Process Modeling and Bottleneck Mining in MXML-based Course Training Event Logs

Parham Porouhan


The paper is divided into two main parts. In the first part of the study, we applied two process mining discovery techniques (i.e., alpha and heuristic algorithms) in order to extract knowledge from an event log previously collected from an information system —during a project management training course at a private university in Thailand. The event log was initially consisted of 548 process instances and 5,390 events in total. Using alpha algorithm we could reconstruct causality (in form of a Petri-net) from a set of sequences of events, while through heuristic algorithm we could derive XOR and AND connectors (in form of a C-net) based on the dependency, significance and correlation metrics/coefficients. The results showed 80% of the applicants/students managed to achieve the project management certificate successfully, while 6% of them fail to achieve any certificate (after maximum number of 3 attempts to re-take the course). Surprisingly, 14% of the applicants (77 cases) neither achieved a certificate nor failed the course. Therefore, in the second part of the study, we used conformance checker and performance analysis techniques in order to further analyze the points of non-compliant behavior (i.e., bottlenecks) for every case in the log. Subsequently, we could detect and identify the number of missing tokens, as well as the activities that were not enabled, or remained enabled. 


Process Mining, Model Discovery, Alpha algorithm, Heuristic algorithm, Process Simulation, ProM, Bottleneck Mining, Conformance Checker, Performance Analysis, MXML.

Full Text:



Professor James G. Williams (University of Pittsburgh)


Dr. Nipat Jongsawat (Rajamangala University of Technology Thanyaburi)



  • There are currently no refbacks.

Progress in Applied Science and Technology

Faculty of Science and Technology