DSV Blog

Today Paul, Martin Henkel, Birger, and myself submitted a project proposal to the Swedish Research Council.

Title: Service Oriented Systems Architecture for Knowledge Intensive Processes

Abstract: Software support for well structured business processes is today
provided through workflow technology and process management tools. Tailored to
support well structured processes, these tools provide only poor support for loosely
structured work activities such as knowledge intensive processes. Knowledge intensive
processes are heavily reliant on professional knowledge, deal with large amounts of data,
and tasks that can be redone several times. The purpose of the project is to bring together
state-of-the-art research in business process management systems and knowledge management
to design an open service oriented architecture for supporting knowledge intensive processes.
This architecture will enable flexible process enactment, configurable and context-aware user
interfaces, and service based task support.

Project Proposal
Popular Description (in Swedish)

Business Process Reengineering, Business Process Management, Workflow Management Systems and Process Aware Information Systems are a few concepts circulating around during the last decade and denoting a special kind of information systems or activities relevant for their development and utilization. These concepts address the intersection of the Information Systems and Business Management areas. As in any young discipline, this area suffers from a terminological blur, i.e. different people define the concepts above differently and sometimes the definitions are overlapping. Therefore, we start with a brief overview of the concepts we have adopted.

Terminology

During mid 90 the ideas of Business Process Rengineering (BPR) from the U.S. gained increased attention. BPR denoted an organized effort of a company to streamline its processes. It was defined by Hammer and Champy [2]

“… the fundamental rethinking and radical redesign of business processes to achieve dramatic improvements in critical contemporary measures of performance, such as cost, quality, service, and speed.”

BPR can be carried out at organizational level only. However, as most companies today are dependent on IT-systems that also influence the way of working, a BPR effort usually result in new needs and requirements put on the underlying IT support. If satisfied, these needs can significantly increase the benefit of the undergone business reengineering. One of the main critiques towards BPR regards its focus on a major solitary change, which can not sufficiently capture the continuous dynamics and evolution of an enterprise. As a result of this, in late 90s and early 00s the attention shifted from BPR to Business Process Management (BPM). Business Process Management includes, according to Weske [4],

“.. concepts, methods, and techniques to support the design, administration, configuration, enactment, and analysis of business processes.”

A Business Process is the ordering of a set of activities in an organization for achieving certain goals. A case is a specific instantiation of a business process. The administration of cases, hence business process execution, can be supported through software systems. A system specifically developed for this purposed is called a workflow management system (WFMS). A Workflow is defined by the Workflow Management Coalition (WfMC) as the

“.. automation of a business process, in whole or part, during which documents, information or tasks are passed from one participant to another for actions according to a set of procedural rules” [5]

Because of present confusion between the terms (business) process and workflow, we will refer to a workflow as the technical implementation of a (business) process in a Workflow Management System. Workflow models, which are usually graphical, are used to specify the process under consideration. A Workflow Management System (WFMS) is a system which can read, interpret, and execute processes by scheduling and distributing work to different agents according to workflow models. More precisely a WFMS is according to the Workflow Management Coalition (WfMC):

“A System that defines, creates and manages the execution of workflow through the use of software, running on one or more workflow engine, which is able to interpret the process definition, interact with workflow participants and, where required, invoke the use of IT tools and applications.” [5]

Figure 1 below shows the Workflow Reference Model by WfMC, defined approximately ten years ago. During these ten years also the term Business Process Management Suits (BPMS) has appeared and been adopted by some of the former WFMS vendors who now market their products as BPMS. Some (e.g. Gartner) claims that BPMS adds to the WFMS (which traditionally were build to support the document routing) the aspect of system connectivity. Indeed, while the development WFMS started thirty years ago as document routing systems, they have grown to exactly workflow management (and not document management) systems (which can also be seen from the ten years old Workflow Reference Model in Figure 1). We consider therefore the term BPMS as just a modern denotation for the traditional WFMS, simply aligned with the BPM trend terminology.

Fig 1 The Workflow Reference Model by WfMC

Recently, in 2005 the term process aware information system was coined. A Process Aware Information Systems (PAIS) is a

“software system that manages and executes the operational processes involving people, application, and/or information sources on the basis of process models” [1].

PAIS incorporates a wider spectrum of tools including WFMS, Case Handling Systems, Groupware (such as project management tools and process-aware collaboration tools), process/services composition etc. The lifecycle of a PAIS is depicted in Figure 2. Interesting to note is that in addition to the WFMS there is a category of tools focusing exclusively on supporting the business process analysis and modeling phase.

Fig 2 The PAIS life cycle (copied from [1])

Languages and Systems

As apparent from the previous section the area of BPM and Workflow system has been developing rapidly during the last decade. New business process modeling and implementation notations have been established and join the existing ones in continues development. Figure 3 chart the mainstream languages in the area. The obvious question is how these languages, which are created with the same purpose, i.e. business process modeling, compare to each other.

Since 2002, in cooperation with A/Prof. Arthur ter Hofstede, BPM Group, Queensland University of Technology (QUT) and Prof. Wil van der Aalst, BPM Center, Technical University of Eindhoven (TUE) and QUT, we have been working on a number of deep analyses and comparisons of BPM languages. The method of work has been to study the languages through one same evaluation framework, The Workflow Patterns framework, developed at QUT, was used for this purpose. The framework was developed through a bottom-up analysis of more than 15 WFMS, during which a number of patterns were extracted and systemized into three categories: control-flow, data, and resource patterns.

Fig 3 Business process modeling languages (copied from [3])

The green shading in Figure 3 show the languages in which analyses we have been participated. (The results are listed below.) The blue shading show the analyses performed by our partners. The red shading show an analysis (with the same workflow patterns framework) performed by other researchers.

• BML
• BPEL4WS
• BPML (WSCI)
• UML 2.0 AD
• BPMN

Recently, our work was extended to with the workflow patterns framework analyse some mainstream workflow management systems. The suitability of the following open-source offerings for BPM was studied:

• jBPM
• OpenWFE
• Enhydra Shark

Preliminary results from this work are reported here. The report is currently under review by the vendors.

References

[1] M. Dumas, W.M.P van der Aalst and A. ter Hofstede, Process Aware Information Systems: Bridging People and Software through Process Technology, John Wiley & Sons Inc., 2005

[2] M. Hammer and J. Champy, Reengineering the Corporation: A Manifesto for Business Revolution, Harper Business, 1993

[3] 5. M. Josuttis, SOA in Practice, OReilly, 2007

[4] M. Weske, Business Process Management: Concepts, Languages, Architerctures, Springer, 2007

[5] The Workflow Management Coalition, “Terminology and Glossary”, WFMC-TC-1011, http://www.wfmc.org/standards/docs/TC-1011_term_glossary_v3.pdf, 1999.

In December 2007 we established a contact with Tibco, Sweden.

On 3rd of December, we had a meeting with:

Mark Rattley - Vice President UKI, Nordics & MEA Region,
Jan Hygstedt - Country Manager Nordics, and
Mårten Nilsson - Solution Architect;

all representatives from Tibco. From DSV Prof. Paul Johannesson, Birger Anderson, Michael Persson and myself attended the meeting. I gave a brief presentation of DSV and Paul introduced Syslab. Mark introduced the company and gave an overview of Tibco’s University Relations Program.

I contacted Tibco, because I am interested in investigating their workflow management system called iProcess Suite for teaching purposes. After the meeting I was invited to join two courses on the product: BPM200 Essentials of TIBCO BPM 10.x and BPM205 Designing with the TIBCO iProcess tm Modeler 10.x., which I did during week 50. Bram van Tol, from the Netherlands, was course leader. He provided a high quality training on the tool. I would like to thank Mark, Jan and Mårten for providing me with the opportunity to join the course. I am looking forward to a continued co-operation.

During the period 20th-26th of September 2007, Nick Russell from TUE visited us at DSV. Nick wrote his PhD thesis within the Business Process Management group (BPM) at the Faculty of Information Technology at the Queensland University of Technology (QUT) in Brisbane, Australia, where he worked under the supervision of A/Prof. Arthur ter Hofstede and Prof. Wil van der Aalst. Nick’s research interest is in the area of suitability of workflow languages for process modeling. He has done major contributions within the Workflow Pattern Initiative with his work on data, resource and exception handling patterns, as well as extensions and formalization of the control-flow patterns. He has also been involved in the work with newYAWL. Nick moved recently from QUT to TUE, where he will do his postdoc in the group of Prof. Wil van der Aalst.

Nick initiated his visit at DSV with giving a seminar “An Introduction to the Workflow Patterns and newYAWL” on the 20th of September 2007. During his stay we worked on finalizing the evaluation of two open source workflow management systems – a work which was initiated earlier during the year and which also involves Arthur and Wil.

Last week A/Prof Arthur ter Hofstede from QUT visited us at DSV. Arthur is co-leader of the Business Process Management group (BPM) at the Faculty of Information
Technology at the Queensland University of
Technology (QUT) in Brisbane,
Australia.
Together with Prof. Wil van der Aalst from TUE and QUT, he is leading the work
of the Workflow Patterns Initiative and the work on development and
implementation of the language YAWL. I started cooperating with Arthur’s research
group during my postdoc studies and I have visited QUT a numerous
times since 2001 (latest in May this year). Our cooperation is on suitability analysis of
languages and standards for Business Process Management. Earlier results
include analyses of BML, BPEL4WS, UMN2.0 AD and BPMN. During Arthur’s visit we continued
working on an analysis of two open source workflow management systems.

Today 23 May 2007, I am flying back to Stockholm after a four weeks trip to Brisbane, Australia and
an informal visit to the BPM group at QUT. During my stay in Brisbane I cooperated with A/Prof. Arthur ter
Hofstede and Nick Russell on a deep analysis of OpenWFE, which is one of the
mainstream open-source workflow management systems (WFMSs). We also briefly discussed
an initial analysis of jBPM which is another open-source WFMS.

I used the opportunity to consult Lachlan Aldred, who is the main developer of the YAWL Engine, about custom YAWL services and the development of a database YAWL service.

During these weeks, I also had the opportunity to attend the final PhD seminars of Nick Russell and Michael Adams. Nick presented and defended his thesis on: (i) extensions to the Workflow Patterns framework with
Data, Resource and Exception handling patterns as well as a revision and
formalisation of the original Control-flow patterns; and (ii) a further
development of the language YAWL, which was developed based on the Workflow
Patterns, into newYAWL – a version capturing the extensions introduced into the
patterns framework.

Michael presented his work on exception handling and dynamic workflow. In his thesis he proposes a framework enabling the on-the-fly change of executing business processes. The framework takes some ideas from Activity
Theory, which is a theory about work organisation developed in the USSR during the
1920s, and uses them to develop guiding principles for workflow management
systems. It uses the concept of Worklet, which is a small independent (sub)
process. During execution of a task, the appropriate worklet is selected based
on context information organised in the form of so-called Ripple Down Rules. As
time goes by, these rules may be extended and also new worklets may be added.
In addition, the same framework is used for exception handling where Exlets
form exception handling processes using concepts developed by Nick Russell that
may be invoked depending on the type of exception occurring. The whole
framework has been implemented in YAWL but is general enough so that it could
be transferred to any workflow management system offering the required
interfaces.

Today 23 May 2007, I am flying back to Stockholm after a four weeks trip to Brisbane, Australia and
an informal visit to the BPM group at QUT. During my stay in Brisbane I cooperated with A/Prof. Arthur ter
Hofstede and Nick Russell on a deep analysis of OpenWFE, which is one of the
mainstream open-source workflow management systems (WFMSs). We also briefly discussed
an initial analysis of jBPM which is another open-source WFMS.

I used the opportunity to consult Lachlan Aldred, who is the main developer of the YAWL Engine, about custom YAWL services and the development of a database YAWL service.

During these weeks, I also had the opportunity to attend the final PhD seminars of Nick Russell and Michael Adams. Nick presented and defended his thesis on: (i) extensions to the Workflow Patterns framework with
Data, Resource and Exception handling patterns as well as a revision and
formalisation of the original Control-flow patterns; and (ii) a further
development of the language YAWL, which was developed based on the Workflow
Patterns, into newYAWL – a version capturing the extensions introduced into the
patterns framework.

Michael presented his work on exception handling and dynamic workflow. In his thesis he proposes a framework enabling the on-the-fly change of executing business processes. The framework takes some ideas from Activity
Theory, which is a theory about work organisation developed in the USSR during the
1920s, and uses them to develop guiding principles for workflow management
systems. It uses the concept of Worklet, which is a small independent (sub)
process. During execution of a task, the appropriate worklet is selected based
on context information organised in the form of so-called Ripple Down Rules. As
time goes by, these rules may be extended and also new worklets may be added.
In addition, the same framework is used for exception handling where Exlets
form exception handling processes using concepts developed by Nick Russell that
may be invoked depending on the type of exception occurring. The whole
framework has been implemented in YAWL but is general enough so that it could
be transferred to any workflow management system offering the required
interfaces.

Today Paul, Jelena and myself submitted a project proposal to the Swedish Research Council.

Title: A Universal Repository of Process Models

Abstract: The rapid development of Internet during the last decade has
supported enterprises in building novel infrastructures, setting up
virtual organisations, and operating in larger geographical spaces.
To manage this new environment, enterprises need to align their IT
infrastructures to the business processes. Therefore, the interest
in business process management using Process Aware Information
Systems (PAIS) has been rapidly increasing. Solutions implemented in
PAISs are often complex and time-consuming to develop. One way to
address this problem is to utilize repositories of reusable process
models. However, while repositories have proved to be successful
within object-oriented and component-based development, similar
success has not yet been achieved in the area of PAIS. This is
because we still lack the critical mass of process models within a
single repository and we lack transparency between different
repositories. The main goal of the project is, therefore, to design
the architecture of a universal process repository, i.e. a
repository that is independent of process modelling languages,
comprises a large number of existing process repositories, and is
open for change and growth by any potential user. The long term goal
of the project is to lay the foundations for a Business Process
Management Wikipedia, which will become a universal knowledge
resource on process models that can be used by researchers for
empirical investigations in the business process management area.

Project Description
Popular Description (in Swedish)