An operational support approach for Mining Unstructured Business Processes

Abstract

The refined process mining framework contains a set of activities that use extracted information from event logs, discovered models and normative ones. Among these activities, we find those dealing with running events in a Structured Business Process (SBP) context, which are the Detect, the Predict and the Recommend activities. These three activities are nominated as operational support system that performs well on SBP while, it stills a challenging task for an Unstructured Business Process (UBP), because of its complex structure. In this regard, a special interest is given to the use of existing process mining techniques to analyse unstructured processes, from the extraction of a process model based on event data to recommendations at a later stage. To this end, we propose the orchestration of process mining activities into an UBP operational support approach, through the following phases: 1.Preparing Normative model, 2.Detect violations, 3.Preparing predictive model and Predictions and 4.Preparing the recommender model and Recommendations.