Business processes run at the core of an organisation's value creation and are often the target of optimisation efforts. Organisations aim at adhering to their optimised processes. However, deviations from the optimised process still occur and may potentially impede efficiency in process executions. Conformance checking can provide valuable insights regarding past process deviations, but it cannot identify deviations before they occur. Outcome-oriented predictive business process monitoring (PBPM) provides a set of methods to predict process outcomes, e.g., key performance indicators. We propose an outcome-oriented PBPM method for predictive deviation monitoring using conformance checking and deep learning to draw the most out of the two domains. By leveraging early intervention, the method supports the proactive handling of deviations, i.e., inserted and missing events in process instances, to reduce their potential harm. Our evaluation shows that the method can predict business process deviations with high predictive quality, particularly for processes with fewer variants.
Weinzierl, Sven; Dunzer, Sebastian; Tenschert, Johannes Christian; Zilker, Sandra; and Matzner, Martin, "Predictive Business Process Deviation Monitoring" (2021). ECIS 2021 Research Papers. 131.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.