Information Models¶

What are Information Models¶

In the SMARTUNIFIER, an Information Model is defined as the data related to communication that a device or IT system can provide. Each device or IT system is represented by an Information Model. An Information Model is composed of elements known as Node Types, and these models are structured hierarchically like a tree. This means that elements within an Information Model can include additional elements. This hierarchical structure is essential for accurately modeling the data structure of devices in a way that reflects their real-world complexity.

Before starting the modeling process for the data structure of, for example, equipment or an IT system, it’s crucial to have a clear understanding of the overall use case. Generally, each piece of equipment or IT system that needs to be integrated will have its own Information Model. However, if there are multiple pieces of equipment or IT systems of the same type, a single Information Model can be used for all of them.

During the later stages of configuring a SMARTUNIFIER instance, the Information Model that is included in a Mapping, as well as the Mapping itself, can be applied to a Device Types. This allows for the creation of multiple instances from a single device type. Therefore, the Information Model serves as a blueprint for the data structure of a device or IT system and needs to be created only once. Understanding the reusability of the Information Model is crucial.

This concept emphasizes that once an Information Model is established, it can be effectively reused across different devices or IT systems of similar types, significantly streamlining the configuration process and enhancing efficiency.

Contextualization¶

Contextualization describes the process and functionality of connecting and combining correlated data, often across system boundaries. This context data can be stored and used internally within SMARTUNIFIER by utilizing Information Models to define the data structures, in combination with the InMemory Communication Channel. The context data can then be reused and mapped to other systems wherever the data is needed.

How to create a new Information Model¶

Follow the steps described below to create an Information Model:

• Select the SMARTUNIFIER Information Model Perspective (1).

• You are presented with the following screen containing a list view of existing Information Models.

• In order to add a new Information Model, select the “Add Model“ button at the top right corner (2).

• On the following screen provide the following mandatory information: Group and Name (3).

• The “Apply” button at the top right corner is enabled after all mandatory fields are filled in. Click the button to create a new Information Model (4).

• The newly created Information Model is now visible as a node on the left side of the screen.

• After the root model node is created, a new Information Model can be built up using definition types.

• Perform a right click on the root model node and select “Add Node” (5). Select a Definition Type from the dialog (6).

Node Types¶

Basics¶

Based on the specific use case and the communication channels involved, it is important to select the appropriate Node Type. Node Types are key components of an Information Model and include various elements such as variables, properties, events, commands, and collections like arrays and lists.

Each Node Type is associated with a Data Type, which specifies whether it is a predefined data type (such as String, Integer, Boolean, etc.) or a custom data type.

Example of a Node of Type Variable:

Naming Restrictions¶

There are specific restrictions on the naming conventions for Node Types. For instance, Scala keywords are not permissible. Moreover, a Node Type name cannot begin with a number, and the underscore character “_” is the sole permitted separator.

If the name of the Node, for example, the name of a Variable, requires a specific naming convention that cannot be used as a Node Type name, it can be customized using the “Field Name” option within the corresponding Communication Channel. Supported Communication Channels with the option to set a custom field name include:

• File Reader (CSV, JSON, XML)

• File Tailer (CSV, JSON, XML)

• File Writer (CSV, JSON, XML)

• MQTT (CSV, JSON, XML)

• SFTP File Writer (CSV, JSON, XML)

• WebSocket Client (CSV, JSON, XML)

List of Scala Keywords Prohibited as Node Type Names in the Information Model:

Available Node Types¶

Data Types¶

There are two different types of Data Types that can be used in the Information Model:

Predefined Types¶

These are the standard (primitive) data types that are available in the SMARTUNIFIER. They include,

Data types like these can be applied to the following Node Types: Variables, Properties, Arrays, Lists.

Example - Variable with a Predefined Data Type:

• Add a new Variable

• Enter an ID (Name)

• select a predefined data type e.g., “String” (1)

Custom Types¶

Custom data types are user-defined data types that can be created and used in the Information Model. They can be applied to the following Node Types: Variables, Properties, Arrays, Lists, Events, Commands. They come in handy, for example, when a complex data structure is required.

The name of the Data Type is to be defined by the user. It is recommended to name custom data types in a way that reflects the data structure they represent.

Example - Variable with a Custom Data Type:

• Add a new Variable

• Enter an ID (Name) e.g., “MyFirstComplexVariable”

• Enter a custom name for the Data Type e.g., “MyFirstComplexVariableType” (1)

Now another Variable can be added under the “MyFirstComplexVariable” - Variable:

• Perform a right click on the Custom Variable - “MyFirstComplexVariable”

• Select “Add Node” and choose Variable (2)

Structures Required by Channels¶

The structure of an Information Model is influenced by the Communication Channel used in the integration. Communication Channels can be classified into two types: data-driven and event-driven communication.

• Event-driven involves integration triggered by an event within a system, such as the receipt of goods. This event activates a Rule within the Mapping of SMARTUNIFIER.

• Data-driven, on the other hand, is initiated by a change in the data within a system.

• Additionally, there is command-driven integration, where an event in one system immediately requires a response from another system. This differs from event-driven integration, which does not anticipate a direct reply.

Below, a table presents some examples of use cases along with the Information Model structures they require:

Importing Data Structures¶

Data structures can be imported by using extensions, which is especially convenient when dealing with complex structures that include many variables. The following import options are available:

• OpcUa Model Import : Data structures can be imported from an OPC UA server.

• JSON Model Import: JSON structures can be imported directly.