Curriculum Computer Science and Innovation for Societal Challenges, XXXVI series  Grant sponsor IBM, Dipartimento di Matematica - Università degli Studi di Padova Supervisor s Massimiliano De Leoni Contact alessandro.padella@phd.unipd.it

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 focus of Process Mining has largely been on Process Discovery and Conformance Checking, its ultimate goal is to improve the processes that organizations perform. A first form of improvement is related to the so-called Process-aware Recommender System, which – given a KPI of interest, aims to predict which executions are not going to meet the minimum KPI requirements, and consequently to determine which recovery actions to enact to put those executions back “on track”. However, if a large portion of the executions requires recovery actions, it is likely that the procedure behind the process itself is structurally characterized by issues that prevent from reaching the KPI requirements. In this case, instead of recovering single process executions, the model that encode the process procedures need to be repaired at the design time to ensure that future executions are going to perform satisfactorily. My project will touch upon both of scenarios: implementing a Recommender System to put running cases back on track, and repairing the process model to obtain a new process that structurally will lead to better KPI values. Changing single executions at run-time or the model at the design-time require commitment from process stakeholders, who need to be explained of the reason of these changes. Therefore, I will also investigate techniques to associate recovery actions and model changes with human-intelligible explanations to convince about the goodness of the changes. During the project, Process-Mining techniques will be ensembled with Machine- and Deep-Learning techniques. The techniques will result in software tools with advanced Graphical User Interfaces to provide process actors and stakeholders with the suggestions for improvement and the corresponding explanations. It follows that techniques from human-computer interaction will also be leveraged on.