Analysis And Optimization Of Change Domain In Business Process Models Based On Petri Net

Petri net exhibits abundant particular background. Since it as modeling tool, business process problems can be projected into a network model, and thus it can translate an analysis of business process problems into the process models. In the model application represented by e-commerce business process analysis, not only analyze model’s accessibility, and study its execution behavior. Especially, when the source model is so difficult to obtain or low maneuverability that need to reconstruct a target model to analyze the problem, it becomes particularly important to analyze the consistency of the target model. As a result, the behavior analysis of the model can be summarized into two aspects. On the one hand, the behavior consistency analysis of the target model, including the establishment and seeking of change domain, etc. On the other hand, the analysis of choosing behavior execution path. In order to improve the consistency degree of the target model and to ensure the optimality of execution path, the content of this paper mainly studies the following four aspects.(1) Aiming at the problem of consistency comparison between the source model and the target model, the developed methods determine the difference between models according to analyze the matching degree of the structure or behavior between model tasks, has some limitations to solve complex interaction sequences existing in the model. The method of analyzing the behavioral deviation between business process models based on the Petri net kinetic expressions is proposed, using expressions to calculate the behavioral distance of each pair of sequences and the behavioral inheritance between models, and the calculation method of the behavioral deviation between business process models is given out, in the paper. This method, combining the matching degree of behaviour with model structure, realizes the analysis of the behavior deviation between models.(2) Aiming at the problem of searching the change domain of the target model, former studies search the change range using change nodes, from sequences perspective, which have a lot of limitations. The analysis method of searching the smallest change domain based on the Petri net kinetic expressions is proposed, and the method of looking for suspected non-efficient sequences of business processes from a model perspective as well as the method of determining the smallest change domain of business process model are given out in the paper.(3) Aiming at the problem of modifying the change domain of target model using the method of modular substitution, the developed researches of modular substitution only consider the restriction for activities from control flow, ignoring the role of the data flow of models, which have many limitations. The method of analyzing the modular substitution in business process models based on differential Petri net is proposed, in the paper. Also, Liveness detection in control flow and data flow of models is examined by the evolution expression of differential Petri net through semantics, and stability analysis of substitution module and its discrete and continuous parts is done by differential Petri net, respectively.(4) Aiming at the problem of searching the optimal path of behavior execution in business process, The existing proposed methods were to search the optimal path on the basis of static analysis, ignoring the impact of behavioral constraints on the effectiveness of the execution behavior, which have some limitations. The analysis method of searching the optimal path based on modal behaviour effective range in the paper, using the ordering relation of Petri nets behavior profile to describe constraints and determine behavioral effective ranges. This method uses behavioral effective ranges to substitute for the existing fixed value, easy to better describe behavior constraints of business processes and effective behavior under constraints.