Main Content

System Composer Concepts

System Composer™ combines concepts from systems engineering with concepts from Simulink®. This page defines these concepts and their respective applications to help you understand how these domains overlap in System Composer.

Use this page to learn about System Composer concepts and how they apply to systems engineering design. Each section defines a concept, explains how the concept is used in System Composer, then links to more information in the documentation. Based on your architectural modeling goal, review the corresponding section to learn more about key concepts associated with that goal.

Author Models to Describe Structure of System and Software Architectures

An architecture model in System Composer consists of the common Simulink constructions: components, ports, and connectors. An architecture represents a system of components.

An architecture model overview showing a model with components, ports, and connectors labeled

TermDefinitionApplicationMore Information

A System Composer architecture represents a system of components and how they interface with each other structurally and behaviorally.

Different types of architectures describe different aspects of systems. You can use views to visualize a subset of components in an architecture. You can define parameters on the architecture level using the Parameter Editor.


The root is at the top of an architecture hierarchy. A root architecture has a boundary defined by its architecture ports surrounding the system of interest.

The root architecture has a system boundary surrounding your architecture model. You can add architecture ports that define interfaces across the boundary.

Compose Architectures Visually


A System Composer model is the file that contains architectural information, such as components, ports, connectors, interfaces, and behaviors.

Perform operations on a model:

  • Extract root-level architecture.

  • Apply profiles.

  • Attach interface data dictionaries.

  • Generate instances from model architecture.

A System Composer model is stored as an SLX file.

Create Architecture Model with Interfaces and Requirement Links

A component is a replaceable part of a system that fulfills a clear function in the context of an architecture. A component defines an architectural element, such as a function, another system, hardware, software, or other conceptual entity. A component can also be a subsystem or subfunction.

Represented as a block, a component is a part of an architecture model that can be separated into reusable artifacts. Transfer information between components with:

Compose Architectures Visually


A port is a node on a component or architecture that represents a point of interaction with its environment. A port permits the flow of information to and from other components or systems.

  • Component ports are interaction points on the component to other components.

  • Architecture ports are ports on the boundary of the system, whether the boundary is within a component or the overall architecture model. The root architecture has a boundary defined by its ports.

Compose Architectures Visually


Connectors are lines that provide connections between ports. Connectors describe how information flows between components or architectures.

A connector allows two components to interact without defining the nature of the interaction. Set an interface on a port to define how the components interact.

Compose Architectures Visually

Represent Design Alternatives Using Variant Components

Create variant components and implement multiple design alternatives or variants, chosen based on programmatic rules. Add variant choices to a component to make a variant component. The active choice represents the original component.

Two variants and an active choice inside a variant component.

TermDefinitionApplicationMore Information

A variant is one of many structural or behavioral choices in a variant component.

Use variants to quickly swap different architectural designs for a component while performing analysis.

Create Variants
variant control

A variant control is a string that controls the active variant choice.

Set the variant control programmatically to control which variant is active.

Set Variant Control Condition

Describe Port Specifications Using Interfaces

Define interfaces to represent the kind of information that flows through a port. Assign interfaces to ports using the Interface Editor in Dictionary View. Use an Adapter block to reconcile differences between interfaces on a connector between ports.

An architecture model with the Interface Editor open. The interface data dictionary archDictionary.sldd, data interface Charger, data element Voltage, and value type VoltageType are labeled.

Manage owned interfaces local to a port using the Interface Editor in Port Interface View.

An architecture model with the Interface Editor open. The Power Source port and owned element elem0 are labeled.

TermDefinitionApplicationMore Information
interface data dictionary

An interface data dictionary is a consolidated list of interfaces and value types in an architecture and where you use them in the architecture.

You can save local interfaces on a System Composer model in an interface data dictionary using the Interface Editor. You can reuse data dictionaries between models that need to use a given set of interfaces, elements, and value types. Data dictionaries that you attach to models are stored in separate SLDD files.

data interface

A data interface defines the kind of information that flows through a port. The same interface can be assigned to multiple ports. A data interface can be composite, meaning that it can include data elements that describe the properties of an interface signal.

Data interfaces represent the information that is shared through a connector and enters or exits a component through a port. Use the Interface Editor to create and manage data interfaces and data elements and store them in an interface data dictionary for reuse between models.

data element

A data element describes a portion of an interface, such as a communication message, a calculated or measured parameter, or other decomposition of that interface.

Data interfaces are decomposed into data elements:

  • Pins or wires in a connector or harness.

  • Messages transmitted across a bus.

  • Data structures shared between components.

value type

A value type can be used as a port interface to define the atomic piece of data that flows through that port and has a top-level type, dimension, unit, complexity, minimum, maximum, and description.

You can also assign the type of data elements in data interfaces to value types. Add value types to data dictionaries using the Interface Editor so that you can reuse the value types as interfaces or data elements.

Create Value Types as Interfaces
owned interface

An owned interface is an interface that is local to a specific port and not shared in a data dictionary or the model dictionary.

Create an owned interface to represent a value type or data interface that is local to a port.

Define Owned Interfaces Local to Ports

An adapter connects two components with incompatible port interfaces by mapping between the two interfaces. An adapter can act as a unit delay, rate transition, or merge. You can also use an adapter for bus creation. Use the Adapter block to implement an adapter.

With an adapter, you can perform functions on the Interface Adapter dialog box:

  • Create and edit mappings between input and output interfaces.

  • Apply an interface conversion UnitDelay to break an algebraic loop.

  • Apply an interface conversion RateTransition to reconcile different sample time rates for reference models.

  • Apply an interface conversion Merge to merge two or more message or signal lines.

  • When output interfaces are undefined, you can use input interfaces in bus creation mode to author owned output interfaces.

Author Physical Behaviors for Components Using Simscape

Author physical models in System Composer by using subsystem components. A subsystem component is a Simulink subsystem that is part of the parent System Composer architecture model. Physical information can extend past subsystem component boundaries. You can describe physical components with physical ports, connectors and physical interfaces. You can use Simscape™ to describe the behavior of such physical components.

The Power component displays the physical interface WirelessPower, physical element ElectricalConn, and domain foundation.electrical.electrical.

TermDefinitionApplicationMore Information
physical subsystem

A physical subsystem is a Simulink subsystem with Simscape connections.

A physical subsystem with Simscape connections uses a physical network approach suited for simulating systems with real physical components and represents a mathematical model.

Implement Component Behavior Using Simscape
physical port

A physical port represents a Simscape physical modeling connector port called a Connection Port (Simscape).

Use physical ports to connect components in an architecture model or to enable physical systems in a Simulink subsystem.

Define Physical Ports on Component
physical connector

A physical connector can represent a nondirectional conserving connection of a specific physical domain. Connectors can also represent physical signals.

Use physical connectors to connect physical components that represent features of a system to simulate mathematically.

Architecture Model with Simscape Behavior for a DC Motor
physical interface

A physical interface defines the kind of information that flows through a physical port. The same interface can be assigned to multiple ports. A physical interface is a composite interface equivalent to a Simulink.ConnectionBus object that specifies a number of Simulink.ConnectionElement objects.

Use a physical interface to bundle physical elements to describe a physical model using at least one physical domain.

Specify Physical Interfaces on Ports
physical element

A physical element describes the decomposition of a physical interface. A physical element is equivalent to a Simulink.ConnectionElement object.

Define the Type of a physical element as a physical domain to enable use of that domain in a physical model.

Describe Component Behavior Using Simscape

Extend Architecture Modeling Language with Profiles and Stereotypes

Create a profile in the Profile Editor and add stereotypes to it with properties. Apply the stereotype to a component, and set the property value in the Property Inspector.

The Profile Editor and Property Inspector

TermDefinitionApplicationMore Information

Stereotypes provide a mechanism to extend the core language elements and add domain-specific metadata.

Apply stereotypes to core element types. An element can have multiple stereotypes. Stereotypes allow you to style different elements. Stereotypes provide elements with a common set of properties, such as mass, cost, and power.


A property is a field in a stereotype. You can specify property values for each element to which the stereotype is applied.

Use properties to store quantitative characteristics, such as weight or speed, that are associated with a model element. Properties can also be descriptive or represent a status. You can view and edit the properties of each element in the architecture model using the Property Inspector.


A profile is a package of stereotypes.

You can use profiles to create a domain of specialized element types. Author profiles and apply profiles to a model using the Profile Editor. You can store stereotypes for a project in one or several profiles. When you save profiles, they are stored in XML files.

Manage and Verify Requirements to Prove System Compliance to Stakeholder Needs

In the Requirements Perspective, you can create, manage, and allocate requirements. View the requirements for an architecture model. This functionality requires a Requirements Toolbox™ license.

The Transport Load requirement link on the robot variant component

Use Simulink Test™ to create a test harness for a System Composer component to validate simulation results and verify design in the Simulink Test Manager (Simulink Test). This functionality requires a Simulink Test license.

The test harness example in the Small UAV architecture model.

TermDefinitionApplicationMore Information

Requirements are a collection of statements describing the desired behavior and characteristics of a system. Requirements help ensure system design integrity and should be achievable, verifiable, unambiguous, and consistent with each other. Each level of design should have appropriate requirements.

To enhance traceability of requirements, link system, functional, customer, performance, or design requirements to components and ports. Link requirements to each other to represent derived or allocated requirements. Manage requirements from the Requirements Perspective on an architecture model or through custom views. Assign test cases to requirements using the Simulink Test Manager (Simulink Test) for verification and validation.

requirement set

A requirement set is a collection of requirements. You can structure the requirements hierarchically and link them to components or ports.

Use the Requirements Editor (Requirements Toolbox) to edit and refine requirements in a requirement set. Requirement sets are stored in SLREQX files. You can create a new requirement set and author requirements using Requirements Toolbox, or import requirements from supported third-party tools.

requirement link

A link is an object that relates two model-based design elements. A requirement link is a link where the destination is a requirement. You can link requirements to components or ports.

View links by using the Requirements Perspective in System Composer. Select a requirement in the Requirements Browser to highlight the component or the port to which the requirement is assigned. Links are stored externally as SLMX files.

test harness

A test harness is a model that isolates the component under test with inputs, outputs, and verification blocks configured for testing scenarios. You can create a test harness for a model component or for a full model. A test harness gives you a separate testing environment for a model or a model component.

Create a test harness for a System Composer component to validate simulation results and verify design. To edit the interfaces while you are testing the behavior of a component in a test harness, use the Interface Editor.

Define Relationships Between Elements of Different Architecture Models Using Allocations

In the Allocation Editor, allocate components between two architecture models, based on a dependency or a directed relationship.

The allocation set browser displays an allocation set with one allocation scenario between two models and their allocations.

TermDefinitionApplicationMore Information

An allocation establishes a directed relationship from architectural elements — components, ports, and connectors — in one model to architectural elements in another model.

Resource-based allocation allows you to allocate functional architectural elements to logical architectural elements and logical architectural elements to physical architectural elements.

allocation scenario

An allocation scenario contains a set of allocations between a source and a target model.

Allocate between model elements in an allocation scenario. The default allocation scenario is called Scenario 1.

Systems Engineering Approach for SoC Applications
allocation set

An allocation set consists of one or more allocation scenarios that describe various allocations between a source and a target model.

Create an allocation set with allocation scenarios in the Allocation Editor. Allocation sets are saved as MLDATX files.

Scope Complex Architectures into Simpler Diagrams by Creating Filtered Views

Apply a view filter to generate an element group of components for the view in the Architecture Views Gallery.

A generated view from a filter where the selected components have a stereotype.

TermDefinitionApplicationMore Information

A view shows a customizable subset of elements in a model. Views can be filtered based on stereotypes or names of components, ports, and interfaces, along with the name, type, or units of an interface element. Create views by adding elements manually. Views create a simplified way to work with complex architectures by focusing on certain parts of the architectural design.

You can use different types of views to represent the system. Switch between a component diagram, component hierarchy, or architecture hierarchy. For software architectures, you can switch to a class diagram view. A viewpoint represents a stakeholder perspective that specifies the contents of the view.

element group

An element group is a grouping of components in a view.

Use element groups to programmatically populate a view.


A query is a specification that describes certain constraints or criteria to be satisfied by model elements.

Use queries to search elements with constraint criteria and to filter views.

Find Elements in Model Using Queries
component diagram

A component diagram represents a view with components, ports, and connectors based on how the model is structured.

Component diagrams allow you to programmatically or manually add and remove components from the view.

Inspect Components in Custom Architecture Views
hierarchy diagram

You can visualize a hierarchy diagram as a view with components, ports, reference types, component stereotypes, and stereotype properties.

There are two types of hierarchy diagrams:

  • Component hierarchy diagrams display components in tree form with parents above children. In a component hierarchy view, each referenced model is represented as many times as it is used.

  • Architecture hierarchy diagrams display unique component architecture types and their relationships using composition connections. In an architecture hierarchy view, each referenced model is represented only once.

Display Component Hierarchy and Architecture Hierarchy Using Views

Use Analysis to Perform Trade Studies and Validate Architectures Against Constraints

Create an analysis function to analyze power consumption in the RobotDesign architecture model.

function RobotDesign_1(instance,varargin)

if instance.isComponent() && ~isempty(instance.Components)...
 && instance.hasValue('RobotProfile.ElectricalComponent.Power')
    sysComponent_power = 0;
    for child = instance.Components
        if child.hasValue('RobotProfile.ElectricalComponent.Power')
           comp_power = child.getValue('RobotProfile.ElectricalComponent.Power');
           sysComponent_power = sysComponent_power + comp_power;

Analyze the robot design by using the analysis function to determine total power usage.

The robot model Instance Properties are open, displaying total power consumption.

TermDefinitionApplicationMore Information

Analysis is a method for quantitatively evaluating an architecture for certain characteristics. Static analysis analyzes the structure of the system. Static analysis uses an analysis function and parametric values of properties captured in the system model.

Use analyses to calculate overall reliability, mass roll-up, performance, or thermal characteristics of a system, or to perform a size, weight, and power (SWaP) analysis to increase efficiency.

analysis function

An analysis function is a MATLAB® function that computes values necessary to evaluate the architecture using the properties of each element in the model instance.

Use an analysis function to calculate the result of an analysis.

instance model

An instance model is a collection of instances.

You can update an instance model with changes to a model, but the instance model will not update with changes in active variants or model references. You can use an instance model, saved in a MAT file, of a System Composer architecture model for analysis.

Run Analysis Function

An instance is an occurrence of an architecture model element at a given point in time.

An instance freezes the active variant or model reference of the component in the instance model.

Create a Model Instance for Analysis

Specify Operational Constraints Between Components Using Executable Sequence Diagrams

Create a sequence diagram in the Architecture Views Gallery to describe system interactions.

A lifeline, message, fragment, and operand in a sequence diagram for the robot design model.

TermDefinitionApplicationMore Information
sequence diagram

A sequence diagram represents the expected interaction between structural elements of an architecture as a sequence of message exchanges.

Use sequence diagrams to describe how the parts of a system interact.

Describe System Behavior Using Sequence Diagrams

A lifeline is represented by a head and a timeline that proceeds down a vertical dotted line.

The head of a lifeline represents a component in an architecture model.

Add Lifelines and Messages

A message sends information from one lifeline to another. Messages are specified with a message label.

A message label has a trigger and a constraint. A trigger determines whether the message occurs. A constraint determines whether the message is valid.

Create Messages in Sequence Diagram

A gate in a sequence diagram represents the root architecture of the corresponding architecture model.

Connect messages to gates to represent architecture ports.

Create Sequence Diagram Gates

An annotation describes the elements of a sequence diagram.

Use annotations to provide detailed explanations of elements or workflows captured by sequence diagrams.

Use Annotations to Describe Elements of Sequence Diagram

A fragment indicates how a group of messages specified within execute or interact.

A fragment is used to model complex sequences, such as alternatives, in a sequence diagram.

Author Sequence Diagram Fragments

An operand is a region in a fragment. Fragments have one or more operands depending on the kind of fragment. Operands can contain messages and additional fragments.

Each operand can include a constraint to specify whether the messages inside the operand execute. You can express the precondition of an operand as a MATLAB Boolean expression using the input signal of a lifeline.

Add Fragments and Operands

Simulate Integrated Architecture by Implementing Component Behaviors

Use a reference component to decompose and reuse architectural components and Simulink model behaviors. Use a subsystem component or state chart to implement Simulink and Stateflow® behaviors.

A reference architecture, a subsystem component, and a state chart behavior in the robot design model.

TermDefinitionApplicationMore Information
reference component

A reference component is a component whose definition is a separate architecture model, Simulink behavior model, or Simulink subsystem behavior. A reference component represents a logical hierarchy of other compositions.

You can synchronize and reuse reference components as Reference Component blocks. There are three types of reference components:

  • Model references are Simulink models.

  • Subsystem references are Simulink subsystems.

  • Architecture references are System Composer architecture models or subsystems.


A parameter is an instance-specific value of a value type.

Parameters are available for inlined architectures and components. Parameters are also available for components linked to model, subsystem, or architecture references that specify model arguments. You can specify independent values for a parameter on each component.

subsystem component

A subsystem component is a Simulink subsystem that is part of the parent System Composer architecture model.

Add Simulink subsystem behavior to a component to author a subsystem component in System Composer. You cannot synchronize and reuse subsystem components as Reference Component blocks because the component is part of the parent model.

state chart

A state chart diagram demonstrates the state-dependent behavior of a component throughout its state lifecycle and the events that can trigger a transition between states.

Add Stateflow chart behavior to describe a component using state machines. You cannot synchronize and reuse Stateflow chart behaviors as Reference Component blocks because the component is part of the parent model.

Author and Simulate Activity Diagrams to Allocate to Components

Activity diagrams authored in System Composer describe high level functionality of the system in modeling system dynamics. Use activity diagrams to describe the behavior of systems as a transformation of inputs to outputs through actions that process token flows. You can also simulate and visualize activity diagrams to validate system behavior.

You can allocate activity diagram elements to elements of a System Composer architecture model using the Allocation Editor to more fully describe your functional architectural design. For more information, see Design Architectures and Activity Diagram for Mobile Robot.

Activity Diagram Concepts

TermDefinitionApplicationMore Information
activity diagram

An activity diagram is the primary diagram to describe an activity. An activity describes system behavior that models the flow of tokens from inputs to outputs through a controlled sequence of actions. An activity diagram contains action nodes with pins connected by flow lines. Activity diagrams represent the flow of tokens through various actions that describe the execution of activities.

Use activity diagrams to conceptualize a system, visualize functional flow through actions or decisions, and understand how system components interact with one another. You can use activity diagrams to describe the behavior of systems as a transformation of inputs to outputs through actions that process token flows.


Tokens are objects that flow in the activity diagram. A token can represent data such as structures and integers, or simply pass on the control.

Use a token to move data or control across the activity diagram. These are the types of tokens:

  • Object token — Represents an object such as a piece of data.

  • Control token — Represents a control or a triggering event that does not carry any data.

action node

An action node is a key building block in an activity diagram. An action node represents an action to be executed. Action nodes consume input tokens and produce output tokens on pins.

Use a MATLAB function or a nested activity diagram to describe the behavior of an action node.

control node

A control node routes a logical flow of tokens through the system.

Use control nodes and flows to route tokens. Control nodes can be used to initialize, split, merge, and terminate token flows.

Use Control Nodes to Manipulate Token Flows

A pin directs tokens in or out of an action node. The directionality of the pin represents input or output. You can connect pins by object flows.

Use pins to route an object token to or from an Action Node. Pins are also used to store object tokens before or during execution. You can use pins only for object flows.

Author Activity Diagrams

A flow in an activity diagram connects two nodes. A dashed line represents a control flow and a solid line represents an object flow.

You can use object flows to route input or output tokens to carry information or physical items between object nodes. You can use control flows to model transfer of control from one Action Node to another. These are the types of flows:

  • Object flow — Tokens in an object flow contains token data on which actions operate.

  • Control flow — Tokens in a control flow trigger the execution of actions.

Simulate, Visualize, and Validate Activity Diagrams

Author, Simulate, and Deploy Software Architectures

Design a software architecture model, define the execution order of the functions from the components, simulate the design in the architecture level, and generate code.

A software architecture with a software component linked to an Export-Function model.

View the software architecture diagram as a class diagram in the Architecture Views Gallery.

A class diagram view.

TermDefinitionApplicationMore Information
software architecture

A software architecture is a specialization of an architecture for software-based systems, including the description of software compositions, component functions, and their scheduling.

Use software architectures in System Composer to author software architecture models composed of software components, ports, and interfaces. Design your software architecture model, define the execution order of your component functions, simulate your design in the architecture level, and generate code.

software component

A software component is a specialization of a component for software entities, including its interfaces.

Implement a Simulink export-function, rate-based, or JMAAB model as a software component, simulate the software architecture model, and generate code.

software composition

A software composition is a diagram of software components and connectors that represents a composite software entity, such as a module or application.

Encapsulate functionality by aggregating or nesting multiple software components or compositions.

Model Software Architecture of Throttle Position Control System

A function is an entry point where a transfer of program control occurs and can be defined in a software component.

You can apply stereotypes to functions in software architectures, edit sample times, and specify the function period using the Functions Editor.

Author and Extend Functions for Software Architectures
function element

A function element describes the attributes of a function in a client-server interface.

Edit the function prototype on a function element to change the number and names of inputs and outputs of the function. Edit function element properties as you would edit other interface element properties. Function argument types can include built-in types as well as bus objects. You can specify function elements to support:

  • Synchronous execution — When the client calls the server, the function runs immediately and returns the output arguments to the client.

  • Asynchronous execution — When the client makes a request to call the server, the function is executed asynchronously based on the priority order defined in the Functions Editor and Schedule Editor and returns the output arguments to the client.

function argument

A function argument describes the attributes of an input or output argument in a function element.

You can set the properties of a function argument in the Interface Editor just as you would other value types: Type, Dimensions, Units, Complexity, Minimum, Maximum, and Description.

service interface

A service interface defines the functional interface between client and server components. Each service interface consists of one or more function elements.

Once you have defined a service interface in the Interface Editor, you can assign it to client and server ports using the Property Inspector. You can also use the Property Inspector to assign stereotypes to service interfaces.


A server is a component that defines and provides a function.

A server component is where the function is defined. You can implement function behavior in a Simulink export-function model.

Service Interfaces Overview

A client is a component that sends a request to the server.

A client component is where the function is called. The implementation of function call behavior is dependent on the synchronicity of the function execution.

Service Interfaces Overview
class diagram

A class diagram is a graphical representation of a static structural model that displays unique architecture types of the software components optionally with software methods and properties.

Class diagrams capture one instance of each referenced model and show relationships between them. A component diagram view can be optionally represented as a class diagram for a software architecture model.

Class Diagram View of Software Architectures

Related Topics

External Websites