Bachelor Thesis - Visualization of record logs

Kontakt

Name

Miriam Wagner

Wissenschaftliche Mitarbeiterin

Telefon

work
+49 241 80 21916

E-Mail

E-Mail
 

Description

Most organisations, in a variety of fields such as industry, insurance and healthcare, execute several different (business) processes. Modern information systems allow us to track, store, and retrieve the data related to the execution of such processes, in the form of record logs. A record log is a combination of an event log with time series data. Therefore, a record can be an event or a measurement. Measurements are continuous time series data, e.g., in the record log, which is used for this thesis, the wind speed is measured every 10 minutes and is stored as a measurement in the record log.

The field of process mining is concerned with the study and analysis of the events stored in record logs. The main goal of process mining is to improve the organisation’s knowledge of its own processes, by analysing its execution, as captured in the data. The insights retrieved show the company how the process is actually functioning.

One of the initial steps to analyse data is getting a deeper understanding of the data and the structural attributes. A commonly used method is the visualization with a dotted chart to get a first idea of different properties. A dotted chart shows different attributes over a time axis, e.g., if the cases are on the y-axis, time on the x-axis and the dots display activities, badge executions can be easily detected. In addition, other distributions of work are made visible, e.g., activity ‘a’ is always executed on Mondays. Furthermore, if the resources are on the y-axis, time on the x-axis and activity present the dots, the workload of the resources over time is visualized and some patterns may be found, e.g. resource ‘1’ just works 3 days a week or resource ‘2’ executes activity ‘b’ just every two weeks.

The known tools, e.g., ProM dotted chart, do not considerate the continuous measurements recorded along with the events in a record log. The goal of this Bachelor thesis is to fill this gap by combining the classical dotted chart with a visualization of the measurements. This new mixed chart should be as interactive as the dotted chart is, e.g., the user should be able to zoom in or out. The result should be highly user-friendly.

Prerequisites

  • Good programming skills (Java or Python)
  • Front-End development

Pointer

Supporting Process Mining by Showing Events at a Glance

Supervisor

Prof. Dr. Ir. Wil van der Aalst

Advisor

Miriam Wagner MSc

For more Information

Send an e-mail to . Make sure to include detailed information about your background and scores for completed courses.