Discovering and Optimizing Processes from Event Logs of Document Management Systems

Modern information management systems are often driven by workflow engines, which require accurate, detailed and structured process description. Instead of theoretical modelling such processes we propose a technique which can be a connection between traditional management (derived from organizational structure) and business process management approach. In our work we present a case study where the automatic process mining techniques are applied to administrative processes in an environment of event logs recorded by document management systems (DMS). We use event log data to discover and describe process map of the organization and to model structured process descriptions needed by workflow systems. Our approach allows for the extraction of models based on facts and it is a direct connection between the process model and raw data generated by the organization. Such scenario increases consistency between the model mapped in the IT system and the reality of the organization.