Riccardo Galanti

Ritratto Riccardo Galanti

Computer Science and Innovation for Societal Challenges, XXXV series
Grant sponsor

myInvenio S.r.l.

Massimiliano DeLeoni

Luciano Gamberini

Project description
Process Mining aims to extract knowledge about business processes that are executed within organizations, by reasoning on the actual process executions, which are recorded in so-called event logs. Event logs group process events in traces, each of which contains the events related to a specific process-instance execution (e.g. for a given customer). While the main focus of Process Mining has been on Process Discovery, Process Mining aims at a broader application. In this project, this event-log knowledge will be exploited by Process-aware Recommender systems (PAR), which aim to predict how the executions of process instances are going to evolve in the future, to determine those that have higher chances to not meet desired levels of performance (e.g., costs, deadlines, customer satisfaction). I will develop PAR systems that, after alerting on the customers with higher chances of being dissatisfied, will suggest the mitigation actions to put in place to try to recover those customers. This will rely on simulating (a subset of) the possible customer-journey continuations (e.g. those observed in the journeys of similar customers). Then the system will recommend those continuations that will likely lead to higher satisfaction, according to a predictive model, which is learnt from the data recorded in the event log. I will have to return results that non-technical users can interpret, analyse and employ. In particular, the root causes that drive certain improvements need to be provided and understood; otherwise, the process’ stakeholders and actors will unlikely trust them and, hence, use. Here, I will develop techniques that leverage on the state of the art of process mining and machine-learning. Since the aim is to also provide models that are understandable by non-technical stakeholders, the aspects related to visualization, human-computer interaction and cognitive science will play a predominant role; here, we will exploit the recent advances in the field of “Explainable AI”. The intermediate and final research results will be developed in a form of software modules that I will integrate in the process-mining suite of myInvenio. These modules ought to become solid software that can be marketed in a short time and with little effort.