Process Mining

• Explain the behavior of a Petri net (e.g., “token game”) and corresponding basic properties (e.g., deadlock-free, bounded)
• Model a given business process using Petri nets in a correct way
• Compare the different process mining family of techniques (e.g., control-flow discovery and conformance checking) and select which fits given scenarios
• Interpret the quality criteria for evaluating process mining outcomes, in particular concerning control-flow discovery algorithms
• Compare the available algorithms and tools for process mining techniques and determine which fits given problems
• Utilize the state-of-the-art of available technologies, both commercial products and open source tools
• Construct event logs starting from raw data recordings or as simulations of process models preserving certain properties (to be used in controlled environments)
• Extract the control-flow of the activities recorded in an event log with commercial and open source tools
• Calculate the conformance of an event log with respect to a given reference process model
• Assess (business) related questions using process mining techniques
• Explain the current trends in process mining research

The course consists of lectures and tutorials/exercises. All theoretical aspects will be supported by practical sessions consisting of both exercises and software tests.

The topics presented are:
▪ Formalizations and graphical representations of business processes
▪ Extraction of data for process mining purposes
▪ Simulation of models for process mining purposes
▪ Algorithms for control-flow discovery
▪ 4 quality dimensions of process mining
▪ Algorithms for conformance checking

02105/02158/02141, Understanding of algorithms and basic data structures, together with some general computer science modeling knowledge. Experience with notions of concurrent programming is also advisable.

Lectures, group work and student presentations.