Logo des Repositoriums
 
Zeitschriftenartikel

Detecting Common Errors in Event-Driven Process Chains by Label Analysis

Lade...
Vorschaubild

Volltext URI

Dokumententyp

Text/Journal Article

Zusatzinformation

Datum

2011

Zeitschriftentitel

ISSN der Zeitschrift

Bandtitel

Verlag

Gesellschaft für Informatik e.V.

Zusammenfassung

In this article, we discuss several classes of error patterns that can frequently be found in Event-Driven Process Chains (EPC). Instances of these patterns can be detected by using a pattern-matching approach: The model is translated into a set of Prolog rules and potential modelling problems are located by querying the Prolog fact base for certain problem patterns. In particular, this article presents patterns for problems that can be detected by analysing the labels of events and functions in EPCs. To make reasoning about the contents of the labels possible, the labels are transformed into a normal form. By taking synonyms, antonyms and negating words (no, not) into account, we locate labels that contradict each other. This leads to the detection of some classes of errors like modelling of an event and its negation occurring at the same time. Our method has been applied to 1253 EPC models in German language. We have been able to detect a large number of errors in those models that remain undetected using traditional approaches for EPC validation.

Beschreibung

Gruhn, Volker; Laue, Ralf (2011): Detecting Common Errors in Event-Driven Process Chains by Label Analysis. Enterprise Modelling and Information Systems Architectures - An International Journal: Vol. 6, Nr. 1. DOI: 10.18417/emisa.6.1.1. Berlin: Gesellschaft für Informatik e.V.. PISSN: 1866-3621. pp. 3-15. Special Issue on Label Quality Assurance for Business Process Models

Schlagwörter

Zitierform

Tags