A Framework for Generating Domain-specific Rule for Process
Model Customisation
Neel Mani
1
, Markus Helfert
1
and Claus Pahl
2
1
ADAPT Centre for Digital Content Technology, Dublin City University, School of Computing, Dublin, Ireland
2
Free University of Bozen-Bolzano, Free University of Bozen-Bolzano, Bolzano, Italy
Keywords: Domain Model, Domain-Specific Rule, Rule Language, Rule Generation, Business Process Model,
Variability Model, Model Driven Architecture, Model Transformation, Domain-Specific Language.
Abstract: The domain-specific model-driven development requires effective and flexible techniques for implementing
domain-specific rule generators. In this paper, we present a framework for rule generation through model
translation with feature model, a high-level of the domain model to translate into low-level of rule language
based on the paradigm of software reuse in terms of customisation and configuration with domain-specific
rule strategies benefit mode-to-text translations. This framework is domain-specific where non-technical
domain user can customise and configure the business process models. These compositions support two
dimensional of translation modularity by using software product line engineering. The domain engineering is
achieved by designing the domain and process model as a requirement space, it is also called template model,
connecting with feature model through weaving model. The feature model is a high-level input model to
customise the template model to an implementation. The application engineering is achieved by supporting
the rule definition and configuring the generated rules. We discuss the development approach of the
framework in a domain-specific environment; we present a case study in a Digital Content Technology (DCT)
domain.
1 INTRODUCTION
Nowadays the reuse of software is one of challenges
to implement to customise the application based on
the end-user requirements. The traditionally model-
driven engineering use for extracting the knowledge
from a high-level design model to low-level langua-
ge. How to effectively fill the gap between the softw-
are product line engineering and the models to extract
the knowledge from high-level models to low-level
language. The primary requirement of the framework
where end user can opt their requirement to adapt the
desirable models. For achieving the desirable model,
there are several steps to require the implementation
of configuration-based systems where translate the
high-level design to low level execution. Enterprises,
usually have high level legacy models and they are
domain and process based models. Automatic code
generation (Edwards, Brun et al. 2012, Prout, Atlee et
al. 2012, Ringert, Roth et al. 2015) is a well-known
approach for getting the execution code of a system
from a given abstract model. Rule is an extended
version of code that requires compiling and building
it in a configurable mode. Rule generation is an
approach by which the higher-level design model as
input and the lower level of execution code as output
are shaped. It may be platform independent or
platform specific (Bergmayr and Wimmer 2013)
approach. In Model-Driven Architecture (MDA), the
techniques are expressed by design models as the
primary artefact of development and use them as a
basis for obtaining a configurable domain-specific
rule for business process model customisation in
different ways (Gonçalves 2015). But it does not talk
about the variability of models (domain model and
process model).
The major motivations is to address the automated
customise the models based on the end-user and then
generate the domain-specific rule(DSR) (Mani,
Helfert et al. 2016) from high-level complex domain
models at run time. The domain models are
characterised by the complexity of their structures as
specified in the metamodel or the non-trivial nature
of constraints imposed on them or a combination of
these two factors. It is a problem by which the
existing fully-automated model generators fail.
Mani N., Helfert M. and Pahl C.
A Framework for Generating Domain-specific Rule for Process Model Customisation.
DOI: 10.5220/0006512201630171
In Proceedings of the International Conference on Computer-Human Interaction Research and Applications (CHIRA 2017), pages 163-171
ISBN: 978-989-758-267-7
Copyright
c
2017 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved
Our contribution is a framework for customising
and configuring process model in digital content
technology domain in dynamic environment. The
production capability is based on reusable model as
core assets that configure the constraints of the
domain problems by non-technical domain user. At
runtime, we propose the model-based rule generation
for process models that use the variability models.
They capture the user requirement to process
automatically the rule generation and configuration of
the process model’s composition. Mass customisation
updates are possible through software Product Line
Engineering (SPLE), which provides a platform to
domain expert for designing and developing the
domain template (domain model and process model).
The paper is organised as follows. We compare
the related work Section 2. Section 3 is the motivation
of research paper. In Section 4, we discuss the overall
proposed approach, in which we discuss DSLs,
feature model, domain model and the rule. Section 5
introduces and describe the design of the framework
and its component and In Section 6, we evaluate the
overall approach in terms of usability The Section 7
concludes the paper reflecting the future work.
2 RELATED WORKED
In this section, we review the present concept,
technologies and techniques that motivate us to
present the relevant problem under domain constraint
study. It includes a short description of automatic
code generation, language and applications based on
Model-driven principles.
Our research work uses the SPLE life cycle for
customising the feature based on the user requirement
and MDA framework that uses in generating the do-
main-specific rule of the customised domain model.
The domain expert performing on the high level
of abstraction, the machine works low level of
language (Gupta 2015). The code generation
technique accelerates the transformation process
between designing the application and its implement-
tation in terms of executable code (Prout, Atlee et al.
2012, Gurunule and Nashipudimath 2015). The code
generation is an approach for converting the high
level of design constructs into low level of executable
code constructs. The automatic code generation is
converting software design into executable code
without little bit intervention of programmer or
developer (Gurunule and Nashipudimath 2015). The
generation technique reduces the effort of the manual
writing of programming code, avoiding the manual
programming errors like syntactical (spelling
mistake) and programmatically semantical errors.
The web application development is description
of processes, techniques and selecting appropriate
models for web engineering. The web engineering
uses automatic web application methodologies such
as UWE (Koch, Knapp et al. 2008), WebML (Ceri,
Fraternali et al. 2000) and WebDSL (Groenewegen,
Hemel et al. 2008) approaches. The design and
development of Web applications and tools provide
mainly conceptual models (Ceri, Fraternali et al.
2002), focusing on content, navigation and presenta-
tion models (Linaje, Preciado et al. 2007, Moreno,
Meliá et al. 2008). Now , the model driven approach
for dynamic web application, based on MVC and
server is described by Distante et al. (Distante,
Pedone et al. 2007). However, these methods do not
consider as the user requirement of the variability mo-
del. To simplify our description, we have considered
the user requirement and according to the need of the
user, the user selects the feature and customises the
enterprise application in the dynamic environment.
Currently, there is no such type of methodology
or process of development for creating a rule-based
system in a web application (semantic). Dioufet et al.
(Diouf, Maabout et al. 2007, Musumbu, Diouf et al.
2010) propose a process which merges UML models
and OWL ontologies for business rule generation.
The solution for semantic web is ontologies, UML
and applying the MDA approach for generating or
extraction rules from high level of models. Although,
the proposed combination of UML and semantic
ontologies are for extracting the set of rules in target
rule engine, but they only generate the first level of
the abstraction of the rules.
We have proposed a classification of several
quality and governance constraints elsewhere (Pahl
and Mani 2014): authorisation, accountability,
workflow governance and quality. The domain-
specific rule language (DSRL) (Mani and Pahl 2015)
is a combination of rules and BPMN. Our approach
provides a framework architecture for generating the
domain-specific rule and configuring the generated
rule for process model customisation using the
variability model.
3 MOTIVATIONS
The enterprise needs to change and adapt
progressively the dynamic behaviour at runtime in
response to changing conditions, updating new
features in support of enterprise and business
applications and in the surrounding business process
models. Nowadays, the requirements for developing
Figure 1: Framework for SPLE: domain and application engineering, problem and solution space (Mani, Helfert et al. 2017).
and customisation process are increasing. However,
the process model languages (Standard , Reichert and
Dadam 1998, White 2004, Van der Aalst and Ter
Hofstede 2005, List and Korherr 2006) restrict
domain user and designer to describe explicitly the
process execution plan as pre-defined task control
flow, data flow and work/process allocation schema.
The changes reflect at modelling stage or design
phase which make the process model rigid
(Boukhebouze, Amghar et al. 2011, Gromoff,
Kazantsev et al. 2012, Rangiha and Karakostas 2013).
Thus, adaptability is emerging as crucial and an
underlying capability in highly dynamic
environment.
During business process customisation and
adaptation in dynamic and domain-specific environ-
ments, the end users have only domain knowledge
without technically customising and adapting the
changes and configuration is done in process models
at runtime. For example, process activities and sub
process could be activated and deactivated (removed)
(Mani, Helfert et al. 2016) based on user requirement
at application engineering level.
4 SOFTWARE PRODUCT LINE
ENGINEERING
The key contribution is the use of a feature model to
bridge the gap between an assumed domain model
(here in ontology form) and the domain-specific rule
extension of a business process. The feature model
streamlines the constraints customisation for business
processes. It acts as a bridge between the domain
model and the rule language.
We propose to abstractly support the different
SPLE aspects from MDA. The MDA aims at
capturing all the relevant aspects of the framework
through appropriate models. The stakeholders’
motives are more prominently captured by models
than the implementation codes. Models capture the
requirements or the intentions of the end users more
effectively, help to avoiding accidental implementa-
tion details and also are more suited for analysis.
Models, in MDA context, are much more than being
supportive artifacts; rather, these are actually source
artifacts which can be utilized for automated analysis
and/or rule generation.
In our approach, the aim is the solve the need for
expressive and easy-to-understand adaptation.
Therefore, the domain model is implemented as a
variability model which describes the variants in
which the domain expert designs the domain template
composition. Since the domain model abstracts the
rule generation, the definition of a bridge between the
elements in the variability model and the elements in
the domain model could be used to support the
dynamic rule generation in the underlying domain-
specific environment.
To this end, we use a weaving model as an
additional software asset input to project the changes
in the features of the variability model, on abstract
elements in a domain model. In other words, the
weaving model works as a bridge between the
elements in these models.
The goal of Software Product Lines Engineering
(SPLE) is developing a set of software components
and systems with similar characteristics and catering
to the requirements of a domain through management
of certain features (Kang, Cohen et al. 1990). SPLE
effectively tunes down the development cost and
Table 1: Weaving model.
Variablity Model Feature Model Process Model Domain Model
1. V1 F1 Task1 Class1
2. V2 F2 Task2 Class 2
3. V3 F3 Task3 Class 3
Figure 2: Models that support the SPL for autonomic rule generation.
market time of the software, enhances the overhead
quality and engineering by reusing assets strategically
within the domain. SPLE uses adopted techniques to
manage reusability with commonality and variability
model that effectively categorizes the common assets
and their variabilities. The software product line
framework has two phases (see Figure 2): (i) the
Problem Space for describing the problem
description, the type of applications, or an individual
application in the category; (ii) Solution Space for
providing the software components to solve that
problem; iii) Domain engineering phase may be
defined as a formally represented platform in which
development and implementation products take
place. In SPLE, the variability modeling technique
known as “Feature Models” are used for the purpose
of portraying variability in hierarchical manner
differentiation or simplification of features in
hierarchy of products belonging to a software family.
The DSRs are products, based on activate or
deactivate features in the feature model to manage the
requirement of domain user or stack holder during
feature selection. Therefore DSRs depend upon the
feature of the feature model. The rule generation in
the variability model produces the adaptation space
with 1) and all possible constraint configurations of
the process model (in terms of active and inactive
features in the feature model with parametric value)
and 2), customising the process model in terms of the
functional and operational use. In order to avoid the
problem or an interruption (error, system halt,
malfunctioning, wrong interpretation, etc.) during
rule generation and or configuration in critical service
application, we argue that the feature model and its
possible configurations should be validated and
verified at the runtime.
The SPLE and MDA are not only complementary,
but their integration may lead to significant gains in
various applications. On one side MDA provides for
abstractly representing various aspects of a product
line, while on the other SPLE provides for a well-
defined application scope. This provides a sound
basis for the development and selection of appropriate
modeling languages. Further, the automated genera-
tion of system configurations is made possible by
accurate models as a result of automated analysis and
rule. MDA provides effective techniques for
conveying the results of specifying variability as
follows:
Metamodeling: It refers to type of systems that
express, for specific domains, having the
constraints that are associated with a product
line, with key abstract syntax characteristics and
static semantic constraints
Domain-specific languages (DSLs): In order to
formalize the specifics of structure of the
product line, its behavior and requirements with
respect to domain, the DSLs provide notations
governed by extendable metamodels.
Model transformations and rule generators: It
refers to ensuring the consistency of
implementations of the product line along with
the corresponding analysis. The analysis may be
retrospect to functional and QoS requirements.
Key advantages of using MDA in conjunction
with variability of SPLs are (1) rigorously capturing
the commonalities and variabilities in a family of
systems and (2) helping automated repetitive tasks
that must be accomplished for each product instance.
Figure 2 shows how to combine modelling and
model transformations to develop an SPLE-MDA for
Digital Content Technology. First, the assets of the
SPLE-MDA model elements are described with a
family of Digital Content Technology. These model
elements are conformed to the metamodel (of Domain
Model), which is a DSL for Digital Content
Technology. Second, the decision model is another
model which specifies the aspects or characteristics
(named features) of a particular Digital Content
Technology. Third, a weaving model projects the
features on the DSL for scoping the domain model.
Finally, the output system is obtained through a
model (scoped DSRL) to text (domain-specific rule)
the translation.
5 DESIGNS OF FRAMEWORK
The dynamic process model adaptation is possible by
customising and configuring the overall architecture
of the framework at runtime through predefine
domain template (domain model and process model)
models. This approach works on principal of active-
tion and deactivation of models. However, the
proposed framework follows strategy for the dynamic
activation and deactivation of features in the domain
template, based on the end-user requirement. After,
the domain template is modeled and the weaving
model has connected the feature model and the
domain template (via a virtual relational table) at
design time, the customisation and adaptation may
take place at run time. The customisation of the
process model and rule generation achieve at runtime.
This approach works under the domain-specific
environment.
An overview of our approach, enabling
customisation and configuration of the component
connections for the rule is shown in Figure 3. As
illustrated, the approach covers both dynamic process
model adaptation and development compositions.
The aim is to design appropriate architecture modes
of the feature selection based on user requirements
and generate the configurable rule for process model
customisation. At run time, the users configure the
Figure 3: Detailed Framework of Rule Generation and Process Model Customisation.
Figure 4: Feature model of Machine translation.
rule, thereby facilitating the framework to quality
configuration in terms of effective, efficient and
satisfactory usability. The feature model
requirements are continually analysed and validated
every feature by target feature validator with respect
to the runtime feature section. It reflects in the
template model environment as an activation and
deactivation through weaving model. In the case of
the feature selection violation, the target feature
validator gets prompted and validated accordingly on
basic criteria of the feature model. For example, the
user can remove the mandatory feature from the
system. In the following, we briefly discuss each
phase and describe their actives.
The framework offers a dynamic solution for rule
generation and process model customisation. The
framework consists of two sections: Development
Composition and Dynamic Process Adaptation. Dev-
elopment composition consist of different elements
such Weaving Model, Process model, metamodel,
DSRL configuration generator. We have developed
two building blocks to provide utilisation of the
domain model variability and community at run time:
(1) the domain model customiser uses a weaving
model (table 1). The weaving models are models
which capture different kind of relationships between
models (Del Fabro and Valduriez 2007). It is about
finding a similar type of transformation patterns
between model elements integrating transformations
in the tabular form.
The weaving helps to execute the customisation
process base on validate feature. The validate feature
is a set of analysed and verified features, which is
captured by the requirements of a domain user or
stakeholder. (2) In the Rule generator, the models and
metamodel that created in the domain model
customiser are used in the MDA at different level M1
and M2 (Figure 3). The generated rule configuration
has two different types: per-configured (at design
time) and post-configurable (dynamic) metamodel
In the Dynamic Process Adaptation, the proposed
structure carries out the following steps to support
dynamics adaptation. First, the model-based
configurator collects the information from various
models. If the requirement of target feature is
violated, the feature selection or customisation
activities such as: rename, move, update and delete
feature, affect the activation and deactivation process.
The execute customisation models pass the captured
requirement to weaving model for further activities.
The customised and generated sets of rules are output
from the Development Composition.
We propose an approach that offers a solution for
the dynamic adaptation of rule generation and, BPM
customisation and configuration composition. The
variability model services are responsible for carrying
mandatory or optional feature of the BPM and
metamodel. The feature of a process can be activated
or deactivated at any moment of time, i.e. design or
run time. We introduce a mechanism where domain
expert and user can perform their tasks in a simply
way. A domain expert can design high-level of
solution for domain, based on that solution, domain
user can modify and customise model elements
(activities) in any process over time.
The model elements activation and deactivation
depend on the variability model and what are the
requirement of end user. In this research, there are
two different groups of users, involving one expert in
domain with modelling knowledge and other have a
functional domain knowledge, but they are non-
technical. Therefore, we select software product line
engineering (SPLE) platform where the user can
perform the tasks. The SPLE is a standard model to
develop software applications using platform and
mass customisation (Böckle, van der Linden et al.
2005).
5.1 Feature Model
Figure 4 shows the feature model used in our case
study. For instance, the Language Model, Transfer
based MT, Interlingua MT, Direct MT and
Translation Model features are variants that can be
used during execution to accomplish the machine
translation functionality in the Model point.
5.2 Weaving Model
A product line feature model represents variabilities
and commonalities. The features in a feature model
are simply symbols with their type. Mapping features
to other models (feature model, domain model and
process model,) expressed in Table 1. Next, we show
how to perform the mapping by means of a weaving
model (Geyer and Becker 2002). We use a static
weaving model for managing the variability
relationships among all models. The principle
argument for using the static weaving model is with
domain-specific environment, when the domain
experts have significant domain knowledge. They
design and develop the domain template at design
(static) time. This weaving approach enables us for
scoping and configuring the Domain Models from a
set of given Features.
6 EVALUATIONS
We evaluate the DCT, analysing the time and
efficiency of its configuration including satisfaction
and operational compliance based on computing
human interaction by the end user. The goal of the
analysis is to get a comparative analysis of the time
and efficiency of configured DCT and its sub-process
systems. The emphasis is on analysing the relative
benefit of the proposed framework and a manual or
traditional or baseline approach regarding the
efficiency, effectiveness, and satisfaction with
function and operational compliance support. The
feature selection and configuration scenarios involve
to modifications resulting from improvement of the
complex process activity that affected the function
and operation of the process.
The current example is a part of a digital content
processing process model as a sample process for the
rule composition of business processes and domain
constraints that conduct this process. The source text
is translated into target language by the machine
translation activity. The translated text quality
decides whether further post-editing activity is
required. Usually, these constraints are domain-
specific, e.g., referring to domain objects, their
properties and respective rules.
The capability of the solution is effective,
efficient and satisfactory by the user when used under
specified conditions. We evaluate our contribution in
usability and forcing it on the following aspects
which is illustrated in Figure 5.
There are different component in the ISO standard
9241, applying the specification of usability into both
hardware and software designs. We discuss the
usability criteria and its goals.
In this overall framework, the usability criteria define
as effectiveness, efficiency and satisfaction. It
specified that end-users achieve specified goals are in
domain-specific environments.
Figure 5: Usability evaluation criteria.
Efficiency. The comparison between the time
taken of configuring domain constraints in
manual and semi-automatic process, based on
that find which process is more efficient.
Effectiveness. The generated rule configuration
in terms of accuracy to prevent or protect errors
to achieved the configuration of domain
constraints goal.
Satisfaction. The measure of end-user’s
comfort and acceptability of the overall
framework.
7 CONCLUSION AND FUTURE
WORK
In this paper, we have presented a model-based
framework, generating the domain-specific rule for
process model configuration and customisation in
dynamic environment. A case study on digital content
technology shows the applicability of this framework,
a realisation development approach and a proof of
concept prototype validating the feasibility of the
proposed approach. The main benefit of the
framework has: 1) Non-technical domain experts can
customise and configure the business process without
knowing any technical knowledge;2) The framework
can configure the domain constraints in dynamic
environment; and 3) A third is to evaluate the
framework architecture mechanisms in terms of
usability. It can be handled the completeness of
configuration of rule in terms: effectiveness,
efficiency and satisfaction base on computer human
interface by end use. The completeness of rule
configuration means performance, error free rule,
syntactically and semantically correctness after
configuring the rule.
We plan to extend this approach in combination
with our existing work on business process model
customisation based on user requirement (feature
model, domain model and process models), so that a
complete development life cycle for the customisati-
on and configuration of business process models is
supported. We also see the need for further research
that focuses on how to define the DSRL in terms of
abstract and concreate syntactical definition with
grammar formation across different domains and how
to convert conceptual models into generic domain-
specific rule language which are applicable to other
domains. So far this is a model to text translation, but
shall be improved with a system that learns from
existing rules and domain models, driven by the
feature model approach with automatic constraints
configuration, and to result in an automated DSRL
generation.
ACKNOWLEDGEMENTS
This research is supported by Science Foundation
Ireland (SFI) as a part of the ADAPT Centre at Dublin
City University (Grant No: 12/CE/I2267).
REFERENCES
Bergmayr, A. and M. Wimmer (2013). Generating
Metamodels from Grammars by Chaining
Translational and By-Example Techniques. MDEBE@
MoDELS.
Böckle, G., F. J. van der Linden and K. Pohl (2005).
Software product line engineering: foundations,
principles and techniques, Springer Science & Business
Media.
Boukhebouze, M., Y. Amghar, A. c.-N. Benharkat and Z.
Maamar (2011). A rule-based approach to model and
verify flexible business processes. International Journal
of Business Process Integration and Management 5(4):
287-307.
Ceri, S., P. Fraternali and A. Bongio (2000). Web Modeling
Language (WebML): a modeling language for
designing Web sites. Computer Networks 33(1): 137-
157.
Ceri, S., P. Fraternali and M. Matera (2002). Conceptual
modeling of data-intensive Web applications. IEEE
Internet Computing 6(4): 20-30.
Del Fabro, M. D. and P. Valduriez (2007). Semi-automatic
model integration using matching transformations and
weaving models. Proceedings of the 2007 ACM
symposium on Applied computing, ACM.
Diouf, M., S. Maabout and K. Musumbu (2007). Merging
model driven architecture and Semantic Web for
business rules generation. International Conference on
Web Reasoning and Rule Systems, Springer.
Distante, D., P. Pedone, G. Rossi and G. Canfora (2007).
Model-driven development of web applications with
UWA, MVC and JavaServer faces. International
Conference on Web Engineering, Springer.
Edwards, G., Y. Brun and N. Medvidovic (2012).
Automated analysis and code generation for domain-
specific models. Software Architecture (WICSA) and
European Conference on Software Architecture
(ECSA), 2012 Joint Working IEEE/IFIP Conference
on, IEEE.
Geyer, L. and M. Becker (2002). On the influence of
variabilities on the application-engineering process of a
product family. Software Product Lines, Springer: 1-14.
Gonçalves, R. C. A. (2015). Parallel programming by
transformation.
Groenewegen, D. M., Z. Hemel, L. C. Kats and E. Visser
(2008). WebDSL: a domain-specific language for
dynamic web applications. Companion to the 23rd
ACM SIGPLAN conference on Object-oriented
programming systems languages and applications,
ACM.
Gromoff, A., N. Kazantsev, K. Evina, M. Ponfilenok and
D. Kozhevnikov (2012). Modern era in business
architecture Design. Far East Journal of Psychology
and Business 9(2): 15-34.
Gupta, G. (2015). Language-based software engineering.
Science of Computer Programming 97: 37-40.
Gurunule, D. and M. Nashipudimath (2015). A Review:
Analysis of Aspect Orientation and Model Driven
Engineering for Code Generation. Procedia Computer
Science 45(0): 852-861.
Kang, K. C., S. G. Cohen, J. A. Hess, W. E. Novak and A.
S. Peterson (1990). Feature-oriented domain analysis
(FODA) feasibility study, DTIC Document.
Koch, N., A. Knapp, G. Zhang and H. Baumeister (2008).
UML-based web engineering. Web Engineering:
Modelling and Implementing Web Applications,
Springer: 157-191.
Linaje, M., J. C. Preciado and F. Sánchez-Figueroa (2007).
Engineering rich internet application user interfaces
over legacy web models. IEEE internet computing
11(6): 53-59.
List, B. and B. Korherr (2006). An evaluation of conceptual
business process modelling languages. Proceedings of
the 2006 ACM symposium on Applied computing,
ACM.
Mani, N., M. Helfert and C. Pahl (2016). Business Process
Model Customisation using Domain-driven Controlled
Variability Management and Rule Generation.
International Journal on Advances in Software
9(Numbers 3 & 4, 2016): 179 - 190.
Mani, N., M. Helfert and C. Pahl (2017). A Domain-specific
Rule Generation Using Model-Driven Architecture in
Controlled Variability Model. Procedia Computer
Science 112: 2354-2362.
Mani, N. and C. Pahl (2015). Controlled variability
management for business process model constraints.
ICSEA 2015, The Tenth International Conference on
Software Engineering Advances, IARIA XPS Press.
Moreno, N., S. Meliá, N. Koch and A. Vallecillo (2008).
Addressing new concerns in model-driven web
engineering approaches. International Conference on
Web Information Systems Engineering, Springer.
Musumbu, K., M. Diouf and S. Maabout (2010). Business
rules generation methods by merging model driven
architecture and web semantics. 2010 IEEE
International Conference on Software Engineering and
Service Sciences, IEEE.
Pahl, C. and N. Mani (2014). Managing quality constraints
in technology-managed learning content processes.
Prout, A., J. M. Atlee, N. A. Day and P. Shaker (2012).
Code generation for a family of executable modelling
notations. Software & Systems Modeling 11(2): 251-
272.
Rangiha, M. E. and B. Karakostas (2013). Goal-driven
social business process management. Science and
Information Conference (SAI), 2013, IEEE.
Reichert, M. and P. Dadam (1998). ADEPTflex—
Supporting dynamic changes of workflows without
losing control. Journal of Intelligent Information
Systems 10(2): 93-129.
Ringert, J. O., A. Roth, B. Rumpe and A. Wortmann (2015).
Code Generator Composition for Model-Driven
Engineering of Robotics Component & Connector
Systems. arXiv preprint arXiv:1505.00904.
Standard, O. Web services business process execution
language version 2.0.
Van der Aalst, W. M. and A. H. Ter Hofstede (2005).
YAWL: yet another workflow language. Information
systems 30(4): 245-275.
White, S. A. (2004). Business process modeling notation.
Specification, BPMI. org.