Uniﬁed Quality Training Process

Hind Dahar and Ounsa Roudies

Mohammed V University in Rabat, EMI, Siweb Team E3S, Av. Ibn. Sina BP 765, Rabat-Agdal, Maroc, Hungary

Keywords:

Quality Standard, Quality References, Mapping, Harmonization, Co-deployment, MDE.

Abstract:

The development of quality in the global market has made standards and repository indispensable tools. Each

standard has its own strengths and covers speciﬁc perimeters in the company, which necessitates the co-

deployment of several standards simultaneously. However, the diversity of standards can lead to different

problems such as heterogeneity and redundancy, which results in increased implementation costs in terms of

time, resource mobilization and budget. We propose a quality approach called“Uniﬁed Quality” to address

these problems. This paper describes the conﬁguration process that creates The common repository, that en-

compasses all requirements of the co-deployed standards without redundancy and ambiguity. It is an iterative

and MDE based process. Finally, the training process is illustrated by its application to both ISO 9001 and

CMMI standards.

1 INTRODUCTION

Quality development is a key competition process in

the global market. This requires companies to use

standards and benchmarks of good practices in man-

agement and information systems. Several standards

cover the main activities of IT (information tech-

nology) systems such as ISO 9001 (Marques et al.,

2017), CMMI (noa, 2015) and ITIL (noa, 2012). Each

standard has its own strengths but does not satisfy the

collection of IT system requirements independently

of other standards (Jomaa, 2015). For example, ITIL

ensures the post-production quality, while CMMI im-

proves the maturity level of company’s software en-

gineering processes. However, ITIL or CMMI cannot

satisfy, by its self, the whole company needs.

The problem remains in the implementation of

several standards within the same company. Sev-

eral works have attempted to approach this sub-

ject. Some authors believe that this issue is due to

heterogeneity of standards that it is manifested in

their entities and their processes structures (Mu

noz

et al., 2014)(Kerzazi, 2015), It can also appear in

terms of terminology, size or complexity level (Pardo-

Calvache et al., 2015)(Dahar and Roudies, b). For

instance, heterogeneity causes difﬁculties for orga-

nizations in interpreting components of several stan-

dards (Pardo et al., 2015)(Pardo et al., 2012). while

for others, the issue is related to redundancies that

generate additional costs in terms of human resources,

time and budget (Baldassarre et al., 2012).

Many types of approaches have addressed the

challenges resulting from the co-deployment of sev-

eral quality standards, namely:

• Ontological approaches that present ontolo-

gies (Pardo et al., 2012) and Baldassar (Baldas-

sarre et al., 2012) to overcome the problem of ter-

minological difference in standards.

• Approaches based on model engineering and con-

struction of a general meta-model (Ponce et al.,

2014) which is based on the Meta models of stan-

dards.

• Approaches of integrating (Marques et al., 2017)

one model into another, or into a common frame-

work adopted by the organization.

Our literature review shows that these approaches are

speciﬁc and focus on managing two or three well-

deﬁned standards. We therefore conclude that re-

search ﬁeld is lacking of an approach not limited to

a little number of standards. In other words, we need

an approach that allows to align a set of undeﬁned and

unlimited standards.

Trying to remedy the inefﬁciency of the simulta-

neous use of several standards, we propose Uniﬁed

Quality approach. It is supported by the MDE (Model

Driven Engineering) approach. It aims to align sev-

eral standards in order to facilitate their integration

and to minimize the efforts related to their simulta-

neous usage in the IT ﬁeld. In addition, it optimizes

audit process.

Dahar, H. and Roudies, O.

Uniﬁed Quality Training Process.

DOI: 10.5220/0010147502410248

In Proceedings of the 12th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2020) - Volume 3: KMIS, pages 241-248

ISBN: 978-989-758-474-9

241

Our approach consists of two processes: Conﬁg-

uration and Audit. The approach’s conﬁguration pro-

cess aims to integrate several standards into the same

optimal quality repository. The standards are inte-

grated in an incremental way, which capitalizes on

the building of the quality repository and its train-

ing. The conﬁguration process is divided into two

sub-processes for this purpose:

• First, the building the pivot meta-model that

presents the structure of the quality repository.

• Second, the training process of this repository.

This repository allows to unify different structures of

the quality standards in order to solve their hetero-

geneity and redundancy issues. The second process

of Uniﬁed Quality approach is devoted to the quality

audit of a speciﬁc organization. First, the organiza-

tion chooses the collection of standards it wishes to

co-deploy. Then, Uniﬁed Quality system provides a

set of elementary requirements that the auditor has to

check. These are represent all the chosen standards in

a clear and not redundant manner. The audit process

will be addressed in subsequent study.

This paper is devoted to training process. Building

the pivot Meta Model has been described previously.

The objective of this paper is to describe how a new

adopted standard is integrated in the quality reposi-

tory and how the requirements are optimised.

To validate our approach, we apply it to ISO 9001

and CMMI to ensure the functioning of the proposed

models and transformations. We choose the ISO 9001

case that is already integrated in the quality repository

and consider the integration of the CMMI model.

The rest of the paper is organized as follows. The

second section is a overview of different aspects of

the article, the third section summarizes the construc-

tion process of the pivot meta-model, the fourth one

describes the training process of Uniﬁed Quality ap-

proach, followed by the ﬁfth section that describes the

experiment of the approach application. Lastly the

conclusion highlights the strengths of this approach,

some limits and leads to new perspectives.

2 UNIFIED QUALITY OVERVIEW

Uniﬁed Quality consists of two processes: Conﬁgura-

tion and Audit.

2.1 Conﬁguration Process

The conﬁguration process presents the backbone of

the UQ approach. Its main objective is to unify the

standards chosen by the company into a single reposi-

tory called the UQ repository, and this to facilitate and

optimize the simultaneous use of the standards within

the company and to eliminate the problems caused by

this issue. To this end, UQ repository contains all

the requirements of integrated standards. These re-

quirements are grouped by sector and are elementary

and unique. The conﬁguration process is subdivided

into two sub-processes, namely, building of the pivot

meta-model sub-process entitled UQ pivot building

and the training sub-process entitled TPUQ.

• UQ-Pivot Building Process: UQ Pivot meta-

model is a meta-model that supports communi-

cation between standards and the UQ repository.

It presents the structure of the UQ repository on

which the training process is based. The process

occurs only once in the life cycle of the system,

except in future need.

• Training Process: the role of this process is to

train the new standard in the UQ repository. It

is an iterative process. The number of iterations

equals the number of standards to be integrated.

It is ﬂexible and dynamic, depending on the stan-

dard.

2.2 Audit Process

The second process of Uniﬁed Quality approach is de-

voted to the quality audit of a speciﬁc organization.

First, the organization chooses collection of standards

it wishes to co-deploy. Then Uniﬁed Quality system

provides a set of elementary requirements that the au-

ditor has to check. These represent all the chosen

standards in a clear and not redundant manner. The

audit process will be addressed in subsequent work.

We propose in the following a detailed descrip-

tion of each sub-process composing the conﬁguration

process, to present on the one hand their phases, steps

and tasks and explain on the other hand their concepts

and theoretical foundations.

3 PIVOT META-MODEL

Our goal is to analyze the structures and concepts of

existing standards in order to build a common model.

It aims to align the structures of quality standards in

order to facilitate their communication and minimize

the efforts related to their simultaneous use. For this

purpose, we follow an MDE approach. We rely on

Meta modeling to deduce the ﬁnal structure of the

model. In our approach, we call the common model

UQ pivot model and its meta-model UQ pivot meta-

model.

KMIS 2020 - 12th International Conference on Knowledge Management and Information Systems

242

This section describes the UQ pivot meta-model.

The details of the process of this meta-model building

are given in a previous article(Dahar et al., )(Dahar

and Roudies, a).

Figure 1 presents the UQ pivot meta-model. The

UQ pivot meta-model consists of the Standard class

. Each standard is composed of a set of rubrics. It

allows the standard to be subdivided into several areas

or specialties. In addition, the requirements underline

the need to be met in order to obtain certiﬁcation of a

standard.

Figure 1: Pivot meta-model.

• Requirement: After analysis of the concepts of the

standards leaders, we noticed that in the standards,

the main concept is Requirement for ISO9001 and

Practice for CMMI. These two concepts have the

same intention at the level of organizations, it

presents the action or need to be satisﬁed in order

to obtain the level of quality required.

• Rubric: In addition, the other classes of the two

standards generally allow the grouping of these

requirements according to one or several axis.

Therefore, there are two types of rubrics, elemen-

tary and compound rubrics.

• Compound Rubric: The grouping contains groups

of requirements in order to have a well-structured

hierarchy, for example Composite item for ISO

9001 and Process area for CMMI.

• Elementary Rubric: The grouping of require-

ments is directly linked to the requirements, for

example objective for CMMI and elementary arti-

cle for ISO 9001.

4 TRAINING PROCESS

UQ Training is an agile process. It allows to unify

standards in a UQ repository. This repository elim-

inates misunderstanding and redundancy in the re-

quirements of standards, and inconsistency in their

structures. The process integrates the standards into

the UQ repository incrementally. It consists of three

phases: the loading phase of the new standard, the

reﬁnement phase and the integration phase. It relies

on the quality engineer who is a required analyst, and

his main role is to make sure that the products com-

ply with the company’s requirements, and to manage

the balance between productivity and quality. In this

process, he is in charge of judging the elementary and

reﬁning. The ﬁgure shows the steps and tasks of the

training process as well as their input and output mod-

els. Figure2

Figure 2: UQ training process.

4.1 Step 1. Loading a New Standard

The standard loading step consists of studying each

standard separately. It is based on model engineering.

Its goal is to load the new standard into the UQ pivot

model. As explained above, the main actor is the soft-

ware engineer. This step consists of two tasks. The

ﬁrst one is the loading of the standard original model

and the second one is its transformation into UQ pivot

model.

4.1.1 Task 1. Creation of a Standard Original

Model

It consists of representing the new standard by a

model based on the standards concept. For exam-

ple CMMI is represented in terms of processes areas,

goals and practices. The result is the standard orig-

inal model. This standard original model is related

only to the standard, so it is the same model regard-

less of the enterprise. Thus, it is maybe reused. This

step is usually carried out by the quality engineer of

the enterprise.

4.1.2 Task 2. Updating UQ Pivot Model

The original model of each standard is transform in

order to express in the UQ pivot Model. UQ pivot

model encompasses the representation of each se-

lected standard expressed by common language. For

example CMMI is expressed in the pivot model in

terms of requirements and rubrics as shown in ﬁgure1.

This perform by deﬁning transformation rules.

Uniﬁed Quality Training Process

243

Our method uses a (1 to 1) transformation (Mens

et al., ). The transformation system used is said to be

exogenous because the source and target models do

not come from the same metal-model. This system

is also called horizontal because the source and target

models involved in the transformation are at the same

level of abstraction. The transformation mechanism

used to load the UQ pivot model is schematized in

Figure 3.

Figure 3: Transformation from standard to Pivot repository.

4.2 Step 2. Reﬁnement of a Standard

Before analyzing the correspondence between re-

quirements of the UQ pivot model and those of the

UQ repository, it is necessary to ensure that they are

placed at the same level of granularity. In this phase,

we use an analytical approach based on requirements

engineering. The requirements of the quality stan-

dards are functional. They deﬁne one or more func-

tionalities to be applied. The proposed solution to

unify the level of granularity is the function-oriented

reﬁnement. The goal is to rewrite a requirement into

several reﬁned requirements. A reﬁned requirement

is elementary, associated with a single functionality.

The reﬁnement is done by the quality engineer. The

latter is in charge of analyzing requirements, deﬁning

their functionalities and rewriting them into elemen-

tary requirements.

Given

RP : The set of UQ pivot model requirements.

RR : The set of reﬁned requirements.

r ∈ RP and r

∈ RR.

We express this reﬁnement by the function named

”F” presented by two equations; Equation 1 is deﬁned

from RP to RR

. Indeed, if r is an elementary require-

ment then it is assumed to be already reﬁned. There-

fore the image of r is the same. Otherwise, we decom-

pose r into several elementary requirements r

. The

union of r

is equivalent to r, as illustrated in Equa-

tion 2.

F(r) =

(

r, r is elementary

, r

, , r

), otherwise

(1)

∪

i=1

⇐⇒ r (2)

4.3 Step 3. Construction of the UQ

Repository

This step is composed of two main tasks.

4.3.1 Task 1. Requirement Mapping

In this phase, we create mapping links between the

reﬁned requirements and the UQ repository require-

ments. This is done by the company’s quality en-

gineer. Analyzing each repository requirement with

each UQ pivot model requirement is a long and cum-

bersome process. We deﬁne two levels of mapping,

namely rubrics and requirements. At the ﬁrst level,

we analyze the mapping between the rubrics of the

UQ pivot model and those of UQ repository. This

allows us to identify the rubrics that cover similar ar-

eas or that may have points of intersection and also

those that cover completely separate areas. This level

allows us to detect all requirements that will be ana-

lyzed at the second level. It is obvious that there is no

mapping link between requirements that belong to di-

vergent rubrics, and only those that belong to aligned

rubrics will be analyzed.

The function RM represented the requirements

mapping takes the following values:

• 1 in the case where there is a mapping,

• 0 otherwise.

The function RQM The function RM represented the

rubrics mapping takes the following values:

• 1 in the case where there is a mapping,

• 0 otherwise.

Many Natural Language Processing (NLP) applica-

tions allow to compute the similarity in meaning be-

tween two short texts. There is a wide range of meth-

ods for calculating the similarity in meaning between

two sentences like Pre-trained sentence encoder(Cer

et al., ) and Smooth Inverse Frequency(Karipbayeva

et al., ) or more advanced algorithms like Siamese

Manhattan(Amin et al., ).

These methods can be used to automate the com-

parison of two requirements and to estimate the error

distance at the equivalent level. The quality engineer

will intervene only if the error distance is low in or-

der to conﬁrm the equivalence between the two com-

pared requirements. The comparison of the efﬁciency

of each similarity algorithm and the implementation

KMIS 2020 - 12th International Conference on Knowledge Management and Information Systems

244

of the most relevant one remains to be studied as part

of my future work.

4.3.2 Task 2. Merging the Standard and the UQ

Repository

This composition makes it possible to merge the re-

quirements of two models by taking into account

the mapping links established between these require-

ments. And this for ensuring that these requirements

are not duplicated in the new UQ repository. In addi-

tion to the ﬁnal model of the UQ repository.

The requirements of the source models are divided

into two main groups. The ﬁrst group presents those

that have a mapping in the model opposite, while the

second group presents those that do not. Indeed, there

are two activities that produce elements in the target

model which is the UQ repository. The elements that

have been identiﬁed as mapping are merged into a se-

quence of model elements in the target model and a

selection of the elements for which a mapping has not

been found in the opposite model is transformed into

elements of the target model. Each merge rule deﬁnes

the types of elements it can merge, as well as a list

of elements it produces in the target model. It also

deﬁnes exactly how the source elements are linked to

the newly created elements in the target model. Sim-

ilarly, each transformation rule deﬁnes the instances

it can transform, and a list of the model elements it

produces in the model.

5 EXPERIMENT

Let us consider SoftChallenge, a software company

producing logistic systems. At 2017 it has chosen to

be certiﬁed ISO 9001, in order to improve its busi-

ness processes. Therefore the quality team needs a

check list of the ISO 9001 requirements. The qual-

ity engineer has integrated ISO 9001 requirements

in SoftChallenge UQ repository. At 2020, SoftChal-

lenge decided to improve its software engineering

activities by the CMMI standard certiﬁcation. SO

CMMI requirements must be integrated in SoftChal-

lenge UQ repository.

We assume that the ISO 9001 standard is the ﬁrst

quality standard deployed in the company, and that

our main objective is the deployment of the CMMI

standard while pursuing our UQ approach. In our

case, we will follow a single iteration of CMMI in-

tegration.

We are currently implementing a framework that

helps the quality engineer in training the new stan-

dards in the UQ repository. In this article we will use

the framework in the application of the ﬁrst part of

the process, the second and third part will be applied

manually, and the ﬁnal version of the framework will

be described in details in a future work.

In view of the volume of ISO 9001 and CMMI

standards, we chose just the two-part of both stan-

dards. We have chosen the two most convergent parts

in order to visualize the approach and its beneﬁts.

We have chosen Article 4 Quality management

system of the ISO 9001 standard[REF]. And the

generic goals of the CMMI standard[REF]. !

5.1 Step 1. Loading the CMMI Original

Model

In this task, we use the UQ framework in order to

instantiate the CMMI standard, the ﬁgure 4 presents

the generic goals and practices of the CMMI standard.

Figure 4: Generic goals and practices of the CMMI stan-

dard.

The transformation rules is illustrated in the table1

The result of the transformation gives us the pivot

model enriched by the instances of the CMMI stan-

dard.

5.2 Step 2: CMMI Reﬁnement in UQ

Pivot Model

We elaborate the reﬁnement of the requirements of

the CMMI UQ Pivot model by illustrating an exam-

ple of a requirement, we have chosen the GP2.5 re-

quirement, which is: GP2.5 Identify and involve the

parties concerned. As shown in Figure 6, this require-

ment has two features, the identiﬁcation of the parties

concerned and their involving.

F(GP2.5) = (GP2.51, GP2.52) (3)

Uniﬁed Quality Training Process

245

Table 1: The transformation rules from CMMI to the UQ

Pivot.

Source target Attribute

Standard - id-s: CMMI

- label-s: Capability

Maturity Model In-

tegration

- validity-s:

Maturity

Level

Compound

rubric

- id-r: id-m

- label-r: label-m

- type-r: maturity

level

Process

Area

Compound

rubric

- id-r: id-pa

- label-r : label-pa

- type-r: process

area

Goal Elementary

rubric

- id-r: id-g

- label-r: label-g

- type-r: type-g

Practice Requirement - id-rp: id-p

- label-rp: label-p

- type-rp: type-p

- validity-rp : vali-

dity-p

Figure 5: Result of the transformation from CMMI model

to PIVOT model.

5.3 Step 3: Updating UQ Repository

The experimentation of this part is done in two parts:

5.3.1 Task 1. Requirement Mapping

In this part we will elaborate the mapping between

the requirements of the UQ repository and those of

the UQ pivot Model. As explained in the require-

ments mapping paragraph, this task is done in two

steps, rubric mapping and requirements mapping.

As a reminder, the example chosen ISO 9001

Figure 6: Example of a reﬁnement of a requirement.

rubrics are

• 4.4 Quality management system and its processes.

And the example chosen CMMI rubrics are:

• GG1 Achieving speciﬁc objectives

• GG2 Institutionalize a managed process Plan the

process

• GG3 Institutionalize a deﬁned process Establish a

deﬁned process

The result of the RQM function between the rubric

4.4 and the GG1 rubric is RQM(4.4, GG1) ) = 0, but

for the example of the two rubrics 4.4 and GG2 the

RQM (4.4, GG2) equals to 1.

We conclude from the rubric mapping result that,

it is not necessary to process the mapping between

the requirements of rubric 4.4 and GG1 because there

is no reconciliation between the two rubrics, and in

this case the QM mapping function between these re-

quirements equals to 0. for each requirement of 4.4

and r requirement of GG1 MQ(r,x) = 0. Which gives

us the beneﬁt of analyzing the requirements of these

two rubrics.

On the other hand it is necessary to analyze the

mapping between the requirements of the rubric com-

binations (4.4 GG2) and (4.4, GG3). To illustrate,

using the example of (4.4 GG.2); The result of the

requirements mapping of the rubric 4.4 and GG2 is

shown in table2. The case of requirement GP2.1 (Pro-

vide resources) and the requirement d1(determine the

resources needed for these processes), these two re-

quirements cover the same functionality, which is the

determination of the resources for the processes. Then

there is a mapping between the two requirements and

the function RQM(d1, GP2.1) = 1. On the other

hand, the example of requirement pair GP2.3 (Edu-

cating people) and g1 (Evaluate these processes), il-

lustrates the case of absence of mapping and the func-

tion RQM(g1,GP2.3) = 0.

KMIS 2020 - 12th International Conference on Knowledge Management and Information Systems

246

Table 2: Result of the mapping of the requirements of rubric 4.4 and GG2.

RQM a b c d1 d2 e f g1 g2 h

GP2.1 0 0 0 1 0 0 0 0 0 0

GP2.2 0 0 0 0 0 1 0 0 0 0

GP2.31 0 0 0 0 0 0 0 0 0 0

GP2.32 0 0 0 0 0 0 0 0 0 0

GP2.4 0 0 0 0 0 0 0 0 0 0

GP2.5 0 0 0 0 0 0 0 0 0 0

GP2.6 0 0 0 0 0 0 0 1 0 0

GP2.7 0 0 0 0 0 0 0 0 0 0

5.3.2 Task 2. Integration of the UQ Pivot Model

to the UQ Repository

We proceed in this task to integrate the pivot UQ

model in the UQ repository.

The pivot model requirements that have a mapping

of the requirements from the UQ repository will not

be integrated into the UQ repository but all the others

will be integrated, in the table3 shows the result of the

example integration of the GG.2 ﬁeld.

6 CONCLUSION

Even if, software quality assurance approaches have

reached a maturity and their standards are popular,

their deployment is mastered when only one of them

is deployed. But in the case of the co-deployment of

several quality standards, many problems remain and

are the subject of extensive research. The originality

of UQ approach is ﬁrst to provide an gradual and ag-

ile integration of standards according to the company

needs and vision. Secondly the approach is based on

formalisation and MDE. The quality engineer have a

unique quality check list to drive his audit, avoiding

redundancy and ambiguity.

The approach consists of two processes, conﬁgu-

ration and audit. The approach’s conﬁguration pro-

cess aims to integrate several standards into the same

optimal UQ repository. The standards are integrated

in an incremental way capitalizes on the building of

the UQ repository and its training, It is divided into

two sub-processes for this purpose. The ﬁrst one is

the building of the pivot meta-model that presents the

structure of the UQ repository and second one is the

training process of this repository which is the sub-

ject of this paper. This repository allows to unify the

different structures of the quality standards in order to

solve their heterogeneity and redundancy issue. The

second process of Uniﬁed Quality approach is de-

voted to the quality audit of a speciﬁc organization.

. The audit process will be addressed in subsequent

Table 3: The UQ repository containing article 4 of the ISO

9001 standard.

Rubric RequirementP

Quality manage-

ment system and

its processes

a) determine the inputs re-

quired and the outputs ex-

pected from these processes;

b) determine the sequence

and interaction of these pro-

cesses;

c1) Determine the criteria and

methods ...

c2) Apply the criteria and

methods.

d1) determine the resources

needed for these processes

d2) ensure their availability;

e) assign the responsibilities

and authorities for these pro-

cesses;

f) address the risks and op-

portunities ...

g1) valuate these processes

g2) implement any changes

needed...

h) improve the processes and

the QMS.

GG 2 Institution-

alize a managed

process Plan the

process

GP2.3) Educating people

GP2.4) Work control prod-

ucts

GP2.51) Identify the parties

concerned

GP2.52) involve the parties

concerned

GP2.7)Review Status with

Higher Level Management

work. The models of the UQ approach are reusable

and independent of the company. We are in the pro-

cess of implementing a ﬁnal application that facili-

tates the application of the UQ approach as well as

we are still working on the audit process that will be

the subject of the next article(Kitchenham, 2013).

Uniﬁed Quality Training Process

247

REFERENCES

(2012). ITIL V3 Foundation Exam.

(2015). CMMI for Development, Version 1.3 CMMI-DEV,

V1.3.

Amin, K., Lancaster, G., Kapetanakis, S., Althoff, K.-D.,

Dengel, A., and Petridis, M. Advanced similarity

measures using word embeddings and siamese net-

works in CBR. In Bi, Y., Bhatia, R., and Kapoor, S.,

editors, Intelligent Systems and Applications, pages

449–462. Springer International Publishing.

Baldassarre, M. T., Caivano, D., Pino, F. J., Piattini, M.,

and Visaggio, G. (2012). Harmonization of ISO/IEC

9001:2000 and CMMI-DEV: from a theoretical com-

parison to a real case application. Software Quality

Journal, 20(2):309–335.

Cer, D., Yang, Y., Kong, S.-y., Hua, N., Limtiaco, N., John,

R. S., Constant, N., Guajardo-Cespedes, M., Yuan, S.,

Tar, C., Sung, Y.-H., Strope, B., and Kurzweil, R. Uni-

versal sentence encoder.

Dahar, H. and Roudies, O. Mapping of quality standards.

8(11):2406–2412.

Dahar, H. and Roudies, O. Measurement of co-deployment

of IT quality standard: Application to ISO9001,

CMMI and ITIL. In Rocha, l. and Serrhini, M., ed-

itors, Information Systems and Technologies to Sup-

port Learning, pages 451–459. Springer International

Publishing.

Dahar, H., Roudies, O., and Younoussi, S. Process of build-

ing of the pivot meta-model. pages 139–143. ACM.

Jomaa, H. (2015). Contribution de l’usage des syst

emes

d’information

a la performance des organisations.

page 239.

Karipbayeva, A., Sorokina, A., and Assylbekov, Z. A Cri-

tique of the Smooth Inverse Frequency Sentence Em-

beddings.

Kerzazi, N. (2015). Conceptual alignment between SPEM-

based processes and CMMI. In 2015 10th Interna-

tional Conference on Intelligent Systems: Theories

and Applications (SITA), pages 1–9, Rabat. IEEE.

Kitchenham, B. (2013). Guidelines for performing Sys-

tematic Literature Reviews in Software Engineering.

page 44.

Marques, P. A., Meyrelles, P. M., Saraiva, P. M., and

Frazao-Guerreiro, F. J. (2017). Integrating Lean Six

Sigma with ISO 9001:2015. In 2016 IEEE Interna-

tional Conference on Industrial Engineering and En-

gineering Management (IEEM), pages 894–898, Bali,

Indonesia. IEEE.

Mens, T., Gorp, P. V., Varr

o, D., and Karsai, G. Applying a

model transformation taxonomy to graph transforma-

tion technology. 152:143 – 159.

noz, M., Mejia, J., and Gasca-Hurtado, G. P. (2014).

A Methodology for Establishing Multi-Model Envi-

ronments in Order to Improve Organizational Soft-

ware Processes. International Journal of Software En-

gineering and Knowledge Engineering, 24(06):909–

933.

Pardo, C., Pino, F. J., Garcia, F., Baldassarre, M. T., and

Piattini, M. (2015). From chaos to the systematic har-

monization of multiple reference models: A harmo-

nization framework applied in two case studies. Jour-

nal of Systems and Software, 86(1):125–143.

Pardo, C., Pino, F. J., Garc

ıa, F., Piattini, M., and Bal-

dassarre, M. T. (2012). An ontology for the harmo-

nization of multiple standards and models. Computer

Standards & Interfaces, 34(1):48–59.

Pardo-Calvache, Universidad del Cauca, C. J., Garc

ıa-

Rubio, F. O., Universidad del Cauca, Piattini-Velthuis,

M. G., Universidad del Cauca, Pino-Correa, F. J., Uni-

versidad del Cauca, Baldassarre, M. T., and Univer-

sidad del Cauca (2015). A 360-degree process im-

provement approach based on multiple models. Re-

vista Facultad de Ingenier

ıa Universidad de Antio-

quia, (77).

Ponce, J., Garc

ıa-Borgo

non, L., Garc

ıa-Garc

ıa, J.,

Escalona, M. J., Dominguez-Mayo, F. J., Alba, M.,

and Aragon, G. (2014). A Model-Driven Approach

for Business Process Management. 1(2):22.

KMIS 2020 - 12th International Conference on Knowledge Management and Information Systems

248