Process Mining of RFID-Based Supply Chains

Abstract

Process mining facilitates the analysis of business processes by extracting a process model from event logs. Most mining algorithms perform well on single-system event logs that explicitly refer to a process instance or case. However, in many operational environments such case identifiers are not directly recorded. In supply chain processes there are even further challenges, since different identification numbers and numerous aggregation steps prevent individual work items to become traceable as a case. In this paper, we investigate how the EPCglobal standard for processing Radio Frequency Identification (RFID) events can make supply chain data accessible for process mining. Our contribution is an algorithm that is able to deal with challenges of case identification and focus shifts. We present a prototypicalimplementation and use a process based on the Supply Chain Operations Reference (SCOR) Model to evaluate our implementation.