Towards Goal-Oriented Business Process Model Repair

Process mining is a technology that extracts useful information from event logs output from information systems and utilizes it for business process analysis. Conventional process mining approaches do not take into account information about organizational goals, and thus may output process models that inhibit the goal of the organization, for example, “to reduce cost”. To solve such problems, goal-oriented process discovery method was proposed as an existing research. On the other hand, one of the methods to generate process models is process model repair. This method uses a process model and an event log as input, and repairs the existing process model to reproduce all the behaviors observed in the event log. However, the existing process model repair methods do not take into account the information about goals. Therefore, in this paper, we propose a goal-oriented process model repair method. It is easy to visually identify and compare which parts of the existing process model satisfy the goal satisfaction and which parts do not satisfy the goal. In addition, the structure of the model is simpler and closer to the initial model, and its effectiveness can be demonstrated. The effectiveness of the model was confirmed by experimental results.