Abstract: To ensure the correctness of a complex BPMN collaboration model,not only complex elements such as multiple instances and subprocesses should be covered,but also data flow errors should be detected as well as control flow errors.However,Business process Modeling Notation (BPMN2.0) lacks formal semantics,which hinders the verification of the correctness of the model.In this paper,the definition of hierarchical Colored Petri Net with arc weight (HCPN) is given,which can not only formally represent data flow,but also model the structure of multiple instances and sub-processes.Furthermore,a formal mapping method from BPMN collaboration model to HCPN model is proposed.Then,based on the arc weight of HCPN model,the definitions of missing,losing and redundant data flow errors are given,and the corresponding detection algorithms are proposed.Finally,an automated modeling and verification framework is designed,and a case study is given to illustrate the effectiveness of the proposed approach.

Key words: colored Petri net, BPMN collaboration model, data flow error, model verification, formalization

摘要： 为了保证复杂BPMN协作模型的正确性,不仅要涵盖多实例和子进程等复杂元素,还要在检测控制流错误的同时检测数据流错误。但业务流程建模标注（BPMN2.0）缺乏形式化语义的描述,这对模型正确性的验证造成了阻碍。故给出了一种具有弧权重的层次化着色Petri网（HCPN）的定义,它既可以对数据流进行形式化表示,又可以对多实例和子进程结构进行建模。进一步提出了从BPMN协作模型到HCPN模型的形式化映射方法。然后基于HCPN模型的弧权重给出了缺失、丢失和冗余三种数据流错误的定义,并提出了对应的检测算法。最后设计了一个自动化建模与验证的框架,通过一个案例研究说明了该方法的有效性。

关键词: 着色Petri网, BPMN协作模型, 数据流错误, 模型验证, 形式化