Lecture summary
Lecture 1: General knowledge BPM and first stage of BPM lifecycle
Business process management is:
- A systematic, structured approach to analyze, improve, control and manage processes with
the aim of improving the quality of products and services.
- A customer-focused approach to the systematic management, measurement and
improvement of all company processes through cross-functional teamwork and employee
empowerment.
- Supporting business processes using methods, techniques and software to design, enact,
control and analyze operational processes involving humans, organizations, applications,
documents and other sources of information.
Second industrial revolution (1865 -1900): Taylor’s scientific management:
- Rationalization
- Division of labor
- Specialist
- Manager
- Functional units
Drawbacks: departmental suboptimization, client dissatisfaction, lack of workforce commitment and
inflexibility.
Process wave (1990-1999):
- Improve customer service by improving cross-functional processes, using IT and
redesigning/reengineering/innovating process structures.
- Process awareness & broad scope
- Improves overall performance, inter-departmental cooperation and ‘esprit de corps’.
BPM lifecycle:
1. Process identification:
- Designation phase
o Identifying organization processes and their relationships
- Evaluation phase:
o Prioritizing the identified processes for process management activities
Process architecture: a representation of the processes in an organization and their relations
, - Which processes exist?
- Where does each process start/end?
- How are the processes related?
- At which level of detail should they be described?
Concrete and practical approach for process architecture:
1. Identify case types
a. Customer type
b. Product type
c. Service type
d. Channel
2. Identify functions
a. Something a company can do
3. Construct case/function matrix
a. Split up when:
i. different flow-objects
ii. change in flow-object multiplicity
iii. change in transactional state
iv. logical separation in time (happens once per…)
v. logical separation in space
vi. logical separation in …
vii. using a reference model
viii. significant difference in functions per case type
4. Identify processes
Lecture 2: Second stage of BPM lifecycle
2. Process discovery:
- Interviews-based
- Workshop-based
- Evidence-based (documentation, data or observation)
Process mining: a family of data analysis techniques aimed at distilling valuable knowledge about a
process from its event log.
Alpha algorithm:
- Extracting ordering relations among log activities
- Building the process model describing the inferred relations
1. Identify all tasks
2. Identify first tasks
3. Identify last tasks
4. Identify x -> y, x -> (y # z), (x # y) -> z relations
5. Remove duplicate relations
6. Connect first and last tasks to start and end events
7. Construct the flow (note the duplicate arrows)
, + Intuitive logic - Short loops
+ Simple algorithm - Noise
+ Can discover fundamental process constructs - Strict assumptions
- Advanced patterns
Other process discovery techniques:
- Heuristic based
- Genetic algorithms
- Region space theory
- …
Conformance analysis: checking how well the actual executions match the process model. How well
is a business process followed in practice? What is the quality of a model produced by an algorithm?
- Fitness: extent to which the log can be correctly replayed
¿ Fitting traces
o
¿ Total traces
1 m 1 r
o
2 (1− + 1−) ( )
c 2 p
- Appropriateness: extent to which only traces that are in the log can be replayed
m l m l
1 |S F ∩ S F| 1 |S P ∩ S P|
o +
2 |SmF| 2 |S lP|
Lecture 3: Modelling
Process model purposes: communication (with all
stakeholders) (1) and execution (by workflow/BPM
system) (2).
Syntactic quality: conform to the rules of the
notation (correctness)
Semantic quality: making true statements about the
world (validity & completeness)
Pragmatic quality: usable by stakeholders
(understandability)
Lecture 1: General knowledge BPM and first stage of BPM lifecycle
Business process management is:
- A systematic, structured approach to analyze, improve, control and manage processes with
the aim of improving the quality of products and services.
- A customer-focused approach to the systematic management, measurement and
improvement of all company processes through cross-functional teamwork and employee
empowerment.
- Supporting business processes using methods, techniques and software to design, enact,
control and analyze operational processes involving humans, organizations, applications,
documents and other sources of information.
Second industrial revolution (1865 -1900): Taylor’s scientific management:
- Rationalization
- Division of labor
- Specialist
- Manager
- Functional units
Drawbacks: departmental suboptimization, client dissatisfaction, lack of workforce commitment and
inflexibility.
Process wave (1990-1999):
- Improve customer service by improving cross-functional processes, using IT and
redesigning/reengineering/innovating process structures.
- Process awareness & broad scope
- Improves overall performance, inter-departmental cooperation and ‘esprit de corps’.
BPM lifecycle:
1. Process identification:
- Designation phase
o Identifying organization processes and their relationships
- Evaluation phase:
o Prioritizing the identified processes for process management activities
Process architecture: a representation of the processes in an organization and their relations
, - Which processes exist?
- Where does each process start/end?
- How are the processes related?
- At which level of detail should they be described?
Concrete and practical approach for process architecture:
1. Identify case types
a. Customer type
b. Product type
c. Service type
d. Channel
2. Identify functions
a. Something a company can do
3. Construct case/function matrix
a. Split up when:
i. different flow-objects
ii. change in flow-object multiplicity
iii. change in transactional state
iv. logical separation in time (happens once per…)
v. logical separation in space
vi. logical separation in …
vii. using a reference model
viii. significant difference in functions per case type
4. Identify processes
Lecture 2: Second stage of BPM lifecycle
2. Process discovery:
- Interviews-based
- Workshop-based
- Evidence-based (documentation, data or observation)
Process mining: a family of data analysis techniques aimed at distilling valuable knowledge about a
process from its event log.
Alpha algorithm:
- Extracting ordering relations among log activities
- Building the process model describing the inferred relations
1. Identify all tasks
2. Identify first tasks
3. Identify last tasks
4. Identify x -> y, x -> (y # z), (x # y) -> z relations
5. Remove duplicate relations
6. Connect first and last tasks to start and end events
7. Construct the flow (note the duplicate arrows)
, + Intuitive logic - Short loops
+ Simple algorithm - Noise
+ Can discover fundamental process constructs - Strict assumptions
- Advanced patterns
Other process discovery techniques:
- Heuristic based
- Genetic algorithms
- Region space theory
- …
Conformance analysis: checking how well the actual executions match the process model. How well
is a business process followed in practice? What is the quality of a model produced by an algorithm?
- Fitness: extent to which the log can be correctly replayed
¿ Fitting traces
o
¿ Total traces
1 m 1 r
o
2 (1− + 1−) ( )
c 2 p
- Appropriateness: extent to which only traces that are in the log can be replayed
m l m l
1 |S F ∩ S F| 1 |S P ∩ S P|
o +
2 |SmF| 2 |S lP|
Lecture 3: Modelling
Process model purposes: communication (with all
stakeholders) (1) and execution (by workflow/BPM
system) (2).
Syntactic quality: conform to the rules of the
notation (correctness)
Semantic quality: making true statements about the
world (validity & completeness)
Pragmatic quality: usable by stakeholders
(understandability)
