TradSNCF: AI to help rail staff welcome Olympic Games travellers

| minute read

TradSNCF, rail operator SNCF’s AI-powered translation tool, enhances the travel experience for millions of passengers from around the world by ensuring smooth communication and optimal traveller information during and beyond the Paris Olympic and Paralympic Games.

Welcoming the Olympic and Paralympic Games in Paris is a great pride for the SNCF, but it posed significant logistical challenges, particularly in terms of mobility and traveller information.

For SNCF, this event was a unique opportunity to accelerate its innovation programmes and sustainably improve the travel experience.

AI-powered passenger traffic management

Each year around 80 million foreign tourists visit France, many of whom do not speak French. This is what led the SNCF to launch its project, to facilitate the mobility of all its customers, regardless of their language.

In stations, time and efficiency are crucial, both for travellers who want to reach their destination without obstacles and for SNCF staff who need to ensure the smooth flow of passengers. These two objectives can now be achieved with a quick and effective translation tool deployed to all 50,000 customer service staff.

TradSNCF is an instant translation solution that’s very ergonomic and easy to use. It has been developed using the best AI translation technologies.

It is a beautiful illustration of our capacity to innovate, which will remain as a 'legacy of the Olympic and Paralympic Games'. Laaziz Aouaa, Project Manager, Digital Department, SNCF Group

The idea is simple: a traveller at a station or on a train asks a question in their language, it’s then automatically translated into French for the staff, and their response is returned in the traveller’s language.

TradSNCF is a concentration of innovation, equipped with algorithms specially trained to best respond to the huge variety of situations encountered in stations.

The solution allows dialogue in 130 languages. It includes over 3,000 words or expressions related to the railway and transport domain in the most common languages, to avoid any misinterpretation. It’s available on mobile devices, smartphones, and tablets.

Sopra Steria masterminding development

Sopra Steria Next supervised the programming, customisation, user experience, and ergonomic aspects of the solution's development, as well as its deployment, while ensuring the training of SNCF teams and change management.

The solution was developed in less than six months, thanks to a previous collaboration with France Travail and the creation of TradEmploi, a multilingual welcome tool for job seekers developed by Sopra Steria in Open Source.

To choose the best solution, SNCF conducted an experiment to compare TradSNCF's performance with other available tools on the market. This experiment took place during the Rugby World Cup, held in France in September and October 2023.

SNCF and Sopra Steria Next then launched a five-week pilot with over 280 beta testers across France, aiming to test the solution's performance and enrich its vocabulary for customer service and traveller information.

The project finally moved to the appropriation phase, organising physical or online meetings with station staff across France to support their adoption of the tool and train them in its use.

AI translation technologies are learning technologies. Their adoption requires specific support to build trust in the tool and make the most of the technology. Rémi Weiller, Director, Transport Consultancy, Sopra Steria Next

New functionalities turbocharge AI potential

Thanks to TradSNCF, customer service staff now have a translation solution that perfectly meets travellers' needs, assisting them before, during, and after their journey.

A series of enhancements and developments are currently being studied, including:

  • The development of a dedicated app (rather than a web app),
  • The addition of an offline mode to support staff working in low-connectivity areas,
  • Automatic language recognition,
  • Document scanning functionality, and
  • The creation of a "favourite phrases" section.

Discussions are also underway to add new languages and continue improving accessibility and ergonomics.

The use of TradSNCF extends beyond interactions between station staff and travellers, it’s also used by community managers to respond to internet users and for contextual information displayed on station screens.

The solution has already been deployed at stations in key cities throughout France and will soon be deployed across the entire country to facilitate travel for all foreign tourists visiting the country each year.

A true mark of its commitment to welcoming visitors, TradSNCF perfectly illustrates SNCF's capacity for innovation in the service of travellers and will remain as a legacy of the Olympic and Paralympic Games.

TradSNCF perfectly illustrates the great diversity of Sopra Steria Group's expertise with active collaboration between specialised contributors: Integration, Infrastructure Management, Industry & Services, People & Change, and Public Service. Roxana Rotaru, Consultant, People & Change, Sopra Steria Next

ai-and-technology

Related content

Sopra Steria Next predicts AI market will double to $1,270 billion by 2028

Sopra Steria Next, the consultancy division of the Sopra Steria Group, a major player in European technology, today publishes a ground-breaking study of the artificial intelligence market.

How meemoo used ethical AI to unmask 3.3 million faces from Flanders’ past

We've partnered with meemoo, the Flemish Institute for Archives, to enhance accessibility and usability of digitised cultural heritage items through AI-driven metadata enrichment.

Big Data London 2023: Navigating the World of Generative AI and LLMs

At the Big Data London 2023 conference we heard about the potential of Generative AI and Large Language Models (LLMs) and explored the current challenges they face. Read more about Vince's findings from the conference here.