Introduction

ArchiMate is an open standard for modeling enterprise architecture. ArchiMate is a modeling language and has a close relationship with the architectural framework Togaf. This combination means that ArchiMate has been introduced by many organizations, both commercial, non-profit and government institutions in the Netherlands and other countries. In addition, ArchiMate modeling is powerfully supported in Sparx Enterprise Architect.

In this article we discuss expanding the ArchiMate modeling language. Not for introducing new concepts but by adding properties to elements and relationships in the language. This is part of the metamodel in ArchiMate open standard and offers organizations the opportunity to register c features of the organization-specific enterprise architecture in close connection with the concepts in the language.

ArchiMate metamodel

ArchiMate is a rich modeling language with about sixty elements and twelve relationships. This makes it possible to model all aspects of an organization and its environment. In some situations, ArchiMate is seen as too rich for specific goals and target groups. ArchiMate has also introduced a mechanism for this and described it in detail. Namely views and viewpoints.

Image metamodel

This means that ArchiMate's metamodel closely matches the needs of many organizations, ranging from just the open standard to making extensions in the language for the benefit of their own organizational needs.

ArchiMate in Enterprise Architect

Enterprise Architect supports multiple modeling languages such as UML, BPMN, BABoK. ArchiMate is also well supported in Enterprise Architect. In addition, more and more mechanisms of the ArchiMate language have been implemented. This includes extensions in addition to the ArchiMate diagrams such as the toolbox, working method for the ArchiMate viewpoints and the option to view viewpoints using the model wizard and then generate an initial view based on the viewpoint.

Image ArchiMate in EA

Since version 16 of Enterprise Architect, it has also been easy to restrict the metamodel for elements and relationships. This makes it possible to limit the language in elements, relationships, diagrams and quick linkers for certain user groups. A nice functionality for organizations that want to use a limited set of the ArchiMate language to simplify the language for the various stakeholders.

ArchiMate also describes expanding the language with properties. If you want to make organization-specific extensions, that is possible. There are also functionalities available in Enterprise Architect that support this. The MDG technology builder in particular has the option to expand the language with elements and relationships with their own properties. This allows you to introduce your own modeling language. In itself a powerful functionality, but also with a major disadvantage. You introduce a new modeling language based on ArchiMate, but IN ADDITION to ArchiMate. In many situations this is not desirable.

In this article I will therefore introduce a method that makes it possible to introduce organization-specific properties within the ArchiMate language without creating a new modeling language. For this we use an example based on the description available on the website with the ArchiMate standard (see the useful links below).

Combine functionalities in Enterprise Architect

As mentioned, you can use the MDG Technology builder to expand the metamodel for a language. This functionality uses the Tagged Values concept in the background. This functionality makes it possible to extend the Enterprise Architect data model.

To expand the ArchiMate language, we can suffice with the use of Tagged Values. The description of the ArchiMate language does mention specializations of concepts. However, if we look at the description, we see that no new mutual connections arise. New specializations can therefore be developed using Tagged Values.

You can elaborate these Tagged Values within the UML types, but they can also be used for ArchiMate. What is powerful is that a script can be defined for a Tagged Value. This allows us to set additional behavior and conditions for the different Tagged Values. In the image below you see the screen where you can define the Tagged Values in combination with the result in the tags of an element. I'll use a simple naming convention here. The name of the Tagged Values all start with ArchiMate. I will come back to this later.

Image: Tagged Values

We can now easily expand the model based on the stereotypes and the tagged values without having to introduce new stereotypes. This means that the ArchiMate metamodel remains available unchanged and we can continue to exchange ArchiMate models based on the AMEF exchange format.

However, there is a functionality missing that makes the modeler’s life easy. A bottleneck with the tagged values is that the tagged values have to be added manually via the property screen. This is repetitive and slowing down when working with your own tagged values.

Fortunately, there is something that can be done about that. In Sparx Enterprise Architect you can catch events that occur in Enterprise Architect via a (model-based) add-in. This also applies to a context changed event. This means that if the context changes to another element or another connector, a command is executed. You can then use the Enterprise Architect API to check whether all tagged values are present for a certain stereotype and, if not, create these tagged values for the element or connector.

This functionality has been realized in the IDEA Add-on. In the settings of the IDEA addon you can prefix which tagged values should be created. In our case this is ArchiMate as a prefix, then the IDEA addon is used based on the handling of an event in EA. In the image below you can see the piece of software that underlies this within the IDEA AddOn. If you want to create a model-based add-in yourself, you can do so based on the code in the example below.

Public Shared Function DefaultTaggedValues(Element As EA.Element, _

 Repository As EA.Repository) As Boolean

           Dim Tag, oTV As EA.TaggedValue

           Dim oDef As New IDEADefinitions()

           Dim Prefix As String

           Dim oPT As EA.PropertyType

           Dim strPTTag As String

           Dim strNames As String = ";"

           Try

               Prefix = oDef.GetSettingValue("TaggedValuePrefix")

               If Prefix.Length > 0 Then

                   For Each oTV In Element.TaggedValues

                       strNames += oTV.Name + ";"

                   Next

                   For Each oPT In Repository.PropertyTypes

                       Dim blnCreateTag As Boolean = False

                       strPTTag = oPT.Tag.ToString()

                       If strNames.Contains(strPTTag) Then

                           blnCreateTag = False

                       Else

                           If strPTTag.Contains(Prefix) Then

If oPT.Detail.ToString().Contains("BaseStereotype") = False And _             Element.Stereotype.Length > 0 Then

                                   blnCreateTag = True

                               End If

                               If (oPT.Detail.ToString().Contains("BaseStereotype") And _

                                    oPT.Detail.ToString().ToUpper().Contains(Element.Stereotype.ToUpper())) Then

                                   blnCreateTag = True

                               End If

                               If oPT.Detail.ToString().Contains("AppliesTo=" + Element.Type) Then

                                   blnCreateTag = True

                               End If

                               If blnCreateTag Then

                                   Tag = Element.TaggedValues.AddNew(oPT.Tag, "")

                                   Tag.Update()

                               End If

                           End If

                       End If

                   Next

                   Element.Update()

                   Return True

               Else

                   Return False

               End If

           Catch ex As Exception

               DLA2EAHelper.Error2Log(ex)

           End Try

           Return False

End Function

Source code

I have created an example for the combination of the tagged values and the IDEA, including a short video on how this can be achieved.

Summary

This whitepaper describes how you can extend the ArchiMate metamodel with properties. In Sparx Enterprise Architect this is possible with the tagged value functionality. In addition, the IDEA AddOn describes how repetitive and slow work for the modeler can be automated. With this we introduce a powerful method to extend the ArchIMate metamodel without having to introduce new stereotypes into the model. Particularly for organizations that see working based on open standards as an important requirement.

Useful Links

• ArchiMate language modifications
• IDEA AddOn productpage
• Introduction video about ArchiMate in Sparx
• Resource file of the tagged values
• Webvideo about tagged values and ArchiMate

Introducting an Architecture Repository is an opportunity to professionalize the architecture modeling process and products. There a publication is prepared  for supporting an approach for introducing Sparx Enterprise Architect as An Architecture Repository. This book focuses on introducing an architecture repository. An architecture repository contributes to a more mature approach to introducing architecture into an organization. This is a combination package of a book, online training and a sample repository.

The book, training and sample repository are prepared based on the following subjects

• Document driven approach as baseline architecture
• Architecture Repository as target architecture
• Solution architecture for the transfer from baseline to architecture
• Step by step guides
• Aids for successfully introducing an architecture repository

The combination package is available from Leanpub. Please follow the link where you see some sample pages of the book and a sample video of the online training. For more information please visit:

https://leanpub.com/architecturerepository

The sample repository included in the combination package is used to write the book, a nice example of the power of Sparx Enterprise Architect and an approach also know as "Eat you own Dogfood". However it gives you an option to make a jumpstart in introducing an architecture repository.

Resume

ArchiMate is a powerful enterprise modelling language for the architect's expressiveness. As a result, ArchiMate is sometimes experienced as complex. However, there are a number of parts in the metamodel that can simplify the views.

In this white paper, we discuss two forms of simplification, namely nesting of elements to simplify the visual representation. In addition, working with derived relationships to reduce the complexity in a diagram.

Both forms of simplification are supported in Sparx Enterprise Architect. However, with two extra routines in the IDEA Addon, the modeller is automatically supported in implementing changes in the model and validating whether these transformations comply with the rules in the ArchiMate metamodel

Introduction

ArchiMate is a powerful language with several dozen concepts, elements and relations. This gives the architect expressiveness to express an architectural elaboration. For architects among themselves, ArchiMate is important as a language with a rich vocabulary. However, this also has a number of drawbacks. The main disadvantages are that the language is experienced as difficult for a number of stakeholders.

Fortunately, a number of concepts have been added in ArchiMate that allow for a simplification of the language with a relatively small limitation in expressiveness. Two of these simplifications are nesting elements to simplify that way. The other set-up is to work with derivative relations. However, both simplifications are included in the meta model and are structured in the language.

In this white paper we discuss these two simplifications with a number of helpers in the IDEA addon for Sparx Enterprise Architect. In these helpers, the simplifications based on the meta model are automatically supported. A nice dimension is that this also offers possibilities to validate the model in accordance with the rules that apply to these simplifications.

The helpers are within the IDEA addon under the ArchimAid functionality. In this article we will discuss this functionality. The addon is available as open source and can be downloaded from https://data-docent.nl/IDEANL.aspx#gsc.tab=0. In addition, a web video has been created to demonstrate the functionality of ArchimAid.

Nested elements

Archimate perspective

In ArchiMate, a large number of elements can be related to each other. In that case, a diagram will soon be regarded as quite full and well-organized for the reader. Consider, for example, hierarchical structures such as representing the organizational structure or the classification of information objects divided into domains. Below is an example of a diagram that shows a simple hierarchy.

This diagram can be greatly simplified by using nesting. Nesting places the child elements within the parent element. In the example, therefore, a nesting is created twice. See below

This makes the classification easier to interpret, even for stakeholders without in-depth ArchiMate knowledge.

The metamodel does, however, impose some restrictions on this method. Nesting is mainly used for structural relationships and for specializations. In addition, it is desirable to include in your own modelling convention that only the same relationship is used in a nested elaboration. So no combination of, for example, composition and aggregation or specialization in one diagram, then it is no longer possible to deduce which relations have been used.

Sparx Enterprise Architect implementation

In Sparx Enterprise architect the design of nesting is adequately implemented. If you connect two elements and then place one inside the other, the connector will be made invisible as expected, provided you choose the relationship types defined in ArchiMate that allow nesting.

The bottleneck in this, however, is that if you do not create a connector between two elements in advance and apply nesting, there will be no notification that a connector based on ArchiMate is relevant. In one diagram it goes well, but in the repository and in other diagrams it is not clear that there should be a relationship. The ArchimAid is helpful here.

ArchimAid Helper

The ArchimAid helper is a screen in the IDEA addon that supports the ArchiMate modeller. This way you can easily search for ArchiMate stereotypes, layers and aspects. However, the Embed tab provides an overview of the connectors that are there, including the stereotypes, or not.

You can then use the Create Embedded button to create the connectors based on the selected stereotype to create the connectors. See also the web video that shows how create embedded works.

Derived relationships

Archimate perspective

Do you work part of an the enterprise in ArchiMate in detail, you see that a model emerges that has many elements and relationships. This provides a level of detail that is most expressive for the community of architects. However, these detailed elaborations are too complex for other stakeholders. Compare it with the construction world where the specifications are relevant for this target group. However, the future residents can suffice with an artist's impression of the house to be built.

Within ArchiMate it is possible to simplify the model by using derived relations. In the image below you can see an example of this. Instead of a diagram consisting of several elements and relationships, a new relationship is established between two elements. The other elements can then be removed.

This diagram shows the simplification by establishing a derivative relationship between two elements. The elements that can be removed as a result are colored grey. If desired, you could remove a few more elements if this simplifies the model.

Sparx Enterprise Architect implementation

Sparx Enterprise Architect offers you the freedom to establish relationships based on the meta model. This also supports the addition of derived relationships. However, this is an activity that must be performed manually by the modeler.

An additional challenge is that the modeller must be aware that adding a derivative relationship to one diagram can also have an effect on other diagrams on which the elements both occur.

ArchimAid Helper

The ArchimAid functionality does support the modeller automatically in determining derived relationships. You can see an example of this on the screen. You choose a start and end entity and determine based on the ArchiMate metamodel rules which relationship arises as a derived relationship. This can be done via multiple paths through the model. You will see it in the diagram.

Then, in the list of possible simplifications, you can choose the paths you want to replace with the simplified relation. The other relationships are then made invisible in the chart, not from other charts. See the web video with an interactive example.

More information

For more information about the  IDEA ArchimAid screens please visit the following online content:

• https://youtu.be/7gogT2O2Kf8
• https://data-docent.nl/ideaen.aspx#gsc.tab=0
• https://youtu.be/mkFNc6MCZq8
• https://community.sparxsystems.com/white-papers/1355-using-archimate-viewpoints-in-nterprise-architect
• https://data-docent.nl/frmSearchList.aspx?module=webcontent&mode=direct&searchfor=idea#gsc.tab=0

Introduction

Frequently there are legacy systems/applications with no previous documentation, and the lack of it makes maintenance and further development of those systems/applications stagnant and demanding. This particular article outlines the ability of graphs (created through the Neo4j database) to automate the creation of documentation in the form of ArchiMate model in Sparx Enterprise Architect or Archi.

We will create an ArchiMate model for the legacy application that will describe basic entities, such as screens, fragments, roles, files and relations between them. An HTML version of the automatically generated ArchiMate model is available here. It's an export from Sparx Enterprise Architect to HTML.

In a next article, we use the graph to identify:

• application communities
• communities where fragments are used
• communities with fewer screens than roles
• the missing role setting in a linked screen
• unused parts of the application
• risk business processes

Example

An illustrative example of two screens and its relations to other screens, fragments, roles and files is below: An illustrative example of the relation between two screens and other fragments, roles and files is below:

Even if there is only the source code of the application, there is a way (by source code scrapping) to obtain all required information. The source code scrapping should be implemented accordingly to the specific application. However, it is not covered in this article.

The example of an outcome from the scrapping for one screen (Screen_414) in a json format is below:

Introduction

Modelling enterprises and their architectures is important for most organisations. For example for defining business processes, organisational structures, data and information modelling. But also for modelling the goals and drivers of the organisation, the alignment of business and IT need to be modelled due to the complexity of organisations in modern societies. In medium and large scale organisations the role of enterprise- or information architect is introduced to create these models of the enterprise.

What do we need for modelling enterprise architectures? An important aspect is to introduce a coherent modelling language that enables the enterprise architect to model all the relevant aspects of the organisation. This can be done by using a combination of modelling languages like UML, BPMN and DMN. Another approach is to create your own modelling language based on the MDG module in Enterprise Architect. Fortunately there is a specific enterprise architecture modelling language ArchiMate that you can use.

In this whitepaper we will introduce ArchiMate as modelling language for enterprises, its advantages and disadvantages and the possibility to apply the language in the context of a specific organisation. ArchiMate is in basic available as MDG within Enterprise Architect. But when you want to apply the language effectively, especially within large organisations, you need to modify and extend the configuration in Enterprise Architect. This is done by effectively configuring a number of default functionalities and by introducing Add-ons with specific functionalities relevant for ArchiMate. Last but not least it is important to define working processes for architects and procedures within teams to successfully apply ArchiMate as an enterprise modelling language.

ArchiMate

We start with the description of ArchiMate as stated in Wikipedia:

ArchiMate (originally from Architecture-Animate) is an open and independent enterprise architecture modelling language to support the description, analysis and visualization of architecture within and across business domains in an unambiguous way.

ArchiMate is a technical standard from The Open Group and is based on the concepts of the IEEE 1471 standard. It is supported by various tool vendors and consulting firms. ArchiMate is also a registered trademark of The Open Group. The Open Group has a certification program for ArchiMate users, software tools and courses.

ArchiMate distinguishes itself from other languages such as Unified Modelling Language (UML) and Business Process Modelling and Notation (BPMN) by its enterprise modelling scope.

Source Wikipedia

ArchiMate is available as an open standard and is maintained by The Open Group. Since it is open all the information about the language is available from the The Open Group website via https://pubs.opengroup.org/architecture/archimate31-doc/chap03.html#_Toc489945967. Therefore we will limit the description of the language here by a short description of the most important aspects.

Framework

ArchiMate is based on a framework with layers and aspects. This framework is a matrix of the layers and aspects. There is a core framework for modelling the basic aspects of an enterprise from business layer via application layer to technology layer. The full framework is adding some extra dimension like strategy, motivation and project dimensions. In the figure below the full framework.

Source: The Open Group

The framework reduces the complexity of modelling since certain characteristics are modelled with a cell (a combination of a layer and an aspect). Within each cell a limited set of concepts and relationships is available for the modeller.

Layers

As you see in the figure there are a 5+1 layers in the framework. Each layer can be used to create a model with a limited scope. Interesting is that the layers can be seen as tiers. Between these tiers there are a limited number of defined connections possible. The layers are modelled in a horizontal manner.

Aspects

The aspects are the vertical structuring of the enterprise. There are 3+1 aspects. The main aspects are active structure (who), behaviour (what) and passive structure or informational (about).  Also between the aspects a limited number of relationships are available to create connections between these aspects. Furthermore as within the layers the aspects can be seen as tiers where you preferably only connect an aspect only with the elements in the aspects direct aside it.

Elements

Within the cells of the framework specific elements are introduced. Elements are drawn with qualified stereotypes. Each element has a symbol in the right top corner of the element. Elements can be seen as metamodel elements with a definition described in detail in the language. This is used as a classification for the relevant enterprise dimensions. Creating these elements introduces classified libraries of elements all with the same definition.

Relationships

Relationships are used to connect the different elements within a layer, aspect or cell. Relationships are drawn with stereotyped lines and connect two elements. More important it creates connections between elements that are positions in different cells, layers and aspects. Also the relationships have a well described definition in the language. Due to this a relationship can be used only in a limited set of combinations of elements.

In the image below the most important elements and relationships of the core framework are displayed.

Viewpoints

One of the drawbacks of ArchiMate is that the language can be extensive and complex. Since we want to be able to model all the characteristics of an enterprise we have numerous layers, aspects and especially elements and relationships. Since we want to use the models to inform our stakeholders with  specific views on the organisation in ArchiMate viewpoints are introduced. Viewpoints are creating a template for a view or diagram that is based on a subset of the available elements and relationships. This makes it possible to create relative easy to understand views that are used to “tell” a specific part or goal of the organisation.

In ArchiMate there is a set of generic viewpoints defined that can be used as templates for your own diagrams. Also these viewpoints are well described in the language specification. However the language can be extended with viewpoints. This means that you can create your own viewpoints for specific organisational situations or stakeholders. This is an approach applied by many organisations. Please be aware that when doing so an organisation should describe the applied viewpoints in a structured manner.

In the image below an example from the Enterprise Architect Model Wizard is given. This is one of the defined viewpoints in the ArchiMate language.

ArchiMate challenges

Although ArchiMate is probably the best solution for enterprise architecture modelling there are numerous challenges when introducing the language in a (large) architecture team. Finding solutions to prevent for failures when introducing a modelling approach is essential. Below you find a short list of possible solutions to prevent for pitfalls:

• Introduce relevant viewpoints for the context of your organisation. Every organisation is operating in a specific context. This means that not all elements and relationships are relevant for the organisations. Therefore a selection of the relevant elements and relationships is made. This selection is implemented in various viewpoints. Sometimes you can use the standard viewpoints available in the language. More often an organisation will create its own viewpoints to support modelling the specific organisational context.
• Create modelling conventions. Not only viewpoints are relevant, it has to be combined with modelling conventions. Things like naming conventions, colour usage, status of elements, diagrams and relationships. Furthermore conventions on baseline and target architectures are relevant in modelling conventions
• Prevent the emergence of ArchiMate dialects. When you have a large team of architects that are working on different locations in your organisation the risk of ArchiMate dialects is present. This means that every architect creates his or her own modelling approach, creates own viewpoints or even uses elements and relationship not relevant for the context of your organization.
• Usage of patterns and building blocks, reusable models or parts of models is a great way to introduce standardisation of models. Furthermore creating architectural models will speed up enormously by introducing patterns and building blocks. As with viewpoints patterns and building blocks are available from suppliers or sectoral organisations, but creating specific patterns is also often applied.
• Develop a modelling community, when you involve your architecture team in the development of your (ArchiMate) modelling language in your organisation. Creating ArchiMate models is difficult so discussing the best approach in stead on make decisions on your own will in the end improve the quality of the models in your team. Create modelling meetings with your team where you discuss problems and solutions. Another option is to introduce “pair modelling”.
• Train your architects, in many organisations the architecture team is a combination of internal and hired (external) architects. Although most hired architects have experience in using ArchiMate they are not familiar with the specific modelling conventions and viewpoints in your organisation. Take the effort in the training of these architects when they start working in your team. This can be done by face to face training but also instruction videos are very effective (and less time consuming in the end)
• Introduce a validation process. When an architect delivers his or her ArchiMate models a check for the correctness of these models is necessary. Not only a check on the ArchiMate rules in general but especially on the specific rules within your own organisation architecture.

The above described solutions are mainly focused on organizing the modelling process. However the supporting modelling tool is also important. It has to support the modelling process in an effective and efficient manner and reduce the workload of the modelling architects. Enterprise Architect out of the box (without configuration and Add-ons) is a tool less supportive for your architectural team. Therefore take time to configure the modelling tool. It has sufficient possibilities to do so.   In the next chapters we will describe the available functionalities in Enterprise Architect and two Add-ons.

Functionalities in EA

Enterprise Architect (EA) offers extensive functionalities for creating ArchiMate models. In every new version of the tool we see new functionalities, extra validations in the ArchiMate MDG. Creating a configuration effective for architectural teams using ArchiMate is relatively easy. This chapter describes a number of available techniques for effectively configuring Enterprise Architect.

Model wizard

Enterprise Architect supports a large number of modelling languages, ranging from generic like UML to very specific like Autosar. For creating enterprise architecture in most situations the ArchiMate modelling language is sufficient (or even oversized). Since version 15 perspectives are available. This makes it possible to limit the available languages to a certain field of interest. For modelling enterprise architectures there are four languages available in the perspectives in EA. ArchiMate is the most detailed one and the one with the most functionalities.

When selecting the ArchiMate perspective the Model Wizard  gets activated. This offers you a list with a description and examples of ArchiMate models based on the standard viewpoints in the language. I think this is a great starting point for novices in ArchiMate but also as starting point for ArchiMate modelling teams when they want to evaluate viewpoints or develop organisation  specific viewpoints.

In the image below you see an example of the model wizard with an ArchiMate viewpoint

ArchiMate Diagrams

Experienced ArchiMate modellers will need a more direct approach in creating diagrams than the Model Wizard and the generation of example models. They create diagrams directly from the browser or the wizard.

Apart from the default diagram types for the layers and the most important extensions In de diagram dialog there is an interesting (rather hidden) feature present. Actually the list of diagram types is a tree view and when expanding this tree you can see the generic ArchiMate viewpoints.

Unfortunately the toolbox for the diagram is not modified on the selected viewpoint. Another nice feature would be to create your own model wizard pages for the organisation specific viewpoints or modifying the Diagram tree and configure your organisation specific set of viewpoints. Maybe in a future version of EA

Perspective Sets

An interesting new feature in EA is the possibility to create Personal or Model Based perspective sets. This feature let you limit the modelling languages available for the user or the model. From the perspective of organisation specific viewpoints and modelling conventions an even more interesting feature is the possibility to restrict the available elements and relationships in a modelling language.

In the figure below you can see how easy it is to restrict ArchiMate elements and relationships. This will restrict the various ArchiMate toolboxes for the diagrams. Particularly relevant when your organisation uses a subset of ArchiMate concepts to prevent the evolution of ArchiMate dialects.

When you develop perspective sets you can combine this with the authorisation module in EA. You can assign a perspective set to  an authorisation group to limit the modelling possibilities in languages and when defined in the ArchiMate language. Users in the authorisation group can now only use the defined subset.

ArchiMate Add-ons

On top of the default functionalities of EA there are two Add-ons offering extra functionalities for ArchiMate based architecture teams. In the chapters below we will briefly discuss the functionalities in these Add-ons.

Model Expert

Model Expert is an EA Add-on developed by Ability Engineering in the United Kingdom. Model Expert offers multiple functionalities mainly focussed on model validation and model quality. In this chapter we describe the most relevant functionalities for the ArchiMate perspective modelling team.

Package dashboard

A nice feature in Model Expert is the option to create a package dashboard. This dashboard calculates the complexity of the diagrams and elements within that packages and gives the modellers an idea about the quality of the package content.

Below you see an example of the dashboard

Reference model

The basis of Model Expert is that you create a reference model or meta model of (parts of) your model. This means that you can define a package in your repository in which you place a number of sample diagrams that are modelled based on correct usage of your modelling conventions.

For ArchiMate the approach with viewpoints makes creating a Model Expert reference model is relatively easy. When you have created example diagrams of your viewpoints, and often this is already present, these can be the source for the metamodel created by Model Expert

This metamodel can later be used to validate new diagrams, models and package content. So when a modeller has created diagrams that do not apply to the metamodel of the viewpoints he or she is able to validate the models and gets information about which rules are not or wrong applied in the diagram.

In Model Expert you can create a metamodel on full package content but also on diagram level. From ArchiMate perspective this is relevant. When you create a organisation specific model with a subset of elements and relationships from ArchiMate the package level metamodel will be used. However when you have multiple viewpoints with a subset of ArchiMate concepts a metamodel of the specific diagrams is the correct approach. Luckily you can combine the both approaches in your repository and validate on both metamodels.

When you have generated an ArchiMate Metamodel in Model Expert you have the possibility to finetune your convention rules by adding extra rules or modify the generated rules.

In the figure below you see an example of the reference model maintenance screen

Diagram validation

When you have created a meta - or reference model in Model Expert you can check your diagrams and validate them against the reference model. Nice feature is that a new diagram will be generated in which the validated  diagram is combined with the relevant warnings and errors.

This approach makes it possible to validate the ArchiMate diagrams by the individual architect to do a check on his or her diagrams before the approval is done. When there is a validation process in place with the architectural team a model manager can do this validation before the model is approved.

In the figure below you see an example of a diagram with validation messages. You can see that the diagram is a combination of the original elements, the warning and error messages and an overview and score of the severity of the errors.

Managed packages and diagram

Checking the validity of a diagram is an relevant feature when the diagram is already created. Would it not be interesting to prevent a modeller from making mistakes when creating the ArchiMate model in a diagram. This approach is especially relevant when you have an architecture team working on different locations or when you have architects that are only temporarily part of your modelling team.

With managed packages and diagrams the options of diagram types but also on elements and relationships is validated during the creation of the diagram content.

In the figure below you see the Model Expert dialog when a modeller selects an element that is not in the reference model of the managed diagram or package. In the example an element is added to the diagram that is not present in the reference model. The modeller gets a warning and the option to select one of the stereotypes available in the reference model.

MDG generator

The most advanced feature is combining the model validation with a MDG based on the metamodel of the viewpoints. This means that you can create diagrams based on the generated MDG. This diagram has a toolbox with only the elements and relationships relevant in the viewpoint. This means that the modeller knows what elements are present and is unable to select an element from the toolbox that is unavailable in the viewpoint.

When he or she accidentally drags an element from the repository to the diagram not relevant in this viewpoint he or she will get a warning. See the figure below with a viewpoint based limited toolbox and a warning when selecting a irrelevant element

IDEA

IDEA is an Open Source Add-on developed by EAxpertise. It is mainly focussed on supporting data modelling teams but since conceptual data modelling is often done with ArchiMate there are also relevant functionalities for ArchiMate modellers

ArchimAID

In EA there is a clear separation of reusing existing elements in the repository and creating new elements. For existing elements a modeller uses the browser or the search windows and then drags the elements to the diagram he or she is creating. For creating elements the usage of the toolbox is the general approach.

Disadvantage of this separation of functionalities is, especially in large teams or large models the change of introducing duplicates is large. It will be more easy to introduce duplicates when the modelling and naming conventions are not fully in place in a modelling team.

In this situation the ArchiMAID screen can be helpful. It is actually a combination of search window and creating new elements in one wizard. The first step is to select one or more ArchiMate stereotypes and search for a keyword within the repository filtered on the selected stereotypes. When found you can add the elements to the diagram as a link. When not found you can go to an entry screen and then add the element to the repository and the active diagram.

In the figure below you see the three pages in the ArchiMAID wizard

Deduplicator

In large modelling teams, and not only ArchiMate based teams, it is unavoidable that duplicate elements will exist in a repository. Therefore in the IDEA Add-on various deduplicate routines are available. The first one is a warning when a modeller creates a new element in the repository with the same name, stereotype and version.

The second one is a deduplicator screen that searches for duplicates in a repository from the scope of a package. This means that for all the elements in a package and its sub packages a check is done for duplicates in the repository. These duplicates are displayed in a grid view. When the modeller wants to deduplicate he or she can set various parameters for this functionality and merges the duplicate elements and connectors.

Below you see an image of the two screens within the package deduplicator

The deduplicator will merge all the elements to the elements present in the selected package. You can save the deduplicated elements in a duplicates package as a backup of these elements

Summary

In this whitepaper we introduced an approach to use ArchiMate for Enterprise Architecture modelling in large teams. The usage of viewpoints makes it possible to reduce complexity in modelling. Generic functions in EA and specific more advanced functionalities in Add-ons makes it possible to support an architecture modelling team in creating validated and quality models of the enterprise.

Configuring enterprise architect and the Add-ons for organisation specific viewpoints can be complex and requires knowledge on both the ArchiMate modelling techniques and configuring the functionalities and add-ons.

EAxpertise has created multiple example repositories for specific ArchiMate viewpoints. In these repositories the functionalities and add-ons are configured in an extensive manner. When you want to introduce this approach in your organisation please contact us. We can help you making a jump start with the introduction of ArchiMate viewpoints and a repository configured in an effective manner to support your ArchiMate modelling team.

Hyperlinks

https://pubs.opengroup.org/architecture/archimate3-doc/

https://www.sparxsystems.com/enterprise_architect_user_guide/15.1/model_domains/archimate.html

https://www.sparxsystems.com/enterprise_architect_user_guide/15.1/user_interface/perspectives.html

https://www.eamodelexpert.com/

http://eaxpertise.nl/ideaen.aspx

Sparx Systems is pleased to announce a new Sparx Systems University Week scheduled June 3^rd-7^th.

During Sparx Systems University week, a range of both free and paid training sessions are hosted at training locations around the world and online.  All sessions are supported by Sparx Service providers and qualified trainers in a number of different languages.

This week-long event is a unique opportunity to gain knowledge and improve skills around Enterprise Architect. June's course line-up covers a broad range of topics, and has been developed with both introductory and advanced users in mind.

The first Sparx University Week for 2018 will be run during March, with most training sessions being held during the week of March 2-9.

University Weeks are hosted by Sparx Systems Japan, Sparx Systems Central Europe, Sparx Services North America, Sparx Systems India and Sparx Services UK (Hippo Software).

The delivery format for each course varies depending on the subject, audience and location, with a mix of both online and face-to-face seminars and courses.

It is anticipated that Sparx University Week will be run on a global basis every few months, in conjunction with Sparx Systems Sister and Services Partner network.

Bookings are essential, please review the course schedule below and visit the course provider's website for more details and to register your place.

Sparx University Week Schedule:

Introduction

One of the challenges of a teambased architecture repository based on the ArchiMate language is to reduce the number of duplicate entities. With duplicates I mean an Archimate concept with the same name and stereotype that has two or more instances in the repository.

Especially in large repositories with a complex project browser configuration and a large number of modellers the changes of duplicates are substantial. Therefore mostly a procedure is introduced based on the steps that a modeller has to do a repository search before he or she can actually add an element from the toolbox and modify it

Search

For searching for elements in the repository there is a search dialog availabe that opens by default in a simple common search that searches for a text in all the items. This works fine but has a number of limitations for an ArchiMate modeller:

• The search is relatively slow
• The modeller is not interested in the full content but only in ArchiMate elements of a specific type.

Luckily we can define our own searches in EA and an example of a ArchiMate specific search is shown bin the image below:

In this image you see that we can search for application components only and the list is by default sorted by objectname. This will help us in a later stage in user friendliness. The SQL statement is given below.

SELECT t_object.ea_guid AS CLASSGUID, t_object.name as ObjectName, t_object.Stereotype, ParentPkg.Name as ParentName, t_object.ModifiedDate, t_object.Author, t_object.[Version], t_object.[Alias], t_object.Object_Type as CLASSTYPE,t_object.Note as Notes FROM t_object, t_package as ParentPkg WHERE (t_object.name LIKE '#WC#[Search Term]#WC#' OR '[Search Term]'='') AND t_object.stereotype IN('ArchiMate_ApplicationComponent') AND t_object.Package_ID = ParentPkg.Package_ID ORDER BY t_object.Name

The trick is in two parts of the select statement:

• The first one is that we filter on the stereotype of the element so we only see a limited number of elements based on that stereotype
• The second is that we do a trick with the search term. We combine a search term search with a full search so you can see all the elements when the search term is empty

Model view

This helps us a lot in finding the relevant elements relatively fast however the search screen has a number of limitations in which selecting all the time the right query is relatively user unfriendly. However luckily Enterprise Architect has another screen for implementing this user friendly approach in a convenient way: the model view.

In the modelview you can define your own search view and link these views to a search in the search component of EA. See the picture below

In this screen you see a sample of a number of ArchiMate specific searches based on their stereotype. When you expand one of these searches you get the list of elements based on this stereotype. Since we have a combination of a keyword search and a full search (on the stereotype) the modeller can select the approach that best fits his needs. In the view you can select an element by typing in the first characters another option is to open the conext menu and go to the search function in EA and now we open the richt search immediately.

Download

For download of this solution or for downloading other scripts or add ons please visit the eaxpertise website.

The second Sparx University Week will be run during October, with most training sessions being held during the week of October 23-27.

University Weeks are hosted by Sparx Systems Japan, Sparx Systems Central Europe, Sparx Services North America, Sparx Systems India and Sparx Services UK (Hippo Software).

The delivery format for each course varies depending on the subject, audience and location, with a mix of both online and face-to-face seminars and courses.

It is anticipated that Sparx University Week will be run on a global basis every few months, in conjunction with Sparx Systems Sister and Services Partner network.

Bookings are essential, please review the course schedule below and visit the course provider's website for more details and to register your place.

Sparx University Week Schedule:

Enterprise Architect User Group

London 2017; 18th - 19th May

The London

2017 meeting of the Enterprise Architect User Group sees a shakeup to the agenda in the form of an additional day being added to the roster. In additional to the traditional presentation day of User Stories, How to's etc the extra day added to the event is taking the form of a training day.

The training day adds to the event a selection of six, three hour training sessions on a variety of subjects from BPMN to TOGAF and Model Curation.

Location

Code Node, 10 South Place, London, EC2M 7BT

Get Directions

Agenda; Thursday 18th May

You can find information on these training sessions over at the EA User Group website.

Agenda; Friday 19th May

You can find a synopsis for each of these presentations over on the EA User Group website.

How to buy your tickets...

Tickets for the event are available directly from the EA User Group website and are priced as follows:

• Full two day event ticket; £550.00 +Vat
• Friday only ticket; £75.00 +Vat