BUSINESS PROCESS MODELING AND SOA IN INDUSTRIAL
O&M APPLICATION DEVELOPMENT
David Hästbacka, Petri Kannisto and Seppo Kuikka
Department of Automation Science and Engineering, Tampere University of Technology
P.O. Box 692, FI-33101 Tampere, Korkeakoulunkatu 3, Tampere, Finland
Keywords: BPM, SOA, Application development, Orchestration, Operation and Maintenance, Industry.
Abstract: While striving to increase profits in global competition, companies are trying to improve efficiency and
reduce costs by outsourcing and focusing on their core functions. For operation of industrial plants this often
results in provision of services even for high-priority activities such as maintenance. Integration of external
information systems and service providers to business processes and information workflows brings new
challenges to application development in order to support introduction of maintenance services as efficiently
as possible. This paper discusses the approach of applying business process modeling and service-oriented
concepts to development of supporting software applications. Business process modeling is proposed for
describing service interactions and information flows, and to function as a foundation for the application
development. To satisfy required flexibility in changing business environments, the applications represented
as services are composed into executable process workflow orchestrations using standard Internet
technologies. To validate the approach a scenario consisting of a condition monitoring process and an
environment footprint estimator is presented.
1 INTRODUCTION
Operating models in manufacturing and process
industries have evolved in such a way that industrial
companies have to concentrate on their primary
production related functions in order to stay
competitive and profitable in the global
environment. This has led to provision of services
for non-core activities in the quest to improve
efficiency and reduce costs by outsourcing and
relying on services from business partners.
Operation and maintenance (O&M) of industrial
plants involves operative production and
maintenance of devices and systems as a whole.
Maintenance is an important factor in achieving high
utilization and productivity, and its activities depend
highly on operating processes and production
schedules. For task co-ordination, asset management
and sharing of mutually important data becomes
even more important when services of external
service providers are being used. This information is
required in performing the service or produced as a
result of it.
Unfortunately, a lot of this information exchange
is still transferred manually between enterprises, i.e.
from person to person or by email. Development of
applications to support O&M tasks is complicated
and the developed applications are challenging to
maintain due to heterogeneity among technologies
and practices. The work processes should be taken
into account for information exchange requirements,
and management of work processes and resources in
service collaborations should also be considered.
Our previous work in the area of Web service
technologies and O&M have shown inefficiencies in
the development of O&M applications. In order to
provide applications that better support operations in
distributed service environments, the degree of
automation in development has to increase. This
paper presents an approach based on business
process modeling (BPM) and utilization of service-
oriented architecture (SOA) in development of
applications to support O&M services. With process
modeling a mutual understanding of services and
information flows between participants can be
achieved. These models are then essential in
implementing the software applications that support
activities in the distributed environment.
SOA is an important facilitator in loosely
coupled information system integration and is well-
suited for cross-enterprise integration. The
277
Hästbacka D., Kannisto P. and Kuikka S..
BUSINESS PROCESS MODELING AND SOA IN INDUSTRIAL O&M APPLICATION DEVELOPMENT.
DOI: 10.5220/0003507202770285
In Proceedings of the 13th International Conference on Enterprise Information Systems (ICEIS-2011), pages 277-285
ISBN: 978-989-8425-55-3
Copyright
c
 2011 SCITEPRESS (Science and Technology Publications, Lda.)

autonomy and composability of services in SOA
provide means for management of service processes
and rapidly changing business environments.
To validate the approach, two different kind of
executable business processes were implemented to
illustrate integration of O&M activities and data
gathering. The first process is an activity triggering
condition monitoring process while the
environmental footprint estimator is used for
aggregating data to be used in decision support. The
results and findings of the implementations are
presented as well as examined for future research.
Section 2 presents background and related work
in the area of O&M in industry, BPM and SOA
advances. Section 3 presents the approach based on
business process modeling of O&M activities in
application development, and the service-oriented
aspects of information systems implementing
executable processes. The condition monitoring and
environmental footprint estimator processes are
presented in section 4, and the paper is concluded
with future work in section 5.
2 BACKGROUND AND RELATED
WORK
2.1 O&M in Industry
Current industrial manufacturing practices have
evolved into efficient and rationalized operation
processes. Utilization of expertise and supporting
services from various service providers has
increased as the emphasis of manufacturing is
focused more on core competencies. This has
resulted in extensive use of services, and in many
cases, from a large number of service providers.
Maintenance is an important supporting activity
that has been increasingly outsourced in
manufacturing and process industry. Levery (1998)
states that maintenance of productive assets is
outsourced for specialist maintenance where the
expertise is not available in-house or when repetitive
routine work is performed. He continues that
between these there are tasks, where maintenance
requirements cannot be defined clearly or be related
to certain assets of parts of the production process,
that require knowledge of in-house maintenance and
production process related information.
There are many different types of maintenance
strategies used that all require a lot of planning of
both preliminary schedules and tasks to be executed.
Whether it is reactive maintenance, in case of a
sudden failure, or a scheduled preventive
maintenance operation, the tasks need to be
allocated to an unambiguous location in the plant
hierarchy. In addition to the plant model, also
previous incidents and operations are of relevance
when planning and executing the work.
As condition based maintenance is becoming
general this brings another valuable source of
information that is used in planning of maintenance
operations. Condition monitoring systems can
provide notifications and trigger events for starting
required maintenance tasks either in advance or
when a sudden failure occurs.
2.2 Towards Service Systems
Process device manufacturers have shown growing
interest in providing after-sales services for their
devices, e.g. maintenance throughout the lifecycle.
Auramo et al. (2004) have analyzed a process
equipment supplier shifting its operation from
selling equipment and services to offering long-term
partnerships, and taking more responsibility of the
operations of the customer. The authors argue that
the supplier’s role as a co-producer of value results
in increased operational efficiency while
simultaneously providing better service to the
customer. Baines et al. (2007) reviewed the state-of-
the-art in the integrated combination of products and
services and concluded that provision of services
requires understanding of how customers value the
service and how products, technologies and
operations support this value offering.
Managing the resulting service network requires
administration similar to in-house processes.
However, as work practices may differ, it is more
challenging to integrate and, for instance, ensure
sufficient communication and information exchange.
It is characteristic to a service network that the
business environment may change, i.e. when service
providers are replaced or a new participant enters the
network. Introducing the new circumstances requires
flexibility in adapting the processes accordingly.
This requirement is especially challenging from an
information systems integration point of view due to
the various data exchanges required in the actual
services performed.
2.3 Information Exchange
Straightforward access to maintenance data and
related information systems is of significant value in
planning, conducting and reporting maintenance
activities. Information on running operations is
ICEIS 2011 - 13th International Conference on Enterprise Information Systems
278

constantly accumulated and stored in various
systems. Maintenance related information is
typically stored in enterprise resource planning
(ERP) and manufacturing execution systems (MES).
However, with more sophisticated control systems,
i.e. DCS and PLC systems - and intelligent field
devices with condition monitoring features, for
instance - the diversity in integrations is increasing.
In addition, there can be other information sources,
e.g. provided by device vendors or authorities, that
are of relevance in maintenance operations.
Environmental aspects, for instance, are becoming
more important for sustainable production and there
are external data sources providing reference data
that can be used in assessments.
In organizationally and geographically
distributed business environments it should be of
mutual interest to have access to up-to-date data in
order to make the most of existing information and
to be able to receive and provide the services as
efficiently as possible. It is still quite common that
information is not shared automatically between
information systems as part of the operations, and
data required for the tasks need to be gathered
manually and reported after completion.
E-maintenance is a term used for software
platforms and integration of physical processes with
enterprise tools. In the PROTEUS initiative the aim
was to create a software platform to support web
based e-maintenance practices using Internet
technologies and middleware (Szymanski et al.,
2003; Bangemann et al., 2006). Business process
integration to the platform has also been considered
by Hausladen & Bechheim (2004). They state that
for successful integration the workflow must be
defined clearly and that the processes must be stable
and not change on short term perspective in order to
be implemented on the platform.
Viinikkala et al. (2005) have discussed the use of
value added Web services to enhance O&M
information management. In their approach, they
propose a service framework using Web Service
technology to address the challenges of information
flow and integrated plant operations. The value
added services are categorized based on complexity
and the added value of O&M information.
Zeeb at al. (2008) have considered Web services
for providing well-defined interfaces to support
communication in maintenance procedures. In their
approach on the LOMS platform, a maintenance
procedure is modeled as a workflow and Web
service based technologies with service templates
used for creating services. However, BPEL was not
used due to DPWS compliancy issues.
Hutchinson et al. (2007) state that there are, in
general, significant challenges when trying to
integrate services from different sources and
migrating existing systems to service oriented
computing. In the research, they identified typical
architectural mismatches and suggest that a
migration process addressing these challenges
should include detailed business process analysis as
well as architectural analysis.
3 DEVELOPMENT APPROACH
BASED ON BPM AND SOA
The approach to be presented emphasizes the role of
business process modeling of services when
developing applications that support the service, e.g.
in maintenance. A business process oriented
approach supports operations driven development
and process models are used to steer the
implementation of respective software applications.
The other important aspect is service-orientation
in the supporting IT infrastructure in order to
facilitate exchange of data required in and produced
as a result of services. Service-oriented architecture
in information systems enables aligning applications
with the business processes and not vice versa.
3.1 Plant Lifecycle Management
Plant lifecycle management stands for all the
information related to a plant throughout its lifecycle
from early design, construction, operation and
maintenance to disposal and recycling. As a concept
it integrates the logical plant to the physical and the
technical model. In practice, this is integration of
schemata and diagrams from e.g. the manufacturing
process design, mechanical models for construction,
e.g. CAD, and information systems such as ERP,
MES and DCS.
The emergence of the so called digital factories
enable extended utilization of information produced
in all lifecycle phases. Ideally, all activities are
defined and the outcome of all tasks is recorded in
the plant wide model. The plant information model
then becomes an almost equally important asset as
the physical plant which it logically represents.
A logical plant information model is beneficial
for plant operations, and especially for services
performed by external service providers. The
challenges of managing mutual information in
distributed service networks can be met with open
systems and standardization to make integration of
BUSINESS PROCESS MODELING AND SOA IN INDUSTRIAL O&M APPLICATION DEVELOPMENT
279

applications easier. For example, allocating
maintenance activities to assets, such as devices and
equipment in a plant, can be specified if open and
standardized access to the plant model can be
provided for the service providers.
Despite standardization efforts, such as ISO
15926-1 (2004) and its other parts, there is still
diversity in plant information model descriptions.
Therefore, service integration platforms should not
restrict models or data, and instead provide tools that
are versatile for developing the needed applications.
3.2 Business Process Modeling of
Services with BPMN
Managing activities in a distributed service network
requires careful specification of responsibilities and
interactions between participants. Traditionally
modeling of business processes has been used as a
tool to represent processes of an enterprise for
analysis and process improvement. More recently,
partly due to the process thinking paradigm, BPM
has been used to present interweaved activities in
complex business scenarios.
In contrast to many other approaches for
application development in O&M and related
services, lean management of processes and
flexibility required in business environments of
today should, according to our opinion, be
emphasized. Applications that support O&M
activities in services are required in order to enable
efficient operation, and BPM can assist in specifying
required interactions and information flows.
Especially when operations are distributed and tasks
of different service providers are interweaved, the
benefits of process modeling are evident.
First of all, business process modeling is
documentation and understanding of how operations
are being carried out. When we understand how our
processes work, identify who is involved and how
activities flow we can improve our way of working.
Often it is easy to improve efficiency by removing
manual work with automated information systems.
Modeling of business processes using a notation
providing mutual understanding enables
specification of operations and tasks on a suitable
level for the participants involved. BPM can be used
to clarify the responsibilities and workflows in
outsourced maintenance service scenarios. In
addition to tasks, also information flows between
participants and required resources can be described.
Business Process Modeling Notation (BPMN) is
a standard notation for describing business process
interactions and information flow (OMG, 2009). The
graphical notation intends to be both intuitive and
expressive for modeling of various types of
processes. The diagrams consist of pools
representing participants and inside them tasks
represent activities. Tasks are connected by arrows
indicating process flows and may include loops,
sequences, conditions and alternative paths. Tasks
connected between pools represent communication
or events that in turn can affect the process.
3.3 Executable Processes - WS-BPEL
BPM diagrams present processes on an abstract level
and are primarily intended for human interpretation.
Efficiency of business processes can be improved by
automation and implementation as executable
processes integrating various information systems.
For operation and maintenance services it
provides the following benefits:
• Automation of laborious processes and
routine activities that can be defined clearly
• Processes can be executed periodically or
started as a result of an event
• Activities can be automatically documented
for future reference and analysis
Executable processes can be controlled by
humans, include decision making based on
conditions or utilize external services or composite
processes that provide the required control and
decision making. Although automation is preferred,
human participation in some form may often be
required to perform the correct operations.
In addition to programmatic system integration
approaches, there are platforms and technologies for
creating executable processes. For example, WS-
BPEL (OASIS, 2007) as well as many commercial
proprietary solutions provide means for managing
workflow between systems with e.g. Web services.
By orchestrating Web services, for example, it is
possible to create programmatically executable
business processes by composition of services.
WS-BPEL is a standardized, XML-based
language for describing service orchestrations by
linking Web service interactions and message flows.
A WS-BPEL process can be composed inside
another process and the process itself can be invoked
as any other Web Service due to its standard WSDL
interface description.
The executable processes can be implemented
manually based on the abstract process descriptions.
However, with the use of standard notations
including sufficient data in the models, much of the
work can be eased, and executable processes can
even be automatically generated. Some
ICEIS 2011 - 13th International Conference on Enterprise Information Systems
280

transformations, for example mapping of certain
BPMN structures to corresponding WS-BPEL
elements, can be done automatically (see figure 1).
Figure 1: Abstract BPM diagrams are used as a basis for
creating executable processes based on Web Service
technology.
Advanced tools for specifying executable
processes, e.g. in WS-BPEL, not only take
implementing processes from a programming
construct level to a more expressive level, but also
bring reconfigurability to the process management.
WS-BPEL processes can be used to invoke services,
control execution flow and data management using a
declarative approach. Reconfiguration is especially
useful in O&M business processes when the
business environment changes and processes need
updating. For example, a maintenance service
provider integrating a new service to plant
operations might only need small modifications to
an existing business process template, e.g. new data
mappings and information routing.
Integration of systems requires well-defined
interfaces and available services. Additionally, it is
also important to have the data accessible in a
machine-interpretable format preferably based on
acknowledged standards. Ideally, the specification of
automatic business processes mitigates to composing
reusable service units and connecting information
flows into executable processes.
3.4 SOA and XML Based Technologies
to Facilitate Integration
In distributed environments, such as operation and
maintenance of industrial plants, there are many
partners involved with various information systems
that would require integration in order to fully
support the use of external services. When required
information in maintenance tasks is transferred
manually, it is often lost for future reference and
analysis. If integration was flexible and systems
were designed open, information exchange between
participants could be significantly improved.
SOA characterizes assets as service components
that for the O&M application development domain
offer the following benefits:
• Application development in a business
process driven manner
• Flexible applications suitable for dynamic
business environments
• A foundation for integrating and managing
data from various applications and sources,
both horizontally and vertically
• Integration of operations and services of
business partners and service providers
SOA provides many advantages of which
especially abstraction, reusability and composability
are of interest in O&M application development.
Abstraction of the inner logic of a service promotes
contract based development of interfaces, i.e.
standardization, and obliges interactions to focus on
information flows. Reuse of service components or
service encapsulated legacy systems justify
expensive development efforts and provides means
to integrate systems also in the future.
System integrations are challenging due to
heterogeneity in both technologies and data
representations. Modern information systems such
as many enterprise resource planning and
maintenance information systems offer integration
options using Web service and XML technologies.
Where legacy systems are needed, migrations to
SOA using wrappers can be used to expose data and
functionality as services. However, requirements for
implementing and integrating services are not
limited to data access, but more importantly to its
format and semantics. As different standards and
practices are used, it is important for the application
development method to support management of
various contents with, for example, middleware or
XML transformations such as XSL and XQuery.
When information systems involved in the
process are available as services, they can be
programmatically invoked, and a process flow
between activities can be established. The
supporting message processing and transformations
required can also be included and controlled e.g.
from WS-BPEL orchestrations. Human interaction
can be implemented through input in existing tools
or new interfaces for acknowledgment, acceptance
or denial of actions in service processes.
3.5 Service Composition
Service based operation is composition by nature.
From a plant operations perspective its main process
is production. The main process consists of sub
BUSINESS PROCESS MODELING AND SOA IN INDUSTRIAL O&M APPLICATION DEVELOPMENT
281

processes and is supported with activities that are
related to the core functions. Similar to composing
operational services, such as maintenance work, also
the related processes in information systems can be
composed to support the provided service.
Figure 2 illustrates the outline of business
processes, service interfaces and applications. The
top level in the figure models actual business
processes, i.e. the interactions and tasks performed
in daily operations on an abstract level of detail.
In order to create applications that either fully
automate or support O&M processes, an approach
based on SOA can be applied. The orchestration
service layer represents an implemented, executable
business process supporting and corresponding to
the actual maintenance work processes. The
orchestration is a composition controller that links
services into workflows through a process service
model. An orchestration can therefore also link other
orchestrations into compositions.
The service interface layer can be divided, in
addition to orchestration services, to business
services and application services. Business services
represent business logic and business models of
maintenance service activities whereas application
services provide reusable functions for processing
data in various application environments. In addition
to these, utility and adapter services are typically
required to support the service compositions.

Business
service layer
Application
service layer
Application
layer
Orchestration
service layer
Service
interface
layer
Business
process layer
Figure 2: Vertical alignment of business processes, service
compositions, services and applications. Erl (2006).
The composition of services based on business
processes in a top-down manner enables
development of applications that support actual
business processes. When systems and applications
are seen as services, process compositions can be
created in a flexible manner satisfying the
requirements of changing business demands in
O&M services. The layered approach also enables
service providers to implement their processes in
their own way, and in turn utilize other service
providers while still maintaining interoperability.
4 EXPERIMENTAL RESULTS
In this study, two business processes were created
using the chosen technologies. The first example is a
condition monitoring process that collects condition
information from devices. The second process is an
environmental footprint estimator for approximating
the environmental load a device causes during its
lifetime. The business processes can be used
together; if the condition monitoring process
indicates a need for a device replacement, different
device candidates can be compared using the
footprint estimator.
4.1 Development and Execution
Environment
The software used to develop and run executable
business processes in this study is Intalio BPMS. It
provides tools for modeling business processes and
their interaction with Web services in BPMN. Intalio
BPMS can partially convert BPMN diagrams to WS-
BPEL and execute the resulting processes. There is
both a free and a commercial edition available.
Intalio BPMS has been used for managing
clinical workflows (Nugrahanto & Morrison, 2008)
and modeling of banking related business processes
(Mpardis & Kotsilieris, 2010). Martin & Bagnoud
(2009) used it in a case in which the interoperation
of different business process platforms was tested.
Jallow et al. (2010) introduced a requirements
change management system for building industry.
In this study, BPMN is used to model the
business processes. Based on these models service
orchestrations are then created as executable WS-
BPEL processes. WS-BEPL interacts with services
using WSDL documents describing both SOAP and
HTTP interfaces, including also REST resources.
4.2 Condition Monitoring
The condition monitoring business process retrieves
the condition information from the devices of one
manufacturer. There are three main problems. First,
there has to be a means to find out which devices
have been supplied by the manufacturer. Second,
there has to be a way to find a device so that its
condition information can be retrieved (the device
network is dynamic so it can change any time).
ICEIS 2011 - 13th International Conference on Enterprise Information Systems
282

Third, there has to be a means to integrate the
devices to business processes.
The first problem is solved by using a plant
information model. It can provide the serial numbers
of the devices of a manufacturer. The second
problem is solved by using a service to provide the
endpoints of devices according to their serial
numbers. The third problem, integration, is solved
by using a technology called DPWS (Devices Profile
for Web Services; Nixon et al., 2009).
The architecture of the condition monitoring
service is presented in figure 3. Squares represent
DPWS devices, circles are business processes and
the rectangle represents the plant information model.
GetConditionInfoBySerialNr
GetConditionServiceEndpoints
GetEndpoints
(DPWS)
GetConditionInfo
(DPWS)
GetConditionInfo
Plant information
model
ConditionInfoOfManufacturersDevices
GetSerialNrsOfManufacturer
Figure 3: Composition of hierarchical processes in the
condition monitoring example.
There are two types of DPWS services in the
hierarchy. GetEndpoints takes device serial numbers
as its input and finds the endpoints of the
corresponding devices. GetConditionInfo provides
the actual condition information for devices.
There is a total of five business processes. When
ConditionInfoOfManufacturersDevices is called, it
calls GetSerialNrsOfManufacturer that uses the
plant information model to retrieve the serial
numbers of the devices delivered by the chosen
manufacturer. Then, GetConditionInfoBySerialNr is
called. It calls GetConditionServiceEndPoints to
retrieve device endpoints according to device serial
numbers. Now that device endpoints are known,
GetConditionInfoBySerialNr retrieves condition
information from the devices by calling
GetConditionInfo. Finally, the return value is
returned to ConditionInfoOfManufacturersDevices
which shall further return it to its caller.
Figure 4 demonstrates the use of BPMN and
describes the flow of the GetConditionInfo business
process. The business process retrieves the condition
information of each device given in its input.
Depending on the input of the business process,
either the condition information of all the devices or
only the condition information of devices with
problems is assigned to the return value. To achieve
this, conditional execution paths are used.
Figure 4: BPMN presentation of the GetConditionInfo sub
process that requests and processes data from devices.
4.3 Environmental Footprint Estimator
The environmental footprint estimator estimates the
emissions output of devices during their lifetime
based on consumption of specific resources. For
example, comparing consumption of electricity in
alternative devices can give an estimate of emissions
that can be taken into account in decision making.
As there may be hundreds of emissions caused by a
production process, the relevant resources consumed
are chosen by the user when the business process is
started. The emissions from only one resource are
calculated at a time; if a device consumes more than
one resource, the business process is run once for
each resource. To enable the comparison of different
devices, the business process can process several
consumption amounts at a time.
A public European Union based environmental
database (ELCD, 2010) is used to provide
information about the emissions caused in resource
production. If the duration of device life and its
consumption of resources are known, it is possible to
estimate the footprint of the equipment or device by
using reference data from the database.
The architecture of the business process is
presented in figure 5. It is a hierarchy that consists of
business processes (circles) and environmental data
resources (squares). The environmental data is
provided in XML format, which facilitates
integration to WS-BPEL business processes.
There are four XML document resources in the
hierarchy. Process data sets provide the reference
amounts of emissions caused by the production of
one resource. To retrieve the units of emissions,
three additional data sets must be retrieved for each
emission: flow, flow property and unit group.
BUSINESS PROCESS MODELING AND SOA IN INDUSTRIAL O&M APPLICATION DEVELOPMENT
283

Figure 5: Composition of processes as services that are
used in gathering environmental footprint estimation data.
There are five business processes in the
hierarchy. GetFootprints calls GetSubstance-
ExchangesWithUnits to retrieve the reference
amounts and the reference units of the emissions of
one production process. To perform that,
GetSubstanceExchangesWithUnits uses two other
business processes. GetSubstanceExchanges
retrieves the reference amounts of emissions from a
process data set, and GetFlowProperties retrieves
the related reference unit by utilizing flow data sets,
flow property data sets and unit group data sets.
Once the reference emissions are known, there may
be a need for unit conversions by ConvertUnit.
4.4 Future Refinement and Research
The examples illustrate how a plant information
model, including engineering data from design, can
be utilized in maintenance services. Because the
condition monitoring process is fully automatic, it
can be executed periodically and trigger events of
assessing the maintenance need or replacement of a
device. The environmental footprint estimator
process, on the other hand, is invoked manually and
can be used to gather data to support decision
making e.g. in choosing new equipment.
In order to improve the examples the condition
monitoring process could, either automatically or by
human operator acceptance, invoke processes such
as device inspections or ordering of maintenance
services. The devices being monitored might also
have different maintenance service providers
requiring either separate processes or additional
logic in looking up the service provider.
The condition monitoring example could also be
upgraded to use a standard message exchange
format. A good candidate for this would be the Open
Systems Architecture for Condition-Based
Maintenance (OSA-CBM; MIMOSA, 2006) that is
intended for transferring information in a condition-
based maintenance system.
Regarding processes in general, and especially
automation and event-driven execution, the WS-
Notification standard family (Graham et al., 2004) is
interesting. WS-Notification provides a topic-based
publish/subscribe pattern for registration to events of
interest that could be used to implement event-
driven process execution. Beneficial to service
compositions in real business environments and
characteristic to the pattern is dynamic registration
to the data source and separate messages to
registered subscribers.
In the long run, the emergence of Semantic Web
technologies opens up new possibilities especially in
classification of information and applying reasoning
to further automate processing and workflow. Also
as further standardization in the O&M domain is
needed the importance of semantics is increasing.
5 CONCLUSIONS
In efforts to make industrial plant operations more
efficient, maintenance among other operations has
been outsourced. Integration of mutually valuable
information could improve service collaboration by
giving the service provider more insight in providing
the service and the customer in receiving it.
Especially in distributed service networks,
information on maintenance operations and
maintenance history is shared inefficiently partly
due to laborious integrations that would be required.
In our opinion, application development should take
business processes and information exchange
requirements better into consideration when
designing new applications and integrating
information systems. An operation driven O&M
application development approach suitable for
services can be achieved with business process
modeling and implementation based on SOA. The
approach presented is a top-down development
approach that enables business demands and service
operations to steer the application development. For
technologies and migration of existing information
systems, a standards based service-oriented
approach is suggested for providing service
interfaces for data exchange and service
composition.
The experiments indicate that business process
modeling with BPMN and integration of O&M
application services using WS-BPEL provides
means for composing services efficiently while
maintaining the required flexibility and
reconfigurability of services. The approach probably
also saves time compared to traditional
ICEIS 2011 - 13th International Conference on Enterprise Information Systems
284

programming approaches, and promotes
standardization of data, application interfaces and
operation practices.
An interesting future research topic from a
services point of view is management of service
collaborations where more details and participants
are involved. This, as well as previously presented
challenges, call for standardization to aid in
integration of services and information systems.
Also advances in the use of intelligent agents,
utilization of technologies of the Semantic Web, and
the increased capabilities of plant floor devices
provide interesting research topics in the future.
REFERENCES
Auramo, J., Tanskanen, K., Småros, J., 2004. Increasing
Operational Efficiency Through Improved Customer
Service: Process Maintenance Case, International
Journal of Logistics: Research and Applications,
Volume 7, Issue 3, September 2004, pages 167 - 180.
Baines, T, Lightfoot, H, Evans, S, Neely, A, Greenough,
R, Peppard, J, Roy, R, Shehab, E, Braganza, A,
Tiwari, A, Alcock, J, Angus, J, Bastl, M, Cousens, A,
Irving, P, Johnson, M, Kingston, J, Lockett, H,
Martinez, V, Michele, P, Tranfield, D, Walton, I,
Wilson, H, 2007. State-of-the-art in product-service
systems, Proceedings of the Institution of Mechanical
Engineers, Part B: Journal of Engineering
Manufacture, Volume 221, Issue 10, 2007, pages
1543-1552. doi: 10.1243/09544054JEM858
Bangemann, T., Rebeuf, X., Reboul, D., Schulze, A.,
Szymanski, J., Thomesse, J-P, Thron M., Zerhouni, N.
PROTEUS--Creating distributed maintenance systems
through an integration platform, Computers in
Industry, Volume 57, Issue 6, E-maintenance Special
Issue, August 2006, Pages 539-551, ISSN 0166-3615,
doi: 10.1016/j.compind.2006.02.018.
ELCD, 2010. ELCD core database version II. European
Commission, DG Joint Research Centre - Institute for
Environment and Sustainability. Available:
http://lca.jrc.ec.europa.eu/lcainfohub/datasetArea.vm
Erl, T., 2006, Service-Oriented Architecture: Concepts,
Technology & Design. Prentice Hall, Pearson PTR,
760 p. ISBN: 0131858580
Graham, S., Niblett P., Chappell, D., Lewis, A.,
Nagaratnam, N., Parikh, J., Patil, S., Samdarshi, S.,
Sedukhin, I., Snelling, D., Tuecke, S., Vambenepe,
W., Weihl, B., Publish-Subscribe Notification for Web
services (WS-Notification whitepaper), version 1.0,
2004.
Hausladen, I, Bechheim, C., E-maintenance platform as a
basis for business process integration, 2nd IEEE
International Conference on Industrial Informatics,
2004. INDIN '04. pp. 46-51, 26-26 June 2004.
Hutchinson, J., Kotonya, G., Walkerdine, J., Sawyer, P.,
Dobson, G., Onditi, V., 2007. The Challenge of
Evolving Existing Systems to Service-Oriented
Architectures, 5th IEEE International Conference on
Industrial Informatics. INDIN’07. Vol. 2, pp.773-778,
23-27 June 2007.
Intalio BPMS. Available: http://www.intalio.com/bpms
ISO, 2004. 15926-1 Industrial automation systems and
integration — Integration of life-cycle data for process
plants including oil and gas production facilities —
Part 1: Overview and fundamental principles. ISO
Standards.
Jallow, A. K., Demian, P., Baldwin, A. N. & Anumba, C.
J., 2010. An Integrated Requirements Management
System for Construction Projects. International
Society for Computing in Civil and Building
Engineering (ISCCBE).
Levery, M., 1998. Outsourcing Maintenance - A Question
of Strategy, Engineering Management Journal , vol.8,
no.1, pp.34-40, Feb 1998.
Martin, S. & Bagnoud, L., 2009. A Case Study of
Business Process Interoperability in a Logistics
Supply Chain Environment. In Lecture Notes in
Business Information Processing, 2009, Volume 17,
Part 8. Springer.
MIMOSA, 2006. Open Systems Architecture for
Condition-based Maintenance (OSA-CBM) Primer,
August 2006, Machinery Information Management
Open Standards Alliance.
Mpardis, G. & Kotsilieris, T., 2010. Bank Loan Processes
Modelling Using BPMN. In DESE 2010,
Developments in E-systems Engineering. IEEE.
Nixon, T., Regnier, A., Driscoll, D. & Mensch, A., 2009.
Devices Profile for Web Services Version 1.1. OASIS.
Nugrahanto, S. & Morrison, I., 2008. A Design Science
Approach to Modelling and Facilitating Clinical
Workflow and Decision Making. In HIC 2008,
Australia's Health Informatics Conference. Health
Informatics Society of Australia Ltd (HISA).
OASIS, 2007. Web Services Business Process Execution
Language (WS-BPEL) Version 2.0, OASIS Standard.
OMG, 2009. Business Process Model and Notation
(BPMN) Version 1.2. OMG specification.
Szymanski, J., Bangemann, T., Thron, M., Thomesse, J.-
P., Reboeuf, X., Lang, C., Garcia, E., PROTEUS - a
European initiative for e-maintenance platform
development, 9th IEEE International Conference on
Emerging Technologies and Factory Automation.
ETFA'03. Vol. 2, pp. 415- 420, 16-19 Sept. 2003.
Viinikkala, M., Jaakkola, V-P., Kuikka, S., Value Added
Web Services for Industrial Operations and
Maintenance. 7th International Conference on
Enterprise Information Systems, 2005, ICEIS’05. pp.
361-365, 25-28 May 2005.
Zeeb, E., Priiter, S., Golatowski, F., Berger, F., A Context
Aware Service-Oriented Maintenance System for the
B2B Sector, 22nd International Conference on
Advanced Information Networking and Applications -
Workshops, 2008. AINAW 2008. pp.1381-1386, 25-28
March 2008.
BUSINESS PROCESS MODELING AND SOA IN INDUSTRIAL O&M APPLICATION DEVELOPMENT
285