Researches On Conformance Checking Of Process Models And Model Repair Based On Workflow Net

Business processes described by formal or semi-formal models have been wildly realized via information systems.The availability of event logs generated from these systems and the requirements of appropriate process models are increasing.On the one hand,most organizations make substantial efforts to document their processes;while on the other hand,these processes leave footprints in their information systems.Process mining technique relates observed behavior to modeled behavior.Event logs generated from these systems are probably not consistent with the existing models due to insufficient design of the information system or the system upgrade.Approaches to check conformance between the process model and event logs are proposed.The information from conformance checking can be used for improving the process model.Petri net is used as a modeling tool in this thesis.The main work and innovation are as follows:1.Deviation detection between event logs and a process model without loop.The inconsistencies between an existing process model and the corresponding event logs are called deviations.To detect the deviations,the business processes and the deviations are formally defined based on Petri nets.An approach to detect the deviations between event logs and models,which are described by causal nets are proposed.The deviations detection method is extended for a process model that described by a general Petri net.Some clinical processes in a healthcare information system are used as a case study to illustrate our work.Experimental results show the efficiency and scalability of the proposed approach.2.Deviation detection between event logs and a process model with loop.The definition of the type of the loop structure in a process is introduced.A method to judge the type of the loop structure is given.The ways to determine the fragment of the loop structure and the repetition sub-trace are presented.An efficient approach to detect deviations between the process model with loop structures and event logs is proposed.A standard clinical process of cerebral hemorrhage in a healthcare information system is used as a case study to illustrate our work.Experimental results show the effectiveness and efficiency of the presented approach.3.Aligning observed and modeled behavior based on workflow decomposition.For the complex and large process models and event logs,divide and conquer method is a good strategy to implement the conformance checking.An efficient approach is proposed to align the observed and modeled behavior.Based the refine process structure tress of the process model,a decomposition approach is proposed to divide large process models and event logs into several separate parts that can be analysed independently.An alignment approach based on a state equation method in Petri net theory enhances the performance of conformance checking4.Process model repair based on firing sequencesBy the information from the conformance checking,the process models can be improved.Based on the type of the deviations,process models can be repaired from three aspects:removing behavior,adding behavior and changing behavior.The method to determine and insert a "redo" activity in a process model is given.A real-world process model of hospitalization in a hospital and the corresponding event logs are employed to evaluate the proposed approaches.The correctness and effectiveness of the proposed methods are illustrated through experiments.