Diagnosing Industrial Business Processes: Early Experiences

摘要

Modern day enterprises rely on streamlined business processes for their smooth operation. However, lot of these processes contain errors, many of which are control flow related, e.g., deadlock and lack of synchronization. This can provide hindrance to downstream analysis like correct simulation, code generation etc. For real-life process models other kind of errors are quite common, - these are syntactic errors which arise due to poor modeling practices. Detecting and identifying the location of occurrence of errors are equally important for correct modeling of business processes. We consider industrial business processes modeled in Business Process Modeling Notation (BPMN) and use graph-theoretic techniques and Petri net-based analyses to detect syntactic and control flow related errors respectively. Subsequently based on this, we diagnose different types of errors. We are further able to discover how error frequencies change with error depth and how they correlate with the size of the subprocesses and swim-lane interactions in the models. Such diagnostic details are vital for business process designers to detect, identify and rectify errors in their models.