AI applications have the potential to transform business models, but first they must overcome some structural challenges

Innotrans 2024, which took place from September 24 to 27 in Berlin, was a “fireworks display of innovations” in all areas related to the railway industry. However, one of the star elements in this journey into the future has been digitalization, specifically the applicability of AI as an innovative and disruptive technology.

In most debates, stands and presentations, AI has been the main character, mainly because of its versatility: it adds value in optimizing the use of trains, in operational improvement of maintenance and in user experience. In fact, the exhibition had, for the first time, a specific area dedicated to AI and a tour called “AI Mobility Lab” which offered guided visits to exhibitors presenting their innovations in this field.

At the opening ceremony, under the title ‘From Hype to Reality: AI in the Mobility Sector’, representatives of leading operators and constructors discussed the opportunities and conditions for the successful use of AI. Through specific use cases, they explored the large window of opportunities offered by digitalization and this new technology: AI cameras that evaluate facilities and enable an appropriate sequence for maintenance, intelligent monitoring of asset status, predictive analytics applied to potential failures, unified international ticketing, occupancy-based pricing…

It should be noted, however, that to materialize these digitization efforts it is necessary to re-equip vehicles and address the physical refurbishment of the rail infrastructure. The fact that less than 10% of the suggested AI use cases have reached the production phase shows that AI still faces significant hurdles, although we can affirm that its use in the railway sector is already a reality.

One of the most important challenges is that of data management, ultimately the source of these AI tools. The quality of the data and the creation of a model that allows them to be shared is essential to obtain positive results. This requires the creation of reliable and secure environments in which interoperability between all the actors in the process is essential, even between competitors. This is the only way to maintain effective and ‘cybersecure’ control over the desired data and to guarantee the shielding of AI systems from the design phase.

In addition to the technical aspect, the accelerated development of AI also demands a legal response. It is essential to develop a clear and reliable regulatory framework for the use of AI that facilitates sectoral cooperation.

AI has the potential to generate movements in the sector’s business models, which in turn demands new competencies, capabilities, and skills. Cooperation with universities and training facilities that provide specialized and qualified personnel is therefore important. The railway sector is 150 years old, but it is facing a unique opportunity to transform itself into a very attractive sector for future professionals.

Adapting to the new reality represented by AI must involve all players in the railway sector: operators, infrastructure managers, builders and, of course, technology providers. As the spearhead of digital innovation, it is essential that technology providers equip our solutions and products with the necessary capabilities to enable intelligent self-diagnosis through analytical algorithms, interoperability with the systems it interacts with, and the security of the data and algorithms used.

Quique Sánchez Candorcio.

Railway & Industry business Director of Trebide