Predictive and Prescriptive Business Process Monitoring with Reinforcement Learning

Nowadays, more and more process data are automatically recorded by information systems, and made available in the form of event logs. In this context, process mining enables business processes analysis based on their observed behaviour recorded in event logs by providing the means to discover, monitor, and improve processes. During the last years, there has been an explosion of research works proposing approaches which adopt machine learning algorithms in order to provide flexibility, explainability and predictive capabilities. Further, prescriptive business process monitoring has the credentials to increase data analytics maturity and lead to optimized decision making, ahead of time, for business performance improvement. In this paper, we propose an integrated predictive and prescriptive business process monitoring approach with the use of Reinforcement Learning (RL). The proposed approach is evaluated in the context of a use case from the banking sector.