Utility-based control flow discovery from business process event logs

Abstract

Process Aware Information Systems (PAIS) are IT systems which support business processes and generate event-logs as a result of execution of the supported business processes. Fuzzy-Miner (FM) is a popular algorithm within Process Mining which consists of discovering a process model from the event-logs. In traditional FM algorithm, the extracted process model consists of nodes and edges of equal value. However, in real-world applications, the actors, activities and transition between activities may not be of equal value. In this paper, we propose a utility-based Fuzzy Miner (UBFM) algorithm to e ciently mine a process model driven by a utility threshold. The term utility can be measured in terms of pro t, value, quantity or other expressions of user's preference. The focus of this paper is to take into consideration the statistical (based on frequency) and semantic (based on user's objective) aspect into account. We conduct experiments on realworld dataset and demonstrate the effectiveness of our approach.