Research Of Process Diagnose System Base On The Mining Of Enent Log

With the development and application of workflow technology, a variety of business process management system (BPMS) or workflow management systems (WFMS) are getting more and more concern of the modern enterprise. As a software platform of process definition, implementation and tracking, BPMS effectively support the business process management throughout the life cycle, which including process design, process configuration, process entactment and process diagnostic phase. While most of the process management system focused primarily on the function of process design and configuration phase, they ignoring the process of data collection and analysis during the operational phase, which leading directly to the process diagnostic phase abandoned. The process diagnostic phase is based on analysis and monitoring the actual process and provide input of for the next stage of the process re-engineering to achieve process improvement, that is to say business process reengineering (BPR) and continuous process improvement (CPI), it is of the important phase. The lack of effective tools for process improvement would bring hardness for the practical application of process management. Therefore, it is worthy of studying on how to make the diagnostic process support process continuous improvement.Process mining is a branch of data mining in the area of process management, since the late 90s last century, it firstly proposed by the Agrawal. Since then it gain widespread concern among the academia and industry circle. Process mining techniques is to collect and use performance data obtained through information system. a variety of process mining algorithms canbe used in the event log records to gain the order and relations of actual activities or tasks in process instances, which can support process design and analysis. This paper focuses on the application of process mining in the diagnosis phase to monitor and analyze process. Introducing process mining based diagnostic method to an important subsystem in process management software system which is named Process Diagnosis System and building the process Diagnosis system model as well as BPMS's integration framework. The research is of great significance for improving the analysis and design level of BPMS, also the continuous process improvement within enterprise. The main results can be summarized as:First, On perspective of BPM life-cycleand based on application of business process management system status and problems, the paper defined the connotation and extension of process diagnosis.Second, from different dimensions of process mining, which are the control flow dimension, organizational dimension and the case dimension, the paper discussees its application in the establishment of process models, analysis of process performance, construction of organizational structures and social networks in a systematic way. On this basis, it divides the field in event log according to requirements of different application types of process mining. Then an example is described.Third, it identifys the process issues for mining related to the process diagnosis phase. Afterwards, it highlights the event log -based process mining diagnostics method and expands the preparation stage, model construction and model analysis phase of the implementation strategy.Fourth, it builds process diagnosis system's functional model and framework.The function of each module are analyzed. Finally the paper builds the whole business process management system framework. With details of the XML format, each node is analyzed. Entity-relationship (E-R) diagram of required fields in event log is constructed and as well as the database table.Fifth, the existing process mining tool as ProM framework is used to simulate the relevant functional modules of process diagnosis system, it introduces two process cases to verify the result.