A Colored Petri Net Executable Modeling Approach for a Data Flow Well-Structured BPMN Process Model
Fenglan Huang,Feng Ni,Jiang Liu,Fan Yang,Jiayi Zhu
DOI: https://doi.org/10.1109/access.2022.3198969
IF: 3.9
2022-09-04
IEEE Access
Abstract:BPMN process models have been widely used in software designs. The BPMN process models are characterized by a static graph-oriented modeling language and a lack of analytical capabilities as well as dynamic behavior verification capabilities, which not only leads to inconsistencies in the semantics of the BPMN process models, but also leads to a lack of model error detection capabilities for the BPMN process models, which also hinders the correctness verification and error correction efforts of the models. In this study, we propose an executable modeling approach for CPN-based data flow well-structured BPMN (dw-BPMN) process models, and consider both control-flow and data-flow perspectives. First, we present a formal definition of the dw-BPMN process model, which is formally mapped into a CPN executable model in three steps: splitting, mapping and combining. Then, we discuss four types of data flow errors that can occur in the model: missing, lost, redundant, and inconsistent data error. To detect these four data flow errors, we propose a detection method based on the execution results of the CPN model. Subsequently, we propose correction strategies for these four data flow errors. Finally, a dw-BPMN process model of a robot's temperature detection system for COVID-19 prevention and control in a kindergarten was used as an example to verify the validity of the method.
computer science, information systems,telecommunications,engineering, electrical & electronic