Research On The Method Of Change Domain Based On Merged Model Of Petri Net Configuration

With the complexity and diversity of business needs, business process management is becoming increasingly difficult. In the business process systems, a multiple of models need to be merged in order to meet the complex business systems functional requirements. However, the merged model always runs abnormally. After identifying the change regions or the suspected changes caused the abnormality, we can change the area to be repaired to accomplish desired goals ultimately. Therefore, analysis of the change domain has become the important prerequisite in the area of business process management. While one of the effective methods to determine the change is to extract the digest of input models.Currently, the extracting digest is focused on the reversibility and trace-ability of the merged model, mainly considering merging the input models into a single model, and then anglicizing the digest. However, the obtained digest is always not complete. Other words, it is not very accurate, what’s more, it does not combine the change regions of the upset models and the responsible party model closely. Just because of the above backgrounds, this paper, based on series of basic properties and extended process tree of Petri net with configuration places, considers combining the comprehensive behavior profile and the comprehensive structural tree of the input models to mine the digest. And a method of optimizing change region of business process model based on Petri net and module is proposed, mostly importantly, with the trace-ability of the merged model, the change regions and the responsible party model are combined closely. The main contributions of this paper are followed.(1) A method of extracting the digest of the merged model based on Petri net configuration is proposed. The digest is the most recurring fragments across a collection of input models with the same target, which is intended for analysts to optimize these fragments through identifying the same and different properties. But the obtained digest is not complete. A novel method of digest extraction based on Petri net configuration is presented in this paper. Firstly, algorithms for the comprehensive behavior profile and comprehensive process tree of Petri nets are given, and the digest is mined through the behavior and structure aspects. Then, the digest is optimized based on configuration transition. Finally, a simple example is used to verify the correctness of the method.(2) An analysis of the suspected changes based on the merged model is proposed. Due to the gradual technical upgrading, the users’ concepts are becoming open-minded. However, there is always a correlation main line between the series models. In order to avoid excessive number of input models in one-time extraction and thus result in excessive computing, an analysis method of the change region based on merged Petri model in the paper. Firstly, several complex models are merged into a model with identifiers. Then, an algorithm for extracting the digest from the merged model is given, and the changes are optimized based on module behavior profile. Finally, the s changes are analyzed by the trace-ability of the merged model.