Investigating Completeness of Coding in Business Process Model and
Notation
Carlos Habekost dos Santos
1
, Lucin
´
eia Heloisa Thom
1
and Marcelo Fantinato
2
1
Department of Informatics, Federal University of Rio Grande do Sul, UFRGS, Porto Alegre, Brazil
2
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
1
.
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
1
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
Copyright
c
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 ve 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
2
.
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
2
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