Converting unstructured into semi-structured process models

Rik Eshuis, Akhil Kumar

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Business process models capture process requirements that are typically expressed in unstructured, directed graphs that specify parallelism. However, modeling guidelines or requirements from execution engines may require that process models are structured in blocks. The goal of this paper is to define an automated method to convert an unstructured process model containing parallelism into an equivalent semi-structured process model, which contains blocks and synchronization links between parallel branches. We define the method by means of an algorithm that is based on dominators, a well-known technique from compiler theory for structuring sequential flow graphs. The method runs in polynomial time. We implemented and evaluated the algorithm extensively. In addition we compared the method in detail with the BPStruct method from literature. The comparison shows that our method can handle cases that BPStruct is not able to and that the method coincides with BPStruct for the cases that BPStruct is able to handle.

Original languageEnglish (US)
Pages (from-to)43-61
Number of pages19
JournalData and Knowledge Engineering
Volume101
DOIs
StatePublished - Jan 1 2016

Fingerprint

Process model
Business process model
Modeling
Synchronization
Directed graph
Polynomials
Graph

All Science Journal Classification (ASJC) codes

  • Information Systems and Management

Cite this

@article{2affef80dbe24cb18df4c7f85698ae6c,
title = "Converting unstructured into semi-structured process models",
abstract = "Business process models capture process requirements that are typically expressed in unstructured, directed graphs that specify parallelism. However, modeling guidelines or requirements from execution engines may require that process models are structured in blocks. The goal of this paper is to define an automated method to convert an unstructured process model containing parallelism into an equivalent semi-structured process model, which contains blocks and synchronization links between parallel branches. We define the method by means of an algorithm that is based on dominators, a well-known technique from compiler theory for structuring sequential flow graphs. The method runs in polynomial time. We implemented and evaluated the algorithm extensively. In addition we compared the method in detail with the BPStruct method from literature. The comparison shows that our method can handle cases that BPStruct is not able to and that the method coincides with BPStruct for the cases that BPStruct is able to handle.",
author = "Rik Eshuis and Akhil Kumar",
year = "2016",
month = "1",
day = "1",
doi = "10.1016/j.datak.2015.10.003",
language = "English (US)",
volume = "101",
pages = "43--61",
journal = "Data and Knowledge Engineering",
issn = "0169-023X",
publisher = "Elsevier",

}

Converting unstructured into semi-structured process models. / Eshuis, Rik; Kumar, Akhil.

In: Data and Knowledge Engineering, Vol. 101, 01.01.2016, p. 43-61.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Converting unstructured into semi-structured process models

AU - Eshuis, Rik

AU - Kumar, Akhil

PY - 2016/1/1

Y1 - 2016/1/1

N2 - Business process models capture process requirements that are typically expressed in unstructured, directed graphs that specify parallelism. However, modeling guidelines or requirements from execution engines may require that process models are structured in blocks. The goal of this paper is to define an automated method to convert an unstructured process model containing parallelism into an equivalent semi-structured process model, which contains blocks and synchronization links between parallel branches. We define the method by means of an algorithm that is based on dominators, a well-known technique from compiler theory for structuring sequential flow graphs. The method runs in polynomial time. We implemented and evaluated the algorithm extensively. In addition we compared the method in detail with the BPStruct method from literature. The comparison shows that our method can handle cases that BPStruct is not able to and that the method coincides with BPStruct for the cases that BPStruct is able to handle.

AB - Business process models capture process requirements that are typically expressed in unstructured, directed graphs that specify parallelism. However, modeling guidelines or requirements from execution engines may require that process models are structured in blocks. The goal of this paper is to define an automated method to convert an unstructured process model containing parallelism into an equivalent semi-structured process model, which contains blocks and synchronization links between parallel branches. We define the method by means of an algorithm that is based on dominators, a well-known technique from compiler theory for structuring sequential flow graphs. The method runs in polynomial time. We implemented and evaluated the algorithm extensively. In addition we compared the method in detail with the BPStruct method from literature. The comparison shows that our method can handle cases that BPStruct is not able to and that the method coincides with BPStruct for the cases that BPStruct is able to handle.

UR - http://www.scopus.com/inward/record.url?scp=84957691704&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84957691704&partnerID=8YFLogxK

U2 - 10.1016/j.datak.2015.10.003

DO - 10.1016/j.datak.2015.10.003

M3 - Article

AN - SCOPUS:84957691704

VL - 101

SP - 43

EP - 61

JO - Data and Knowledge Engineering

JF - Data and Knowledge Engineering

SN - 0169-023X

ER -