I suspect that this is a typo. I don't think that ARP4751 exists BUT ARP4761 exists. Review and make sure we are referencing the correct standard.
There is a need to augment the RAAML existing concepts with a set of extensions and refinements to accommodate security related information as it impacts the measures that RAAML seeks to model (e.g., safety, reliability).
There is coordination with the UAF team, by Mary Tolbert (UAF Lead on Security Viewpoints), to ensure there is not any overlap with security concepts. Mary has attended all RAAML technical meetings along with providing support for STPA support for security consistent with SAE J3187 extensions for security.
In addition, Bob Martin (SAACM Chair), has been involved in the RAAML RTF meeting with the following comments on inclusion of security concepts.
“The first was the June 2022 meeting in Orlando where we discussed the underlying security concepts - mainly of weaknesses and vulnerabilities - and how weaknesses at the design, architecture, code, or deployment levels can lead to undesired behaviors. These may be undesirable from many perspectives - from a reliability one, a safety one, or a security perspective. The same concept of weaknesses and vulnerabilities underlies the quality work that has come from the Architecture Driven Modernization Task Force and was recently republished by ISO/IEC as ISO/IEC 5055 but is also the underpinnings of the Systems Assurance Task Force's work in the Software Fault Pattern Meta-model.
This vulnerability and weakness model was created by MITRE in our development of the Common Vulnerabilities and Exposures (CVE) effort which is now captured in ITU-T's X.1520 standard and our creations of the Common Weakness Enumeration (CWE) which is captured in ITU-T's X.1524. I have led MITRE's work in these efforts and their publishing as international standards but also in getting them consistently and compatibly into OMG's work described above and in RAAML.
The second in-depth meeting I participated in was the one in Chicago where the result of formulating additions to RAAML for security analysis extensions to safety and reliability were presented and approved for adding to the specification resulting in Jira issue RAAML11-31 and the change that resolved it.”
Augment RAAML risk analysis with foundational security analysis concepts
Add foundational concepts to support security risk analysis methods by adding a security library and profile along with tieing them into the existing STPA method.
Add "Weakness", "Vulnerability", and "Threat" to a general security concept library to support security risk analysis and assessment methods which builds upon the RAAML common core.
Add "Limitation" to the general concept library to bridge between "Weakness" and "Vulnerability". Add "Threat", "PresentedBy", "Impacts", "Asset", "SecurityActor", and "Valuates" stereotypes to the general security profile.
Revise STPA method to move common elements to the general library/profile for use by other security methods and update STPA to include STPA-Sec.
Many parts of RAAML are broken in to a Library and Profile. The Profile contains the Stereotypes, for example the definition of the <<AND>> stereotype for an AND gate. The Library contains a definition, for example that an AND Block with the <<AND>> stereotype is a type of Gate. In other words, RAAML often has two elements in the model in different namespaces but with the same name.
In the Gaphor tool, we autocode the model in to a Python datamodel. So far while implementing UML and SysML, we didn't have to name each Python class with the full namespace because the element names were always unique. With RAAML this is no longer the case, there are often two elements with the same name.
I think this can also cause confusion for users by duplicating names. For example, if someone is talking about the AND element, are they referring to the the stereotype or the block?
SysML v2 is starting to use two similar names for the definition and the usage, but the definition includes "def" in the name. For example, the definition of a part is a "part def" and the usage is a "part". I would recommend we use something similar for RAAML. the definition of the AND gate could be "AND_Def", and the stereotype could be "AND".
The duplicated names include:
Gate, AND, BasicEvent, Cause, ConditionalEvent, DormantEvent, Early, FMEAItem, FailureMode, HouseEvent, INHIBIT, Late, MAJORITY_VOTE, NOT, OR, SEQ, TopEvent, UnsafeControlAction, XOR, ZeroEvent.
Generator implementation should ensure element name uniqueness in python
Generator implementation should ensure element name uniqueness in python or any other language. Lumping two UML namespaces (Profile and Library) into a single python namespace causes element name uniqueness problems. This can be solved by adding prefixes or postfixes to generated python names or by separating generated content into separate packages/files.
The section focusses on probability as an attribute from FTA elements.
This is arguably the least important attribute from FTA, particularly in system modeling and given the simplistic treatment of probability in this section.
A far more valuable attribute from FTA is the cut set, or minimal cut set at any gate within the FTA hierarchy.
This is a major shortfall that I believes renders the use of these profiles within system modeling very limited, particularly in early system work before the base event probabilities can be known. Even in later lifecycle phases, the simplistic treatment of probabilities that ignores real world interdependence renders the results from such modeling misleading at best.
Cut set calculation is tool vendor responsibility
The goal of the RAAML specification is to provide data structures that are sufficient to support FTA so that cut sets can be calculated. RAAML does not provide algorithms or implementations for cut set calculations. This is to be done by tool vendors implementing the RAAML specification.
The GSN Community standard and metamodel, as of version 2 of GSN, are based on extending OMG’s Structured Assurance Case Metamodel (SACM) version 2. In a sense, all GSN elements are sub-classes of elements defined in SACM so that tools supporting GSN can leverage the SACM metamodel for exchanging GSN content. The material in RAAML that discusses GSN and define RAAML GSN constructs does not reflect this approach and appears at odds to the GSN community standard and its use and leveraging of OMG’s SACM standard.
*RAAML provides a translation to SCSC GSN *
RAAML provides a translation from GSN notation on top of custom metamodel of SACM to SCSC GSN notation on top of UML metamodel plus profile. The RAAML group believes that this translation faithfully represents the GSN core part of SACM. In the future if the SCSC GSN and/or SACM changes then it can be reflected in RAAML.
The Automotive Industry Action Group and the "Verband der deutschen Automobilindustrie" published a FMEA handbook in 2019 where they discourage the use of the Risk Priority Number (RPN) and instead suggest the use of an Action Priority to determine order and importance of identified failure modes and their effects.
Deferred to the next version of RAAML
The support for AIAG & VDA FMEA is not within the scope of RAAML 1.1. However, we will consider incorporating it in the future if there is a significant amount of interest from the industry.
There should be an explanation of what the previousRPNValues attribute means.
previousRPNValues attribute documented in AbstractFMEAItem
Current specification describes previousRPNValues attribute in AbstractFMEAItem. FMEAItem inherits from AbstractFMEAItem and redefines previousRPNValues attribute by specifying its type as Real by keeping the definition unchanged. This is why an additional definition is not necessary and would duplicate an already existing definition.
Define and propose modidfications to RAAML 1.1 to include reliability block diagrams
Define and propose modidfications to RAAML 1.1 to include reliability block diagrams
Minor changes in front matter (Chapters 1 through 7), add a new section on RBDs at the end of Chapter 9 (profile and library) and a new section on RBDs at the end of Chapter 10 (Views). See the attached .docx files for content of these new sections.
A Reliability Block Diagram (RBD) is a graphical representation used in reliability engineering to analyze the reliability and availability of complex systems. It is a method for assessing the performance of systems composed of multiple components, sub-systems, or processes. In an RBD, system elements (components or groups of components defined as an assemblies, subsystems, or other system architectural elements) are depicted as blocks, and these blocks are connected by lines to represent how they are interconnected or dependent on each other. The goal is to evaluate the overall reliability and availability of the entire system by considering the reliability characteristics of each component and their interconnections
An instance of a part can be part of exactly one composite instance, but there could be many options for the composite. For example the same ScenarioStep could be part of many Scenarios. Of course, each instance of the ScenarioStep will occur only in one Scenario instance.
In order to allow this, the multiplicity on the whole side of the composite relationship in figure 9.22 must be 0..1.
In this figure this multiplicity is not shown. This leads to two possible interpretations. According to UML the default is 1, but according to SysML the default is 0..1. Now, the General Concepts Library is a SysML model, hence we could assume that it is 0..1.
However, this default is just a notational thing: "SysML: These multiplicities may be assumed if not shown on a diagram.". In order to make it into the xmi-file the multiplicities must be set explicitely. And in the current xmi-file the multiplicity is not defined. When it is opened by Cameo, it sets the value explicitely to 1. You might consider this a bug in Cameo, but I guess this behavior is within the specification.
So my question is: What was the intention? Should a scenario step only be part of one scenario? Then the multiplicity must be 1 and this should be made clear in the diagrams. Or can it be part of many scenarios? Then the multiplicity must be 0..1 and this should also be shown in the diagrams. And it should be defined in the xmi-file.
The same question needs to be answered for all composite associations in the libraries, e.g. Risk to Effect, Event and HarmPotential. ScenarioStep is just one example for the issue.
Composite relationships should have 0..1 multiplicity at the composite end
Most of composite relationships should have 0..1 multiplicity at the composite ends. This is a larger effort for the working group with almost no impact to end users. Thus the working group will focus on it in the next release.
o Table 9.1 lists «ControllingAction» as “A stereotyped UML dependency” and states in the preceding text that “Situations can be mitigated, detected, and prevented via the ControllingAction. The use of this relationship introduces new safety requirements.” This term does not appear anywhere else.
o The Core Profile does, however, describe «ControllingMeasure» as “A measure taken to address (mitigate severity, reduce probability of occurrence, increase probability of detection) a potential or real adverse situation.” which is similar, but not identical to «ControllingAction»
*Update Controlling Action with Controlling Measure *
Replace controlling action with controlling measure in domain model and corresponding text.
The current version of ISO 26262 is version 2018, the RAAML specification is referencing version 2011.
Update references to ISO 26262:2018
The current version of ISO 26262 is ISO 26262:2018 and the version referenced in RAAML 1.0 is ISO 26262:2011. Need to revise reference to latest version and check for any additional changes with the new version.
We suggest to add a new method for risk management for medical devices according to ISO 14971.
In the context of the medicussy project we have created a profile and a library. These could be used as a starting point. The files can be found in a github-repository: https://github.com/oose/medicussy. There you also find further example models that illustrate the application of the method.
We invite everybody to contribute to the github repository.
Out of scope of RAAML 1.1
ISO 14971 is out of scope of RAAML 1.1. We will consider supporting it in the future if there enough interest from the industry.
Additional files to enhance RBD method proposal are necessary (XMI files and examples).
*Add XMI files and an example for RBDs *
Add XMI files for RBDs (see RBD.zip
).
Add section 2.6 to illustrate RBD library and profile usage with an example (see RAAML 11 examples - omg jira.docx).
Considering its significance, Situation is defined rather ambiguously in Table 5.1: “A situation describes a set of situation occurrences of some type. The system, place, time and state parameters are described by classifiers rather than individual descriptions. A situation occurrence is a system being in a given place at given time and in a given state.” Section 9.1 adds that “The system, place, time and state parameters are described by classifiers rather than individual descriptions.” This prompts a number of questions, which are unanswered by the specification:
• Why is there no formal definition of ‘situation occurrence’ in the Core Profile? (it is only informally defined as part of the definition of ‘situation’) The lack of a formal definition seems to be something of an oversight, given their central importance, especially considering that the stated aim of RAAML to “avoid inconsistent model-based solutions [that] prohibit direct model sharing between organizations and across the various tools”.
o My recommendation would be to add this to the Core Concepts e.g. “Figure 9.1 – Core concepts domain model”, a definition in Section 9.1.x and an entry in the 5.1 definitions table.
• In the absence of the above definition of ‘situation occurrence’, the RAAML user is left to guess how ‘situation occurrences’ should be represented in the model?
o One might initially assume that a ‘situation occurrence’ is intended to be an instances of ‘situation’, but this assumption would appear to contradict the definition of ‘situation’, which states it is a set of ‘situation occurrences’. In other words, every instance of ‘situation’ must be associated with a set corresponding instances of ‘situation occurrence’
o What ‘types’ of ‘situation occurrence’ exist (definition of ‘situation’ states “…situation occurrences of some type”)
o What is the nature of the set of ‘situation occurrences’ that are described by a ‘situation’?
Should this be represented in UML with shared or composite aggregation from whatever represents ‘situation occurrence’ to ‘situation’ – maybe not, because ‘situation’ only “describes a set”, so perhaps this should be a plain association labelled ‘describes’?
If the relationship is aggregation, should/could a ‘situation’ comprise anything else alongside the set (such as a description of the relationships between the members of the set)?
o Regardless of whether ‘situation’ aggregates ‘situation occurrences’ or merely describes them, what range of multiplicities is valid (e.g. can a ‘situation’ be an empty set, have only a single constituent ‘situation occurrence’, or must it be comprise two or more)?
• RAAML deliberately relies on UML classifiers to represent “system, place, time and state”.
o State is a UML type and RAAML shows e.g. «FailureState» extending the «State» metaclass, but which classifiers are intended to be used for ‘system’, ‘time’, and ‘place’?
o I am also left wondering why RAAML relies on classifiers for this purpose, rather than making them explicit in the Core Concepts – no rationale for this decision is provided. This seems almost certain to result in various interpretations in different organizations' tools, thus violating the stated aim "avoid inconsistent model-based solutions"
Situation occurence is defined in the body text
The situation occurence is defined in the body text of the section describing the domain model. It is not a specific defined element and situation occurrences could be modeled using instances although none of the provided methods leverage this capability and it is therefore not specifically defined. If the need for instances is necessary for any future added methods, the definitions will be revisted at that time.
How are risk matrices supported that visually relate risks to the probability and severity of their occurrence as well as to the defined risk acceptance criteria levels (e.g. acceptable/tolerable/intolerable)?
Out of scope of RAAML 1.1
Risk matrices is a useful capability in S&R domain however support for them is out of scope for RAAML 1.1. We will consider adding support for risk matrices in upcoming versions of the specification.
How is issue/task/responsibility tracking supported for an (unsatisfied) ISO 26262 safety case?
The RAAML specification does not cover issue tracking
Issue tracking is out of scope for the RAAML specification. There are tools existing which track issues and would be expected to be used for ISO 26262.
Sometimes the transition from one situation to another situation needs additional properties which cannot be captured by a simple association. The common solution for that is to use an association class to express causality allowing for additional properties to be added to the transition. This pattern is already used in the STPA method. The proposal is to create a general pattern so other methods can access this capability. This would include creating an association class on the common core Situation.
In Enterprise Architect 15.2 the CoreRAAML.xmi is not loaded completely. The generalization relation from Situation to SysML Block is not loaded. It seems to me that the xmi export has tool-specific content.
RAAML claims to implement STPA, but some important elements are missing.
For example,
In Step-1 of the STPA process (in the 2018 Leveson/Thomas STPA Handbook – Figure 2.3) it calls out the need to derive “Safety Constraints” from the Losses. I can find no reference to Safety Constraints in the RAAML.
Furthermore the RAAML does not provide any linkage from Losses to Hazards, which is expected in the Handbook.
An abstract stereotype ‘SupportingInformation’ is used to cover ContextStatement, Assumption and Justification. This is an odd choice since these 3 elements cannot be connected to the argument by the ‘SupportedBy’ relationship. This is therefore confusing and potentially misleading. Perhaps ‘ContextualInformation’ would be a better phrase (the GSN standard refers to Context, Assumption and Justification elements collectively as ‘contextual elements’
Rename SupportingInformation abstract stereotype
Rename the abstract stereotype ‘SupportingInformation’ to prevent confusion or misunderstandings.
Section 3.3 cites www.goalstructuringnotation.info as the source of the GSN specification. In fact this location has not been the formal source of the GSN standard since Version 1 was superseded in January 2018. The formal source for version 2 is https://scsc.uk/scsc-141B. It should also be noted that this will be replaced by version 3 imminently (due to be released on 3rd May 2021, and available at https://scsc.uk/scsc-141C). A new information site for GSN is available at https://scsc.uk/gsn and the goalstructuringnotation.info domain will be remapped shortly to point to this new location
Reference is out of date for SCSC standard
Update reference to the latest version of SCSC standard
ContextStatement has been used as an element name which is an odd choice given that the GSN standard uses <context statement> as the term for the <element statement> of a GSN context element. ‘Context’ would be more consistent with the selection of terms for other GSN elements
*Rename ContextStatement abstract stereotype *
Rename the abstract stereotype ‘ContextStatement’ to avoid confusion and misunderstandings.
At the moment it is not possible to connect the transfer in to a gate with a connector properly-typed-by-association, because
TransferIn type is FTATree, not the Event
and gate source association currently connects Gate and Event;
To address this, Gate.sourceEvent type needs to be widened to FTAElement (which allows both Event and FTATree)
Gate.sourceEvent type needs to be widened to FTAElement
At the moment it is not possible to connect the transfer in to a gate with a connector properly-typed-by-association, because
TransferIn type is FTATree, not the Event
and gate source association currently connects Gate and Event;
There is a user need to model FTA's in the IBD view directly, which may be covered by tool vendor implementation, but in addition there is the need to take elements from other analyses and tie them to elements in the FTA directly. RAAML allows this but the icons are lost from the initial FTA element when they are typed by other elements (for example: Failure Mode). This has been demonstrated in the FTA and FMEA cross analysis use case where FTA IBD elements are typed by FMEA elements.
Add a Property meta class to FTA Event stereotype
To be able to create fault trees that are using situation from other methodologies as events like FMEA failure modes.
Therefore, constraints in the fault trees need to be updated.
Comment from Gaphor UML/SysML implementation reported by Dan Yeaw. Editorial comment: The figure name does not have a space after ISO. The following "Figure 10.16 - ISO26262 Profile" should be changed to "Figure 10.16 - ISO 26262 Profile".
*The Figure 10.16 name is missing a space after "ISO" *
The figure name does not have a space after ISO. The following "Figure 10.16 - ISO26262 Profile" should be changed to "Figure 10.16 - ISO 26262 Profile".
Comment from Gaphor UML/SysML implementation reported by Dan Yeaw. Editorial comment: Misspelled encompassing in figure title "Figure 10.14 - All-Encompasing Operational Situations"
Fix spelling for Figure 10.14
Editorial comment: Misspelled encompassing in figure title "Figure 10.14 - All-Encompasing Operational Situations"
Comment received from Gaphor UML/SysML implementation reported by Dan Yeaw. Figure 10.4 - General Concepts Profile. Editorial comment on note that the note should say "From RAAML Core" instead of "From S&R Core".
Figure 10.4 Note text update
Figure 10.4 - General Concepts Profile. Editorial comment on note that the note should say "From RAAML Core" instead of "From S&R Core".
Comment received from Gaphor UML/SysML implementation reported by Dan Yeaw. Editorial comment. The note should read "From RAAML core" instead of "From S&R core".
FTA Profile Note text update
The note in figure 10.9 should read "From RAAML core" instead of "From S&R core".
AbstractEvent uses the attribute 'probability' as a placeholder. The term 'probability' will intuitively be associated with a number between 0 and 1. ISO/IEC guide 73 uses the term 'likelihood' with the following note:
"NOTE 2 The English term “likelihood” does not have a direct equivalent in some languages; instead, the equivalent of the term “probability” is often used. However, in English, “probability” is often narrowly interpreted as a mathematical term. Therefore, in risk management terminology, “likelihood” is used with the intent that it should have the same broad interpretation as the term “probability” has in many languages other than English"
Given the abstract definition of AbstractEvent, and the attribute statement that method developers can derive more specialized ways to characterize probability, it may be helpful to use the subtlety of the difference in terms here to allow more clarity between the concept (likelihood) and its utilization (probability)
Rename to likelihood
The working group agrees to rename the probability property to "likelihood". In addition to that, the probability property in FTAElement shall redefine the likelihood property and AbstractOperationalSituation shall inherit from AbstractEvent and also redefine the likelihood property.
The following Process Model situations have spaces in their names:
Inadequate Controller Decisions
Inadequate Control Execution
Inadequate Process Behavior
Inadequate Feedback and Inputs
They should be in upper camel case / Pascal case for consistency with the other elements in the specification.
Spaces in classifiers for process models
Unnecessary spaces in the classifier names.
The current specification only provides an ID property for GSN elements. Hazard Analysis methods typically use the element names or unique identifiers and not having an explicit property in the specification will lead to inconsistencies across tools. The proposal is to add an ID property to the core <<Situation>> which would be inherited by all the RAAML method elements.
Add ID carrier to Core Profile
There is not an ID property that is consistent across the specification for elements. This leaves the implementer to adding an ID which will be inconsistent from implementation to implementation.
UnsafeControlAction should inherit from Situation Stereotype. It is currently inheriting from the FailureMode stereotype.
Inherit from Situation Stereotype
Inherit UnsafetControlAction from Situation Stereotype
For some reasons, the Library of the FMEA Profile is not working. It's not possible to open it with and XMI-Viewer, with Magic Draw or any other tool.
You can prove it with opening the file in the explorer.
I try to implement the RAAML-Profiles in Papyrus, and it would be great if I can use this file, too.
Best Regards
*Fix syntactic errors on .xmi file *
RAAML file in question (the ad/20-07-12, FMEALib.xmi) has two simple XML problems
at line #107 and line #150
namely – missing the closing slash in the <type> element
The FTA Majority Vote Gate does not calculate as expected which was identified from usage of the imported FTA .xmi.
Error in .xmi calculation for Majority Vote
The Majority Vote calculation in the FTA.xmi is not calculating correctly.
It looks like a copy & paste error me.
The Javascript used for the inihibit gate constraint block is the same as the Jvascript used for the AND constraint block.
The control value input is not even used in the Javascript which triggered me to look more closely
Question on Inhibit constraint calculation
Question if the Inhibit Gate probability calculation should be similar to the And Gate. According to the underlying FTA standard the probability is calculated the same for the Inhibit Gate and the And Gate with some differences on the inputs. The calculation was checked and performs as expected.
The spec has an inconsistency in it - the text says there is one (a) cause - but the graphic shows 1..*. Text:
"An AbstractFMEAItem is a scenario (more specifically - AbstractRisk scenario) composed of a failure mode, a cause and
(potentially multiple) effect(s). It stores assessed and mitigated risk priority numbers."
Figure: 9.32
Update text description to match plural causes in FMEA library diagram
Update singular "cause" to "causes" in FMEA Library Item text 9.3.1.
Under ‘GSNNode’ it is stated that the ‘short phrase’ (referred to in the GSN standard as the <element statement>) shall be captured as the UML model element name, whilst the Human-readable identifier (referred to in the GSN standard as the <element identifier>) shall be stored as a separate tag. This appears an odd choice since the <element identifier> is assured to be unique, whilst the <element statement> is not. It may be appropriate to swap these allocations.
GSN Statement Text and Identifier
The uniqueness or non-uniqueness of the field was not a primary reason for the mapping choice.
The reason for mapping the GSN statement text to a UML element name is to make it easier for the end user to see/find/manipulate the element in the tool UI. UML tools primarily expose the model element names everywhere in the GUI (element visualization in the diagrams, various tables, containment trees),
while tag information is secondary.
The GSN standard is made available under creative commons licence version 4. This allows free use but requires appropriate attribution, which could be accommodated in the reference in section 3.3, or in section 4.
“GSN Standard V2 License: This work is licensed under the Creative Commons Attribution 4.0 International License.
To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ or send a letter to Creative Commons, PO Box 1866, Mountain View, CA 94042, USA.
You are free to share the material in any form and adapt the material for any purpose providing you attribute the material to the SCSC ACWG, include the license details above and indicate if any changes were made. See the license for full details.”
Reference creative commons license
Add creative commons acknowledgement for GSN standard.
It is noted that the GSN profile only addresses the core GSN notation (e.g. it does not address patter or modular extensions). A note to this effect in 9.6.1 would be worthwhile for clarity
Initial support for core elements comment
Add note to the GSN section’s introduction that the GSN UML profile only covers the GSN core notation in this release/edition. Also add that same note to the Related Documents section.
In figure 10.12 there is a note against ‘solution’ that states a solution could also be some artefact of the system. The GSN standard is clear that a solution presents a reference to evidence. It is not understood how this can be an artefact of the system. Similarly a statement in the note attached to ‘contexturalStatement’ states that context can be either context statement or some artifact of the system; this distorts the definition in the GSN standard which states that the <context statement> can be either an a reference to contextual information, or a statement. This implies that there is not a direct relationship between the modelling in RAAML and the GSN standard.
*Remove note from figure 10.12 solution *
We should remove the ability to directly substitute a GSN solution with a SysML element because the result is not a valid GSN model. However, what we should allow is for GSN solution to be able to reference a SysML element. We need to decide whether that is by a tag reference property, or by a «trace» relationship, or a dedicated relationship.
A relationship of GSN to the SACM metamodel has been developed and is publicly available. It is not clear why the GSN UML profile is required separately to this metamodel, what relationship or utility is intended
Merge with RAAML-49
Same solution as RAAML-49. "Add a statement to the introduction of the GSN UML profile section to explain that whilst GSN is an extension of the OMG SACM 2.1 standard, which has a defined meta-model based on the OMG MOF standard, the objectives of RAAML to integrate with SysML 1.5 & 1.6 necessitate the use of a UML profile interpretation of the GSN standard. In doing this there are certain design choices born out of the constraints of conforming to UML that may be considered imperfect from a pure GSN implementation point of view."
While a UML profile can provide a mechanism for establishing the primitives and components for capturing the safety and reliability aspects of a system within a MBSE tool, the real goal would appear to be the exchange of MBSE models that include safety and reliability aspects. A metamodel would seem to be a more appropriate mechanism for capturing/conveying the details of what is to be exchanged. Any UML profile necessary could be derived from that metamodel when needed. How does RAAML support the exchange of safety and reliability aspects of a system?
UML XMI used as an exchange format
RAAML WG took a lightweight approach to base RAAML on UML+profiles+libraries to promote exchange of S&R information between tools. This approach has two advantages over having separate metamodel approach. First, any UML-compliant tool (including SysML tools) can load and store these models. A separate metamodel would necessitate a special tool that would need to know how to load and store these models. Second advantage is that connectivity between safety and reliability information and the systems model is easy to achieve for tool vendors that already support UML.
What is the envisioned use and utility of the RAAML UML libraries with respect to exchanging safety and reliability aspects of a system from one MBSE modelling tool to another?
RAAML is extending SysML to support interchange via XMI
RAAML is extending SysML thus exchange between different MBSE tools is expected to happen via XMI.
Section "2. Conformance" is more explicit about that:
"Type 1 Conformance: RAAML model interchange conformance. A tool demonstrating model interchange conformance can import and export conformant XMI for all valid RAAML models."
Currently the LossScenario is only defined in the STPA library and does not include a defining stereotype. The LossScenario is similar in purpose to the FMEAItem where it is typically the focus of a line item in an analysis table. The proposal is to add a LossScenario stereotype to the STPA Profile to be consistent with the FMEA and give a handle for table generation.
Introduce LossScenario Stereotype in STPA Profile
Add stereotype to be consistent with which stereotypes are defined for other methods. This is similar to FMEAItem Stereotype
Comment received from Gaphor UML/SysML implementation reported by Dan Yeaw. Editorial comment:
The figure name "Figure 10.13 - ISO26262 Library" should have a space after ISO as the following "Figure 10.13 - ISO 26262 Library".
*Add Space in Figure 10.13 name after "ISO" *
The figure name "Figure 10.13 - ISO26262 Library" should have a space after ISO. Change the Figure name to include a space after ISO to "Figure 10.13 - ISO 26262 Library".
The representation of Figure 9.59 could be taken to imply the conditioning event connects to the bottom of the gate. By convention the conditioning event connects to the side connector with the bottom connection used for the enabling condition.
This confusion is also conveyed in Figure 10.8 (page 111)
Connect the condition association to the right of the INHIBIT gate
Connect the condition association to the right of the INHIBIT gate for clarity (figure 9.59 and 10.8).
'Undeveloped' seems to be defined for the exclusive use with the GSN profile, so it seems more appropriate to define it in 9.6.1.
There are also some typographical corrections that should be considered ...express the fact that the goal or strategy is not fully developer... should be replaced with ...express the fact that support for the goal or strategy is not fully developed...
Change wording to be relevant in FTA and GSN context
Wording needs to be changed to be relevant in FTA and GSN context.
Updated model, Updated stereotype description text, Update .xmi
The failure modes covered by this section are exclusively Functional failure modes, they are one of a number of taxonomies of functional failure and would not be suitable for component level FMEA, or FMEA integration to higher level items.
This should be acknowledged in the introduction to this section.
RAAML is not restricted to functional or component FMEAs only
The RAAML working group will elaborate in the specification text that RAAML supports both functional and component FMEA. The target of the RelevantTo relationship can be any element, the source needs to be connected to a situation.
Use of GSN without reference to SACM
Issue Description: We believe that it would be appropriate for an OMG standard to be referencing the definitive work on assurance modeling language published by OMG, and therefore suggest that SACM should be included.
Reference to SACM
Add text to explain use of GSN UML profile in RAAML is based on extension of SACM.
In Figures 9.104 and 9.105 the RAAML specification makes suggestions on the concrete syntax (in this case the graphical notation) of GSN elements in which the metamodels (sets of abstract syntax) are defined. Defining concrete syntax for GSN elements would seem to be out of the scope, where SACM and GSN itself should be leveraged for these definitions rather than defining them in RAAML.
Revise the text so that it is clear that SCSC/GSN defines the graphical notation.
The acknowledged simplistic treatment of probability can easily be replaced with a more correct definition that addresses independence [P(a.b) = (P(a) - P(cc)) x (P(b) - P(cc)) + P(cc)] where cc represents the common/non-independent failure & [P(a+b) = P(a) + P(b) - P(a.b)]
Any simplification should be taken by the modeler not by the profile.
Current specification supports common cause probabilities
A common cause event can be addressed by combining the independent probabilities of the events and adding a OR gate to include the common cause component.
The RAAML spec allows for changing the math for any gate.
How are risk matrices supported that visually relate risks to the probability and severity of their occurrence as well as to the defined risk acceptance criteria levels (e.g. acceptable/tolerable/intolerable)?
Adding support for risk matrices
Risk matrices are important in RAAML context. However, the focus of RAAML 1.0 is to provide the foundation for modeling safety and reliability aspects of systems. We will be adding support for risk matrices in the upcoming versions of RAAML.
Hi folks, I am part of the Society of Automotive Engineers STPA workgroup with Kyle Post and others. I did reach out to Kyle and he didn't have a solution for my issue, thought I'd ask the group. I am using Rhapsody v8.4 developing SysML models (using their SysML profile) and trying to import these RAAML profiles, but the Rhapsody xmi importer doesn't get past the first step, saying 'Cannot determine uml version'. Wondered if this is an issue others have run into and whether there might be a simple solution, Thanks, Jerry
Current .xmi's are correct and standard compliant
Was an issue with using an older tool version.
Data that the xmi is correct:
-Correct from MagicDraw's test
-Correct from Enterprise Architect's test
-Correct by comparison to other OMG normative documents:
For example, UML 2.5.1 (e.g. StandardProfile.xmi)
namespaces, used for UML models are:
xmlns:xmi="http://www.omg.org/spec/XMI/20131001"
xmlns:uml="http://www.omg.org/spec/UML/20161101"
and those are exactly the namespaces that we used across the *.xmi documents in RAAML
Additionally, the Rhapsody version used was an older version and if the UML version is manually changed to an older version it imported into Rhapsody.
How is issue/task/responsibility tracking supported for an (unsatisfied) ISO 26262 safety case?
Out of scope for current version of the RAAML specification
The current version of the RAAML specification does not specify a way to link to issues in issue tracking systems. That may either be described in another specification or may be a focus area in upcoming versions of the RAAML specification.
Comment received from Gaphor UML/SysML implementation reported by Dan Yeaw. Figure 10.3 - General Concepts Library. Editorial comment. There is a lot on the diagram, why was everything merged?
Concerns on general concepts library image
It is an overview on all concepts included in the general library and providing a summary on releations among concepts. Without this overview you don't get the big picture.
Issue reported from Gaphor UML/SysML tool implementation notes by Dan Yeaw. Editorial comment is on Figure 9.122 referring to showing the AnySituation line on the document asking why the line is repeated here.
Why repeat AnySituation
It is a visualization of the text above the image. The OperationalSituiation is inherited from AnySituation.
The section focusses on probability as an attribute from FTA elements.
This is arguably the least important attribute from FTA, particularly in system modeling and given the simplistic treatment of probability in this section.
A far more valuable attribute from FTA is the cut set, or minimal cut set at any gate within the FTA hierarchy.
This is a major shortfall that I believes renders the use of these profiles within system modeling very limited, particularly in early system work before the base event probabilities can be known. Even in later lifecycle phases, the simplistic treatment of probabilities that ignores real world interdependence renders the results from such modeling misleading at best.
RAAML provides sufficient information to calculate cut sets
Operations such as calculation of cut sets are meant to be implemented by tool vendors. The RAAML specification needs to provide enough data for these operations to be able to return results. Current version of the RAAML specification provides enough data for that. That was demonstrated in the pilot implementation by export to the R fault tree package contained in CRAN R repository by means of the generic tables provided in Cameo Systems Modeler.
ErrorRealization is used to define the relationship between Error and Failure, however no such relationship between fault and error or failure appear to be described, leaving the concept unclear.
The Activation relationship connects error and fault
The Activation relationship is defined in that same diagram and it connects to AbstractCause and DysfunctionalEvent which play the fault and error roles respectively in that association.
AbstractEvent is introduced as a derivative of AnySituation. 9.1 defines the situation concept based on state. It is not obvious how an event can be a state when interpreted from the perspective of the fault modeling techniques in scope of this document.
Current description is sufficient
The word "state" as used in the definition of Situation can apply not only to the system under analysis but also to its environment. The states of the environment can be used as events triggering changes in system states.
Many parts of RAAML are broken in to a Library and Profile. The Profile contains the Stereotypes, for example the definition of the <<AND>> stereotype for an AND gate. The Library contains a definition, for example that an AND Block with the <<AND>> stereotype is a type of Gate. In other words, RAAML often has two elements in the model in different namespaces but with the same name.
In the Gaphor tool, we autocode the model in to a Python datamodel. So far while implementing UML and SysML, we didn't have to name each Python class with the full namespace because the element names were always unique. With RAAML this is no longer the case, there are often two elements with the same name.
I think this can also cause confusion for users by duplicating names. For example, if someone is talking about the AND element, are they referring to the the stereotype or the block?
SysML v2 is starting to use two similar names for the definition and the usage, but the definition includes "def" in the name. For example, the definition of a part is a "part def" and the usage is a "part". I would recommend we use something similar for RAAML. the definition of the AND gate could be "AND_Def", and the stereotype could be "AND".
The duplicated names include:
Gate, AND, BasicEvent, Cause, ConditionalEvent, DormantEvent, Early, FMEAItem, FailureMode, HouseEvent, INHIBIT, Late, MAJORITY_VOTE, NOT, OR, SEQ, TopEvent, UnsafeControlAction, XOR, ZeroEvent.
Deficiency in Gaphor tool
The AND Stereotype in the Profile and the AND block in the Library are distinguishable because of id, type and namespace are different.
The GSN Community standard and metamodel, as of version 2 of GSN, are based on extending OMG’s Structured Assurance Case Metamodel (SACM) version 2. In a sense, all GSN elements are sub-classes of elements defined in SACM so that tools supporting GSN can leverage the SACM metamodel for exchanging GSN content. The material in RAAML that discusses GSN and define RAAML GSN constructs does not reflect this approach and appears at odds to the GSN community standard and its use and leveraging of OMG’s SACM standard.
SACM mapping RAAML GSN
RAAML provides a translation from GSN notation on top of custom metamodel of SACM to SCSC GSN notation on top of UML metamodel plus profile. The RAAML group believes that this translation faithfully represents the GSN core part of SACM. In the future if the SCSC GSN and/or SACM changes then it can be reflected in RAAML.
In Figure 9.127 of the ISO-26262 part of RAAML the concept of ASIL was misread from the ISO-26262 2018 standard. In ISO-26262 Clause 1, it states that QM is not an ASIL. Therefore, it suggests that Figure 9.127 should be reviewed/revised to align with Clause 1 of ISO-26262.
Consistent with the use in the industry and ISO 26262 standard
As per Table 4 - ASIL determination in the ISO 26262 chapter 3, QM is classified under ASIL. RAAML is consistent with what is specified in ISO 26262 specification even if QM does not have ASIL requirements.
ASIL decomposition levels should not be included in the ASIL stereotype. Maybe ASIL decomposition belongs to a sub-type of ASIL in RAAML.
ASIL decomposition levels
The ASIL decomposition levels are included as they are commonly used when specifying the safety integrity levels for requirements and elements. This is referred to as the ASIL in industry and breaking it into a separate or subtype does not seem to have a specific use case for the added complexity.
Does RAAML include the definition for concrete syntax?
RAAML of concrete syntax
RAAML uses UML and SysML for the concrete syntax. The reference to UML is made on page 3 and page 5 of the proposal.
There are 5 instances in the informative description that says the Functional Safety can be represented in a SysML IBD, but the diagrams are SysML BDDs.
I recommend changing the text to SysML BDD.
Update RAAML Informative Example document Editorial Errors
Example text refers to IBD where the diagram is a BDD in 5 places in the document. Need to change text from IBD to BDD.
