一种改进的流程图相似度检索算法及实现

Improved business process similarity retrieval algorithm and realization

面对企业的大量业务流程管理问题,探究了基于Petri网对工作流模型的表达,对有向图邻接矩阵定义规则作了改进,成功地建立了描述流程模型间关系的邻接模型矩阵,对业务流程模型检索具有重要意义。在此基础上基于模糊理论的贴近度概念,计算邻接模型矩阵间的相似度。为了更方便地存储流程模型,提出了模型矩阵库概念。通过实例证明了该算法在流程模型相似度检索方面的合理性和有效性。

Facing with a large number of business process management problems for companies, this paper made a research on the expression of workflow model based on Petri net, and made an improvement on the definition rule of directed gragh adjacency matrix, successfully set up the adjacency matrix model to describe the relationship between process models. This was significant to the retrieval of business process model. On the basis of the concept of close degree based on the fuzzy theory, it solved the similarity of adjacency matrix models. Secondly, to storage the process model better, this paper proposed the concept of model matrix library. Finally, an example show that the algorithm is rational to solve the problem about the retrieval of process model similarity, and it is the validity of algorithm.