Date of Award
Doctor of Philosophy (PhD)
College of Computing and Engineering
Francisco J. Mitropoulos
Michael J. Laszlo
Fault-tolerant inter-organizational workflow processes help participant organizations efficiently complete their business activities and operations without extended delays. The stalling of inter-organizational workflow processes is a common hurdle that causes organizations immense losses and operational difficulties. The complexity of software requirements, incapability of workflow systems to properly handle exceptions, and inadequate process modeling are the leading causes of errors in the workflow processes.
The dissertation effort is essentially about diagnosing errors in stalled inter-organizational workflow processes. The goals and objectives of this dissertation were achieved by designing a fault-tolerant software architecture of workflow system’s components/modules (i.e., workflow process designer, workflow engine, workflow monitoring, workflow administrative panel, service integration, workflow client) relevant to exception handling and troubleshooting. The complexity and improper implementation of software requirements were handled by building a framework of guiding principles and the best practices for modeling and designing inter-organizational workflow processes.
Theoretical and empirical/experimental research methodologies were used to find the root causes of errors in stalled workflow processes. Error detection and diagnosis are critical steps that can be further used to design a strategy to resolve the stalled processes. Diagnosis of errors in stalled workflow processes was in scope, but the resolution of stalled workflow process was out of the scope in this dissertation. The software architecture facilitated automatic and semi-automatic diagnostics of errors in stalled workflow processes from real-time and historical perspectives. The empirical/experimental study was justified by creating state-of-the-art inter-organizational workflow processes using an API-based workflow system, a low code workflow automation platform, a supported high-level programming language, and a storage system. The empirical/experimental measurements and dissertation goals were explained by collecting, analyzing, and interpreting the workflow data. The methodology was evaluated based on its ability to diagnose errors successfully (i.e., identifying the root cause) in stalled processes caused by web service failures in the inter-organizational workflow processes.
Fourteen datasets were created to analyze, verify, and validate hypotheses and the software architecture. Amongst fourteen datasets, seven datasets were created for end-to-end IOWF process scenarios, including IOWF web service consumption, and seven datasets were for IOWF web service alone. The results of data analysis strongly supported and validated the software architecture and hypotheses. The guiding principles and the best practices of workflow process modeling and designing conclude opportunities to prevent processes from getting stalled. The outcome of the dissertation, i.e., diagnosis of errors in stalled inter-organization processes, can be utilized to resolve these stalled processes.
Mudassar Habib Ghazi. 2022. Diagnosis of Errors in Stalled Inter-Organizational Workflow Processes. Doctoral dissertation. Nova Southeastern University. Retrieved from NSUWorks, College of Computing and Engineering. (1174)