Business Process Analytics

Process Mining
Chapter I : Introduction

2019-2020 Master 2 STIN

Mathilde Boltenhagen

[email protected]

LSV, CNRS, ENS Paris-Saclay, Inria, Université Paris-Saclay

!1
Calendar 2019-2020
| Master 2 STIN • BPA • 2019 • Boltenhagen |

SESSION 1 : 25/09/2019 SESSION 5 : 14/11/2019

Lecture : Introduction Lecture : Conformance Checking
+ TP + TP
SESSION 2 : 03/10/2019 SESSION 6 : 11/12/2019
Lecture : Model discovery Exam
+ TD + TP
SESSION 3 : 10/10/2019 SESSION 7 : 16/12/2019
Lecture : Model discovery Lecture : Data analytics
+ TD + presentations
SESSION 4 : 08/11/2019 SESSION 8 : 18/12/2019
Lecture : Conformance Checking Lecture : Data analytics
+ TD + presentations

!2
Markings 2019-2020
| Master 2 STIN • BPA • 2019 • Boltenhagen |

> Exam : 11/12/2019 (sujet en français ;) )

-> 40 %
-> handwritten double-sided sheet authorized
> TP : 14/11/2019 -> 15/12/2019
-> 30 %
> 10-min presentations : 16-18/12/2019
-> 30 %
-> a package of ProM OR a Process Mining tool OR a scientific paper
-> mail with name members and subject before : 09/10/2019
-> outlines of the presentation before : 14/11/2019 (1 page maximum)

!3
Outlines
| Master 2 STIN • BPA • 2019 • Boltenhagen |

Chapter I. Chapter II. Chapter III. Chapter IV.

Introduction Model discovery Conformance Data Analytics

1. What is Process Mining ?

2. What and where can be processes/systems ?
3. Data as Event log
4. Process Models

!4
Chapter I : Introduction
1. What is Process Mining ?

!5
Chapter I : Introduction 1. What is Process Mining ?
| Master 2 STIN • BPA • 2019 • Boltenhagen |

> Recent field

> IEEE Task Force : 2009

> BPI workshops : 2009

> First Book : 2011

Wil Van Der Aalst
> First International Conference of Process Mining : 2019

!6
| Master 2 STIN • BPA • 2019 • Boltenhagen |
Chapter I : Introduction 1. What is Process Mining ?
Field graph | Master 2 STIN • BPA • 2019 • Boltenhagen |

Optimization Data
Conflicts Business
Mining Data
Cost/profit Business Data
Intelligence Visualization
Risk Management Science
Process
architecture Process Machine
Big Data
Business Mining Verification, Learning
Process Specification of
Modeling testing/training

Formal
Game theory Models Proofs

Complex systems Models Checking

Concurrency
!7
Chapter I : Introduction 1. What is Process Mining ?
Definition of Manifesto | Master 2 STIN • BPA • 2019 • Boltenhagen |

Support/Control

Reality Systems

Models ent
Models lem
design analyses Imp Record,
Store

Discovery
Models
Conformance Databases

Enhancement

!8
Source : https://www.win.tue.nl/ieeetfpm/downloads/Process%20Mining%20Manifesto.pdf
Chapter I : Introduction 1. What is Process Mining ?
Definition of Manifesto | Master 2 STIN • BPA • 2019 • Boltenhagen |

Support/Control

Reality Systems

Models ent
lem Close to
analyses Imp Record,
Store Business
Intelligence
Discovery
Models
Conformance Databases

Enhancement

!9
Source : https://www.win.tue.nl/ieeetfpm/downloads/Process%20Mining%20Manifesto.pdf
Chapter I : Introduction 1. What is Process Mining ?
Definition of Manifesto | Master 2 STIN • BPA • 2019 • Boltenhagen |

Support/Control

Reality Systems

Models ent
lem
analyses Imp Record,
Store

Discovery
Close to Data Mining,
Models Use Data Science
Conformance Databases techniques and
Formals Models
Enhancement

!10
Source : https://www.win.tue.nl/ieeetfpm/downloads/Process%20Mining%20Manifesto.pdf
Chapter I : Introduction 1. What is Process Mining ?
Definition of Manifesto | Master 2 STIN • BPA • 2019 • Boltenhagen |

Support/Control

Reality Systems

Models ent
lem
Business analyses Imp Record,
Process Store
Management
Discovery
Models
Conformance Databases

Enhancement

!11
Source : https://www.win.tue.nl/ieeetfpm/downloads/Process%20Mining%20Manifesto.pdf
Chapter I : Introduction 1. What is Process Mining ?
Examples | Master 2 STIN • BPA • 2019 • Boltenhagen |

a patient arrives at the emergencies e1

first aid is done e2 the patient is urgently operated

What aremust
the patient thebe behaviors ? x
moved to the department e3

the patient must be moved to the department y

Hospital the doctor fills a form about the patient e4

< e1, e2, e3, e4 >

!12
Chapter I : Introduction 1. What is Process Mining ?
| Master 2 STIN • BPA • 2019 • Boltenhagen |

ali ty
u s
Ca
Systems Generate behaviors

Models

m ps
tim esta
Databases Conc
u rren
cy
Nois
e

!13
 Chapter I : Introduction
2. What and where can be processes/systems ?

!14
Chapter I : Introduction 2. What and where can be processes/systems ?
| Master 2 STIN • BPA • 2019 • Boltenhagen |

> Behaviors of persons

- in health care
- demographic studies
- Users data of apps

> Behaviors of objects Systems Generate behaviors

- industrial companies
- Connected objects

> Anywhere
- recorded data are everywhere

!15
Chapter I : Introduction 2. What and where can be processes/systems ?
| Master 2 STIN • BPA • 2019 • Boltenhagen |

> Activities

> Participants

> Ressources Systems Generate behaviors

> Other data

———> Complex databases

———> Complex questions

!16
Chapter I : Introduction 2. What and where can be processes/systems ?
| Master 2 STIN • BPA • 2019 • Boltenhagen |
s
r o cesse
pa st p
ing ?
tand d first cesse
s
ders ee st p r o
Un proc ng p
a d?
Who ersta
n d i
f ini s he
n d y A
U
a c tivit c a tion o
n fi t
Whe eri
a n ce/V llowed
c o n form
n t Xa A?
Semanti ear ? C ip a ity
y B c a n app pa rtic s activ
tivit Is es
Does ac s t p r o cesses proc
a
rs t a nding p
Un d e
eve nt ? sses
a u s e s this Present proc
e
What c
bo u t strea ming
t u r e events W h at a
ng fu
Predicti ap p ear ? data ?
is even t
Will th
!17
Chapter I : Introduction 2. What and where can be processes/systems ?
| Master 2 STIN • BPA • 2019 • Boltenhagen |

2.2.What
Whatand
andwhere
wherecan
canbe
beprocesses/systems
processes/systems
<< Process mining is an important tool for

?? modern organizations that need to manage non-

Generate behaviors
trivial operational processes. >> Systems

Manifesto of Process Mining

!18
Chapter I : Introduction
3. Data as Event Log

!19
Chapter I : Introduction 3. Data as Event Log
| Master 2 STIN • BPA • 2019 • Boltenhagen |

> Event data

> Recorded from systems

- may contain noise ps
e s tam
- may be incomplete ti m
- may contain outliers, exceptional behaviors Databases

> Have other ressources informations Nois

- very complex e

!20
Chapter I : Introduction 3. Data as Event Log
| Master 2 STIN • BPA • 2019 • Boltenhagen |

User ID Timestamp Activity Age Location

2093 2019-02-15 T 10:45 UTC Register 18 40.741895,-73.989308
2093 2019-02-15 T 11:10 UTC Complete profile 18 40.741895,-73.989308
3004 2019-02-17 T 12:43 UTC Register 22 40.4167047,-3.7035825
2093 2019-03-08 T 11:23 UTC Register 18 48.8566969,2.3514616
2014 2019-03-09 T 12:09 UTC Upload file 39 41.8755616,-87.6244212
3004 2019-04-17 T 09:33 UTC Send a request 22 37.3886303,-5.9953403
3004 2019-04-17 T 09:56 UTC Delete profile 22 37.3886303,-5.9953403
1009 2019-08-01 T 07:03 UTC Register 50 34.6937569,135.5014539

!21
Chapter I : Introduction 3. Data as Event Log
| Master 2 STIN • BPA • 2019 • Boltenhagen |

User ID Timestamp Activity Age Location

2093 2019-02-15 T 10:45 UTC Register 18 40.741895,-73.989308
2093 2019-02-15 T 11:10 UTC Complete profile 18 40.741895,-73.989308
3004 2019-02-17 T 12:43 UTC Register 22 40.4167047,-3.7035825
2093 2019-03-08 T 11:23 UTC Register 18 48.8566969,2.3514616
2014 2019-03-09 T 12:09 UTC Upload file 39 41.8755616,-87.6244212
3004 2019-04-17 T 09:33 UTC Send a request 22 37.3886303,-5.9953403
3004 2019-04-17 T 09:56 UTC Delete profile 22 37.3886303,-5.9953403
1009 2019-08-01 T 07:03 UTC Register 50 34.6937569,135.5014539

> a row is an event

!22
Chapter I : Introduction 3. Data as Event Log
| Master 2 STIN • BPA • 2019 • Boltenhagen |

User ID Timestamp Activity Age Location

2093 2019-02-15 T 10:45 UTC Register 18 40.741895,-73.989308
2093 2019-02-15 T 11:10 UTC Complete profile 18 40.741895,-73.989308
3004 2019-02-17 T 12:43 UTC Register 22 40.4167047,-3.7035825
2093 2019-03-08 T 11:23 UTC Register 18 48.8566969,2.3514616
2014 2019-03-09 T 12:09 UTC Upload file 39 41.8755616,-87.6244212
3004 2019-04-17 T 09:33 UTC Send a request 22 37.3886303,-5.9953403
3004 2019-04-17 T 09:56 UTC Delete profile 22 37.3886303,-5.9953403
1009 2019-08-01 T 07:03 UTC Register 50 34.6937569,135.5014539

> each case is a sequence of several events

!23
Chapter I : Introduction 3. Data as Event Log
| Master 2 STIN • BPA • 2019 • Boltenhagen |

User ID Timestamp Activity Age Location

2093 2019-02-15 T 10:45 UTC Register 18 40.741895,-73.989308
2093 2019-02-15 T 11:10 UTC Complete profile 18 40.741895,-73.989308
3004 2019-02-17 T 12:43 UTC Register 22 40.4167047,-3.7035825
2093 2019-03-08 T 11:23 UTC Register 18 48.8566969,2.3514616
2014 2019-03-09 T 12:09 UTC Upload file 39 41.8755616,-87.6244212
3004 2019-04-17 T 09:33 UTC Send a request 22 37.3886303,-5.9953403
3004 2019-04-17 T 09:56 UTC Delete profile 22 37.3886303,-5.9953403
1009 2019-08-01 T 07:03 UTC Register 50 34.6937569,135.5014539

> each case is a sequence of several events

> each event of a case usually has different timestamp

!24
Chapter I : Introduction 3. Data as Event Log
| Master 2 STIN • BPA • 2019 • Boltenhagen |

User ID Timestamp Activity Age Location

2093 2019-02-15 T 10:45 UTC Register 18 40.741895,-73.989308
2093 2019-02-15 T 11:10 UTC Complete profile 18 40.741895,-73.989308
3004 2019-02-17 T 12:43 UTC Register 22 40.4167047,-3.7035825
2093 2019-03-08 T 11:23 UTC Register 18 48.8566969,2.3514616
2014 2019-03-09 T 12:09 UTC Upload file 39 41.8755616,-87.6244212
3004 2019-04-17 T 09:33 UTC Send a request 22 37.3886303,-5.9953403
3004 2019-04-17 T 09:56 UTC Delete profile 22 37.3886303,-5.9953403
1009 2019-08-01 T 07:03 UTC Register 50 34.6937569,135.5014539

> each case is a sequence of several events

> each event of a case usually has different timestamp
> an activity name describes the event
!25
Chapter I : Introduction 3. Data as Event Log
| Master 2 STIN • BPA • 2019 • Boltenhagen |

User ID Timestamp Activity Age Location

2093 2019-02-15 T 10:45 UTC Register 18 40.741895,-73.989308
2093 2019-02-15 T 11:10 UTC Complete profile 18 40.741895,-73.989308
3004 2019-02-17 T 12:43 UTC Register 22 40.4167047,-3.7035825
2093 2019-03-08 T 11:23 UTC Register 18 48.8566969,2.3514616
2014 2019-03-09 T 12:09 UTC Upload file 39 41.8755616,-87.6244212
3004 2019-04-17 T 09:33 UTC Send a request 22 37.3886303,-5.9953403
3004 2019-04-17 T 09:56 UTC Delete profile 22 37.3886303,-5.9953403
1009 2019-08-01 T 07:03 UTC Register 50 34.6937569,135.5014539

> other resources often exist to describe the events

!26
Chapter I : Introduction 3. Data as Event Log
| Master 2 STIN • BPA • 2019 • Boltenhagen |

User ID Timestamp Activity Age Location

2093 2019-02-15 T 10:45 UTC Register 18 40.741895,-73.989308
2093 2019-02-15 T 11:10 UTC Complete profile 18 40.741895,-73.989308
3004 2019-02-17 T 12:43 UTC Register 22 40.4167047,-3.7035825
2093 2019-03-08 T 11:23 UTC Register 18 48.8566969,2.3514616
2014 2019-03-09 T 12:09 UTC Upload file 39 41.8755616,-87.6244212
3004 2019-04-17 T 09:33 UTC Send a request 22 37.3886303,-5.9953403
3004 2019-04-17 T 09:56 UTC Delete profile 22 37.3886303,-5.9953403
1009 2019-08-01 T 07:03 UTC Register 50 34.6937569,135.5014539

> a sequence of event of 2093’s case can simply be represented by :

< Register, Complete profile, Register >

!27
Chapter I : Introduction 3. Data as Event Log
| Master 2 STIN • BPA • 2019 • Boltenhagen |

User ID Timestamp Activity Age Location

2093 2019-02-15 T 10:45 UTC Register 18 40.741895,-73.989308
2093 2019-02-15 T 11:10 UTC Complete profile 18 40.741895,-73.989308
3004 2019-02-17 T 12:43 UTC Register 22 40.4167047,-3.7035825
2093 2019-03-08 T 11:23 UTC Register 18 48.8566969,2.3514616
2014 2019-03-09 T 12:09 UTC Upload file 39 41.8755616,-87.6244212
3004 2019-04-17 T 09:33 UTC Send a request 22 37.3886303,-5.9953403
3004 2019-04-17 T 09:56 UTC Delete profile 22 37.3886303,-5.9953403
1009 2019-08-01 T 07:03 UTC Register 50 34.6937569,135.5014539

< Register, Complete profile, Register > : case 2093

< Register, Send a request, Delete profile > : case 3004
< Upload file > : case 2014
< Register > : case 1009
!28
Chapter I : Introduction 3. Data as Event Log
| Master 2 STIN • BPA • 2019 • Boltenhagen |

> Format .xes for

Standard for eXtensible Event Stream :

- xml-based format
- invented in 1849
- officially published in 2016
- standardized by IEEE Task Force of Process
Mining in 2010
- replace .mxml files

!29
- www.xes-standard.org
Chapter I : Introduction 3. Data as Event Log
| Master 2 STIN • BPA • 2019 • Boltenhagen |

Definition (Event) : An event e is defined by an activity and a timestamp and

can contain other ressources informations.

!30
http://www.processmining.org/_media/presentations/event_logs_the_input_for_process_mining.pdf
Chapter I : Introduction 3. Data as Event Log
| Master 2 STIN • BPA • 2019 • Boltenhagen |

Definition (Case) : A case c is an ordered sequence of events <e1, … en> such

that ei ∈ ∑, an event universe. Events are ordered by timestamp.

!31
Chapter I : Introduction 3. Data as Event Log
| Master 2 STIN • BPA • 2019 • Boltenhagen |

Definition (Event Log) : An event Log L is a set of cases. As every event are
different, cases are different.

!32
http://www.processmining.org/_media/presentations/event_logs_the_input_for_process_mining.pdf
Chapter I : Introduction 3. Data as Event Log
| Master 2 STIN • BPA • 2019 • Boltenhagen |

Definition (Trace variant) : A trace variant σ is a singular sequence of activities

<a1, … an> such that all cases c ∈ L that correspond to the same permutation of
activities are represented by σ.

Example of trace variants :

< Register Client, Mail rejection> 100
< Register Client, Mail Accepted> 500

In this lecture, log are then be shorthanded by :

< RC, MR>100
< RC, MA>500

!33
http://www.processmining.org/_media/presentations/event_logs_the_input_for_process_mining.pdf
Chapter I : Introduction 3. Data as Event Log
| Master 2 STIN • BPA • 2019 • Boltenhagen |

What is the .xes file of those data event ?

User ID Timestamp Activity Age Location

2093 2019-02-15 T 10:45 UTC Register 18 40.741895,-73.989308
2093 2019-02-15 T 11:10 UTC Complete profile 18 40.741895,-73.989308
3004 2019-02-17 T 12:43 UTC Register 22 40.4167047,-3.7035825
2093 2019-03-08 T 11:23 UTC Register 18 48.8566969,2.3514616

!34
Chapter I : Introduction 3. Data as Event Log
| Master 2 STIN • BPA • 2019 • Boltenhagen |

<?xml version="1.0" encoding="UTF-8" ?> <trace>

<!-- This file ha………[…]……ent Log"/> <string key="concept:name" value="3004"/>
<trace> <event>
<string key="concept:name" value="2093"/> <string key="concept:name" value="Register"/>
<event> <string key="lifecycle:transition" value="complete"/>
<string key="concept:name" value="Register"/> <date key="time:timestamp"
<string key="lifecycle:transition" value="complete"/> value="2019-02-17T00:00:00.000+01:00"/>
<date key="time:timestamp" value="2019-02-15T00:00:00.000+01:00"/> <int key="Age" value="22"/>
<int key="Age" value=« 18"/> <string key="Location" value="40.4167047,-3.7035825"/>
<string key="Location" value="40.741895,-73.989308"/> </event>
</event> </trace>
<event> </log>
<string key="concept:name" value="Complete profile"/>
<string key="lifecycle:transition" value="complete"/>
<date key="time:timestamp" value="2019-02-15T00:00:00.000+01:00"/>
<int key="Age" value="18"/>
<string key="Location" value="40.741895,-73.989308"/>
</event>
<event>
<string key="concept:name" value="Register"/>
<string key="lifecycle:transition" value="complete"/>
<date key="time:timestamp" value="2019-03-08T00:00:00.000+01:00"/>
<int key="Age" value="18"/>
<string key="Location" value="48.8566969,2.3514616"/>
</event>
</trace>

!35
Chapter I : Introduction 3. Data as Event Log
| Master 2 STIN • BPA • 2019 • Boltenhagen |

What are the different trace variants ?

User ID Timestamp Activity Age Location

2093 2019-02-15 T 10:45 UTC Register 18 40.741895,-73.989308
2093 2019-02-15 T 11:10 UTC Complete profile 18 40.741895,-73.989308
3004 2019-02-17 T 12:43 UTC Register 22 40.4167047,-3.7035825
2093 2019-03-08 T 11:23 UTC Log out 18 48.8566969,2.3514616
2014 2019-03-09 T 12:09 UTC Register 22 41.8755616,-87.6244212
3004 2019-04-17 T 09:33 UTC Send a request 22 37.3886303,-5.9953403
3004 2019-04-17 T 09:56 UTC Delete profile 22 37.3886303,-5.9953403
2014 2019-08-01 T 07:03 UTC Complete profile 50 34.6937569,135.5014539
2014 2019-08-01 T 07:05 UTC Log out 22

!36
Chapter I : Introduction 3. Data as Event Log
| Master 2 STIN • BPA • 2019 • Boltenhagen |

• Traces :
< Register, Complete profile, Log out > : 2093
< Register, Send a request, Delete profile > : 3004
< Register, Complete profile, Log out > : 2014

• Trace variants :
< Register, Complete profile, Log out > 2
< Register, Send a request, Delete profile > 1

!37
Chapter I : Introduction
4. Process Models

!38
Chapter I : Introduction 4. Process Models
Some Process Models | Master 2 STIN • BPA • 2019 • Boltenhagen |

FlowCharts Data-Flow Diagram BPMN Petri Nets

!39
http://www.bpmn.org/ ; https://www.businessprocessincubator.com ; https://www.visual-paradigm.com/ ;
Chapter I : Introduction 4. Process Models
Business Process Modeling Notation | Master 2 STIN • BPA • 2019 • Boltenhagen |

Event : happens in the process, has a trigger and a result.

!40
Chapter I : Introduction 4. Process Models
Business Process Modeling Notation | Master 2 STIN • BPA • 2019 • Boltenhagen |

Activity : is a generic term for work that company performs

in a Process (task, choreography task, collapsed sub-
process…)
Register

!41
Chapter I : Introduction 4. Process Models
Business Process Modeling Notation | Master 2 STIN • BPA • 2019 • Boltenhagen |

Definition (Causality) : Two activities have a causal relation if one appears

before the other.

Log in Read mails …

!42
Chapter I : Introduction 4. Process Models
Business Process Modeling Notation | Master 2 STIN • BPA • 2019 • Boltenhagen |

Gateway : is used to control the divergence and

convergence of Sequence Flows in a Process

Accepted

Register

Rejected

!43
Chapter I : Introduction 4. Process Models
Business Process Modeling Notation | Master 2 STIN • BPA • 2019 • Boltenhagen |

Definition (Concurrency) : Two activities are concurrent if they can be executed

at the same time.

Use
Services

Register
+ ++
Complete
profile

!44
Chapter I : Introduction 4. Process Models
Business Process Modeling Notation | Master 2 STIN • BPA • 2019 • Boltenhagen |

Connecting objects : is used to control the divergence and

convergence of Sequence Flows in a Process
Default Flow

Conditional Flow

!45
Chapter I : Introduction 4. Process Models
Business Process Modeling Notation | Master 2 STIN • BPA • 2019 • Boltenhagen |

Pool : represents a Participant in a Collaboration

Accepted
Customer

Register

Rejected

!46
Chapter I : Introduction 4. Process Models
Business Process Modeling Notation | Master 2 STIN • BPA • 2019 • Boltenhagen |

Lan : A Lane is a sub-partition within a Process, sometimes within a Pool, and will
extend the entire length of the Process

Accepted

Log in
registered
Already
Customer

Rejected
customer

Register Rate App

New

!47
Chapter I : Introduction 4. Process Models
Business Process Modeling Notation | Master 2 STIN • BPA • 2019 • Boltenhagen |

Lan : is a sub-partition within a Process, sometimes within a Pool, and will extend
the entire length of the Process

Accepted

Log in
registered
Already
Customer

Rejected
customer

Register Rate App

New

!48
Chapter I : Introduction 4. Process Models
Business Process Modeling Notation | Master 2 STIN • BPA • 2019 • Boltenhagen |

Data Object : provides information about what Activities require to be performed

and/or what they produce

Input Output Collection

!49
Chapter I : Introduction 4. Process Models
Business Process Modeling Notation | Master 2 STIN • BPA • 2019 • Boltenhagen |

Message : is used to depict the contents of a communication between two Participants

Group : is a grouping of graphical elements that are within the same Category

Text Annotation : provides additional text information to the reader

Text

!50
Chapter I : Introduction 4. Process Models
Business Process Modeling Notation | Master 2 STIN • BPA • 2019 • Boltenhagen |

Create the BPMN model of the following process:

To be hired in the I.N.V.E.N.T.E.D. company, one has to apply by sending a email with a CV. When the
recruitment manager receives the document, he/she analyses it and either send a email for a rejection
either call the applicant for an appointment. The process continues if the applicant goes to I.N.V.E.N.T.E.D.
to meet the team leader. Then the team leader approves or not the application and informs the
recruitment manager who will contact the applicant. Moreover, the team leader has to fill a form. If
succeed, the applicant will sign the contract.

!51
Chapter I : Introduction 4. Process Models
Business Process Modeling Notation | Master 2 STIN • BPA • 2019 • Boltenhagen |

Applicant
Apply to Get an
Rejection Sign the
I.N.V.E.N.T.E.D appointment contract

CV
X
I.N.V.E.N.T.E.D. company
Recruiter

Analyse
X Meeting with
the CV the applicant
Received
the application

Inform
Team Leader

the recruiter

Meeting with the

applicant + Fill a form

!52
Chapter I : Introduction 4. Process Models
(Labeled) Petri Nets | Master 2 STIN • BPA • 2019 • Boltenhagen |

Definition ((Labeled) Petri Net) : A Process Model defined by a Labeled Petri

Net is a tuple N = <P,T,F, λ, ∑>, where P is the set of places, T is the set of
transitions, F : (P * T) U (T * P) is the flow relation, ∑ is an alphabet of actions and
λ: T -> ∑ U τ, labels every transition by an action or as silent (τ).

!53
Chapter I : Introduction 4. Process Models
(Labeled) Petri Nets | Master 2 STIN • BPA • 2019 • Boltenhagen |

places T2
C T4
τ transitions
T3
T1 F T7
S A
T5
B
Activity names
markings T6 T8
D S : start the process
G
C : write a comment
F : upload a file
…
!54
Chapter I : Introduction 4. Process Models
(Labeled) Petri Nets | Master 2 STIN • BPA • 2019 • Boltenhagen |

Definition (Firing transition) : In a Petri Net, a transition t is enable and can fire if all the
places before the transition (noted •t) have a token. When a transition fires, all the tokens
in •t are removed and all the places after the transition (noted t•) are marked.
T2 T2
C T4 C T4

T3
τ T3
τ
T1 F T7 T1 F T7
S A S A

T5 Firing T1 T5
B B
T8 T8
T6 T6
D D
G G

!556
Chapter I : Introduction 4. Process Models
(Labeled) Petri Nets | Master 2 STIN • BPA • 2019 • Boltenhagen |

Definition (Reachability) : In a Petri net N, a marking m’ is reachable if we can

find a firing sequence <t1…tn> of transitions to get the marking m’. We note
m[t1…tn>m’.
T2
Find a firing sequence to get that marking : C T4
τ
T3
< T1, T2, T5 > T1 F T7
<T1, T3, T5 > S A
<T1, T5, T3 > T5
<T1, T5, T2 > B
T8
T6
D
G

!56
Chapter I : Introduction 4. Process Models
(Labeled) Petri Nets | Master 2 STIN • BPA • 2019 • Boltenhagen |

Definition (Workflow system) : A workflow system <N, m0, mf> is a Petri net N with an
initial marking and a final marking such as m0 and mf contain a unique place with
respectively, no entry arc and no exit arc (variant definitions exist).
T2 T2
C C

T3 T3
T1 F T7 T1 F T7
S A S A

T5 T5
B B

Initial marking Final marking

!57
Chapter I : Introduction 4. Process Models
(Labeled) Petri Nets | Master 2 STIN • BPA • 2019 • Boltenhagen |

T2
T2
C T4

T3
τ C T4
τ
T3
T1 F T7
T1 F T7
S A
S A
T5
T5
B
B
T8
T6 T8
D T6
G D
G

Initial marking Final marking

< T1, T2, T5, T7, T8 > < T1, T3, T6, T4, T8 > < T1, T3, T6, T4, T8 >
<T1, T3, T5, T8, T7 > <T1, T2, T6, T4 > <T1, T2, T6, T4 >
<T1, T6, T3, T4 > <T1, T2, T6, T4, T8 > <T1, T2, T6, T4, T8, T8, T8, T8 >
<T1, T6, T3, T4, T8, T8 > <T1, T2, T6, T4, T8, T8, T8 > <T1, T2, T6, T4, T8, T8, T8, T8, T8, T8 >
!58
Conclusion
| Master 2 STIN • BPA • 2019 • Boltenhagen |

Chapter I. Chapter II. Chapter III. Chapter IV.

Introduction
Introduction Model discovery Conformance Data Analytics

1. What is Process Mining ?

2. What and where can be processes/systems ?
3. Data as Event log
4. Process Models
1. Process mining fill the gap between Data Science, Business Process Management and Formal Methods
2. Anywhere, log are hugely recorded in 2019
3. Three notions are needed : case, timestamp and activity
4. BPMN models and Petri nets

!59