Download Will van der Aalst - EIT Digital Master School

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Document related concepts
no text concepts found
Transcript
Text
Text
Wil van der Aalst
Professor Information Systems TU/e
A new profession is emerging, just like
computer science in the early 1980-ties!
Industry and Society need Data Scientists!
t
t
EIT-Digital Data Science Major
5 universities involved
EIT-Digital Data Science Major
3 entry universities
EIT-Digital Data Science Major
5 exit universities
(specializations)
EIT-Digital Data Science Major
5 specializations
Process Mining in High Tech Systems,
Healthcare, Visual Analytics, or Big
Software at TUE
Distributed Systems & Data Mining
for Really Big Data at KTH
Design, Implementation, and Usage
of Data Science Instruments at TUB
Internet of Things (IoT) at UPM
Multimedia & Web Science for Big
Data at UNS
Statistics on the first intake
• Total number of students: 46
• EU students 54%
• Age :20-29
• Female students: 17%
DSC
DSC/e
t
t
Data Science Center Eindhoven
http://www.tue.nl/dsce/
DSC/e: Competences and Research Programs
28 groups and 420+ people involved
Analysis: How to turn data into real value
(models, answers/decisions, and
visualizations/insights)?
Probability and Statistics
Data Mining
Stochastic Networks
Process Mining
Visualization
Enabling technologies: How to get the
data and deal with computational/
infrastructural challenges (big data and
hard questions)?
Internet of Things
Large-Scale Distributed
Systems
Data-Intensive Algorithms
[RP1] Process Analytics:
Improving Service While Cutting Costs
[RP2] Customer Journey:
Correlating Events to Learn and Influence Customer Behavior
Context: Why are we using data science,
does itMaintenance
have the intended effect,
will
[RP3] Smart
&and
Diagnostics:
people accept it?
Safeguarding Availability
Human and Social Analytics
[RP4] Quantified Self:
Security, Ethics,
ImprovingPrivacy,
Performance
andand
Well-Being
Governance
Data-Driven
[RP5] Data Value
andOperations
Privacy:
Management
Economic and Legal Aspects of Data Science
Data-Driven Innovation and
Business
[RP6] Smart Cities:
Ensuring Safety and Convenience for Citizens
[RP7] Smart Grids:
Data Intensive Infrastructures
11
Data Science Flagship (Philips & DSC/e)
4 Strategic topics
•
•
•
•
Data Driven Value Propositions
Healthcare Smart Maintenance
Optimizing Healthcare Workflows
Continuous Personal Health
4 TU/e departments
16 PhD students
30 Data science specialists
Many more organizations
• BrandLoyalty
• Vanderlande Industries
• ASML
• SynerScope
• Magnaview
• Fluxicon
• Adversitement
• Rabobank
• ING
• SAP
• IBM
• PwC
• AMC
• …
Process
Mining
Example:
Process Mining as the Bridge
Between Data Science and
Process Science
Process Mining: Spreadsheet for behavior
resource
row = event
case
identifier
activity
name
timestamp
• Input: events (“things that
have happened”)
• Mandatory per event:
• case identifier
• activity name
• timestamp/date
• Optional
• resource
• transaction type
• costs
• …
Process Mining: Spreadsheet for behavior
208 cases
5987 events
74 activities
Process Mining: Spreadsheet for behavior
batching for activities
“opstellen eindnota” and
“archiveren”
Loesje van
der Aalst
desire line
Process Discovery
Process Mining: Spreadsheet for behavior
process discovery
NO
modeling
needed!
Process Mining: Spreadsheet for behavior
process discovery
NO
modeling
needed!
74 act.
11 act.
3 act.
process
model
Conformance Checking
event data
desire line
very safe
system
Conformance Checking
Process Mining: Spreadsheet for behavior
conformance checking
?
discovered or
hand-made
Process Mining: Spreadsheet for behavior
conformance checking
fitness of
93.5%
Process Mining: Spreadsheet for behavior
conformance checking
final inspection is
skipped 40 times
Process Mining: Spreadsheet for behavior
conformance checking
move on model
(something should have
happened, but did not)
move on log
(something happened that
should not happen)
Process Mining: Spreadsheet for behavior
performance analysis
NO
modeling
needed!
bottleneck
average
flowtime is
1.92 months
Process Mining: Spreadsheet for behavior
performance analysis
waiting time of
15.74 days
NO
modeling
needed!
Process Mining: Spreadsheet for behavior
NO
modeling
needed!
animating reality
real cases
Process Mining: Spreadsheet for behavior
animating reality
16 cases are
queueing
Process Mining: Spreadsheet for behavior
What?
Deviations
Where?
Why?
time
costs
…
Conclusion
•Need for Data Scientists!
•Wonderful Data Science Master Program with 3
entry points and 5 specializations
• Ask Farideh Heidari ([email protected]) for details!
•Zoomed-in on the Data Science ecosystem in
Eindhoven: Data Science Center Eindhoven (DSC/e)
•Zoomed-in on a particular Data Science topic:
Process Mining (linking processes and data)
32
masterschool.eitdigital.eu
More information?
http://www.masterschool.eitdigital.eu/programmes/dsc/
https://www.coursera.org/course/procmin/
http://www.processmining.org/
http://www.tue.nl/dsce/
http://vdaalst.com/
Related documents