Investigating Completeness of Coding in Business Process Model and

Notation

Carlos Habekost dos Santos

, Lucin

eia Heloisa Thom

and Marcelo Fantinato

Department of Informatics, Federal University of Rio Grande do Sul, UFRGS, Porto Alegre, Brazil

School of Arts, Sciences and Humanities, University of S

ao Paulo, S

ao Paulo, Brazil

Keywords:

Business Process Management, Process Implementation, Business Process Model and Notation.

Abstract:

One of the ways to represent a business process graphically is using the Business Process Model and Notation

(BPMN). One of the things deﬁned by the BPMN speciﬁcation is a textual rule and a correspondent XML

Schema for each notational element. However, there are some limitations regarding textual rules of notational

elements and their XML Schema. For example, the XML Schema of end event element do not have any control

to not connect any element after it, which can lead to a modeling issue. This paper introduces an approach to

increment the XML Schema in a set of notational elements. The approach considers the BPMN textual rules

and compares with the current XML schema, proposed by BPMN. To evaluate the approach, we will develop

a prototype, to verify the completeness of the developed XML Schema allows better understanding compared

with the current schema and will use mathematical formalism to verify the correctness of this new schema.

We expect that our approach facilitate the understanding of business process by users and minimize possible

implementation problems (e.g. deadlocks, lack of synchronization, livelocks, etc). Altogether, the results of

this research can be interesting for users who want develop the BPM tools.

1 INTRODUCTION

Independently of the application domain, the activ-

ities of an organization are related to business pro-

cesses such as buying products, participating in a pub-

lic tender, manufacturing automobiles, or treating of

patients.

The Business Process Management (BPM) allows

the representation and management of its activities. In

addition, it provides a set of techniques for the anal-

ysis and continuous improvement of organizations’

business process (Weske, 2012), (Dumas et al., 2013)

and (WfMC, 1999).

The life cycle of BPM includes the phases of mod-

eling, conﬁguration, implementation, executing and

validation of business process (Weber et al., 2009),

(Dumas et al., 2012). In particular, BPMN helps in a

better documentation and standardization of business

processes, as well as to increase efﬁciency and quality

in execution (Thom, 2012).

In the modeling phase, business processes can be

represented graphically by the Business Process Mod-

eling and Notation (BPMN). BPMN provides a very

large set of notational elements such as activities,

events, gateways, etc. Besides BPMN, there are other

notations that also allow process modeling such as

Event Process Chain (EPC), Activity Diagram of the

Uniﬁed Modeling Language (UML) and Petri Nets.

However, BPMN is a standard for process modeling,

adopted by the Object Management Group (OMG)

(BPM, 2013), (Aalst, 2013), (Thom et al., 2009) and

used by the majority of vendor’s environment. For

this reason, BPMN is the object of study in this paper.

The notational elements of BPMN are encoded in

the XML Schema. It is also known as XSD (XML

Schema Deﬁnition), the ofﬁcial recommendation of

the World Wide Web Consortium (W3C) for XML

document validation, structure deﬁnition, content and

semantics

For each notational element, BPMN speciﬁcation

offers a textual rule, a graphical representation and

its XML Schema (See Figure 1). However, there is a

lack of completeness between the textual rule and the

corresponding XML Schema of notational elements.

For example, in BPMN speciﬁcation there is a tex-

tual rule deﬁning which in the message ﬂow element

must connect elements from different pools, but its

More information available in: http://www.w3.org/

XML/Schema.html

328

Habekost dos Santos C., Thom L. and Fantinato M..

Investigating Completeness of Coding in Business Process Model and Notation.

DOI: 10.5220/0005464603280333

In Proceedings of the 17th International Conference on Enterprise Information Systems (ICEIS-2015), pages 328-333

ISBN: 978-989-758-098-7

 2015 SCITEPRESS (Science and Technology Publications, Lda.)

XMLSchema do not represent this textual rule. Other

examples:

- In end event element can not be placed any outgo-

ing sequence ﬂow.

- In loop activity, executes the inner activity as long

as the loop condition evaluates as true.

- In parallel gateway, this wait for all incoming ﬂow

before triggering the ﬂow through its outgoing se-

quence ﬂows.

In these examples, all XML Schema do not have

any code to represent such textual rules. Only there

are basic information about the notational element

such as name, type, etc.

In BPMN speciﬁcation, there exists ﬁve categories

of notational elements: ﬂow objects, data, connecting

elements, swimlanes and artifacts. We investigate the

elements of these categories, identifying limitations

regarding textual rules and respective XML Schema

of the notational elements. There are, depending on

the notational element, modeling problems such as

deadlock, livelock or lack of synchronization as de-

scribed below:

• Lack of Synchronization: happens when parallel

ﬂows, which are started with a gateway divid-

ing the ﬂow, do not converge to another gateway

which join the ﬂow.

• Deadlock: occurs when it is not possible to per-

form any activity due to a crash in a point of the

process.

• Livelock: when the process enters an inﬁnite loop,

recurrently performing the same activities.

In this context, the main goal of this paper is to

report an on-going research that aims to increase the

XML Schema completeness, proposed by the BPMN

speciﬁcation, in order to have more adherence to the

textual rules described for each notational element.

The work occurs based on the analysis of these rules,

identifying and implementing those that are not in-

cluded in the current XML Schema. In addition,

the research also aims to create XML Schema vari-

ants, containing the textual rules already deﬁned by

the BPMN speciﬁcation (BPM, 2013), and modeling

rules suggested by the literature, such as those pro-

posed in (Mendling et al., 2010).

We believe that with our approach users will have

a better understanding of the XML Schema of nota-

tional elements, from the most complete codiﬁcation

of BPMN.

For the development of the XML Schema com-

pleteness of notational elements, we used this

methodology:

• Selection of notational elements: deﬁnition of a

set of notational elements that will have the XML

Schema incremented.

• Validation Perspective: deﬁnition of validation

perspectives. We plan to do a prototype for users

interact with the XML code generate from XML

Schema, through forms that show a textual rule,

the original code and the incremented code. The

users will identify the code best represent the tex-

tual rule.

The remainder of this paper is structured as fol-

lows. Section 2 provides the state of art in busi-

ness process modeling and implementation. Section

3 shows the proposed approach for the XML Schema

completeness and perspectives for validation. Finally,

Section 4 concludes the paper.

2 STATE OF ART

To obtain the state of the art related to modeling and

implementation of business process we conducted a

systematic literature review using (Kitchenham and

Charters, 2007). A systematic literature review is a

means to identify, assess and interpret research pa-

pers available and relevant to a speciﬁc research ques-

tion, a thematic area, or a phenomenon of interest

(Kitchenham et al., 2010).

Our aim with the systematic review was to gather a

better understanding of modeling and implementation

so that we could identify areas for improvement quote

(Santos and Thom, 2014).

The systematic literature review occurred in the

ﬁrst semester of 2014. We analyzed the title, key-

words and abstract of works considering the period

from 2009 until 2014. Our approach focused on the

digital libraries:

• ACM Digital Library;

• IEEE Xplore Digital Library;

• Springer Link;

• Google Scholar;

• Author’s page with related publications to BPM

area and h-index relevant.

To select the works, we considered “string search”

to get it. Terms with quotation marks (“ and ”) indi-

cates these should be considered the complete term:

1. “business process” AND modeling AND execu-

tion;

2. “business process execution” OR “workﬂow exe-

cution”;

InvestigatingCompletenessofCodinginBusinessProcessModelandNotation

329

Figure 1: A example modeling rule deﬁned by BPMN.

3. (“business process” OR workﬂow) AND (model

OR modeling OR design) AND (implementation

OR implemented OR execution OR executable

OR enactment);

From the string search in the digital libraries,

adding search in author’s BPM page, it reached a set

of 350 works.

We performed the analysis with the title and ab-

stract of each article. As result, we selected 30 can-

didate articles, which the approach was related to the

string search. Through the analysis of these articles,

19 were selected as best works to represent the string

search. This article, we select the works which repre-

sents the results of systematic literature review.

The works studied started on the BPMN speciﬁca-

tion (BPM, 2013). For each notational element, there

exists a textual rule, the graphical representation and

the XML Schema. Figure 1 illustrates an example of

this deﬁnition in the BPMN speciﬁcation. The de-

scribed example shows the exclusive gateway (XOR)

element used for routing the process ﬂow.

From this example, it is possible to note that there

is low consistency between the default XML Schema

of BPMN and the textual rule. There are not some

rules allowed in the XSD, for example, the evaluation

of only a condition which have no corresponding code

in the “XML Schema”. Furthermore, BPMN deﬁnes

rules indicating which elements can be connected by

a sequence ﬂow (e.g. a gateway can not connect to

a start event), however, there is no XSD solution to

allow this control.

In (Mendling et al., 2010), was presented a re-

search on the relationships between structured mod-

els and their error probability and understanding. As

a result, seven modeling rules were proposed to guide

users to avoid problems of modeling and understand-

ing business processes models.

In the approach of (Stroppi et al., 2011) was

proposed an extension of the BPMN notation called

BPMN + X. This work extends the XSD encoding of

notational elements. Also it discusses about the lack

of guidelines and methodologies to enable the exten-

sion, by the BPMN speciﬁcation.

In (Dijkman et al., 2008) is proposed an analy-

sis of business process models, which are mapped to

petri nets in order to identify possible problems such

as deadlock and livelock with the help of some tech-

niques.

Altogether, the studies analyzed aim to verify is-

sues related to modeling problems. Currently, these

works focus on the use of external techniques such as

petri nets to solve these problems, rather than focus-

ing on improvements directly in the encoding of the

BPMN speciﬁcation.

3 COMPLETENESS OF XML

SCHEMA

In this section we discuss the XML Schema com-

pleteness of notational elements. Initially, we stud-

ied the structure of the XML Schema available in the

BPMN speciﬁcation version 2.0 (BPM, 2013). We

proposed, then to increase the completeness of the

XML Schema concerns by adapting this schema to

ICEIS2015-17thInternationalConferenceonEnterpriseInformationSystems

330

represent conditions and other rules that are missing

in the original XML Schema. The adaption follows

three phases:

1. Deﬁnition of the group of notational elements to

be incremented.

2. XML Schema increment to make it more com-

plete regarding the textual rules inherent to each

notational element being studied.

3. Deﬁnition of possible perspectives of how to vali-

date the XML Schema incremented.

First, we deﬁned the group of notational elements

that will be incremented. For this selection, we con-

sidered only the basic set of elements deﬁned in the

BPMN speciﬁcation: events, activities and gateways.

In this paper, we present the proposal of the exclu-

sive gateway. However, we are also investigating the

inclusive gateway (OR) and parallel gateway (AND).

We started the investigation on these notational ele-

ments because there are studies in the literature that

highlight them as the most problematic in terms of

user understandability (Kossak et al., 2012), (Figl

et al., 2013), (Dehnert and Aalst, 2004) and model-

ing (Mendling et al., 2010).

Next, we modiﬁed the current XML Schema,

adding new code to it. We developed a new XML

Schema based on the analysis of the textual rules de-

scribed in the Chapter 13 of the BPMN speciﬁcation

which describes the semantics of the notational ele-

ments (BPM, 2013). The change occurs in the XML

Schema of the notational element. In (Stroppi et al.,

2011) it is proposed a way to modify the XML schema

to extend the notational elements. The proposed com-

pleted XML Schema is illustrated in Figure 2.

Figure 2a shows the XML Schema suggested

for the XOR element (see “Extension code XSD”).

The completeness is achieved by adding the

<xsd:element> tag. Figure 2b shows the corre-

sponding XML code generated.

The code added, in this example, refers to the tex-

tual rule deﬁned in Figure 1 (see “Textual Rule”).

This rules says that only one condition evaluates to

true. The branch corresponding to this condition re-

ceives the token and no other condition is tested. If no

condition evaluates to true, the token is passed to the

default sequence ﬂow. In case there is no default ﬂow,

an exception is generated. The logic used to develop

the code shown in “XML code” is (See Figure 2b):

- While the counter true condition is zero (trueCon-

ditions == 0):

- If the tested condition is true (condition ==

true), then the counter true conditions is incre-

mented (trueConditions = 1), otherwise it veri-

ﬁes if all conditions were tested (allConditions

== 0).

- If all conditions were tested, then it checks if

there is a default condition (conditionDefault).

If so, then the counter true conditions is incre-

mented (trueConditions = 1), otherwise an ex-

ception is generated informing that the condi-

tions are not valid.

The number of iterations of the algorithm varies

according to the number of conditions.

The variables used in the example, trueCondi-

tions, condition, allConditions and conditionDefault,

were added exclusively for this example. In practice,

the tool that implements this XML Schema must also

implements their conditions so that the operation oc-

curs as expected.

In addition to the XML Schema completeness, we

plan to provide alternative XML Schema based, for

example, on the seven rules and standardization of the

labels format deﬁned by (Mendling et al., 2010). The

goal of alternative XML schema is to provide sugges-

tions to the user who wants to avoid potential prob-

lems with deadlock, livelock, or enable a correct mod-

eling of the elements labels.

3.1 Perspectives of Validation

In this section, we describe the case study designed

to validate the proposed XML Schema completeness.

For this purpose, a prototype in which users can inter-

act with the XML code generate from this schema,

will be developed. This will occur through forms

that show the textual rule, the original XML code of

BPMN speciﬁcation and the incremented XML code.

The purpose of this case study is to identify, through

the users, which XML Schema best represents the tex-

tual encoding rule.

The forms will be designed using Google Forms

The target audience is students and people from the

industry, who have some knowledge in programming

logic, but not necessarily in XML. The idiom for de-

velopment will be Portuguese and English.

We decided to use this type of case study because

it can helps to identify if the the users can have a

better understanding of XML Schema, and if have

more adherent to the textual rules. Therefore, the goal

with the validation is to compares the current XML

Schema and the improved XML Schema that we are

proposing, to identify which one best applies the no-

tational elements textual rules.

Another way to validate the work, refers to the use

of mathematical formalism to verify the correctness

http://www.google.com/forms/about/

InvestigatingCompletenessofCodinginBusinessProcessModelandNotation

331

Figure 2: Proposal for completeness of BPMN encoding.

of the proposed XML Schema. The aim of this vali-

dation is to verify to the reliability and robustness of

the proposed schema. The formalism to be applied is

being analyzed.

ICEIS2015-17thInternationalConferenceonEnterpriseInformationSystems

332

4 CONCLUSIONS

This paper presented an ongoing work regarding the

completeness XML Schema of notational elements.

The current XML Schema proposed in the BPMN

speciﬁcation does not cover all the deﬁned textual

rules, as a result each vendor’s environment imple-

ments it in a different way.

Though the BPMN notation should not have re-

strictions on the XML Schema of notational elements

to implement different scenarios of the business pro-

cess models, the textual rules should be implemented,

to reduce the possibility of problems to happen in

modeling, such as deadlocks and livelocks.

The completeness XML Schema proposed allows

to increase semantic in the schema of notational ele-

ments. Textual rules that could be only viewed in the

BPMN speciﬁcation, can now be also viewed in the

XML Schema, so that users can have a more adherent

schema.

As current work limitation refers to the fact that

not all textual rules can be applied. For example, in

Figure 1, “Textual Rule” deﬁnes that each token ar-

riving in a gateway is sent to exactly one the gateway

output streams. This textual rule can not be repre-

sented in XML Schema because there is not a tag that

can express this rule.

As main contribution, the present work will pro-

vide a more adherent XML Schema from textual

rules, adding more semantic of the notational ele-

ments. Users who plan to develop a new tool for

modeling business processes will be using a more ac-

curate XML Schema, reducing the possibility of po-

tential problems.

As future work, we plan to complete the develop-

ment of alternative XML Schema. After, we will de-

velop a parsing, which enables a more detailed anal-

ysis of the coding notation, allowing a more effective

scrutiny of possible failures of implementation.

Finally, we propose graphical representation of

notational elements that facilitate user understanding

of the business process, so that can represent when a

gateway is fork or join, using different graphics. This

work is focused on enabling the minimization of pos-

sible modeling problems.

ACKNOWLEDGEMENTS

The authors would like to thank Capes and CNPq,

Brazil, for partially supporting this work.

REFERENCES

(2013). Business Process Modeling Notation (BPMN), V.

2.0.2. BPMN.

Aalst, W. M. P. v. d. (2013). Business Process Management:

A Comprehensive Survey. ISRN Software Engineer-

ing, 2013:1–37.

Dehnert, J. and Aalst, W. M. P. v. d. (2004). Bridging the

Gap between Business Models and Workﬂow Speciﬁ-

cations. Journal of Cooperative Information Systems,

pages 1–39.

Dijkman, R. M., Dumas, M., and Ouyang, C. (2008).

Semantics and analysis of business process models

in BPMN. Information and Software Technology,

50(12):1281–1294.

Dumas, M., Rosa, M. L., Jan, M., and Reijers, H. A.

(2013). Fundamentals of Business Process Manage-

ment. Springer-Verlag, Berlin, Germany, ﬁrst edition.

Dumas, M., Rosa, M. L., and Mendling, J. (2012). Un-

derstanding business process models: the costs and

beneﬁts of structuredness. Advanced Information . . . ,

pages 31–46.

Figl, K., Recker, J., and Mendling, J. (2013). A study

on the effects of routing symbol design on process

model comprehension. Decision Support Systems,

54(2):1104–1118.

Kitchenham, B. and Charters, S. (2007). Procedures for

performing systematic reviews. Keele university. tech-

nical report tr/se-0401, Department of Computer Sci-

ence, Keele University, UK.

Kitchenham, B., Pretorius, R., Budgen, D., Pearl Brereton,

O., Turner, M., Niazi, M., and Linkman, S. (2010).

Systematic literature reviews in software engineering

– A tertiary study. Information and Software Technol-

ogy, 52(8):792–805.

Kossak, F., Illibauer, C., and Geist, V. (2012). Event-

Based Gateways: Open Questions and Inconsisten-

cies, pages 53–67.

Mendling, J., Reijers, H. A., and Aalst, W. M. P. v. d.

(2010). Seven process modeling guidelines (7PMG).

Information and Software Technology, 52(2):127–

136.

Santos, C. H. and Thom, L. H. (2014). Uma revis

ao sis-

tem

atica sobre modelagem e execuc¸

ao de processos

de neg

ocio. Individual work, Institute of Informatics,

Federal University of Rio Grande do Sul, BR.

Stroppi, L. J. R., Chiotti, O., and Villarreal, P. D. (2011).

Extending BPMN 2.0: Method and Tool Support.

Business Process Model and . . . , pages 59–73.

Thom, L. H. (2012). Gerenciamento de Processos de

Neg

ocio e Aplicabilidade na Sa

ude e na Rob

otica.

lbd.dcc.ufmg.br.

Thom, L. H., Reichert, M., and Iochpe, C. (2009). On

the Support of Workﬂow Activity Patterns in Process

Modeling Tools: Purpose and Requirements.

Weber, B., Sadiq, S., and Reichert, M. (2009). Beyond

rigidity – dynamic process lifecycle support. Com-

puter Science - Research and Development, 23(2):47–

65.

Weske, M. (2012). Business Process Management. Springer

Berlin Heidelberg, Berlin, Heidelberg.

WfMC (1999). Workﬂow Management Coalition Terminol-

ogy & Glossary. (3):1–65.

InvestigatingCompletenessofCodinginBusinessProcessModelandNotation

333