An Automatic Business Process Modeling Method Based on Markov Transition Matrix in BPM

Recently a novel process modeling method is presented in the field of business process management, which is called as process mining. Process mining technology can automatically deduce process models based on given mining algorithm from large numbers of process logs. The new modeling technology makes modeling efficiency, objectivity and quality be greatly improved than before. However the research of this new modeling technology is currently on primary stage. And the known process mining algorithms are not enough perfect. Therefore a new process mining algorithm is designed in this paper. In the algorithm a Markov transition matrix is set up based on process logs. And according to the matrix the eight mining rules of process logical relations are designed. This new mining algorithm can not only effectively reduce the effect of logs quality on process mining but also greatly enhance the automatic process modeling capability.