Detecting and identifying data drifts in process event streams based on process histories

Florian Stertz, Stefanie Rinderle-Ma

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

12 Scopus citations

Abstract

Volatile environments force companies to adapt their processes, leading to so called concept drifts during run-time. Concept drifts do not only affect the control flow, but also process data. An example are manufacturing processes where a multitude of machining parameters are necessary to drive the production and might be subject to change due to e.g., machine errors. Detecting such data drifts immediately can help to trigger exception handling in time and to avoid gradual deterioration of the process execution quality. This paper provides online algorithms for concept drift detection in process data employing the concept of process histories. The feasibility of the algorithms is shown based on a prototypical implementation and the analysis of a real-world data set from the manufacturing domain.

Original languageEnglish
Title of host publicationInformation Systems Engineering in Responsible Information Systems - CAiSE Forum 2019, Proceedings
EditorsCinzia Cappiello, Marcela Ruiz
PublisherSpringer Verlag
Pages240-252
Number of pages13
ISBN (Print)9783030212964
DOIs
StatePublished - 2019
Externally publishedYes
Event31st International Conference on Advanced Information Systems Engineering, CAiSE 2019 - Rome, Italy
Duration: 3 Jun 20197 Jun 2019

Publication series

NameLecture Notes in Business Information Processing
Volume350
ISSN (Print)1865-1348
ISSN (Electronic)1865-1356

Conference

Conference31st International Conference on Advanced Information Systems Engineering, CAiSE 2019
Country/TerritoryItaly
CityRome
Period3/06/197/06/19

Keywords

  • Concept drifts
  • Online process mining
  • Process technology

Fingerprint

Dive into the research topics of 'Detecting and identifying data drifts in process event streams based on process histories'. Together they form a unique fingerprint.

Cite this