PhD Defense: 'Efficient alignment-based conformance checking of procedural and declarative process models'

Process mining is a set of methods that examine event data to reveal how processes work in practice and how they can be optimized. One of the main areas within this field is conformance checking, which compares observed behavior with process models to identify and measure deviations. Due to the complexity of large real-life processes, existing conformance checking techniques fail to produce results for some cases and exhibit poor performance for others.

To address this, new scalable algorithms for computing optimal alignments between event logs and process models were developed. The contributions include an efficient alignment algorithm for procedural models, a fast approach for declarative models, and a technique for data-aware declarative models with support for flexible data conditions. These advances improve both speed and diagnostic power, enabling organizations in healthcare, IT, finance, and other sectors to better understand and optimize their processes.

Supervisors: Manuel Lama Penín and Manuel Mucientes Molina