Back to the Order: Partial orders in streaming conformance checking

Most organizations are built around their business processes. Commonly, these processes follow a predefined path. Deviations from the expected path can lead to lower quality products and services, reduced efficiencies, and compliance liabilities. Rapid identification of deviations helps mitigate such risks. For identifying deviations, the conformance checker would need to know the sequence in which events occurred. In this paper, we tackle two challenges associated with knowing the right sequence of events. First, we look at out-of-order event arrival, a common occurrence in modern information systems. Second, we extend the previous work by incorporating partial order handling. Partially ordered events are a well-studied problem in process mining, but to the best of our knowledge it has not been researched in terms of fast-paced streaming conformance checking. Real-life and semi-synthetic datasets are used for validating the proposed methods.