Cameo Simulation Toolkit: v19SP3: GOTCHA/BUG: When a Property is typed by an abstract Block an instance of a concrete specialising Block (where available) will automatically be created and assigned even when the lower multiplicity is 0
MagicDraw/Cameo 19SP3: Activity Diagrams: Pin display mode 'Position of Labels' does not work correctly for 'Name And Type Labels Inside', only shows the Name inside (not the Label)
Cameo Simulation Toolkit: You can assign the results of simulations calculations on instances to the default values of properties of the Classes/Blocks that type the instance.
UML/SysML: MagicDraw/Cameo: 19SP3: Instance Table Diagram sometimes does not display the values for nested value properties
UML/SysML: MagicDraw/Cameo: Instance Table Diagrams are one of the most powerful features of MagicDraw/Cameo! Learn to use them to progressively test, debug, and evolve your models as you work!
UML/SysML: MagicDraw/Cameo 19SP3: GOTCHA/TIP: An Instance Table will not display a row for instances typed by an abstract Class/Block (although one can show such instances as InstanceSpecification symbols on a Class Diagram or Block Definition Diagram)
UML/SysML: Cameo Simulation Toolkit 19SP3: GOTCHA/TIP: ConstraintBlock constraints: Not every available constraint language can handle Enumeration literals (if in doubt choose 'English')
UML: Cameo Simulation Toolkit 19SP3: GOTCHA: CreateObjectAction ignores an Artifact as 'classifier' even though Artifact is a Classifier
GOTCHA: UML/SysML: Cameo Simulation Toolkit 19SP3: A [true] or [false] guard on an ActivityEdge MUST be a LiteralBoolean, not just characters typed in, or the evaluation may give unexpected or ill-defined results
UML/SysML: Cameo Simulation Toolkit 19SP3: A parent Activity with a DecisionNode that uses a Activity as a decisionInput Behavior terminates immediately after the decisionInput terminates (but OpaqueBehavior works)
UML/SysML: Cameo Simulation Toolkit 19SP3: GOTCHA: Will not evaluate a guard using a token from a decisionInputFlow UNLESS a decisionInput Behaviour is explicitly defined (but Alf does)
Webel electronics modelling recipe for SysML: Diagrams can be made less verbose by using stereotype icon shape display mode, by not showing names and types on obvious Ports, by hiding context-specific values, and by omitting {net} tagged values.
MagicDraw/Cameo: Rule: Internal Block Diagram and Block structure compartment: only one symbol for a given Property may be shown.
MagicDraw/Cameo: Display option: You can choose whether to show Domain Specific Language (DSL) stereotypes: 'None', 'All', 'Only Last'
Webel: SysML: Electronics modelling: TIP: Use a custom Connector stereotype '<>' with the keyword '<>' to carry a 'net' property. (Use of punctuation in Stereotype names is not usually recommended and the keyword is not usually displayed.)
Webel Parsing Analysis: MagicDraw/Cameo: SysML1.6+: ElementGroup-based Snippets do not list in the compartment of Package/Model symbols because ElementGroup is based on the UML Comment, which does not list in Package/Model symbols.
Webel Parsing Analysis: An "index" Parsing Analysis Diagram (PAD) showing a collection of Snippets may optionally (but need not always) show the /member tagged value of each Snippet.
UML/SysML: MagicDraw/Cameo: WARNING: If you assign a value to a Slot for an untyped Property then assign a ValueType later it will DELETE your assigned value! Assign a type first .
The SysML1.6 derived /tracedTo is only available on AbstractRequirement (but in MagicDraw/Cameo you can use derived relationships to achieve the same thing on other kinds of NamedElement).
The SysML Trace relationship can be used as a quick way to traceably elicit model elements from an identified diagram or table from a domain source document. You may visually remove the Trace symbols as each element is elicited to reduce clutter.
The UML «Trace» and SysML Trace can't be applied to a Slot as target (because Slot is just an Element, not a NamedElement).
Webel Parsing Analysis: Track and display alternative names, human friendly names, organisation-specific names, and identifiers using tagged values for a custom stereotype «pa:term» (adapt or extend as required).
Webel Parsing Analysis: It does not matter whether you use a Package or a Model package. (The informal Webel convention is that Models are used for systems engineering analysis with SysML and Packages are reserved for code-related software engineering.)
Webel Parsing Analysis: A stereotype with keyword «pa:from» may be applied to a Dependency from a Package to another Package within the Source Input Zone to indicate that all of its Elements were directly or indirectly elicited from source Documents.
Webel Parsing Analysis: The name of a Parsing Analysis Diagram (PAD) may be drawn from a focus Snippet OR may simply indicate a topic of interest to the analysis
Webel Parsing Analysis: Where too many dashed line "anchors" from Snippets to elicited members lead to clutter they may be selectively omitted from a Parsing Analysis Diagram (PAD) once each member has been collected.
Webel Parsing Analysis: While Associations may be collected as elicited members of a Snippet this can quickly lead to clutter in the /member list display. Often just collecting an end Property as member is sufficient.
Webel Parsing Analysis: While Generalizations may be collected as elicited members of a Snippet this can quickly lead to clutter in the /member list display.
Webel Parsing Analysis: An anonymous Element may be collected as a /member of a Snippet (it is not important whether collected elements list with a clear name under /member, only that they are traceably elicited).
Webel Parsing Analysis: A "focus" Parsing Analysis Diagram (PAD) for one or more Snippets SHOULD always show the /member tagged value of every Snippet.
Webel Parsing Analysis: Acronym: PAD = Parsing Analysis Diagram (may be nearly any diagram type, except those types that must be owned by an elicited model element)
Webel Parsing Analysis: As StateMachine Diagrams are always owned by an element that is not eventually under the Source Input Zone they should not be used as «pa» diagrams. Elicit States instead via the /member section of the specification dialog.
Using Model-Based Systems Engineering with ANY version of Systems Modeling Language (SysML) is better than just doing drawings in a graphics tool or presentation tool
You DO NOT have to wait for SysMLv2 to do excellent Model-Based Systems Engineering with SysML! You can do MOST of what you need to do with SysMLv1.6/1.7 and you WILL be able to migrate models to SysMLv2 (once it matures). Start your SysML models NOW!
MagicDraw/Cameo v19SP3: UML/SysML: HOWTO Set an OpaqueExpression on a Slot that seems to be stuck on a numerical value.
MagicDraw/Cameo v19SP3: vs SysPhS-1.1: OpaqueExpression for Slot value exports to Modelica as 'null' if only uses a single variable. WORKAROUND/HACK prefix the variable with '1 *' [FIXED in v2021x]
Webel Parsing Analysis: The symbol for a Parsing Analysis Container MUST be able to indicate the unique domain source document from which its analysed text extract was quoted.
Webel Parsing Analysis: The symbol for a Parsing Analysis Container MUST be able to OPTIONALLY display a list of names of elicited model elements (a.k.a. members)
Webel Parsing Analysis: It SHOULD be possible to relate one Parsing Analysis Container to another and stereotype that relationship as an RDF/OWL-like semantic triple
Webel Parsing Analysis: It MUST be possible to collect elicited members of a Parsing Analysis Container graphically in diagrams using relationship-like path drawing (not just via a specification dialog or other indirect modelling means)
Webel Parsing Analysis: The symbol for the relationship between a Parsing Analysis Container and elicited model elements SHOULD be a dashed line (like the anchor/handle line used with a UML Comment)
Webel Parsing Analysis: The symbol for a Parsing Analysis Container MUST be text-friendly and evocative of quoting a domain source text extract
Webel Parsing Analysis: The symbol of a Parsing Analysis Container MUST be usable on every possible diagram type
SysMLv2: On the v2 Comment extension of the v2 AnnotatingElement as a candidate Parsing Analysis Container
SysMLv1.x: Q: Why can't a Package with PackageImports be used as a Parsing Analysis text container? Why is the SysML1.6 ElementGroup (extended and customised as the Webel «snippet») far better suited for text-driven model element elicitation?
The targeting of the Modelica and Simulink simulation language families by the SysML Extension for Physical Interaction and Signal Flow Simulation (SysPhS) encourages development of SysML models aligned with known practices for a wide class of problems!
MagicDraw/Cameo v19SP3: vs SysPhS-1.1: Modelica export: Direct binding from a PhSVariable value property within a FlowProperty of a Port to an inner value property does not flatten. WORKAROUND: Use an intermediate constraint property.
MagicDraw/Cameo v19SP3: Does not support display of context-specific values on a FlowProperty symbol within a Port in an IBD
SysML vs Modelica: GOTCHA: Terminology: A 'connector' in Modelica is equivalent to the Type of a Port in SysML. A Connector in SysML-1.x is equivalent to a 'connect(source,target)' in Modelica.
MagicDraw/Cameo: Modelica export: Need environment option to disable generation of layout annotations (not just on individual export dialog)
MagicDraw/Cameo v19SP3+SysPhS-1.1: Does not cleanly export INLINE Modelica 'if/then/else' statements (a.k.a. "switching" form)
SysPhS: MagicDraw/Cameo v19SP3: Export to Modelica does not interpret as 'start' the default on a PhSVariable assigned via ElementValue to a PhSConstant
MagicDraw/Cameo: GOTCHA: When applying a numerical default to a value property make sure the value property has already been typed by Real or Integer (or a ValueType that extends one of them) otherwise it will assign a LiteralString as default.
SysML: Naming: You may include Block, ValueType, and Signal names in the names of Behaviors (such as Activities) as long as this does not undermine the principles of functional analysis and allocation.
SysML: Naming: Including Block, ValueType, and Signal names in the names of Behaviors (such as Activities) can sometimes undermine purist functional allocation (because it may presuppose the element of the physical solution that carries out the function).
MagicDraw SysML/Cameo: If you have a SINGLE Allocate from an Activity to a Block it will set that Block to be the Behavior::/context for that Activity.
SysML: Having a Behavior owned by a Block it is allocated to may undermine purist functional allocation (because it presupposes the element of the physical solution that carries out the function)
UML/SysML: In Internal Block Diagrams: If you have a Port with a name that indicates a unique role AND and if there is an ItemFlow on a Connector that implies or suggests the Type of the Port, consider hiding the Type on the Port symbol.
SysML+SysPhS: Flows: If you intend to use a Classifier to type the itemProperty of an ItemFlow on a Connector for a physical interaction you MUST use a Block (not a ValueType or Signal) so your can extend ConservedQuantityKind.
SysML: Flows: If you intend to use a Classifier to type the itemProperty of an ItemFlow on a Connector use a ValueType or Block rather than a Signal, otherwise you'll get probably-unwanted signal receptions on the owning context Block.
SysML: Whether you use a Block, ValueType, or Signal to represent something that flows (and can be applied to an ItemFlow) depends on what you want to achieve. If you want to indicate something structured with value properties with Units use a Block.
SysML: The placement of usages of Blocks, their Ports, and Connectors in an Internal Block Diagrams DOES NOT necessarily represent physical geometry!
In the Webel terminology for a basic control loop there is an 'aim' value (controlled via an actuator) and a 'got' value (a value read from a sensor). The 'got' value is NOT necessarily exactly the same as the actual physical value.
Additional Dependency relationships between ValueTypes and their Units on some SysML diagrams on this site are for educational illustration only (you don't need them in your own SysML models).
MagicDraw/Cameo: SysML Parametrics: To colour instance table cells to indicate broken constraints you need to create at least one Simulation Configuration Diagram, which loads a validation profile with colour coding rules.
MagicDraw/Cameo v19SP3 vs SysML&SysPhS: Export to Modelica: The exported layout annotations appear to be completely broken (at least vs Wolfram SystemModeler) just de-select generation of them on export.
GOTCHA: MagicDraw SysML/Cameo 19SP3: Export to Modelica: The name of a redefining Property must be exactly the same as the Property it redefines or it will not export properly!
Webel vs SysPhS-1.1: Annex A.5: Humidifier: Where ValueTypes involving litre are defined, the Unit symbol "L" is used rather than the Modelica-preferred "l" (in combination with an explicit additional unit converter).
Webel vs SysPhS-1.1: Annex A.5: Humidifier: The water temperature from TemperatureIncreaseConstraint and HeatingCalculationConstraint starts at 0 °C (should probably be the environment temperature 20 °C). Needs an additional parameter and initial value.
Webel vs SysPhS-1.1: Annex A.5: Humidifier: Where custom ValueTypes are defined, Modelica-friendly Unit symbols are used. Examples: "m3" not "m^3"; "degC" not "°C"; "J/(K.L)" (full stop as multiplier) not "J/(K⋅L)"; (EXCEPT "L" for litre not "l").
Webel: SysML: DO NOT sacrifice modelling naming conventions for the mere sake of carrying organisation-specific names! Instead use tagged values of custom stereotypes as metadata to carry alternative names in parallel with systematic model element names.
Webel: In some SysML trails ValueType names with Unit indicator suffixes have been used for dimensional analysis and illustrative purposes. This practice is NOT otherwise recommended here. Instead just use consistent custom ValueTypes across your system!
MagicDraw/Cameo: If you drag a Behavior from the model browser onto the SYMBOL of a State (that does not already own the Behavior) and set it as an 'entry', 'doActivity', or 'exit' Behavior a "wrapper" Behavior owned by the State will be created.
MagicDraw SysML/Cameo 19SP3: GOTCHA: If you drag a Behavior from the model browser onto the 'entry', 'doActivity', or 'exit' field in a specification dialog of a State it WILL BECOME OWNED by the State (which "steals ownership")!
MagicDraw/Cameo: Display options: On an 'entry', 'doActivity', or 'exit' Behavior of a State you may choose to display the specification code of an OpaqueBehavior or just its name.
MagicDraw/Cameo v19SP3 vs SysML&SysPhS: GOTCHA: Will NOT correctly export a SysML-1.6 «~interfaceBlock» to Modelica (but does cope with a tilde ~ prefix in the name of a plain InterfaceBlock)
MagicDraw/Cameo vs SysML&SysPhS: GOTCHA: Will NOT export a non-normative System «system» block to Modelica, it only recognises a plain Block «block»!
Sample problems and example diagrams in graphical language specifications are NOT necessarily indications of how one should model on a real-world project! They often just serve to demonstrate a particular aspect of the specified language.
SysML: When using Property::defaultValue and Property::redefinedValue to carry a "shadow hierarchy" of redefinitions across an entire system hierarchy, considering using a user-defined keyword such as «configuration» or «scenario» on each redefining Block
UML/SysML: Navigation: Try to offer a way out of a diagram (usually up a hierarchy, but possibly across) using a navigable symbol (linked to a diagram) and/or a diagram symbol. Avoid "cul-de-sacs"!
MagicDraw SysML/Cameo: You can customise the generic query table diagram kind to check that the context Blocks of your Parametric Diagrams only contain BindingConnectors (not Connectors)
Block naming: If you find you've got similar blocks named 'Thing' and 'Thing2' then 'Thing' is probably better renamed 'Thing1'
SysPhS-1.1: Annex A.5: Humidifier: Use of UML-style direct Port conjugation not permitted since SysML-1.6, prefer ~InterfaceBlock type-based conjugation (example requires migration)
MagicDraw/Cameo: v19SP3: Property created by dragging onto a Class or Block a symbol of a Classifier named with a single letter capital name 'N' has a poor name 'N:N'
MagicDraw/Cameo: You can drag a Class symbol onto a Class symbol (in a Class Diagram) or a Block symbol or ValueType symbol onto or Block symbol (in a Block Definition Diagram) to create a new part property or value property
MagicDraw/Cameo: You can drag out a Property (or Port) from a Class or Block symbol to create an Association with the Property (or Port) as one end
MagicDraw/SysML vs SysPhS-1.1: Can't reproduce the "shortcut" property path representation of some properties nested within Ports as shortcut symbols fully inside the diagram frame. (Might be a specification diagram style issue.)
Webel vs SysPhS-1.1: Diagramming style: DO NOT recommend overlapping Connectors with "phantom" fork (or junction) as shown in in 'Figure 48: Internal structure of the signal processor'
SysPhS: MagicDraw/Cameo: In the sysphs_profile the properties for PhSVariable have multiplicity [1], so the defaults always appear explicitly (but may be overridden): isContinuous: Boolean = true, isConserved: Boolean = false, changeCycle: Real = 0
SysML: HOWTO Safely incorporate ConstraintBlock equations from a library into a project with local ValueType variants: CASE: SysPhS vs ISO-80000 ModelLibrary
SysPhS-1.1: When RealSignalInElement or RealSignalOutElement are used to type a SysML Port, assignments to the FlowProperty 'rSig' map to the Port name (only) when mapped to Modelica.
GOTCHA: MD SysML/Cameo 19SP3 vs SysPhS-1.1: Export to Modelica does not see a 'doActivity' on a State (does not write it to a Modelica algorithm), you MUST use an 'entry'!
SysML: Although not encouraged, you can still use a DataType to type a Property of a Block, it just won't be listed in the 'values' compartment. Prefer the SysML ValueType versions of primitives!
GOTCHA: MagicDraw SysML/Cameo 19SP3: Export to Modelica: 'entry', 'doActivity', or 'exit' Behaviors of a State must be directly owned (not just "wrapped") or they won't be seen on export!
SysML-16: Taken literally the text and OCL of constraint 'Block::6_valueproperties_composite' imply that every FlowProperty typed by a ValueType should have AggregationKind 'composite'
MDSysML19SP3: Validation engine does not report value properties with AggregationKind other than composite
MagicDraw SysML/Cameo: 19SP3: Does not seem to support export to Modelica from a Package [use a SysML Block as root instead]
MagicDraw SysML/Cameo: SysPhS: Export to Modelica: It does not seem to matter whether you give the language of Constraints as 'Modelica' or 'sysphs' (although SysPhS specifies a restricted SysPhS expression grammar)
MagicDraw SysML/Cameo: GOTCHA: Export to Modelica only includes owned Constraints (not just applied Constraints) under the 'equation' export
MagicDraw SysML/Cameo: GOTCHA: A Constraint created and applied to a ConstraintBlock via the sidebar menu is NOT owned by the ConstraintBlock!
MagicDraw SysML/Cameo: 19SP3: Export to Modelica from TestBed (for SignalProcessor) sample does not validate in Wolfram SystemModeler
MagicDraw SysML/Cameo: 19SP3: Export to Modelica from SysPhS sample for HumidifierSystem does not execute in Wolfram SystemModeler Simulation Centre
Webel Best Practice: SysML/SysPhS: DO NOT use overlapping Connector lines from/to one Port (can be misunderstood as "summed" flow and/or physical node/fork/junction).
UML/SysML: A Boolean "state flag" attribute corresponding to a State can be useful for indicating states in some diagram types (but must be synchronised with Transitions to States carefully).
UML/SysML: During modelling it can really help to create little instance (object) diagrams as you go! This can also help communicate with non-SysML stakeholders.
UML/SysML: If you have a Boolean "state flag" attribute corresponding to a State you MUST set it on an 'entry' of the State, not on the 'effect' of a Transition into the State (otherwise with multiple incoming Transitions it could be WET and breaks SSoT).
UML/SysML: When you use an AddStructuralFeatureValueAction to set one end of a bi-directional Association it also sets the other end.
Cameo Simulation Toolkit matches the Parameters of an effect Behavior and a trigger Operation on a Transition under the hood.
Webel Twin Pattern: Once a PotentialPhysicalAsset has been integrated with a real, existing (has mass) ActualPhysicalAsset under twinning control, its job is done forever (although it may be kept as an historical tracking record).
Webel Twin Pattern: A DigitalTwin (even a "process twin" or "finance twin"), always, without exception, BY DEFINITION HERE either directly or indirectly involves at least one existing or anticipated PhysicalEntity (even if only via another DigitalTwin)!
Webel Twin Pattern: A DigitalEntity (a.k.a. «digital» @Entity) used as a digital variant must not be directly bound to a «physical» PhysicalEntity (may not directly actuate/sense/affect)
Webel Twin Pattern: A DigitalTwin may use one or more variantEntity:@Entity[0..*] to explore the impact of changes (optimisation studies, trade off studies) and then use them to drive the PhysicalEntity into a desired state (via its ControlSystem).
Webel Twin Pattern: The primary aim of the digitalEntity:@Entity of a DigitalTwin is to replicate its physical:PhysicalEntity as closely as possible (not to explore variants).
Webel Twin Pattern: A «digital» DigitalEntity (a.k.a. @Entity) is NOT a representation! It is a digital encapsulation (it can only be perceived at the level of "bits and bytes"). It HAS many representations (such as views)!
WARNING: In natural language casual conversation one often hears people speaking of a digital twin "replicating" or "twinning" a physical entity. If you do it that way literally, you will NOT have anything to manage the control system loop!
Webel Twin Pattern: A PotentialPhysicalAsset is «digital», not «physical». It is used to acquire (an existing) or create (build, manifest, make) an «physical» ActualPhysicalAsset, which is a special case of «physical» PhysicalEntity.
Webel Twin Pattern: A DigitalTwin manages a control loop including a PhysicalEntity and a DigitalEntity (a.k.a. @Entity) that «replicates» it. It typically includes at least one Sensor and at least one Actuator.
Webel Twin Pattern: It is a DigitalEntity (a.k.a. @Entity) - not the DigitalTwin that owns it - that directly «replicates» geometrical, spatial, and material aspects (only) of a single PhysicalEntity.
Webel Twin Pattern: A «digital» DigitalEntity (a.k.a. @Entity) encapsulates strictly geometrical, spatial, and material aspects of a «physical» PhysicalEntity only. BY DEFINITION of this pattern it DOES NOT encapsulate processes!
Webel: SysML-1.7/SysMLv2: WISHLIST: Constraint: A BindingConnector used for pure proxying MUST NOT be typed by an AssociationBlock by definition, because the associated information can be mis-appropriated to undermine the proxy equality!
WISHLIST: Webel has suggested that future SysML should support the display of Classifier-level :features in compartments on a rectangular itemProperty symbol on an ItemFlow on a Connector in an IBD.
Webel: SysMLv1.x: AVOID (where possible) SysML Unit names that are the same as unit symbols. Unit names SHOULD start with a lower case Latin alpha letter. Custom Unit names should be a single lower case word or lowerCamelCase.
MagicDraw/Cameo v19SP3: uses the Block compartment name 'signal receptions' instead of just 'receptions'. [CLAIMED FIXED in v2021x]
UML/SysML: In Internal Block Diagrams: Consider hiding the name of a named Port or Property in a Diagram if its Type is sufficient to indicate its role.
Dependency/Usage relationships in some referenced UML/SysML diagrams on this site are for educational illustration only, they are NOT part of the actual model.
