Panel on SysML v2 (Part 1): Without conformance, users will suffer the most

Last week, I moderated a panel with four well-known experts from the SysML world: Andreas Pollom, Daniel Siegl, Robert Karban and Vince Molnár explained why SysML conformance is important, how we can achieve it, what we get out of it and who suffers the most if we don't ensure conformance quickly. Below is a summary in German. The article also contains the one-hour English-language recording of the discussion.

The idea for this topic was born at TdSE 2023, where Daniel Siegl and I in particular had already worked intensively on the topic. Thanks to Rüdiger Maier from LieberLieber, who didn't let up, we were able to hold this panel online on February 15, 2024.

In this panel, we assumed that the audience is already familiar with SysML v1 and knows what will change in SysML v2. Here at SE-Trends we already have a lot of material on SysML.

Become an activist!

First of all: we have organized this panel because SysML v2 conformance is currently in acute danger. If we do not find a solution to this problem in the near future, this could be a major setback for SysML in particular, but also for MBSE in general. The victims will be the users who expect their work results to be interoperable.

The victims of a lack of SysML v2 conformity will be the users!

Daniel Siegl (00:49:03)

Hence the panelists' appeal to readers (if you would like to become active, you can me or contact the linked panelists via LinkedIn):

• Tool manufacturer: Conformity will save you time and money. Therefore, get involved with money, or better with personnel, for example in setting up a test suite.
• User: Hope for the best, but prepare for the worst! In particular: Standardize your modeling today, avoid customization and profiling as much as possible.
• User: Ask your tool manufacturers explicitly about conformity. Make it clear that the topic is very important to you.
• Users in large corporations: Conformity can even become a problem within the same organization. Corporations such as ESA, Mercedes, thyssenkrupp can benefit from compliance through internal interoperability alone. Therefore, there is a business case for such companies and we are happy to help you develop it.
• Students and teachers: There is a lot of potential here for interesting master's or doctoral theses.
• All readers: Even small amounts of money (for companies) help to finance the work on conformity. Is your company in a position to make a contribution? Active cooperation from competent people often helps even more than money.

The panelists

Andreas Pollom joined the company in 2015 as a student assistant at Chair of Virtual Product Development (VPE) at RPTU for the first time with SysML. Since 2019, he has been working on MBSE in conjunction with variant management at the Fraunhofer Institute for Experimental Software Engineering IESE. He has gained experience in many industries and has developed his own approach to integrating system engineering.

Robert Karban has been working in the field of MBSE for a long time and with SysML since 2007. Much of his work at that time was incorporated into the SysML v1 standard, for example proxy ports. He was also part of the SysML Tool Submission Team. He is co-founder of OpenMBEE and worked in the past for the NASA Jet Propulsion Laboratory and appears independently in this panel.

Vince Molnár is assistant professor at the Budapest University of Technology. He has been giving training courses on MBSE since 2015 and his research areas also include formal methods, in particular formal verification. This is what makes SysML v2 so interesting for him, as he knows first-hand the many challenges of SysML v1 in this regard. He has worked directly with the SysML v2 Submission Team and the Execution Working Group, becoming a co-author of the standard.

Daniel Siegl is responsible for business development for DearLovera small company that develops tools to support MBSE. He also co-chaired a SysML group working to establish a standard for model exchange based on SysML v1 for the automotive industry. He therefore has first-hand knowledge of the necessity and challenges regarding conformity. He also understands the importance of conformance from the perspective of a small tool manufacturer. He is also Deputy Director of Standards at INCOSE.

SysML v2 Conformance: View now

The discussion lasted around an hour and is available in full here. As always with SE-Trends, a summary of the most important findings can be found below.

What is SysML v2 conformance? (6:37)

Vince gave an overview of the origins of SysML v2 conformance, which was already called for in the SysML v2 RFP with a corresponding requirement. Therefore, Standard had to define conformance criteria from the beginning, which fall into five areas:

• Abstract syntax
• Concrete syntax
• Semantic conformance
• Modeling Exchange
• Domain Library

Vince's team has collected further criteria that go beyond these five areas.

SysML v1 has not managed to realize conformity in interoperability. Test suites would help to realize this for SysML v2 right from the start.

Robert Karban

Robert supplemented this with the perspective of a user who has already gained 20 years of experience with SysML v1 conformance. One of the biggest problems with SysML v1 was interoperability, so this is where he would hope to see the most benefit. Robert would like to see test suites, as the OMG sometimes does with other standards.

From the user's point of view, he sees several dimensions. In particular, he pointed out that users generally do not make full use of the standard and that partitioning therefore makes sense. Tool manufacturers also do not always implement the full standard (especially small manufacturers). The conformity of the specific subset of the standard used is therefore relevant for the user. For small tool manufacturers in particular, it would be helpful if there were different subsets of the standard based on the usage scenarios. For example, variability is an aspect that not all users necessarily need. Robert made a few specific suggestions:

• Behavior and structure in textual and graphical notation
• Variability
• API

Robert also pointed out that we can forget about the interoperability of the diagrams in SysML v1 anyway. This is a completely different issue that we will certainly not be able to solve with SysML v1.

Some vendors claim to have interoperability just because SysML v2 syntax highlighting works. This is not yet SysML v2 conformance, this is only a part regarding the textual representation!

Daniel Siegl

Daniel Siegl gave the specific example that some manufacturers already describe a tool with syntax highlighting as "compliant". Of course, this is not conformance to the SysML v2 standard. This specific example shows how important it is to have several conformance levels.

Andreas underlined the urgency of interoperability by citing teams that were not even able to exchange models with each other within their own organization. Of course, this was not yet about SysML v2, which Daniel confirmed. He pointed out that profiling can also destroy interoperability between the same tools.

Methodology or language-What is the problem?

At this point, the question arose as to whether interoperability is caused by SysML itself or by the methodology used. Andreas saw a 50:50 split here. There is also the fact that many tools do not require a formally correct SysML model. This can lead to teams, consciously or unconsciously, working with non-standard models.

This also means that a clear demarcation between language and methodology is necessary to ensure conformity.

Next week: How do we achieve conformity and what happens if we don't?

Now that we've heard from the panelists what we mean by SysML v2 conformance, next week we'll get down to the real question: What does SysML v2 conformance (or lack thereof) mean for users? How have other standards created conformance? And what can we learn from the standards that have conformance issues?

If you want to find out now, you can watch the recording of the panel today. Everyone else: The article next week do not miss!