Font Size: a A A

Business Process Model Configuration Mining Analysis Based On Business Capabilities

Posted on:2020-10-11Degree:MasterType:Thesis
Country:ChinaCandidate:L YingFull Text:PDF
GTID:2428330575971903Subject:Applied Mathematics
Abstract/Summary:PDF Full Text Request
Business process management(BPM)is a systematic and structured method for managing,analyzing,controlling,and improving business processes.BPM can be divided into two main aspects:one is the establishment and specification of the business process model,the other is the optimization of the business process model.However,both aspects of BPM require a certain amount of mining of the business process.Therefore,business process mining has become an important application of business process management.However,in the actual mining process,there is often some objective or subjective factors,which cause problems in the execution of the log sequence.For example,some transitions in the model are prohibited from being executed at runtime(blocking transition),and some transitions are hidden at runtime(hidden transition).Therefore,in order to effectively restore specific behaviors in the process model,mining configuration information such as blocking transitions has become one of the key points of process mining.At present,the mining of business process configuration information is mainly aimed at the hidden transitions of business processes,but seldom involves the mining of blocking transitions.The mining of hidden transitions is to analyze and mine the hidden transitions of process model under the complete log,and rarely consider the blocking transition mining problem of process model under complete or incomplete log.Because the current approach is only for mining certain configuration information of a process model,it is not able to fully display the specific behavior of a process model,and to a certain extent,it also seriously affects the applicability and rationality of the process model.In this paper,on the one hand,the idea of entropy are used to mine the blocking transition of business process under the complete log,and the idea of conditional behavior probability are for mining the blocking transition under the incomplete log.But on the other hand,by combining the methods of business capability,in the case of a given configurable process,the change region of the configurable model is analyzed from the perspective of business capability,which is beneficial to verify the rationality of the configuration information for a given configurable process model.The main contributions of this paper are as follows:(1)An entropy-based business process blocking transition mining method is proposed.Look for blocking transitions in the business process model for a given complete event log.The existing method is mainly to analyze the behavior profile relationship between logs under the complete log and find the blocking transitions of the process model.This method can not accurately locate the blocking transitions.In this paper,based on the Merging operation of Petri net,the incremental model and the initial model are merged,and the possible active position of blocking transition is analyzed.And then the precise active position of blocking transition is determined by entropy analysis method.This method is not only easy to calculate,but also helpful to find out the precise location of blocking transition.(2)A method of mining business process blocking transition based on conditional behavior probability is proposed.Look for blocking transitions in the business process model for a given incomplete event log.The current method mainly solves the mining problem of process model blocking transition under the complete event log,but little research on the incomplete event log.In this papper,firstly,establish an incremental model based on the incomplete event logs classified by the co-occurrence class,and use the behavioral profile relationship of the region to merge the model.And then query the activity in the process model through the pull rule,and find that there may be active positions of blocking transition.Finally,the precise activity position of the blocking transition is found by the calculation of the conditional behavior probability.This method can effectively and quickly mine the blocking transition of the process model under incomplete logs.(3)A method for changing region analysis of configurable business process models based on business capabilities is proposed.The existing methods mainly focus on the analysis of business process model change region,but pay little attention to the problem of configurable process model change region.For the change region analysis problem of configurable process model,this paper combines the theoretical knowledge of business capability,and the business capability aspect of the transition analyzes the change region of the configurable process model.Firstly,according to the categories and attributes of the business capabilities,the two configurable source models are configured and merged.And then analyze between the obtained business capability-annotated configurable modes to find the change region of the configurable process model.Finally,the mapping relationship between process models finds the exact change region of the configurable source model.This method is helpful to improve the correctness of process model configuration.Figure[22]table[18]reference[79]...
Keywords/Search Tags:Event log, Entropy, Conditional behavior probability, Blocking transition, Business capability, Configurable process model, Change region
PDF Full Text Request
Related items