Intelligent monitoring of business processes using case-based reasoning

The work in this thesis presents an approach towards the effective monitoring of business processes using Case-Based Reasoning (CBR). The rationale behind this research was that business processes constitute a fundamental concept of the modern world and there is a constantly emerging need for their efficient control. They can be efficiently represented but not necessarily monitored and diagnosed effectively via an appropriate platform. Motivated by the above observation this research pursued to which extent there can be efficient monitoring, diagnosis and explanation of the workflows. Workflows and their effective representation in terms of CBR were investigated as well as how similarity measures among them could be established appropriately. The monitoring results and their following explanation to users were questioned as well as which should be an appropriate software architecture to allow monitoring of workflow executions. Throughout the progress of this research, several sets of experiments have been conducted using existing enterprise systems which are coordinated via a predefined workflow business process. Past data produced over several years have been used for the needs of the conducted experiments. Based on those the necessary knowledge repositories were built and used afterwards in order to evaluate the suggesting approach towards the effective monitoring and diagnosis of business processes. The produced results show to which extent a business process can be monitored and diagnosed effectively. The results also provide hints on possible changes that would maximize the accuracy of the actual monitoring, diagnosis and explanation. Moreover the presented approach can be generalised and expanded further to enterprise systems that have as common characteristics a possible workflow representation and the presence of uncertainty. Further work motivated by this thesis could investigate how the knowledge acquisition can be transferred over workflow systems and be of benefit to large-scale multidimensional enterprises. Additionally the temporal uncertainty could be investigated further, in an attempt to address it while reasoning. Finally the provenance of cases and their solutions could be explored further, identifying correlations with the process of reasoning.

Name dc:type.qualificationname

phd

Level dc:type.qualificationlevel

doctoral

Grantor dc:publisher.institution

University of Greenwich

Year dc:date.issued

2012

Author dc:creator

• Kapetanakis, Stylianos

Advisors dc:contributor.advisor

• Petridis, Miltiadis
• Bacon, Liz
• Parrott, Kevin

• QA76 Computer software

Language dc:language

en