Category: Trends

MBSE is too important not to have a podcast

It’s September 2020 – a lazy late summer evening: Tim Weilkiens and I are holding a virtual meeting to discuss what we would like to present in our joint contribution to Systems Engineering 2020. It will boil down to an example from the current development status of SysMLv2 – something new – a live demo would be great – hardly anyone in Germany knows this yet. But why? You should actually start a podcast!

That’ s how you can summarize the birth of our podcast idea “The MBSE Podcast – Trust us we are Systems Engineers”

Why choose MBSE?

But what is it all about? Tim, author and board member of oose e.G. and I, MBSE & PLM consultant and team leader at CONTACT Software, are both passionate about systems engineering. With this podcast we set out on a mission to spread the word about MBSE (Model-based Systems Engineering). MBSE has gained momentum in Germany over the last ten years. This means that it has arrived in many industries or at least the awareness that systems engineering is a key competence for mastering the increasing complexity of interdisciplinary product systems.

However, finding a solution to a complexity problem is certainly not easy in itself. This applies very well to the topic MBSE. MBSE is very multifaceted and not very easy for newcomers to grasp: Methods, processes, languages, roles, architectures, frameworks, tools and training – to name just a few.

MBSE from A to Z – for beginners and advanced

This is where we start to explore the topic piece by piece in a relaxed atmosphere. There are plenty of books, training courses and tools available. But in our podcast we would also like to take a look behind the scenes and pass on our knowledge between the lines of specifications. We benefit from our private and professional engagement with the Gesellschaft für Systems Engineering (GfSE), the International Council on Systems Engineering (INCOSE) and the Object Management Group (OMG). Thus, we are very close to current developments in the (MB)SE field and are in close exchange with national and international key players from application, consulting, research and tools.

The start of the continuous podcast

On October 30, 2020 the podcast started with a teaser on YouTube. The first content episode entitled “The History of SysML” followed on November 5, 2020. We plan to publish one or two episodes per month.

Why on YouTube?
The platform offers extensive infrastructure for us and very easy access for viewers.

Why live?
Tim and I run the podcast in our spare time. We simply do not have the time to cut episodes for evenings. The live stream and the recording available afterwards are as they are: 100 percent authentic. In the future, we plan to include interaction with the listeners via live chat.

For those who prefer to listen to a podcast, we publish in addition to YouTube the audio track on Spotify and Apple Podcasts.

Risks and safety – (just) an issue in times of crisis?

Germany and almost all other countries are facing a crisis! Within a very short time we had to realize that an infectious organic structure with a size of a few nanometers is causing gigantic damage worldwide and is turning our lives upside down. We suddenly realize how vulnerable our existence is and how our own need for security always comes to the fore when everything around us can no longer be taken for granted.

History teaches us that it doesn’t matter what the name of a disaster is – COVID-19, 9/11, Fukushima or Tsunami, safety and risks have to be reconsidered and reassessed in each case. Because too much security makes any system slow and cumbersome. By contrast, too many risks, however, entail a high degree of danger – not only for the individual, but also for society as a whole.

The consideration and evaluation of product risks is a familiar topic in the PLM world. Risk management is often perceived as an annoying accessory, which essentially causes a lot of paper to be filled. However, the concrete added value is not always immediately apparent. This is where the change of perspective from producer to consumer helps: Countless test and quality seals in every area of daily life ensure that products are safe to use, for example the TÜV for the car, the CE mark on the notebook or the GS seal on the office chair. Nobody wants to do without the fact that the product he uses is sufficiently safe and that he can use it without major risks.

Three fundamental insights have become apparent for my consulting work:

  1. Think the unthinkable
    Science and technology make us feel more and more that nothing can happen to us. But this certainty can be deceptive. Sometimes we only feel safe because we are not (yet) aware of risks or because we misjudge them.
  2. You cannot think as stupidly as it may come
    It will never be possible to anticipate every risk and to be prepared for all eventualities or even be insured against them. Nevertheless, risk management trains the eye for possible dangers and effective countermeasures. Regularly thinking about risks and sounding out parameters for security in times of no crisis trains thought structures and action patterns. These consequently form a solid basis for sound crisis management.
  3. Necessity is the mother of invention
    Unknown situations paralyze one’s own thinking and acting at first, but as a consequence they stimulate creativity in particular. There is enormous potential here for innovative and efficient ideas on how to achieve great things with existing onboard resources. So every crisis also offers the chance for change and improvement.

Corona ensures that the golden mean between safety and risk in all areas of life can once again be balanced under the given parameters. Perhaps the changed perspectives will even make it possible to take a more positive view of risk management in PLM and give free rein to one’s own creativity both in assessing risks and developing safe products?

AI – Where we are in the Hype Cycle and how it continues

While the artificial intelligence index shows that the increase of research articles and conferences in the field of AI continues, the media is slowly showing some fatigue in the face of the hype. So it’s time to take stock: What has been achieved? What is practically possible? And what is the way forward?

What has been achieved?

In the years 2018 and 2019 the previously developed methods for the application of neural networks (this is how I define AI here) were further refined and perfected. Whereas the focus was initially (2012-2016, Imagenet competition) on methods for image classification and processing and then on audio methods (2015-2017, launch of Alexa and other language assistants), major advances in text processing and generation were made in 2019 (NLP = natural language processing). Overall, the available technologies have been further improved and combined with a great deal of effort, especially from the major players (Google, Facebook, OpenAI, Microsoft).

What is practically possible?

The use of AI is still essentially limited to four areas of application:

  • Images: image recognition and segmentation
  • Audio: Conversion from speech to text and vice versa
  • NLP: word processing and generation
  • Labeled Data: Prediction of the label (e.g. price) from a set of features

This list is surprisingly short, measured by the attention AI receives in the media. The most impressive successes of AI, however, result from a combination of techniques such as speech assistants using a combination of audio, NLP and labeled data to convert the input into text, recognition of text intention with NLP and prediction of the speaker’s wish by using huge amounts of labeled data, meaning previous evaluations of similar utterances.

Decisive for the development of precisely these AI application fields were

  1. the existence of large quantities of freely available benchmark data sets (data sets for machine learning) on which algorithms have been developed and compared
  2. a large community of researchers who have jointly agreed on the benchmark data sets and compared their algorithms in public competitions (GLUE, Benchmarks AI, Machine Translation, etc.)
  3. a free availability of the developed models, which serve as a starting point for the practical application (exemplary Tensorflow Hub)

Based on these prerequisites one can quickly assess how realistic some marketing fantasies are. For example, there are neither benchmark data sets nor a community of researchers for the often strikingly presented field of application of predictive maintenance, and accordingly there are no models.

What’s next?

On the one hand, it is foreseeable that the further development in the AI area will certainly continue initially in the above-mentioned fields of application and continue to develop in the peripheral areas. On the other hand, areas are emerging which, similar to the above-mentioned fields of application, will be driven forward at the expense of large public and private funds (e.g. OpenAI and Deepmind are being subsidised by Elon Musk and Google with billions of euros respectively). An example of large investments in this area is certainly autonomous driving, but also the area of IoT. In total, I see the following areas developing strongly in 2020-2022:

  • The combination of reinforcement learning with AI areas for faster learning of models
  • A further strengthening in the area of autonomous driving resulting from the application and combination of AI and reinforcement learning
  • Breakthroughs in the generalization of the knowledge gained from image processing to 3D (Geometric Deep Learning and Graph Networks)
  • A fusion of traditional methods from statistics with neural networks
  • IoT time series (see below)

I see a big change coming with the rise of IoT and the associated sensor technology and data. By their very nature, IoT data are time series that must be filtered, combined, smoothed and enriched for evaluation. Relatively little specific has been done to date for this purpose. It could be that from 2020 – 2022, this topic could hold some surprising twists and breakthroughs for us. German industry in particular, which has benefited rather little from the initial developments in the field of AI, should find a promising area of application here.