Automated Discovery of Workflow Models from Hospital Data

Workflow nets, a subclass of Petri nets, are known as attractive models for analysing complex business processes. In a hospital environment, for example, the processes show a complex and dynamic behavior, which is difficult to control; the workflow net which models such a complex process provides a good insight into it, and due to its formal representation offers techniques for improved control. We propose a method whose main advantage consists in discovering the workflow Petri nets automatically from process logs. We illustrate the functioning of our method on simulated hospital process logs, containing information about medical actions over time. The results of our experiments indicate that this method is able to discover processes whose underlying models are acyclic and sound WF nets, involving parallel, conditional and sequential constructs. We argue that solutions have to be found for cyclic and freechoice /non-free-choice workflow nets