Source code partitioning using process mining

  • Authors:
  • Koki Kato;Tsuyoshi Kanai;Sanya Uehara

  • Affiliations:
  • Software Innovation Lab., Fujitsu Laboratories Ltd., Japan;Software Innovation Lab., Fujitsu Laboratories Ltd., Japan;Fujitsu Ltd., Japan

  • Venue:
  • BPM'11 Proceedings of the 9th international conference on Business process management
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

Software maintenance of business application software such as adding new functions and anti-aging should be performed cost-effectively. Information such as grouping of business activities that are executed as a unit, source code which corresponds to the activities, and the execution volume of the activities is useful for deciding on what areas of business application software to invest in, and prioritizing maintenance requests. We propose a new method which extracts such information using the BPM-E process mining tool we have developed. The method was applied to in-house business systems; the results showed that the method successfully extracted the grouping of events, but that there are accuracy issues in associating events with source code.