<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.0 20120330//EN" "JATS-archivearticle1.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink">
  <front>
    <journal-meta />
    <article-meta>
      <title-group>
        <article-title>A UML Pro le for the e3-Value e-Business Model Ontology</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>C. Huemer</string-name>
          <email>huemer@big.tuwien.ac.at</email>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>A. Schmidt</string-name>
          <email>al.schmidt@sap.com</email>
          <xref ref-type="aff" rid="aff1">1</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>H. Werthner</string-name>
          <email>werthner@ec.tuwien.ac.at</email>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>M. Zapletal</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>Institute of Software Technology and Interactive Systems, Vienna University of Technology</institution>
          ,
          <country country="AT">Austria</country>
        </aff>
        <aff id="aff1">
          <label>1</label>
          <institution>SAP Research CEC</institution>
          ,
          <addr-line>St. Gallen</addr-line>
          ,
          <country country="CH">Switzerland</country>
        </aff>
      </contrib-group>
      <pub-date>
        <year>2008</year>
      </pub-date>
      <abstract>
        <p>Shorter life cycles of products and services require faster changing business models. Information systems must quickly adjust to the adapted business models. Business models are usually described by their own proprietary notation, which is incompatible with UML - the de-facto modeling standard in software engineering. In order to allow a straight-through modeling approach from business models over business process models to software artifacts, it is desirable to use a common modeling approach. Thus, we suggest to map existing concepts to describe business models onto the UML notation. In our work we mainly focus on inter-organizational systems. A promising approach describing a business model for an inter-organizational network of actors is delivered by e3-Value. In this paper, we present a discussion of di erent approaches to represent the e3-Value concepts by means of UML. A UML notation for e3-Value is a precondition to future work on aligning e3-Value to UMLbased approaches specifying inter-organizational business processes.</p>
      </abstract>
      <kwd-group>
        <kwd>Business modeling</kwd>
        <kwd>e3-Value</kwd>
        <kwd>UML pro le</kwd>
        <kwd>B2B</kwd>
      </kwd-group>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>-</title>
      <p>
        Service-oriented Architectures (SOA) are said to provide a means of quickly
adapting IT to changing business needs. However, most approaches focus only on
the IT layer. Instead, we propose an integrated approach for inter-organizational
systems spanning from business models over business process models to their
execution in a SOA. An integrated approach starts with analyzing the economic
drivers for a business collaboration. This means describing the business models
in terms of economic values that are exchanged between business partners.
Candidate approaches to be used on this level of abstraction are e3-Value [
        <xref ref-type="bibr" rid="ref1">1</xref>
        ], REA
[
        <xref ref-type="bibr" rid="ref2">2</xref>
        ] or BMO [
        <xref ref-type="bibr" rid="ref3">3</xref>
        ].
      </p>
      <p>
        In order to guarantee that each partner deserves its economic value they have
to agree with each other on the inter-organizational business processes to realize
the value exchanges. A resulting global choreography may be described by the
UN/CEFACT modeling methodology (UMM) which is speci cally designed for
this purpose [
        <xref ref-type="bibr" rid="ref4">4</xref>
        ]. The UMM is de ned as a UML pro le. In a following step
the orchestration of the internal business processes is de ned and bound to the
UMM choreography. This orchestration may also be described by the means of
UML leading to a software development process that ends up with the automatic
generation of machine-interpretable work ows.
      </p>
      <p>In order to allow a straight-through modeling approach, it is desirable to
base the di erent steps in developing inter-organizational systems on a single
modeling paradigm. Most of the steps described above are already based on
UML. This means they customize the general purpose language UML by means
of stereotypes, tagged values and constraints for their speci c purpose. Only the
de nition of the value exchanges is not based on UML. Thus, the de nition of
the UML pro le for value exchanges completes an overall UML based approach
for inter-organizational systems. Since the e3-Value approach speci cally targets
business models in an inter-organizational environment, it is our goal to
transform its concepts to a UML pro le. In this paper we discuss di erent options
for representing e3-Value in UML. A UML-based e3-Value notation is a
precondition to seamlessly integrate it with the UMM. However, a detailed analysis of
this integration is out of scope for this paper and is up to future work.
2
2.1
e3-value at a glance</p>
      <p>
        Basic concepts
e3-Value is an ontology-based methodology for modeling and designing business
models for business networks [
        <xref ref-type="bibr" rid="ref1">1</xref>
        ] incorporating concepts from requirements
engineering and conceptual modeling (including a graphical notation). Its main focus
is on identifying and analyzing how value is created, exchanged and consumed
within a multi-actor network, hence, taking the economic value perspective and
visualizing what is exchanged (which kind of economic value) by whom [
        <xref ref-type="bibr" rid="ref5">5</xref>
        ]. An
economic value exchange, and consequently the e3-Value ontology as a whole,
is based on the principle of reciprocity emphasizing the dual character of
business transactions. This "give and take"-approach denotes that every actor o ers
something of economic value, such as money, physical goods, services, or
capabilities, and gets something of economic value in return.
      </p>
      <p>
        The e3-Value ontology de nes a number of concepts that will be brie y
outlined in this section. The concepts are described in more detail in [
        <xref ref-type="bibr" rid="ref1">1</xref>
        ] as well as
in [
        <xref ref-type="bibr" rid="ref6">6</xref>
        ] and [
        <xref ref-type="bibr" rid="ref5">5</xref>
        ].
      </p>
      <p>Actors represent parties engaged in a value exchange. They are considered
as independent economic entities that strive for pro tability (in case of an
enterprise) or maximizing their economic utility (in case of an end-consumer) by
carrying out value activities. These pro table or utility-increasing value
activities are intended to be directly and unambiguously assigned to the
corresponding actors. By conducting value activities actors exchange value objects that are
valuable to one or more actors of the business network. As mentioned before,
these objects are "things of value" - either material, such as physical goods,
products or money, or non-material (e.g. services, capabilities or experience).</p>
      <p>Actors signal their will to provide or request value objects through interfaces.
In the e3-Value ontology these interfaces are called value ports. The rationale
of value ports is to abstract from an actor's internal processes and instead
concentrate exclusively on the external connection to other business partners (i.e.
actors) and other components. Two value ports are connected to each other
via a value exchange. The latter depicts one or more potential trades of value
objects between value ports. The values o ered to or requested from the
environment are represented by so-called value o erings, representing sets of equally
directed value ports. The concept of value o erings allows the mapping of
"object bundling" in case that objects are of value and can be requested or provided
only in combination. This means that an o ering may consist of several value
object s to be exchanged. One or a maximum two value o erings - in general
one incoming and one outgoing - are subsumed by a value interface typifying
the concept of economic reciprocity and showing which value object is o ered
in return for another. Each actor may have multiple value interfaces grouping
individual value ports. The exchange of values is atomic - i.e., value objects in an
o ering are exchanged through the value interface on all of its value ports (each
exchanging exactly one value object ) or none of them.</p>
      <p>For creating appropriate visual representations of the value models a
graphical notation is provided - the stated elements are represented in gure 1.</p>
      <p>For creating appropriate visual representations of the value models a graphical notation is
provided; the stated elements are represented in Figure 1.</p>
      <p>Actor</p>
      <p>AVcatilvuiety InVtearlfuaece VPaolrutse ExVcahlaunege with OVbajelucets</p>
      <p>Payment</p>
      <p>Goods
foArkN/jDoin forOk/Rjoin StSimtaurltus StiSmtouplus
2.2</p>
    </sec>
    <sec id="sec-2">
      <title>Example: waste management</title>
      <p>In this subsection we demonstrate the e3-Value concepts by means of an example
from the waste management domain. We consider the international - i.e.
crossborder - trading of waste. In this case study waste is traded between an exporter
and an importer. In principle we must distinguish two di erent kinds of trading.
In the majority of cases the exporter has to pay the importer for the waste
handling, i.e. for the recovery or disposal of the waste. Accordingly, waste and
money is given to the importer and the exporter gets the service of waste handling
in return. In the minority of cases, the waste is traded like a regular good. The
trading of re-cycled paper is a typical example of such a case. This means the
importer has to pay for the waste. In other words the importer gets the waste,
and the exporter receives money in return.</p>
      <p>Moreover, the international trading of waste has some legal implications.
Competent authorities in the countries of the exporter and the importer control
the trading. Accordingly, the exporter has to inform the export authority about
a waste transport. The exporter delivers relevant environmental information
about the transport and the export authority issues a transport allowance
in return. The value of a transport allowance is considered equally to ful ll
the legal regulations and, thus, to avoid possible nes. Similarly, the importer
has to provide environmental information about the transport to the import
authority in order to get the transport allowance for importing the waste. The
export authority and the import authority have a natural interest in providing
each other with the relevant environmental information collected on each side.
Thereby, both competent authorities are able to obtain the full information
about the waste transport on each side.
[Environmental Information] [TransportAllowance]</p>
      <p>Exporter
[Environmental Information] [TransportAllowance]</p>
      <p>Importer
Request Waste Transport</p>
      <p>Receive Waste
Request Waste Transport
Transfer Waste</p>
      <p>[Waste]
[Waste Handling]</p>
      <p>
        [Waste]
This section focuses on mapping the e3-Value concepts to a UML pro le. In
addition to a prose description of the concepts, e3-Value comes with a MOF-like
meta model specifying the concepts and the relations between them (c.f. [
        <xref ref-type="bibr" rid="ref5">5</xref>
        ]).
Furthermore, e3-Value de nes its own graphical notation. The MOF-like meta
model is signi cantly di erent from the UML meta model. Developing a UML
pro le means to represent the e3 concepts on top of the UML meta model by
means of stereotypes, tagged values and constraints.
      </p>
      <p>A mapping to a UML pro le is not straightforward due to the signi cant
di erences between the e3-Value meta model and the UML meta model. Since
UML originates from software engineering, none of the UML standard diagrams
has originally been intended to capture e3-Value semantics. It is necessary to
analyze which of the existing UML standard diagrams and corresponding model
elements are best suited for a customization towards UML. In the following
subsections we discuss ve di erent alternatives by means of the waste management
example and state their strengths and shortcomings.
3.1</p>
    </sec>
    <sec id="sec-3">
      <title>The activity parameter variant</title>
      <p>Value activities are a cornerstone of e3-Value. At a rst glance it seems to be
consequent to map them to activities and activity diagrams, respectively. A
possible solution for the waste management scenario based on activity diagrams
is depicted in Figure 3.</p>
      <p>Each e3-Value actor results in a UML activity partition assigned to a
corresponding UML actor. The activity partition shows his area of responsibility. In
the waste management example the activity diagram includes four activity
partitions for the following actors: exporter, export authority, import authority,
and importer. Each value activity is mapped to a UML activity showing the
stereotype value activity. In the waste management example, each party
performs an activity request waste transport. Furthermore, the exporter performs
transfer waste and the importer performs receive waste.</p>
      <p>We use UML activity parameters to model value exchanges. An activity
parameter describes the input or output to/from a UML activity. It follows, that an
o ering activity carries an output parameter, whereas a consuming activity
carries an input activity. The ow of a value object is described by an exchange from
the output parameter to the input parameter. The value object itself is a
stereotype based on the UML metaclass class. This value object is assigned to both, to
the input parameter and to the output parameter. To illustrate these concepts
we take a look at the value exchanges between the exporter and the export
authority on the left hand side of gure 3. The value exchanges are realized
between the activities called request waste transport on each partner's side.
The value object environmental information is assigned to the output
parameter of the exporter's activity as well as to the input parameter of the export
authority's activity in order to realize the value exchange from the exporter
to the export authority. The ow of the value object transport allowance goes
the other way round.</p>
      <p>A major drawback of this variant is the fact, that it lacks the concepts of
value ports and value interfaces. These concepts are required to group value
exchanges for denoting the atomicity of a set of value exchanges. There is no
concept in UML activity diagrams that corresponds to an e3-Value value port.
For representing value interfaces one might think of UML parameter sets to
group multiple parameters. However, the semantics of a parameter set allow to
group either input or output parameters, but does not allow a mix of input and
output parameters. Furthermore, multiple parameter sets of the same activity
are in an XOR relationship with each other - which does not correspond to the
e3-Value semantics. Due to these two reasons, UML parameter sets do not t the
requirements of our mapping. A workaround denoting the atomicity of two or
more value exchanges is attaching a constraint to the respective object ows. In
gure 3 we de ned such a constraint - mandating an AND relationship between
the value exchanges of the exporter and the export authority. For the sake of
readability we refrain from showing this kind of constraint between other value
exchanges in gure 3.</p>
      <p>For mapping e3-Value scenario paths we utilize the pseudo states for
modeling ows in UML activity diagrams. The start stimulus and the stop stimulus
of e3-Value are mapped to UML initial and nal states, respectively. The AND
fork/joins are mapped to the UML concepts of fork/joins, whereby e3-Value OR
fork/joins correspond to UML decisions/merges in our mapping. In gure 3,
the decision node within the transfer waste activity of the exporter indicates
a choice of two di erent scenarios: either exchanging waste for money or
exchanging waste and money for the service of waste handling. Similarly, the fork
node immediately after the initial state denotes that the exporter must
conduct a value exchange with the export authority (environmental information
against transport allowance) and a value exchange with the importer as
explained above.</p>
      <p>It is important to note that a scenario path does not specify a control ow
amongst the activities nor does it mandate any sequences in time. In order to
stress this fact we refrain from using the UML connector type control ow.
Instead, we use the concept of a trace - a stereotyped UML dependency - to
describe the scenario paths. Usually a trace should lead to a value interface.
Since the concept of a value interface cannot be represented in this variant a
trace leads to any value port being part of a set of value ports that are graphically
grouped to a value interface.
3.2</p>
    </sec>
    <sec id="sec-4">
      <title>The boundaries variant</title>
      <p>The boundaries variant is somewhat similar to the activity parameter variant. It
is also based on UML activity diagrams. The representation of e3-Value actors
and e3-Value value activities still remains the same. However, as shown in gure
4 value exchanges are modeled using the UML object node notation instead of
the activity parameter notation. In other words, the exchange of a value object is
modeled by an object ow starting from the o ering activity to the value object
modeled as object node leading to the consuming activity. Although the
seman:EnvironmentalInformation :TransportAllowance</p>
      <p>:EnvironmentalInformation :TransportAllowance
:ExportAuthority
«ValueActivity»
Request Waste
Transport
:Exporter
«ValueActivity»
Request Waste
Transport
«ValueActivity»
Transfer Waste
tics is the same as in the activity parameter variant, the object node notation
allows grouping the exchanged value objects. We use the concept of interruptible
regions provided by UML for describing the atomicity of value exchanges being
part of a value scenario - e.g., the exchange of transport allowance in return to
environmental information between the exporter and the export authority. In
activity diagrams, interruptible regions usually denote boundaries for exception
handling. Nevertheless, we are able to use boundaries to express the semantics
of economic reciprocity of two or more value exchanges.</p>
      <p>The usage of interruptible regions results in a big drawback. Since
interruptible regions must not be the source or the target of any UML connector, we
are not able to represent scenario paths. Similar to the activity parameter
variant, the inability to represent the e3-Value concepts of value ports and value
interfaces is a major shortcoming. Although the boundary variant bene ts from
compact and simple diagrams, it is inadequate for describing more complex
e3Value scenarios.
3.3</p>
    </sec>
    <sec id="sec-5">
      <title>The interface variant</title>
      <p>
        The interface variant is also based on UML activity diagrams, but provides
another solution for value exchanges. As shown in gure 5 this solution makes
use of UML interfaces. A value interface is a stereotyped UML interface that
is now used to bind value exchanges to an atomic unit. A value activity may
have one to many value interfaces - each connected with a UML dependency.
According to our example in gure 5, the exporter has a total of three value
interfaces. One is bound to request waste transport, the remaining two are
bound to the value activity transfer waste. A value exchange between two value
interfaces is described by the UML concept of an information ow. Information
ows describe circulation of information in a system in a general manner [
        <xref ref-type="bibr" rid="ref8">8</xref>
        ]. In
our mapping, they are used to exchange value objects. The value object being
exchanged is speci ed as the information ow's classi er. As shown in gure
5, the value interfaces of exporter and export authority exchange values via
two information ows. One information ow sends environmental information
from the exporter to the export authority in exchange for another information
ow carrying the transport allowance. All classi ers of information ows are
stereotyped as value objects.
      </p>
      <p>The interface variant also provides scenario paths - described by similar
concepts as introduced for the activity parameter variant. Although the de nition
of this variant lacks the explicit notion of value ports, it is the rst mapping
variant described so far that allows for modeling implicitly all e3-Value concepts
by means of UML. In terms of value ports, UML purists might correctly argue,
that the concept of a connector end as de ned in the UML meta model
corresponds to a value port in e3-Value. However, for the sake of simplicity - and
since connector ends are not supported by most UML tools - we refrain from
stereotyping the connector ends of an information ow as a value port.
«ValueInterface» «ValueO«bfjleocwt»» Waste «ValueInterface»
«ValueObject» Money</p>
      <p>«flow»
«ValueObject» WasteHandling
«ValueInterface» «ValueO«fblojewc»t» Waste «ValueInterface»</p>
      <p>«flow»
«ValueObject» Money</p>
      <p>«flow»
:ExportAuthority
:ImportAuthority
Similar to the interface variant, the signal variant (c.f. gure 6) covers all
e3Value concepts. However, it di ers from the interface variant in terms of modeling
value interfaces and value ports. In the signal variant, value interfaces are
represented as stereotypes based on UML activities. Value interfaces are modeled
as child elements of the value activity they belong to.</p>
      <p>As shown in our waste management example in gure 6, we place the value
interfaces inside of the value activities. Unlike in the interface variant, value
ports are modeled explicitly using the concept of UML signals. We use send
signal actions to indicate outgoing value ports and receive signal actions for
representing incoming value ports. Each signal action - no matter if incoming
or outgoing - is stereotyped as a value port. As we know from e3-Value , value
ports must be part of exactly one value interface. Correspondingly, we place the
stereotyped UML signals within the value interfaces they are part of.</p>
      <p>The concepts for representing value exchanges and scenario paths look alike
the ones used in the interface variant. Value exchanges are described using
information ows. The value object that is conveyed through the value exchange
is assigned as the information ow's classi er. For de ning the scenario path
we apply the concept of the trace dependency combined with some elements
borrowed from UML diagrams in order to describe AND and OR fork/joins.
:ExportAuthority</p>
      <p>:ImportAuthority
«ValueActivity»
Request Waste Transport
«ValueInterface»
«ValuePort»</p>
      <p>«ValueActivity»</p>
      <p>Request Waste Transport
«EVnvailruoenO«mfbleojenwct»at»lInformation«Va«luVeaI nluteerPfoarct»e»
«ValuePort»</p>
      <p>«flow»
«ValueObject»</p>
      <p>EnvironmentalInformation</p>
      <p>The signal variant provides the user with all e3-Value concepts when using
UML for business modeling. Some UML purists may disagree with the nested
UML activities and actions in order to model value ports and value interfaces of
a value activity. Those nested activities - missing start and end nodes - do not
result in a well de ned sequence of activity ows. However, this is a result of the
di erent semantics of ows in e3-Value and in UML activity diagrams.</p>
    </sec>
    <sec id="sec-6">
      <title>3.5 The use case variant</title>
      <p>In order to overcome the incompliance in the ow semantics between e3-Value and
UML activity diagrams, we propose another solution based on UML use case
diagrams (c.f. gure 7). An e3-Value actor is again represented by a UML actor.</p>
      <p>ExportAuthority
An actor may be connected to one or more use cases representing value activities.</p>
      <p>Thus, each use case gets the stereotype value activity assigned. Considering
our example in gure 7, we model four business partners named exporter, export
authority, import authority and importer. Each business partner is connected
to his stereotyped value activities. For example, the exporter is associated with
two use cases - request waste transport and transfer waste - both stereotyped
as value activity.</p>
      <p>
        In order to represent the value interfaces of a value activity we use UML
ports. According to the UML2 speci cation, ports represent interaction points
between a classi er and its environment [
        <xref ref-type="bibr" rid="ref8">8</xref>
        ]. Since the concept of a use case
inherits from the concept of a classi er, a use case is eligible to have embedded
ports. Each port in our mapping is stereotyped as value interface. In UML, each
port may de ne zero to many interfaces. Each interface is either providing or
requiring objects. Thus, the concept of a UML port interface perfectly ts the
needs of an e3-Value value port. Accordingly, outgoing ports become providing
interfaces and incoming ports become consuming interfaces. Each port interface
gets the stereotype value port assigned. Due to readability reasons, we do not
show the stereotypes value interfaces and value ports in our example gure 7.
      </p>
      <p>Consider the value activity transfer waste of our example: This use case
has two ports representing its value interfaces. One port shows the exchange of
waste and money in return to waste handling. Thus, this port de nes three port
interfaces - two providing ones indicating that waste and money is transfered
to the value activity receive waste and one interfaces consuming the service
of waste handling. The other port of transfer waste has two interfaces - one
provides waste and the other one consumes money in return.</p>
      <p>The value exchanges between the ports and their interfaces are described
like in the interface and signal variant. We denote them by information ows
leading from providing interfaces to consuming interfaces. The classi er of this
information ow corresponds to the value object sent in this value exchange.
Scenario paths are again described by the concept of a trace dependency. We
use trace dependencies to describe possible paths of value interfaces with UML
pseudo states representing AND/OR fork and joins.</p>
      <p>In summary, we prefer the use case variant for mapping e3-Value models to
UML. It results in comprehensible business models representing the whole set
of e3-Value semantics. In addition, the use case variant is fully compliant to the
UML meta model. Furthermore, e3-Value is a method for requirements gathering
and will be used as such as part of the UMM. In the UML - and also in the UMM
- the concept of a use case serves as a mean for eliciting the requirements of a
system. Thus, it is reasonable to model e3-Value concepts by means of use case
diagrams rather than by activity diagrams.
4</p>
      <sec id="sec-6-1">
        <title>Related Work</title>
        <p>
          There have been a lot of di erent approaches to model business by means of
UML. However, most of the processes focus on modeling business processes. A
good evaluation of di erent approaches of modeling business processes by means
of UML activity diagrams is provided in [
          <xref ref-type="bibr" rid="ref9">9</xref>
          ]. However, there is a signi cant
di erence between business modeling and business process modeling [
          <xref ref-type="bibr" rid="ref10">10</xref>
          ]. The
former describes the value perspective, whereas the latter speci es the process
perspective. Hence, our approach may sit on top of the presented business process
modeling approaches. The only all-embracing UML-based approach to model
businesses and their processes is provided by Penker and Eriksson [
          <xref ref-type="bibr" rid="ref11">11</xref>
          ]. They
show how to use UML for documenting the entire enterprise. It is outlined how
to model businesses, from business architecture to processes, business rules, and
goals. Furthermore they de ne business patterns that provide re-usable solutions
to common business problems.
        </p>
        <p>
          In the research eld of ontology-based e-business modeling other
non-UMLbased concepts, similar to e3-Value, have been proposed in the past. The REA
(Resource-Event-Agent) business model ontology evolved from a generalized
framework for modeling accounting information systems [
          <xref ref-type="bibr" rid="ref12">12</xref>
          ] to an ontology
for enterprise information systems [
          <xref ref-type="bibr" rid="ref13">13</xref>
          ]. Their main components - agents,
resources, events and exchanges - are essentially equivalent to the corresponding
e3-Value concepts [
          <xref ref-type="bibr" rid="ref5">5</xref>
          ]. The (e-)Business Model Ontology (BMO), in turn,
possesses a much wider scope conceptualizing a variety of internal resources, assets
and capabilities [
          <xref ref-type="bibr" rid="ref3">3</xref>
          ] [
          <xref ref-type="bibr" rid="ref14">14</xref>
          ]. Finally, in [
          <xref ref-type="bibr" rid="ref15">15</xref>
          ] the authors introduce a reference ontology
by incorporating concepts from e3-Value, REA and BMO.
        </p>
        <p>
          The identi cation of the objects of value exchanged within a business network
is not always straightforward. In our example, participants exchange documents
ful lling legal regulation and avoiding nes. These documents are objects of
value. A discussion about the notion of value objects is given in [
          <xref ref-type="bibr" rid="ref16">16</xref>
          ].
        </p>
        <p>
          Further ontological approaches comprise the AIAI Enterprise Ontology [
          <xref ref-type="bibr" rid="ref17">17</xref>
          ],
which de nes business model-related concepts, including activities and sales (in
accordance to e3-Value), but is much more complex with a large number of
concepts and relationships. Likewise, the Toronto Virtual Enterprise (TOVE)
ontology covers a wide area of company-related concepts [
          <xref ref-type="bibr" rid="ref18">18</xref>
          ]. Still, important
cross-organizational aspects (e.g. an enterprise's interfaces to its environment)
are missing and eliminate the possibility of mapping transactions beyond
company boundaries. One of the major inconveniences of both ontologies - in addition
to their increased complexity - is their lack of an adequate graphical
representation.
5
        </p>
      </sec>
      <sec id="sec-6-2">
        <title>Conclusion</title>
        <p>The contribution of this paper are ve di erent variants for de ning a UML
pro le for e3-Value. UML is considered as the "lingua franca" in software
engineering. However, UML lacks standardized means for describing business models
from a value perspective. e3-Value - an approach for value-based requirements
engineering - is an ideal complement for designing e-business systems with UML.</p>
        <p>When mapping e3-Value to UML we had to consider two somewhat con
icting aspects: On the one hand side, all e3-Value concepts should also be present in
a corresponding UML pro le. On the other hand side, resulting UML diagrams
must be easily understood by business experts with limited UML knowledge.
Inasmuch these diagrams must not be overloaded. In our research work, we
started o with a mapping towards activity diagrams that include value
activities with input/output parameters. At rst glance, this solution seemed to be
the obvious choice, since the resulting diagram results in the same look and feel
as e3-Value diagrams. Unfortunately, not all e3-Value concepts may be mapped
in this solution. So we continued with developing alternative variants based on
activity diagrams. However, we learned that all the proposed solutions either fail
in mapping all e3-Value concepts or result in overloaded diagrams that are hard
to read for business people.</p>
        <p>Finally, we based the UML pro le for e3-Value on use case diagrams. We
prefer this solution to the ones based on activity diagrams. It also provides a
better integration with the UN/CEFACT modeling methodology (UMM) which
serves as a starting point of our research. Being editors of the UMM speci
cation - which is de ned as a UML pro le for inter-organizational systems - we
are looking for an approach that captures the business justi cation for
interorganizational systems. We believe that the UML pro le for e3-Value satis es
these needs. Due to page constraints we were not able to outline the dependencies
between UMM and e3-Value. Accordingly, a detailed description of integrating
e3-Value into UMM is up to future work.</p>
        <p>A UML pro le for e3-Value may be used in any UML based software
development process and not just in relationship with UMM. Others may prefer
a di erent variant according to their speci c needs. Whatever variant is chosen
users bene t from our work since they are able to use standard UML tools for
modeling e3-Value. We are convinced that this further potentiates the di usion
of e3-Value due to a better integration into the software engineering process.</p>
      </sec>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          1.
          <string-name>
            <surname>Gordijn</surname>
            ,
            <given-names>J.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Akkermans</surname>
          </string-name>
          , H.:
          <article-title>E3-value: Design and Evaluation of e-Business Models</article-title>
          .
          <source>IEEE Intelligent Systems</source>
          <volume>16</volume>
          (
          <issue>4</issue>
          ) (
          <year>2001</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          2.
          <string-name>
            <surname>Geerts</surname>
            ,
            <given-names>G.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>McCarthy</surname>
            ,
            <given-names>W.E.</given-names>
          </string-name>
          :
          <article-title>The Ontological Foundation of REA Enterprise Information Systems</article-title>
          .
          <source>Technical report</source>
          , Michigan State University (
          <year>2000</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          3.
          <string-name>
            <surname>Osterwalder</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Pigneur</surname>
            ,
            <given-names>Y.</given-names>
          </string-name>
          :
          <article-title>An e-Business Model Ontology for Modeling eBusiness</article-title>
          .
          <source>In: 15th Bled Electronic Commerce Conf</source>
          . (
          <year>2002</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref4">
        <mixed-citation>
          4. UN/CEFACT:
          <article-title>UN/CEFACT's Modeling Methodology (UMM)</article-title>
          ,
          <source>UMM Meta Model - Foundation Module. (September 2006) Technical Speci cation V1</source>
          .0, http://www.unece.org/cefact/ umm/UMM Foundation Module.pdf.
        </mixed-citation>
      </ref>
      <ref id="ref5">
        <mixed-citation>
          5.
          <string-name>
            <surname>Gordijn</surname>
            ,
            <given-names>J.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Akkermans</surname>
          </string-name>
          , H.:
          <article-title>Value based requirements engineering: Exploring innovative e-commerce idea</article-title>
          .
          <source>Requirements Engineering Journal</source>
          <volume>8</volume>
          (
          <issue>2</issue>
          ) (
          <year>2003</year>
          )
          <volume>114</volume>
          {
          <fpage>134</fpage>
        </mixed-citation>
      </ref>
      <ref id="ref6">
        <mixed-citation>
          6.
          <string-name>
            <surname>Gordijn</surname>
            ,
            <given-names>J.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Akkermans</surname>
          </string-name>
          , H.:
          <article-title>Does e-Business Modeling Really Help?</article-title>
          <source>In: Proc. of the 36th Hawaii Intl. Conf. On System Sciences, IEEE</source>
          (
          <year>2003</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref7">
        <mixed-citation>
          7.
          <string-name>
            <surname>Buhr</surname>
            ,
            <given-names>R.J.A.</given-names>
          </string-name>
          :
          <article-title>Use Case Maps as Architectural Entities for Complex Systems</article-title>
          .
          <source>IEEE Trans. Softw. Eng</source>
          .
          <volume>24</volume>
          (
          <issue>12</issue>
          ) (
          <year>1998</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref8">
        <mixed-citation>
          8. Object Management Group (OMG): Uni ed Modeling Language: Superstructure. (
          <year>February 2007</year>
          )
          <article-title>Version 2</article-title>
          .1.1, http://www.omg.org/docs/formal/07-02-05.pdf.
        </mixed-citation>
      </ref>
      <ref id="ref9">
        <mixed-citation>
          9.
          <string-name>
            <surname>Russell</surname>
          </string-name>
          , N.,
          <string-name>
            <surname>van der Aalst</surname>
          </string-name>
          , W.M.,
          <string-name>
            <surname>ter Hofstede</surname>
            ,
            <given-names>A.H.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Wohed</surname>
            ,
            <given-names>P.</given-names>
          </string-name>
          :
          <article-title>On the Suitability of UML 2.0 Activity Diagrams for Business Process Modelling</article-title>
          .
          <source>In: 3rd Asia-Paci c Conf. on Conceptual Modelling</source>
          , Australian Computer Society, Inc. (
          <year>2006</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref10">
        <mixed-citation>
          10.
          <string-name>
            <surname>Gordijn</surname>
            ,
            <given-names>J.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Akkermans</surname>
            ,
            <given-names>H.</given-names>
            , Van Vliet, H.
          </string-name>
          :
          <article-title>Business Modelling is not Process Modelling</article-title>
          .
          <source>In: ER '00: Proc. of the Workshops on Conceptual Modeling for EBusiness and the Web</source>
          , Springer LNCS (
          <year>2000</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref11">
        <mixed-citation>
          11.
          <string-name>
            <surname>Penker</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Eriksson</surname>
            ,
            <given-names>H.E.</given-names>
          </string-name>
          :
          <article-title>Business Modeling With UML: Business Patterns at Work</article-title>
          . Wiley (
          <year>2000</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref12">
        <mixed-citation>
          12.
          <string-name>
            <surname>McCarthy</surname>
            ,
            <given-names>W.E.</given-names>
          </string-name>
          :
          <article-title>The REA Accounting Model: A Generalized Framework for Accounting Systems in a Shared Data Environment</article-title>
          .
          <source>The Accounting Review</source>
          <volume>57</volume>
          (
          <issue>3</issue>
          ) (
          <year>1982</year>
          )
          <volume>554</volume>
          {
          <fpage>578</fpage>
        </mixed-citation>
      </ref>
      <ref id="ref13">
        <mixed-citation>
          13.
          <string-name>
            <surname>Hruby</surname>
            ,
            <given-names>P.</given-names>
          </string-name>
          :
          <string-name>
            <surname>Model-Driven Design</surname>
          </string-name>
          Using Business Patterns. Springer (
          <year>2006</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref14">
        <mixed-citation>
          14.
          <string-name>
            <surname>Osterwalder</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          :
          <article-title>The Business Model Ontology. A Proposition in a Design Science Approach</article-title>
          . Dissertation, University of Lausanne (
          <year>2004</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref15">
        <mixed-citation>
          15.
          <string-name>
            <surname>Andersson</surname>
            ,
            <given-names>B.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Bergholtz</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Edirisuriya</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Ilayperuma</surname>
            ,
            <given-names>T.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Johannesson</surname>
            ,
            <given-names>P.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Gordijn</surname>
            ,
            <given-names>J.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Gregoire</surname>
            ,
            <given-names>B.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Schmitt</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Dubois</surname>
            ,
            <given-names>E.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Abels</surname>
            ,
            <given-names>S.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Hahn</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Wangler</surname>
            ,
            <given-names>B.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Weigand</surname>
          </string-name>
          , H.:
          <article-title>Towards a Reference Ontology for Business Models</article-title>
          .
          <source>In: Proc. of the 25th Int. Conf. on Conceptual Modeling</source>
          , Springer LNCS (
          <year>2006</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref16">
        <mixed-citation>
          16.
          <string-name>
            <surname>Weigand</surname>
            ,
            <given-names>H.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Johannesson</surname>
            ,
            <given-names>P.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Andersson</surname>
            ,
            <given-names>B.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Bergholtz</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Edirisuriya</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Ilayperuma</surname>
            ,
            <given-names>T.</given-names>
          </string-name>
          :
          <article-title>On the Notion of Value Object</article-title>
          .
          <source>In: Proc. of the 18th Intl. Conf. on Advanced Information Systems Engineering</source>
          , Spring LNCS (
          <year>2006</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref17">
        <mixed-citation>
          17.
          <string-name>
            <surname>Uschold</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>King</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Moralee</surname>
            ,
            <given-names>S.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Zorgios</surname>
            ,
            <given-names>Y.</given-names>
          </string-name>
          :
          <article-title>The Enterprise Ontology</article-title>
          .
          <source>The Knowledge Engineering Review</source>
          <volume>13</volume>
          (
          <issue>1</issue>
          ) (
          <year>1998</year>
          )
          <volume>31</volume>
          {
          <fpage>89</fpage>
        </mixed-citation>
      </ref>
      <ref id="ref18">
        <mixed-citation>
          18.
          <string-name>
            <surname>Fox</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Gruninger</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          : Enterprise Modelling.
          <source>AI Magazine</source>
          <volume>19</volume>
          (
          <issue>3</issue>
          ) (
          <year>1998</year>
          )
          <volume>109</volume>
          {
          <fpage>121</fpage>
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>