Organisations have processes to manage their business activities, often referred to as business processes. In today’s competitive global economy, automation of processes with appropriate technology is advantageous. However, the paradox of processes automation is the continuous evolution and change that occurs in business processes. As the business processes evolve and change, the underpinning automated systems need to reflect those changes. Even after a decade of research in the areas of business process automation (BPA) and business process evolution management (BPEM), organisations still find it challenging to manage evolution of automated processes. Therefore, this thesis finds answers to the question of “How can business process evolutions be accurately and effectively reflected in already implemented web-based workflow systems?” In order to provide a holistic solution to the above research question, this research introduces a framework named paradigm of process automation – PoPA framework and discusses its role in managing process evolution. This framework embodies a business process at four levels as pragmatic, semantic, syntactic, and implementation. Each of these levels deals with a distinctive representation of a business process. For example, the pragmatic level represents the contextual artefact elements such as Acts, policies, organisational structures, rules, and guidelines; that define a process, and the syntactic level denotes the models created for the purposes of automation. When a change takes place in any one of the levels of the PoPA framework, it creates a propagating impact on elements in the above-mentioned four levels. This propagation of impact takes place due to constraints, associations, dependencies (CAD) among elements within and across the levels (intra and inter-level CAD). When analysing intra and inter-level CAD most correlations are found to be hierarchical; therefore, a relational database structure is appropriate to capture these hierarchical associations. However, operational processes at the semantic level have complex associations, which are not hierarchical. Therefore, this research proposes to use Kleene Algebra with Test (KAT) for representing CAD at the semantic level. Propagating impact does not exclusively depend on inter and intra-level CAD, but is also closely associated with the nature of evolution. Depending on the nature of evolution, the propagating impact can be categorised as direct, indirect, secondary, and non-cautionary (DISN) impact. These DISN impacts suggest the severity of the propagating impact. The core contribution of this research is the Process Evolution and Change Impact Analysis (PECIA) Model, which enables the management of process evolution accurately and effectively in automated systems. In this research, a process automation project named Online Courses Approval System (OCAS) is used as an exploratory case study. The practical utility of the PECIA Model is validated using evolution scenarios of OCAS and epistemic utility is analysed based on a study of the literature. Amidst a plethora of literature on BPA and BPEM, this research is significant due to the following theoretical contributions that facilitate in managing automated processes in tandem with organisational process evolution: ���� PECIA Model holistically captures inter and intra-level CAD of process elements facilitating the propagating impact analysis within and across the four levels of the PoPA framework. • A novel use of KAT to capture CAD among process elements cohesively and completely into linear expressions, in order to analyse the impact propagation. • An algorithm that analyses KAT expressions of a process, to locate DISN impacts so that evolutions can be carried out accurately and effectively. The future works that arise from this work are manifold. These may include improving the use of the PECIA Model as a corporate process knowledge repository, and exploring possible other uses of the PECIA Model and KAT based process expressions. / Doctor of Philosophy (PhD)
| Identifer | oai:union.ndltd.org:ADTP/204263 |
| Date | January 2008 |
| Creators | Ginige, Jeewani A., University of Western Sydney, College of Health and Science, School of Computing and Mathematics |
| Source Sets | Australiasian Digital Theses Program |
| Language | English |
| Detected Language | English |
Page generated in 0.0077 seconds