Introduction to Process Mining and its applicability to enterprise technology customer support. Please visit my blog at http://www.haimtoeg.com/?p=1881 for further discussion.
2. Process mining is a discipline positioned at the
crossroads of computational intelligence, data
mining, and process modeling and analysis. The idea
of process mining is to discover, monitor and improve
real processes by extracting knowledge from the
event logs produced as part of on-going business
activity.
Adapted from Process Mining Manifesto
What Is Process Mining?
4. Some
reasons
support
may
adapt
well
to
Process
Mining:
* Highly
disciplined
* Heavily
instrumented
and
regimented
* Fairly
well
documented
processes
* Culture
of
metrics
and
improvement
Why
Support?
5. * Business:
Do
More
With
Less
* Efficiency
and
Effectiveness
* Reduce
cost
/
case
or
/
customer
-‐>
Operational
metrics
* Case
life,
backlog,
time
since
last
touched
* Cases
/
installation
* Improve
effectiveness
-‐>
Result
metrics
* NPS,
CSAT
* Correlate
and
analyze
data
* Prioritized
action
list
Process
Improvement
Goals
6. * Greater
granularity
<-‐>
abstraction
* Drill
down
to
individual
process
elements
* Segregate
data
by
operational
element
(geo,
product)
* Combine
process
elements
to
study
overall
activity
Improving
Process
Improvement
9. Weaknesses?
* Reliance
on
generalized
metrics
* Anecdotal
review
of
one
or
small
number
of
process
instances
* Limited
ability
to
drill
down
and
investigate
* Not
much
abstraction
ability
either
* Event
logs
used
only
sporadically
and
reactively
Familiar?
11. * Model
the
executed,
real-‐world,
processes
* Identify
and
review
process
variants
* Discover
informal
collaboration
and
knowledge
* Investigate
and
measure
every
action
and
transition
* Support
for
Lean
perspectives
* Workload
/
People
perspectives
Discovery
Benefits
12. Discovery
Results
Case
flow
via
activities:
Process
flow:
• Activity
Frequency
Bottlenecks:
• Critical
activities
and
paths
• Investigate
operational
metrics
17. Combine
your
existing
and
discovered
process
models:
* Find
degree
of
conformance
between
the
two
* Understand
repeat
offences
and
take
corrective
action
* Systems
* Processes
* Metrics
* People
* Estimate
cost
and
benefit
of
alignment
vs.
no
action
Conformance
Benefits
19. * Leverage
existing
data
using
recently
developed
leading
edge
techniques
* Review
executed
processes
in
detail
to
get:
* Excessively
repeated
activities
* Process
bottlenecks
* Critical
resources
* Knowledge
hoarding
* Drive
systematic,
prioritized
process
and
organizational
improvement
Process
Mining
for
Support