=Paper=
{{Paper
|id=Vol-460/paper-13
|storemode=property
|title=An Ontological Framework for Integrated Public Service Delivery
|pdfUrl=https://ceur-ws.org/Vol-460/paper13.pdf
|volume=Vol-460
}}
==An Ontological Framework for Integrated Public Service Delivery==
https://ceur-ws.org/Vol-460/paper13.pdf
An Ontological Framework for
Integrated Public Service Delivery
Sietse Overbeek1, Marijn Janssen1, and Patrick van Bommel2
1
Faculty of Technology, Policy and Management, Delft University of Technology,
Jaffalaan 5, 2600 GA Delft, The Netherlands
{S.J.Overbeek, M.F.W.H.A.Janssen}@tudelft.nl
2
Institute for Computing and Information Sciences, Radboud University Nijmegen,
Toernooiveld 1, 6525 ED Nijmegen, The Netherlands
P.vanBommel@cs.ru.nl
Abstract. Present-day governments are experiencing a shift from supplying
common, non-electronic services towards supplying demand-driven and
personalized e-services. An ontological framework for integrated public service
delivery is presented to anticipate on the need to match supply and demand of
public services. This framework contributes to distinguish key concepts and
relations that form the basis for coordinating the activities necessary for
integrated service delivery. The ontology has been realized by studying
organizational processes, by conducting interviews as part of an expat case, and
by literature study. In the case of this research, expats are persons who live in
another country and want to come over to The Netherlands for their work. For
this purpose, they need to request services that are integrated in a whole. The
proposed ontology provides a foundation for an architecture blueprint that can
enable demand-driven integrated service delivery in practice.
1 Introduction
Governments are experiencing a shift from supplying common, non-electronic
services towards more demand-driven and personalized electronic service (e-service)
delivery. To accomplish these goals, governments are becoming more externally
oriented instead of focusing on internal functions. They are focusing more on their
client’s needs and less on their own functionality, organizational structure, and
boundaries. Initially, public organizations focused on recurring client needs instead of
on incidental needs. As such, assessing needs and reacting to needs do not provide the
flexibility to react to new needs or even changes in laws and regulations. Government
functions are fragmented due to constitutional, legal, and jurisdictional limitations. As
a consequence, governments are often acting in silo structures, but nowadays are
forced to cooperate with other government agencies and partners in the private sector.
One of the initiatives that anticipates on the aforementioned governmental
developments is the long-term ‘B-dossier’ research project [1]. Results of this project
are specifically aimed at providing computer-based support for government agencies
to realize a more Integrated Service Delivery (ISD). For ISD the tasks of several types
Proceedings of ONTOSE 2009 129
of actors need to be coordinated. In this case, an actor can be defined as an entity that
is able to perform a task, such as a human or a computer.
The focus of the research reported in this paper is to determine relations and
interdependencies between the main concepts for ISD. ISD requires that public
organizations collaborate with each other, which in turn requires insight in the
relationships among the functionalities and services provided by such organizations.
This understanding contributes to distinguish concepts and relations that form the
basis for coordinating the activities necessary for ISD. This is realized by the
development of an ontology for ISD, which aligns and abstracts domain knowledge
found in a case in the public domain and by studying relevant literature. Section 2
clarifies background knowledge necessary to develop an ontology for ISD. From an
organization-centric view, knowledge is acquired from the case by applying a bottom-
up approach which consists of studying organizational processes that include
information on service supply during process fulfillment. Several process models
have been created as a result of this study. Besides studying organizational processes,
we have analyzed how actors would fulfill their part of a process. This has been
realized by conducting interviews. Thus, we have applied an organizational view and
an actor-centric view when studying processes in public organizations. These views
have been extended by incorporating other views on public processes from current
literature before proceeding to the realization of an ontological knowledge framework
in section 3. Finally, section 4 concludes this paper.
2 Background
Ontologies are becoming essential for organizations, because ontologies are machine-
processable semantic resources for many application areas [2]. An ontology is an
agreed understanding of a certain domain, formally represented as logical theory in
the form of a computer-based resource. By sharing an ontology, autonomous and
distributed software applications can meaningfully communicate to exchange data and
thus make such data transactions interoperate independently of their internal
technologies. Relating the notion of ontology to the research described in this paper, it
can be noticed that organizations sharing an ontology which includes semantics
related to the public domain create a starting point for realizing ISD.
To understand the relations and interdependencies between main concepts for ISD
we have studied processes involved in the expat case in detail. Actor involvement in
those processes has been studied by interviewing expats who search and request
public services. Expats are, in this case, persons who live in another country and want
to come over to The Netherlands for their work. For this purpose they will at least
need a (temporary) residence permit, a registration in the citizens’ registry, a bank
account, a job, a health insurance, and housing. Process models have been developed
for each of these scenarios. The expat case contains typical problems of ISD and
involves organizations that need to collaborate. An example of a problem related to
ISD is to bridge the digital divide [3]. Citizens lacking Internet access at home should
be able to use e-services by other means, such as community self-service terminals.
130 Proceedings of ONTOSE 2009
A process model containing processes that are required to fulfill when requesting a
residence permit for expats has been developed and is shown in figure 1.
Fig. 1. High-level process model for requesting and receiving a residence permit.
This model is based on the Business Process Modeling Notation (BPMN) [4], which
is an industry standard graphic notation for representing organizational processes. The
model is based on information concerning the processes related to the acquisition of a
residence permit provided by the Dutch Immigration and Naturalization Service
(INS). Figure 1 shows that there are three composite processes involved when
obtaining a residence permit. First, a temporary residence permit has to be requested
by an expat followed by the request of a permanent residence permit before
registering at the municipality. To comprehend what the composite processes exist of,
detailed process models for the composite processes have also been developed. It has
been noticed that several parties are involved in the process to let an expat obtain a
temporary residence permit. The process starts by the expat requesting to obtain a
residence permit from the INS. The remaining process steps can then be fulfilled until
the expat collects the residence permit from the INS. An ontology that is shared by
public organizations is an agreed understanding of the public domain and as such
enables to identify essential concepts and relations between concepts.
Studying public processes in which multiple parties interact is but one of the ways
to achieve a better understanding of ISD. This can be regarded as an organization-
centric view on ISD, because public processes are arranged by public organizations
such as the INS, municipalities, and embassies. As part of the expat case, eleven
interviews have been conducted with expats to understand how they participated in
process fulfillment during their attempts to acquire a residence permit, a registration
in the citizens’ registry, and so on. Lessons learned from this actor-centric view can
not only be used to improve current governmental processes and service delivery, but
they can also be used to understand ISD from an actor’s point of view.
Central issues related to the residence permit process that were experienced by the
interviewed expats are concerned with: Information that is presented in Dutch only,
governmental Web sites that are not functioning properly, contradictory information
presented by multiple public organizations, and serious human mistakes during
service delivery. The latter is related with loss of documents and failing / forgetting to
inform other parties in the process. These issues obviously appeal to improve ISD, of
which the ontological framework presented in section 3 can act as a step in the right
direction. Our research contributes to at least partly resolve the aforementioned issues.
The issue of presenting contradictory information can be resolved by letting public
organizations share an ontology such as is presented in the next section, so that an
agreed understanding is realized concerning the public domain. Based on the
ontology, integrated public e-services can communicate to exchange data and thus
make such data transactions interoperate independently of technology. This can at
least partly resolve the issue of malfunctioning Web sites. Human mistakes during
Proceedings of ONTOSE 2009 131
service delivery can be diminished when more insight is provided how services can be
offered and integrated for repeatable service requests.
Now that we have gathered both organization-centric and actor-centric insights on
processes in the public domain with respect to a case in which typical problems for
ISD are surfaced, it is possible to develop an ontology that contributes to distinguish
the key concepts and relations for coordinating the activities necessary for ISD.
3 Ontological framework for integrated public service delivery
To accurately describe key concepts and relations between such concepts in the public
domain, the domain knowledge that we have aggregated in the previous section is
used as a basis for the creation of an ontological knowledge framework. Besides these
insights the ARIS EPC (Event-Driven Process Chains) model [6] can be used for
ontology creation, because it provides a description of a public process that extends
our organization-centric and actor-centric views with service-centric, resource-centric,
and event-centric views. The service-centric view describes concrete services offered
by organizations that are required by actors for successful process fulfillment. The
resource-centric view describes resources belonging to an organization and which
processes consume which resources. To understand the event-centric view an
explanation of this view is provided in the next section. First, an ORM representation
of the ontological framework for ISD is presented in section 3.1. Second, an OWL
specification of the ORM model is presented in section 3.2.
3.1 ORM representation of the ontological knowledge framework
Figure 2 shows an Object-Role Modeling (ORM) model of the proposed ontological
framework for ISD. ORM is a conceptual data modeling technique, which can not
only be used for the conceptual modeling of database models, but for a variety of
modeling purposes such as the modeling of ontologies.
132 Proceedings of ONTOSE 2009
Fig. 2. An ontological framework for integrated public service delivery.
In an ORM model, ovals represent object types (which are counterparts of classes),
whereas boxes represent relations between object types. These relations are dubbed as
fact types. For more details on Object-Role Modeling, see e.g. [7].
There are eight central concepts that are part of the ontology. These are the
concepts of role, actor, service, process, resource, organization, event, and an Event-
Driven Service-Oriented Architecture (EDSOA) concept. A description of a public
process from an actor-centric and organization-centric view, such as discussed in
section 2, forms the basis of introducing the ‘actor’ and ‘organization’ concepts in the
ontology. The concepts of service, resource, and event are part of the ARIS EPC
model [6]. The role concept is introduced in the ontology to be able to denote a
specification of an actor enactment. An actor is a resource of an organization that
enacts a role during process performance or, on a more granular level, task
performance. An employee enacting the role of registrar at a municipality is an
example of such an actor at a public organization. These actors can use services that
are offered by organizations. Services are on their turn required by actors during
Proceedings of ONTOSE 2009 133
process performance to assist actors in process fulfillment. For example, an expat that
performs the process to acquire a residence permit is provided with an e-service to
request a permit online and to provide those digital documents to government
agencies that are necessary for the permit request.
Finally, the concepts of event and event-driven service-oriented architecture need
to be introduced. An EDSOA in the context of this research defines a methodology
for designing and implementing computer-based applications and systems in which
events are transmitted between a set of integrated and interacting services [8]. Such
events are consumed or produced by actors in organizations. An actor that consumes
an event can subscribe to an architecture that manages such events, and an actor that
produces an event publishes to this architecture. When an event is broadcasted by an
actor, the architecture facilitates that this event is forwarded to a demanding actor. If a
demanding actor is unavailable, the architecture can facilitate the storage of the event
and try to forward it later. This architecture-based coordination of events can be
dubbed as event orchestration [1]. An example of an event in the residence permit
process shown in figure 1 can be an event ‘residence permit form received’. A
subsequent event that can be produced by a receiving expat is a ‘residence permit
signed’ event. Building applications and systems based on an EDSOA allows these
applications and systems to be more responsive, since such systems are more oriented
to unpredictable and asynchronous environments. Eventually, implementation of an
EDSOA based on the ontological framework shown in figure 2 can enable ISD and
orchestration of events between services in practice.
To increase usability for public organizations that wish to adopt the ontological
framework shown in figure 2, the ontology might be specified in multiple
specification languages, such as XML, RDF, RDF-S, OWL, etc. (see e.g. [2]). These
languages are specifically designed for use by computer-based applications that need
to process the content of information instead of just presenting information to human
actors. However, the Web Ontology Language OWL facilitates greater machine
interpretability of Web content than that supported by e.g. XML, RDF, and RDF
Schema (RDF-S) by providing additional vocabulary along with a formal semantics
[2]. To increase successful adaptation and machine interpretability of our ontological
framework, an OWL specification of the ontological framework that has been
visualized in ORM so far is presented in the next section.
3.2 OWL representation of the ontological knowledge framework
A partial OWL representation of the ontological framework for integrated public
service delivery is shown in figure 3. This representation verbalizes the concepts,
relations, and constraints of the ORM model shown in figure 2.
134 Proceedings of ONTOSE 2009
....
1
Fig. 3. A partial OWL representation of the ontological framework for
integrated public service delivery.
Representing the ontological concepts and relations between those concepts by means
of OWL yields differences compared to modeling it in ORM. However, both
languages intend to express the same meaning. For example, the ORM uniqueness
constraint that spans over ‘Enacts / IsEnactedBy’ cannot be expressed in OWL, as it is
implied by definition [2]. I.e., the formalization of ObjectProperties in OWL does not
allow the same tuple to appear twice in the same set, such as Enacts =
{,}. The other uniqueness and mandatory constraints are
all expressed as a cardinality restriction in OWL. For instance, the mandatory
constraint on ‘Coordinates’ is expressed in OWL by the constraint
‘owl:minCardinality’. An ‘owl:minCardinality’ constraint of one or more means that
all instances of the class must have a value for the property.
The differences in modeling the ontology as described above illustrate different
ways of characterizing the ontology. The contrast in formalizations and constructs of
both languages causes such differences. The choice of which language is more
suitable for specifying an ontology depends on the application scenario and
perspectives of the ontology [2]. For example, ORM and EER are suitable for
database and XML-based application scenarios since they are extensive in their
treatments of data set integrity. Description logic based languages such as OWL seem
to be more applicable for deductive and reasoning-based application scenarios, as
they focus on the expressiveness and the decidability of axioms.
As a next step in this research, an event-driven service-oriented architecture will be
developed that adopts the proposed ontological framework for ISD as the knowledge
base. The ontological framework provides a foundation for describing actor context,
public processes, resources, etc. Therefore, the ontology helps the architecture to
dynamically compose a personalized process flow and automate the execution of the
process flow.
Proceedings of ONTOSE 2009 135
4 Conclusions
Governments are experiencing a shift from supplying common, non-electronic
services towards more demand-driven and personalized electronic service (e-service)
delivery. To anticipate on these developments, an ontological framework for
integrated public service delivery is described in this paper. The key concepts and
relations that form the basis for coordinating the activities necessary for integrated
public service delivery can be distinguished by means of this ontology. It is also a
foundation for an EDSOA that can integrate services and orchestrate events between
services in practice. Knowledge to develop the ontology is acquired by studying a
case in which expats search and request public services, and by studying relevant
literature. Expats are persons who live in another country and want to come over to
The Netherlands for their work. Public processes have been studied from an
organizational viewpoint resulting in detailed process models. This has been
illustrated by a process for requesting a residence permit. Furthermore, several expats
have been interviewed to understand how actors participating in such processes would
fulfill their part of the process resulting in an actor-centric view on public processes.
Next, the studied processes in the expat case have been hierarchically structured by
identifying the core processes, the actors that participate in each core process, and the
tasks those actors fulfill as part of a core process. Finally, an ontological framework
for ISD is realized by extending the results from the expat case with findings from
related literature. The resulting ontology is represented as an ORM model and
described in OWL.
References
1. Klievink, A.J., Janssen, M.F.W.H.A., Lankhorst, M.M., Leeuwen, D. van: An Event-
driven Service-Oriented Architecture for Coordinating Flexible Public Service Networks.
In: Wimmer, M., Jochen Scholl, H., Ferro, E. (eds.) Electronic Government, 7th
International Conference, EGOV 2008, Turin, Italy, August 31 - September 5, 2008,
Proceedings. LNCS, vol. 5184, pp. 133-140. Springer, Berlin (2008)
2. Jarrar, M., Meersman, R.: Ontology Engineering – The Dogma Approach. In: Dillon, T.,
Chang, E., Meersman, R., Sycara, K. (eds.) Advances in Web Semantics I. LNCS, vol.
4891, pp. 7-34. Springer, Berlin (2008)
3. Ke, W., Wei, K.K.: Successful E-Government in Singapore. Communications of the
ACM. 47, 95-99 (2004)
4. OMG, Business Process Modeling Notation (BPMN) Version 1.0. OMG Final Adopted
Specification, Object Management Group (2006)
5. Dang, J., Hedayati, A., Hampel, K., Toklu, C.: An Ontological Framework for Adaptive
Medical Workflow. Journal of Biomedical Informatics. 41, 829–836 (2008)
6. Wikipedia. Event-driven Process Chain. Available from:
http://en.wikipedia.org/wiki/Event-driven_process_chain
7. Halpin, T.: Information Modeling and Relational Databases, from Conceptual Analysis to
Logical Design. Morgan Kaufmann, San Mateo, CA (2001)
8. Yuan, S.-T., Lu, M.-R.: An Value-Centric Event Driven Model and Architecture: A Case
Study of Adaptive Complement of SOA for Distributed Care Service Delivery. Expert
Systems with Applications. 36, 3671-3694 (2009)