Learning to act: a Reinforcement Learning approach to recommend the best next activities

The rise of process data availability has led in the last decade to the development of several data-driven learning approaches. However, most of these approaches limit themselves to use the learned model to predict the future of ongoing process executions. The goal of this paper is moving a step forward and leveraging data with the purpose of learning to act by supporting users with recommendations for the best strategy to follow, in order to optimize a measure of performance. In this paper, we take the (optimization) perspective of one process actor and we recommend the best activities to execute next, in response to what happens in a complex external environment, where there is no control on exogenous factors. To this aim, we investigate an approach that learns, by means of Reinforcement Learning, an optimal policy from the observation of past executions and recommends the best activities to carry on for optimizing a Key Performance Indicator of interest. The potentiality of the approach has been demonstrated on two scenarios taken from real-life data.