PRISM - A predictive risk monitoring approach for business processes

Abstract

Nowadays, organizations face severe operational risks when executing their business processes. Some reasons are the ever more complex and dynamic business environment as well as the organic nature of business processes. Taking a risk perspective on the business process management (BPM) lifecycle has thus been recognized as an essential research stream. Despite profound knowledge on risk-aware BPM with a focus on process design, existing approaches for real-time risk monitoring treat instances as isolated when detecting risks. They do not propagate risk information to other instances in order to support early risk detection. To address this gap, we propose an approach for predictive risk monitoring (PRISM). This approach automatically propagates risk information, which has been detected via risk sensors, across similar running instances of the same process in real-time. We demonstrate PRISM’s capability of predictive risk monitoring by applying it in the context of a real-world scenario.