Process-Mining-Based Workflow Model Fragmentation for Distributed Execution

• Published in: IEEE transactions on systems, man and cybernetics. Part A, Systems and humans man and cybernetics. Part A, Systems and humans, 2011-03, Vol.41 (2), p.294-310
• Main Authors: Sun, S X, Qingtian Zeng, Huaiqing Wang
• Format: Article
• Language: English
• Subjects: Analytical models; Data mining; Data models; Distributed workflow; Partitioning algorithms; Petri net; Petri nets; process mining; Servers; task assignment; Timing; workflow management; workflow model fragmentation
• ISSN: 1083-4427; 2168-2216; 1558-2426; 2168-2232
• DOI: 10.1109/TSMCA.2010.2069092

A complex workflow is often executed by geographically dispersed partners or different organizations. As a solution for dealing with the decentralized nature of workflow applications, a workflow can be fragmented into small pieces and scheduled to different servers for its execution. An important challenge in distributed workflows is to optimize the fragmentation and distribution to achieve efficiency with respect to time and server resources. To tackle this challenge, we propose the application of process mining to the fragmentation of a workflow for distributed execution. The workflow model discovered through process mining records the actual execution of a workflow and allows in-depth analysis of its temporal behavior. Based on examination of the model resulting from process mining, we demonstrate how to determine the minimum time to finish a workflow and how to partition the workflow in order to achieve efficient server usage.