Discovering pareto-optimal process models: a comparison of MOEA techniques

Process mining aims at discovering the workflow of a process from the event logs that provide insights into organizational processes for improving these processes and their support systems. Ideally a process mining algorithm should produce a model that is simple, precise, general and fits the available logs. A conventional process mining algorithm typically generates a single process model that may not describe the recorded behavior effectively. Recently, Pareto multi-objective evolutionary algorithms have been used to generate several competing process models from the event logs. Subsequently, a user can choose a model based on his/her preference. In this paper, we have used three second-generation MOEA techniques, namely, PAES, SPEA-II, and NSGA-II, for generating a set of non-dominated process models. Using the BPI datasets, we demonstrate the efficacy of NSGA-II with respect to solution quality over its competitor algorithms.