Business Process Mining
Lecture 3: Concepts & Capabilities of
Process Mining
Course Outline
Introduction
Process performance measurement & dashboards
Process mining
• Automated process discovery
• Conformance checking
• Performance mining
• Variant analysis
Simulation and “what-if” process mining
Data extraction & preparation for process mining
Predictive process monitoring
Business case analysis & project management for process mining
Trends: prescriptive monitoring, causal process mining, robotic process mining
2
How does Process Mining work?
Process Mining
An event log is a collection of timestamped
event records. Each event record tells us
something about the execution of a work
item (and hence a task) of the process (e.g.,
that a task has started or has been
completed), or it tells us that a given
message event, or other relevant event has
event log
occurred in the context of a given
case in the process.
4
Structure of a Business Process Event Log
Event log
Process Mining
Automated Process
Discovery discovered
process model
event log
8
Automated Process Discovery
An automated process discovery technique
takes as input an event log and produces as
output a process model that captures the Process Map
behavior of the log in a representative way. (directly follows graph)
CID Task Time Stamp …
13219 Enter Loan Application 2007-11-09 T [Link] -
13219 Retrieve Applicant Data 2007-11-09 T [Link] -
13220 Enter Loan Application 2007-11-09 T [Link] -
13219 Compute Installments 2007-11-09 T [Link] -
13219 Notify Eligibility 2007-11-09 T [Link] -
13219 Approve Simple Application 2007-11-09 T [Link] -
13220 Compute Installements 2007-11-09 T [Link] -
BPMN process model Notify
Retrieve Rejection
Applicant
… … … … Data
Enter Loan
Application
Approve
Compute Simple
Installments Application
Notify
Eligibility
Approve
Complex
9 Application
Process Mining
Automated Process
Discovery discovered
process model
event log
Conformance
Checking
✓/
Business rules /
normative model 15
Conformance checking
Modeled process Actual process
(Expected: 8 hours) (In reality: 18 hours)
Rule-Based Conformance Checking
•Oftentimes, we don’t have a full process model available
•Instead, we have a set of compliance rules, e.g.
• Certain tasks MUST be executed in every case
• Certain tasks can be executed at most once
• When a task A occurs, another task B must also occur (within e.g. 2 hours)
• Two tasks A and B CANNOT be executed by the same person (four-eyes
principle)
17
Process Mining
Automated Process
Discovery discovered
process model
Performance
Mining
event log
Enhanced
process model
Conformance
Checking
✓/
Business rules /
normative model 19
Performance Mining
•Process mining tools typically support performance mining
in thee ways:
• Performance-enhanced process maps
• Dashboards
• Log animation
20
Peformance-Enhanced Process Map
Activity label
(and its duration)
Directly-follows relation between two
labels (with waiting time)
Bottleneck node
(all incoming
arcs are slow)
Process Mining
23
Automated Process
Discovery discovered
process model
Performance Variants Analysis
Mining
event log
Enhanced Difference
process model diagnostics
event log’
Conformance
Checking
✓/
Business rules /
normative model 23
Variant Analysis
Simple cases Complex cases
Find differences in 75% 25%
performance and control-
flow between multiple
variants of a process
A process variant is a subset
of the executions of a
process corresponding to a
given product, customer,
etc.
24
Variant Analysis
In process mining tools, such as Apromore, variant analysis can be
approached as follows:
1. First, use filtering to “slice” the event log into multiple logs (one
per variant).
2. Then compre the multiple logs using:
▪Side-by-side comparison of process maps or process models
▪A comparative multi-log dashboard
▪A multi-log animation
25
Summary: Four Process Mining Capabilities
EVENT LOGS
