Program

In this presentation we discuss how systems with Artificial Intelligence (AI), in particular Machine Learning (ML), can undergo safety assessment. This is relevant, if ML is used in safety related applications e. g. railway automation. Taking a deeper look into ML models, we argue that many models of machine learning are statistical models. Safety assessment would then have to concentrate on the model that is used in ML, besides the normal assessment procedure. Part of the budget of dangerous random failures for the relevant safety integrity level needs to be used for the random failure behavior of the ML system. We demonstrate our thoughts with a simple example and propose a research challenge that may be decisive for the use of ML in safety-related systems.

Jens Braband joined Siemens Mobility in 1993 after a PhD in Stochastic Modeling. As a key principal expert he is responsible for RAMSS in Railway Automation. He is accredited as an independent safety assessor by the German Federal Railway Authority (EBA). At TU Braunschweig he lectures and researches as a honorary professor for Risk and Safety Analysis of Transportation Systems.

I will discuss why federating knowledge in the field of R&I is key and in particular for the development of AI application for rail. I will share the experience of Shift2Rail Joint Undertaking, the first and biggest EU R&I integrated Programme, and highlight ideas for the next steps. Some of the challenges addressed will be how to create an effective test bed for innovation and where AI fit with railways. I will provide a high-level view of the possible future rail system that use artificial intelligence to optimise its system of systems.

Giorgio Travaini as Shift2Rail Head of Research & Innovation is responsible for coordination of the JU’s programme with a view to ensuring that the activities contribute to the fulfilment of the objectives of the S2R Master Plan. Prior to this posting, Giorgio Travaini was Senior Corporate advisor at UNIFE – the association of the European rail industry – advising the General Director on strategic topics, and liaising with the Shift2Rail Founding Members. He had been coordinating since 2010 the preparation phase of Shift2Rail, that lead to the JU regulation proposals of the European Commission. Since 2007, Giorgio Travaini was responsible as Senior Technical Manager for the supervision and coordination of more than 30 successfully EU-funded research projects, the UNIFE dedicated team and technical committees. He had also been significantly involved in ERRAC – the European Technology Platform for Rail Research – where he covered the secretary role. In 2006, Giorgio Travaini assisted Mr Paolo Costa, Chairman of the Transport and Tourism Committee of the European Parliament, in all legislative dossiers both in Brussels and Strasbourg. He joined the EU environment in 2004, working for the European association of Chamber of Commerce in Brussels and for the Brussels Regional office representation of his native region – Valle d’Aosta (IT). Prior to that, Giorgio Travaini had other international experiences from 2001 in Rome working for Procter&Gamble and in Paris working for BNP Paribas Cardiff. Giorgio Travaini holds an Engineering Degree (Polytechnic of Turin) and a Master Degree in public affairs (London South Bank University).

Every day we rely on a wide range of machines, but every machine eventually breaks down unless it’s being maintained. Predictive maintenance lets you estimate when machine failure will occur. This way, you can plan maintenance in advance, better manage inventory, eliminate unplanned downtime, and maximize equipment lifetime.

Stefano received a Master’s Degree in Electrical Engineering at University of Bologna, Italy, in July 1995, and got a Post Graduate Advanced Degree in Information Technology at CEFRIEL, Polytechnic of Milan the same year. He’s been with MathWorks since 2005. After spending eight years as a Senior Application Engineer in the field of Signal Processing and Communication Systems, supporting companies in the Communications, Electronics, Semiconductors and Aerospace and Defense industry segments, Stefano is currently working with the Academia Group to help the top universities with the adoption of MathWorks tools for effective teaching and research. Before that, he worked with R&D labs in STMicroelectronics and Philips Research, where he dealt with the design and development of wireless communication and video processing systems. Stefano has also been Contract Professor with the University of Milano for three years, where he was teaching Transmission Theory for the Telecommunication Software Engineering Bachelor Degree.