Cross-Instance Regulatory Compliance Checking of Business Process Event Logs

• Published in: IEEE transactions on software engineering Vol. 49; no. 11; pp. 4917 - 4931
• Main Authors: van Beest, Nick, Groefsema, Heerko, Cryer, Adrian, Governatori, Guido, Tosatto, Silvano Colombo, Burke, Hannah
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.11.2023; IEEE Computer Society
• Subjects: Business; Business process; Combinatorial analysis; Complexity; Compliance; Computational complexity; Computational modeling; cross-instance; event log; instance-spanning; Monitoring; Patent applications; Regulation; Regulations; Runtime; runtime verification; Sequences
• ISSN: 0098-5589; 1939-3520
• DOI: 10.1109/TSE.2023.3319086

Event logs capture the execution of business processes, such that each task is represented by an event and each individual execution is a chronological sequence of events, called an event trace. Event logs allow after-the-act and runtime analysis of deployed business processes to verify whether their execution complies with regulations and business requirements. Checking the compliance of a single sequence of events in a trace is straightforward and a number of approaches have been proposed to address this. However, some regulations or business rules span multiple process instances, requiring a cross-instance analysis. In order to check whether such requirements are maintained at all times, multiple traces need to be analysed together, which can result in a combinatorial computational complexity. In this article, we present a novel approach that efficiently checks runtime regulatory compliance based on event logs, while supporting cross-instance rule evaluation and extensible function evaluation over sequences of attribute data values. The efficiency and applicability of the proposed method is tested in a two-pronged evaluation, showing a significant improvement over existing techniques with respect to capabilities as well as computational complexity. The approach presented in this paper is subject to a patent application, with patent number WO2021/248201.