Sei sulla pagina 1di 47

Chapter 7:

Process Analysis
Contents

1 Introduction

2 Process analysis and simulation

3 Process performance management and warehousing

4 Process mining

5 Process compliance

6 Summary & outlook

References
W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 2
1 Introduction

− Let us recall the different analysis perspectives

© 2015 Springer-Verlag Berlin Heidelberg

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 3


1 Introduction

− Process analysis helps to understand the operational perspective


based on process models or event data

Taking a fresh
approach to data
science
 Process Science

© 2015 Springer-Verlag Berlin Heidelberg


W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 4
1 Introduction

Goals:
The aim of this chapter is to convey process analysis methods and
techniques, in particular:
− Understanding the difference between static and dynamic process
analysis
− Introducing the concept of process warehousing
− Introducing the basic idea of process mining
− Presenting selected process mining algorithms
− Pointing to further process mining scenarios
− Summarizing the key challenges of business process compliance

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 5


1 Introduction

© 2015 Springer-Verlag Berlin Heidelberg

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 6


1 Introduction

On top of static/dynamic analysis distinction into qualitative and


quantitative analysis questions, for example:
Examples:
− Qualitative:
 Static: „Which activities are not assigned a role?“
 Dynamic: „Given a set of process instance executions, how does the
underlying process model look like?“
− Quantitative:
 Static: „Average duration of process task X based on simulation data?“
 Dynamic: „How many running instances do currently exceed the
imposed deadline?“

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 7


Contents

1 Introduction

2 Process analysis and simulation

3 Process performance management and warehousing

4 Process mining

5 Process compliance

6 Summary & outlook

References
W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 8
2 Process analysis and simulation

− Several analysis tasks at the process model, without any


simulation, for example, estimated process costs

© 2015 Springer-Verlag Berlin Heidelberg

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 9


2 Process analysis and simulation

− Dynamic analysis at design time is based on simulation data, i.e.,


artificial runs of process instances based on a given process model
− The goal is to predict their behavior
− Several input data can be typically specified, e.g.:
 Number of simulated instances
 Probabilities at alternative branchings
 Capacities of resources
 Costs and processing times for tasks (fixed, based on probability distribution)
 Strategies for resolving bottlenecks
− Expected output comprises
 Overall process duration
 Processing time/costs for certain tasks
 Number of possible process executions in given time frame

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 10


2 Process analysis and simulation

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 11


2 Process analysis and simulation

− Tools:
 Signavio
 BOC
 ARIS
 Bonapart
 CPN Tools

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 12


2 Process analysis and simulation

Process optimization
− Can be based on insights from static and dynamic analysis
− Example: eliminate identified bottlenecks
− Optimization dimensions1: cost, time, quality, flexibility
− Partly conflicting
− Structural optimizations include, e.g., eliminating, combining,
reordering activities
− Organization optimizations, e.g., empowering users (specialist 
generalist)
− No automatism, only best practices
− Important to analyze process after optimizations again
1Selma Limam Mansar, Hajo A. Reijers: Best practices in business process redesign: use and impact. Business Proc. Manag.
Journal 13(2): 193-213 (2007)

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 13


Contents

1 Introduction

2 Process analysis and simulation

3 Process performance management and warehousing

4 Process mining

5 Process compliance

6 Summary & outlook

References
W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 14
3 Process performance management and warehousing

Now real process execution data is analyzed (not simulated)


− First „level“: process monitoring
 Observe process behavior based on events emitted by the process instances
during runtime
 “Predictive monitoring that aims at forecasting potential problems during
process execution before they occur”2
 Definition and monitoring of Key Performance Indicators such as throughput
time
 Tools:
 IBM Business Modeler Advanced:
https://www.ibm.com/developerworks/downloads/ws/wbimod/index.html
 ARIS Process Performance Manager:
https://www.softwareag.com/corporate/products/aris_alfabet/bpa/aris_ppm/default
2Andreas Metzger, Philipp Leitner, Dragan Ivanovic, Eric Schmieders, Rod Franklin, Manuel Carro, Schahram Dustdar, Klaus
Pohl: Comparing and Combining Predictive Business Process Monitoring Techniques. IEEE Trans. Systems, Man, and
Cybernetics: Systems 45(2): 276-290 (2015)

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 15


3 Process performance management and warehousing

Now real process execution data is analyzed (not simulated)


− Second „level“: process warehousing
− Offline analysis of process execution data
− Data is structured in a multi-dimensional way (cf. Chapter 3)

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 16


3 Process performance management and warehousing

Dimension instance InstanceState

InstanceID

Dimension processType Dimension time


Process

Duration
ProcessVersion ProcessName ProcessID Day Month Quarter Year
Cost

Week

Activity Dimension organization


Dimension activityType
Duration

ActivityType ActivityName ActDefID Quality Actor Role RoleType

© 2015 Springer-Verlag Berlin Heidelberg


OrgUnit OrgUnitType
W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 17
3 Process performance management and warehousing

− Application of OLAP operations on process cube


− Application of cross-sectional analysis methods (cf. Chapter 5)

Decision tree:
The activities with a
duration above 40 are
performed by Actor XY

© 2015 Springer-Verlag Berlin Heidelberg

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 18


Contents

1 Introduction

2 Process analysis and simulation

3 Process performance management and warehousing

4 Process mining

5 Process compliance

6 Summary & outlook

References
W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 19
4 Process mining

Process mining has three key tasks3:


− Process discovery
− Process conformance
− Process enhancement

3Wil M. P. van der Aalst: Process Mining - Data Science in Action, Second Edition. Springer 2016, ISBN 978-3-662-49850-7,
pp. 3-452

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 20


4 Process mining

 Particularly exploration („finding“) process models is often a


cumbersome and errorneous task
 Are there alternatives?
 Observation: Processes are often implicitly executed (maybe
distributed over different systems)
 Prerequisite: Log data of processes available
 Process / Workflow mining offers techniques to automatically derive
process / workflow models from such log data

Prepare Ship
Start A shipment goods

Start B
End A Register
order
(Re)send
bill
Archive
order
….
Receive
payment

Process discovery Contact


customer

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 21


4 Process mining

Different process mining algorithms


− Focus on discovering the control flow, i.e., the tasks and their
relations
 Alpha Miner4
 Heuristic Miner5
 Genetic Miner6

4Ana Karla A. de Medeiros, Wil M. P. van der Aalst, A. J. M. M. Weijters: Workflow Mining: Current Status and Future
Directions. CoopIS/DOA/ODBASE 2003: 389-406
5A. J. M. M. Weijters, J. T. S. Ribeiro: Flexible Heuristics Miner (FHM). CIDM 2011: 310-317

6Wil M. P. van der Aalst, Ana Karla A. de Medeiros, A. J. M. M. Weijters: Genetic Process Mining. ICATPN 2005: 48-69

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 22


4 Process mining Alpha Miner

© 2015 Springer-Verlag Berlin Heidelberg


W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 23
4 Process mining

EXERCISE: For the following log, apply the α-algorithm and derive
the corresponding Petri Net:

- Case1: <A, C, D, E, F, G>


- Case2: <A, B, D, E, F, G>
- Case3: <A, C, D, F, E, G>
- Case4: <A, B, D, F, E, G>
- Case5: <A, B, D, E, F, G>

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 24


4 Process mining

Result:

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 25


4 Process mining
Result produced using ProM 5.2

− Alpha algorithm might lead to complex models

Alpha Miner
applied to Higher
Education Data

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 26


© 2015 Springer-Verlag Berlin Heidelberg
4 Process mining

Heuristics Miner6
1. Read a log
2. Get the set of tasks
3. Infer the ordering relations based on their frequencies
4. Build the net based on inferred relations
5. Output the net

6Weijters,AJMM (Ton), van der WMP (Wil) Aalst, und de AKA (Ana Karla) Medeiros. 2006. Process mining with the
HeuristicsMiner algorithm. Technische Universiteit Eindhoven. http://repository.tue.nl/615595.

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 27


4 Process mining

Heuristics Miner6:
Let W be an event log over T, and a, b ∈ T:
• |𝑎𝑎 > 𝑊𝑊 𝑏𝑏| is the number of times 𝑎𝑎 > 𝑊𝑊 𝑏𝑏 occurs in W,
𝑎𝑎 >𝑊𝑊 𝑏𝑏 −|𝑏𝑏 >𝑊𝑊 𝑎𝑎|
• 𝑎𝑎 ⟹ 𝑊𝑊 𝑏𝑏 =
𝑎𝑎 >𝑊𝑊 𝑏𝑏 + 𝑏𝑏 >𝑊𝑊 𝑎𝑎 +1

Insight:
The more frequently a task A directly follows another task B, and
the less frequently the opposite occurs, the higher the probability
that A causally follows B!

6Weijters,AJMM (Ton), van der WMP (Wil) Aalst, und de AKA (Ana Karla) Medeiros. 2006. Process mining with the
HeuristicsMiner algorithm. Technische Universiteit Eindhoven. http://repository.tue.nl/615595.

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 28


4 Process mining
Heuristic Miner applied
to example log

© 2015 Springer-Verlag Berlin Heidelberg

Result produced using ProM 5.2

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 29


4 Process mining
Heuristic Miner applied
Result produced using ProM 5.2
to HEP log

© 2015 Springer-Verlag Berlin Heidelberg

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 30


4 Process mining

Genetic Miner7 steps:


1. Read event log
2. Encoding individuals: causal matrix
3. Creating initial solution: all activities in the log, create causal
relationshops randomly
4. Fitness and selection of individuals: based on conformance
(preciseness and completeness)
5. Creating new offspring: mutation and crossover on causal
matrices
6. Evaluating offspring

7A.K.A.Medeiros, A.J.M.M. Weijters, und W.M.P. Aalst, “Genetic process mining: an experimental evaluation,” Data Mining
and Knowledge Discovery, 14:245-304 (2007)

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 31


4 Process mining

Genetic Miner, Step 2:


− Abstraction from Petri Nets  WHY?
− However, no information should be lost!
− Representation as causal matrix:

DEFINITION (CAUSAL MATRIX): A Causal Matrix is a tuple CM = (A, C, I, O),


where
- A is a finite set of activities,
- C ⊆ A × A is the causality relation,
- I: A  P(P(A)) is the input condition function
- O: A  P(P(A)) is the output condition function,
such that
- C = {(a1, a2)∈ A × A | a1 ∈ ∪ I(a2)}
- C = {(a1, a2)∈ A × A | a2 ∈ ∪ O(a1)}
- C ∪ {(ao, ai) ∈ A × A | ao •= ∅ ∧ •ai = ∅} is a strongly connected graph,

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 32


4 Process mining

© 2015 Springer-Verlag Berlin Heidelberg


W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 33
4 Process mining

Genetic Miner, Step 3:


− Individuals are causal matrices
− Given a log, all individuals in any population of the genetic algorithm have the
same set of activities (or tasks) A. This set contains the tasks that appear in the
log.
− The setting of the causality relation C can be done via a completely random
approach or a heuristic one.
− The random approach uses 50% probability for establishing (or not) a causality
relation between two task in A
− The heuristic approach uses the information in the log to determine the
probability that two tasks are going to have a causality relation set:
− “The more often a task t1 is directly followed by a task t2 (i.e. the subtrace “t1,
t2" appears in traces in the log), the higher the probability that individuals are
built with a causality relation from t1 to t2 (i.e., (t1, t2) ∈ C)“7
7A.K.A.Medeiros, A.J.M.M. Weijters, und W.M.P. Aalst, “Genetic process mining: an experimental evaluation,” Data Mining
and Knowledge Discovery, 14:245-304 (2007)

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 34


4 Process mining

Genetic Miner, Step 4:


− Conformance as fitness function f8

1 ∑𝑘𝑘
𝑖𝑖=1 𝑛𝑛𝑖𝑖 ∗𝑚𝑚𝑚𝑚 1 ∑𝑘𝑘
𝑖𝑖=1 𝑛𝑛𝑖𝑖 ∗𝑟𝑟𝑟𝑟
𝑓𝑓 = ∗ 1 − + ∗ 1− where
2 ∑𝑘𝑘
𝑖𝑖=1 𝑛𝑛𝑖𝑖 ∗𝑐𝑐𝑐𝑐 2 ∑𝑘𝑘
𝑖𝑖=1 𝑛𝑛𝑖𝑖 ∗𝑃𝑃𝑃𝑃

− ni is the number of traces


− ci is the number of consumed tokens
− pi is the number of produced tokens
− mi is the number of missing tokens
− ri is the number of remaining tokens

8Anne Rozinat, Wil M. P. van der Aalst: Conformance checking of processes based on monitoring real behavior. Inf.
Syst. 33(1): 64-95 (2008)

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 35


4 Process mining

Genetic Miner, Step 5:


− Elitism: certain percentage of the best individuals is copied into next
population
− Crossover:
− two parents produce two offsprings
− Selection of parents: tournament
− Mutation: Insertion of new material into current population
− Randomly choose subset and add a task (∈ A) into subset
− Randomly choose subset and remove task
− Randomly redistribute elements in the subsets of I/O into new subsets
− Example for Mutation: I(D) = {{F, B, E}, {E, C}, {G}}
− Mutation by adding tasks: {{F, B, E}, {E, C}, {G, D}}
− Mutation by removing tasks: {{F, B, E}, {C}, {G}}
− Mutation by redistribution: {{F}, {E, C, B}, {G}, {E}}

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 36


2.4 Heuristic and Genetic Miner

Genetic Miner, Step 6:


Stop Criteria:
− n generations allowed
− Fittest individual has not changed for n/2 generations in a row

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 37


4 Process mining

© 2015 Springer-Verlag Berlin Heidelberg


W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 38
4 Process mining

Process mining has three key tasks3:


− Process discovery
− Process conformance
− Process enhancement

3Wil M. P. van der Aalst: Process Mining - Data Science in Action, Second Edition. Springer 2016, ISBN 978-3-662-49850-7,
pp. 3-452

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 39


4 Process mining

Conformance checking8
− Problem: given a process model P and a set of traces T,
determine how „good“ P reflects the behavior set out by T
− Based on fitness function f
1 ∑𝑘𝑘
𝑖𝑖=1 𝑛𝑛𝑖𝑖 ∗𝑚𝑚𝑚𝑚 1 ∑𝑘𝑘
𝑖𝑖=1 𝑛𝑛𝑖𝑖 ∗𝑟𝑟𝑟𝑟
𝑓𝑓 = ∗ 1 − + ∗ 1− where
2 ∑𝑘𝑘
𝑖𝑖=1 𝑛𝑛𝑖𝑖 ∗𝑐𝑐𝑐𝑐 2 ∑𝑘𝑘
𝑖𝑖=1 𝑛𝑛𝑖𝑖 ∗𝑃𝑃𝑃𝑃

− ni is the number of traces


− ci is the number of consumed tokens
− pi is the number of produced tokens
− mi is the number of missing tokens
− ri is the number of remaining tokens
8Anne Rozinat, Wil M. P. van der Aalst: Conformance checking of processes based on monitoring real behavior. Inf.
Syst. 33(1): 64-95 (2008)

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 40


4 Process mining

© 2015 Springer-Verlag Berlin Heidelberg

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 41


Contents

1 Introduction

2 Process analysis and simulation

3 Process performance management and warehousing

4 Process mining

5 Process compliance

6 Summary & outlook

References
W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 42
5 Process compliance

− Problem: given a process model P or a set of traces T and a set


of compliance constraints C: does P or T respectively comply with
the constraints in C
− Comply means does not violate any constraint in C
− Basic dinstinction into design and runtime compliance checking
(see next slide)

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 43


5 Process compliance

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 44


Contents

1 Introduction

2 Process analysis and simulation

3 Process performance management and warehousing

4 Process mining

5 Process compliance

6 Summary & outlook

References
W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 45
6 Summary & outlook

 So far only discovery of control flow aspects


 Development of approaches for discovery of other important
aspects, for example:
 Organizational mining9, e.g., social network mining (e.g., who is
working with whom)?
 Decision mining10: Mining of transition conditions determining routing
in alternative branchings (e.g., age > 20)  combines process mining
and data mining (decision trees)
  combined approaches, see Chapter 8
 Change mining based on change logs
9W.M.P. van der Aalst and M. Song: Mining Social Networks: Uncovering interaction patterns in business processes.
International Conference on Business Process Management (BPM 2004), LNCS 3080, pages 244-260 (2004)
10A. Rozinat and W.M.P. van der Aalst: Decision Mining. International Conference on Business Process Management

(BPM 2006), Vienna, pages 420 425 (2006)

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 46


6 Summary & outlook

W.Grossmann, Stefanie Rinderle-Ma, University of Vienna – Chapter 7: Process Analysis 47

Potrebbero piacerti anche