Process models are used to describe and reason about the execution of a process (e.g., package delivery) where a process instance (package), also called as a case, moves through the system. A case in a process is often subject to interaction with other cases and/or resources (e.g. deliverer), impacting the workflow. Event logs record which activity is executed at which time for which case and which resources are involved. These can be used to analyze whether and where the actual process behavior recorded in an event log deviates from the behavior prescribed by a process model both regarding individual cases and multiple cases simultaneously. Deviations can be of different nature: skipping prescribed steps, swapping their order, multi-tasking when it is forbidden to multitask, etc.
The Master student will analyze real-life data from several processes, including a production process from Omron (omron.nl) and the loan application process from BPI 2012/2017, and apply an existing technique to expose deviations. The student will develop a method for bringing information about deviations to the business user by generating insights which deviations frequently occur together, which deviations can be predicted based on the choices made in the process or on the resources involved, etc. The project builds on the knowledge and skills obtained in the Foundations of Process Mining course combined with the knowledge and skills from courses on data mining and/or artificial intelligence.