Diagnosing Conformance Between Object-Centric Event Logs and Models

Abstract

Conformance checking techniques can diagnose discrepancies between the behavior observed in an event log and the behavior allowed by a reference model. These discrepancies might indicate undesirable deviations, fraud, inefficiencies, or other issues in a business process. Today, most organizations use object-centric systems such as ERP and CRM systems, which generate and store data in an object-centric manner. Unfortunately, existing conformance checking techniques often employ process-centric models, e.g., Petri nets, which typically consider process instances in isolation (ignoring interactions among them) and are more focused on the behavioral perspective of processes. Accordingly, the conformance diagnosis results are only on the behavioral perspective and cannot deal deviations related to the data perspective and interactions, failing to reveal some useful insights related to the undiscovered deviations. In this paper, we aim to check conformance based on object-centric logs and models, which combines data and behavior perspectives, and the resulting diagnostic information contains conformance problems that would have remained undetected using conventional techniques. At last, our conformance checking approach is evaluated based on the data generated in a real ERP system.

Keywords

• Process mining
• conformance checking
• object-centric
• OCEL
• OCBC models
• ERP