Detection of processing model basic elements in intellectual analysis of flexible processes through business process intelligence

Abstract

The problem of constructing flexible process models is very urgent nowadays. The structure of flexible processes may be changed either at configuration or runtime stages. Therefore, the flexible process model can be further adapted to the specific implementation conditions in static and dynamic modes. The process model can be constructed by traditional methods, involving domain experts, and by process mining techniques as well. Process mining technique is used for process modeling based on their execution recorded in event logs. The current work is based on the idea of constructing a flexible process model by process mining techniques using algebra of finite predicates (AFP) and logical networks. AFP is intended to formalize the logical processes and represents the discrete unit to describe arbitrary finite relations. Solving AFP equations is implemented by logical networks that convert algebraic equations into a binary system. The proposed approach represents augmentation of a consecutive model at the stage of log analysis for various executions of the process. Identification of different elements is performed by logical networks.