Table of Contents

Performance Analysis of Business Processes Through Process Mining

Peter Hornix
Master's Thesis. Technische Universiteit Eindhoven, Eindhoven, The Netherlands, 2007.

Abstract

Process Mining (or workflow mining) is an important research area of the section Information Systems (IS/TM) of the department of Technology Management at Eindhoven University of Technology. The starting point of process mining is registered information about tasks that are performed in a process (handling of an insurance claim, treatments of a patient in a hospital etc.). This registered information is typically called an event log. The goal of process mining is deducing knowledge (e.g. a process model, average throughput time) from such event logs. The knowledge obtained this way can increase understanding of the processes within an organization and can help with redesigning them if needed.

Within the section IS/TM a process mining framework has been developed, called the ProM framework. So far, emphasis within this framework has been on the discovery and validation of process models (workflows). Herein, the ProM framework is practically unique. The analysis functionality that is currently present in the framework mainly consists of qualitative analysis. Main concern of qualitative analysis is the logical correctness of processes (absence of anomalies, like deadlocks and livelocks). For the practical usability of the framework however, functionality which supports analysis of the performance of processes (quantitative analysis) is also required.

The goal of this master’s project is to extend the ProM framework with plug-ins that support process performance analysis. In this thesis, the research that was done during the project will be discussed and the plug-ins that have been implemented will be described.

Download PDF (3.4 MB) (See TU/e catalogue)