Resilient Process Modeling and Execution Using Process Graphs

Abstract

Unreliable communication challenges the execution of business processes. Operation breaks down due to intermittent, delayed or completely failing connectivity. The widely used Business Model and Notation 2.0 (BPMN) provides limited flexibility to address connectivity-related issues and misses a technique to verify process resilience. This paper presents a graph-based approach to identify resilient process paths in BPMN business processes. After a process-to-graph transition, graph-based search algorithms such as shortest-path and all-paths are applied to list resilient configurations. Evaluation of the approach confirms reasonable performance requirements, good scalability characteristics, and a significant resilience improvement. Recommendations for the practical insert of algorithms and metrics conclude the paper. © 2021, Springer Nature Switzerland AG.