基于三维行为关系图的模型一致性检测方法

Consistency Detction Method of Models Based on Three-dimensional Behavior Relation Graph

在业务流程模型的相似性分析过程中,有时会出现环结构。已有的方法一般不考虑环结构,忽略了环结构对模型一致性分析的影响。以Petri网的多重变迁集为基础,提出一种新的一致性度测量方法。首先通过分析5种行为序列关系来刻画Petri网的变迁之间的相互关系,给出了三维行为关系图的概念;然后对Petri网的三维行为关系图间的关系进行研究,提出了基于Petri网的三维行为关系图的一致性检测方法。理论分析和实例测试均表明了该方法的有效性。

In the similarity analysis process of business process models,sometimes there may be loop structure in the business process model.Existing methods do not consider the loop structure,and ignore the influence of loop on consistency.A behavior consistency measure method was proposed based on transition multi-sets of Petri nets.Firstly,this paper analyzed five kinds of behavior relations of transitions,and proposed a three-dimensional behavior relation graph based on Petri net branching processes to compare the behavior relations between two models.Secondly,by analyzing the relations between two three-dimensional behavior relation graphs,this paper proposed a consistency measure method based on three-dimensional behavior relation graphs.The theoretical analysis and specific examples show that the method is very effective.