Investigation of business process optimization using design of experiments

Abstract

Business process optimization can be considered as the problem of constructing business process design with optimum attributes. Vergidis (2008) proposed an approach for the Evolutionary Multi-objective business process optimization, the business process optimization framework (bpoF). The bpoF utilizes as a main component the proposed business process representation and EMOAs in order to generate alternative optimized designs. The business process representation is described by mathematical parameters and the composition of a business process design is based on an algorithm that is named Process Composition Algorithm (PCA). This thesis provides a complete and extended investigation of the business process optimization framework (bpoF). Employing a series of scalable tests the main elements of bpoF, are examined in-depth. In addition, this thesis applies the statistical approach of Design of Experiments (DoE) in the business process optimization problem. This approach provides the necessary tools needed to analyze and interpret the results.