2
Desc ip ion: Redesign o
p ocess asks
Name: E ec .5
Cos :(+) posi i e
Time: (+) posi i e
Quali y: (0) neu al
Flexibili y: (0) neu al
IIoT Technology
Name: Sensingand Ac ua ing.1
Sma De ices: P ocess-awa e
de icescombining senso sand
ac ua o s.
Da a P ocessing: Sensing
p ocess- ela edda a and
p ocessing i on edge de iceso
cloud. Da a-based eac ion ia
ac ua o sacco ding on p ocess
low.
Pe o mance Indica o
Mechanism
Name: Au oma ion.1
Ins uc ion: Re o i ing
machines wi h senso sand
ac ua o s; ins alling new
equipmen .Au oma ingmanual
p ocessac i i ies.
Value P oposi ion
Name: Au onomous Sys ems.1
Key Capabili ies:Comp ehensi e
pe cep ion
IIoT Ma u i y: Au onomous
sys ems
Name: Time.3
Pe o mance Measu es:
Wo king ime
Pe o mance Indica o
P ocessPe spec i e
Fig. 11 Ac i i y au oma ion pa e n
305
1 3
Le e aging heindus ial in e ne o hings o business…
which he p oduc ion objec can au oma ically coo dina e wi h he p oduc ion and
anspo a ion machines and plan an op imal p oduc ion p ocess. The objec o be
p ocessed is equipped wi h a mic ochip o his pu pose and can ecei e indi idual
p oduc ion ins uc ions om he cloud o enable a maximally lexible and ully au o-
ma ed p ocess. O he examples desc ibe he au oma ion o whole ac i i ies wi hin
ood supply chains (Pang e al. 2015). This comp ehensi e s udy shows he band-
wid h o applica ion possibili ies o his pa e n.
7 Summa y e alua ion
To e alua e he use ulness o he pa e ns and he unde lying me amodel, hey we e
applied in a eal-wo ld case s udy a a leading ma ke playe in he chemical p od-
uc s indus y. In his espec , a p ojec eam used he pa e ns o de elop a no el
dis ibu ion channel o chemical p oduc s in he Scandina ian egion, and hus, un-
damen ally imp o ing he exis ing dis ibu ion p ocesses. The summa i e e alua ion
aimed o assess i he esea ch app oach enabled a sys ema ic and success ul iden-
i ica ion, speci ica ion, and implemen a ion o IIoT-based BPI applica ions. This
di ec ly e e s o he o mula ed RQ1 and he p e alen challenges ha indus ial
o ganiza ions ace.
7.1 Use case se up
Changing cus ome beha io and expec a ions ha e p esen ed subs an ial challenges
o he case s udy o ganiza ion wi hin he Scandina ian egion. P e iously, cus om-
e s we e limi ed o pu chasing p oduc s exclusi ely o e he coun e om ce i ied
e aile s o ia ending machines. Howe e , hese wo p ima y dis ibu ion channels
we e no longe su icien o ca e o he di e se needs o bo h p i a e and business
cus ome s. Nowadays, cus ome s expec online se ices, a single s anda dized dis-
ibu ion p ocess, 24/7 a ailabili y, o ex ensi e guidance. These and o he cus ome
equi emen s we e iden i ied wi hin an ex ensi e su ey. Al hough each o he wo
channels b ings ce ain ad an ages, none combines all equi emen s. As a esul , a
comp ehensi e edesign and imp o emen p ojec ega ding he dis ibu ion chan-
nels and he unde lying p ocesses was ini ia ed.
Ini ially, he o ganiza ion o med an in e disciplina y p ojec eam comp ising
se en indi iduals, i.e., p ocess owne s, sales s a , enginee s, and IT specialis s. The
au ho s o his a icle emained in a consul ing posi ion. The ini ia i e commenced
in 2019 and concluded in 2022, culmina ing in a ho oughly new dis ibu ion chan-
nel, including a edesigned dis ibu ion p ocess. Based on he cus ome su ey, a se
o design objec i es had been de i ed ha cons i u e undamen al equi emen s o
he new dis ibu ion p ocess. The no el solu ion should be au oma ed, s able, eco-
nomical, scalable, and enable online se ices and ea u es. Fu he mo e, i should
eplace he exis ing dis ibu ion channels in he long un while combining bo h
ad an ages.
306
C.S oibe , S.Schönig
1 3
7.2 De eloping hesma ending cabine
A comp ehensi e echnology analysis de e mined ha he IIoT’s capabili ies
could acili a e he design and implemen a ion o he en isioned dis ibu ion
channel. None heless, a sys ema ic app oach was essen ial o success ully iden-
i y sui able IIoT solu ions aligned wi h he design objec i es. In his espec , he
p esen ed pa e ns o IIoT-based BPI applica ions we e applied. Du ing a se ies
o wo kshops, he design objec i es we e ho oughly examined and mapped wi h
he pa e n ca alog. Beginning wi h de ailed desc ip ions o he design objec-
i es, he eam diligen ly sough p oblem desc ip ions wi hin he pa e ns o
add ess hem. E en ually, ou pa e ns we e success ully iden i ied, acili a ing
he a ainmen o all design objec i es. A combina ion o ha dwa e and so wa e
design and de elopmen was equi ed o ealize he newly designed dis ibu ion
channel, e e ed o as he Sma Vending Cabine .
The Sma Vending Cabine comp ises i e essen ial componen s: a s anda d
20- ee con aine , a modula IIoT ki encompassing senso s, ac ua o s, and elays,
as well as edge nodes, a cloud se e hos ing he business p ocess managemen
sys em (BPMS), o e ing APIs and handling he analysis and s o age o IIoT da a
o mul iple se ices. The 20- ee con aine was ou i ed wi h he IIoT senso and
ac ua o ki , p o iding su icien space o accommoda e a di e se ange o chemi-
cal p oduc s. I ope a es au onomously, wi hou needing on-si e sales ep esen a-
i es, and adhe es o he loca ion equi emen s o adi ional ending machines.
This new dis ibu ion channel and he no el dis ibu ion p ocess a e explained in
he ollowing.
The i s s ep wi hin he new dis ibu ion p ocess is he pu chase o chemical
p oduc s by he cus ome . Cus ome s can sea ch o speci ic p oduc s a hei nea -
es Sma Vending Cabine loca ion using an al eady exis ing sma phone applica-
ion. Subsequen ly, he cus ome s mus be physically p esen a he app op ia e loca-
ion and eques access o he cabine . The cabine le e ages he Au hen ica ion and
Au ho iza ion pa e n o de ec he cus ome ’s p esence ia Blue oo h. The cabine
doo is opened by ac ua o s only i he cus ome has placed a pu chase o de , ensu -
ing cos e iciency by elimina ing he need o addi ional human in e aces. This
is based on he Ac i i y Au oma ion pa e n. These p ocess s eps a e illus a ed in
Fig.12 on he le .
Once he cabine doo is open, he cus ome is p esen ed wi h he pu chased p od-
uc and can p oceed o emo e i . The P ocess Guidance pa e n uses ligh ba s o
isually guide he cus ome o he co ec p oduc . This ea u e p o es pa icula ly
help ul o cus ome s who a e new o he p ocess o un amilia wi h he p oduc s.
Figu e12 in he cen e illus a es his s ep.
To ensu e a s able and adap able p ocess, e o s a e de ec ed and communica ed o
he cus ome using he De ia ion De ec ion pa e n. Fo ins ance, i a cus ome mis-
akenly emo es he w ong p oduc , he ligh ba changes o ed, and push messages
a e sen o he cus ome ’s sma phone. The cus ome ecei es ins uc ions h ough he
ligh ba s and he sma phone applica ion o ec i y he e o . Figu e12 on he igh
shows wo exempla y push no i ica ions. The i s indica es ha he cus ome has
e ie ed he w ong p oduc and should e u n i o he espec i e slo . The second
307
1 3
Le e aging heindus ial in e ne o hings o business…
no i ica ion ale s he cus ome ha he cabine was no closed co ec ly a e e ie ing
he p oduc .
7.3 Resul s ande alua ion
The Sma Vending Cabine is cu en ly in he ollou phase, wi h se en exis ing loca-
ions in May 2023. An ini ial cus ome su ey e ealed a signi ican inc ease in sa -
is ac ion le els. E en cus ome s who had p e iously elied on a single es ablished
dis ibu ion channel exp essed hei sa is ac ion wi h he new channel and indica ed a
p e e ence o i in he u u e. Fo he indus ial o ganiza ion, a i s analysis showed a
36% dec ease in dis ibu ion cos s compa ed o he e aile p ocess and 9% compa ed
o ending machines. This is due o lowe pe sonnel and main enance cos s compa ed
wi h he legacy channels.
To e alua e he use ulness o he de eloped pa e ns, he au ho s conduc ed semi-
s uc u ed in e iews wi h all p ojec eam membe s. In his ega d, he in e iew ques-
ions we e o mula ed o indica e he e ec i eness and applicabili y o he pa e ns. The
in e iews showed consis en ly posi i e eedback and con i med he use ulness o he
pa e ns. Pa icula emphasis was placed on hei suppo o de eloping IIoT-based
echnical solu ions o exis ing design objec i es. Thanks o he gene ic ep esen a ion,
he p ojec pa icipan s could apply he pa e ns well o he eal scena io and map he
exempla y desc ip ions wi h he scena io a he chemical o ganiza ion. In gene al, he
case s udy p o ed he use ulness o he me amodel and pa e ns, whe e o e he esea ch
app oach is c i ically discussed in he subsequen sec ion.
8 Discussion
Exis ing esea ch does no p o ide adequa e suppo o he iden i ica ion o IIoT
applica ions ha enable app op ia e BPI. Fu he mo e, he p ecise de ini ion o
he "Ac o Imp o emen ," i.e., how exis ing business p ocesses a e p ecisely
Fig. 12 Sma ending cabine applica ion
308
C.S oibe , S.Schönig
1 3
ans o med o he imp o ed a ge s a e h ough IIoT in eg a ion, emains elusi e
in mos cases. To add ess hese challenges and expand he unde s anding o IIoT, a
me amodel o IIoT-based BPI and an ini ial se o i e pa e ns we e de eloped. The
subsequen subsec ions ou line he heo e ical and p ac ical con ibu ions and impli-
ca ions ha eme ged du ing he de elopmen , demons a ion, and e alua ion ac i i-
ies. Addi ionally, he limi a ions o he esea ch endea o a e p esen ed o p o ide a
comp ehensi e pe spec i e.
8.1 Theo e ical con ibu ions andimplica ions
F om a heo e ical pe spec i e, his a icle o e s insigh s in o he ole o IIoT as an
enable o impac ul BPI. The heo e ical con ibu ion lies in de eloping a me a-
model ha encompasses he essen ial elemen s and ela ionships be ween IIoT and
BPI. The co e heo e ical implica ions o he a icle can be summa ized in o wo
main aspec s: enhancing he desc ip i e knowledge o IIoT and BPI and es ablishing
a ounda ion o u u e esea ch on pa e ns.
As a heo e ical heo y and design a i ac , he me amodel and pa e ns com-
plemen he exis ing desc ip i e knowledge on IIoT and BPI. Ra he han solely
emphasizing echnology- ela ed cha ac e is ics, hey also ocus on p ocess- ela ed
p oblems. By highligh ing he p ocess- ela ed iew o IIoT applica ions, he me a-
model and pa e ns in oduce no el pe spec i es o he p edominan ly echnical and
enginee ing-cen ic unde s anding o IIoT. Consequen ly, hey p o ide a ounda ion
o heo y-led design and sense-making (G ego and He ne 2013).
The p ima y objec i e o he de eloped me amodel is o se e as a basis o c ea -
ing and illus a ing a comp ehensi e pa e n ca alog. Such a ca alog would be sup-
po i e in speci ying, iden i ying, and implemen ing IIoT applica ions mo e e ec-
i ely and sys ema ically.
8.2 P ac ical con ibu ions andimplica ions
The me amodel and pa e ns o e ele an and applicable insigh s in eal-wo ld
scena ios, add essing a gap in exis ing s udies on IIoT. P e ious esea ch and su -
eys among manage s ha e e ealed ha many IIoT p ojec s emain in he p oo o
concep o ea ly planning s ages and equen ly all sho o deli e ing he an ici-
pa ed bene i s. This unde sco es he need o suppo du ing he p ojec execu ion
phases, including iden i ica ion, speci ica ion, and implemen a ion. In his con ex ,
he demons a ion and e alua ion o he me amodel ha e highligh ed wo p ac ical
implica ions.
Fi s ly, u ilizing pa e ns based on he de eloped me amodel enables manage s o
explo e and iden i y sui able IIoT applica ions. The pa e ns include desc ip ions o
he unde lying p ocess p oblems, he a ec ed p ocess pe o mance indica o s, he
impac on p ocess pe spec i es, and gene ic illus a ions o po en ial solu ions. This
migh acili a e a goal- and p ocess-o ien ed iden i ica ion o applica ions. Addi ion-
ally, he pa e ns help p e en he de elopmen o misguided expec a ions o IIoT
