Dynamically predicting the deadlines in time-constrained workflows

Abstract

In order to predict workflow's deadline, and improve the efficiency of the entire workflow management, this paper proposes an efficient method to dynamically predict the deadline of time-constrained workflows. This method improves the lack of the existing method in which all the paths in selection structures were assigned the same deadline. We apply LMST-invariant decomposition to decompose the selection structures of the workflow. With the checkpoints which are inserted on selection activities, we propose an efficient method to timely predict the workflow's deadline. This method is more suitable for the practical application.