Handling Uncertainty in Predictive Business Process Monitoring with Bayesian Networks

Abstract

Process mining is a growing and promising study area that enables business processes analysis based on their observed behaviour recorded in event logs. Since process mining is a relatively new research area, there are still several challenges, especially related to the emerging big data technologies and methods. Recently, a wide literature about predictive process monitoring techniques has become available. Despite the emergence of predictive business process monitoring application and the exploitation of machine learning algorithms, Bayesian Networks have been underexplored. In this paper, we propose the use of Bayesian Networks for handling uncertainty in predictive business process monitoring, thus providing predictive capabilities in process modelling and execution. The proposed approach is demonstrated in two case studies.