LAB OF SOFTWARE ARCHITECTURES
AND INFORMATION SYSTEMS
FACULTY OF INFORMATICS
MASARYK UNIVERSITY, BRNO
Martin Macák
macak@mail.muni.cz September 26, 2019
PV226: Process Mining seminar
Outline
2
1. Basic overview of Process Mining
2. Course introduction
Processmining
• Process-centric data analysis
• What really happened in the past?
• Why did it happen?
• What is likely to happen in the future?
• When and why do people deviate?
• How to redesign a process to improve it?
• ...
3
Processmining
• Typically working with event logs which represent processes
• These logs have to contain cases (sequences of events)
4
Processmining
• Each event has:
• caseId
• activity
• timestamp (optional)
• resource (optional)
• other data (optional)
5
Processmining
• Sometimes, the mapping is not clear
• For example, the name of the worker can be:
• resource
• activity
• caseId
6
Analysisofthepast
1. Process discovery techniques
• From the event log, we create a model that represents how the process was
executed in reality
• Model can be represented as a petri net, activity diagram, BPMN diagram, …
2. Conformance checking techniques
• We can check the deviations from the created model in historic data
7
Processdiscovery
8
Processdiscoverychallenges
• Concurrency
• Loops
• Noisy behavior
• No negative examples in the log
• Too many allowed behaviors
9
Processdiscoveryactivities
• Explore processes at run-time
• Discover process models
• Compare the model of desired behavior with the model of reality
• Check the deviations in historic data
• Promote the model that shows the desired behavior
10
Addingadditionalperspectives
• Control flow is not the only perspective
• We can enhance the existing process models with:
• Social network analysis
• Organizational structures
• Resource behavior analysis
• Time perspective
• Decision points mining
• …
11
Additionalperspectives
12
• We can add many others
• We can combine them to the integrated model
• Our model is enhanced, we might get better results
Conformancechecking
• We can use the existing model to identify
deviations in the behavior from logs
13
Analysisofthepresent
• Also called operational support
• We use our model to analyze running cases
• We can:
• Detect deviations in real-time data using the model of the desired behavior
• Do real-time predictions (prob. of success, remaining time,…)
• Make recommendations
14
Operationalsupport:Detectdeviations
• We consider only the partial trace of a particular case
• We want immediate response when the deviation occurs
a) Token-based replay
b) Business rules
15
Detectdeviations:Token-basedreplay
• Check the conformance with the model
16
Detectdeviations:Businessrules
• Specific rules we want to follow
• To define them, we can use Declare
• Constraint-based workflow language that uses graphical notations
and semantics based on Linear Temporal Logic
• Example:
• a and b cannot happen in the same case
• a cannot happen before c has happened
• every d have to be eventually followed by a or b
17
Operationalsupport:Predict&Recommend
• We can apply data mining techniques (supervised learning, …)
• Examples of predictions:
• Total cost of the current case
• Total service time for the current case
• Probability of meeting the deadline
• Remaining flow time
• Examples of recommendations:
• Minimize the total costs
• Maximize the number of accepted cases
• Minimize resource usage
• Minimize the remaining flow time
18
PV226 Course information
• e-learning (recommended: 2. – 7. week)
• https://www.coursera.org/learn/process-mining
19
PV226 Course information
• Project
• Application of Process Mining to a problem
• You can come up with your own topic, set your own difficulty
• You can work in pairs
• We will have a meeting (24.10.?) where we will discuss your topics
• 5.12. – presentation of your work in Lasaris seminar (A319)
• Optional consultations of your project / email communication 
• Examples of project types:
• Process discovery in tool Disco (https://fluxicon.com/disco/)
• Process analysis in tool ProM (http://www.promtools.org/)
• Process analysis using Python (https://github.com/pm4py/pm4py-source)
• Survey research paper about the specific usage of Process Mining
20
Questions?
Sources
• Presentation based on the book Process Mining: Data Science in
Action
• https://www.springer.com/gp/book/9783662498507
• Use school VPN and you can download it! 
22