Transformation of Business Process Model and Notation models onto Petri nets and their analysis

Business Process Model and Notation is known as a widely used standard for business processes modeling. However, its main drawback is that it lacks formal semantics, leading to some undesirable properties, such as livelocks and deadlocks, such that it creates models with semantic errors. In order to...

Full description

Bibliographic Details
Main Authors: Umair Mutarraf, Kamel Barkaoui, Zhiwu Li, Naiqi Wu, Ting Qu
Format: Article
Language:English
Published: SAGE Publishing 2018-12-01
Series:Advances in Mechanical Engineering
Online Access:https://doi.org/10.1177/1687814018808170
Description
Summary:Business Process Model and Notation is known as a widely used standard for business processes modeling. However, its main drawback is that it lacks formal semantics, leading to some undesirable properties, such as livelocks and deadlocks, such that it creates models with semantic errors. In order to formally verify them, we need to transform it onto a formal language, for example, Petri nets. The approach proposed in this article is an extension of previous approaches stated in the literature by adding probability to gateways and time to transitions. The first aim is to transform the Business Process Model and Notation process diagram onto Petri nets automatically using a developed software package. The developed software package is capable of transforming the XML file of a Business Process Model and Notation process diagram into “m” files of a Petri net. The “m” files of the Petri net are then coupled with the General Purpose Petri Net Simulator (GPenSIM) for analysis in MATLAB. The second aim is to manually transform the Business Process Model and Notation process diagram using mapping figures onto Petri nets and then analyze it using Timed Petri Net Analyzer tools. The advantage of transforming a Business Process Model and Notation diagram automatically is that we can add time to transitions and probability to gateways. Furthermore, the simulation time can be checked using MATLAB.
ISSN:1687-8140