A Thing Called ”Fluid Process” –
Beyond Rigidity in Business Process Support
Manfred Reichert
Institute of Databases and Information Systems, University of Ulm, Germany
Abstract: This keynote reports on a new class of processes - so called fluid processes
- whose ”engineering” and ”use” is indistinguishable. Fluid processes are continually
being adapted and reformed to fit the actual needs and constraints of the situation in
hand and to fulfill the overall goals of the involved actors in the best possible way.
We present a detailed review of challenges and techniques that exist for the support of
fluid processes. We give insights into their nature, discuss fundamental challenges to
be tackled, summarize basic technologies enabling fluid processes at the information
system level, and describe advanced applications of fluid processes.
1 Motivation
The economic success of an enterprise increasingly depends on its ability to flexibly sup-
port its business processes by information systems and to react to changes in its environ-
ment in a quick and flexible way [RRMD09, WRRMW09]. However, today’s process
engineering technologies and tools are ill equipped to meet this challenge because of their
inherent brittleness and inflexibility [WRRM08, RD98]. The current generation of BPM
tools [Wes07] implicitly embrace the ”engineer use” dichotomy inherited from traditional
approaches to software and database engineering. This, in turn, is based on the classic
engineering principle that systems are first ”engineered” and then, once deemed fit for
purpose, are ”used” (or ”operated”). Maintenance and evolution activities are not regarded
as part of operation but rather as interruptions to the ”in use” state which temporarily
return the system to the ”being engineered” state. However, in scenarios in which require-
ments come and go on a much more dynamic and ad hoc basis (e.g. healthcare [LR07] or
automotive engineering [MHHR06]), this ”engineer use” strategy is unworkable.
The only feasible way to cope with dynamism is to dissolve the fundamental distinction
between ”engineering” and ”use” and seamlessly merge the whole service and process
lifecycle into a single encompassing framework [WRRMW09, WSR09]. This will lead
to a new class of processes - so called fluid processes - whose ”engineering” and ”use”
is indistinguishable. Such processes are continually being adapted and reformed to fit the
actual needs and constraints of the situation in hand and to fulfill the overall goals of the
involved actors in the best possible way. Since enterprise workflows can be viewed as
special forms of processes, the notion of fluid workflows is also important.
2 Characterization of Fluid Processes
Fluid processes and services should be not confused with ”(self-)adaptive systems” as
recognized by the adaptive systems research community. Fluid processes and services
are adaptive in the sense that they are continually evolving and reshaping to fit the sit-
uation in hand, but unlike classical adaptive systems (as understood in adaptive system
research) they are not expected to do this themselves. On the contrary the whole point of
fluid process is that the adaptation is performed with the help of the human user/engineer
where needed. In other words, in fluid processes, human engineers and users are ”part of
the loop”, and processes use and adaptation are seen as two sides of the same coin. In
this sense, fluid process have more in common with agile software development methods,
which are focused on encouraging human developers to evolve software in as rapid and
effective a way as possible, than adaptive systems which are responsible for all dynamic
adaptation themselves.
Adaptive systems research is still essentially based on the engineer use distinction, the
only difference to traditional approaches is that the systems are given much more intelli-
gence than usual in order to exhibit a wider range of behaviors. This notwithstanding, the
techniques and ideas of adaptive systems research are very relevant to fluid process and
services, because all forms and techniques of adaptation need to be exploited. In fact, the
notion of fluid process and services subsumes standard adaptive systems, and is based on
a more general approach where humans are tightly bound into the loop as well as artificial
intelligence.
Fluid processes also fundamentally change the way in which human stakeholders interact
and collaborate because they dissolve the distinction between process engineers and pro-
cess users. To date, business process support technologies have focused on enhancing and
automating the way in which process users collaborate and interact, but have not signifi-
cantly changed the way in which the processes themselves are engineered (i.e. defined and
maintained). The assumption has been that this is done by IT specialists in a distinct engi-
neering phase with little or no connection to the execution of the processes or the normal
operation of the enterprise. However, with fluid processes this distinction will blur (if not
entirely disappear) and process engineers will also be process users and vice versa. Stated
differently, process engineering will also he regarded as a normal, fluid business process
involving the collaboration of multiple stakeholders. In short, process engineering will
become reflective.
3 Outline of the Keynote
This keynote presents a detailed review of challenges and techniques that exist for the sup-
port of fluid processes [RRMD09]. We give insights into the nature of fluid processes, dis-
cuss fundamental technological challenges to be tackled, give an overview on basic tech-
nologies enabling fluid processes at the information system level, and describe advanced
applications of fluid processes. In this context we revisit some of our research projects
on fluid processes like ADEPT [RD98, RDB03, RRD03], CEOSIS [RMR07, RMR09],
Corepro [MRH08], and Philharmonic Flows [KR09].
References
[KR09] V. K¨
unzle and M. Reichert. Integrating Users in Object-aware Process Management
Systems: Issues and Challenges. In Proc. Business Process Management Workshops
2009. Springer, 2009.
[LR07] Richard Lenz and Manfred Reichert. IT support for healthcare processes - premises,
challenges, perspectives. Data Knowledge Engineering, 61(1):39–58, 2007.
[MHHR06] Dominic M¨
uller, Joachim Herbst, Markus Hammori, and Manfred Reichert. IT Sup-
port for Release Management Processes in the Automotive Industry. In Business
Process Management, LNCS 4102, pages 368–377, 2006.
[MRH08] Dominic M¨
uller, Manfred Reichert, and Joachim Herbst. A New Paradigm for the
Enactment and Dynamic Adaptation of Data-Driven Process Structures. In Proc.
CAiSE’08, LNCS 5074, pages 48–63, 2008.
[RD98] M. Reichert and P. Dadam. ADEPTflex - Supporting Dynamic Changes of Work-
flows Without Losing Control. JIIS, 10(2):93–129, 1998.
[RDB03] M. Reichert, P. Dadam, and T. Bauer. Dealing with Forward and Backward Jumps
in Workflow Management Systems. Software and Systems Modeling, 2(1):37–58,
2003.
[RMR07] S. Rinderle-Ma and M. Reichert. A Formal Framework for Adaptive Access Control
Models. In Journal of Data Semantics, IX, LNCS 4601, pages 82–112, 2007.
[RMR09] S. Rinderle-Ma and M. Reichert. Comprehensive Life Cycle Support for Access
Rules in Information Systems: The CEOSIS Project. Enterprise Information Sys-
tems, 3(3):219–251, 2009.
[RRD03] S. Rinderle, M. Reichert, and P. Dadam. Evaluation Of Correctness Criteria For Dy-
namic Workflow Changes. In W.M.P. v.d. Aalst, A.H.M. ter Hofstede, and M. Weske,
editors, Proc. Int’l Conf. on Business Process Management (BPM’03), LNCS 2678,
pages 41–57, Eindhoven, The Netherlands, June 2003.
[RRMD09] Manfred Reichert, Stefanie Rinderle-Ma, and Peter Dadam. Flexibility in Process-
Aware Information Systems. T. Petri Nets and Other Models of Concurrency, 2:115–
135, 2009.
[Wes07] M. Weske. Business Process Management: Concepts, Languages, Architectures.
Springer, 2007.
[WRRM08] Barbara Weber, Manfred Reichert, and Stefanie Rinderle-Ma. Change patterns and
change support features - Enhancing flexibility in process-aware information sys-
tems. Data Knowledge Engineering, 66(3):438–466, 2008.
[WRRMW09] Barbara Weber, Manfred Reichert, Stefanie Rinderle-Ma, and Werner Wild. Pro-
viding Integrated Life Cycle Support in Process-Aware Information Systems. Int.
Journal of Cooperative Information Systems, 18(1):115–165, 2009.
[WSR09] Barbara Weber, Shazia Wasim Sadiq, and Manfred Reichert. Beyond rigidity - dy-
namic process lifecycle support. Computer Science - R&D, 23(2):47–65, 2009.
