Tackling Dierent Business Process Perspectives

Business Process Management (BPM) has emerged as a discipline to design, control, analyze, and optimize business operations. Conceptual models lie at the core of BPM. In particular, business process models have been taken up by organizations as a means to describe the main activities that are performed to achieve a specific business goal. Process models generally cover different perspectives that underlie separate yet interrelated representations for analyzing and presenting process information. Being primarily driven by process improvement objectives, traditional business process modeling languages focus on capturing the control flow perspective of business processes, that is, the temporal and logical coordination of activities. Such approaches are usually characterized as “activity-centric”. Nowadays, activity-centric process modeling languages, such as the Business Process Model and Notation (BPMN) standard, are still the most used in practice and benefit from industrial tool support. Nevertheless, evidence shows that such process modeling languages still lack of support for modeling non-control-flow perspectives, such as the temporal, informational, and decision perspectives, among others. This thesis centres on the BPMN standard and addresses the modeling the temporal, informational, and decision perspectives of process models, with particular attention to processes enacted in healthcare domains. Despite being partially interrelated, the main contributions of this thesis may be partitioned according to the modeling perspective they concern. The temporal perspective deals with the specification, management, and formal verification of temporal constraints. In this thesis, we address the specification and run-time management of temporal constraints in BPMN, by taking advantage of process modularity and of event handling mechanisms included in the standard. Then, we propose three different mappings from BPMN to formal models, to validate the behavior of the proposed process models and to check whether they are ...