Chapter 2Process Modeling and Analysis
prof.dr.ir. Wil van der Aalstwww.processmining.org
Overview
PAGE 1
Part I: Preliminaries
Chapter 2 Process Modeling and Analysis
Chapter 3Data Mining
Part II: From Event Logs to Process Models
Chapter 4 Getting the Data
Chapter 5 Process Discovery: An Introduction
Chapter 6 Advanced Process Discovery Techniques
Part III: Beyond Process Discovery
Chapter 7 Conformance Checking
Chapter 8 Mining Additional Perspectives
Chapter 9 Operational Support
Part IV: Putting Process Mining to Work
Chapter 10 Tool Support
Chapter 11 Analyzing “Lasagna Processes”
Chapter 12 Analyzing “Spaghetti Processes”
Part V: Reflection
Chapter 13Cartography and Navigation
Chapter 14Epilogue
Chapter 1 Introduction
Productivity improvements
• Adam Smith (1723-1790) showed the advantages of the division of labor.
• Frederick Taylor (1856-1915) introduced the initial principles of scientific management.
• Henry Ford (1863-1947) introduced the production line for the mass production of “black T-Fords”.
• Since 1950 computers and digital communication infrastructures are the most dominant factor influencing business processes and their management.
PAGE 2
Role of models
• Operations management, and in particular operation research, is a branch of management science heavily relying on modeling.
• Models are used to reason about processes (redesign) and to make decisions inside processes (planning and control).
• A process model may be used to discuss responsibilities, analyze compliance, predict performance using simulation, and configure a WFM system.
PAGE 3
Problems of models
• The model describes an idealized version of reality.• Inability to adequately capture human behavior.• The model is at the wrong abstraction level.• Therefore, we advocate the use of event data:
− Process mining allows for the extraction of models based on facts.
− Moreover, process mining does not aim at creating a single model of the process.
− Instead, it provides various views on the same reality at different abstraction levels.
− For example, users can decide to look at the most frequent behavior to get a simple model (“80% model”).
− However, they can also inspect the full behavior by deriving the “100% model” covering all cases observed.
PAGE 4
Transition systems
PAGE 5
[start]
register request
examine thoroughly
examine casually
checkticket
decide
pay compensation
reject request
reinitiate request
[end][c1,c2]
[c1,c4]
[c2,c3]
[c3,c4]
checkticket
examine casually
examine thoroughly
[c5]
Petri nets
PAGE 6
astart register
request
bexamine
thoroughly
cexamine casually
d
check ticket
decide
pay compensation
reject request
reinitiate request
e
g
h
f
end
c1
c2
c3
c4
c5
t
p
transition
place
token
AND-split AND-join
XOR-joinXOR-split
AND-split
XOR-split
XOR-join
Reachability graph
PAGE 7
astart register
request
bexamine
thoroughly
cexamine casually
d
check ticket
decide
pay compensation
reject request
reinitiate request
e
g
h
f
end
c1
c2
c3
c4
c5
t
p
transition
place
token
AND-split AND-join
XOR-joinXOR-split
AND-split
XOR-split
XOR-join
[start]
register request
examine thoroughly
examine casually
checkticket
decide
pay compensation
reject request
reinitiate request
[end][c1,c2]
[c1,c4]
[c2,c3]
[c3,c4]
checkticket
examine casually
examine thoroughly
[c5]
How many states?
PAGE 8
t t2
p
i2
t3i3
t4i4
t5i5
t1i1
out
(a) (c)
t
p
(b)
WF-nets and soundness
PAGE 9
start end
YAWL
PAGE 10
start
register request
examine thoroughly
examine casually
checkticket
decide
pay compensation
reject request
new information
end
c1
c2
OR-split OR-join
c3
condition (like a place in a Petri net)
task (i.e., an atomic activity)
AND-split
XOR-split
OR-split
AND-join
XOR-join
OR-join
start end
multiple instance task
composite task
cancelation region
BPMN
PAGE 11
task/activity
AND-splitgateway
XOR-split gateway
OR-split gateway
AND-join gateway
XOR-join gateway
OR-join gateway
startevent
endevent
intermediate events
y
z
x
deferred choice pattern using the event-based XOR gateway
register request
examine casually
examine thoroughly
check ticket
decide
pay compensation
reject request
reinitiate request
start end
Event-Driven Process Chains (EPCs)
PAGE 12
function
AND-split connector
XOR-split connector
OR-split connector
AND-join connector
XOR-join connector
OR-join connector
startevent
endevent
AND AND
XOR
OR OR
XOR
intermediate event
start
register request
examinethoroughly
examine casually
checkticket
decide
pay compensation
reject request
end
e1
AND
OR
XOR
OR
AND
XOR end
e2
e3
e4 e5
e6
Vicious cycle paradox
PAGE 13
e1
OR
f1
AND
e3
e2
OR
f2
AND
e4
Causal nets (C-nets)
PAGE 14
a
register request
b
examine thoroughly
c
examine casually
d
checkticket
decide
pay compensation
reject request
e
g
h
f
end
reinitiate request
z
XOR-split AND-split OR-split
XOR-join AND-join OR-join
Why C-nets?
• Similar to heuristic nets and representation used by genetic miners.
• Fits well with mainstream languages (BPMN, EPCs, YAWL, BPEL, etc.).
• Model XOR, AND, and OR, but no silent steps or duplicate activities needed.
• Loose interpretation (focus on replay semantics rather than execution semantics).
PAGE 15
Another C-net
PAGE 16
a
start booking
c e
complete booking
bookcar
d
bookhotel
b
bookflight
WF-net interpretation of C-nets(only valid sequences!)
PAGE 17
bookflight
a
start booking
c
b
d
bookcar
bookhotel
e
complete booking
a
start booking
c e
complete booking
bookcar
d
bookhotel
b
bookflight
Non-sound C-nets
PAGE 18
a
start booking
c e
complete booking
bookcar
d
bookhotel
b
bookflight
(a) unsound because there are no valid sequences
a
start booking
c e
complete booking
bookcar
d
bookhotel
b
bookflight
(b) unsound although there exist valid sequences
Unbounded C-net
PAGE 19
a
c
b d e
Verification
PAGE 20
astart register
request
bexamine
thoroughly
cexamine casually
d
check ticket
decide
pay compensation
reject request
reinitiate request
e
g
h
f
end
c1
c2
c3
c4
c5
deadlock
c6
Example: YAWL
PAGE 21
Performance analysis, e.g., simulation in BPM|one
PAGE 22
Limitations of model-based analysis
• Verification and performance analysis heavily rely on the availability of high quality models.
• When the models and reality have little in common, model-based analysis does not make much sense.
• There is often a lack of alignment between hand-made models and reality
• Process mining aims to address these problems by establishing a direct connection between the models and actual low-level event data about the process.
• Process discovery techniques allow for viewing the same reality from different angles and at different levels of abstraction.
PAGE 23