Process Mining and Predictive Monitoring: an overview
Chiara Ghidini, Senior Research Scientist @ Fondazione Bruno Kessler
According to one of the reference courses on the topic, Process Mining is Data Science in Action: the “missing link between model-based process analysis and data-oriented analysis techniques”. Indeed it consists of a series of techniques that enable the (i) discovery, (ii) monitoring, and (iii) improvement of real processes (i.e., not assumed processes) leveraging the knowledge contained in today’s (information) systems and represented in the format of so-called event logs.
In this presentation I will provide an overview of the main components of process mining, namely: the automated process discovery from event logs, the checking of conformance (or deviations) of the process model and the event log, and the improvement of models based on the actual process executions contained in the event log. I will also illustrate in detail some of the work we carry out on predictive process monitoring, that is, the ability of foreseen the future unfolding of current process executions.