Prover

On Dec 4, the first HLL forum organized by Prover was held in Paris. The event brought together the major users of the HLL language.

The forum was an exciting opportunity to learn how HLL is used today, and to discuss how HLL can evolve in a controlled way to meet the challenges of tomorrow. We hope that HLL will thereby be stronger and more attractive for new users. Already today, HLL is used by several infrastructure managers and signaling suppliers to state safety requirements and make mathematical models of systems for formal verification. With an active standardization community, the language will gain even more credibility and become the natural choice for many more users.

We are already looking forward to the next HLL Forum to create a thriving ecosystem of end users, suppliers, tool vendors, and academia.

Share this article

Learn to build a solid safety case for rail control systems using formal verification

Fill out your information here.

Do you want news and upcoming events from Prover?

Fill out your information here.

More News & Articles

  • Why open signaling belongs in the first meeting, not the last

    When replacing relay interlockings, most conversations start with technology. The better starting point is governance: how do you make sure you can still change parts of the system 20 years from now? We look at what Open Signaling means in practice, and how our Stockholm Metro pilot was designed for replaceability from day one.

  • Prover, RATP and Systerel publish unified HLL 4.0 specification

    Prover, RATP and Systerel publish the unified HLL 4.0 specification, keeping the rail industry's formal verification language open and free of dialects.

  • CENELEC is amending EN 50716 for AI. Here is what the draft says

    Generative AI has entered railway software engineering ahead of the standards governing it. A new CENELEC draft amendment to EN 50716 doesn't add requirements. It clarifies which existing rules apply when AI drafts specs, writes code, or generates test cases, and why those rules are harder to satisfy with generative models than teams might assume.