Validating and Enhancing Declarative Business Process Models Based on Allowed and Non-occurring Past Behavior

摘要

Contemporary organizations have been implementing a wide variety of process-aware information systems in order to streamline their operations. The current organizational environment is often characterized by a multitude of internal and external directives which impose restrictions through business rules on the operations and as such define declarative business process models. We present a twofold methodology which can be applied towards the validation and enhancement of process models which are expressed in a declarative form in order to improve their correctness and completeness. Our approach is based on validation of real-life behavior using rule property checking, and on allowed behavior by the process model which was not encountered in real-life cases by matching rule-generated rejected activity occurrences with absent behavior in the event log. Our methodology retains the ability to correspond retrieved findings to decision-makers in a clear and comprehensible manner (i.e. in the form of a new rule), rather than a formal revision of an implemented procedural model, which is a significant advantage when considering business-IT alignment concerns.