8. Process Mining Flashcards
Today’s challenges for businesses (6)
- high customer expectations
- shorter product cycles
- mass customization -> variety of customers from all
over the world have different demands - digital transformation
- disruptive technologies
- global supply chains
=> remaining competitive and on top of the market for the long-run is difficult nowadays:
"Processes are the engine of every experience"
What is a process? (Definition, Problem in practice)
= a business process is a collection of related, structured activities or tasks that in a specific sequence produce a service or product
Problem: In practice, there are many things that can go wrong!
- problems mean direct costs for the business
(employees, material, etc.)
- problems mean indirect costs/opportunity costs
(customer satisfaction, time factor, etc.)
Old vs. New Way: Process Mapping (2 Characteristics and Problem)
Process Mapping:
- subjective and partial -> consultants reconstruct
processes by doing personal interviews, on
premises analysis and come up with a model
- very lengthy and costly
Problem: Full complexity of the problem is difficult to understand, only fractions/symptoms/snapshots are being analyzed and adopted => one time understanding
Old vs. New Way: Process Mining (3 Characteristics)
Process Mining
- objective and complete by having a data driven approach
- immediate information available (time efficient) and self-servicing
- continuously enhancing -> real time data upload tells us continuously about the process as oppose to being restricted to only snapshots with Data Mapping
Interdisciplinary Approach (5 Factors for each components)
- Process Science
- Process Management
- Automation
- Process Control
- Process Improvement
- Operations Management
- Data Science
- Data Mining
- Statistics
- Machine Learning
- Databases
- Predictive Analytics
=> intersection of both sciences defines Process Mining
Process Mining
= data analytics tool to reconstruct, analyze, and improve business processes based on log data from transactional IT systems
-> Process Mining bridges the gap between model-based analyses and data-centric analysis techniques.
Businesses heavily rely on IT systems (Data Storage, Examples)
Business data is stored in a variety of IT systems such as workflow management systems, ERP and CRM systems, supply chain management systems etc.
Event Logs (Basis of Process Mining, Definition of Event Log, Minimum, Characteristics (5)
- > structured data is the basis of Process Mining as well as looking at the digital footprints
- > an event log stores the data that is required for Process Mining (at the minimum, the event log covers three columns: caseID, activity name, timestamp)
Characteristics:
- CaseID: indicates which process instance the event
belongs to (a case usually consists of multiple
events)
- Activity: Describes the action that is captured by the
event
- Timestamp: indicates the time when the event took
place
- Trace: A sequence of events, ordered by timestamp
that belong to the same case
- Variant: The traces of all the different cases with the
same activity sequence
Process Mining Techniques (3)
Discovery = uncovers how processes are executed vy replicating a process model from the event log
Conformance Checking = compares the “as-is” process from the log data with the “to-be” model
Enhancement = uses insights from discovery and conformance to improve the process
Business Impact of Process Mining (4 overall impacts, 3 direct impacts)
Efficiency
- perfect order
- automation
Quality and fulfillment
- rework
- rejections
Speed and Agility
- On-time delivery
- throughput time
Compliance
- “Should-be” vs. “As-Is”
- Process monitoring
=> affects direct costs
=> affects indirect costs
=> affects opportunity costs
Advantages of Process Mining (Regarding organizational change)
+ reduce process cost
+ increase revenue
+ reduce working capital
+ improved risk-management
