Determining process model using Time-based Process Mining and control-flow pattern

Riyanarto Sarno, Widyasari Ayu Wibowo, Kartini, Yutika Amelia, Kelly Rossa

Research output: Contribution to journalArticlepeer-review

34 Citations (Scopus)

Abstract

Determining right model of business process from event log is the purpose of process discovery. However some problems i.e the inability to discover OR, noise and event log incompleteness are emmerged while determining right model of business process. First, OR relation is often discovered as AND relation. Second, noise problem is occured when there are truncated and low frequency traces in event log. Thus control-flow pattern is used to solve issues of same noise relation frequency hence it discovers relation based on transaction function of activity. Consequently, it can refine non noise relation in business process model. Third, incompleteness leads to incorrect discovery of parallel process model; therefore we used Timed-based Process Mining which utilized non-linear dependence to solve the incompleteness. Finally this paper proposed combination of Timed-based Process Mining and control-flow pattern to discover OR and handle same frequency noise and incompleteness. From the experiment in section 3, this proposed method manages to get right process model from event log.

Original languageEnglish
Pages (from-to)349-359
Number of pages11
JournalTelkomnika (Telecommunication Computing Electronics and Control)
Volume14
Issue number1
DOIs
Publication statusPublished - 1 Mar 2016
Externally publishedYes

Keywords

  • Conditional OR
  • Control-flow pattern
  • Incompleteness
  • Noise
  • Timed-based Process Mining

Fingerprint

Dive into the research topics of 'Determining process model using Time-based Process Mining and control-flow pattern'. Together they form a unique fingerprint.

Cite this