Exploiting logical formulations for workflow trace monitoring

7^th July 2015, 13:00 Ashton Lecture Theater
Dr Chiara Ghidini
Fondazione Bruno Kessler (FBK)
Trento
Italy

Abstract

Modern information systems that support complex business processes generally maintain significant amounts of process execution data, particularly records of events corresponding to the execution of activities (event logs). The capability to analyze such event logs to (predictively) monitor workflow (aka, process) executions has gone to notably increase in the last decades and has brought to the diffusion of several reasoning-based tools for the analysis of process executions.
In this talk I will present two different scenarios where the logical formulation of business processes is exploited, possibly in combination with learning techniques, to monitor process executions. In the first scenario I will present an approach to recover incomplete/missing information about process executions, relying on a logical formulation in terms of a planning problem which is applied to post-mortem data (process model and event logs).
In the second scenario I will present an approach to analyze such event logs in order to predictively monitor business constraints during business process execution. At any point during an execution of a process, the user can define business constraints in the form of linear temporal logic rules. When an activity is being executed, the framework identifies input data values that are more (or less) likely to lead to the satisfaction of each business constraint.
For both scenarios I will also present a validation based on real-life data pertaining to the management of birth data in an italian hospital (scenario 1) and treatment of cancer patients in a large hospital (scenario 2).