Sys ema ic De elopmen o ERP Modules using a Model-D i en
S a egy Focusing on he Use s
M. A. Oli e o1,2 a, L. Mo ales-T ujillo1b
, F. J. Domínguez-Mayo3c
and M. Mejías3d
1
Resea ch G oup IWT2, Uni e si y o Se ille, Spain
2
Is i u o di Scienza e Tecnologie dellIn o mazione, Consiglio Nazionale della Rice che, I aly
3
Depa men Languages and Compu e Sys ems, Uni e si y o Se ille, Spain
Keywo ds: Design Sp in , ERP, NDT, Odoo.
Abs ac : ERP sys ems a e composed o di e en unc ional modules on which each one add esses a di e en business
a ea. De elopmen s on hese modules a e managed independen ly on each one, which allow o handle and
add ess he managemen o many ela ed in o ma ion equi emen s. In his con ex he s a up G7Inno a ion
wo ks o i s p oduc iMEDEA. In his s udy we ha e combined wo me hodologies, Design Sp in and NDT
4.0; and sys ema ized he de elopmen o ERP sys em modules. This combina ion allows you o use Design
Sp in o gene a e and alida e p o o yping, and NDT 4.0 o do he s udy, analysis, and design o he so wa e
o be de eloped. In addi ion, acco ding o he speci ica ions de ined in NDT, he code gene a ion o he ERP
module can be au oma ed. This p oposal has been alida ed in a case s udy in collabo a ion wi h he s a up
G7Inno a ion, whe e we ha e applied bo h me hodologies on Odoo, an open sou ce ERP sys em based in
Py hon language. Thanks o he use o hese wo me hodologies we ha e p oduced a module ela ed o he
needs o he clinic by educing cos s, imes and human ailu es.
1
INTRODUCTION
The iMEDEA p ojec is a solu ion de eloped by
G7Inno a ion (G7Inno a ion, 2019). This p ojec is
based on he Odoo ERP (Odoo, 2019) and i s goal is
o acili a e he managemen o human ep oduc ion
uni s. I is composed o di e en modules ha can be
used independen ly, which o e a be e suppo o
business cus ome s, gi ing as esponse imes o hei
p oblems and wi h e icien in o ma ion managemen
(Delgado and Ma ín, 2000). All hese ea u es allow
imely decision making and lowe ope a ing cos s.
Wi hou a clea guidance, modules a e no p o-
duced acco ding o clien expec a ion, and delays a e
in oduced when modi ica ions a e needed a e e-
qui emen s has been de ined.
Wi h he aim o educing changes in la e s ages
o he p ojec and o p oduce modules acco ding o
clien expec a ion, his wo k p esen s a me hodolog-
ical p oposal ha encompasses wo me hodologies,
a
h ps://o cid.o g/0000-0002-6627-3699
b
h ps://o cid.o g/0000-0001-9554-1173
c
h ps://o cid.o g/0000-0003-3502-8858
d
h ps://o cid.o g/0000-0002-0353-6391
Design Sp in (Ban ield e al., 2015), (Knapp, 2018)
and NDT (Escalona e al., 2004), (Na iga ional De el-
opmen Techniques). These al eady exis ing me hod-
ologies ocus on he ea ly alida ion h ough p o o-
yping and he code gene a ion espec i ely. This e-
duce he lea ning cu e needed o simple de elop-
men s, inc ease p oduc i i y, and educe he need o
changes in he equi emen s.
In his wo k we ocus on he Odoo modules, being
he main p oduc de eloped by he s a up
G7Inno a ion. We ha e alida ed his p oposal by
gene a ing a new module wi h his company in a join
wo k.
The es o he a icle is o ganized as ollows: The
R&D p ojec is de ined in sec ion 2, he used ools in
sec ion 3 and he esea ch ends wi h he conclusions
in sec ion 4.
Figu e 1: R&D P ojec sequence.
The p oposal concludes wi h he ope a ions phase.
This phase is esponsible o ensu ing he p ope
2
R&D PROJECT
This guide is di e en ia ed in h ee s ages: (1) So -
wa e P oduc Disco e y, whe e we apply Design
Sp in o di ec ly know he eal needs o use s; (2)
De elopmen , whe e he p ojec is echnically spec-
i ied, we build and es he sys em in an una ended
way by using NDT and model ans o ma ions; and
(3) Ope a ion, o he deploymen and main enance
Figu e.1 summa izes his p ocess.
The so wa e p oduc disco e y s age aims o
know wha a e he needs and mo i a ions ha need a
solu ion. To his end, Design Sp in me hodology is
applied. This me hodology is s uc u ed in a se o
miles ones ha a e scheduled o be comple ed in i e
days. As a summa y o his me hodology, di e en
solu ion p oposals a e pu o wa d wi h he aim o
ob aining di e en app oaches ha help in decision-
making. Final decision o how he solu ion shall be
ely on he in o ma ion ha has been ob ained om
each al e na i e. Once he clien ag ees on he solu-
ion o be achie ed, he p o o ype is ske ched. The
p o o ype mus be ealis ic enough o inally be able
o alida e he solu ion p oposal wi h he use . The use
o p o o ypes o alida e he p ojec allows o dis- ca d
he al e na i es a low cos .
De elopmen o he p oduc based on he p o o-
ype is di ided in o h ee miles ones: echnical speci-
ica ion, cons uc ion and e i ica ion and alida ion.
The echnical speci ica ion de ines he s uc u e and
beha io o ou sys em using UML (OMG, 2015)
models and IFML (B ambilla e al., 2014) models.
Knowing he esul ha we wan o ob ain, hanks o
he p o o ypes, and once de ined he echnical a chi-
ec u e, he cons uc ion s age begins.
Du ing cons uc ion, we make use o model o ex
(M2T) ope a ions, which allow us o ob ain code om
design models. These ans o ma ions gene a e code
om he da a model o ou sys em, aking he diag am
o classes o he sys em om NDT as he sou ce
model. Simila ly, unc ion codes can be ob ained
h ough he unc ional equi emen s and use cases
model o e ed by NDT. The gene a ion o au oma ed
code inc eases p oduc i i y and educes he
possibili y o human ailu e.
Ve i ying and alida ing he quali y o he p ojec
is he las s ep be o e launching he p oduc in o p o-
duc ion. Du ing his s age, unc ional and accep ance
es s mus be ca ied ou , among o he s. To each a
s able es ing en i onmen , ano he ype o ans-
o ma ion is used, he Model o model (M2M) ans-
o ma ions o e ed by NDT. Using he CASE ool o
NDT, allow you o gene a e he documen a ion o he
es plan.
wo king o he sys em and i s echnological in-
as uc u e in p oduc ion en i onmen s when he ap-
plica ion is deployed (He ik, 2014).
3
USED TECHNOLOGIES
In he disco e y phase o he p oduc , he “Powe -
Poin ” ool has been used o de elop and alida e p o-
o ypes. Al hough he e a e nume ous al e na i es, o
quickly p o o ype, we ha e chosen his ool o i s
g ea lexibili y. Using he pa e n mode o his ool i
is possible o c ea e empla es o he in e ace o he
so wa e o be de eloped, which can be la e com-
ple ed by adding he con en ha we wan o alida e
wi h clien s o cus ome s. In addi ion, his ool allows
you o add links be ween objec s so ha you can sim-
ula e na iga ion among he slides by using i ems ha
simula es he bu ons o he p oposed inal sys em.
This ool p oduced e y p ecise esul s ha allowed us
o make he designed p o o ypes qui e ealis ic, ha
we e used o alida e he solu ion p oposal wi h he
use s in some in e iews. Gi en he ac ha we know
abou he Odoo in e ace, and he echnological limi-
a ions o he ea u es ha Odoo o e s, we ha e gen-
e a ed ealis ic p o o ypes a low cos . This allowed us
o easily disca d some al e na i es ha did no mee
use o cus ome expec a ions. This p ocess could be
also be done wi h he Odoo S udio ool, a ool ha
acili a es he gene a ion o modules wi h a g aphical
in e ace like D ag D op. Despi e i is ully in eg a ed
wi h Odoo and allow au ogene a ing he module, we
did no con empla e his op ion because o he di i-
cul ies i could in oduce when eusing he gene a ed
code. Fo G7Inno a ion i is no enough ha ing a sin-
gle module, bu also be able in eg a e i wi h hi d pa -
ies APIs which o he bes o ou knowledge i is no
easible wi h gene a ed module wi h Odoo S udio.
Once he p oduc is de ined, he phase o a o mal
de ini ion o he equi emen s. The echnical speci-
ica ion, wi hin he de elopmen phase has been ca -
ied ou using NDT-Sui e (Ga cía-Ga cía e al., 2014).
NDT-Sui e is a se o ools ha acili a es he use o
he NDT me hodology, which is in eg a ed on op o
he En e p ise A chi ec ool (A chi ec , 2019). Wi h
his ool, and s a ing om he ag eed p o o ype, we
ha e been able o de ine he sys em class diag ams,
ac o s, use cases wi h hei in e ac ion equi emen s
and some non- unc ional equi emen s.
Thanks o he abili y o pe o m M2M (OMG,
2008a) and M2T (OMG, 2008b) ans o ma ions o -
e ed by NDT-Sui e we au oma ically ob ained some
o hese equi emen s, Py hon code ha was used as
he basis o he cons uc ion phase and he documen-
a ion o he es plan o e i ica ion and alida ion.
The code sel -gene a ed by he M2T ans o ma-
ions has been used as he basis o he cons uc ion.
Howe e , i is s ill necessa y o adap some code con-
igu a ion pa ame e s in Odoo o each an in eg a ion
wi h he sys em. Ne e heless, gi en he ac ha a
g ea p opo ion o he code has been gene a ed au o-
ma ically he p oduc i i y is inc eased and he possi-
bili y o human ailu es has been educed.
The e i ica ion and alida ion o he sys em ha
is gene a ed is ca ied ou using he Ka alon ool, by
execu ing he sys em es plan ha gene a es NDT
om he de ined ones. I is impo an o no e ha he
in eg a ion es s o he au ogene a ed modules wi h
he exis ing ones ha e no been possible o be sel -
gene a ed wi h he use o hese me hodologies so a ,
despi e he impo ance hese kinds o modula de el-
opmen s.
Once G7Inno a ion deli e ed he alida ed and
e i ied sys em o he esponsible o he iMEDEA
p ojec , he in o ma ics eam o he clinic was he e-
sponsible o ca ying ou he phase o ope a ions.
Each one o he desc ibed phases is highly inde-
penden om he o he s. This allow ha ing wo k
eams ha can be wo king in pa allel.
4
CONCLUSIONS
The p oposal o his me hodology a ises as a need o
acili a e communica ion wi h cus ome s, and o al-
low a de elopmen ha is use - iendly wi hou im-
plying an inc ease in he cos s.
A e applying hese me hodologies, we ha e been
able o gene a e a p oduc close o he use and clien
expec a ions, au oma ing i s cons uc ion which e-
duce human ailu es and inc eases p oduc i i y. The
p o o ypes de ined using he Design Sp in me hodol-
ogy we e e y ealis ic as we knew abou he echnol-
ogy ea u es on which we wan ed o de elop he inal
p oduc and i s echnological limi a ions.
The use o hese me hodologies educed he need
o code modi ica ions in ad anced p ojec s ages be-
cause o he alida ion wi h he cus ome ’s needs du -
ing he in e iews on which hey could es he p o o-
ype. Also, human e o s a e educed since mos o he
code was au oma ically gene a ed. Howe e , he
au oma ed gene a ion equi es adap ing he code and
execu e mo e in ensi e in eg a ion es s o e i y he
compa ibili y o he gene a ed code wi h he al eady
deployed modules.
As esul , using his combina ion o me hodolo-
gies we p oduced a module in collabo a ion wi h he
s a up G7Inno a ion.
Finally, we plan o apply ou p oposal in o he
case s udies o di e en business a eas and scien i ic
a eas in collabo a ion wi h he esea ch eam o his
s a up. In pa icula , ou p oposal can be en iched
om a join wo k in a eas on which he esea che s o
G7Inno a ion RD eam has been wo king p e iously
as: (i) he design and de elopmen o echnological
solu ions in he a ea o equi emen s enginee ing
(Ga cía-Ga cía e al., 2012), (Escalona e al., 2013b),
(ii) he So wa e P ocess Managemen (Ga cia-Ga cia
e al., 2017), (Meidan e al., 2018), (iii) he PLM
sys ems (P oduc Li ecycle Managemen ) (Escalona
e al., 2013a) and (i ) he CPM (Clinical P ocess
Managemen ) (Ga cía Ga cía e al., 2015).
ACKNOWLEDGEMENTS
This esea ch has been pa ially suppo ed by he
POLOLAS p ojec (code TIN2016-76956-C3-2-R) o
he Spanish Minis y o Science and Inno a ion.
REFERENCES
A chi ec ., E. (2019). Tool. A ailable a : h ps://
spa xsys ems.com/p oduc s/ea/. Las accessed: July
2019.
Ban ield, R., Lomba do, C. T., and Wax, T. (2015). De-
sign sp in : A p ac ical guidebook o building g ea
digi al p oduc s. ” O’Reilly Media, Inc.”.
B ambilla, M., F a e nali, P., e al. (2014). The in e ac ion
low modeling language (i ml).
Delgado,
J.
and
Ma ín,
F.
(2000).
E olución
en
los
sis emas
de
ges ión
emp esa ial.
del
m p
al
e p.
Economía
in-
dus ial, 331(1):51–58.
Escalona, M., Mejías, M., and To es, J. (2004). De el-
oping sys ems wi h nd & nd - ool. In 13 h In e -
na ional con e ence on in o ma ion sys ems de elop-
men : me hods and ools, heo y and p ac ice, Vilna,
Li huania, pages 149–59.
Escalona, M. J., Ga cía-Ga cía, J. A., Mas, F., Oli a, M.,
and Del Valle, C. (2013a). Applying model-d i en
pa adigm: Calipsoneo expe ience. In CAiSE Indus- ial
T ack, pages 25–32. Ci esee .
Escalona, M. J., U bie a, M., Rossi, G., Ga cia-Ga cia, J.,
and Luna, E. R. (2013b). De ec ing web equi emen s
con lic s and inconsis encies unde a model- based
pe spec i e. Jou nal o Sys ems and So wa e,
86(12):3024–3038.
G7Inno a ion (2019). Technology-based company, based
in Se ille (Spain) and wi h a high le el o inno a ion in
he ield o de elopmen o ICT solu ions applied o
clinical and heal h managemen . A ailable a :
h p://g7inno a ion.com/. Las accessed: July 2019.
amewo k o imp o e he p ocess managemen : he
empowe p ojec . In 2017 IEEE 15 h In e na ional
Con e ence on Indus ial In o ma ics (INDIN), pages
553–558. IEEE.
Ga cía-Ga cía, J., Escalona, M., Ra el, E., Rossi, G., and
U bie a, M. (2012). Nd -me ge: a u u e ool o con-
cilia ing so wa e equi emen s in mde en i onmen s.
In P oceedings o he 14 h In e na ional Con e ence on
In o ma ion In eg a ion and Web-based Applica ions &
Se ices, pages 177–186. ACM.
Ga cía-Ga cía, J. A., Escalona, M. J., Domínguez-Mayo, F.
J., and Salido, A. (2014). Nd -sui e: a me hod- ological
ool solu ion in he model-d i en enginee ing pa adigm.
Jou nal o So wa e Enginee ing and Ap- plica ions,
7(04):206.
Ga cía Ga cía, J. A., Escalona, M. J., Ma ínez-Ga cía, A.,
Pa a,
C.,
and
Wojdyński,
T.
(2015).
Clinical
p ocess
managemen : a model-d i en & ool-based p oposal.
He ik, J. (2014). Wha Does IT Ope a ions Managemen
Do? (ITOps)). Joe He ik: Tech Machinis . A ailable
a : h ps://joehe ik.com/. Las accessed: July 2019.
Knapp J., Ze a sky J., K. B. (2018). Sp in : El mé odo pa a
esol e p oblemas y es a nue as ideas en solo 5 das.
España: Conec a.
Meidan, A., Ga cía-Ga cía, J. A., Ramos, I., and Escalona,
M. J. (2018). Measu ing so wa e p ocess: A sys-
ema ic mapping s udy. ACM Compu ing Su eys
(CSUR), 51(3):58.
Odoo (2019). Managemen so wa e. A ailable a : h ps://
www.odoo.com/. Las accessed: July 2019.
OMG (2008a). Documen s Associa ed wi h Me a Objec
Facili y (MOF). A ailable a : h ps://www.omg.o g/
mo /. Las accessed: July 2019.
OMG (2008b). MOF Model o Tex T ans o ma ion
Language (MOFM2T). A ailable a : h ps://www.omg.
o g/spec/MOFM2T/Abou - MOFM2T/. Las accessed:
July 2019.
OMG (2015). OMG Uni ied Modeling Language, 2.5.
A ailable a : h ps://www.omg.o g/spec/UML/. Las
accessed: July 2019.
Ga cia-Ga cia,
J.,
En íquez,
J.
G.,
Ga cia-Bo goñon,
L.,
A é alo,
C.,
and
Mo illo,
E.
(2017).
A
mde-based