Information technology is a powerful ally in the quest to create and manage the connected objects of tomorrow. The Flexy project—a road/rail transport system now under development at SNCF—is a perfect example of these emerging mobilities, because digital support is critical to both developing and operating it.

How can a shuttle move smoothly between rail and road? How should it react if a car approaches just as it’s about to leave the tracks? What kind of road-rail transition platform should it have? Solving these problems requires very advanced simulations—and collaboration between experts in many different fields. We need automation and mechanics expertise for contact between train wheels and track. Pneumatics for contact with the road. Plus all of the fields that study behaviour and the human factors involved in introducing these new mobilities.

To manage the complexity of these emerging mobilities and the new technologies of tomorrow’s rail system—such as self-driving trains and the TELLi innovative light train—we need new engineering methods.

“At SNCF, complex systems engineering means 3 things: a shared language, a shared method, and shared modelling tools. This is called model-based systems engineering,” says Marc Sango, a rail system physics researcher in our Technology, Innovation and Group Projects Division.

“We take a collaborative, interdisciplinary approach that brings together rolling stock, network and operations specialists from SNCF Voyageurs and SNCF Réseau. All of them are members of a complex systems engineering group within Synapses, our in-house network of experts,” says Marc. “Complex systems engineering helps us manage the butterfly effect from changing regulations. Our experts can quickly identify the consequences—including the financial impact—of even a tiny shift at the other end of the chain,” adds Édouard Chabanier, a Synapses member and head of standardization and regulation at SNCF Réseau’s Industrial and Engineering Division.

A shared language lets SNCF talk about modelling with partners in train manufacturing and other industries. It also helps minimize risk and improve product quality. At SNCF, for example, we use the Systems Modelling Language (SysML) standard, and particularly ARCML, a form of SysML. Both grew out of ARCADIA (ARChitecture Analysis and Design Integrated Approach), a shared method used successfully in avionics, the automotive industry and other competitive transport sectors.

Take TELLi, Draisy and Flexy, 3 innovative light train projects. Because we have a shared language, we and our partners can work together to model a train or shuttle system—both rolling stock and infrastructure—and simulate its behaviour. This lets us identify the risks of new technologies before we ever draft the specifications.

ARCADIA is recognized by AFNOR, the French standards authority. With this shared method, we can put system architecture and collaboration at the heart of systems engineering. That lets us break down the silos separating safety, security and integrated logistics support, and encourage these specialized engineering fields to work together.

And our shared tool is Capella, an open-source software program that supports our common language and standardized method.

“But a language, a method and a tool aren’t enough,” warns Marc Sango. “If you want them to be widely adopted by engineers in key fields, you also need a library of reusable frames and instructions for using them.” These resources will let us optimize risk analysis, which is currently done “by hand” with office automation tools.

In short, the task is to “derisk” new technologies and identify ways of using them that will create genuine value for the Group in the future. And that’s something we look forward to.

Yes. We’re trying to solve interoperability problems between countries that have different operating rules. We partner with SNCF Réseau’s Industrial & Engineering Division as part of the Shift2Rail and LinX4Rail initiatives. We also work with SNCF’s Interoperability, Standardization & Research – Europe Division as part of the System Pillar programme under the Europe’s Rail Joint Undertaking. Through these initiatives, we plan to establish new European standards that we can use to design the rail system architecture of tomorrow. That’s why we’re adopting innovative methods such as model-based systems engineering.

Learn more about LinX4Rail

Learn more about the ERJU/System Pillar programme

We took the initiative on creating the first version of rail system architecture modelling, which incorporates all other command-and-control and signalling subsystems. And that’s just one example.

We also proposed a shared European platform for system architecture modelling. Europe’s Rail is now studying ways of acquiring and rolling out a platform along these lines.

Under the Research4Future programme, we’re working to identify new methods and other emerging technologies and “derisk” them. After that, we plan to share them within SNCF Group’s innovation programmes, in our rolling stock engineering units and at SNCF Réseau. We’re also working with the self-driving freight train team on a proof of concept that will let us apply the project’s MBSE model to optimize risk analysis using a new method: model-based safety analysis.