SWB Process

A Business Process Management System driven by Semantic Technologies

Hasdai Pacheco, Karen Najera, Hugo Estrada and Javier Solis

Fund of Information and Documentation for the Industry INFOTEC, Mexico DF, Mexico

Keywords:

Business Process Management System, Semantic Technologies, Ontology-Driven Development.

Abstract:

Constant changes in the market force enterprises to continuously deﬁne and redeﬁne their business processes,

and the technology that supports them, in order to fulﬁll the organizational objectives. Business Process

Management Systems (BPMS) are intensively used in organizations as a useful tool to face those changes.

However, as stated in literature and practice, a BPMS still has to cope with the agility to adapt to changes and

the low degree of automation of the BPM life-cycle. These issues have been faced through the integration

of BPM with Semantic Technologies. Some proposals are focused in conceptual approaches while others

involve tools to cover part of the BPM life-cycle. Nevertheless, there are no works that have implemented a

BPMS that exploits Semantic Technologies for covering the whole BPM life-cycle. In this paper we present

SWB Process, an industrial and Open Source BPMS completely driven by Semantic Technologies that uses

ontologies to agilely support constant changes in the processes of organizations, increasing the degree of

automation of the BPM life-cycle. Moreover, we take advantage of ontologies to quickly adapt it to new BPM

needs. SWB Process has been validated through real projects in several government agencies in Mexico.

1 INTRODUCTION

Constant changes in the market force enterprises to

continuously deﬁne and redeﬁne their business pro-

cesses, and the technology that supports them, in or-

der to fulﬁll the organizational objectives. In this

context, the paradigm of Business Process Manage-

ment (BPM) has been widely accepted in industry

and research to optimize enterprise resources and core

activities, since it encompasses methods, techniques

and Information Technologies (IT) to manage busi-

ness processes involving humans, applications, docu-

ments and other sources of information (van der Aalst

et al., 2003). BPM is directed by a life-cycle that com-

prises four phases: modeling, implementation, exe-

cution, and analysis (Wetzstein et al., 2007). These

phases can be covered by systems known as Busi-

ness Process Management Systems (BPMS). Thus, a

BPMS allows to generate business processes models

and provides mechanisms to translate those models

into an executable system that helps process perform-

ers to accomplish their business tasks.

Several BPMS have been proposed in academy and

industry (Butti et al., 2013; Domingue et al., 2013;

Jain et al., 2013; Calkins et al., 2013). However, as

stated in literature and practice (Hepp et al., 2005;

Wetzstein et al., 2007; Filipowska et al., 2011), there

are still some issues that a BPMS has to address, such

as the agility to adapt to changes and the low de-

gree of automation of the BPM life-cycle. In order

to overcome these issues, researchers are tackling the

integration of BPM with Semantic Technologies since

this integration offers inherently more ﬂexibility for

supporting and increasing the degree of automation

of the BPM life-cycle in changing scenarios (Davis,

2005; Filipowska et al., 2011). Some proposals are

focused in conceptual approaches, that is, BPM on-

tologies and formalizations (Panos and G

omez, 2012;

Oro and Ruffolo, 2012; Mueller, 2012), while oth-

ers involve tools using ontologies to cover part of the

BPM life-cycle, or architectures and functional re-

quirements for a Semantic BPMS (Wetzstein et al.,

2007; Karastoyanova et al., 2008; Domingue et al.,

2013). Nevertheless, there are no works that have im-

plemented a BPMS that exploit Semantic Technolo-

gies for covering the whole BPM life-cycle.

In this paper, we present an industrial and Open

Source BPMS called SWB Process

completely

driven by Semantic Technologies (ontologies, triple-

stores, query languages and reasoners) that supports

the whole BPM life-cycle. SWB Process has been de-

http://www.semanticwebbuilder.org.mx/SWBProcess

525

Pacheco H., Najera K., Estrada H. and Solis J..

SWB Process - A Business Process Management System driven by Semantic Technologies.

DOI: 10.5220/0004714705250532

In Proceedings of the 2nd International Conference on Model-Driven Engineering and Software Development (MODELSWARD-2014), pages 525-532

ISBN: 978-989-758-007-9

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

veloped following the Ontology-Driven Information

Systems (ODIS) approach (Guarino, 1998; Uschold,

2008). Accordingly, ontologies drove the develop-

ment of SWB Process, and they also play an impor-

tant role during the supported BPM life-cycle. By us-

ing ontologies, SWB Process provides a solution to

agilely support constant changes in organizations in-

creasing the degree of automation of modeling, im-

plementation, execution, and analysis phases of BPM

life-cycle. Moreover, we take advantage of Ontology-

Driven Development, such as maintenance and update

tasks improvement, making SWB Process ﬂexible to

be quickly adapted to new BPM needs.

SWB Process has been successfully validated through

real projects, supporting the business processes of

several government agencies in Mexico.

The paper is structured as follows: Section 2 de-

scribes the development of SWB Process. Section 3

presents how SWB Process supports BPM life-cycle

along with an example. Section 4 describes the bene-

ﬁts of SWB Process. Section 5 presents related works.

Finally, section 6 presents our conclusions and ongo-

ing work.

2 SWB PROCESS

DEVELOPMENT

In this section, the development of SWB Process

is presented. SWB Process is a Semantic BPMS

that uses the Business Process Model and Notation

(BPMN) 2.0 (OMG, 2011) as business processes

modeling language and a Web architecture for busi-

ness process execution and deployment. Ontology-

Driven Information Systems ideas (Guarino, 1998;

Uschold, 2008) were followed to deﬁne SWB Pro-

cess. Therefore, its development was addressed

by using an Ontology-Driven Development Frame-

work called SemanticWebBuilder (SWB) (Solis et al.,

2013), which provides the mechanisms to generate

the base source code of Web applications starting

from an ontology that deﬁnes system requirements.

For the development of SWB Process we used the

following components provided by SWB: a) A do-

main ontology, deﬁned with the Web Ontology Lan-

guage OWL (OMG, 2004a), that describes system

requirements for Web applications (SWBOntology);

b) a code generator to transform ontology deﬁnitions

(extending from the SWBOntology) into Java source

code (SWBCodeGen); and c) a platform with libraries

and utilities to accelerate software development and to

encourage software reuse (SWBPlatform).

We describe, in the following subsections, the three

major steps followed to generate SWB Process

(Fig. 1): Modeling the SWB Process Ontology, Au-

tomatic code generation and Speciﬁc development.

Figure 1: SWB development methodology.

2.1 Modeling the SWB Process

Ontology

The SWB Process ontology

(SWBPOntology), is

an OWL ontology that captures the functional re-

quirements of SWB Process, as well as the relevant

concepts and behavioral aspects from the Business

Process Model and Notation (BPMN) 2.0 speciﬁca-

tion (OMG, 2011). For the SWBPOntology mod-

eling, two previous activities were achieved: First,

the identiﬁcation of relevant classes from the BPMN

speciﬁcation and second, the identiﬁcation of con-

cepts and behaviors coming from system require-

ments for the implementation of the ﬁnal BPMS.

For the ﬁrst activity, elements for process modeling

and execution coming from the BPMN speciﬁcation

were considered (expressed as UML class diagram

fragments). This included the elements involved only

in BPMN process orchestration and BPMN collabo-

rations i.e. omitting those elements for BPMN chore-

ographies and BPMN conversations. As a result,

we had a set of candidate BPMN classes from the

Common, Foundation, Collaboration, Activities, Ar-

tifacts, Events, Data, Gateways and Process packages.

Examples of these classes are the Activity, Process,

Event, Subprocess, CallActivity, Gateway, BaseEle-

ment, Documentation and Task classes, among oth-

ers. It is important to point out that as our intent was

to provide process model persistence and data inter-

change through RDF and OWL formats, the section

on BPMN Diagram Interchange and Exchange for-

mats of the speciﬁcation were not considered in the

candidate classes identiﬁcation.

For the second activity, the gathering of additional re-

quirements for the BPMS implementation was needed

because the BPMN speciﬁcation is mainly focused in

deﬁning the notation and semantics for process mod-

eling and interchange using XML, thus it lacks of

http://www.semanticbuilder.com/SemWB4/SWB4/

swbp/WEB-INF/owl/ext/swp.owl

MODELSWARD2014-InternationalConferenceonModel-DrivenEngineeringandSoftwareDevelopment

526

implementation details for a ﬁnal BPMS. This ac-

tivity led us to the identiﬁcation of ontology con-

cepts and behaviors needed to support the BPM

life-cycle through the BPMS components. Exam-

ples of the identiﬁed concepts are, service concepts

(such as Web Service (WS), SPARQL querying and

DataBase querying), documents, templates, user in-

teraction components and process execution concepts

(such as process instance and ﬂow node instance).

With the candidate classes of the BPMN speciﬁca-

tion and the concepts from the BPMS requirements

the SWBPOntology modeling was carried out extend-

ing the SWBOntology provided by the SWB Frame-

work (Solis et al., 2013). Thus, OWL classes and

properties corresponding to the identiﬁed concepts

and candidate classes were created. We reused the

SWBOntology since it deﬁnes classes and properties

to support the automatic transformation from the on-

tology to Java source code and concepts for Web com-

ponents and Web site features development needed

since SWB Process performs process execution in a

Web architecture.

The main challenge in the modeling step was the

mapping between the BPMN class structure (which

includes multiple inheritance of classes) and sepa-

rated OWL classes (concepts and behaviors) in the

SWBPOntology as required by the SWB Framework

(where multiple inheritance is allowed only through

behaviors). At this point, the SWBOntology played a

fundamental role, since it includes deﬁnitions useful

to ﬁll the gaps in the implementation of the BPMN

speciﬁcation, for instance, participants are deﬁned in

BPMN as roles or resources but only at conceptual

level, on the other hand, SWBOntology has a deﬁ-

nition of roles and users that can be reused by SWB

Process along with their source code for user manage-

ment, user registration and user validation.

At the end of the modeling step, a total of 164 OWL

classes for the SWB Process implementation were de-

ﬁned in the SWBPOntology, of which 151 are concept

deﬁnitions and 13 are behavioral concepts.

2.2 Automatic Code Generation

This step consisted in the generation of SWB Process

base source code (in Java programing language) by

executing the SWBCodeGen. Therefore, it was con-

ﬁgured to use the SWBPOntology as input to get as

output the base source code of SWB Process, which

corresponds to a domain-speciﬁc and high-level Java

API (SWBP API). The generated SWBP API encap-

sulates the source code (classes and methods) neces-

sary to achieve the persistence of the objects involved

in the different components of SWB Process, includ-

ing: 1) a set of Java classes and interfaces correspond-

ing to the OWL concepts and behavioral aspects de-

ﬁned in the SWBPOntology; and 2) a set of class

methods to access the corresponding OWL properties

deﬁned in the SWBPOntology.

The SWBP API (comprised by 302 java classes and

13 java interfaces) is supported by the SWBPlatform

which provides a set of libraries that allows develop-

ers to use connectors to several triplestores for RDF

storage (such as Bigdata and Apache Jena). In this

way, the SWBP API helps developers to reduce the

complexity of managing RDF persistence in a stan-

dardized way in which data persistence mechanisms

are separated from the business logic of the end ap-

plication. Thus, accelerating the application develop-

ment.

2.3 Speciﬁc Development

This step consisted in the development of the compo-

nents and the operational business logic by using the

SWBP API to cover the SWB Process functionalities.

The operational business logic of each BPMN ele-

ment was developed taking into account the BPMN

execution semantics from the BPMN 2.0 speciﬁca-

tion and previous experience in the development of

workﬂow systems. This execution semantics served

as the basis for the deﬁnition and implementation of a

state-based process execution engine, as well as user

interaction components for the modeling, conﬁgura-

tion and management of business processes. These

components, whose architecture is shown in ﬁgure 2,

allow end users to manage the BPM life-cycle in a

generic way.

The SWB Process components are described below.

Process Modeler (SWBP Modeler). The SWBP

Modeler is a Web based component that supports

to graphically design business processes by using

BPMN 2.0 notation taking into account the BPMN

execution semantics (e.g. which BPMN elements can

be connected in a process ﬂow). It includes a mech-

anism to relate ontologies with a business process to

deﬁne the structure of business artifacts, such as data,

documents or Web Services. The SWBP Modeler

maps in a transparent manner to the user, each graphi-

cal element of a particular business process to its cor-

responding concept in the SWBPOntology, generat-

ing as output a SWBPOntology instance which de-

ﬁnes a Semantic Process Model.

Conﬁguration and Deployment Module (SWBP

Conﬁgurator). This component provides a Web

based user interface that takes as input the Semantic

Process Model for its conﬁguration in order to make it

executable and deployable on a Web site. The conﬁg-

SWBProcess-ABusinessProcessManagementSystemdrivenbySemanticTechnologies

527

Figure 2: SWB Process Architecture.

uration comprises two parts: a) the execution conﬁgu-

ration, which consists of capturing the values of each

property of the graphical elements of the process. The

properties are those deﬁned in the SWBPOntology

belonging to the Semantic Process Model. More-

over, it includes the relation of BPMN data objects de-

ﬁned in the Semantic Process Model, with classes of

the SWBPOntology or other ontologies that describe

business artifacts used along the process ﬂow. b) the

deployment conﬁguration, which consists of the deﬁ-

nition of Web form templates of process activities for

user interaction. The output of this component corre-

sponds to the Semantic Executable Process Model.

Business Process engine (SWBP Engine). The busi-

ness process engine empowers process execution and

monitoring. It implements a state machine that co-

ordinates business process ﬂow taking into account

properties, behavior and constraints deﬁned in the

SWBPOntology for each BPMN element, as well as

all the speciﬁc conﬁguration deﬁned in the Semantic

Executable Process Model. It also manages the ex-

ecution of Service and Script tasks. Moreover, the

SWBP Engine manages the persistence of all data of

the process and its execution in RDF format.

Management Module (SWBP Management). This

module allows the instantiation of the Semantic Exe-

cutable Process Model to generate Semantic Process

instances. It provides a business task inbox to accom-

plish the execution of Semantic Process instances.

The business task inbox allows users to perform their

tasks and to reallocate human resources for the tasks.

Moreover, the SWBP Management provides a mod-

ule for process tracking that retrieves process instance

data (stored in RDF format by the SWBP Engine) and

generates tables and graphs to show Semantic Pro-

cess instances execution performance. Furthermore,

it provides an SPARQL Endpoint to query process in-

formation, not only from a process instance, but also

from the process structure itself, such as, process ﬂow.

3 SWB PROCESS AND BPM

LIFE-CYCLE

In this section is described how Semantic Technolo-

gies, as basis of SWB Process, support the BPM life-

cycle. To do this, ﬁrst, we present the deﬁnition pro-

posed in (Wetzstein et al., 2007) for each phase of the

BPM life-cycle. Then, we explain how SWB Process

gives support to the phases.

• Modeling: in this phase, business analyst creates

a business process model with help of a modeling

tool by specifying the order of tasks in the busi-

ness process. This phase is covered by SWB Pro-

cess through the SWBP Modeler, which supports

the design of BPMN diagrams and transforms

those diagrams into a Semantic Process model.

• Implementation: in this phase, the business pro-

cess model, created in the modeling phase, is

transformed and enriched by IT engineers into an

executable process model. This phase is covered

by the SWBP Conﬁgurator which leads the con-

ﬁguration of the Semantic Process model (gener-

ated in the modeling phase) for its execution and

deployment on a Web site. After the conﬁgura-

tion, the Semantic Process model becomes a Se-

mantic Executable Process model which is ready

to be executed.

• Execution: in this phase, a process engine ex-

ecutes a process instance (an speciﬁc execution

of the executable process model), by navigating

through the control ﬂow of the executable process

model. This phase is covered by SWB Process

through the SWBP Engine and the SWBP Man-

agement. The SWBP Engine empowers the Se-

MODELSWARD2014-InternationalConferenceonModel-DrivenEngineeringandSoftwareDevelopment

528

mantic Process instance execution coordinating

the process ﬂow, taking into account the conﬁgu-

ration deﬁned in the Semantic Executable Process

model. The SWBP Management manages the ex-

ecution of Semantic Executable Process models

on a Web site, for instance, generating new Se-

mantic Process instances and providing a business

task inbox to list tasks to be performed by a user.

• Analysis: in this phase, process analysis com-

prises monitoring of running process instances

and process mining. This phase is covered by

SWB Process through the SWBP Management. It

includes a tracking component that displays: pro-

cess execution performance and information of

the running Semantic Process instances. More-

over, an SPARQL Endpoint is provided for query-

ing process information, such as, process ﬂow or

process instance execution and data.

3.1 SWB Process in Practice

In this section, the workﬂow with SWB Process is

presented, which consists of seven steps that cover

the BPM life-cycle. The ﬁrst ﬁve steps must be per-

formed by the business analysts or engineers to de-

ﬁne a Semantic Executable Process Model along the

modeling and implementation phases, the sixth and

seventh steps are useful during execution and analy-

sis phases, where process orchestration is performed.

Following, an example of a process generated with

SWB Process is described that comprises the men-

tioned steps. The example corresponds to the abstract

of one of the processes implemented in our research

center (INFOTEC), whose objective is to manage the

employee vacation requests. The process ﬂow is as

follows:

An ‘employee’ sends a vacation request, the ‘dept. su-

pervisor’ can approve the vacation request, reject it,

or ask the ‘employee’ to reschedule dates. In case

the ‘dept. supervisor’ approves the vacation request,

‘human resources’ department has to validate it. If

‘human resources’ department validates the request,

the ‘employee’ is notiﬁed via e-mail about the ap-

proval, otherwise, he is notiﬁed about the rejection;

in case the ‘dept. supervisor’ rejects the request, the

‘employee’ is notiﬁed via e-mail about the rejection;

and ﬁnally, in case the ‘dept. supervisor’ ask for

rescheduling dates, the ‘employee’ may modify dates

and send the vacation request again.

Process participants are: employee, dept. supervisor

and human resources department. Whereas data in-

volved in the process are: vacation start date, vacation

end date, request comment, reject comment, vacation

Figure 3: Vacation request data modeled in Protege.

request status and validation.

As a ﬁrst step, business artifacts are deﬁned in terms

of ontologies. In the example we refer to process

participants and data. For this purpose, an ontol-

ogy editor such as Protege

or TopBraid

can be

used. We have taken the SWBOntology (Solis et al.,

2013) as basis to reuse its User deﬁnition for pro-

cess participants, therefore, a new ontology that ex-

tends the SWBOntology is generated. On this ontol-

ogy, the VacationRequest class was created to deﬁne

data as data type properties of the class. The object

properties: user who request, user who approves and

user who validates, were created to relate the User

deﬁnition in the SWBOntology with the VacationRe-

quest class. Fig. 3 shows the implementation of the

VacationRequest class in Protege. The second step

corresponds to graphically model the BPMN diagram,

which includes the business process ﬂow and business

rules by using the SWBP modeler. Fig 4 shows the Va-

cation request process (the database symbol located at

the bottom represents the association of the Vacation-

Request class deﬁned in the ontology). In the third

and fourth steps, participants and data are related with

process activities and execution properties are conﬁg-

ured for each process element, this includes the asso-

ciation of ontology classes to the process data objects.

The ﬁfth step is related to the deﬁnition of Web form

templates of process activities for user interaction. In

Fig. 5, the employee request task conﬁguration is pre-

sented. Data deﬁned in the ontology and related to

process data objects is listed and the user can select

which property will appear in the Web form template

for this task, in this case: start date, end date and re-

quest comment from the VacationRequest class deﬁ-

Protege. http://protege.stanford.edu/

TopBraid. http://www.topquadrant.com/products/

SWBProcess-ABusinessProcessManagementSystemdrivenbySemanticTechnologies

529

Figure 4: Vacation request process modeled with SWB Process modeler.

Figure 5: Employee request task conﬁguration.

nition. Moreover, the user can deﬁne the type of form

element to be used for each concept property, such as

Text area or calendar date selector. After the ﬁfth step,

the process is ready to be deployed in a wrapper Web

page. In the sixth step, the employee participant can

generate a process instance to perform a vacation re-

quest through the User task inbox of the SWBP Man-

agement. Fig. 6 displays the employee request task

deployment.

Finally, in the seventh step, monitoring, tracking

and querying process data can be performed through

the SWBP Management.

4 SWB PROCESS BENEFITS

Using an Ontology-Driven approach in the devel-

opment of SWB Process allowed us to provide ad-

vantages over other industrial BPMS in several as-

pects. In terms of changes in requirements, ontolo-

gies provide a ﬂexible way to adapt to new BPM

needs, helping to increase the automation level of

the BPM life-cycle. For example, if the BPMN no-

tation changes and more primitives are added, it is

possible to quickly adapt SWB Process including the

new primitives in the SWBPOntology and in con-

sequence, in the SWB Process source code. The

same is true for changes in the execution seman-

tics of the BPMN speciﬁcation or for changes in

the way the BPM life-cycle works on business pro-

cesses. Moreover, using ontologies to deﬁne the sys-

tem architecture and data model enables the inter-

change of process deﬁnition and execution informa-

tion between other components, agents or IT systems

in a standard machine-readable format (OWL and

RDF (OMG, 2004a; OMG, 2004b)) without the need

for additional data treatment. In terms of dynamic

workﬂow management, ontologies in SWB Process

provide a mechanism to change process deﬁnitions

and execution properties on the ﬂy, maintaining the

consistency of the running process cases without the

need of process model (re)compilation. This is possi-

ble because all process models are a set of individuals

of the SWBPOntology and the SWBP Engine acts as

a model interpreter for the process models, instead of

an application compiler for each model.

On the other hand, by using ontology concepts as data

structures for business objects conﬁguration in a busi-

ness process model, SWB Process provides a way

to make explicit the knowledge involved in business

MODELSWARD2014-InternationalConferenceonModel-DrivenEngineeringandSoftwareDevelopment

530

Figure 6: Employee request task deployment.

process execution, enabling automatic process space

querying and reasoning tasks over process execution

data for business analysis, intelligence and improve-

ment. Besides these beneﬁts, as business processes

are described with Semantic Web standards, interop-

erability between business processes is enabled. Fur-

thermore, business process knowledge can be pub-

lished through a SPARQL endpoint on the Seman-

tic Web, and other Semantic Web paradigms such as

Linked Data can be applied to exploit business knowl-

edge and to enrich business process deﬁnitions.

5 RELATED WORKS

Research works are tackling the integration of BPM

with Semantic Technologies from different perspec-

tives. Some proposals are focused in theoretical

and conceptual approaches, for instance: in Oro

work (Oro and Ruffolo, 2012), a framework to cre-

ate business process ontologies is presented. Ontolo-

gies can be queried and exploited to monitor pro-

cess models, extract information from documents,

execute processes and monitor the execution, and

ﬁnally, analyze process instances. The work of

Mueller (Mueller, 2012) provides three ontologies

with important concepts of existing BPMS, such as

classes and properties of the BPMN 2.0 speciﬁcation

and the Service Component Architecture (SCA) as-

sembly model. Some applications of this ontologies

in the BPM life-cycle are described.

On the other hand, some approaches are focused

on the deﬁnition of system requirements for a Se-

mantic BPMS, or involve tools to partially cover

the BPM life-cycle, for instance: the work of Wet-

zstein (Wetzstein et al., 2007) describes functional re-

quirements for a Semantic BPMS, according to the

BPM life-cycle: modeling (semantic annotation, and

process fragments), implementation (process compo-

sition), execution (dynamic SWS discovery and in-

vocation) and analysis (process mining and monitor-

ing). The work of Karastoyanova (Karastoyanova

et al., 2008) presents a reference architecture for a Se-

mantic BPMS. The architecture comprises function-

alities for each phase of the BPM life-cycle. More-

over, the authors propound a Semantic Execution En-

vironment and show how existing BPMS components

can be extended with semantic features. The SU-

PER Project (Domingue et al., 2013) applies seman-

tic technology to acquire, organize, share and use the

stakeholders knowledge and knowledge embedded in

business processes within existing IT systems, in or-

der to make companies more adaptive.

However, there are no works that have implemented

a BPMS that exploit Semantic Technologies for cov-

ering the whole BPM life-cycle, which is the case

of our approach. Our BPMS encompasses Seman-

tic Technologies at conceptual and implementation

levels. At conceptual level, our BPMS has an on-

tology (SWBPOntology) that deﬁnes elements of the

BPMN 2.0 speciﬁcation along with BPMS system

requirements (similar to Mueller (Mueller, 2012)),

moreover, additional ontologies can be created and

associated to speciﬁc business processes to deﬁne

business objects making explicit the knowledge in-

volved in a business process. In contrast to Oro

work (Oro and Ruffolo, 2012), where a framework is

provided to create process ontologies, in our BPMS,

business processes are automatically represented in

terms of ontologies, when they are modeled with

BPMN 2.0, since business processes are SWBPOn-

tology instances. Thus, having the same ontologies

advantages without needing to generate them sepa-

rated nor giving process data treatment. At imple-

mentation level, we present an implemented BPMS

that covers the whole BPM life-cycle, in contrast to

SWBProcess-ABusinessProcessManagementSystemdrivenbySemanticTechnologies

531

SUPER (Domingue et al., 2013) which is focused

on Semantic Web Services, and in contrast as well

to Wetzstein (Wetzstein et al., 2007) and Karastoy-

anova (Karastoyanova et al., 2008) works, where

BPMS functional requirements are proposed without

system implementation.

6 CONCLUSIONS

In this paper, we have presented an industrial and

Open Source Semantic BPMS called SWB Process.

It has been developed following the Ontology-Driven

Information Systems approach. Accordingly, on-

tologies were directly involved in the development

of SWB Process through Ontology-Driven Develop-

ment, and ontologies also play an important role dur-

ing the supported BPM life-cycle. By using ontolo-

gies as the basis of SWB Process, we provide a so-

lution with ﬂexible and agile mechanisms to adapt to

new BPM needs and continuous changes in organiza-

tions, increasing the degree of automation and better

supporting the BPM life-cycle. Moreover, the infor-

mation implicitly represented in a business model has

explicit meaning, therefore, machines as well as peo-

ple are enabled to understand, share and reason over

business processes models and information. In addi-

tion, other Semantic Web paradigms can be applied to

exploit business information such as Linked Data.

SWB Process has been successfully implemented and

validated through real projects, supporting the busi-

ness processes of several government agencies in

Mexico, for instance, the Federal Electricity Com-

mission (CFE)

and the National Institute of Women

(INMUJERES)

. Moreover, it has been used to imple-

ment processes of our research center (INFOTEC).

In tandem with SWB Process, we provide the follow-

ing support services: consultancy, mentoring, techni-

cal support, training and customization.

In our ongoing work, we are addressing the semantic

annotation of business processes with external ontolo-

gies, to clarify processes through generic concepts,

providing additional support to business analysis dur-

ing modeling and enabling reuse of information.

REFERENCES

Butti, S. et al. (2013). WebRatio tool. Online.

http://www.webratio.com/. Last access: 05/07/2013.

Calkins, M. W. et al. (2013). Apian tool. Online.

http://www.appian.com/. Last access: 05/07/2013.

http://www.cfe.gob.mx/

http://www.inmujeres.gob.mx/

Davis, M. (2005). The Business Value of Semantic Technol-

ogy, from vision to mainstream markets 2000 - 2010.

In Semantic Web Applications for National Security.

Domingue, J. et al. (2013). SUPER Project. Online.

http://www.ip-super.org/. Last access: 05/07/2013.

Filipowska, A., Kaczmarek, M., Koschmider, A., Stein, S.,

Wecel, K., and Abramowicz, W. (2011). Social soft-

ware and semantics for business process management

- alternative or synergy? Journal of Systems Integra-

tion, 2(3):54–69.

Guarino, N. (1998). Formal ontology and information sys-

tems. In Proceedings of Formal Ontology in Informa-

tion System, pages 3–15. IOS Press.

Hepp, M., Leymann, F., Domingue, J., Wahler, A., and

Fensel, D. (2005). Semantic business process man-

agement: a vision towards using semantic web ser-

vices for business process management. In e-Business

Engineering, ICEBE 2005, pages 535–540.

Jain, R. et al. (2013). Intalio tool. Online.

http://www.intalio.com/. Last access: 05/07/2013.

Karastoyanova, D., Lessen, T. V., Leymann, F., Ma, Z.,

Nitzsche, J., Wetzstein, B., Bhiri, S., Hauswirth, M.,

and Zaremba, M. (2008). A reference architecture for

semantic business process management systems. In

Multikonferenz Wirtschaftsinformatik.

Mueller, R. (2012). Enterprise applications of semantic

technologies for business process management. Jour-

nal of Zhejiang University SCIENCE C, 13(4):308–

310.

OMG (2004a). OWL Web Ontology Language. On-

line. http://www.w3.org/TR/2004/REC-owl-features-

20040210/.

OMG (2004b). Resource Description Framework (RDF).

Online. http://www.w3.org/RDF/.

OMG (2011). Business Process Model and Notation 2.0.

http://www.omg.org/spec/BPMN/2.0/.

Oro, E. and Ruffolo, M. (2012). A knowledge representa-

tion formalism for semantic business process manage-

ment. In Advances in Knowledge Representation.

Panos, A. and G

omez, J. (2012). Dealing with Vagueness

in Semantic Business Process Management through

Fuzzy Ontologies. In International Workshop on Se-

mantic Business Process Management.

Solis, J., Pacheco, H., Najera, K., and Estrada, H. (2013).

A MDE Framework for Semi-automatic Development

of Web Applications. In 1st International Conference

on MDE and Software Development, pages 241–246.

Uschold, M. (2008). Ontology-driven information systems:

Past, present and future. In FOIS’08, pages 3–18.

van der Aalst, W., Hofstede, A. H. M. T., and Weske, M.

(2003). Business process management: A survey.

In 1st International Conference on Business Process

Management, volume 2678 of LNCS, pages 1–12.

Wetzstein, B., Ma, Z., Filipowska, A., Kaczmarek, M.,

Bhiri, S., Losada, S., Lopez-Cob, J.-M., and Cicurel,

L. (2007). Semantic business process management:

A lifecycle based requirements analysis. In Workshop

on Semantic Business Process and Product Lifecycle

Management.

MODELSWARD2014-InternationalConferenceonModel-DrivenEngineeringandSoftwareDevelopment

532