WEB BASED INTEGRATION OF MES AND OPERATIONAL BI
Tom Hänel, Marco Pospiech and Carsten Felden
TU Bergakademie Freiberg, Faculty of Business Administration, Lessingstraße 45, 09599, Freiberg, Germany
Keywords: Web Services, Integration, Flexibility, Decision Support.
Abstract: Internet and web-based technologies increase the informal networks within and across organizations. This
implies an acceleration of changing basic conditions and forces the embracement of innovative technologies
to achieve flexible business processes. However, current research indicates a lack of flexibility in enterprise
applications with negative effects on customer satisfaction and service orientation. Hence, the paper’s
arguments build up on the hypothesis that a web service (WS) based integration approach is beneficial for a
conjoint process oriented and flexible decision support oriented infrastructure. A combination of
Manufacturing Execution Systems (MES) and Operational Business Intelligence (OpBI) is discussed,
because both concepts are promising support of process flexibility. In result an architecture scheme
demonstrates a WS oriented interaction of MES and OpBI functions. In conclusion the effort of information
gathering can be reduced. The use of web-based technologies in context of operational decision making
facilitates a comprehensive synchronization of business processes with flexibility-enhancing effects.
1 INTRODUCTION
The design and control of business processes based
on coherent information is a determining
competitive factor. Internet and web-based
technologies influence the business processes of an
organization as well as its relationships to customers
and suppliers. These increasing informal networks
accelerate the changing basic conditions. Companies
are forced to embrace the emerging web
technologies in order to keep their business
processes flexible by corresponding adjustments.
However, research indicates that organizations are
not able to meet flexibility demands. According to a
survey of the Aberdeen Group, 85 percent of
companies do not provide an adequate flexibility
within their applications (Rodriguez, 2007). This
lack of flexibility implies high cost due to delayed
decisions and low productivity associated with
negative effects in terms of customer satisfaction
and service orientation. With respect to the given
issue, the position paper discusses whether a web
service (WS) based integration approach is
beneficial for a conjoint process oriented and
flexible decision support oriented infrastructure.
The range of conformable concepts allowing
efficient support is large and manifold. Recently,
MES and OpBI came into the discussion promising
both support of process flexibility. These concepts
are integrative approaches for operational process
control and analysis, but they come from different
perspectives – the engineering and the decision
support point of view. A combined approach of
MES and OpBI facilitates overarching analyses to
comprehensively coordinate and optimize processes
so that organizations are able to react fast and
flexible on business occurrences (Hänel and Felden,
2011). Considering the state-of-the-art, the
integration potential of WS to combine MES and
OpBI is not investigated, although especially
service-oriented architectures (SOA) are beneficial
to support flexibility (Erl, 2009). Therefore, we
contribute to the research of WS and decision
support by proposing a discussion about
opportunities and potentials of a web-based
integration regarding to engineering and economic
driven systems.
Chapter 2 sheds light on the integration potential
of OpBI and MES. Furthermore, the ability of WS
and SOA in context of a flexible operational
decision making is discussed. Chapter 3 joins the
separately considered aspects and presents an
architecture scheme for a web based integration of
MES and OpBI. Finally, the paper is summarized to
give conclusions and further research perspectives.
195
Hänel T., Pospiech M. and Felden C..
WEB BASED INTEGRATION OF MES AND OPERATIONAL BI.
DOI: 10.5220/0003959701950200
In Proceedings of the 8th International Conference on Web Information Systems and Technologies (WEBIST-2012), pages 195-200
ISBN: 978-989-8565-08-2
Copyright
c
2012 SCITEPRESS (Science and Technology Publications, Lda.)
2 STATUS QUO
The scope of OpBI and MES is the analysis of
processes to recognize weak points, malfunctions or
business interruptions to improve the management
of business processes continuously and to generate
overarching process information. This chapter
explains the concepts of OpBI and MES as well as
its potential of complementation. Thereafter, WS are
put in context to the demonstrated decision support
to support flexible architecture requirements.
2.1 OpBI and MES
OpBI is aiming for an integrated ensemble of
analytical activities and operational processes
(Eckerson, 2007). The main focus is on reducing
times to collect, report, and analyze data as well as
to take appropriate decisions (White, 2006).
Information regarding the process states during
progress is provided (Bauer and Schmid, 2009). Due
to this reason, OpBI analyzes, controls, and
improves organizational core processes in a fast and
flexible manner (Cunningham, 2005). Thereby, a
Corporate Performance Management is facilitated
(Schwingel, 2010) considering the organization as
closed-loop system, where strategic-tactical and
operational management is interrelated (Golfarelli et
al., 2004). Figure 1 comprises the functionality of
OpBI.
Figure 1: Functions of OpBI.
OpBI provides analytical capabilities in order to
control the organizational value creation in favor of
a continuous improvement of process design and
execution. Thereby, it is a moderator between the
analytical intention and the actual occurrence of a
process. A timely adequate relation between process
performance and states of target achievement get
communicated to the corresponding audience.
A comparable approach to support the decision
making on the shop-floor is the MES (Younus et al.,
2010). It is placed between the layer of Enterprise
Resource Planning (ERP) and the layer of process
execution (ISA, 2000). A vertical integration by
enabling task-oriented compaction, communication
and access of data is realized (Kletti, 2007). The
ERP-system responsible for order and resource
planning communicates desired quantities to the
MES executing a permanently target-performance
comparison and a feedback to ERP. This is to be
done over the full production cycle using real-time
data (MESA, 1997). The MES-architecture consists
of application layer, functional layer, and data
interface layer (Fei, 2010). The data interface layer
enables the access of MES-database on machines
and plants to gather relevant data. The application
layer presents the information generated out of a
MES-database on several clients. Users are able to
send requests and to get desired results. Therefore,
MES are covering eight functions (VDI, 2007):
Figure 2: Functions and architecture of MES.
The functions of OpBI and MES consider an
integrated provision of data as well as its purposive
reporting and analysis. If the MES gets more
complex by including a high number of operational
processes, the similarity to OpBI will grow. This is
associated with a performance lost and limitation of
decision-making in real-time, because an increasing
complexity requires a higher degree of interfaces
(Saenz de Ugarte et al., 2009). Furthermore, the
limited analysis capabilities of the MES (Alpar and
Louis, 2007) question the benefits of such a strategy.
OpBI also forces the decision-making in real-time,
has comprehensive analysis capabilities and
facilitates company-wide process control. But, this
concept is seldom applied in manufacturing
(Eckerson, 2007). A possible reason is that the MES
covers more functions than operational BI, because
it is especially designed for production environments
(Meyer et al., 2009). Hence, OpBI cannot
compensate a MES and vice versa, but they have
beneficial intersections to support enterprise-wide
decision making. Table 1 demonstrates this
functional complementation potential.
WEBIST2012-8thInternationalConferenceonWebInformationSystemsandTechnologies
196
Table 1: Overlapping and complementation possibilities
for MES and OpBI.
OpBI functions
Decision Support
Business relevant information
Data preparation
Data collection
Information description
MES functions
Scheduling
M M
Quality management
M M
Labour Management
M M
Materials management
M M
Management of operating
resources
M M
Data collection
O
Performance analysis
O O
Information management
O O
An integration approach is required to provide a
basis for a conjoint process oriented and flexible
decision support oriented infrastructure using MES
in combination with OpBI. Therefore, the presented
monolithic driven and data warehouse oriented
architecture of the concepts are contradictory.
Flexibility and reduction of complexity is possible
by modularization, which means to structure a
system in semi-autonomous and straightforward
subsystems (Aier and Dogan, 2005). To support
such a modular design Section 2.2 discusses WS to
gain a flexible information architecture.
2.2 Web Services in context of MES
and OpBI
The technology of WS is commonly implemented in
a SOA, representing the current state-of-the-art for
flexible IT structures (Erl, 2009). A SOA can be
understood as „ … a way of designing and
implementing enterprise applications that deals with
the intercommunication of loosely coupled, coarse
grained (business level), reusable artifacts (services).
Determining how to invoke these services should be
through a platform independent service interface …“
(Wilkes and Harby, 2004). It is possible to combine
(orchestration) the existing components depending
on flexibility requirements in consequence of the
loosely coupled services. Thus, business
functionalities are to be abstracted and assembled
according to the process logic. This leads to new
processes, which can be easily redefined or created.
Especially, vendor independent specifications enable
a consistent integration of heterogeneous systems
and an open communication between various
components. A SOA is understood as an abstract
concept, supportable by WS (Erl, 2009). According
to the W3C, a WS is defined as follows:
A Web service is a software system designed
to support interoperable machine-to-machine
interaction over a network. It has an interface
described in a machine-processable format
(specifically WSDL). Other systems interact
with the WS in a manner prescribed by its
description using SOAP-messages, typically
conveyed using HTTP with an XML
serialization in conjunction with other Web-
related standards. (W3C, 2004)
Nowadays, SOAP and WS Description Language
(WSDL) specifications are state-of-the-art based on
Extensible Markup Language (XML) (Melzer,
2010). WSDL describes the WS-interface and
specifies service functionalities, restrictions and
conventions. Furthermore, it refers to an endpoint
that corresponds to a software component. Those
descriptions can be stored in a Universal,
Description, Discovery and Integration (UDDI)
repository. The UDDI is based on XML and serves
as directory service that is responsible for WS
publication and detection. SOAP allows as message
exchange format the discovery, searching, finding
and usage of WS. Thereby, the transport occurs by
an underlying protocol like HTTP or FTP. In this
context, a WS gets described by a WSDL document
and published in a UDDI. A potential service-
consumer will search through SOAP in the UDDI
and retrieves the matching WSDL. Afterwards, the
service consumer will start the communication to the
service provider by the SOAP protocol.
The main concept of a SOA is the encapsulation
of business functionalities. In this context, OpBI and
MES functionalities have to be encapsulated into
WS, so that they act either as service user or service
provider. Performance analysis services provide
analytical business logic and correspond to mature
BI techniques (Martin, 2011). The data collection
services provide the required operational, tactical
and strategic data. They are divided into data access,
transformation and infrastructure services,
corresponding mainly to encapsulate extract,
transform and load (ETL) functionalities (Dinter and
Stroh, 2009). The data access service includes four
basic operations, namely create, read, update and
delete. They allow universal access to all connected
systems and all systems to request demanded
WEBBASEDINTEGRATIONOFMESANDOPERATIONALBI
197
analyses (Vogt et al., 2008). Source and target exist
likewise as service or database e.g. the DWH. If
persistently storing is not required, the data get
promptly accessed by a service from any operational
system on demand and will be used for further
analysis (Dittmar, 2007). Transformation services
represent the encapsulated transformation phase of
ETL. Thus, tasks like aggregation, encoding,
filtering, conversion, join and mapping have be
fulfilled by services. (Martin 2011; Dinter 2008)
There are existing cross-cutting tasks in addition to
the presented services that are realized through
infrastructure services (Dinter and Stroh, 2009)
including data security and data protection features
as well as aspects of data quality, master-data, and
meta-data management (Gordon et al., 2006).
Furthermore, the data collection services provide all
MES relevant data regarding to materials, operators,
machines and processes in real time through specific
decision support services. On this basis, the labour
management service ensures that every shift is
properly organized and recorded. The material
management handles the need driven supply and
disposal with material on schedule, as well as the
management of work in process. Here, quality
information, schedule and material status data are
considered. In this context, the quality management
services sustain the guarantee of the product quality
and the capability of the process, by quality planning
and inspection. The service for management of
operating resources provides a demand-actuated
availability on schedule and functionality of
equipment (machines, operating utilities) in a
historical, current and further view. The scheduling
service plans operational sequences in
manufacturing under consideration of available
resources and capacities. (VDI, 2007)
3 WEB BASED INTEGRATION
PLATFORM
The platform is divided into five components, which
gets successively refined. The data storage
component holds data sources and serves as
fundament for further analyses. Here, process
changes get noticed by the event engine and will be
instantly inserted by orchestrated data collection
services placing the target data into the data source.
In addition, performance analysis and decision
support services accessing the available data through
the same services. The enterprise-wide consolidation
of information exceeds in most of the cases the IT
budget and the manageable complexity (Melzer,
2010). Therefore, it appears adversely to follow the
traditional BI architecture concept, since a DWH
exists no longer in a monolithic pillar, but rather in
an embedded IT infrastructure (Martin, 2011). Thus,
the DWH lose its role as central data storage
component (Dittmar, 2007). The historical data
stored by the DWH are combined and synchronized
with real-time information from the operational
processes. This provides an input flow for decision
support and performance analysis services. There is
the danger that redundant and distributed storage
lead to individual application terminologies ending
in inconsistencies and duplication. Due to this
reason, a central meta data repository is required.
The service platform component reflects all
encapsulated and presented functionalities of OpBI
and MES. In this context, both have to be
implemented in a WS oriented way. Here, OpBI and
MES take the role as service provider or service
consumer. This integrates analytical functions into
processes and operational applications affecting the
operational detailed planning. All services are event-
driven and transfer real time data. Thereby, WS are
endpoints, which react to or produce new events.
(Vogt et al., 2008) In this context, a WS oriented
platform promises the integration of OpBI services
into MES processes and systems, avoids redundant
implementations, enables an unproblematic
integration, improves scalability, and allows an open
communication between all components.
The integration component realizes the
coordination of all services and events in the total
system. The service repository (UDDI) manages and
publishes all service descriptions, which are presented
in form of WSDL. Furthermore, the service repository
selects WS in cooperation with the orchestration
engine. The selection is based on the service
description, which makes the service available for the
orchestration engine. The orchestration executes a
service sequence that can be embedded in other
systems or processes. In addition, the orchestration
engine contains mechanisms that allow state
management, logging and monitoring of sequences.
The orchestration engine is requested by events that
are triggered from the event engine. Here, the event
engine is based on publish-and-subscribe. It receives
and processes events from all components, which are
sent to registered users. Furthermore, the event engine
holds analytical operations.
The analysis is enriched by business process
events (Vogt et al., 2008). Therefore, defined
business rules are required, which are provided by
the corresponding repository. This prevents a
redundant implementation of business rules,
WEBIST2012-8thInternationalConferenceonWebInformationSystemsandTechnologies
198
Figure 3: Web-based integration architecture for OpBI and MES.
increases their reusability and separates the process
logic from the decision logic.
The core of the process component is the
process. Here, the event engine analyzes continually
business process events in real time. The required
process states are received by data collection
services and passed through decision support and
performance analysis WS. Hereby, a timely
adjustment of the manufacturing process in
consequence of unpredictable circumstances can be
established. Complex events or causal, temporal or
spatial relationships are detected by the event engine
and leading to defined reactions in process
execution. Thus, decisions can be automated and
done in shortest time. Consequently, the approach
enables the monitoring of critical process key figures
in MES, OpBI analyses and hereupon tailored
events. A so called closed-loop can be established
that provides reduced decision latency.
The top level is represented by the presentation
component. This enables a flexible integration of
different visualization options. The fundament is
performance analysis and decision support services
which allow graphical representations in portals,
dashboards, or office applications. (Vogt et al.,
2008) The user can be informed about circumstances
by event or send its decision directly to the event
engine.
4 CONCLUSIONS
The position paper discusses an integration platform
to consolidate economic and engineering driven
information across the whole value creation. The
proposed architecture joins the concepts of MES and
OpBI, while their domain specific functions are
conserved. Thereby, during the paper`s discussion
the advantages of WS are evident in favour of a
flexible process oriented decision support. Process
performance indices are enriched by information
from all segments of the value chain as part of
technical-economic analyses. This reduces the effort
of information gathering and provides the ability for
a comprehensive synchronization of business
processes with flexibility-enhancing effects.
Organizations are able to consider changing
requirements of customers and suppliers
contemporary to adjust their production, distribution
and purchase processes. This implies an increased
process transparency so that improvements and
innovations are possible in fields of process and
product design. Next to the introduction of
completely new products, there are for example
opportunities to improve quality characteristics of
products or to tailor the use of resources according
to business needs. Due to the knowledge about the
WEBBASEDINTEGRATIONOFMESANDOPERATIONALBI
199
process performance in all segments of the value
creation, the necessity as well as technical and
economic feasibility of innovation is assessable in
detail. In order to realize the presented benefits of
the integration platform, a practical validation of
these arguments is required. Thereby, the discussed
insights of this position paper serve as guidance and
basis for further research.
REFERENCES
Aier, S., Dogan, T., 2005. Indikatoren zur Bewertung der
Nachhaltigkeit von Unternehmensarchitekturen. In
Wirtschaftsinformatik 2005, 607-626.
Alpar, P., Louis, J.P., 2007. Eine empirische
Untersuchung der Softwareunterstützung bei der
Fertigung und Qualitätssteuerung. Philipps-
Universität Marburg.
Bauer, A., Schmid, T., 2009. Was macht Operational BI
aus? BI-Spektrum, 4 (1), 13-14.
Cunningham, D., 2005. Aligning Business Intelligence
with Business Processes. TDWI, 20, 50-51.
Dinter, B., 2008. Einsatzmöglichkeiten serviceorientierter
Architekturen in der Informationslogistik. In Töpfer, J.
and Winter, R. (Eds.), Active Enterprise Intelligence,
1st Edition. Springer, Berlin.
Dinter, B., Stroh, F. 2009. Design Factors for Service-
oriented Architecture Applied to Analytical
Information Systems: an Explorative Analysis. In
Proceedings of the 17th European Conference on
Information Systems ECIS, Verona.
Dittmar, C., 2007. Latenzzeiten von Business Intelligence-
Systemen. In Gluchowski, P. et al. (Eds.),
Schlaglichter der Wirtschaftsinformatik, 1st Edition.
GUC,Chemnitz, 131-142.
Eckerson, W.W., 2007. Best Practices in Operational BI:
Converging Analytical and Operational Processes.
TDWI Best Practices Report. Renton (WA).
Erl, T., 2009. SOA Design Patterns, 1st Edition, Prentice
Hall, Boston.
Fei, L., 2010. Manufacturing Execution System Design
and Implementation, 2
nd
International Conference on
Education Technology and Computer (ICETC).
Golfarelli, M., Rizzi, S., Cella, I., 2004. Beyond data
warehousing: what's next in business intelligence? In
Proceedings of 7th ACM international workshop on
Data warehousing and OLAP, ACM Press, New York.
Gordon, S. et al., (2006). Service-Oriented Business
Intelligence. In The Architecture Journal. 6, 23-32.
Hänel, T., Felden, C., 2011. Limits or Integration? -
Manufacturing Execution Systems and Operational
Business Intelligence. In Americas Conference on
Information Systems AMCIS 2011, Detroit, USA.
ISA ANSI/ISA-95.00.01-2000, 2000. Enterprise Control
System Integration, Part 1: Models and Terminology,
ISA technical paper.
Martin, W., 2011. Performance Management and
Analytics – Business Intelligence meets Business
Process Management, Retrieved 2011-11-21, from:
http://www.wolfgang-martin-team.net/BI-BPM-
SOA.php.
Melzer, I., 2010. Service-orientierte Architekturen mit
Web Services, 4th Edition, Spektrum Verlag,
Heidelberg.
MESA, 1997. MES Explained: A High Level Vision for
Executives, MESA International – White Paper
Number 6, Pittsburgh.
Meyer, H., Fuchs, F., Thiel K., 2009. Manufacturing
Execution Systems (MES), McGraw Hill, Colum.
Kletti, J., 2007. Manufacturing Execution System – MES,
Springer, Berlin.
Rodriguez, R. A., 2007. Aligning IT to Business
Processes: How BPM is Complementing ERP and
Custom Applications. Aberdeen Group.
Saenz de Ugarte, B., Artiba, A., Pellerin, R., 2009.
Manufacturing execution system – a literature review.
Production Planning and Control, 20, 6, 525-539.
Schwingel, J. (2010). Strategien für Process Performance
Management. BI-Spektrum 4 (3), 14-18.
VDI 5600, 2007. Manufacturing Execution Systems, VDI
Verlag, Duesseldorf, Germany.
Vogt, T. et al., 2008. Business-Intelligence-Konzept auf
Basis einer Event-Driven Service-Oriented
Architecture. In Dinter, B. et al. (Eds.), Synergien
durch Integration und Informationslogistik
Proceedings DW 2008.
Wilkes, S., Harby, J., 2004. A move to drive industry
standardization of SOA. Retrieved 2012-01-21, from:
http://xml.coverpages.org/SOA-
BlueprintsConceptsV05.pdf.
W3C (2004), W3C Web Services Glossary. Retrieved
2012-01-27, from: http://www.w3.org/TR/2004/NOTE-
ws-gloss-20040211.
White, C., 2006. The Next Generation of Business
Intelligence: Operational BI. (White Paper). Retrieved
2012-01-21, from: http://certification. sybase.com.
Younus, M. et al., 2010. MES Development and
Significant Applications in Manufacturing – A
Review, 2
nd
International Conference on Education
Technology and Computer (ICETC).
WEBIST2012-8thInternationalConferenceonWebInformationSystemsandTechnologies
200
