View
3
Download
0
Category
Tags:
Preview:
DESCRIPTION
The Business Process Management (BPM) conference series celebrates its tenth anniversary. This is a nice opportunity to reflect on a decade of BPM research. This talk will describe the history of the conference series through the prism of typical BPM use cases and six key BPM concerns: Process Modeling Languages, Process Enactment Infrastructures, Process Model Analysis, Process Mining, Process Flexibility, and Process Reuse. Although BPM has matured as a research discipline, there are still various important problems that remain open. Moreover, despite the broad interest in BPM, there is significant room for improvement when it comes to the the adoption of state-of-the-art results by software vendors, consultants, and end-users. The BPM discipline should not shy away from the key challenges and set clear targets for the next decade. Keynote BPM 2012: http://bpm2012.ut.ee/ Prof.dr.ir. Wil van der Aalst is a full professor of Information Systems at the Technische Universiteit Eindhoven (TU/e). Currently he is also an adjunct professor at Queensland University of Technology (QUT) working within the BPM group there. His research interests include workflow management, process mining, Petri nets, business process management, process modeling, and process analysis. Wil van der Aalst has published more than 150 journal papers, 17 books (as author or editor), 300 refereed conference/workshop publications, and 50 book chapters. Many of his papers are highly cited (he has an H-index of more than 92 according to Google Scholar, making him the European computer scientist with the highest H-index) and his ideas have influenced researchers, software developers, and standardization committees working on process support. He has been a co-chair of many conferences including the Business Process Management conference, the International Conference on Cooperative Information Systems, the International conference on the Application and Theory of Petri Nets, and the IEEE International Conference on Services Computing. He is also editor/member of the editorial board of several journals, including the Distributed and Parallel Databases, the International Journal of Business Process Integration and Management, the International Journal on Enterprise Modelling and Information Systems Architectures, Computers in Industry, Business & Information Systems Engineering, IEEE Transactions on Services Computing, Lecture Notes in Business Information Processing, and Transactions on Petri Nets and Other Models of Concurrency. In 2012, he received the degree of doctor honoris causa from Hasselt University. He is also a member of the Royal Holland Society of Sciences and Humanities (Koninklijke Hollandsche Maatschappij der Wetenschappen) and the Academy of Europe (Academia Europaea).
Citation preview
A Decade of Business Process Management Conferences: Reflections on a Developing Discipline
Wil van der Aalstvdaalst.com
Just before BPM 2003
• Petri nets conference 2003• Grzegorz Rozenberg• Earlier events:
− W. van der Aalst, G. De Michelis, and C.A. Ellis, editors. Workflow Management: Net-based Concepts, Models, Techniques and Tools (WFM’98), Lisbon, June 1998.
− J. Desel, A. Oberweis, W. Reisig, G. Rozenberg, editors. Petri Nets and Business Process Management, Dagstuhl Seminar 98271, July 1998.
− W. van der Aalst, J. Desel, and R. Kaschek, editors. Software Architectures for Business Process Management (SABPM’99), Heidelberg, June 1999.
− W. van der Aalst, J. Desel, and A. Oberweis, editors. Business Process Management: Models, Techniques, and Empirical Studies, LNCS 1806, 2000.
How it all started …
PAGE 2
PAGE 3
77 submissions, 25 papers accepted
BPM 2003, June 26-27, 2003, Eindhoven
PAGE 4
PAGE 5
PAGE 6
PAGE 7
PAGE 8
PAGE 9
BPM 2003 in Eindhoven
PAGE 10
PAGE 11
PAGE 12
PAGE 13
BPM 2003, Eindhoven
PAGE 14
Eindhoven
Potsdam
614 km
PAGE 15
BPM 2004, Potsdam
PAGE 16
Nancy
Potsdam
818 km
PAGE 17
BPM 2005, Nancy
PAGE 18
Nancy
Vienna
903 km
PAGE 19
BPM 2006, Vienna
PAGE 20
Brisbane
Vienna
15636 km
PAGE 21
BPM 2007, Brisbane
PAGE 22
Milan
Brisbane
16248 km
PAGE 23
BPM 2008, Milan
PAGE 24
Milan
Ulm
429 km
PAGE 25
BPM 2009, Ulm
PAGE 27
BPM 2010, Hoboken
PAGE 29
BPM 2011, Clermont-Ferrand
PAGE 31
Tallinn
Beijing
6741 km
PAGE 32
PAGE 33
A Decade of Business Process Management Conferences
Reflections on a Developing Discipline
Another variant of the BPM lifecycle
PAGE 34
diagnosis/requirements
configuration/implementation
enactment/monitoring
adustment
(re)designmodelsdata
insightdiscussion
verification
performance analysisanimation
specificationdocumentation
configuration
Four main activities related to BPM
PAGE 35
model
enact
analyze
manage
creating a process model to be used for discussion, training, analysis or enactment
using a process model to control and supportconcrete cases
analyzing a process using a process
model and/or event logs (verification,
simulation, process mining, etc.)
all other activities, e.g., adjusting the
process, reallocating resources, or
managing largecollections of related
process models
History and Origins of the Domain
PAGE 36
database system
user interface
database system
user interface
database system
ap
plic
atio
n
BP
M s
yste
m1960 1975 1985 2000
ap
plic
atio
n
ap
plic
atio
n
ap
plic
atio
n
BPM
WFM
office automation
data modeling
operations management
scientific management
data/process mining
software engineering
formal methods
business process
reengineering
Skip Ellis, Office Talk, 1979
Michael Zisman, SCOOP, 1977
Anatol Holt, Information Systems Theory Project, 1968
Carl Adam Petri, Petri nets, 1962
PAGE 37
20 BPM Use Cases
20 BPM Use Cases
• Use cases to obtain a model [1-5]• Use cases to obtain a configurable model [6-8]• Use cases related to enactment [9-13]• Use cases for model-only-based analysis [14-15]• Use cases for log&model-based analysis [16-17]• Use cases to repair, extend or improve process
models [18-20]
Notation:
PAGE 38
MD|N|E
CMD|N|E
S E D
human model configurable model
information system
eventdata
diagnosticsD=descriptiveN=normativeE=executable
Use Case 1: Design model (DesM)
PAGE 39
MD|N|E
design model
(DesM)
Use Case 2: Discover model from event data (DiscM)
PAGE 40
E MD|E
discover model from event data
(DiscM)
Use Case 3:Select model from collection (SelM)
PAGE 41
MD|N|E
select model from collectionMMM
D|N|E
(SelM)
Use Case 4:Merge models (MerM)
PAGE 42
MD|N|E
merge modelsMMM
D|N|E
(MerM)
Use Case 5:Compose model (CompM)
PAGE 43
MD|N|E
compose modelMD|N|E
MD|N|E
MD|N|E
(CompM)
Use Case 6:Design configurable model (DesCM)
PAGE 44
CMD|N|E
design configurable model
(DesCM)
Use Case 7: Merge models into configurable model (MerCM)
PAGE 45
CMD|N|E
merge models into configurable modelMMM
D|N|E
(MerCM)
variant 1
variant 2
aa bb
dd
gg hh
ff
aa
dd
ee
gg hh
cc
ff
aa bb
dd
ee
gg hh
cc
ff
Use Case 8:Configure configurable model (ConCM)
PAGE 46
MD|N|E
configure configurable model
CMD|N|E
(ConCM)
aa bb
dd
ee
gg hh
cc
ff
aa
dd
gg hh
ff
Use Case 9:Refine model (RefM)
PAGE 47
refine model
MD|N
ME
(RefM)
Use Case 10:Enact model (EnM)
PAGE 48
enact model
ME
S(EnM)
Use Case 11:Log event data (LogED)
PAGE 49
log event data
ES (LogED)
Use Case 12:Monitor (Mon)
PAGE 50
monitor
S D(Mon)
Use Case 13:Adapt while running (AdaWR)
PAGE 51
Sadapt while running
ME
(AdaWR) SME
Use Case 14: Analyze performance based on model (PerfM)
PAGE 52
analyze performance based on model
ME
PD(PerfM)
Use Case 15:Verify model (VerM)
PAGE 53
verify model
ME
CD(VerM)
Use Case 16: Check conformance using event data (ConfED)
PAGE 54
check conformance using event data
ME
CDE (ConfED)
Use Case 17: Analyze performance using event data (PerfED)
PAGE 55
analyze performance using event data
ME
E PD(PerfED)
Use Case 18: Repair model (RepM)
PAGE 56
repair model
MD|N|E
CD MD|N|E
(RepM)
Use Case 19:Extend model (ExtM)
PAGE 57
extend model
ME
E ME
(ExtM)
astart
b
c
d
e
g
h
f
end
timestamps in the event log can be used to analyze waiting
times in-between activities
1391
1537
566
15371537
566
1391
1391
146146
971 9711537
1537
146
930 930
461 461
Sue Mike
Pete
Mary
Norman
check="OK" and report="Approved"
resource information in the event log can be used for social network analysis, role
discovery, and performance analysis
attributes in the event log can be used for decision point analysis
Use Case 20:Improve model (ImpM)
PAGE 58
improve model
MD|N|E
MD|N|E
PD (ImpM)
Overview Use Cases
PAGE 59
diagnosis/requirements
configuration/implementation
enactment/monitoring
adustment
(re)designmodelsdata
insightdiscussion
verification
performance analysisanimation
specificationdocumentation
configuration
1
2
34
56
7
8
9
10
11 12
13
1415
16 17
1819
20
• Use cases to obtain a model [1-5]
• Use cases to obtain a configurable model [6-8]
• Use cases related to enactment [9-13]
• Use cases for model-only-based analysis [14-15]
• Use cases for log&model-based analysis [16-17]
• Use cases to repair, extend or improve process models [18-20]
BPM proceedings (10 years)
PAGE 60
-
surprising focus on
verification
enactment is broad topic
not surprising
weakness: topics related to process improvement and
performance analysis score low
289 papers, 367 tags
Detailed view
PAGE 61
PAGE 62
6 BPM Key Concerns
6 BPM Key Concerns (not detailed here)
1. Process modeling languages (suggesting new languages, comparing/evaluating languages, etc.)
2. Process enactment infrastructures (focus on software and systems to execute, support, and monitor processes)
3. Process model analysis (e.g., verification and simulation; no event data is used)
4. Process mining (analysis based on event data)
5. Process flexibility (ability to deal with foreseen and unforeseen changes)
6. Process reuse (configurable models, reference models, process repositories, similarity search, etc.)
PAGE 63
PAGE 64
BPM proceedings (10 years) -
PAGE 65
Observations
• Disclaimer: tagging of 289 papers is highly subjective and obvious classes may be missing (e.g., patterns, process integration, collaboration).
• Perspectives (control-flow, data, resources, etc.) could have been added as an additional dimension.
• Rapidly maturing discipline, but:− Many papers introduce a new modeling language (Needed? Used
again?).
− Several papers cannot be linked to one of the 20 use cases in a straightforward manner.
− Coverage of the domain can be improved.
− Many papers describe implementation efforts; however, frequently the software is not available for the reader.
− Many papers include case studies, e.g., to test a new technique or system, which is good. Unfortunately, most case studies seem rather artificial.
PAGE 66
PAGE 67
Relating
Flexibility, Configuration, and Mining
BPM Challenges are Related
PAGE 68
processflexibility
processmining
processconfiguration
flexibility versus conformance
cross-organizational mining
variability at different levels
Process Flexibility
PAGE 69
processflexibility
processmining
processconfiguration
flexibility versus conformance
cross-organizational mining
variability at different levels
Taxonomy of Flexibility
PAGE 70
flexibility by definition
process definition
process instance
degr
ee o
f im
pact
design time runtime
time at which flexibility is added
flexibility by deviation
flexibility by underspecification
flexibility by change
Helen Schonenberg, Ronny Mans, Nick Russell, Nataliya Mulyar, Wil M. P. van der Aalst: Process Flexibility: A Survey of Contemporary Approaches. Lecture Notes in Business Information Processing, 2008, Volume 10, Part 1, 16-30, DOI: 10.1007/978-3-540-68644-6_2
Flexibility by Definition
a b cstart end
flexibility by definition
process definition
process instance
degr
ee o
f im
pact
design time runtime
time at which flexibility is added
flexibility by deviation
flexibility by underspecification
flexibility by change
Flexibility by Deviation
a b cstart end
flexibility by definition
process definition
process instance
degr
ee o
f im
pact
design time runtime
time at which flexibility is added
flexibility by deviation
flexibility by underspecification
flexibility by change
Flexibility by Underspecification
a cstart end?
x y z
flexibility by definition
process definition
process instance
degr
ee o
f im
pact
design time runtime
time at which flexibility is added
flexibility by deviation
flexibility by underspecification
flexibility by change
Flexibility by Change
a b cstart end
a cstart end
flexibility by definition
process definition
process instance
degr
ee o
f im
pact
design time runtime
time at which flexibility is added
flexibility by deviation
flexibility by underspecification
flexibility by change
Procedural Versus Declarative
PAGE 75
non co-existence: activities b and d cannot happen both for the same case
response: every occurrence of b should be eventually followed by c or d
precedence: every occurrence of d needs to be preceded by a
response
non co-existence
precedence
eat food feel bad
drink beer
drink wine
c
d
a bc1
c4
c2
c3
Declarative = anything is possible unless it is explicitly forbiddenProcedural = everything is impossible unless it is triggered explicitly
Process Configuration
PAGE 76
processflexibility
processmining
processconfiguration
flexibility versus conformance
cross-organizational mining
variability at different levels
Variants of the same process
PAGE 77
aa bb
dd
ee
gg hh
cc
ff
aa bb
dd
gg hh
ff
aa
dd
ee
gg hh
cc
ff
Configurable process models
• Reference models revisited, but now better (correct, executable, etc.).
• Examples:− 430 Dutch municipalities need to execute the same collection
of processes, but value their "couleur locale"− Hertz has 8,650 rental locations in about 150 countries
worldwide all executing essentially the same set of processes (but with local differences)
− All 94 U.S. District Courts in the United States share the same set of workflows
• Process sharing will increase (cf. cloud computing, SaaS, etc.)
• "Content" is often missing in BPM approaches!
PAGE 78
Process Mining
PAGE 79
processflexibility
processmining
processconfiguration
flexibility versus conformance
cross-organizational mining
variability at different levels
PAGE 80
Process Discovery
PAGE 81
Conformance Checking
desire line expected or normative path
PAGE 82
BPM Challenges are Related
PAGE 83
processflexibility
processmining
processconfiguration
Process Flexibility and Process Mining
PAGE 84
degree of flexibility
low high
degr
ee o
f exa
min
ation
low
high
uncontrolled chaos
potential waste
Process and Case Dimensions
PAGE 85
timelocation
group
concept drift
analysiscross-organizational
process mining
clustering and classification
acbeabceade
acbeabceade
acbeabceade
Example: Hertz has 8,650 rental locations and different types of customers
PAGE 86
January
Am
ster
dam
gold
silver
normal
Ein
dhov
enP
aris
FebruaryM
arch
Example: All municipalities need to handle building permits
PAGE 87
Q1E
erse
l
>100k
50k
Bla
del
Reu
sel
Q2
Q3
>50k & 100k
Example: Suncorp has different brands and different types of insurance
PAGE 88
before flooding
during flooding
after flooding
Example Questions
• How to detect changes over time (concept drift)?
• How to compare process variants in different organizations (cross-organizational mining)?
PAGE 89
timelocation
group
processflexibility
processmining
processconfiguration
Concept drift (work of JC Bose)
PAGE 90
Cross-organizational mining (work of Joos Buijs)
PAGE 91
• 10 muncipalities: Coevorden, Emmen, Hellendoorn,Gemert-Bakel, Zwolle, Bergeijk, Bladel, Eersel, Reusel-De Mierden, and Oirschot.
• 8 processes: Gemeentelijke Basisadministratie Persoonsgegevens (GBA 3x), Melding Openbare Ruimte (MOR), Wet Algemene Bepalingen Omgevingsrecht (WABO 2x), Wet Maatschappelijke Ondersteuning (WMO), and Waardering Onroerende Zaken (WOZ).
Ingredients:•event logs•models•conformance checking•key performance indicators
Questions:•How similar?•Why better?
PAGE 92
A maturing discipline …
Some books (1/4)
PAGE 93
Some books (2/4)
PAGE 94
Some books (3/4)
PAGE 95
Some books (4/4)
PAGE 96
PAGE 97
Conclusion
10 Years of BPM Conferences
• Business Process Management (BPM) is an important, relevant and interesting topic.
• Provides challenging and fascinating computer science problems (verification, process mining, enactment, flexibility, etc.).
• Requires connections to other sciences (management science, operations research, social sciences, etc.).
PAGE 98
Recommendations
• Avoid introducing new languages without a clear purpose (short lifetime, incomparable results).
• Artifacts (software and data) need to be made available (suggestion: classify papers based on their level of openness).
• Evaluate results based on a predefined criterion and compare with other approaches.
• Many prototypes are developed from scratch and "fade into oblivion", so as a community we should build on shared platforms (and not always ask for something "new").
• Contribution is not always clear; a paper should focus on at least one of the 20 use cases. Suggestions:
− Further develop use case classification− Tag papers based on these use cases
PAGE 99
Recommended